SCRIP: Scholarly Research In Progress 2019 by GeisingerCollege

Volume 3 â&#x20AC;˘ November 2019

525 Pine St. Scranton, PA 18509

570-504-7000

geisinger.edu/gcsom StudentResearch@som.geisinger.edu

Scholarly Research In Progress

Scholarly Research In Progress • Vol. 3, November 2019

Table of contents 2.

Trends in the Acquisition and Distribution of Opioid Drugs among 10 Contiguous States Kristin Feickert, Ceilia Severini, Elizabeth Stackhouse, Michael Belko, Christine Murphy, Alex Mettler, and Zoe Landau

Kathryn T. Volarich, Steven Picozzo, Jacob C. Arnold, and Brian J. Piper

Johanna Dungca

John Orr

Mikael Horissian, Anne Horissian, and Elizabeth Kuchinski

Marc Incitti and Walter DelGaudio

Tian L. Mauer, Anna Bukowski, Maxwell Gruber, and Vikram Siberry

Joshua Emerson Kiddish

40. A Review of Genetic Markers Associated with Penile Cancer

Andrew Denisenko

46. Profile of Dispensary Patients that Substitute Cannabis for Alcohol

Assad Hayat and Brian J. Piper

52. Proliferative Retinopathy Associated with a Case of Hemoglobin C Trait

Karl M. Andersen and Randall R. Peairs

55. Transulnar Approach for Balloon Aortic Valvuloplasty for Severe Aortic Stenosis: A Novel Approach Despite Multiple Comorbidities

Alexandra Cruz-Mullane, Jaclyn C. Podd, Stephanie D. Nichols, Kenneth L. McCall, and Brian J. Piper

36. Understanding the Link Between Energy Production, Symbiosis, and Cancer Metabolism: A Brief Review of Recent Research

28. HPV Vaccination: A Globalized Paradigm

Kristin D. Feickert and Stephanie E. England

Shijo Benjamin and Mary Taglieri

121. Buprenorphine Distribution between 2007 and 2017 in the United States

131. Community-Based Survey to Assess Prescription Affordability in Lackawanna County, Pennsylvania

26. Retroperitoneal Sarcoma in a 55-year-old Female Treated with Immunotherapy

72. Use of Venous Blood Gases for Management of Acid-Base Status in Patients with Severe Septic Shock

21. Physician Use of Patient-Centered Communication: Impact on the Patient Experience

61. Medical Cocaine Use from 2006– 2017 and its Application in U.S. Medicine

Stephen Long, Patrick Roman, Bradley Very, Stephen Voyce, and Yassir Nawaz

17. Evaluation of the Benefits of Skin Cancer Community Health Day

Eric A. Ligotski, Laura M. Loeser, and Alison T. Varano

65. Alternative Treatments for Major Depressive Disorder

14. A Review of Playground Surfacing in Relation to Pediatric Injury

117. Assessing the Effectiveness of the Murine Model in Advancing X-Linked Agammaglobulinemia Research

Entrapment and Fracture of an Irretrievable FFR Pressure Guidewire in a Myocardial Bridge Segment: A Case Report and Review of the Literature

10. Conflict of Interest Disclosure Accuracy among High-Impact Medical Journals

57. Risk Factors Associated with Opioid Overdose Deaths across 30 Counties in Pennsylvania for 2018

Alexandra Cruz-Mullane, Michael A. Freeman, Amalie K. Kropp, and Shane P. Ruddy

Garrison Davis, Marc Incitti, Spencer Davis, and Khaled Sorour

76. A Review of the Effect of Maternal Obesity on Maternal and Fetal Health

Oluwaseyi Olulana

80. Trends in Buprenorphine Prescription in the United States from 2008–2017

Amir R. Pashmineh Azar, Warren S.L. Lam, Suhail H. Kaleem, Laura B. Lockard, Mark R. Mandel, and Brian J. Piper

86. Transcutaneous Vagus Nerve Stimulation for Treatment-Resistant Depression

Brianna Dade, Tina Giutashvili, and Christina Michel

91. Understanding the Opioid Crisis in the Southern Half of the United States

Manraj Sahota and Makayla E. Boyle

98. Using Implementation Science to Address the Opioid Crisis: Deciphering Vocabulary, Adopting Sustainable Practices, and Overcoming Challenges

Niraj J. Vyas

105. Trichostasis Spinulosa Masquerading as Hypertrichosis and Presenting at an Unusual Site in a 13-year-old Female

Kendall Shifflett, Howard Pride, Matthew Palmer, and Eric Hossler

107. Controlled Substance Distribution in West Virginia from 2006 to 2017

Julius A. Hatcher IV, Sneha Vaddadi, and Bradley D. Nafziger

112. Tea as Treatment: Applications of the Kava Plant (Piper methysticum) in Modern Medicine

126. The Cellular Hallmarks of Multiple Sclerosis

Ashley A. Bross, Laura M. Loeser, Elizabeth J. Pavis and Alison T. Varano

Laina Gagliardi, Amalie Kropp, Alice Thompson, Danielle LaPointe, Terrence Habiyaremye, Sneha Vaddadi, Katarina Smigoc, Elizabeth Stackhouse, Michael Belko, Jasmine Santos, and Jennifer Joyce

135. Medication Prescribing Patterns in Patients with Hidradenitis Suppurativa: A Population-Based Study

Andrea J. Borba and Pierce H. Deng

139. Assessment Using the Sorour Airway Visualization Evaluation (SAVE) Score Predicts Difficulty of Tracheal Intubation via an InterObserver Reliability Study

Garrison Davis, Connor Magura, Conor Lynch, Stephanie Tilberry, Sahil Pandya, Julia Shamis, Sharmeen Mian, and Khaled Sorour

143. Assessing Adverse Childhood Experience Education for an Interprofessional Audience

Jacob Arnold, Alayna Craig-Lucas, Mark White, and F. Dennis Dawgert

147. Opioid Mortality Following Implementation of Medical Marijuana Programs in the United States

Daniel E. Kaufman, Asawer M. Nihal, and Janan D. Leppo

151. Prescription Opioid Distribution Before and After Legalization of Recreational Marijuana in Colorado, 2007–2017

Finding Your Way: Opportunities for Student Funding If you are looking for funding opportunities specifically designed for students, you can find assistance at the Office of Research and Scholarship. Funding opportunities can include fellowships, internships, research, programming and collaboration. The Office of Research and Scholarship can help you locate and qualify for funding opportunities, as well as assist in application prep, budgeting and editing. We are here to help you every step of the way! School policy requires that applications are submitted by our office, so call or stop by early so that we can meet your deadline.

Geisinger Commonwealth Student Research Awards The Office of Research and Scholarship is pleased to announce the availability of funds for the 2019–2020 academic year to support student research projects in the areas of basic or clinical science, public/community health, behavioral health and medical education research. The proposed project must be under the supervision of a faculty mentor and be endorsed by the Office of Research and Scholarship. The proposed project period must be no longer than six months and conclude by June 1, 2020. The maximum award for each project is $1,500. Funds cannot be requested for stipends, tuition, travel or wages for the student or faculty mentor. Indirect costs to the sponsoring institution are not allowed. Student Research Awards (SRAs) are intended to foster student scholarship and lead to a tangible deliverable, like an abstract for submission to a regional/national meeting or a manuscript for publication in SCRIP and/or a peerreviewed journal. SRA applications are due Nov. 1, 2019, at 11:59 p.m. EST. Contact Katie Pasqualichio or StudentResearch@som.geisinger.edu if you are interested in applying.

Contact Information:

Amalie K. Kropp, Stephanie D. Nichols, Daniel Y. Chung, Kenneth L. McCall, and Brian J. Piper

158. 2020 Summer Research Immersion Program 159. Finding Your Way: Opportunities for Student Funding

Everett M. Blough

Bradley Very, Stephen Long, Stephen Voyce, and Yassir Nawaz

Katie Pasqualichio, Grants Specialist

Scholarly Research In Progress

Office of Research and Scholarship Phone: 570-558-3955 Internal ext: 5335 Email: kpasqualichio@som.geisinger.edu Geisinger Commonwealth School of Medicine is committed to non-discrimination in all employment and educational opportunities.

159

A message from the editor-in-chief As the Journal of Scholarly Research in Progress (SCRIP) enters its third year of publication, I would like to offer thanks to our readers, to our contributors, to our faculty reviewers and to our student editors for their support of the journal and its mission: to promote and disseminate student scholarly activity at Geisinger Commonwealth School of Medicine. In the last three years, SCRIP has seen unprecedented growth in the form of submissions and accepted published works. This year we received 46 quality submissions from our students— more than four times the number of manuscripts that we received for the inaugural publication. This year’s submissions include literature reviews, case reports and original research manuscripts on topics ranging from balloon valvuloplasty for severe aortic stenosis to examining trends in the acquisition and distribution of opioids throughout the United States. I am heartened by the level of student interest in disseminating their research and scholarly work. In large part, this pursuit is due to the commitment of faculty throughout Geisinger who serve as mentors to our students and who recognize the importance of research and scholarship for their education and career development. Those submitting authors who have had their work accepted should be proud of their achievement. As always, I take the responsibility of sustaining and building upon the SCRIP’s quality and success very seriously. Suggestions from our contributors and readers to further develop and/or improve the journal are more than welcome. If you would like to share your thoughts, please email me at slobo01@som.geisinger.edu. Lastly, I would like to take this opportunity to invite students who have an interest in being involved in the editorial work of the journal. The role of the student editor is to assist in the writing, peer review, publication and editing of content for the journal. It is a great way to build your academic scholarship portfolio and I believe that this would help to ensure the journal’s growth and sustainability. Potential candidates may send their CV (including all relevant research and/or creative scholarship experience, as well as all relevant writing, editing or peer critique experience) to scrip@som.geisinger.edu with the subject “Application for Student Editor.”

Student editors Vaibhav Sharma, MD Class of 2022 Jacob Arnold, MD Class of 2022

Acknowledgments The SCRIP would not be possible without the contributions of faculty and student volunteers committed to the review and assessment of submitted articles. Their feedback provides student authors with an opportunity to strengthen their writing and to respond to critiques. We gratefully acknowledge the following faculty members for their support in providing peer review. Kathleen Doane, PhD Eric Wright, PharmD, MPH Kimberly Miller, PharmD Marika Handakas, LCSW, PsyD Dan Sylvestre, BS Mathangi Rajaram-Gilkes, MBBS, MSc, MS, MEd Anthony Gillott, MD, FACS Gabi Waite, PhD Vicki T. Sapp, PhD Michael Gionfriddo, PharmD, PhD Ying Ju Sung, PhD Mark White, MD, MPH Elizabeth Kuchinski, MPH John Arnott, PhD Mushfiq Tarafder, MPH, MBBS James Caggiano, MD, FAAP Brian Piper, PhD Vaibhav Sharma, BS Diana Callender, MBBS, DM Cyamatare Felix Rwabukwisi, MD, MPH James Caggiano, MD, FAAP Christian Carbe, PhD Youssef Soliman, MD, PhD William McLaughlin, PhD Cathy Wilcox, PhD Mary Triano, MSN, CRNP-C Jennifer Boardman, PhD Carmine Cerra, MD David Averill, PhD

Office of Research & Scholarship Phone: 570-504-9662 Sonia Lobo, PhD, RYT Associate Dean for Research & Scholarship Professor of Biochemistry Michele Lemoncelli Administrative Assistant to the Associate Dean for Research & Scholarship Katie Pasqualichio Grants Specialist Laura E. Mayeski MT(ASCP), MHA Manager, Research Compliance Thomas Majernick, MS Manager, Research Education Resources

On the cover:

Sincerely,

Sonia Lobo, PhD Editor-in-Chief

The cover image shows β3-tubulin in a tibial nerve sample from an frm1 knockout rat model under confocal microscopy. The image was acquired by Cecelia Allison and River Jordan, Doctor of Medicine candidates, class of 2022.

Volume 3 • November 2019

Scholarly Research In Progress

Scholarly Research In Progress • Vol. 3, November 2019

Trends in the Acquisition and Distribution of Opioid Drugs among 10 Contiguous States Kristin Feickert1*, Ceilia Severini1*, Elizabeth Stackhouse1†, Michael Belko1†, Christine Murphy1†, Alex Mettler1†, and Zoe Landau1†

1 Geisinger Commonwealth School of Medicine, Scranton, PA 18509 *Master of Biomedical Sciences Program †Doctor of Medicine Program Correspondence: kfeickert@som.geisinger.edu

Abstract The opioid epidemic is an ongoing crisis that individuals and nations have been battling for more than a decade. Starting in the early 1990s, the situation has since escalated and evolved significantly. In light of the serious threat to public health, the present study’s aim was to determine patterns of retail distribution for several prescription opioids with respect to certain states. Specifically, opioid distribution from 2007 through 2017 for 19 different opioid medications across 10 contiguous states (Delaware, Indiana, Kentucky, Maryland, Michigan, New Jersey, New York, Ohio, Pennsylvania, and West Virginia) were examined. Drug weights were obtained from the U.S. Automation of Reports and Consolidated Orders System (ARCOS), population data was extracted from the U.S. Census Bureau, and oral morphine milligram equivalent (MME) conversion factors were taken from the U.S. Centers for Disease Control and Prevention (CDC). Over the studied period, there was a downward trend in legal acquisition/ distribution of the same 11 opioids post 2011/2012. Statewise, Kentucky had the highest distribution of hydrocodone and New Jersey had the lowest. For morphine and oxycodone, Delaware had the highest distribution while New York and Michigan had the lowest distribution. Summation of 11 analgesic opioids revealed that West Virginia ranked highest for opioid distribution in 2007; this was replaced by Delaware in 2011 and 2017. From 2009 through 2011, Delaware experienced a 21.50% increase followed by a 14.31% decrease in opioid distribution (2011 to 2012). Our data supports a steady overall decline in opioid distribution in the mid to late century. This decline is thought to reflect a growing awareness of and effort to address the opioid epidemic in the United States.

Introduction Opioids have garnered nationwide attention and have been deemed a public health crisis and/or national emergency. An estimated 130 Americans die each day from an opioid overdose (1). Pennsylvania and West Virginia have been particularly impacted by this threat. In 2017, West Virginia had the highest rate of drug overdose deaths (57.8 per 100 000), followed by Ohio (46.3 per 100 000), and then Pennsylvania (44.3 per 100 000) (2). This affected Pennsylvanians in both large cities and rural communities. In light of this, Pennsylvania implemented policy changes aimed at reducing the damage that this epidemic has had on its residents. One example is Pennsylvania’s Prescription Drug Monitoring Program (PA PDMP), which collects information on all filled prescriptions for controlled subjects (3). Through this program, health

care providers can safely give controlled substances while helping patients get the treatment they need. Most recently, Pennsylvania's Secretary of Health signed an updated standing order for prescriptions of naloxone (an opioid antagonist used to revive those who have overdosed), which includes the 2-mg dose auto-injector (3). Another initiative is through the Pennsylvania Department of Health, which released a non-opioid directive. This allows patients to formally communicate that they do not wish to receive opioids as part of their treatment plan (3). The benefit of the directive is that it gives health care providers, prescribers, and patients the option to investigate alternative treatment methods. It also opens the door to discussing past substance use or misuse and may prevent cases of inadvertently prescribing controlled substances to those who could be unfavorably affected (3). In situations of substance abuse, warm handoff is a technique that is already being used in Pennsylvania. After making a direct referral into substanceabuse treatment, a health care provider does a face-to-face introduction between the patient and a substance-abuse specialist before treatment begins (3). Finally, action is being taken to update prescribing laws and there is always continuing education for health care providers (3). Besides Pennsylvania, there are general state policies that help reduce opioid devastation. These include permitting syringe exchange programs and upholding good Samaritan laws, which provide legal protection to bystanders calling in an opioid overdose (4). Despite these positive measures, both West Virginia and Pennsylvania continue to struggle with an opioid epidemic. Therefore, the aim of this study was to examine the pattern of opioid distribution between 2007 and 2017 for 19 different opioid medications across 10 contiguous states. The goal was to see how opioid distribution changed as a function of time and state. With a better understanding of patterns of use, perhaps public policy could be shaped to further mitigate the worst outcomes of the opioid epidemic locally and nationwide.

Materials and Methods Procedures Opioid retail drug distribution by weight was extracted for 10 states (Delaware, Indiana, Kentucky, Maryland, Michigan, New Jersey, New York, Ohio, Pennsylvania, and West Virginia) from the U.S. Department of Justice’s reporting system, ARCOS, for the years 2007 through 2017. Raw data was obtained for 19 drugs (alfentanil, buprenorphine, codeine, dihydrocodeine,

Trends in the Acquisition and Distribution of Opioid Drugs

fentanyl base, hydrocodone, hydromorphone, levorphanol, meperidine, methadone, morphine, noroxymorphone, opium powder, opium tincture, oxycodone, oxymorphone, remifentanil, sufentanil base, and tapentadol). Population data per state was obtained through the U.S. Census Bureau and MME was taken from the Centers for Disease Control (CDC). All graphs were constructed using Prism 8. Procedures were deemed exempt from review by the IRB of the University of New England. Data analysis Raw opioid weight (g) was converted to MME and population corrected. MME conversion factors can be viewed in Table 1 (5). Eleven of the original 19 drugs were then used to construct figures (codeine, dihydrocodeine, fentanyl base, hydrocodone, hydromorphone, levorphanol, meperidine, morphine, oxycodone, oxymorphone, and tapentadol). Certain drugs were excluded because of incomplete data, MME/ population=0, and/or if the drug was not used for mostly analgesic purposes. Years 2007 and 2008 did not include tapentadol, as no data were reported for these years in the ARCOS system. The highest and lowest state distribution was determined for hydrocodone, morphine, and oxycodone by graphically plotting MME/population as a function of time in Excel. Final figures contain the highest and lowest state trendlines as well as Pennsylvania and West Virginia for comparison. Standard error of the mean was calculated via Excel.

Figure 1. Total distribution of 11 opioids by state for 2007

Results Overall opioid use by state was determined for years 2007, 2011, and 2017 by summing the 11 aforementioned opioids (Figures 1, 2, and 3). In 2007, West Virginia had the highest

Figure 2. Total distribution of 11 opioids by state for 2011

Table 1. MME conversion factors

Table 2. Percent increase/decrease of opioid distribution by state

Figure 3. Total distribution of 11 opioids by state for 2017

Trends in the Acquisition and Distribution of Opioid Drugs

opioid distribution. It was replaced by Delaware in 2011 and 2017, where it fell to second and third place, respectively. Pennsylvania experienced the opposite trend. It went from fifth place in 2007 to second place in 2017.

Figure 4. Average distribution of 11 opioids by state

Figure 5. Distribution of hydrocodone as a function of time

The same 11 opioids were averaged by year and state in order to obtain Figure 4. Delaware had the highest opioid distribution and the most notable peak when compared to the other nine states. It had a 21.50% increase from 2009 to 2011 and then a 14.31% decrease from 2011 to 2012. Other noteworthy state increases/decreases are shown in Table 2. States not listed experienced only moderate fluctuations in opioid distribution. Delaware’s maximum distribution occurred in 2011; other states achieved their maximum distribution the following year in 2012. The acquisition of individual opioids was then examined with specific reference to local use. West Virginia and Pennsylvania were selected to represent local patterns of use. In addition, the states with the highest and lowest overall opioid use were also included. The latter two states were depicted graphically alongside local counterparts and varied as a function of prescription opioid. The use of hydrocodone, oxycodone, and morphine can be seen in Figures 5–7. Kentucky had the highest and New Jersey the lowest distribution of hydrocodone. Delaware had the highest distribution of oxycodone and morphine while Michigan and New York had the lowest. Distribution for hydrocodone and morphine peaked in 2012 while oxycodone was more variable depending on the state.

Discussion Opioid use across the United States evolved significantly between the years 2007 and 2017. Despite collecting raw data for 19 different opioids, we chose to focus on the CDC’s list of commonly abused prescription opioids: oxycodone (OxyContin), hydrocodone (Vicodin), and morphine (6). Hydrocodone and morphine’s peak earlier in the course of the decade (on average in 2012) occurred because physicians were prescribing more opioids. The number of opioid prescriptions increased from 112 million to a peak of 282 million in 2012 (9).

Figure 6. Distribution of oxycodone as a function of time

Figure 7. Distribution of morphine as a function of time

Furthermore, the cumulative effect of multiple opioids indicated that West Virginia’s distribution peaked in 2007. This could be attributed to more opioid prescriptions being written or more distributors opening at this time. In 2017, providers in West Virginia wrote 81.3 opioid prescriptions for every 100 people, compared to the average U.S. rate of 58.7 prescriptions (7). This rate was among the top 10 in the U.S. for that year, but it was the lowest rate in the state of West Virginia since 2006 (7). Therefore, despite the higher than average prescribing rate, this decrease relative to other years helps explains the record low in 2017. Unlike West Virginia, Delaware experienced high distribution in 2011 and 2017. This may be due to inadequate access to treatment for those suffering from opioid use disorder. People in a crisis need to be connected to appropriate services in a timely manner and those who have opioid use disorder often need support with basic needs such as employment and housing (8). Finally, Pennsylvania’s distribution increase in comparison to other states could be due to greater public demand and lingering

Trends in the Acquisition and Distribution of Opioid Drugs

misconceptions about opioids. With the advent of synthetic drugs, it is now easier than ever to gain access to and develop opioid dependence. Pharmaceutical companies further complicate the matter by falsely advertising their products as nonaddictive and appropriate for chronic pain. Recently, Pennsylvania Attorney General Josh Shapiro filed a lawsuit against Purdue Pharma for these very reasons (10).

10. Ly L. Pennsylvania attorney general sues Purdue Pharma over opioid epidemic [Internet]. [updated 2019 May 14; cited 2019 May 15]. Available from https://www.cnn. com/2019/05/14/health/pennsylvania-ag-purdue-lawsuitopioids/index.html.

In conclusion, all 10 states experienced a declining trend in opioid distribution with time after 2011/2012. This decline is most likely the result of policy changes, more vigilant monitoring, and growing awareness about opioid risk factors. These changes have been beneficial, but more innovation is needed to help properly control as well as manage the opioid crisis.

Acknowledgments Thanks to feedback from Brian Piper, PhD, MS, at Geisinger Commonwealth School of Medicine and Fahs Beck for funding support.

References 1.

Centers for Disease Control and Prevention. Understanding the epidemic [Internet]. [updated 2018 Dec 19; cited 2019 Dec 19]. Available from https://www.cdc.gov/ drugoverdose/epidemic/index.html.

2. Centers for Disease Control and Prevention. Drug overdose deaths [Internet]. [updated 2018 Dec 19; cited 2019 May 16]. Available from https://www.cdc.gov/ drugoverdose/data/statedeaths.html. 3. Commonwealth of Pennsylvania Department of Health. Opioid epidemic [Internet]. [cited 2019 May 06]. Available from https://www.health.pa.gov/topics/disease/Opioids/ Pages/Opioids.aspx. 4. AmfAR Opioid & Health Indicators Database. West Virginia opioid epidemic [Internet]. [cited 2019 May 07]. Available from https://opioid.amfar.org/WV. 5. Centers for Disease Control and Prevention. Data resources [Internet]. [updated 2019 Apr 10; cited 2018 Oct 19]. Available from https://www.cdc.gov/drugoverdose/ resources/data.html. 6. Centers for Disease Control and Prevention. Opioid basics [Internet]. [updated 2018 Dec 19; cited 2019 May 16]. Available from https://www.cdc.gov/drugoverdose/opioids/ index.html. 7.

National Institute on Drug Abuse. West Virginia opioid summary [Internet]. [updated 2019 Mar; cited 2019 May15]. Available from https://www.drugabuse.gov/opioidsummaries-by-state/west-virginia-opioid-summary.

8. Saloner B, Bachhuber M, Barry CL, Krawczyk N, Pasha O, Sen AP, et al. A blueprint for transforming opioid use disorder treatment in Delaware [Internet]. [cited May 15]. Available from https://dhss.delaware.gov/dhss/files/ johnshopkinsrep.pdf. 9. CNN. Opioid crisis fast facts [Internet]. [updated 2019 Apr 11; cited 2019 May 15]. Available from https://www.cnn. com/2017/09/18/health/opioid-crisis-fast-facts/index.html.

Scholarly Research In Progress • Vol. 3, November 2019

Entrapment and Fracture of an Irretrievable FFR Pressure Guidewire in a Myocardial Bridge Segment: A Case Report and Review of the Literature Stephen Long1†, Patrick Roman1†, Bradley Very1†, Stephen Voyce2, and Yassir Nawaz2 Geisinger Commonwealth School of Medicine, Scranton, PA 18509 Geisinger Community Medical Center, Scranton, PA 18510 †Doctor of Medicine Program Correspondence: slong01@som.geisinger.edu 1

Abstract

Case Presentation

Physiologic measurements such as fractional flow reserve (FFR) or instantaneous wave-free ratio (iFR) have become adjuvants to traditional coronary angiography in the assessment of coronary artery disease and evaluation of the need for coronary revascularization. As these techniques have been employed in more and more complex coronary artery lesions, several case reports have documented facture and retainment of FFR/iFR guidewires. This case is the first reported of entrapment of a fractured FFR guidewire in a myocardial bridge segment of a coronary artery during diagnostic catheterization with subsequent inability to completely retrieve the wire percutaneously or during emergent coronary bypass graft surgery (CABG).

A 75-year-old male with a past medical history of diabetes and hypertension presented as an outpatient to Geisinger Cardiology for evaluation of several months of exertional chest pain and shortness of breath. A previous exercise treadmill stress test showed equivocal ST changes in inferior leads, but no evidence of myocardial ischemia. Due to persistent typical symptoms despite optimal medical management, an elective cardiac catheterization was recommended and conducted at Geisinger Community Medical Center. Vascular access was obtained via the right radial artery. Diagnostic angiography, performed with a 5 French (F) Tiger (Terumo, Somerset, N.J.) catheter, showed 50–60% distal left main coronary artery stenosis extending into the proximal left anterior descending artery (Figure 1) and an area of moderate myocardial bridging in the mid LAD without angiographic stenosis (Figure 2).

Introduction Percutaneous coronary interventions have a reported complication rates of approximately 2%: the incidence of retained interventional equipment is approximately 0.1–0.8% (1). Measuring coronary physiology by either fractional flow reserve (FFR) or instantaneous wave-free ratio (iFR) has been shown to be a useful adjuvant to traditional angiography in determining the need for coronary revascularization (3, 4, 5). As a result, physiologic measurements have become routinely used in the evaluation of coronary artery disease and are associated with low complication rates. Myocardial bridging occurs when a segment of the epicardial coronary artery, usually the mid left anterior descending artery (LAD), dips down into the myocardium. This results in external compression, constriction, and occlusion of this arterial branch during myocardial systole. Myocardial bridging, which has been estimated to have a prevalence of almost 20% of the general population, is typically a benign condition but has been shown to rarely contribute to acute coronary syndrome (6). Little is known about the interaction between myocardial bridging anatomy and interventional equipment such as FFR guidewires. Rare case reports have described instances of fractured pressure wires including use of percutaneous retrieval techniques; however, there have been no reported cases describing entrapment of a fractured FFR guidewire in a myocardial bridge segment (7). We report a case of a fractured pressure guidewire during diagnostic catheterization with possible entrapment due to a myocardial bridge segment followed by subsequent inability to completely retrieve the wire percutaneously or at time of emergent coronary bypass graft surgery (CABG).

Figure 1. Angiography image showing 50–60% occlusive lesion in the left main coronary artery and extended into the proximal left anterior descending artery (arrow).

To determine the hemodynamic significance of left main coronary artery disease, iFR/FFR measurements of the left main coronary artery were performed. After adequate anticoagulation was achieved with heparin, a 5F Tiger catheter was used to advance a Volcano Verrata Plus Pressure (Phillips Volcano, San Diego, Calif.), wire across the left main disease into the mid LAD (Figure 3). The wire was advanced with no resistance or significant torqueing. The iFR measurement was borderline positive at 0.89 (normal>0.9), after which intravenous adenosine infusion was used to perform an FFR measurement. This yielded a significantly positive value of 0.72 consistent with hemodynamically significant left main disease (normal >1.0; FFR values less than 0.8 are indicative of myocardial ischemia). At the conclusion of the case, as minimal traction was applied to remove the pressure wire, the wire

Entrapment and Fracture of an Irretrievable FFR Pressure Guidewire

started to lengthen at the distal, radiopaque portion. The 10-mm radiopaque tip of the wire could be seen lodged in a septal perforator branch of the LAD within the myocardial bridge segment. From that area, the uncoiled thin wire was visualized to extend out to the aorta and completely fractured with the proximal portion in the right subclavian artery (Figure 4).

Figure 2. Angiography images demonstrating the presence of a myocardial bridge segment in the mid left anterior descending artery (brackets). Image B shows significant compression of the LAD during systole when compared to end diastole (A). Arrows indicate the point on EKG tracing that corresponds to the timing of each image.

Due to inability to deliver interventional devices through the 5F system, the decision was made to upgrade to a 6F EBU 3.0 (Medtronic, Dublin Ireland) guide catheter. An Ensnare device (Merit Medical, Jordan, UT) was used to internalize the subclavian end of the thin coil wire into the guide. As we attempted to withdraw the wire, it continued to slip from the Ensnare. Balloon occlusion technique with a 3.0 x 12mm non-compliant balloon (Medtronic, Dublin Ireland) was then attempted. The balloon was inflated inside the tip of the guide catheter in an attempt to trap the wire fragment. The very thin coil wire again slipped as this device was retracted. A gooseneck snare (Merit Medical, Jordan, Utah) was successfully used to snare the thin wire from inside the guide catheter, but while pulling the wire, the coil wire again broke and ultimately, only a 15-cm segment of the proximal coil was able to be retrieved (Figure 5).

Figure 5. Angiography image showing an attempt at snaring the uncoiled distal coil from the right subclavian artery. Figure 3. Angiography image showing the radiopaque tip of the FFR pressure guidewire in the left anterior descending artery.

Figure 4. Angiography images showing the fractured pressure wire lodged in the mid left anterior descending artery. Uncoiling of the distal portion of the wire resulted in marked shortening of radiopaque portion of the FFR wire.

The most proximal end of the coil fragment was now seen in the ascending aorta. At this time, discussion was held with CT surgery. The patient would require ultimately require CABG for his severe left main disease and the decision was made to abandon percutaneous retrieval attempts in order to avoid any injury to the aorta or left main coronary artery and to reduce the patientâ&#x20AC;&#x2122;s radiation and contrast exposure. The patient remained asymptomatic and hemodynamically stable, and the decision was made to surgically retrieve the wire at time of urgent CABG. At operation, a small aortotomy was made and the distal coil was seen protruding from the left main coronary artery into the ascending aorta. The surgeon pulled on the thin, uncoiled wire but noticed significant resistance and decided to cut the wire inside the left main to avoid additional arterial injury. The patient underwent two-vessel bypass to the left anterior descending and left circumflex coronary arteries with the left internal mammary artery anastomosed on the LAD distal to the retained wire segment.

Entrapment and Fracture of an Irretrievable FFR Pressure Guidewire

Figure 6. Post-procedure image of fractured FFR guidewire, showing uncoiled and fractured distal coil wire and fractured central core.

Figure 6 shows the broken wire tip once it was removed and visually inspected. The post-operative course was remarkable for post-operative paroxysmal atrial fibrillation treated with amiodarone and left upper extremity cellulitis with internal jugular and basilic vein thrombosis treated with antibiotics and warfarin. There were no additional cardiac sequelae. He was discharged on post-operative day 9. Anticoagulation and antiplatelet agents included warfarin and 325 mg daily aspirin. Postoperative echocardiogram showed a left ventricular ejection fraction of 55% and no evidence of pericardial effusion. The patientâ&#x20AC;&#x2122;s chest pain and shortness of breath resolved after CABG, and he has had no cardiac complaints at 1-, 3-, and 6-month follow-up.

Discussion Reports of retained equipment in a coronary artery during PCI or diagnostic catheterization are rare and there is a lack of evidence-based guidelines for managing such complications. The first cases of retained interventional guidewire occurred in the 1980s and were typically managed surgically (8). Despite technologic advances in catheterization equipment and technique, the incidence of retained guidewire has gone essentially unchanged, likely due to the increased complexity of coronary catheterization and revascularization procedures (8). It is believed that variations in coronary anatomy such as tortuosity and lesion complexity, increased wire manipulation

and ill-defined patient factors may contribute to the risk of wire fracture. To date, we are unaware of an association between the finding of coronary artery myocardial bridge and guidewire entrapment and fracture. Retained guidewires can be managed percutaneously, surgically, or conservatively, and prior case reports have described several percutaneous techniques that have been used to successfully remove retained equipment. Several percutaneous devices, including snares, filter wires, and retrieval forceps can aid in retrieval of retained devices. Coronary artery stent deployment has also been used successfully to trap retained guidewires along the vessel wall and are felt to reduce the chance of wire migration or thrombus formation (9). Surgical retrieval should be considered in the event of failed percutaneous removal when equipment retention is associated with persistent ischemia or guidewire protrusion into the aorta, which can serve as a nidus for subsequent thrombus formation (8). Conservative management, i.e., leaving the fractured guidewire in place, can be considered when the risk of surgical intervention outweighs potential benefit, such as when the retained fragment is small or located in a chronically occluded coronary vessel. Theoretically, guidewire fragments remaining in the coronary vasculature can serve as a nidus for thrombus formation, endothelial injury, subsequent plaque formation, and arterial narrowing. Moreover, cases of acute coronary syndrome related to retained guidewire fragments have been suspected (9).

Entrapment and Fracture of an Irretrievable FFR Pressure Guidewire

We report a case of a fractured and retained FFR wire fragment in a myocardial bridge segment. The wire fragment was unable to be completely removed by percutaneous or subsequent emergent surgical intervention. The mechanism of wire fracture and the difficulty in removing the wire fragment is unclear. Possibilities including defective wire manufacturing or the presence of myocardial bridge causing compression, entrapment, and fracture of the distal tip. Our case was managed with successful coronary artery bypass grafting distal to the retained wire segment once percutaneous and surgical techniques were unable to completely retrieve the guidewire in its entirety.

8. Singh D, Darbari A. Retrieval of rapped and broken guide wire with immediate rescue off-pump coronary bypass surgery. Interactive Cardiovascular and Thoracic Surgery 2014; 19:529–531. 9. Khan SM, Ho DW, Dinaram T, et al. Conservative management of broken guidewire: Case reports. SAGE Open Medical Case Reports. 2014; 2.

Conclusion Retained pressure wire fragments may be a possible complication of performing hemodynamic assessment of coronary artery disease. There is a lack of data guiding management of this complication. In general, any foreign body in the coronary vasculature should be retrieved, but conservative management can be considered in certain cases. Further investigations regarding the mechanisms of pressure wire fracture and identification of preferred removal techniques will aid decision-making in these challenging cases.

Acknowledgments We would like to thank Stephen Voyce, MD, for his guidance and extensive revisions during the writing of this case. Additionally, we would like to thank Yassir Nawaz, MD, the primary operator on this case, for allowing us to work on this case report and for his assistance in writing this report.

References 1.

Slicker K, Lane W, Oyetayo O, et al. Daily cardiac catheterization procedural volume and complications at an academic medical center. Cardiovasc Diagn Ther. 2016 Oct; 6(5):446–452.

2. Hartzler GO, Rutherford BD, McConahay DR. Retained percutaneous transluminal coronary angioplasty equipment components and their management. Am J Cardiol. 1987;60(16):1260–1264. 3. Tonino P, Bruyne B, Pijls N, et al. Fractional Flow Reserve versus Angiography for Guiding Percutaneous Coronary Intervention. N Engl J Med 2009;360:213–24. 4. Pijls N, Schaardenburgh P, Manoharan G, et al. Percutaneous Coronary Intervention of Functionally Nonsignificant Stenosis: 5-Year Follow-Up of the DEFER Study. JACC 2007;49(21):2105–11. 5. Pijls NHJ. Fractional flow reserve to guide coronary revascularization. Circ J. 2013;77:561–569. 6. Teragawa H, Oshita C, Ueda T. The Myocardial Bridge: Potential Influences on the Coronary Artery Vasculature. Clin Med Insights Cardiol. 2019 May 1;13:1179546819846493. 7.

Surhonne P, Mahla H, Bhairappa S, et al. Successful retrieval of fractures pressure wire tip (FFR) by hybrid technique. J Saudi Heart Assoc. 2015 APR;27(2)118–122.

Scholarly Research In Progress • Vol. 3, November 2019

Conflict of Interest Disclosure Accuracy among High-Impact Medical Journals Kathryn T. Volarich1†‡, Steven Picozzo1†‡, Jacob C. Arnold1†‡, and Brian J. Piper1,2

Geisinger Commonwealth School of Medicine, Scranton, PA 18509 Center for Pharmacy Innovation and Outcomes, Geisinger Precision Health Center, Forty Fort, PA 18704 †Doctor of Medicine Program ‡Authors contributed equally Correspondence: kvolarich@som.geisinger.edu 1

Abstract Background: Accurate conflict of interest disclosure is important for evidence-based medicine. The Physician Payments Sunshine Act was passed in the United States in 2010 as part of the Affordable Care Act. It requires financial payments made to physicians by pharmaceutical or medical device manufacturers to be reported to the Centers for Medicare and Medicaid Services. This information is maintained for public availability in an online database called Open Payments. Conflict of interest self-disclosure forms submitted to two major U.S. medical journals, The New England Journal of Medicine (NEJM) and The Journal of the American Medical Association (JAMA), were examined to review physician-reported financial relationships between physicians and companies. Discrepancies between selfreported and industry reported disclosures were identified. This investigation also identified shortcomings related to the ease of use of Open Payments. Methods: Original research articles published in NEJM (n=206) and JAMA (n=188) from Jan. 1, 2017, to Dec. 31, 2017, were identified, and the first and last author of each was examined. Authors that met inclusion criteria (n=119) were queried in the Open Payments database (https://openpaymentsdata.cms. gov/). Results: The 5 highest-compensated authors received $5.7 million in total. All 5 authors had at least one industry reported payment that was not reported to the journal (Min=$418, Max=$1.4 million). More than one-half of relevant remuneration (55.6%) was not self-reported. Conclusion: These preliminary results indicate that selfreporting may be an inadequate mechanism to communicate conflicts of interest to the readers of high-impact medical journals. Limitations in the Open Payments database and journal disclosure practices impact results and ease of use.

Introduction The Physician Payments Sunshine Act was enacted in the United States in 2010 as part of the Affordable Care Act. It requires financial payments made to physicians by pharmaceutical and medical device manufacturers to be reported to the Centers for Medicare and Medicaid Services (CMS). CMS publishes this data on the Open Payments website. Within the first 5 months of Open Payments’ commencement in August 2013, $430 million of general payments (i.e., payments not related to research funding) were issued by manufacturers to recipients (1). One of the primary

goals of Open Payments was to “prevent inappropriate influence on research, education and clinical decision-making” (2), however, the website alone cannot identify conflicts of interest that may result from financial relationships between companies and physicians. Using conflict of interest selfdisclosure forms submitted to two major U.S. medical journals, The New England Journal of Medicine (NEJM) and The Journal of the American Medical Association (JAMA), this ongoing research assessed the nature of payments made to physicians by drug and device manufacturers, identified discrepancies between Open Payments and author-reported disclosures, and evaluated shortcomings related to the ease of use of Open Payments in identifying these discrepancies. This study extends upon prior research that has quantified conflicts of interest in influential materials like the Diagnostic and Statistical Manual of Mental Disorders (3) and medical textbooks like Goodman and Gilman’s Pharmacological Basis of Therapeutics and Harrison’s Principles of Internal Medicine (4, 5).

Materials and Methods Procedures Original research articles published in NEJM (n=206) and JAMA (n=188) from Jan. 1, 2017, to Dec. 31, 2017, were identified, and the first and last author of each was examined. These journals were selected based on their wide readership and high 2017 impact factors (79.3 and 47.7). Articles were excluded if the first or last author was not a U.S.-based physician (MD or DO) or if either author did not have a record in Open Payments. Data reported to Open Payments for the period from 2013 to 2017 were compiled and compared to self-reported financial disclosures that had been submitted to NEJM or JAMA along with the authors’ published articles. A total of 31 articles from NEJM and 31 articles from JAMA met all criteria for inclusion, with a total of 119 unique authors. Data collection for the 119 authors included sex, specialty, journal(s) of publication, and yearly payment information, broken into general (defined by Open Payments as payments not associated with research), and research payments (adding together Open Payments’ categories for “payments that are associated with a research study,” and “funding for a research project or study where the physician is named as a principal investigator”) (2). The 5 authors with the highest total general payments over this timeframe were further examined, with their payments being sorted by reported category. Physician identity was protected by assigning a number (one to five).

Conflict of Interest Disclosure Accuracy among High-Impact Medical Journals

A payment was considered unrelated if: “It was not disclosed AND the company from which the payment originated does not offer a product which could broadly be considered related to the area of inquiry.” Table 1. Keywords. For each article attributed to the top 5 compensated physicians, keywords were determined to provide a general basis to distinguish between a financial relationship that is related or unrelated to the contents of the article.

The focus of the comparison of self-reported financial relationships was limited to those reported as general payments by Open Payments. We first examined the articles and created a list of “keywords” (Table 1) against which we compared the products offered by the companies (listed in a given author’s Open Payments data) when a payment did not match the disclosure. We then followed the below guides to categorize payments as undisclosed, indeterminate, disclosed, or unrelated. Our definitions are adapted from the International Committee of Medical Journal Editors (ICMJE) disclosure form used by both journals which states:

Study procedures were deemed exempt by the Geisinger Institutional Review Board.

Analysis Analyses were conducted using a 5- and 3-year interval. ICMJE guidelines require disclosure of relevant financial relationships from a period of 36 months prior to publication. For the purpose of this analysis, the 36-month period was counted back from either the date of submission of disclosure forms, the date acceptance of the article by the journal, or the date of publication of the article (the earliest known date for each respective publication was used). Figures were prepared with Excel.

“[authors] should disclose interactions with ANY entity that could be considered broadly relevant to the work. For example, if your article is about testing an epidermal growth factor receptor (EGFR) antagonist in lung cancer, you should report all associations with entities pursuing diagnostic or therapeutic strategies in cancer in general, not just in the area of EGFR or lung cancer (6).” A payment was considered undisclosed if: “The author received a payment during the relevant disclosure period that did not match any disclosures provided to the journal AND the company offers, or offered at the time of the payment, a product which could broadly be considered related to the area of inquiry.” A payment was considered indeterminate if: “The author received a payment during the relevant disclosure period that did not match any disclosures provided to the journal BUT the company was a subsidiary or parent company of a company listed on the disclosure AND/ OR if it could not be determined whether they offer, or offered at the time of the payment, a product which could broadly be considered related to the area of inquiry AND/OR the payment has been disputed.” A payment was considered disclosed if: “The author disclosed a payment from a company which matched the data from Open Payments.”

Figure 1. Payment Types of Top 5 Earners. Each bar represents the monetary sum of the individual physicians’ financial relationships in the period from 2013 through 2017 as reported in Open Payments. Different colors represent different types of payment based on the Open Payments categories for “Nature of Payment.” The largest monetary sum for any type of payment for any of the physicians is “Compensation for Services Other Than Consulting” for Dr. 1.

Conflict of Interest Disclosure Accuracy among High-Impact Medical Journals

Figure 2. A) Payment Amount by Disclosure Status. Different bar colors represent 1 of the 4 established categories of disclosure status, and different bar lengths represent the monetary sum of the individual physicians’ financial relationships that fall into those categories. Not every possible bar appears for every physician if the monetary sum is very minimal or nonexistent. The largest monetary sum for any category of disclosure status for any of the physicians is the “Undisclosed” category for Dr. 1. B) Totals by Disclosure Status. Each slice of the pie represents the monetary sum of all 5 physicians for 1 of the 4 established categories of disclosure status. The disclosure status with the largest monetary sum across all 5 physicians is “Undisclosed,” and the smallest sum is “Unrelated.”

Conflict of Interest Disclosure Accuracy among High-Impact Medical Journals

Results The 5 physicians with the largest Open Payments earnings from 2013 to 2017 were all male MDs, 40% from JAMA, and represented the following specialties, respectively: cardiovascular disease, hematology and oncology, endocrinology, critical care, and interventional cardiology. The total general payments for the top 5 earners over 5 years averaged $1.14 million, and these payments were categorized by payment type (Figure 1). Total general payments for the 36-month period that was determined according to the ICMJE guidelines ranged from $380,000 to $1.48 million. These payments were assigned a disclosure status (Figure 2A and 2B). The vast majority of compensation for two (40%) of the top 5 authors was undisclosed.

Discussion The key finding of this small study was that there were appreciable discrepancies between self-reported and industry reported conflicts of interest disclosure. If replicated, the policies employed by NEJM and JAMA may need to be reconsidered in light of these findings to determine the origins of these differences. Possibilities include inaccurate recall or subjective interpretations of what information is sufficiently relevant to be disclosed. These results are generally congruent regarding the magnitude and prevalence of compensation to these highly influential physician-scholars (4, 5). These findings are not intended to disparage authors or specific journals as collaborations with industry are important for biomedical innovation that could improve health. Our analysis was limited by several factors. NEJM provides readers with access to copies of the disclosure forms where the submission date for the forms is available. JAMA provides a disclosure statement to readers and, as of 2017, provides the date of the article’s acceptance. Therefore, there was some inconsistency in the reference dates for the 36-month reporting period which was outlined in our methods. Our exclusion criteria allowed for consistency of data collection at the expense of excluding the majority of original research articles posted in JAMA and NEJM during 2017. The data collected does not include non-physician authors (PhDs, RNs, PAs), international authors or physicians who co-authored with non-physician first/last authors. Since data was only collected for first and last authors, any payment trends observed here may not be representative of all physician authors publishing in JAMA or NEJM. Moreover, our data was derived from research published in two scientific journals over a 1-year publishing period, making it potentially difficult to extrapolate our results across other journals.

funding. Influential journals should consider disclosing amounts (e.g., >$10,000) or give a direct link to the Open Payments entry. Using the above reported data as a starting point, we intend to further examine each author individually not only for type of payment, but also for how accurately their conflict of interest disclosure forms match the publicly available data from Open Payments as well as ProPublica’s Dollars for Docs. This type of research is important to sustain the venerable reputation of journals like NEJM and JAMA and maintain the highest standards for evidence-based medicine.

Disclosures KV, SP and JA have no relevant disclosures. In the past 5 years, BJP has received travel reimbursement related to medical marijuana and the Hereditary Neuropathy Foundation, research supplies and travel from the National Institute of Drug Abuse and National Institute of Environmental Health Sciences, has a grant in review with Pfizer, and is on the advisory board (pro bono) for the Center for Wellness Leadership.

References 1.

Marshall DC, Jackson ME, Hattangadi-Gluth JA. Disclosure of industry payments to physicians: An epidemiologic analysis of early data from the Open Payments Program. Mayo Clin Proc. 2016;91(1):84–96.

2. Open Payments Data in Context [Internet]. Baltimore: Centers for Medicare & Medicaid Services. Available from: https://www.cms.gov/OpenPayments/About/OpenPayments-Data-in-Context.html 3. Cosgrove L, Krimsy S. A comparison of DSM-IV and DSM5 panel members financial associations with industry: A pernicious problem persists. PLoS Med. 2012;9:e1001190. 4. Piper BJ, Telku HM, Lambert DA. A quantitative analysis of undisclosed conflicts of interest in pharmacology textbooks. PLoS One. 2015;10(7):e0133261. 5. Piper BJ, et al. Undisclosed conflicts of interest among biomedical textbook authors. Am J Bioethics: Emp Bioethics. 2018;9(2):59–68. 6. ICMJE Form for Disclosure of Potential Conflicts of Interest. International Committee of Medical Journal Editors [Internet]. Available from: http://icmje.org/conflicts-ofinterest/

Although there is some degree of variation in the data, and the extreme score of one author may have disproportionately impacted results in our small sample, it has been noted that payments to authors in our sample were markedly higher than their peers based on annual general payment averages for most specialties. Given the growing national discussion around the influence of private-sector money in medical research (e.g., the inaccurate disclosures of Memorial Sloan Kettering Cancer Center oncologist José Baselga, MD, PhD), it is important to evaluate this trend and further examine the nature of this 13

Scholarly Research In Progress • Vol. 3, November 2019

A Review of Playground Surfacing in Relation to Pediatric Injury Johanna Dungca1†

Geisinger Commonwealth School of Medicine, Scranton, PA 18509 †Doctor of Medicine Program Correspondence: jdungca@som.geisinger.edu 1

Abstract Playgrounds provide children with an outdoor environment that facilitates their motor, social and creative development. However, playgrounds may also pose health hazards, ranging from minor lacerations to severe extremity fractures. Playground surfacing material may have a measurable effect on pediatric injury severity. However, many of the studies analyzing playground surfaces and injury rates have not been comprehensively reviewed. A literature review was conducted to assess the relationship between different playground surfaces and pediatric injury. While a number of studies comparing surface-specific injury rates point toward the benefits of sand-based surfacing, the studies are conflicting. Cofounding factors, such as weather and location may explain the conflict. Determining which playground surfaces correlate with high and low injury rates can help provide clinicians, particularly orthopaedic specialists, with injury-specific data that can tailor medical care and help predict post-injury outcomes within the pediatric population.

Introduction Playgrounds provide children with an outdoor environment that facilitates their motor, social, and creative development, but they also pose health hazards. Playground injuries range from minor lacerations to severe extremity fractures. Under a multinational context, playgrounds have been recorded as a significant cause of pediatric injuries and hospitalizations (1, 4, 3, 5, 9, 15). In the United States alone, more than 200 000 children incur a playground injury each year (1, 12, 8). A retrospective cohort study conducted in Switzerland evaluated children who sustained long-bone fractures between 2009 and 2011 (9). Of the 2,716 patients included, 2,840 fractures were recorded and 11% of those fractures occurred on the playground (9). In New Zealand 15% of children who visit the emergency department for a playground injury are hospitalized each year, while 48,000 playground-related injuries are recorded annually by the Leisure Accident Surveillance System of England (5, 14). In Canada, approximately 25,000 children receive some form of emergency room treatment for playground injuries annually (4). These results suggest the widespread, global effect of playground play on children’s health and the need for public safety improvements. The risks associated with playground play call for attention to playground construction,

particularly surfacing. In the U.S., of the approximated 200 000 recorded playground injuries, about 70% are due to a fall (12). Bae et al. (2017) conducted a retrospective analysis of injury mechanisms in 6,110 subjects and found that the leading cause of injury was falls (48.5%). These results suggest the importance of the surfacing material upon which children land post-fall and its role in determining injury severity. This literature review was conducted to assess the relationship between different playground surfaces and pediatric injury.

Methods Twelve studies were identified using database searches of PubMed, NBCI and Google Scholar. Studies were assessed and data was extracted and synthesized.

Results Twelve total studies analyzing playground surfaces injury rates were identified. Five of the 12 studies reviewed suggest that sand-based playground surfaces pose the least risk of injury (2, 4, 6, 8, 17). In the randomized trial conducted by Howard et al. (2009), 37 elementary schools in Toronto, Canada received funding to install either sand or engineered wood fiber surfacing in their school playgrounds. After observing the rate of upper-extremity fractures for 2.5 years across all schools, the authors found that the fracture rate per 100 000 studentmonths was about 5 times greater with engineered wood fiber than that of sand (8). In Ball’s (2002) contract research report, analysis of the consequences of a playground fall showed that the highest rate of fractures occurred on unknown playground surfaces (41.8%), followed by bark/chip surfaces (20.8%), whereas rates were lowest on sand (2.5%) (Table 1). A similar pattern is found for non-fracture consequences of playground falls where the incidence of non-fracture injuries was higher for bark/chip surfaces as compared to sand (Table 1) (2). Ball (2002) also reported that the rate of incurring a fracture on bark/chip was 4.7 times higher than that of concrete playgrounds. The remaining 7 studies had conflicting results where a number of studies suggested bark/chip or

Table 1. Number of surface-specific fracture injuries and non-fracture injuries incurred by playground falls during play with swings, climbing and slides combined. Adapted from (2).

Playground Surfacing in Relation to Pediatric Injury

concrete were implicated with high injury rates. Across the studies, there was no standardized method of data collection which may account for the conflicting results found and many confounding factors were identified.

Discussion Five of the 12 studies suggested that the use of sand-based playground surfacing may be better in attenuating the incidence of pediatric injuries. However, conflicting evidence from other studies makes it difficult to suggest a single playground surface that is associated with the highest injury rates. In a contract research report by Ball (2002) conducted at the Middlesex University in London, reports show higher injury counts (both fracture-related and non-fracture related) on bark/chip as compared to other alternatives such as concrete, rubber and sand (Table 1). A correlational study conducted by Mott et al. (1997) showed that of the injuries of the 330 subjects, injuries occurred more frequently on concrete surfaces as compared to bark or rubberized surfaces. Choi, Kaur, & Robinovitch (2014) studied the peak resultant force of different playground surfaces in relation to distal radius fractures as compared to those of rigid conditions. They found that sand and engineered wood fiber reduced the peak resultant force by 17% and 5% respectively, but concluded that the 17% reduction of sand was clinically insufficient and that the development of safer playground surfacing is needed. The mixed results of these studies are likely due to a number of factors. The randomized trial conducted by Howard et al. (2009) only compared sand and engineered wood fiber surfacing, whereas Ball’s (2002) report accounted for concrete, tarmac, sand, bark/chip, grass/earth, and rubber. Choi et al. (2014) excluded concrete in heir analysis, while Sosin et al. (1993) excluded bark/chip and engineered wood fiber. The discrepancies in the types of surfaces evaluated among these studies suggest the difficulty in generating a safety-based ranking of these surfaces. In addition, the methods of analysis vary across these studies. Howard et al. (2009) measured surface-specific fracture rates. Ball (2002) presented raw injury counts and Choi et al. (2014) measured peak resultant force reduction. The variance of methodology and analysis among the studies poses even greater challenges to establish a choice playground surface. Other confounding factors within the literature also provides obstacles in data interpretation. Inclement weather, humidity and extreme temperatures have been shown to influence the impact-absorption qualities of playground surfacing (4, 11). When compared to sand in wet and dry conditions, wood chips showed better impact attenuation (11). However, this study was not done in-situ, suggesting the limitations of playground attenuation performance in laboratory conditions. Other confounding factors include playground crowdedness, age of playground surfacing, child supervision, child temperament, and preferential use of different playgrounds based on appearance or location (10, 13). Many of these factors are interconnected, making data interpretation even more problematic. Unfortunately, the number of studies that try to account for factors such as playground crowdedness and age of surfacing material in their study design is limited.

Nevertheless, there is merit in the pursuit of these studies. Research in this field has brought attention to the public health standards of playground construction. Against the health hazards posed by playground play, the American Society for Testing and Materials (ASTM) International developed a standard regulatory test that measures the impact attenuation performance of playground surfacing materials. The parameters measured are maximum acceleration (G-max) and the risk of head trauma (Head Injury Criterion [HIC] score) where G-max scores under 200 and HIC scores under 1,000 are considered within safe limits (7, 10, 12, 15, 16). G-max is measured by dropping a head form device that simulates the force of a headfirst fall from different heights and HIC scores are derived from acceleration (10). The literature surrounding the efficacy of this attenuation performance test is conflicting (10, 12). Although Laforest et al. (2001) found evidence supporting an association between G-max and injuring occurrence, they could not validate that 200 was an appropriate G-max threshold for severe head injuries. A study conducted by Mack et al. (2000) concluded that the ASTM International’s impact attenuation performance test results can be unreliable and counterintuitive. The authors conducted 3 head-form device test drops on loosefill playground surfacing such as gravel, sand, wood chips and wood fibers (12). Their results were consistent for wood fibers and wood chips but inconsistent for sand and gravel (12). Counterintuitive to physics principles, they also found that the critical height, which is defined as the height below which severe, life-threatening injuries are unlikely, was greater for thinner surfacing as compared to thicker surfacing for sand (12). Other studies utilizing the ASTM International’s regulatory tests showed differential results as a function of weather where higher temperatures produced erratic HIC scores (4). The implications of these findings suggest cautionary interpretation of G-max and HIC scores as criteria for playground safety and the need for adjustments or alternatives to ASTM International’s impact attenuation tests.

Conclusion Although playgrounds are a prime source of deconstructed learning for children outside of the classroom, these outdoor environments present numerous health hazards. The statistics on playground-related injuries and hospitalization across the globe are alarming. Numerous efforts have gone toward evaluating the role of playground surfacing on injury rates. Some studies comparing surface-specific injury rates point towards the benefits of sand-based surfacing (2, 4, 6, 8, 17). Unfortunately, there are a series of confounding factors such as weather and age of surfacing materials that can affect the physical characteristics of playground surfaces. Non-surface qualities, including child supervision and crowdedness, also pose limits to these studies. Together these factors make it difficult to determine a superior playground surface material. The observed relationship between injury risk and playground play suggests the need for research circumventing the policies and standards of playground surfacing. In particular playground construction and the methods of evaluation impact attenuation of playground surfacing are in question. The lack

Playground Surfacing in Relation to Pediatric Injury

of consistency of the G-max values and HIC scores established by ASTM International across various playground surfaces call for further attention to this field of study. Unfortunately, addressing these concerns takes time and resources that are not always readily available.

8. Howard AW, Macarthur C, Rothman L, Willan A, Macpherson AK. (2009). School Playground Surfacing and Arm Fractures in Children: A Cluster Randomized Trial Comparing Sand to Wood Chip Surfaces. PLoS Medicine, 6(12).

The challenges in this area of research translate to the clinical field as orthopaedic surgeons and pediatricians continue to question the specifics of the relationship between playground play and children’s health. Tools such as injury surveillances and databases can be highly beneficial. They can be used to systematically organize injury data, which can enhance and simplify data analysis in this highly confounded niche topic. At the same time, creating a working list of surface-specific injuries can allow physicians to compare current specifics of playground-based injuries such as fracture severity, number of fractures, treatment plans, and patient outcomes over time. Collecting these clinical measurements, along with associated environmental parameters such as playground surfacing material and the age of playground surfacing, can provide clinically relevant insight that physicians can apply to their practice.

9. Joeris A, Lutz N, Wicki B, Slongo T, Audigé L. (2014). An epidemiological evaluation of pediatric long bone fractures — a retrospective cohort study of 2716 patients from two Swiss tertiary pediatric hospitals. BMC Pediatrics, 14.

Acknowledgments I would like to thank Mark Seeley, MD, for support throughout this literature review.

References 1.

Bae S, Lee JS, Kim KH, Park J, Shin DW, Kim H, … Jeon W. (2017). Playground Equipment Related Injuries in Preschool-Aged Children: Emergency Department-based Injury In-depth Surveillance. Journal of Korean Medical Science, 32(3), 534–541.

2. Ball DJ, Great Britain, Health and Safety Executive, & Middlesex University. (2002). Playgrounds-risks, benefits and choices. Sudbury: HSE Books. Retrieved from http:// www.hse.gov.uk/research/crr_pdf/2002/crr02426.pdf

11. Lewis LM, Naunheim R, Standeven J, Naunheim KS. (1993). Quantitation of impact attenuation of different playground surfaces under various environmental conditions using a tri-axial accelerometer. The Journal of Trauma, 35(6), 932–935. 12. Mack M, Sacks J, Thompson D. (2000). Testing the impact attenuation of loose-fill playground surfaces. Injury Prevention, 6(2), 141–144. 13. Mott A, Evans R, Rolfe K, Potter D, Kemp KW, Sibert JR. (1994). Patterns of injuries to children on public playgrounds. Archives of Disease in Childhood, 71(4), 328–330. 14. Mott A, Rolfe K, James R, Evans R, Kemp A, Dunstan F, … Sibert J. (1997). Safety of surfaces and equipment for children in playgrounds. Lancet (London, England), 349(9069), 1874–1876. 15. Sherker S, Ozanne-Smith J, Rechnitzer G, Grzebieta R. (2005). Out on a limb: risk factors for arm fracture in playground equipment falls. Injury Prevention, 11(2), 120–124.

3. Bond MT, Peck MG. (1993). The risk of childhood injury on Boston’s playground equipment and surfaces. American Journal of Public Health, 83(5), 731–733.

16. Sherker S, Short A, Ozanne-Smith J. (2005). The in situ performance of playground surfacing: implications for maintenance and injury prevention. International Journal of Injury Control and Safety Promotion, 12(1), 63–66.

4. Branson LJ, Latter J, Currie GR, Nettel-Aguirre A, Embree T, Hagel BE. (2012). The effect of surface and season on playground injury rates. Paediatrics & Child Health, 17(9), 485–489.

17. Sosin DM, Keller P, Sacks JJ, Kresnow M, van Dyck PC. (1993). Surface-specific fall injury rates on Utah school playgrounds. American Journal of Public Health, 83(5), 733–735.

5. Chalmers DJ, Marshall SW, Langley JD, Evans MJ, Brunton CR, Kelly AM, Pickering AF. (1996). Height and surfacing as risk factors for injury in falls from playground equipment: a case-control study. Injury Prevention, 2(2), 98–104. 6. Choi WJ, Kaur H, Robinovitch SN. (2014). Measurement of the effect of playground surface materials on hand impact forces during upper limb fall arrests. Journal of Applied Biomechanics, 30(2), 276–281. 7.

10. Laforest S, Robitaille Y, Lesage D, Dorval D. (2001). Surface characteristics, equipment height, and the occurrence and severity of playground injuries. Injury Prevention, 7(1), 35–40.

Eager D, Hayati H. (2019). Additional Injury Prevention Criteria for Impact Attenuation Surfacing Within Children’s Playgrounds. ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems, Part B: Mechanical Engineering, 5(1), 011002.

Scholarly Research In Progress • Vol. 3, November 2019

Evaluation of the Benefits of Skin Cancer Community Health Day John Orr1†

Geisinger Commonwealth School of Medicine, Scranton, PA 18509 †Doctor of Medicine Program Correspondence: orrjohn07@gmail.com 1

Abstract More than 1 million cases of skin cancer are reported in the United States each year. These skin cancers can be found by trained professionals at dermatology office visits, primary care office visits or free skin cancer screening clinics. In June of 2017, Guthrie Health System had a Skin Cancer Community Health Day. Seventy-five individuals were screened for possible skin cancer. Thirty-five of these patients were referred for a follow-up appointment due to concerning skin lesions. This retrospective chart-review study was designed to determine the value of this Skin Cancer Community Health Day. The value of the cancer screening day was determined by looking at the time to follow-up appointment and the outcome of those follow-up appointments. These data points were found by reviewing the patients' charts on Epic between Jan. 28, 2018, and March 28, 2018. The overall average wait time for an appointment for those patients who were referred for follow-up was 48 days. When the 4 patients who were referred to dermatology offices are removed from the calculation, the average wait time falls to only 22 days. Of the 26 patients who have attended follow-up appointments, 57.69% were found to have a precancerous or cancerous lesion and 42.31% were found to have benign lesions. This data shows that when wait times are compared to the national average wait times for a dermatologic visit, the Guthrie Skin Cancer Community Health Day screening event resulted in patients experiencing faster times to referral appointments. Additionally, of the 26 patients who have attended their followup appointments, 57.69% were found to have a lesion that was potentially cancerous. These patients benefited from quicker times to follow-up appointments and their skin lesions were removed. The Guthrie Skin Cancer Community Health Day benefits the Sayre community and would be of value to the community annually.

States. To compare, other common cancers, like breast cancer and colorectal cancer, have estimated incidence numbers reported for 2017 by the American Cancer Society of 252 710 and 135 430, respectively. Skin cancer makes up a large portion of new cancer cases in the U.S., and there is presently no consensus on how or when screening should be done for this common group of cancers (1–2). Evaluation of skin cancer screening clinics in other literature shows that survival from melanoma is related directly to the thickness and stage of the tumor on presentation.

Figure 1. Basal cell carcinoma

Introduction Skin cancer screening initiatives have taken place in the field of dermatology for a number of years now, but these initiatives continue to meet with opposition from regulatory and advisory committees. The actual benefits of this type of screening tool have been brought into question. That said, skin cancer still affects the lives of more individuals than any other type of cancer in the world. More than 1 million people are newly diagnosed with a skin cancer each year in the United States alone. This includes 800 000 new cases of basal cell carcinoma (BCC) (Figure 1), 250 000 new cases of squamous cell carcinoma (SCC) (Figure 2), and 55,000 new cases of melanoma (Figure 3). Together referred to as non-melanoma skin cancers (NMSC), BCC and SCC alone make up more than 1 million new cases of skin cancer each year in the United

Figure 2. Squamous cell carcinoma.

Evaluation of the Benefits of Skin Cancer Community Health Day

Additionally, incidence of skin cancers increases with skin cancer screening clinics, but incidence of thick melanomas and level of mortality decreases. Therefore, skin cancer screening should have the ability to increase the survival rates from melanoma. Furthermore, 31% of patients referred after a skin cancer screening were found to have an additional basal cell carcinoma at follow-up (3-5). Finally, the average wait time for a dermatology appointment in the United States is reported to be about 36 days (6). According to staff at the Guthrie Dermatology Clinic, wait times for standard new patient appointments were approaching 6 months in June of 2017. Therefore, on June 7, 2017, a Skin Cancer Community Health Day screening clinic took place at the Guthrie Internal Medicine Clinic in Sayre, based on the community need for skin cancer screening in Bradford and surrounding counties. Three internal medicine residents and one internal medicine physician administered the screening exams and made the referrals for follow-up care.

Figure 3. Melanoma.

The primary objective of the study is to conduct a retrospective chart review of patients who participated in a community health skin cancer screening day to observe patient outcomes. It is hypothesized that patients who participated in the screening visit had faster referrals for care.

Materials and Methods A retrospective chart review of the 75 patients (28 men, 47 women, mean age 59.99 years) who attended the Skin Cancer Community Health Day at Guthrie Clinic, Sayre was conducted after achieving IRB approval. Subjects were identified by obtaining the list of patients who participated in the quality improvement initiative on June 7, 2017. Records were reviewed over a period of two months from Jan. 28, 2018 to March 28, 2018. Charts were reviewed for the following characteristics: • Responses to 6 survey questions and notation of lesion location on the patient sheet from the Skin Cancer Screening Day (Figure 4) • Location of follow-up (dermatology, general surgery, plastic surgery, outside institution) • Time to follow-up appointment (number of days from June 7, 2017, to follow-up appointment date) • Outcome of follow up appointment/biopsy of lesion • Outcome if not referred for follow-up but found to have a skin lesion requiring biopsy since Skin Cancer Screening Day • Age

Figure 4. Guthrie skin cancer screening form.

• Gender Statistical analysis of the data was then completed using Microsoft Excel.

Results Patient demographics In total, 75 patients (28 men and 47 women) were screened for skin cancer. A total of 35 patients (14 men and 21 women) were referred for follow-up appointments. Fifty percent of the men and 44.7% of the women were referred for follow-up 18

appointments. The average age of screened patients was 59.9 years, referred patients was 59.4 years, non-referred patients was 60.5 years, and pre-cancer or cancer-positive patients was 63.1 years (Figure 5). The number of patients found to have precancerous or cancerous lesions based on follow-up appointment and biopsy per age range were: age less than or equal to 29 years with one, 30–39 years with zero, 40–49 years with two, 50–59 years with three, 60–69 years with two, 70–79 years with five, and greater than or equal to 80 years with two (Figure 6).

Evaluation of the Benefits of Skin Cancer Community Health Day

Discussion Reports of six-month wait times for dermatology appointments in the Guthrie Health System were putting a stress on the nearby communities. Insight into potential ways to remedy this problem needed to be investigated and evaluated. Therefore, when Guthrie Dermatology was excluded, if was found that patients were seen for evaluation and treatment quicker than national average. Patients were seen sooner than they otherwise would have been seen in this community, thus decreasing the burden of their disease. Many of these patients (85%) were evaluated by general surgery or plastic surgery at Guthrie due to schedule availability in these departments.

Figure 5. Average ages of screened patients

Figure 6. Number of patients with pre-cancer or cancer based on age

Average number of days to follow-up appointment The overall average number of days to follow-up examination was 48 days (n = 34). The average number of days to follow-up in those referred to the Guthrie Dermatology Clinic was 223 days (n = 4). The average for those referred to Guthrie General Surgery (n = 18) and Guthrie Plastic Surgery (n = 11) were 26 and 17 days, respectively. Of those referred to an outside (nonGuthrie) institution, the average was 13 days (n = 1). The overall average number of days to follow-up examination excluding those referred to Guthrie Dermatology was 22 days (n = 30) (Table 1).

Table 1. Number of days to follow-up appointment

Outcome of referred patients Of the 35 patients who were referred for follow-up care and had already attended the appointment at the time of the study, 30.77% were found to have precancerous lesions, 26.92% were found to have cancerous lesions, and 42.31 % were found to have benign lesions (Figure 7). Nine patients had yet to attend their follow-up appointment or had an appointment outside of the Guthrie system, so records were not available. There was also a total of 4 patients who were screened during the Guthrie Skin Cancer Community Health Day who were not referred for follow-up care, but who have since had lesions biopsied at dermatology appointments during the study period. All 4 of these patients were found to have only benign lesions.

Figure 7. Outcome of patients referred for follow-up care 19

Evaluation of the Benefits of Skin Cancer Community Health Day

Nevertheless, it still took the 4 patients who were referred to Guthrie Dermatology 223 days to obtain their follow-up exam. This further magnifies the scheduling difficulties that exist in the Sayre community when attempting to obtain a dermatology appointment. Even those who were referred from a screening clinic with potentially harmful lesions found it difficult to schedule their appointment with dermatology in a timely fashion. This does not reflect the quality or recognition of the Guthrie Dermatology Clinic, but it does show that high demand and lack of providers is putting stress on the dermatologic care in the area. The benefits of the Skin Cancer Community Health Day shown in this study illustrate how skin cancer screening clinics like these can potentially act as “triage” clinics for dermatology at Guthrie. If patients are able to attend one of these screening days and then get referred to general surgery or plastic surgery for a biopsy or follow-up care, some of the large patient burden placed on the Guthrie Dermatology Clinic can be lessened. This may help decrease some of the large patient volume “stress” placed on the Dermatology Department.

Conclusion

Furthermore, the information obtained regarding the demographics of those screened for skin cancer showed that the average age of a referred patient was 59.4 years, but the average age of a patient who was found to have a precancerous or cancerous lesion was 63.1 years. Also, the age range that yielded the most precancerous or cancerous skin lesions was the 70–79-year age range. This data would support that cancerous skin lesions are more likely to be found on older individuals during a skin cancer screening, but more research into the utility of these findings needs to be done to allow recommendations on the timing and course of skin cancer screening to be developed.

Additionally, 57.69% of the referred patients were found to have precancerous or cancerous lesions at their follow-up appointments. This shows that screening done by Internal Medicine residents and physicians benefited these individuals. Skin-cancer screening done by non-dermatologists can still benefit the community and not place an added stressor on the already busy dermatology staff. The limitations of this study included not obtaining access to the actual wait times for standard dermatology visits at Guthrie Dermatology during this time period, only having four patients referred to dermatology for follow up, the small overall study population, and only subclassifying biopsied lesions as cancerous or benign. Further evaluation of the current wait times for appointments at Guthrie Dermatology and further analysis of updated patient records for more specific classification of cancerous lesions into melanoma or non-melanoma skin cancer would be areas of future study. Additionally, comparison of skin cancer screening clinics conducted by non-dermatology-trained professionals and those conducted by dermatology-trained professionals and comparison of rural results to urban and suburban clinic results in similar study designs could help extrapolate this data to other populations.

The Guthrie community would benefit from a free skin cancer screening day at least annually. Patients were seen for followup sooner than they otherwise would have been seen in this area. If more of these clinics can be arranged, the community may benefit even more. Individuals were turned away on the day of the Skin Cancer Community Health Day due to time constraints and the fact that only 4 providers were present to conduct the screenings. The day had to be capped at 75 patients. If more providers and more screening clinics are able to be conducted the benefit shown in this study should be repeated. Community skin health would be improved and burden on the health system can be decreased.

Disclosures The study was supported by Guthrie, Robert Packer Hospital, Sayre.

References Marks Jr. JG, Miller JJ. Lookingbill & Marks' Principles of Dermatology. 5th Edition. London: Elsevier; 2013.

2. American Cancer Society. Cancer Facts & Figures 2017. American Cancer Society. 2017. 3. Coory M, Smithers M, Aitken J, Baade P, Ring, I. Urbanrural differences in survival from cutaneous melanoma in Queensland. Australian and New Zealand Journal of Public Health. 2006; 30: 71–74. 4. Brunssen A, Waldmann A, Eisemann N, Katalinic A. Impact of skin cancer screening and secondary prevention campaigns on skin cancer incidence and mortality: A systematic review. Journal of the American Academy of Dermatology. 2017; 76: 129–139. 5. Wakiyama TP, França M, Carvalho LP, Marques M, Miot HA, Schmitt JV. Initial basal cell carcinomas diagnosed in the National Campaign for Skin Cancer Prevention are smaller than those identified by the conventional medical referral system. An Bras Dermatol. 2017; 92(1): 26–9. 6. Kimball AB, Resneck Jr. JS. The US dermatology workforce: A specialty remains in shortage. Journal of the American Academy of Dermatology. 2008; 59(5): 741–745.

Scholarly Research In Progress • Vol. 3, November 2019

Physician Use of Patient-Centered Communication: Impact on the Patient Experience Mikael Horissian1†, Anne Horissian1, and Elizabeth Kuchinski1

Geisinger Commonwealth School of Medicine, Scranton, PA 18509 †Doctor of Medicine Program Correspondence: mhorissian@som.geisinger.edu 1

Abstract Background Patient-centered care has become a primary focus in health care. Physicians’ communication behaviors are a readily modifiable component of patient-centered care. For this reason, understanding the importance of certain behaviors can elucidate ways to improve patient care. The study aims to develop a robust measure of patient-centered communication and to assess the relationship between physicians’ patientcentered communication and several measures of the patient experience. Methods Ninety-two patients (68% female) completed an in-office survey at a primary care office that took 5 to 15 minutes to complete. The survey involved an adapted form of the Patient-Centered Communication Behaviors scale (Cronbach’s alpha=0.98) as the independent variable. Dependent variables included an 8-item trust scale from the Primary Care Assessment Survey (Cronbach’s alpha=0.87), and variables regarding the patient experience including satisfaction, comfort, confidence, honesty, and listening to physician.

to conversation with every staff member. This study focuses on the communication between patient and provider, which has been found to be most consistently rated as important for quality care (2) and one of the most common unmet needs during physician visits (3). Additionally, physicians rate themselves higher for patient-centered communication quality than the patients report (4); therefore, physicians may be unaware of the expectations of quality conversations during visits and may not completely understand the behaviors that define patient-centeredness. Patient-centered communication, satisfaction, and trust Taking the time to explain medications to patients, asking questions and involving patients more in decision-making has been linked to greater satisfaction and higher quality care (5, 6). Higher trust in physician is associated with increased perception of physician’s competence, adherence to treatment, and satisfaction (7). Additionally, nonverbal communications play varying roles in how much patients trust their physician (8). Gender

Higher patient-centered communication was associated with higher trust, satisfaction, confidence and comfort in physician, higher reports of listening to physician, controlling own medical decisions and being honest with physician, and lower reports of keeping information from physician.

Studies show mixed results, with some suggesting that female physicians provide more patient-centered care (9), while others have found that patients rate male physicians higher in patient-centeredness (10). Male physicians were perceived as more empathetic with patient-centered gaze while female physicians were perceived as more empathetic when they had averted gaze (11); therefore, the interaction between gender and perception of patient-centeredness may be multifaceted.

Conclusions

Duration of visit and length of relationship

Higher patient-centered communication was associated with several positive aspects of the patient experience in the ambulatory setting. Clinicians should be aware of the impact that communication behaviors may have on the patient’s experience. Increased education and training in patient-centered communication should be considered by physicians and institutions. Further research is recommended to validate the newly developed measure of patient-centered communication.

Duration of visit and duration of relationship between patient and provider may influence perception of patientcenteredness. Satisfaction with time spent with a surgeon was related more to the surgeon’s level of empathy than to actual visit duration (12). Patients who had a particular medical professional that they saw regularly reported more patientcentered experiences and higher quality in care (13).

Results

Introduction Patient-centered medicine is not a new concept; however, it challenges the traditional medicine model in that it focuses on the principle of autonomy, the outcomes that are important for patients, the patient’s values and beliefs and shared decision-making (1). The definition of patient-centeredness can sometimes be overwhelming, considering patient-centered care encompasses every aspect of medicine from room décor

Positive health outcomes Positive interactions, patient-centered care, and greater trust have been associated with adherence to treatment (7, 14, 15), retention of care (16), and decreased mortality rates (17). Physician friendliness, first-name introduction, and use of open-ended questions were related to patients fully disclosing information (9). These findings illustrate the importance of patient-centered interactions on the health and well-being of patients.

Physician Use of Patient-Centered Communication: Impact on the Patient Experience

Study aims The current literature is helpful to our understanding of the impact of patient-centered care on the patient experience. Although several measures of patient-centered care have been developed, a tool to measure patient-centered communication in completeness is lacking. Our study had three main aims: 1) Develop a robust questionnaire regarding physician patientcentered communication and assess the measure’s reliability; 2) Explore relationships among physician patient-centered communication, patient comfort, confidence, compliance, trust, and satisfaction; and 3) Investigate the influence of age and gender on patient-centered communication and examine the effect of duration of physician-patient relationship on patientcentered communication, trust, and satisfaction.

Materials and Methods Participants One hundred and eleven participants took part in this study. Nineteen were excluded from data analysis, as the surveys were not completed; therefore, the final number of participants was 92 (68% female). Participants’ ages ranged from 18 to 82 years (mean age=37.8; SD=15.4). Twenty-three percent of participants were of Hispanic origin, and the race composition was as follows: 65% White, 20% Black or African-American, 3% Asian, 3% American Indian, and 9% other. Education level was as follows: 6% less than high school, 60% high school, 19% two-year college, 4% four-year college, and 11% graduate/ professional degree. Procedures The participants were patients in the waiting room of a primary-care facility with multiple physicians in a mid-sized city in northeast Pennsylvania. Participants were recruited by researchers who approached them in the waiting room offering a 5- to 15-minute survey. Researchers approached all patients who presented to the clinic during times of survey collection. Participants had to be at least 18 years of age and able to read and understand English. Participants underwent an informed consent process prior to taking the survey, and were offered entry into a raffle for a $150 gift card for completing the survey. The study obtained approval from The Wright Center for Graduate Medical Education Institutional Review Board. Participation remained completely confidential and participants were identified only by an identification number assigned to them. Surveys were stored in a locked file cabinet in a secure office space. Survey data was stored and maintained in an electronic data file via password-protected computers and were solely accessed by members of the study team. Measurement of physician’s patient-centered communication A modified measure of the Physician-Patient Communication Behaviors Scale from the Matched Pair Inventory developed by Campbell et al. was used to measure patient-centered communication (18). This 19-item measure assesses how patients view their physician’s communication habits (e.g., my physician understood what I had to say) and includes a 5-point Likert scale (strongly disagree, disagree, neither agree nor disagree, agree, strongly agree). Thirteen of the items were

validated by Wachira et al. (19). We chose to use all 19 original items, except for the item regarding satisfaction as we will be testing satisfaction as an outcome variable. We added 29 more items encompassing other communication behaviors that were not included in the original scale (e.g., my physician maintains eye contact with me) to make the scale as thorough as possible. These items were either developed by the study investigators or were obtained from scales developed in previous studies (20–27). A final score of patient-centered communication was the average of all items in the measure. Higher scores represented higher levels of patient-centered communication. Patient experience variables Eight items with a 5-point Likert scale (strongly disagree, disagree, neither agree nor disagree, agree, strongly agree) were used to determine how much a patient agreed with each item. Items included satisfaction, comfort and confidence in physician, listening to physician advice, controlling own medical decisions, honesty with physician, feeling free to contradict physician, and keeping information from physician. Patient Trust Scale The Patient Trust Scale is a subscale of the Primary Care Assessment Survey developed by Safran et al. with good reliability (Cronbach’s alpha of 0.86) (28). It is an 8-item questionnaire with a 5-point Likert scale (strongly disagree, disagree, neither agree nor disagree, agree, strongly agree) that measures how much patients trust their physicians. A final score of trust was obtained by taking the average of all items in the scale. Higher scores indicated higher trust. Frequency of patient visit and physician information Participants were asked how often they visit any medical physician, if they have a primary care/family physician, how often they visit their primary care/family physician, how long he or she has been their family physician, the sex of their physician, and the relative age of their physician. Demographic information Participants were asked to report their gender, age in years, education level, ethnicity, and race. Statistical analysis SPSS version 25 was used to perform all analyses. To explore the validity of items in our newly developed measure of patientcentered communication, exploratory factor analysis was conducted using principal axis factoring; oblique rotation (direct oblimin) was utilized as it was expected that factors may be correlated. Cronbach’s alpha ≥0.7 was considered acceptable for measurement of internal consistency; for an item to be included in the final measure, a cut-off level of 0.4 was used for item-total correlation. Linear regression was conducted to determine relationships between variables. A p-value<0.05 was considered statistically significant. The measure of patient-centered communication was used as the independent variable. Dependent variables included trust, satisfaction, comfort, confidence, listening to physician advice, controlling own medical decisions, honesty, feeling free to contradict physician, and keeping information

Physician Use of Patient-Centered Communication: Impact on the Patient Experience

from physician. To examine the relationship among patientcenteredness, satisfaction, and trust, hierarchical multiple regression analyses were conducted with variables entered in the following blocks: (a) Patient-centered communication; and (b) Trust. Criterion variable was Satisfaction. T-test analysis and ANOVA were used to determine if there were differences in patient-centered communication and patient experience variables between age groups, gender, and education level.

Results Measures The measure of patient-centered communication had a Cronbach’s alpha of 0.97. One item, rating of physician as “empathetic,” had a low internal consistency and was removed. The final scale contained 47 items with a Cronbach’s alpha of 0.98. Two factors were extracted which accounted for 66% of the variance. All “positive” variables in the measure (e.g., my physician uses language I can understand) loaded onto Factor 1. All “negative” variables in the measure (e.g., my physician speaks too fast) loaded onto Factor 2. The previously validated 8-item trust scale had a Cronbach’s alpha of 0.87. Relationships Patient-centered communication, trust, and satisfaction Higher patient-centered communication scores were related to higher trust (r = 0.85, p<0.001, 95% CI [0.74, 0.96]). Patientcentered communication scores were also positively correlated with satisfaction (r = 0.75, p<0.001, 95% CI [0.61, 0.89]). Trust was positively associated with satisfaction (r = 0.60, p<0.001, 95% CI [0.43, 0.76]). The mediation model is presented with patientcentered communication as the mediator between trust and satisfaction (Figure 1). Hierarchical multiple regression results are presented in Table 1. In the mediation model, trust was no longer significantly associated with satisfaction.

Age, sex, and duration Patient age was not significantly related to patient-centered communication score (r = 0.02, p = 0.87, 95% CI [-0.19, 0.23]). Patient-centered communication scores were not significantly different between physician age groups – younger than patient, same age as patient, or older than patient (F(2, 89) = 1.00, p = 0.37). There was no significant difference in level of patient-centered communication between patient gender (t = 0.53, p = 0.60, 95% CI [-0.21, 0.36]), or physician sex (t = 0.30, p = 0.73, 95% CI [-0.23, 0.31]). Neither duration of relationship between patient and physician nor frequency of visits with physician were associated with differences in patient-centered communication scores (r = 0.12, p = 0.28, 95% CI [-0.09, 0.33]; r = 0.13, p = 0.21, 95% CI [-0.34, 0.08] respectively). Duration of relationship was not associated with level of trust (r = 0.05, p = 0.62, 95% CI [-0.16, 0.26]) or level of satisfaction (r = 0.11, p = 0.31, 95% CI [-0.10, 0.32]).

Discussion Patient reported measures are a great way to determine physician patient-centeredness, because only patients can determine whether they received the information they wanted, if they understood the physician and if the care falls in line with their personal values (29). The newly developed measure of patient-centered communication had high internal consistency. Additionally, two subscales can be used in the measure: one regarding positive behaviors and the second regarding

Figure 1. Graphical interpretation of the mediation model between trust and satisfaction with patient-centered communication as the mediator N=92 (2016–2017)

Patient-centered communication and patient experience variables Patient-centered communication scores were positively associated with confidence in physician (r = 0.81, p<0.001, 95% CI [ 0.69, 0.93]), comfort (r = 0.85, p<0.001, 95% CI [0.74, 0.96]), controlling own medical decisions (r = 0.45, p<0.001, 95% CI [0.26, 0.63]), listening to my physician (r = 0.62, p<0.001, 95% CI [0.46, 0.79]), being honest with physician (r = 0.58, p<0.001, 95% CI [0.41, 0.75]), feeling free to contradict my physician (r = 0.28, p = 0.006, 95% CI [0.08, 0.48]). There was an inverse relationship between patient-centered communication score and keeping information from physicians (r = -0.31, p = 0.002, 95% CI [-0.51, -0.12]).

Table 1. Hierarchical multiple regression analyses predicting patient satisfaction (2016– 2017)

Physician Use of Patient-Centered Communication: Impact on the Patient Experience

negative behaviors. The only item that had a low internal consistency was rating of physician as “empathetic,” which we believe was most likely due to participants being unaware of the meaning of the word. Further research must be done using the new measure of patient-centered communication to test for internal and external validity. The new measure is longer than prior scales. We believe, however, that it provides a more comprehensive measurement of patient-centered communication. Similar to previous research (5, 6, 8), higher patient-centered communication was associated with higher trust in physician, as well as higher satisfaction with physician. However, when the researchers dissected this relationship further, it was found that patient-centered communication fully mediates the relationship between trust and satisfaction. This suggests that the relationship between trust and satisfaction may be explained by patient-centered communication. Higher ratings of patient-centered communication were associated with patients reporting greater confidence in their physician, greater comfort, higher adherence to physician’s advice, more honesty with their physician, and a greater feeling of controlling their own medical decisions. Considering that adherence to physician’s advice and being honest with your physician has been linked to better health outcomes (30–32), these relationships may suggest that using more patientcentered communication may positively impact the patient’s health. Higher patient-centered communication scores were also related to lower reports of withholding information from their physician. This is very important, as physicians may not be able to adequately help the patient if they are not cognizant of the entire picture. Patient-centered communication did not differ among physicians’ or patients’ gender or age. Contrary to our hypothesis, the duration of the relationship between patient and physician was not related to level of patient-centered communication, trust or satisfaction. This may suggest that while building a life-long relationship with patients is important, patient-centered communication, not longevity, may be the answer to earning a patient’s trust and providing satisfactory service. One limitation is that the survey was only distributed in one office building. Future research should expand the participant pool to examine external validity of the results. Another limitation was language. Several patients in the waiting room could not speak or read in English and therefore could not participate in the study. Having surveys of the most common languages in our area could have prevented this stratification. Further studies should focus on testing internal and external validity of the newly developed measure and examining cultural influences on the effect that patient-centered communication has on patients. One study showed that physicians were rated empathetic in less than 50% of interactions and used open-ended questions in only 30% to 40% of interactions (33); therefore, the need for increased awareness and practice of patient-centered communication is warranted. Short duration communication training has been shown to increase patient-centered communication (34).

Therefore, health care organizations and individuals should place an emphasis on improving physician communication skills in order to maximize the positive impact physicians can make on the patient’s experience.

Conclusion Higher patient-centered communication was associated with greater ratings of satisfaction, trust, confidence in physician, comfort, adherence to physician advice, honesty, and lower reports of withholding information. Patient-centered communication may make a distinct difference in the patient’s experience and their perceived quality of care.

Acknowledgments We would like to acknowledge Mushfiq R. Tarafder, PhD, for his assistance with proofreading the manuscript.

Disclosures The authors have no conflicts of interest. The study obtained approval from The Wright Center for Graduate Medical Education Institutional Review Board.

References 1.

Sacristan JA. Patient-centered medicine and patientoriented research: improving health outcomes for individual patients. BMC Med Inform Decis Mak. 2013;13.

2. Mohammed K, Nolan MB, Rajjo T, et al. Creating a patientcentered health care delivery system: a systematic review of health care quality from the patient perspective. Am J Med Qual. 2016;31:12–21. 3. Smith F, Orrell M. Does the patient-centred approach help identify the needs of older people attending primary care? Age Ageing 2007;36:628–31. 4. Maniaci MJ, Heckman MG, Dawson NL. Physician versus patient perception of physician hospital discharge communication: a preliminary study. Open Journal of Internal Medicine. 2014;4:101–7. 5. Studnek JR, Fernandez AR, Vandeventer S, Davis S, Garvey L. The association between patients' perception of their overall quality of care and their perception of pain management in the prehospital setting. Prehosp Emerg Care. 2013;17:386–91. 6. October TW, Hinds PS, Wang J, Dizon ZB, Cheng YI, Roter DL. Parent satisfaction with communication is associated with physician's patient-centered communication patterns during family conferences. Pediatr Crit Care Med. 2016;17:490–7. 7.

Cousin G, Mast MS. Agreeable patient meets affiliative physician: how physical behavior affects patient outcomes depends on patient personality. Patient Educ Couns. 2013;90:399–404.

8. Hillen MA, de Haes CJM, van Tienhoven G, et al. All eyes on the patient: the influence of oncologists' nonverbal communication on breast cancer patients' trust. Breast Cancer Res Treat. 2015;153:161–71.

Physician Use of Patient-Centered Communication: Impact on the Patient Experience

9. Lewis CC, Matheson DH, Brimacombe CAE. Factors influencing patient disclosure to physicians in birth control clinics: an application of the communication privacy management theory. Health Commun 2011;26:502–11. 10. Duberstein P, Meldrum S, Fiscella K, Shields CG, Epstein RM. Influences on patients' ratings of physicians: physicians demographics and personality. Patient Educ Couns. 2007;65:270–4. 11. Brugel S, Postma-Nilsenova M, Tates K. The link between perception of clinical empathy and nonverbal behavior: the effect of a doctor's gaze and body orientation. Patient Educ Couns. 2015;98:1260–5. 12. Parrish RC, Menendez ME, Mudgal CS, Jupiter JB, Chen NC, Ring D. Patient satisfaction and its relation to perceived visit duration with a hand surgeon. J Hand Surg Am. 2016;41:257–62. 13. Finney Rutten LJ, Agunwamba AA, Beckjord E, Hesse BW, Moser RP, Arora NK. The relation between having a usual source of care and ratings of care quality: does patientcentered communication play a role? J Health Commun. 2015;20:759–65. 14. Johnson MO, Chesney MA, Goldstein RB, et al. Positive provider interactions, adherence self-efficacy, and adherence to antiretroviral medication among HIV infected adults: a mediation model. AIDS Patient Care STDS. 2006;20:258–68. 15. Fuertes JN, Anand P, Haggerty G, Kestenbaum M, Rosenblum GC. The physician-patient working alliance and patient psychological attachment, adherence, outcome expectations, and satisfaction in a sample of rheumatology patients. Behav Med. 2015;41:60–8. 16. Graham JL, Shahani L, Grimes RM, Hartman C, Giordano TP. The influence of trust in physicians and trust in the health care system on linkage, retention, and adherence to HIV care. AIDS Patient Care STDS. 2015;29:661–7. 17. Meterko M, Wright S, Lin H, Lowy E, Cleary PD. Mortality among patients with acute myocardial infarction: the influences of patient-centered care and evidence-based medicine. Health Serv Res. 2010;45:1188–1204. 18. Campbell C, Lockyer J, Laidlaw T, MacLeod H. Assessment of a matched-pair instrument to examine doctor-patient communication skills in practising doctors. Med Educ. 2007;41:123–9. 19. Wachira J, Middlestadt S, Reece M, Peng CJ, Braitstein P. Psychometric assessment of a physician-patient communication behaviors scale: the perspective of adult HIV patients in Kenya. AIDS Res Treat. 2013:1–10.

22. Cegala DJ, Coleman MT, Turner JW. The development and partial assessment of the medical communication competence scale. Health Commun. 1998;10:261–88. 23. Safran DG, Karp M, Coltin K, et al. Measuring patients' experiences with individual primary care physicians. J Gen Intern Med. 2006;21:13–21. 24. Stewart AL, Napoles-Springer AM, Gregorich SE, SantoyoOlsson J. Interpersonal processes of care survey: patientreported measures for diverse groups. Health Serv Res. 2007;42:1235–56. 25. Mercer SW, McConnachie A, Maxwell M, Heaney D, Watt GCM. Relevance and practical use of the consultation and relational empathy (CARE) measure in general practice. Fam Pract. 2005;22:328–34. 26. Haddad S, Potvin L, Roberge D, Pineault R, Remondin M. Patient perception of quality following a visit to a doctor in a primary care unit. Fam Pract. 2000;17:21–9. 27. Wong ELY, Coulter A, Hewitson P, et al. Patient experience and satisfaction with inpatient service: development of short form survey instrument measuring the core aspect of inpatient experience. PLoS ONE. 2015;10. 28. Safran DG, Kosinski M, Tarlov AR, et al. The primary care assessment survey: tests of data quality and measurement performance. Med Care. 1998;36:728–39. 29. Tzelepis F, Sanson-Fisher RW, Zucca AC, Fradgley EA. Measuring the quality of patient-centered care: why patient reported measures are critical to reliable assessment. Patient Prefer Adherence. 2015;9:831–5. 30. Weir MR, Maibach EW, Bakris GL, et al. Implications of a health lifestyle and medication analysis for improving hypertension control. Arch Intern Med. 2000;160:481–90. 31. Elliot WJ, Maddy R, Toto R, Bakris G. Hypertension in patients with diabetes. Overcoming barriers to effective control. Postgrad Med. 2000;107:29–32, 35–36, 38. 32. Raviglione MC, Gupta R, Dye CM, Espinal MA. The burden of drug-resistant tuberculosis and mechanisms for its control. Ann NY Acad Sci. 2001;953:88–97. 33. Pollak KI, Alexander SC, Tulsky JA, et al. Physician empathy and listening: associations with patient satisfaction and autonomy. J Am Board Fam Pract. 2011;24:665–72. 34. Maatouk-Burmann B, Ringel N, Spang J, et al. Improving patient-centered communication: results of a randomized controlled trial. Patient Educ Couns. 2016;99:117–24.

20. Hudon C, Fortin M, Haggerty JL, Lambert M, Poitras M. Measuring patients' perceptions of patient centered-care: a systematic review of tools for family medicine. Ann Fam Med. 2011;9:155–64. 21. Levine R, Shore K, Lubalin J, Garfinkel S, Hurtado M, Carman K. Comparing physician and patient perceptions of quality in ambulatory care. Int J Qual Health Care. 2012;24:348–56.

Scholarly Research In Progress • Vol. 3, November 2019

Retroperitoneal Sarcoma in a 55-year-old Female Treated with Immunotherapy Marc Incitti1† and Walter DelGaudio2

Geisinger Commonwealth School of Medicine, Scranton, PA 18509 Urology Associates of NEPA, Plains, PA 18705 †Doctor of Medicine Program Correspondence: mincitti@som.geisinger.edu 1

Abstract A 55-year-old female presented to the emergency department (ED) with weakness, dyspnea on exertion, abdominal pain, and decreased appetite with an unintentional weight loss of 13 pounds in 3 weeks. She has a strong family history for breast cancer. PET/CT scan for initial strategy revealed a large, abnormal hypermetabolic mass which seemed to arise from the upper pole of the right kidney. A laparotomy with resection of the retroperitoneal mass including a right nephrectomy and right adrenalectomy was performed 2 weeks after the patient’s initial ED visit. Immunohistochemistry further identified the tumor. Postsurgical nivolumab was ordered for 12 doses. We present this case to examine the unique histological characteristics and subsequent treatment course of a retroperitoneal sarcoma.

Case Presentation A 55-year-old female presented to the emergency department complaining of acute dyspnea on exertion and subacute fatigue. She described a dull pain with a fullness in the right upper quadrant (RUQ) of her abdomen despite having a decreased appetite and an unintentional weight loss of 5.8 kg in 3 weeks. An outpatient ultrasound completed prior confirmed a large mass in her RUQ. She was admitted for diagnostic workup of her abdominal pain, treatment for her iron-deficient anemia and possible surgery. Her past medical history includes essential hypertension, diverticulosis, anxiety disorder, tobacco use for 20 years, cholecystectomy, appendectomy, right parotidectomy for unspecified mass, and a Roux-en-Y gastric bypass status post weight loss of 42.2 kg. Multiple surgeries for perianal fistulas, partial bowel obstructions, and adhesions followed in the years after her Roux-en-Y procedure. She has a strong family history for breast cancer; her mother, paternal aunt, and maternal aunt died from breast cancer. She received serum results including normal renal function, BUN and creatinine were 22 and 0.87 mg/dL respectively, and the patient’s GFR was 68 mL/min/1.73m². Having corrected her anemia with ferric carboxymaltose injection and her symptoms managed, the patient was scheduled for a positron emission tomography– computed tomography scan. For initial treatment strategy, the PET-CT scan utilized 11.59 mCi of the tracer FDG, fluorodeoxyglucose. In 60 minutes, the topographic emission and imaging transmission completed, yielding the results of a large, abnormal hypermetabolic mass which seemed to be arising from the right kidney’s upper pole including the renal capsule. Mass effect exerted on the right

hepatic lobe accounted for the patient’s symptoms of fullness and abdominal discomfort. The right adrenal gland was noted to be obscured. Diffuse increase in bone marrow uptake of FDG might have been due to an infiltrative process such as a neoplasm or due to another cause such as inflammation or the iron-deficient anemia. There was no abnormal uptake of FDG anywhere else in the patient’s body. The PET scan demonstrated normal activity of the left kidney and bladder with no uptake into the uterus nor adnexal regions. Urology, surgery, and medical oncology were consulted, the latter starting allopurinol in anticipation for postoperative treatment. Surgery was planned for the following week. A laparotomy with resection of a retroperitoneal mass, measuring 15 cm x 17 cm, a right nephrectomy and right adrenalectomy were performed. Pathology of the resected mass reported the following immunohistochemical characteristics: negative for AE1/3 and CAM5.2 cytokeratin markers, S100, SOX10, smooth muscle actin SMA, desmin, CD34, P504S, EMA, GATA3 and the renal cell carcinoma (RCC) marker. The neoplastic cells were described as having clear epithelioid cytoplasm with spindled nuclei, staining positive for vimentin, CD10, PAX8, CAIX, and CD117. The official diagnosis was made of a malignant neoplasm of the right kidney except the renal pelvis with metastasis to the right adrenal gland; histologic grade G3 (FNCLCC grade 3) designated due to lymph-vascular invasion and presence of pleomorphic sarcomatoid features. During postsurgical follow-up, the medical oncologist prescribed Opdivo (nivolumab) 3 mg/kg IV every 2 weeks for 12 doses. The patient has recovered from her surgery and continues to be compliant.

Discussion Retroperitoneal sarcomas are rare tumors, accounting for 15% of all sarcomas and 1% of all tumors. Generally, after surgery, patients should be followed with imaging every 3 to 6 months for the first 2 to 3 years, then every 6 months for the next 2 years, then annually (1). Five-year survival for any postoperative retroperitoneal sarcoma is 12% to 70%. Staging of retroperitoneal sarcomas is based on the American Joint Committee on Cancer staging system. The patient’s retroperitoneal sarcoma was a grade 3; at 5 years following surgery, high-tumor-grade (grade 3) is a significant predictor of sarcoma-specific survival (2). Immunohistochemistry allowed for the identification of the patient’s tumor as including sarcomatoid features as well as features of a clear renal cell carcinoma (4). Nivolumab

Retroperitoneal Sarcoma

is FDA approved for the treatment of renal cell carcinoma. Nivolumab is an IgG4 anti-PD-1 human monoclonal antibody, and it blocks a signal that would prevent T cells from attempting to clear cancer cells. Treatment with nivolumab comes with risks to develop adverse effects including thyroid dysfunction, pneumonitis, colitis, arthritis and conjunctivitis. In addition to chemotherapy—or in this case, immunotherapy— postoperative radiotherapy is indicated in some cases. This presents particular challenges at this site however, since large abdominal areas generally require irradiation and the occurrence of side effects in many organs limits the doses (3). Although pathology antibody testing led to this specialized diagnosis, one could argue the mainstay choice of treatment, total surgical resection, would not have been altered (5). Indeed, Brown et al. purports all retroperitoneal sarcomas will be large, all will be deep, and all their histologies are treated identically despite the clear differences (6). However, it is the contention of every study that this rare disease deserves collaboration across institutions in order to advance care and refine treatment modalities.

References 1.

Porpiglia AS, Reddy SS, Farma JM. Retroperitoneal Sarcomas. Surg Clin N Am: 2016 Dec;96(1):993–1001

2. Abbott A, Habermann EB, Parsons HM, Tuttle T, AlRefaie W. Prognosis for Primary Retroperitoneal Sarcoma Survivors: A Conditional Survival Analysis. Cancer: 2012 Jul;118(13):3321–9. 3. Nazzani S, Pressier F, et al. A contemporary analysis of radiotherapy effect in surgically treated retroperitoneal sarcoma. Radiother Oncol: 2018 May;127(2):318–325 4. Wilkerson ML, Lin F, Liu H, Cheng L. The Application of Immunohistochemical Biomarkers in Urologic Surgical Pathology. Arch Pathol Lab Med: 2014 Dec;138(12):1643– 1665. 5. Cananzi FC, Ruspi L, Sicoli F, Minerva EM, Quagliuolo V. Did outcomes improve in retroperitoneal sarcoma surgery? Surg Oncol: 2019 Jan; 28(1):96–102 6. Brown RE, St. Hill CR, Greene QJ, Farmer RW, Reuter NP, Callendar GG, Martin RCG, McMasters KM, Scoggins CR. Impact of histology on survival in retroperitoneal sarcomas. Am J Surg: 2011 Sep;202(4):748–753

Scholarly Research In Progress • Vol. 3, November 2019

HPV Vaccination: A Globalized Paradigm Tian L. Mauer1,2†, Anna Bukowski2, Maxwell Gruber2,3, and Vikram Siberry2,4

Geisinger Commonwealth School of Medicine, Scranton, PA 18509 Georgetown University Medical Center, Department of Biomedical Graduate Education, Washington DC 20057 3 Elson S. Floyd College of Medicine, Washington State University, Spokane, WA 99210 4 Georgetown University School of Medicine, Washington DC 20007 †Doctor of Medicine Program Correspondence: tmauer@som.geisinger.edu 1

Abstract In recent years, anti-vaccination movements have gained traction throughout all regions of the world, most notably in Western countries. A rise in anti-vaccination propaganda has sparked a growing “anti-vax” movement that protests the use of any sort of vaccination, based on an unfounded fear of symptomatic manifestations leading to debilitating medical conditions and irreversible damage. Human papillomavirus (HPV) vaccination, in particular, has faced resistance since its inception in the 2000s. Public health concerns relating to a myriad of its purported side effects, ranging from psychological disorder manifestations, motor dysfunction, acquired central nervous system syndromes, autoimmune development, and even the possibility of death, have polarized the masses. However, in light of increasing evidence supporting the overall benefits of HPV vaccination, the authors of this review seek to use a cohort of global research to illustrate the urgency of HPV vaccination as a public health asset and to establish a new dialogue pertaining to the overwhelming benefits of vaccination for people of all demographics.

Introduction Human papillomavirus (HPV) is a primary causative agent for various genotypes of squamous cell carcinomas, with upward of 70% of diagnosed cervical cancers being derived from HPV types 16 and 18 (1). In terms of a global cohort, 265 762 cervical cancer associated deaths occur annually, while cervical cancer remains the second most prominent phenotype of cancer in women ranging from age 15 to 44 (1). Within recent years, the rise in mortality among HPVassociated squamous cell carcinomas has been targeted as a means of drug-related therapy. Currently, three vaccination profiles are available: Cervarix (GSK Biologicals), a bivalent HPV vaccine that prevents infections of the HPV 16 and HPV 18 subtypes; Gardasil (Sanofi Pasteur MSD), a quadrivalent HPV vaccine that, in addition to preventing infections of the HPV 16 and HPV 18 subtypes, also protects against infection of the low-risk HPV 6 and 1 genotypes; and finally Gardasil9 (Sanofi Pasteur MSD), a nonvalent HPV vaccine that acts similar to the quadrivalent Gardasil but protects against five additional genotypes: high-risk HPV 31, 33, 45, 52, and 58 (2). Considering full vaccination coverage and efficacy of 100%, the epidemiological burden of HPV-related squamous cell carcinomas that vaccines could prevent totals 25,267 cervical cancer-related deaths (1). Although the epidemiological variant coverage is vast, the adverse effects of each of the three vaccination profiles have been widely documented. The aim of this review is to 28

identify each area of adverse side effects and refute possible incidence of vaccine-related development regarding: 1) Media and physician influence on vaccine uptake, 2) Gross adverse side effects and symptomology, 3) Central nervous system demyelinating syndromes, 4) Autoimmune diseases.

Results and Discussion Media and physician influence on vaccine uptake Consumers can turn to multiple sources to learn about the HPV vaccine in order to make an informed decision about whether or not to vaccinate. The following section examines characteristics of three different kinds of sources that convey information about HPV vaccination: mass media, local media, and physicians. A study conducted in Europe explored how mass media conveyed information about the HPV vaccine to the public (3). The investigators examined 434 newspaper articles and 102 website reports in Germany and Spain relating to HPV vaccination and cervical cancer prevalence (3). While about two-thirds of website and half of newspaper reports in each country presented numerical estimates of cervical cancer incidence, fewer than 20% of media reports contained numerical estimates of HPV vaccine effectiveness in reducing dysplasia risk (3). Furthermore, fewer than 10% of articles contained accurate estimates of vaccine effectiveness at reducing dysplasia risk (3). From these reports, websites and newspapers in these countries appear to make an effort to inform the public of the link between HPV infection and risk of cervical cancer, but do not elucidate the beneficial protective effect conferred by the HPV vaccine (dysplasia risk reduction). The study also reported that approximately 70% of media reports attempted to explain the HPV vaccine to readers, whereas only around 27% attempted to advertise the vaccine (3). An explanatory tone better serves articles that inform readers about new medications or therapies while an advertising tone may be more effective at communicating the questionable link between serious side effects and the HPV vaccine, such as those we dispute in following sections. Doctors [Germany] and government sources [Spain] presented the majority of vaccine information (3). Information presented by someone with authority on a topic is more likely to be accepted as fact by the reader. While media outlets in Spain and Germany presented ample information about risks of cervical cancer from valid sources, they did not adequately present the role of HPV vaccine in reducing cancer risk. Another study investigated a more local and interactive method of sharing information—HPV vaccine public online forums in Romania (4). The public forums did not require

HPV Vaccination: A Globalized Paradigm

Table 1. Adverse events (AEs) comparing Gardasil (9vHPV) and Gardasil (qHPV) reported from Day 1 to Day 15 after a vaccination visit in males 16â&#x20AC;&#x201C;26 years of age (6).

personal information to make an account and post comments, so forum users were anonymous (4). The investigators divided users from 20 different forums into three categories: information seekers who were undecided with respect to opinion about the HPV vaccine, anti-HPV vaccine, and pro-HPV vaccine (4). The investigators conducted a qualitative analysis of recurring thematic elements in anti-vaccine and pro-vaccine forum comments (4). The anti-vaccine cohort frequently used vivid vignettes and misinformation to elicit fear of HPV vaccine side effects and displayed doubts about the trustworthiness of the government and pharmaceutical companies, arguing that they are most concerned with personal profits (4). Accurate information conveying the risk of cervical cancer and safety of the HPV vaccine could refute scare tactics (such as vivid vignettes) and show people that, even if they may not be able to fully trust companies and the government, the HPV vaccine itself is necessary. Some vaccine dissenters claimed that administering the HPV vaccine in younger girls would encourage them to engage in sexual activity (4). This argument does not discredit the effectiveness of the vaccine itself, yet still may discourage parents from having their daughters take advantage of an intervention for cervical

cancer protection. Countering the increased promiscuity claim would require providing evidence that there is no link between HPV vaccination and earlier sexual activity through conducting studies and making some parents feel obligated to initiate earlier conversations about responsible sex. These arguments against HPV vaccination successfully convinced some undecided people join the anti-vaccination cohort (4). This aspect is troubling, considering that anonymous posters could fabricate their own credentials and personal vignettes yet still be able to prevent someone from taking advantage of a preventive medical intervention. Pro-vaccine bloggers often shared biomedical evidence from widely respected sources such as Food and Drug Administration, detailing effectiveness and the minor sideeffect profile of the HPV vaccine (4). Presenting a carefully constructed, evidence-based argument forms the basis of a scientific discussion. However, many vaccine advocates responded to anti-vaccine posts in a frustrated, demeaning manner, using mockery and name-calling (4). Not only do demeaning comments relating to another personâ&#x20AC;&#x2122;s character detract from scientific discussion, they also call into question

HPV Vaccination: A Globalized Paradigm

Table 2. Percentage of subjects reporting the occurrence of potential immune-mediated diseases symptoms within 1 year of any dose, classified by immune-mediated disorder. Estimated relative risks for controlled studies (total vaccinated cohort) (7).

the character of the person hurling the insult—in this case the pro-vaccine advocate—in the eyes of the undecided consumer looking for information, making the consumer potentially less likely to side with the pro-vaccine stance. Objective, evidencebased arguments can be effective in both promoting benefits and dispelling negative perceptions with regard to HPV vaccine in local forums. A study in Parkland Hospital System (Dallas, Texas) examined communication techniques doctors used when recommending the HPV vaccine to an (unvaccinated) adolescent patient/ parent duo (5). Patients were more likely to receive the HPV vaccine if their doctors used a strong recommendation, meaning an explicit, personally owned recommendation (“I recommend this”) rather than a weak recommendation, meaning a recommendation that is not personally owned (e.g., “It is recommended”) or that includes a reason to opt out of vaccination (e.g., “School doesn’t require it”) (5). However, more doctors used weak recommendations (37.2% of interactions) than strong recommendations (25.6% of interactions) (5),

which could contribute to lack of vaccine uptake. Providing rationales (e.g., cancer risk reduction) during HPV vaccine recommendation counseling was associated with higher vaccine administration (5). Presenting a vaccine as optional detracts from the message that the vaccine is necessary; direct endorsement by a medical professional entrusted to providing care to the patient helps the patient/parent see the necessity of taking the HPV vaccine. Overall, physicians need to make strong, personally owned recommendations with solid rationales and minimal excuses for patient opting out in order to facilitate uptake of HPV vaccine. Gross adverse side effects and symptomology Over the span of 25 years, a myriad of symptoms has been documented in randomized control trials pertaining to Cervarix, Gardasil, and Gardasil9. These adverse side effects range from simple psychosomatic pain, needle injection site pain (erythema, pain, swelling), and vaccination pain (headache and pyrexia) to serious adverse effects such as death.

HPV Vaccination: A Globalized Paradigm

In a longitudinal meta-analysis comparing adverse side effects in the quadrivalent Gardasil versus the quadrivalent plus extra protection Gardasil9, 15,000 young men and women ranging from ages 9 to 26 received >1 dose of Gardasil9 across 9 studies (6). Vaccinations were administered in a monitored pattern with the dosages being administered at Day 1, Month 2, and Month 6 (6). With administration of Gardasil9, myriad symptoms were noted. These symptoms included pain relating to administration at the injection site, vaccination pain, and other symptoms (Table 1) (6). The authors of the meta-analysis concluded that the 9vHPV vaccine (Gardasil9) was generally well tolerated and exhibited a very similar safety profile to qHPV (Gardasil) (6). In an alternative meta-analysis conducted in 2009, the side effects of Cervarix were examined and compiled in an effort to enhance the safety profile of the vaccine. In a trial of 96,704 young girls and adult women, 31,173 comprised the HPV group, 24,241 comprised the co-adjuvant (Coad) group, and 24,241 comprised the control group (7). Each was assigned with measured side effects ranging from unsolicited adverse side effects (Unsol. AEs), medically significant side effects (MSCs), serious side effects (SAEs), and potential immune-mediated diseases (pIMD). Within the first 29 days of being vaccinated by Cervarix, the most reported side effects were upper respiratory tract infections, nasopharyngitis, and headache in each of the groups (7). Throughout the entire study period, the most

Figure 1. Percentage (95% CIs) of all women reporting all unsolicited adverse symptoms (Unsol AEs), medically significant conditions (MSC), potential immune-mediated diseases (pIMDs), and serious adverse events after vaccination (SAE) (Total vaccinated cohort): 30-day and entire study follow-up periods (7).

common Unsol. AEs were gynecological chlamydia infections in the HPV and control groups and nasopharyngitis in the Coad group (Figure 1) (7). In the MSC grouping, the most commonly reported MSCs within the first 29 days of administration were gynecological chlamydia infections and gonococcal infections in the HPV and control groups, along with bronchitis and headache being the most common in the HPV group (7). Throughout the entire study period, the most frequently reported MSCs were gynecological chlamydia infection and depression in the HPV and Coad groups (7). For SAEs, the most frequently recorded adverse event was the propagation

of appendicitis within 30 days of administration in the HPV and Coad groups (7). This occurred in approximately 0.1% of the population (7). Overall 2,448 women throughout the study reported the presence of a SAE (7). The most common SAE throughout the entire study period was spontaneous abortion followed by appendicitis (105 in the HPV group [0.3%]), 111 in controls (0.5%) and 5 in the Coad group (0.2%) (7). The most frequently recorded outcome of pregnancy during the trials was the healthy delivery along with lack of any congenital abnormalities related to the vaccination (7). Finally, the potentiation of immune disorders is of statistical significance. Measured within 1 year of the administration of any dosage, the frequency of immune disease mediated development was similar in each of the groups (<0.3%, Table 2). Although side effects were more prevalent in this Cervarix trial, it should be noted that the parameters of the study in particular increased the likelihood for documentation of minimal and serious side effects. Prior to the beginning of the study, the participants were required to submit a negative pregnancy test along with self-administration of contraceptive, 30 days from the first dosage through 2 months following completion of the vaccination series (7). However, pregnancy did occur in the study, with 10,476 pregnancies being recorded within the duration of the trial (7). Within the parameters of the study, as compared to previous meta-analysis, the candidates were followed up to 8.4 years as opposed to 5.5 years, consistently underwent regular screening for Chlamydia trachomatis and Neisseria gonorrhoeae, as well as underwent consecutive mental examinations (7). It should be also noted that two female populations (ages 15 to 19 and ages 20 to 24) have the highest incidence of Chlamydia trachomatis infection (2,994.4 and 3,730.3 cases per 100â&#x20AC;&#x2030;000 females, respectively) compared to the rest of the U.S. population (8). Also, within the United States, 16.2% of adults exhibited one major episode of depression (9). This statistic further strengthens the argument that consistent screening is correlated to a higher incidence of depression within this particular observed population. In study 1 (study 001), the incidence of spontaneous abortion was found to be 9.1% in the Gardasil9 group, while it was found to be 11.1% in the Gardasil group (6). However, a study solely identifying the incidence of spontaneous abortion in first semester pregnancies (n ~ 237â&#x20AC;&#x2030;755), found 15% or 35,862 spontaneous abortions occurred (10). This provides evidence showing the incidence of spontaneous abortion to be within the normal statistical metrics of a given population. Within each of the studies detailed above, mortality was noted. In the meta-analysis assessing the safety profile of Cervarix, 63 deaths occurred: 25 in the HPV group, 20 in the control, and 18 in groups that were blind (7). Each of these deaths in the HPV group were deemed non-vaccine related, with suicide accounting for the greatest cause (7). In the phase III clinical trials assessing the safety profile and symptomatology comparison of Gardasil9 vs. Gardasil, 7 deaths were reported, with each of the deaths also being ruled out as nonvaccination-related deaths (6). The causes ranged from suicide, malignant lymphoblastic cancer, and vehicular accident to hypovolemia as a result of septic shock (6). As with any therapy, there will also be a balance between therapeutic benefits and side effects. As detailed above, the number of side effects decreases as vaccinations become 31

HPV Vaccination: A Globalized Paradigm

more specific for a given genotype. Substantial risks of potentially lifethreatening symptoms associated with squamous cell carcinoma significantly outweigh the effects of documented minimal side effects of the HPV vaccine. Risk of central nervous system demyelinating syndromes One of the foremost safety concerns with any vaccine, particularly with one that has recently received licensure, is any potential adverse effect in relation to demyelinating diseases such as multiple sclerosis (MS). In response to the licensure of the quadrivalent HPV vaccine (qHPV vaccine) in 2006, a wide range of international scientific studies over the course of the last decade, ranging in both scope and design, systematically addressed these concerns. This section of our review intends to refute the argument that there is a causal relationship between acquired central nervous system demyelinating syndromes (CNS ADS), particularly MS, and the qHPV vaccine, by using statistically significant evidence from the studies detailed below. A BMJ study conducted in 2013 analyzing the neurological outcomes following qHPV vaccination, among other serious adverse effects, provided an early corroboration to the evidence of a 2015 JAMA study detailed below. Both of the papers discussed in this section were of incredible statistical significance, as they made use of the same nationwide vaccination registers of young women numbering in the Figure 2. Association between exposure to quadrivalent human papillomavirus (qHPV) millions in Denmark and Sweden (11, 12). vaccine and adverse events in adolescent girls in Denmark and Sweden, October 2006 Using a powerful sample size nearing through December 2010. Rate ratios are adjusted for country, age in 2-year intervals, calendar 1 million young women aged 10 to 17, year, and parental country of birth, parental education, and paternal socioeconomic status (11). this BMJ study found that of 12 possible neurological outcomes initially proposed for apparent association with qHPV The most compelling study, published in 2015 in JAMA, used vaccination, only 5 neurological outcomes even met the criteria data from two nationwide cohorts of young women in Denmark for further analyses (11). The rate ratios were not significantly and Sweden over the course of a 7-year period who received increased for these five outcomes, leading these researchers to the qHPV vaccine (12). Led by the Department of Epidemiology conclude that there was no evidence to support an association Research at Statens Serum Institut in Copenhagen, Denmark, between qHPV vaccination and neurological adverse events, researchers concluded that there was no evidence to support as detailed in Figure 2 (11). a correlation between qHPV vaccination and MS or any other A study conducted by Kaiser Permanente in southern California, in conjunction with the Department of Neurology at Los Angeles Medical Center, further confirms these findings. Published in 2014, this JAMA Neurology article analyzed the vaccination records of nearly 100 young women aged 9 to 26, correlating with the indicated age range for HPV vaccination (13). Researchers found no association of HPV vaccination with MS or any CNS ADS, up to 3 years post-vaccination (13).

demyelinating diseases of the central nervous system (12). Further, this study supported an overall favorable safety profile of the qHPV vaccine, while also supplementing prior, smallerscale studies examining the risk of MS and CNS ADS following qHPV vaccination (12). The power of this study and its findings stems from its various strengths, some of which were unique. Foremost among these strengths was the studyâ&#x20AC;&#x2122;s large sample size, nearing

HPV Vaccination: A Globalized Paradigm

cofactor in individuals with subclinical autoimmunity because this mechanism would be expected to hasten symptom onset but not change the long-term risk of developing MS.” (13). One could even argue that this aspect of the vaccine proves beneficial to the patient, as accelerating the subclinical to clinical transition may allow the physician to detect the disease earlier, facilitating more timely treatment. Risk of autoimmune diseases A major concern with the HPV vaccines is the possibility of the development of autoimmune diseases. The concern of young and middleaged women developing autoimmune diseases is grounded in the fact that they are the leading cause of death for women in those age groups in the United States (14). Figure 3. Inclusion of girls and women in the cohort analysis and in the self-controlled However, this risk is not associated with the case-series analysis (12). Multiple sclerosis cases were defined as an incident diagnosis HPV vaccination, as this section will discuss. of multiple sclerosis (International Statistical Classification of Diseases, Tenth Revision Among the diseases discussed in this section [ICD-10] code G35) during follow-up. Other demyelinating disease cases were defined are Behcet’s syndrome, Raynaud’s disease, as a first incident diagnosis of optic neuritis (ICD-10 code H46), acute disseminated Type 1 diabetes, central demyelination and encephalomyelitis (ICD-10 codes G040, G378), transverse myelitis (ICD-10 code G373), neuromyelitis optica (ICD-10 code G360), and other central demyelinating diseases MS, connective tissue diseases, Guillain-Barré (ICD-10 codes G368, G369, G379) during follow-up. In the self-controlled case-series syndrome (GBS), autoimmune thyroiditis, and analyses, cases occurring after the 2-year risk period were included in the modeling and Grave’s disease. Because MS and CNS ADS therefore the number of cases differed between the self-controlled case-series analysis were discussed in great detail previously in and the cohort analysis. the CNS ADS section of this review, they will not be addressed in this section. This section will discuss studies from around the world, almost 4 million young women in total, of which 800 000 were focusing on the autoimmune diseases listed above, that will vaccinated during the study period from 2006 to 2013 (12). further prove that the fear of autoimmune disease development Further, this study addressed a major weakness of the 2013 due to the HPV vaccines is unsubstantiated. BMJ study, in that it included a self-controlled case series analysis in addition to a cohort analysis (11, 12). As illustrated The same Swedish and Danish study mentioned in the in Figure 3, the cohort analysis and the self-controlled case previous section, produced by Arnheim-Dahlstrom et al., series analysis were consistent, confirming with statistical also described the relationship of the qHPV vaccine and 23 significance that there was no increased risk of MS or any other different autoimmune diseases (Figure 2). The study evaluated demyelinating disease (12). Lastly, the unique opportunity to data of 296 826 females, aged 10 to 17, who received the HPV analyze a cohort consisting of millions of subjects from two vaccine, and observed that only 23 out of 53 outcomes were countries with similar public health systems and nationwide found to have new autoimmune diseases (11). Twenty of those vaccination registers allowed researchers to not only enhance 23 outcomes did not have rate ratios that were statistically the statistical power of their findings, but also to ensure a high significant (11). The remaining 3 outcomes initially showed degree of generalizability (12). a rate ratio that implied significant association between the qHPV vaccine administration and Behcet’s syndrome, In addition to presenting scientific evidence directly refuting Raynaud’s disease, and Type 1 diabetes. However, after further the misconception that there is a causal relationship between investigation, the association between the qHPV vaccine HPV vaccination and CNS ADS, it is important to address administration and Behcet’s syndrome, Raynaud’s disease, the possible origins of the concerns. The studies detailed and Type 1 diabetes was discovered to not be plausible. above that have examined this purported relationship The absence of significant association between the HPV attribute such worry to several factors, such as the unmasking vaccination and the 3 diseases was due to lack of temporal phenomenon. This theory posits that a clinical visit for the relationship between vaccination and disease, in addition to purpose of receiving a vaccination can ultimately lead to the a weak “risk signal” due to the 3 outcomes only fulfilling 1 of 3 “unmasking” of an asymptomatic disease or condition with predetermined criteria needed to support the significance (11). such unusual symptoms that it would have otherwise gone This was a powerful study, due to the large sample size, that unobserved clinically (11, 12). A different hypothesis postulates found no increased risk of autoimmune diseases due to the that vaccination of any type may accelerate the transition of an qHPV vaccine administration (11). already present subclinical disorder, such as that of CNS ADS, to the status of a clinical disease (13). The study published in Another study tasked with assessing the association of JAMA Neurology referenced above stated that their “findings newly onset autoimmune diseases after HPV vaccination are consistent with vaccines acting as a proinflammatory was conducted in France, using The Pharmacoepidemiologic

HPV Vaccination: A Globalized Paradigm

General Research Extension (PGRx) database, by GrimaldiBensouda et al. This group examined 478 definite cases of autoimmune disease onset after vaccination in 11- to 25-year-old females and matched them to 1,869 referents from the PGRx database (15). The autoimmune diseases investigated were central demyelination/MS, connective tissue diseases, GBS, autoimmune thyroiditis, idiopathic thrombocytopenic purpura, and Type 1 diabetes (15). Grimaldi-Bensouda et al. found that there was a negative association between all of the studied autoimmune diseases together and HPV vaccination. When the diseases were analyzed separately, they found that central demyelination/MS, autoimmune thyroiditis, connective tissue diseases, and Type 1 diabetes were negatively associated to HPV vaccination (15). The data for idiopathic thrombocytopenic purpura and GBS showed no association with HPV vaccination (15). This study further supports the theory that the risk for autoimmune disease development is not elevated by HPV vaccination in females. Lastly, an observational study focused on finding evidence of the “unmasking” effect, which was mentioned previously, and its association with HPV vaccination and autoimmune diseases, particularly Grave’s disease, was conducted using data from Kaiser Permanente in northern and southern California (16). From a group of 6.6 million members, they used 189 629 females, aged 9 to 26 years old, who were vaccinated against HPV (16). The study retrospectively searched for new-onset diagnosis of Grave’s disease within 6 months after vaccination, and found only 32 cases, 18 of which were confirmed (16). Only 6 of the 18 confirmed cases of Grave’s disease were truly new after vaccination, as the others had either abnormal thyroid hormone levels drawn or had symptoms of hyperthyroidism previous to the HPV vaccine administration (16). The “unmasking” phenomenon is important to consider when looking at vaccinations and potential adverse effects. Jacobsen et al. discovered that new diagnoses of autoimmune disease, specifically Grave’s disease, were recorded within days post-vaccination. This finding is consistent with the fact that those cases had thyroid labs drawn the day of the vaccination, leading to the false attribution of the cause of Grave’s disease to the vaccination (16). It is noted by Jacobsen et al. that these coincidental findings of Grave’s disease postHPV vaccination are likely due to the fact that the vaccine is administered at routine doctor’s visits, where the symptoms of the disease could have been discovered, leading to labs being drawn the same day and diagnosis following the vaccination (16). Similarly, the patient may have been brought to the doctor with symptoms of the disease and were found to also be due for vaccination, leading to labs again being drawn the same day and diagnosis following (16). This study adds even more evidence against the misconception that HPV vaccination causes autoimmune diseases, and it is important to screen against the “unmasking” effect when further studying adverse effects of vaccinations. The wide range of autoimmune diseases studied in conjunction with HPV vaccination have been shown to have no association. Large studies are needed to compare the normal incidence of autoimmune diseases—ranging from less than 1 to more than 20 new diagnoses per 100 000 person-years—to the incidence of autoimmune diseases occurring in subjects after receiving

the HPV vaccine (17). The large size of the discussed studies helps to provide more statistically strong evidence of the lack of correlation between the development of newly onset autoimmune diseases and HPV vaccination.

Conclusion First, it is essential to recognize and appreciate the influence of both the media and the physician on HPV vaccine uptake. Since the vaccine's inception, the media has played a role in informing the public about a tremendous range of issues. Reporting of health care news is of particular significance, as the majority of the general population has not received a full medical education and therefore relies primarily on doctors, but also news and media, to interpret developments in the medical field. As a result, the role of the media was of interest to the authors of this review due to the relatively new and somewhat controversial nature of HPV vaccination. Upon review of the current scientific literature surrounding this aspect of HPV vaccinations, the authors of this review have concluded that it is of utmost importance that the media not only report the risks of HPV-induced squamous cell carcinoma proliferation, but further equally highlight the beneficial nature of HPV vaccination. In addition, governments and medical institutions should make objective, accurate information about HPV vaccination readily accessible to the public, to minimize the potentially harmful influence of unsubstantiated social media postings (such as forum comments by anonymous users) on HPV vaccine uptake. The medical opinion of the clinician is traditionally held by the patient to be of vital importance due to the clinician's expertise in a given medical specialty. In the case of the HPV vaccines, the indicated patient demographic is predominantly that of adolescents, who (even more so than adults) rely on the opinions of those in authority, as the majority are still under the supervision of someone older than themselves (i.e., a parent or guardian). Consequently, the influence of the physician recommendation even more greatly impacts the future health of the indicated patient population for this vaccine. In the opinion of the authors, physicians need to provide strong recommendations regarding the HPV vaccine to adolescents. This could be reinforced in physician training through integration in the medical school curriculum or during residency with a training module in the hospital. In light of the influential nature of media coverage and physician recommendations surrounding the HPV vaccines, the authors of this review also found it to be of consequence to highlight the urgency of weighing the benefits and risks of HPV vaccination in order for patients to make informed decisions. Like any vaccine, HPV vaccination indisputably has side effects. However, it is critical that the patient fully understand how a side effect is defined in a given context—is it mere tenderness at the site of injection, or could it be fatal? In the case of the HPV vaccines, it can be concluded with scientific certainty that the benefits outweigh the side effects to such an extent that the World Health Organization’s Global Advisory Committee on Vaccine Safety “considers HPV vaccines to be extremely safe” (18). As elucidated in this review, meta-analyses and studies around the world (some with cohorts ranging in the millions) have demonstrated, with statistical significance, that side

HPV Vaccination: A Globalized Paradigm

effects of the HPV vaccine are minor and non-life-threatening. Any purported fatal side effects from the HPV vaccines have been proven to be either unrelated to the HPV vaccines or of miniscule incidence.

10. Cohain JS, Buxbaum RE, Mankuta D. Spontaneous first trimester miscarriage rates per woman among parous women with 1 or more pregnancies of 24 weeks or more. BMC Pregnancy & Childbirth 2017;17(1):437.

As with all fields of scientific inquiry, there are areas for improvement and further study. The authors of this review noted the disproportionate prevalence of female subjects in nearly all the studies included in this review, and urge future researchers to consider the importance of studying the effects of HPV vaccination in males as well, ranging from its side effects to its benefits. It is the humble opinion of the authors of this review to call for all people, regardless of demographic, to be vaccinated for HPV.

11. Arnheim-Dahlstrom L, Pasternak B, Svanstrom H, Sparen P, Hviid A. Autoimmune, neurological, and venous thromboembolic adverse events after immunisation of adolescent girls with quadrivalent human papillomavirus vaccine in Denmark and Sweden: cohort study. BMJ 2013 Oct 09;347:f5906.

References 1.

Bruni L, Barrioneuvo-Rosas L, Albero G, Serrano B, Mena M, Gómez D, Muñoz J, Bosch FX, de Sanjosé S. ICO/IARC Information Centre on HPV and Cancer (HPV Information Centre). Human Papillomavirus and Related Diseases in the World. Summary Report 27 July 2017.

2. Capra G, Giovannelli L, Matranga D, Bellavia C, Guarneri MF, Fasciana T, et al. Potential impact of a nonavalent HPV vaccine on HPV related low-and high-grade cervical intraepithelial lesions: A referral hospital-based study in Sicily. Human Vaccines & Immunotherapeutics 2017 Aug 3;13(8):1839–1843. 3. Bodemer N, Müller SM, Okan Y, Garcia-Retamero R, Neumeyer-Gromen A. Do the media provide transparent health information? A cross-cultural comparison of public information about the HPV vaccine. Vaccine 2012;30(25):3747–3756. 4. Penţa M, Băban A. Dangerous Agent or Saviour? HPV Vaccine Representations on Online Discussion Forums in Romania. Int J Behav Med 2014 Feb;21(1):20–28. 5. Shay LA, Street RL, Baldwin AS, Marks EG, Lee SC, Higashi RT, Skinner CS, Fuller S, Persaud D, Tiro JA. Characterizing Safety-net Providers’ HPV Vaccine Recommendations to Undecided Parents: A Pilot Study. Patient Education and Counseling 2016;99(9):1452–1460. 6. Moreira J, Edson D, Block SL, Ferris D, Giuliano AR, Iversen O, Joura EA, et al. Safety Profile of the 9-Valent HPV Vaccine: A Combined Analysis of 7 Phase III Clinical Trials. Pediatrics 2016 Aug;138(2):e20154387. 7.

12. Scheller NM, Svanström H, Pasternak B, ArnheimDahlström L, Sundström K, Fink K, et al. Quadrivalent HPV Vaccination and Risk of Multiple Sclerosis and Other Demyelinating Diseases of the Central Nervous System. JAMA 2015 Jan 6;313(1):54–61. 13. Langer-Gould A, Qian L, Tartof SY, Brara SM, Jacobsen SJ, Beaber BE, et al. Vaccines and the risk of multiple sclerosis and other central nervous system demyelinating diseases. JAMA Neurology 2014 Dec;71(12):1506–1513. 14. Walsh SJ, Rau LM. Autoimmune diseases: a leading cause of death among young and middle-aged women in the United States. American Journal of Public Health 2000 Sep 1;90(9):1463–1466. 15. Grimaldi-Bensouda L, Rossignol M, Koné-Paut I, Krivitzky A, Lebrun-Frenay C, Clet J, Brassat D, Papeix C, Nicolino M, Benhamou PY, Fain O, Costedoat-Chalumeau N, Courcoux MF, Viallard JF, Godeau B, Papo T, Vermersch P, Bourgault-Villada I, Breart G, Abenhaim L. Risk of autoimmune diseases and human papilloma virus (HPV) vaccines: Six years of case-referent surveillance. Journal of Autoimmunity 2017;79:84–90. 16. Jacobsen SJ, Sy LS, Ackerson BK, Chao CR, Slezak JM, Cheetham T, Takhar HS, Velicer CM, Hansen J, Klein NP. An unmasking phenomenon in an observational post-licensure safety study of adolescent girls and young women. Vaccine 2012;30(31):4585–4587. 17. Cooper GS, Stroehla BC. The epidemiology of autoimmune diseases. Autoimmunity Reviews 2003 May;2(3):119–125. 18. World Health Organization (WHO) Global Advisory Committee on Vaccine Safety. Global Vaccine Safety: Safety update of HPV vaccines. Available at: http://www. who.int/vaccine_safety/committee/topics/hpv/June_2017/ en/. Accessed May 3, 2018.

Angelo M, David M, Zima J, Baril L, Dubin G, Arellano F, et al. Pooled analysis of large and long-term safety data from the human papillomavirus-16/18-AS04-adjuvanted vaccine clinical trial programme. Pharmacoepidemiology and Drug Safety 2014 May;23(5):466–479.

8. Center for Disease Control. Prevalence of Chlamydia trachomatis Genital Infection Among Persons Aged 14–39 Years — United States, 2007–2012:38. 9. Baxter AJ, Scott KM, Vos T, Whiteford HA. Global prevalence of anxiety disorders: a systematic review and meta-regression. Psychological Medicine 2013 May;43(5):897–910.

Scholarly Research In Progress • Vol. 3, November 2019

Understanding the Link Between Energy Production, Symbiosis, and Cancer Metabolism: A Brief Review of Recent Research Joshua Emerson Kiddish1*

Geisinger Commonwealth School of Medicine, Scranton, PA 18509 *Master of Biomedical Sciences Program Correspondence: jkiddish@som.geisinger.edu 1

Abstract The scientific community’s understanding of cancer is extensive. The extremely dynamic and diverse nature of cancer cells, however, makes a complete understanding difficult at this time. In the 1920s, Professor Otto Warburg proposed his theory that tumor cells depend on glycolysis, whether they are in aerobic or anaerobic conditions. This was revolutionary for the time. It is well-known that cancer cells typically have high rates of proliferation that require significant amounts of energy. Recent research supports and expands upon Warburg’s theory. It is now believed that cancer cells may be capable of a symbiotic relationship involving energy production fuel sources. It is hypothesized that lactate, the byproduct of anaerobic respiration, can reenter the metabolic pathway of an oxygen-rich tumor cell. This concept, in a sense, is revolutionary, as it may help explain the adaptive fitness of cancer cells. This understanding may change the way cancer is treated and managed as new therapeutic targets may be possible. This review seeks to gain more information on the current research that supports and further elaborates on Warburg’s initial theory.

Introduction Understanding the Warburg effect Professor Otto Warburg’s 1956 publication entitled On the Origin of Cancer Cells described his early research and theories on a cancer cell’s ability to thrive in anaerobic conditions. He believed that cancer had two phases. The first he described was one in which the cell’s respiratory system was damaged irreversibly. In the second phase, the injured cells struggled to obtain enough energy through oxidative phosphorylation and therefore may have used other means to produce energy (1). He believed cells that could survive hypoxic conditions by using fermentation should be considered cancerous. Warburg also believed that mitochondria had to undergo changes in order to use “fermentation” instead of traditional “respiration” (1). Considering that mitochondria are the cell’s primary producer of energy, it makes sense to presume that a change in their activity would be necessary to allow a cell to thrive in anaerobic conditions. It is now believed that the mitochondrial changes are driven by oncogene metabolic reprogramming (2). Although Warburg documented these and other assumptions, he failed to obtain significant data, especially on the mitochondria. This is most likely due to the fact that there were technological restrictions at this time, compared to today’s methodologies (3).

The Warburg Effect later became known as aerobic glycolysis, i.e., conversion of glucose to lactic acid in the presence of oxygen (3). Warburg also assumed that cancer can thrive in anaerobic conditions, which is now widely accepted. Interestingly, the scientific community is just now gaining an understanding of the mechanism by which it does so (3). For example, a recent publication by Sebastian et al. proposed that SIRT6, which is a tumor suppressor, regulates aerobic glycolysis in cancer cells (4).

Discussion The effects of hypoxia and acidic conditions on carcinogenic tumor cell growth Recent research suggests that cancer cells do not use glucose as a sole source of energy. It is possible that the byproduct of glycolysis, lactate, can be used as a fuel source by some cancer cells (3). It is now thought that there is a potential symbiotic relationship between cancer cells in well-oxygenated states and those in hypoxic states based on research that demonstrated lactate being used as a fuel source under anaerobic conditions (3). It is well-established that most tumor tissue samples are very acidic when compared to local, healthy samples (5). The ability of tumor cells to continue to proliferate in acidic conditions was the topic of research performed by Helmlinger et al. in 2002. Dr. Helmlinger and his group sought to demonstrate that the low pH associated with carcinogenic tumors is the result of multiple chemical reactions and their respective byproducts (5). Their research did in fact support the previous hypothesis of lactic acid being a source of increasing acidity, but their research also suggested that carbon dioxide that is converted into carbonic acid is also a significant cause of increased acidity (5). This investigation used both in vitro and in vivo techniques. In the in vitro trials, glucokinase activity was inhibited by using mannoheptulose. It is believed that decreased glucokinase activity is commonly associated with cancer cells (5). The in vivo studies were completed by inserting tumor cell tissues into a dorsal skin fold chamber in mice. Tumor growth, blood flow, and pH were monitored (5). The results of the in vitro study demonstrate that cells’ lactate and glucose concentrations decrease with time and are dependent on whether the cell is parental or glycolysisimpaired (5). Parental cells can be defined as those that were originally from healthy tissue. In some other contexts, stem cells can serve as parental cells. The parental cells had higher lactate concentrations, lower glucose concentrations, and an overall basic pH after just 4 hours of incubation (5).

Understanding the Link Between Energy Production, Symbiosis, and Cancer Metabolism

The cell density of glycolysis-impaired cells is also worth noting. The cell density of both cell types (parental and glycolysisimpaired) increased over time, but the glycolysis-impaired cells had an average density just above 4,000 cells per mm2 after just 3 hours, whereas the parental cells had an average cell density of slightly more than 8,000 cells per mm2 (5). The ability of these cells to proliferate was then studied further with the in vivo study. In the in vivo study, glycolysis- impaired cells had a significantly higher ratio of hydronium to lactate production than did the parental tissue samples (5). Lactate yield was considerably higher in glycolysis-impaired than in the parental cells (5). The results of this analysis indicate that impairing a cell’s natural ability to perform efficient glycolysis, by reducing phosphoglucose isomerase activity, may in turn cause a decrease in lactate produced, but will also cause a decrease in pH. This is believed to be the result of carbonic acid (5). Understanding a cancer cell’s ability to thrive in hypoxic conditions has been broadly studied since the time of Warburg. Research published in 2002 by Minchenko et al. demonstrates a possible link between hypoxic states and specific oncogene expression. It is widely accepted that cancer is affected by certain genes becoming activated. Dr. Minchenko and his group paid specific attention to expression of 6-phosphofructo2kinase/fructose-2,6-bisphosphatase-3 (PFKFB3) under hypoxic stress. PFKFB3 is believed to play a significant role in regulating glycolysis. The study also attempted to demonstrate the effect of PFKFB3 in the Pasteur Effect (6), which is defined as a phenomenon in which anaerobic glycolysis is inhibited, typically by the presence of diatomic oxygen. It is known that under hypoxic stress in animals, hypoxiainducible Factor 1 is typically activated, which in turn can activate other genes, or be rapidly degraded once normoxic conditions are present. It is believed that hypoxia-inducible Factor-1 plays a significant role in activating PFKFB3 (6). The results of this study demonstrated that under hypoxic stress, a cell will switch from oxidative phosphorylation to glycolysis. This theory was well-supported prior to this investigation, but the authors elaborated on their finding that it is possible for PFKFB3 to be activated in hypoxic conditions. The activation of PFKFB3 is believed to greatly increase rates of glycolysis (6). The results of this research are important, as understanding these findings may help to explain the proliferation of cancer cells under hypoxia. Understanding the gene cascade that is in part responsible for increasing rates of glycolysis in hypoxic states could be a potential target for cancer treatment therapies (7). Mitochondria: The powerhouses of cells and their role in cancer It is well-known and long accepted that mitochondria are the “powerhouses of the cell,” and are responsible for ATP production. Found within the mitochondria are unique DNA known as mitochondrial DNA (mDNA). Mitochondrial DNA are part of the evidence for the proposed evolutionary history of mitochondria (8). It seems obvious that mitochondria must play a significant role in cancer cell behavior, considering the extremely large energy demand and reproduction rate of carcinogenic tumor cells. What is interesting is the fact that mitochondria in the cytoplasm of cancer cells typically are diminished in number (8). It has been proposed that

the decreased number of mitochondria present are the result of hypoxic states, which call for glycolysis instead of oxidative phosphorylation, which is typically performed by the mitochondria in normoxic states (8). Recent research indicates that mitochondria most likely undergo a conformational change when exposed to different stressors and energy demands. This proposed concept was studied by Rossignol et al. in 2004. The proposed conformational change could potentially help to explain the ability of mitochondria to successfully function in cancer cells under various stressors. The lack of glucose that is typically found in cancer cells commonly forces the cells to use an alternative energy source, which can be glutaminolysis, or the oxidation of glutamine (9). A potential substrate change was hypothesized by the researchers, since glucose is not being used. Therefore, the researchers studied how mitochondria behaved in the presence of glutamine as a substrate in hypoglycemic states (9). Upon evaluation of HeLa cells, oxidative phosphorylation was found to still be utilized by the cancer cells in the absence of glucose, and lactate was still produced as a byproduct (9). The researchers used both glucose and galactose media and found that mitochondrial respiration can be stimulated when in a galactose medium (9). The mitochondria of galactose-grown cells demonstrated a conformational change versus those grown in a glucose medium, which did not exhibit any change (9). Thus, when in the presence of galactose, the mitochondria, which were examined via transmission electron microscopy, revealed a significantly larger number of cristae. They were also found to be in a more condensed and reticular state. Understanding how cancer cells can proliferate aerobically, as well as anaerobically, is important. It is even more important to understand that oxidative phosphorylation is still possible via a substrate change from glucose to galactose. This is significant because it demonstrates that cancer cells that are able to proliferate and produce ATP even under what would be considered oxidative stress in healthy cells. Healthy cells do not proliferate well under oxidative stress, as is the case with galactose versus glucose. A recent review of compiled research on mitochondrial uncoupling states that the uncoupling nature of mitochondria in cancer cells may play a role in chemotherapy drug resistance (3). Further research is being performed on mitochondrial behavior in cancer cells as well as the potential role of gene expression in mitochondria (10). Understanding the link between transforming growth factorbeta signaling, caveolin-1, and cancer metabolism Transforming growth factor-beta (TGF-β) signaling is normally involved in numerous different processes in both developing embryos and adults. Some normal TGF-β homeostatic functions include cell growth, angiogenesis, cellular differentiation, and apoptosis (11). Based on this information alone, one could assume that TGF-β misregulation could play a very significant role in cancer. Caveolin-1 (Cav-1) is a protein coding gene with similar important functions. Guido et al. determined whether hyperactivation of the TGF-β pathway could lead to metabolic reprogramming of the Cav-1 gene. It was hypothesized 37

Understanding the Link Between Energy Production, Symbiosis, and Cancer Metabolism

that hyperactivation of the TGF-β pathway could promote tumorigenesis by causing metabolic changes (11). Previous studies have demonstrated that the loss of Cav1 is an important biomarker for poor prognoses in certain human cancers. Cav-1 typically negatively regulates the TGF-β pathway; thus, when this regulation is suppressed, overexpression of the TGF-β pathway may occur, resulting in favorable carcinogenic conditions. TGF-β can be involved in both paracrine and autocrine loops and has been known to activate fibroblasts, which further help carcinogenic tumors to proliferate in their respective microenvironments (11). This research demonstrated that a loss of Cav-1 could also be associated with metabolic reprogramming of cancer-associated fibroblasts, specifically with regard to aerobic glycolysis. This aerobic glycolysis resulted in the production of pyruvate, lactate, and ketone bodies, which were reentered into the metabolic pathway of neighboring tumor cells to further fuel energy production (11). The possible role of cellular symbiosis between carcinogenic tumor cells in hypoxic and well-oxygenated states A carcinogenic tumor is typically proliferating in close proximity to relatively healthy cells (3). It has been proposed, and is accepted at this time, that cancer cells in a hypoxic state release lactate as a byproduct of anaerobic glycolysis. Lactate, however, may not exclusively be a waste product (3). The released lactate may potentially be involved in an intimate, symbiotic relationship with cancer cells that are in welloxygenated states, as these cells may utilize it as a potential form of convertible energy (3). It is believed that lactate can enter into a well-oxygenated cancer cell’s citric acid cycle and electron transport chain, after being converted into pyruvate via lactate dehydrogenase-1 and under the condition that the mono-carboxylate transporter-1 (MCT-1) gene is activated (12). MCTs typically transport lactate from one cell to another in a process known as a lactate shuttle (12). A 2011 investigation by Whitaker-Menezes et al. demonstrated the effects of MCTs in cancer cells. MCTs act in healthy human tissues in the brain and in skeletal muscle. In the presence of cancer, MCTs were found to be upregulated, therefore causing an increase in lactate shuttles, which move the excessive lactate from a cell that is using aerobic glycolysis to a cell that is using oxidative mitochondrial metabolism (12). The concept of the lactate shuttle further supports the hypothesis of a metabolic symbiosis between hypoxic and well-oxygenated cancer cells. Under optimal conditions, two lactate molecules can be converted into two pyruvate molecules, then acetyl CoA, and after TCA. Thus, a maximum of 29 ATP may be produced (3). The significance of a symbiotic relationship between hypoxic and well-oxygenated cancer cells Symbiotic relationships occur naturally throughout almost every biological discipline. Understanding a potential symbiotic relationship between cancer cells in different states of oxygenation can play a significant role in potentially redefining treatment modalities (3). As of the current time, two major

means of chemotherapy exist: 1) Attacking cellular functions such as DNA replication; and 2) Attacking a tumor’s carbon energy source. To elaborate on the two abovementioned means of chemotherapy: Monocarboxylate transporter 1 (MCT1) for example, is a very common target for pharmacological interventions. Recent studies have demonstrated MCT1 has a relatively high affinity for lactate and is the main means of entry for lactate in oxidative cancer cells (15). A recent study by Le et al. also demonstrated that targeting and inhibiting lactate dehydrogenase may be an effective means of inhibiting tumor growth (14). It is possible that in the future, more effective cancer treatment modalities may surface as a result of an increased understanding of cancer cell behavior (7). One of the many difficulties that cancer treatment poses to modern pharmacology is the inability to completely destroy harmful carcinogenic cells without destroying many healthy cells in the process. Coupling that concept with the rate of reproduction of cancer cells and cancer’s dynamic nature, pharmacological management of cancer is difficult.

Conclusion Cancer is detrimental to animal life as we know it. In 2017, the United States Centers for Disease Control and Prevention listed cancer as the second leading cause of death (16). It has a particularly grave prognosis in humans and takes a financial and emotional toll on millions annually. Biomedical research has given clinicians and researchers alike many new insights and tools to manage cancer. Unfortunately, the biomedical community’s understanding of cancer is far from complete. Fully understanding cancer is an extremely complex undertaking. Cancer has an evolution of its own, and as a result is particularly difficult to manage in a clinical setting. The dynamic nature of cancer makes understanding all of its intricacies difficult. This is especially true in developing treatment modalities such as chemo drug targets. Over the past few decades, since the work of Professor Otto Warburg, we have gained a further understanding which has fueled even more productive and insightful cancer research projects. By progressively gaining a better understanding of cancer on a molecular basis, hopefully a cure to this disease may one day be in sight. The research that has been reviewed as part of this paper holistically evaluated the metabolic properties of cancer. Cancer metabolism, as made evident by recent research, may have not been fully understood. Previous research has shown that cancer cells not only survive, but thrive in environments that most normal cells could never survive in. Further understanding cancer cell metabolism and the possibility of a symbiotic relationship between cancer cells that are in a welloxygenated state with those in a hypoxic state will hopefully lead to advances in the way the medical community treats and manages cancer in the clinical setting. The future of cancer understanding and treatment lies in the hands of researchers and clinicians alike. Progress has indeed been made, but as time advances, so does cancer complexity. It is hoped that a cure can be found as result of a full understanding of carcinogenic processes.

Understanding the Link Between Energy Production, Symbiosis, and Cancer Metabolism

Acknowledgments I wish to thank Michael Bordonaro, PhD, for his expertise and tireless efforts in helping to make this review paper possible. Additionally, I wish to thank Darina Lazarova, PhD, for her contributions and mentorship.

References 1.

Warburg O. Origin of cancer cells. Science 1956; 123: 309–314.

2. Ward P, Thompson C. Metabolic reprogramming: a cancer hallmark even Warburg did not anticipate. Cancer Cell. 2012; 21: 297–308. 3. Nakajima, E, Van Houten, B, Metabolic symbiosis in cancer: Refocusing the Warburg lens. Mol Carcinog. 2013; 52: 329–337.

13. Semenza G. Tumor metabolism: cancer cells give and take lactate. J Clin Invest. 2008; 118 (12): 3835–3837. 14. Le A, Cooper C, Gouw A, Dinavahi R, Maitra A, Deck L, Royer R, et al. Inhibition of lactate dehydrogenase a induces oxidative stress and inhibits tumor progression. Proc Natl Acad Sci USA. 2010; 107: 2037–2042. 15. Vander-Linden, C Corbet, C. Killing two birds with one stone: blocking the mitochondrial pyruvate carrier to inhibit lactate uptake by cancer cells and radiosensitize tumors, Mol Cell Oncol 2018; 5(4). 16. FastStats-Leading Causes of Death [Internet]. Centers for Disease Control and Prevention. [cited 2019 May 21]. Available from: https://www.cdc.gov/nchs/fastats/leadingcauses-of-death.htm

4. Sebastian C, Zwaans B, Silberman D, Gymrek M, Goren A, Zhong L, et al. The histone deacetylase SIRT6 is a novel tumor suppressor that controls cancer metabolism. Cell. 2012; 151 (6): 1185–1199. 5. Helmlinger G, Schell A, Dellian M, Forbes N, Jain R. Acid production in glycolysis-impaired tumors provides new insights into tumor metabolism. Clin Cancer Res. 2002; 8: 1284–1291 6. Minchenko A, Leshchinsky I, Opentanova I, Sang N, Srinivas V, Armstead V, et al. Hypoxia-inducible factor-1mediated expression of the6-phosphofructo-2-kinase/ fructose-2,6-bisphosphatase-3 (PFKFB3) gene —Its possible role in the Warburg effect. J Biol Chem. 2002; 277: 6183–6187. 7.

Granchi, C, Minutolo, F. Anticancer agents that counteract tumor glycolysis. ChemMedChem. 2012; 7: 1318–1350.

8. Samudio I, Fiegl M, Andreeff M. Mitochondrial uncoupling and the Warburg effect: Molecular basis for the reprogramming of cancer cell metabolism. Cancer Res. 2009; 69: 2163–2166. 9. Rossignol R, Gilkerson R, Aggeler R, Yamagata K, Remington S, Capaldi R. Energy substrate modulates mitochondrial structure and oxidative capacity in cancer cells. Cancer Res. 2004; 64: 985–993. 10. Dang C, Le A, Gao P. MYC-induced cancer cell energy metabolism and therapeutic opportunities. Clinical Cancer Research 2009; 21: 6479–6483. 11. Vander-Linden C, Whitaker-Menezes D, Capparelli C, Balliet R, Lin Z, Pestell R, et al. Metabolic reprogramming of cancer-associated fibroblasts by TGF-β drives tumor growth: Connecting TGF-β signaling with “Warburg-like” cancer metabolism and L-lactate production. Cell Cycle. 2012; 11(16): 3019–3035. 12. Whitaker-Menezes D, Martinez-Outschoorn U, Lin Z, Ertel A, Flomenberg N, Witkiewicz A, et al. Evidence for a stromal-epithelial ‘‘lactate shuttle’’ in human tumors: MCT4 is a marker of oxidative stress in cancer-associated ﬁbroblasts. Cell Cycle 2011; 10: 1772–1783.

Scholarly Research In Progress â&#x20AC;˘ Vol. 3, November 2019

A Review of Genetic Markers Associated with Penile Cancer Andrew Denisenko1â&#x20AC;

Geisinger Commonwealth School of Medicine, Scranton, PA 18509 â&#x20AC; Doctor of Medicine Program Correspondence: adenisenko@som.geisinger.edu 1

Abstract Penile cancer is a rare disease with variable incidence based on geographical location and other risk factors. Additionally, pathogenic pathways for the development of penile cancer can be effectively differentiated into human papillomavirus (HPV)-positive pathways and HPV-negative pathways. Advances in medical technology have allowed the identification of biomarkers that have an association with penile cancer. Finding such markers represents a significant advancement in the ability to provide care for patients with this disease. Within the scope of the work presented here, a brief overview of genetic markers found to be associated with penile cancer within the surveyed literature is offered. The form of penile cancer discussed is predominantly penile squamous cell carcinoma, with minor discussion of the rare form known as penile schwannomas. Particular attention will be given to different markers which distinguish HPV-positive penile cancer and HPV-negative penile cancer. Some of the markers discussed for HPV-positive penile cancer include HPV genotype distribution, methylation and chromosomal recombination, expression of P16INK4a, mutations of p53, and mutations of C-Ras and myc. Discussion on HPV-negative penile cancer will include P53 expression, phosphatidylinositol 3-kinase (PIK3CA) copy number variation, inhibitor of DNA binding 1 (ID1) protein expression, laminin subunit gamma (LAMC2) expression, antigen KI-67 (Ki67) expression, and the expression of cyclooxygenase (COX) genes. These and additional biomarkers are listed in Table 1. Advances in the overall understanding of penile cancer and improvements in treatment modalities represent promising developments in our understanding and ability to treat this rare disease.

mean age at which most individuals are found to have penile cancer is 60 years old, with occasional presentation in younger individuals (1). Incidence rates of penile cancer vary based on the geographical distribution of populations. Populations in developing countries throughout Africa, Asia, and South America can have incidence rates for penile cancer comprising 10% of malignancies, while populations in developed countries in Western Europe, as well as within the United States, have an incidence rate of penile cancer that composes 0.4% to 0.6% of malignant diseases (1). Penile cancer exhibits low, intermediate, and high risk based on its histological classification, with basaloid tumors, adenosquamous, sarcomatoid, and poorly differentiated squamous cell carcinoma constituting the highest-risk forms (2). Classical symptoms of penile cancer typically involve visible lesions or abnormalities on the penis, with the majority occurring on the glans penis, corona, or foreskin and occasionally associated infections and ulcerations of the region. Additional symptoms that have been observed include balanitis, rash, or bleeding. The diagnosis of penile carcinoma is achieved via tissue biopsy. (4). Surgical modes of treatment of penile cancer can include limited excision, topical treatment, radiation treatment with adjuvant chemotherapy, laser ablation, or penile amputation for more aggressive forms of penile cancer (4).

Introduction Penile cancer is a rare disease with several potential etiologies and exhibits variability in incidence and prevalence based on geographical location and population. Approximately 26,000 new cases of penile cancer are recorded annually (1). The most common form of penile cancer is penile squamous cell carcinoma, accounting for approximately 50% to 60% of penile carcinoma (2). Rarer forms of penile cancer include penile schwannoma, with fewer than 34 cases recorded in the literature as of 2017 (3). The 40

Table 1. A summary of biomarkers found to be associated with penile cancer in HPV-positive and HPV-negative markers, with additional markers referenced in the literature.

Genetic Markers Associated with Penile Cancer

Risk factors for penile cancer include poor hygiene, phimosis, smegma retention, and HPV infection. This is of particular significance because penile cancers can be grouped into cases that are HPV-positive or HPV-negative, with HPV-independent and nonviral carcinogenesis exhibiting pathologies related to dysfunction of p14/MDM2/P53 or P16/ cyclin D/Rb pathways (1). Other HPV-negative causative sources of penile cancer include inflammation and lichen sclerosis (1). Advances in medical technology have allowed the association of certain genetic markers with various forms of penile carcinoma. Within the scope of this review, markers associated with HPV-positive and HPV-negative forms of penile carcinoma will be discussed, along with markers for rarer forms of penile cancer, such as penile schwannomas. Additional attention will be given to future directions in the prevention and treatment of the various forms of penile cancer. The determination of genetic markers is an important advancement in the understanding of penile cancer pathogenesis. The identification of such biomarkers is likely to be helpful in diagnosis, as well as with personalized treatment protocols. In addition, genetic biomarkers have the potential to be a powerful screening tool in at risk patient populations. Identifying genetic biomarkers for penile cancer is a promising development for the medical advancement of treatment for this rare disease.

Materials and Methods The literature search was conducted by utilizing the PubMed and Google Scholar databases. Search term structured combinations and text were used to find results for penile cancer, genes, and biomarkers. Only English-language publications were used, with publications written up to February 2019. Within each publication, reference lists were evaluated to find relevant studies for this review. Only peerreviewed publications were included. The utilization of publications proceeded through screening based on review of the title, then abstract and evaluation of the full text. Following a review of the literature, a data table was produced (Figure 1) compiling the various discussed putative biomarkers of penile carcinoma.

Results Markers associated with HPV-positive penile cancer As discussed, HPV infection has a significant association with penile cancer, encompassing its own category and carcinogenic pathway of penile malignancy. This association can be quantified with techniques such as detection and genotyping of tumor tissue, allowing the identification of viral genotypes within tumor fragments (5). In addition, the biochemical mechanisms relevant to the carcinogenic properties of penile cancer have been explored in the literature and shed light on interactions between viral and host DNA. Some of these properties include viral DNA methylation, junctions between viral and cellular DNA, and genomic variation (6). In addition, explorations on the association between existing host mutations and HPV infection will be discussed. These findings potentially provide valuable markers in identifying and categorizing HPV-positive penile carcinoma.

HPV genotype distribution in penile carcinoma An approach that localized viral DNA within penile tumor tissue was implemented in a retrospective cohort study of a Brazilian population from 2003 through 2015 (5). Following viral DNA extraction from tumor tissue and statistical analysis, it was found that the prevalence of HPV DNA was 30.6% in tissue samples of penile cancer. The type of viral DNA that was the most common was HPV 16 (62.5%), followed by HPV 18 (5.4%), and HPV 6 with HPV 11 (12.4%). These forms of viral DNA were significantly associated with the variable tumor grade form (II/ III) (5). The authors of the study did not determine any significant difference for survival when comparing individuals with penile cancer that were HPV-positive or HPV-negative, nor did they find any association with HPV infection and lymph node metastasis (5). Despite the determined associations, the authors determined that the use of HPV DNA as a marker for penile cancer requires further elucidation as a viable tool. A 2018 meta-analysis of published by Olesen et al. corroborated some of these findings, showing that pooled HPV DNA prevalence in penile cancer across 52 studies and 4,199 cases was significant at 50.8%, with a high pooled HPV DNA prevalence in the more severe basaloid squamous cell carcinomas (75.7%). These findings highlight a potential benefit for using HPV vaccines in males as a preventive measure against penile cancer (7). DNA methylation and chromosomal recombination While the presence of viral DNA within penile cancer tissue is a compelling finding, an investigation into the mechanisms associated with viral infection provide a wider array of biomarkers and potentially more advanced and targeted therapies. An investigation into some of these properties was carried out by Kalantari et al., and elucidates insights related to the carcinogenic mechanisms of viral infection. These mechanisms include viral DNA methylation, chromosomal recombination, and genomic variation. Of particular interest within this study was the hypermethylation of the viral L1 gene, a major capsid polypeptide, as a potential biomarker for penile cancer progression (6). The study design compared DNA methylation patterns of HPV 16 and 18 in penile cancer to cervical neoplasia and cervical lines. The authors found that there was hyper-methylation of the L1 capsid gene of HPV 16 in 17 of 19 samples where HPV DNA was integrated into penile carcinomas. Specifically, 58% of CpG islands were found to exhibit this hypermethylation (6). Additional methylation sites included the enhancer and promotor of the HPV-16 genome, which is potentially indicative of advancing carcinogenesis (6). The methylation patterns of HPV 16 were analogous to cervical malignancies, showing potential for use as a biomarker of carcinogenesis. An L1 hypermethylation pattern was also seen with HPV 18 DNA, with observation of hypermethylation occurring in 2 penile carcinomas. Another facet of viral DNA oncogenic properties explored was recombination with human DNA at specific locations. Following analysis, it was found that there was HPV 16 DNA integration in chromosomes 3, 8, 9, and 20 (6). The authors of the study remarked that such recombination is a likely prerequisite for the observed methylation patterns of the viral genome. 41

Genetic Markers Associated with Penile Cancer

Genomic variation The study additionally remarked upon differences in carcinogenesis between different variants of HPV 16 as discussed in other epidemiological studies concerning. Namely, AA and Af variants of HPV 16 exhibit more carcinogenic properties than E variants in inducing cervical cancer. While unable to explore HPV variants in asymptomatic individuals, the authors of the study determined that 9 of 17 samples contained AA variants, with 8 samples containing E variants. Due to this small sample size, a significant association between variant and carcinogenic properties could not be elucidated (6). Altogether, the findings presented in this study served to shed light on the mechanisms of causality between HPV and carcinogenesis and showed similarities in etiology between HPV-positive cervical cancer and HPV-positive penile cancer. Expression of p16INK4a A somatic genetic marker that has been associated with the presence of HPV is p16 INK4a. A study carried out by de Andrade Martins et al found that in cases positive for expression of p16 INK4a, 95% (p=0.032) of them contained HPV DNA. These associations were confirmed with histological examinations (8). This significant association was not found with the p16 INK4a protein, which was only associated with tumor subtype. Another study described by Manweiller et al. was able to associate p16 INK4a expression with penile preinvasive and invasive cancer with a specificity of 100% specificity regardless of HPV genotype (9). Findings presented by McDaniel et al. corroborated these findings, which found that 28% of sampled subjects harbored p16 expression. In addition to this, p16 expression (along with presence of HPV infection) showed a significant association with the histological form of penile cancer (10). Interaction between HPV infection and host mutation of P53 Within the literature, other mutations within the host cell line have also been observed to coincide with the carcinogenic properties of HPV infection, though these studies have not been extensively concerned with penile cancer (12). A study published by Stoehr et al. sought to determine an association between single-nucleotide polymorphisms (SNP) (p.Arg 72) in the host p53 and HPV infection. The results of the study found no association between the host SNP and increased risk of penile cancer in the presence of HPV (12). C-ras and myc Another mutation pair that warrants further exploration is concerned with mutations which involve the proto-oncogenes C-ras and myc and the way in which they relate to HPV-positive penile cancer. A study published by Leis et al. addressed this association in a small number of patients with penile cancer, isolating HPV in the primary tumor and linking the presence of HPV to mutations in TP53 and c-ras (13). Additional studies link the presence of HPV 16 and 18 to mutations in c-myc and n-myc (14). These mutations in ras and myc therefore warrant exploration as further markers of HPV positive penile cancer. HPV-negative markers of penile cancer While the presence of HPV infection is a significant source of carcinogenesis in penile cancer, HPV-independent forms of penile cancer constitute a significant alternative pathway 42

in penile malignancies. Discussion here will include genetic markers associated with HPV-negative molecular pathways of penile cancer. In particular, attention will be given to P53 expression as a determining factor of HPV-negative penile cancer, PIK3CA copy number variation, ID1 expression, LAMC2 expression, Ki-67, and COX genes. Additionally, some attention will be given to further exploration of mitochondrial genome heteroplasmy and penile schwannomas. Other genetic markers that were not discussed in detail but have been recorded elsewhere in the literature include the tumor suppressor gene PTEN, apoptosis genes Bcl-2/BAX, growth factor VEGF, and metastasis suppressor genes KA11 and Nm23H1 (15). P53 expression In the absence of P16INK4a, P53 expression can function to act as a genetic marker of penile cancer. A study described by Mannweiler et al. sought to elucidate a means of classification of penile cancers as either HPV-positive or HPV-negative. The researchers determined that P16INK4a negativity and positive expression of P53 was consistent with HPV-independent carcinogenesis (9). This represents a valuable means of classifying the form of penile cancer (HPV-positive or negative) present. PIK3CA copy number aberration PIK3CA is a gene that encodes for a P110Îą, a catalytic subunit of PI 3 Kinase (11). A study piloted by Adimonye et al. explored the association between PIK3CA copy number gain and subsequent activity of PI3K-AKT-mTOR pathway in premalignant penile neoplasms and invasive penile squamous carcinoma. The study found that increased prevalence of PIK3CA copy number gain and increased activation of the PI3K-AKT-mTOR pathway was significantly observed in early aggressive penile carcinogenesis. (17). The researchers concluded that while the clinical applications for these findings is currently limited, further elucidation is required for finding meaningful clinical targets. ID1 overexpression ID1 is a gene which encodes the DNA binding (ID1) protein, which has been implicated as a regulatory protein in terms of tumor progression for various forms of cancer (18). Because of this geneâ&#x20AC;&#x2122;s role as a potential oncogene, the association between overexpression of ID1 and penile cancer was explored by Hu et al. Additional parameters explored by the study were overexpression of ID1 as it pertained to KaplanMeier survival analysis and Cox proportional hazard models. The results of the study showed an association with ID1 overexpression and penile cancer and lymph node metastasis. Additional findings showed poor survival when Kaplan-Meier survival plots were found to coincide with ID1 overexpression and found that ID1 overexpression was an independent predictor in Cox proportional hazard models (18). The authors of this study further investigated the result of ID1 silencing, and found a reduction of invasiveness within the Penl1 cell lines they utilized. Additional findings included the observation that ID1 potentially represses p16 and PTEN expression within cell lines, contributing to carcinogenic processes. These findings point to ID1 overexpression as a potentially valuable genetic marker and therapeutic target.

Genetic Markers Associated with Penile Cancer

LAMC2 expression LAMC2 is a gene that encodes the protein laminin gamma 2, which is a component of a heterotrimeric basement membrane protein subunit. This protein has physiological implications in cell migration and adhesion having been found to coincide with poor prognosis in other malignancies (19). A study described by Zhou et al. investigated the association of LAMC2 overexpression as well as serum LAMC2 with penile cancer for use as a potential biomarker. The results of the study found an association with LAMC2 overexpression and penile squamous cell carcinoma, as well as with LAMC2 overexpression and metastatic lymph node tissues. This association was attributed to LAMC2 and its role as a facilitator of migration, invasion, and the epithelial-tomesenchymal transition associated with carcinogenesis (19). Similar to ID1 overexpression as mentioned previously, LAMC2 overexpression was associated with shorter disease-specific survival, and the predictive value of LAMC2 was found to be better than markers such as C-reactive protein and squamous cell carcinoma antigen as reported in other studies (19). Ki-67 expression Ki-67 is a protein product that is expressed ubiquitously throughout the cell cycle and is not present in cells that occupy the G0 or quiescent state, making it a valuable marker for cells that are actively involved in the cell cycle (15) A study undergone by Gentile et al. investigated Ki-67 association with in a small number of penile cancer samples and revealed positive Ki-67 expression (16). These findings point to Ki-67 expression as another putative biomarker for penile cancer. Cox genes Prostaglandin production via the COX pathway has been shown to be a factor which promotes carcinogenesis in individuals with penile cancer. This is due to prostaglandin, particularly E2, effects on stimulating cell proliferation, preventing apoptotic events, encouraging angiogenesis, increasing cell-cell signaling, and immune suppression (15). This has led to COX gene expression being explored as a potential biomarker for penile cancer. One such study was published by Golijanin et al, who reported their findings on COX-2 and PGE synthase-1 gene expression. Both factors were found to have been upregulated in penile cancer samples, though the small sample size of 7 necessitates further exploration (20). CDKN2A CDKN2A (cyclin-dependent kinase Inhibitor 2A, P16) is a ubiquitously expressed gene encoding the proteins p16 and p14arf, which act as tumor suppressors and cell cycle regulators. As discussed earlier, this gene has been associated with HPV infections, but is also a significant somatic genetic marker of penile cancer. In the study presented by McDaniel et al., 72% of penile cancer cases that were sampled involved CDKN2A (10). In particular, CDKN2A loss coincided with a loss of heterozygosity, contributing to the carcinogenic nature of the mutation (10). EGFR and CDK4 Epidermal growth factor receptor (EGFR) and cyclin-dependent kinase (CDK4) overexpression are markers significantly

associated with penile cancer, which is remarkable due to approved treatments for EGFR- and CDK4-related abnormalities (10). In the aforementioned study, EGFR overexpression was found uniformly and diffusely in all samples when tested by immunohistochemistry (IHC). Despite this, in the aforementioned study, EGFR copy number status was heterogenous and not correlated with IHC, showing EGFR gains or amplifications in approximately 10% of penile cancer cases (10). CDK4 overexpression was also observed at a lower frequency in this study, being found in one study subject (10). Mitochondrial instability Very little exploration has been conducted in the literature regarding changes in the mitochondrial genome and the effect that they may have on the progression and pathogenicity of penile cancer. A recent publication by Araujo et al. demonstrated for the first time some of the impact of mitochondrial variations in penile cancer, with emphasis on the effect of heteroplasmy. The study found that within the examined tumor samples, heteroplasmy was found in significantly higher proportion when compared to the control and exhibited an increased number of protein coding variants. (21). In particular, some of the damaging variants included mutations in ATP6, CYB, ND1, COX3, and ND4L (20). Penile schwannomas In general, schwannomas occur as neoplasms of myelin secreting Schwann cells and are typically associated with a loss of neurofibromin 2 (NF2) gene function (22). Penile schwannomas are one of the rarest forms of penile cancer, but come with a consistent presentation across the literature. Typically presenting with individual painless nodules of the penile shaft and absent of malignant patterns, schwannomas are typically benign and can be treated via surgical excision (23). Given the rarity of this condition and the scarcity of published cases in the literature, exploration of genetic markers could be useful in determining the causative elements of penile schwannomas, potentially starting with NF2 gene function as is observed within schwannomas in other tissues. Potential treatment modalities A number of these molecular targets are the focus of clinical trials that are currently ongoing, showing the utility of biomarkers as promising tools in targeted therapy. The HPV E6/E7 pathway is one such target, which has been targeted through two distinct modalities. One such approach involves vaccination with synthetic plasmids against HPV-16 and HPV-18. This has shown promise in reducing HPV-positive lesions (24). Another HPV-specific therapy involves harvesting patient specific T-cells which can be transplanted back into donor-patients to elicit an anti-tumor response. This approach showed a 33% response rate and represents another potential treatment approach (24). Another molecular target addressed previously is EGFR, with anti-EGFR antibody panitumumab treatment showing objective response in patients (24). Vascular endothelial growth factor (VEGF) also shows promise as a treatment target, due to an observed effectiveness in the treatment of other types of cancer such as lung cancer (24).

Genetic Markers Associated with Penile Cancer

Characterization of genetic biomarkers is therefore important in identifying new targets for treatment. These targets have the potential to provide treatment with a targeted approach based on the genetic variant present within the patient. An expansion of the current understanding of biomarkers of penile cancer can equip clinicians to better understand the disease process and patient they are treating.

Discussion These biomarkers represent potentially valuable novel techniques of identifying penile cancers and for providing therapeutic interventions. The association of specific genetic variations as risk factors for penile cancer provides a framework for developing a targeted approach to developing clinical solutions and provides greater insight about the disease process involved in penile malignancies. More importantly, they give an opportunity to develop new treatment techniques and preventive measures that can improve clinical outcomes in the patient population. Based on the strong interplay between HPV and penile cancer, one of the most promising preventive measures against penile cancers is vaccination against HPV. A trial of the four-valent HPV vaccine showed that it was an effective measure against HPV infection of the external genitalia (25). These results highlight the importance of preventive measures against penile cancer before it becomes a malignant process and has potential to greatly reduce the incidence of penile cancer. Checkpoint inhibitors are a potential therapy option that have grown more prominent in recent years. This mode of therapy has shown great promise in several types of malignancy such as melanoma, lung cancer, bladder cancer, head and neck cancer, renal carcinoma, and lymphoma (26). Importantly, checkpoint inhibitors have been shown to have promising results in other cancers associated with HPV infection (27). These findings lend credibility to the notion that checkpoint inhibitors hold potential promise in addressing malignancies such as penile caner which in have strong associations with HPV infection. Other sources of therapy include an approach called genomic-adjusted radiation dose (GARD) which entails adjusting the radiation dose given in radiotherapy based on genetic properties associated within the tumor (25).

Conclusion The genetic markers and future steps described in the paper constitute a growing body of knowledge related to this rare malignancy. With further elucidation of biochemical and physiological mechanisms, more insights can be developed into understanding the disease process associated with penile cancer. Implementation of preventive measures and novel techniques for treating penile cancer have the potential to both reduce incidence of the disease and improve clinical outcomes for individuals who suffer from penile cancer.

Acknowledgments Special thanks to Heinric Williams, MD, for his support and guidance throughout the production of this work.

References 1.

Bleeker MC, Heideman DA, Snijders PJ, Horenblas S, Dillner J, Meijer CJ. Penile cancer: epidemiology, pathogenesis and prevention. World Journal of Urology (2008); 27, 141–150.

2. Guimaraes G, Rocha R, Zequi S, Cunha I, Soares F. Penile Cancer: Epidemiology and Treatment. Current Oncology Reports (2011)13(3): 231–239. 3. Kumar U, Jha NK. Schwannoma of the penis, presenting as a scrotal mass, rare entity with an uncommon presentation. Urol Ann (2017); 9(3):301–303. 4. Ritchie AW, Foster PW, Fowler S. Penile cancer in the UK: clinical presentation and outcome in 1998/99. BJU International (2004), 94: 1248–1252. 5. de Araújo LA, De Paula A, de Paula H, Ramos J, de Oliveira BR, De Carvalho K, Guimarães RA, de Alencar R, Duarte E, Rabelo Santos SH, Saddi VA, … Carneiro M. Human papillomavirus (HPV) genotype distribution in penile carcinoma: Association with clinic pathological factors. PloS one (2018), 13(6). 6. Kalantari M, Villa LL, Calleja-Macias IE, Bernard HU. Human papillomavirus-16 and -18 in penile carcinomas: DNA methylation, chromosomal recombination and genomic variation. International journal of cancer (2008), 123(8), 1832–40. 7.

Olesen TB, Sand FL, Rasmussen CL, et al. Prevalence of human papillomavirus DNA and p16INK4a in penile cancer and penile intraepithelial neoplasia: a systematic review and meta-analysis. The Lancet Oncology (2019); 20: 145–58.

8. Martins VA, Pinho JD, Teixeira Júnior A, Nogueira LR, Silva FF, Maulen VE, Khayat AS, Calixto J, Costa HA, Ramalho L, … Silva G. P16INK4a expression in patients with penile cancer. PloS one (2018), 13(10). 9. Mannweiler S, Sygulla S, Winter E, Regauer S. Two major pathways of penile carcinogenesis: HPV-induced penile cancers overexpress p16ink4a, HPV-negative cancers associated with dermatoses express p53, but lack p16ink4a overexpression. Journal of the American Academy of Dermatology (2013); 69(1):73–81. 10. McDaniel AS, Hovelson DH, Cani AK, Liu C-J, Zhai Y, Zhang Y, et al. Genomic Profiling of Penile Squamous Cell Carcinoma Reveals New Opportunities for Targeted Therapy. Cancer Res (2015);75:5219–27. 11. Hiles ID, Otsu M, Volinia S, Fry MJ, Gout I, Dhand R, Panayotou G, Ruiz-Larrea F, Thompson A, Totty NF. Phosphatidylinositol 3-kinase: structure and expression of the 110 kd catalytic subunit. Cell (1992). 70 (3): 419–29. 12. Stoehr R, Weisser R, Wendler O, Giedl J, Daifalla K, Gaisa NT, Richter G, Campean V, Burger M, Wullich B, Hartmann A. P53 Codon 72 Polymorphism and Risk for Squamous Cell Carcinoma of the Penis: A Caucasian Case-Control Study. J Cancer (2018); 9(22):4234–4241.

Genetic Markers Associated with Penile Cancer

13. Leis PF, Stevens KR, Baer SC, Kadmon D, Goldberg LH, Wang XJ. A c-rasHa mutation in the metastasis of a human papillomavirus (HPV)-18 positive penile squamous cell carcinoma suggests a cooperative effect between HPV-18 and c-rasHa activation in malignant progression. Cancer (1998); 83: 122–29.

27. Kanaan H, Kourie HR, Awada AH. Are virus-induced cancers more sensitive to checkpoint inhibitors? Future Oncol (2016). 12(23), 2665–2668.

14. Couturier J, Sastre-Garau X, Schneider-Maunoury S, Labib A, Orth G. Integration of papillomavirus DNA near myc genes ingenital carcinomas and its consequences for proto-oncogene expression. J Virol (1991); 65: 4534–38. 15. Kayes O, Ahmed H, Arya M, Minhas S. Molecular and genetic pathways in penile cancer. Lancet Oncol (2007); 8(420–429). 16. Gentile V, Vicini P, Giacomelli L, Cardillo MR, Pierangeli A, Dagener AM. Detection of human papillomavirus DNA, p53, and ki67 expression in penile carcinomas. Int J immunopathol Pharmacol (2006); 19(1):209–215. 17. Adimonye A, Stankiewic E, La Touche S, Kudahetti S, Tinwell B, Corbishley C, Lu Y, Watkin N, Berney D. The Prognostic Value of PIK3CA Copy Number Gain in Penile Cancer. Urology (2018). pii: S0090–4295(18)30560–0. 18. Hu X, Chen M, Li Y, Wang Y, Wen S, Jun F. Overexpression of ID1 promotes tumor progression in penile squamous cell carcinoma. Oncology Reports (2019), 41, 1091–1100. 19. Zhou QH, Deng CZ, Chen JP, Huang KB, Liu TY, Yao K, Liu ZW, Qin ZK, Li YH, Guo SJ, Ye YL, Zhou FJ, Huang W, Liu RY, … Han H. Elevated serum LAMC2 is associated with lymph node metastasis and predicts poor prognosis in penile squamous cell carcinoma. Cancer management and research (2018);10, 2983–2995. 20. Golijanin D, Tan JY, Kazior A, et al. Cyclooxygenase-2 andmicrosomal prostaglandin E synthase-1 are overexpressed in squamous cell carcinoma of the penis. Clin Cancer Res (2004);10: 1024–31. 21. Araujo L, Terra A, Sares C, Sobreira C, Faria E, Machado R, Rodrigues A, Muglia V, Silva W, Reis R. Mitochondrial genome analysis in penile carcinoma. Molecular Biology Reports (2018); 45(4): 591–600. 22. Hanemann CO, Evans DG. News on the genetics, epidemiology, medical care and translational research of Schwannomas. J Neurol (2006); 253:1533–41. 23. Nguyen AH, Smith ML, Maranda EL, Punnen S. Clinical Features and Treatment of Penile Schwannoma: A Systematic Review. Clinical Genitourinary Cancer (2016);14(3), 198–202. doi:10.1016/j.clgc.2015.12.018. 24. Chipollini J, Chaing S, Azizi M, Kidd L, Kim P, Spiess P. Advances in Understanding of Penile Carcinogenesis, The Search for Actionable Targets. Int J Mol Sci. (2017). 18(8):1777. 25. Johnstone P, Spiess P, Giuliano A. New directions in penile cancer. Lancet Oncol (2018). Vol 20:16–17. 26. Eid R, Nemr E, Haddad FG, Kourie HR, Kattan J. Do checkpoint inhibitors provide new hope for management of metastatic penile carcinoma? Future Oncology (2018), 14(8), 677–680. doi:10.2217/fon-2017-0559.

Scholarly Research In Progress • Vol. 3, November 2019

Profile of Dispensary Patients that Substitute Cannabis for Alcohol Assad Hayat1† and Brian J. Piper1,2

Abstract Background: A substitution effect occurs when patients substitute cannabis for another drug. Prior research determined that more than three-quarters (76.7%) of New England dispensary members reported reducing their use of opioids and two-fifths (42.0%) decreased their use of alcohol after starting medical cannabis (MC). The objective of this exploratory study was to identify any factors which differentiate alcohol substituters from those that do not modify their alcohol use after starting MC (non-substituters). Methods: Among dispensary patients (N=1,477) who completed an online survey, more than two-thirds with chronic pain, 7.4% indicated that they regularly consumed alcohol. Comparisons were made to identify any demographic, dosing, or health history characteristics which differentiated alcohol substituters (N=47) from non-substituters (N=65). Respondents selected from a list of 37 diseases/health conditions (e.g. diabetes, sleep disorders) and the total number was determined. Results: Substituters and non-substituters were indistinguishable in terms of sex, age, or prior drug history. A greater percentage of the substituters (59.6%) than nonsubstituters (40.6%, p<0.05) reported using MC two or more times per day. Substituters were more likely to be employed (68.1%) than non-substituters (51.1%, p<0.05). Substituters also reported having significantly more health conditions and diseases (3.3 + 2.0) than non-substituters (2.4 + 1.4, p<0.05). Conclusions: This report offers some insights into the profile of patients whose self-reported alcohol intake decreased following initiation of MC. Alcohol substituters had more other health conditions but also were more likely to be employed. Additional prospective or controlled research into the alcohol substitution effect following MC with a sample with more advanced alcohol misuse may be warranted.

Introduction There has been a substantial amount of clinical and epidemiological research of the potential therapeutic benefits of medical cannabis (MC), particularly in the past two decades (1). MC is currently approved by 32 states and also in Washington, DC; Guam; and Puerto Rico. Further, 19 states condone low tetrahydrocannabinol (THC)/high cannabidiol (CBD) products (2). Ten states and Washington, DC, have legalized cannabis for recreational use. The regulations for the use of cannabis products are state-specific (3). States have approved the use of MC for a variety of diseases and disorders, including pain, spasticity associated with multiple sclerosis, nausea, post-traumatic stress disorder 46

(PTSD), cancer, epilepsy, cachexia, glaucoma, HIV/AIDS, and degenerative neurological diseases (4). Seizures and epilepsy, HIV/AIDS, and chronic pain are the most common approved conditions (3, 5). The 21 certifying conditions in Pennsylvania for MC include intractable pain, cancer, epilepsy, glaucoma, neurodegenerative disorders, autism, inflammatory bowel diseases (IBD), multiple sclerosis, autism, and opioid dependency. The United States is in the midst of an opioid overdose epidemic involving both commercial pharmaceutical (e.g. oxycodone, hydrocodone) and illicitly manufactured (e.g. heroin, fentanyl) drugs (6–9). As a result, there is tremendous interest in whether MC users decrease their use of opioid medications. Three lines of evidence are congruent with a MC-induced opioid substitution effect. First, there have been several well-powered (N>1,000) retrospective surveys from California (5), New England (10), and across North America (11) which have identified a substitution effect. However, although results from these samples collectively indicate that this opioid substitution effect is a robust phenomenon, self-reported data from voluntary (i.e., self-selected) participants regarding this controversial drug should be viewed with great caution. Second, U.S. states with MC had a 25% lower opioid overdose mortality rate relative to states without (12). Third, states with MC have fewer opioid expenditures for Medicare (13) and Medicaid (14). Although the opioid substitution effect is moderately wellestablished (although see (15)), less is known about whether MC patients decrease their use of alcohol (16). Excessive alcohol consumption exerts a significant burden on health care systems (17). Two large (N>35,000 each) nationally representative surveys found that more than 1 out of every 8 (12.7%) adults met criteria for an alcohol use disorder (AUD) (18). Approximately 65,000 people (>80% male) in the U.S. die from alcohol-related causes each year (19). Currently available interventions for AUDs have modest efficacies, and there are limited tools available to identify the subset of patients that may benefit. A Cochrane meta-analysis of Alcoholics Anonymous (AA) and other 12-step programs revealed that the available evidence did not demonstrate effectiveness of AA to achieve abstinence, reduce alcohol use, or improve participation in treatment (20). Further, the AA philosophy is very concerned about recovering alcoholics becoming addicted to prescription drugs or switching to marijuana (21). Although there are Food and Drug Administration approved pharmacotherapies for AUD, there is a need for agents that are both well-tolerated and efficacious. Randomized controlled trials with blind designs showed no efficacy of disulfiram relative to controls (22). The number needed to treat (NNT)

Profile of Dispensary Patients that Substitute Cannabis for Alcohol

with acamprosate such that one person would show decreased likelihood of returning to drinking was 9 (23) or 12 (24). The NNT with the opioid antagonist naltrexone was 20 (24). There is some evidence with different research designs that has evaluated whether cannabinoids may have a therapeutic role for an AUD. Pyrahexyl, a synthetic homologue of THC, was administered to 70 alcoholics. The vast majority (84%) were classified as showing an alleviation of symptoms including calmness and decreased irritability during alcohol withdrawal (25). Ninety-two patients, two-thirds with alcoholism or cirrhosis of the liver, in a naturalistic study self-reported at the one-year follow-up that they used cannabis to decrease their alcohol. Cessation of cannabis was also associated with a return to drinking (26). The rationale for MC substitution for other legal and illegal drugs was fewer side effects and improved symptom management (27, 28). Among New England dispensary patients that also regularly consumed alcohol, twofifths indicated that they reduced their alcohol consumption (10). In addition to clinical research (16), there is also a preclinical evidence base showing that manipulating the endocannabinoid system impacts ethanol consumption (29, 30). The National Academy of Sciences, Engineering and Medicine’s 2017 report concluded that there is no evidence to support, or refute, that MC is an effective treatment for achieving abstinence from addictive substances (1). Therefore, the objective of this exploratory study was to further characterize this substitution effect and identify any factors which differentiated alcohol substituters from non-substituters.

Materials and Methods Participants Participants of this study were members of MC dispensaries located in New England, including the Champlain Valley Dispensary, Southern Vermont Wellness, and the Wellness Connection of Maine. Procedures Participants received an email from their dispensary inviting them to complete surveys hosted on SurveyMonkey followed by up to two weekly reminders. There were three slightly tailored survey versions (Vermont, Maine, Rhode Island) for a subset of items, as different states have different MC laws. Responses were exported to spreadsheets and consolidated into a single file. The survey is available elsewhere (10) but used skip logic to tailor items based on prior responses. An affirmative response to “Do you regularly consume alcohol?” would result in additional items including “Have you noticed a change in your alcohol consumption since you started medical cannabis?” with options of “Yes, I need to drink a lot more,” “Yes, I need to drink slightly more,” “No change,” “Yes, I need to drink slightly less,” or “Yes, I need to drink a lot less.” Among all participants (N=1,477), only those that reported regular alcohol use (N=112, 50.4% male) are the topic of this investigation. Among this subset, the “No change” (N=65) were classified as non-substituters while the “a lot less” (N=28) and “slightly less” (N=19) were collapsed into a substituter group (N=47, note that no respondents reported an increase in alcohol consumption). The survey included a required informed consent, demographic information, MC use

patterns, and items about history of use of recreational drugs. Respondents that identified as employed full-time, part-time, or as students were classified as employed. The total number selections in response to “Which of the following diseases or conditions have you been diagnosed by a medical professional (physician, PA, psychologist) as having?” with 37 options (e.g., diabetes, high blood pressure, chronic pain, sleep disorder) was calculated as an index of overall health. Binge drinking was queried by asking males “How many days per week do you consume five or more alcoholic drinks in a single setting?” with options from 0 to 7. Females received an identical item except with binge drinking defined as four (16). Prior cannabis use was determined by asking “Have you ever used cannabis before your current medical use?” Addiction history was determined with “Have you ever suffered from alcoholism and/ or drug addiction?” with options of no, yes: alcoholism, yes: drug addiction, or both. This study, including the informed consent, was approved by the IRB of Bowdoin College and data collection occurred between August 2015 and April 2016. Dosing was quantified by “How frequently do you use medical cannabis?” with options of more or less than two times per day. Further information on the procedure was previously reported (10, 32, 33). Statistical analysis Differences between substituters and non-substituters were assessed using either an independent samples t-test or chi– square with a p<0.05 considered significant. Variability was reported as the standard deviation in the text and the standard error of the mean in figures. Figures were prepared using GraphPad Prism, v. 6.07 (La Jolla, CA).

Results More than three-fifths of respondents were from Vermont (62.5%), followed by Rhode Island (26.8%) and Maine (5.4%). Table 1 shows that substituters and non-substituters were not significantly different in terms of age, sex, and ethnicity. One-third (33.3%) of respondents reported one or more days per week of binge drinking. Binge drinking five or more days per week was reported by 2.1% of substituters versus 7.8% of non-substituters. Responses to “How would you describe your use of cannabis?” on a continuum, with 10% increments, from 100% recreational/0% medical to 0% recreational/100% medical, were on the medical end (<50% recreational) of the spectrum. The age of first smoking, drinking alcohol, and using marijuana recreationally were not statistically different between substituters and non-substituters. The majority of both substituters (91.5%) and non-substituters (93.4%) denied that they had suffered from alcoholism or drug addiction. However, more of the substituters (38.6%) than non-substituters (19.0%, χ2(1)=5.03, p<0.05) response to “Had you ever used cannabis before your current medical use?” was “Yes, self-prescribed for a medical condition.” Further, a greater percentage of the substituters (59.6%) than non-substituters (40.6%, p<0.05) reported using medical cannabis two or more times per day (Figure 1A). Ratings of “How effective is medical cannabis in treating your symptoms or condition(s)?” with options of 0% (no relief) to 100% (complete relief) were high, but did not differ based on alcohol substitution.

Profile of Dispensary Patients that Substitute Cannabis for Alcohol

alcohol (27). The current sample was composed primarily of chronic pain patients and not individuals who were especially motivated to make a change in their alcohol intake. It is an important question whether these results would generalize to others. Unfortunately, as existing medications are ineffective for the preponderance of alcoholics (23, 24), there is a tremendous need for novel pharmacotherapies, even if they only benefit a subset of patients, which can be employed with evidencebased psychosocial treatments (34).

Table 1. Comparison of patients that did (substituters) or did not (non-substituters) decrease their use of alcohol after starting medical cannabis (defined in results section)

Figure 1B shows that substituters (59.6%) were more likely to be employed (non-substituters=40.6%, p<0.05). The most common disorders and diseases were chronic pain (65.2%), arthritis (43.8%), anxiety (20.5%), hypertension (19.6%), cancer (16.1%), migraines (15.2%), depression (14.3%), a sleep disorder (13.4%), asthma (12.5%), and IBS (11.6%). Substituters had 25.4% more health conditions and diseases than non-substituters (Figure 1C & 2).

Discussion This study documents the self-reported efficacy of MC as a substitute for alcohol. As noted previously (10), more than two-fifths of dispensary members who used alcohol reported a reduction in alcohol use after starting MC. This is consistent with previous studies that documented a similar, or greater, reduction in alcohol (27,28). In a sample from California where the average age was a decade younger and half currently drank alcohol, 40 percent used cannabis as a substitute for

There are at least two mechanisms by which MC could substitute for alcohol: pharmacodynamic and pharmacoeconomic. First, rats that self-administered alcohol increased the endocannabinoid 2-arachidonoylglycerol in a brain area important for reward, the nucleus accumbens (35), indicating that alcohol and THC have some shared neural substrates. Although efforts to employ the cannabinoid (CB1) inverse agonist rimonabant to decrease alcohol consumption were unsuccessful (36, 37), these findings should be interpreted with caution as rimonabant has a nonselective mechanism (38) and may cause dysphoria. Second, MC is not covered by insurance in the U.S. and is expensive (>3,000 $USD/patient/year) (32). However, we view a pharmacoeconomic explanation (i.e., participants spend money that previously was used for alcohol on MC) as unlikely because alcohol substituters were significantly more likely to be employed than non-substituters. Studies with rodents have generally found that administration of cannabinoid receptor antagonists decreases alcohol consumption while cannabinoid agonists cause the opposite effect (29). The limited preclinical findings are opposite the general pattern observed in humans (25â&#x20AC;&#x201C;28, 32, 39, 40), although see (41). However, CBD administration decreases alcohol consumption in mice (30) and blocks the neurotoxic effects of an alcohol binge (42, 43). Therefore, it is important to emphasize that MC is not a single homogenous product. There were 1,987 â&#x20AC;&#x153;strainsâ&#x20AC;? listed in a cannabis database and the ones available for medical use were dispensary specific (10). MC includes both THC, a CB1 and CB2 partial agonist and CBD. The complex pharmacology of CBD includes acting as an indirect CB1 and CB2 antagonist (44), which could counteract the effects of THC. The mechanism of THC is also complex including acting as a serotonin (5-HT1A) partial agonist (45), a G-protein coupled receptor 55 antagonist, stimulating

Figure 1. Frequency of dosing A), employment B), and total number (+ SEM) of health conditions and diseases C) among patients that did or did not reduce their alcohol consumption after starting medical cannabis. * p<0.05

Profile of Dispensary Patients that Substitute Cannabis for Alcohol

more rigorous designs will support the view that addiction is a biopsychosocial disorder where pharmacological interventions that target the endocannabinoid neurotransmitter system (29) are a component of treatment option.

Figure 2. Percent of groups that did (substituters) or did not (nonsubstituters) reduce their alcohol consumption after starting cannabis for medical purposes with diseases or psychiatric disorders including chronic pain, an anxiety disorder (generalized or social), high-blood pressure, cancer, migraine or chronic headaches, major depressive disorder, sleep disorder (insomnia), or irritable bowel syndrome (IBS).

neurogenesis in the hippocampus, and inhibiting the adenosine transporter (46, 47). Perhaps some of the individual differences in MC-associated outcomes may be due to the tremendous variability in products. It is an open question whether the findings from investigations with recreational marijuana use and alcohol, although subtle and contradictory (39, 48), generalize to MC because recreational and medical users may differ on demographic and motivational factors. Further, the products used by MC patients (33) may show only partial overlap with those used for recreational purposes and very limited overlap with preclinical research chemicals. Substituters reported that they have been diagnosed with more diseases and disorders than non-substituters. Given the higher rates of employment among substituters, it is possible that these fit a profile of social drinkers rather than a more advanced alcohol misuse pattern. However, further information about alcohol use patterns (49) would be necessary to support this interpretation, which at this point is only speculation. It is possible that the greater number of conditions, both psychiatric and non-psychiatric, might contribute to a heightened motivation to make lifestyle modifications to improve their quality of life. Despite controversial findings regarding their limited effectiveness (20), AA and other 12-step programs are inexpensive to operate and ubiquitous in the U.S. AA is generally not supportive of pharmacotherapies being employed to treat alcoholism (21). This perspective may stigmatize patients that choose these therapies, form barriers to the implementation of evidence-based medicine (50–51) and may function as an impediment to the successful treatment of addiction. Although the literature for MC substitution effects (11–14, 26–28, 32, 40, 52) infrequently includes prospective investigations (53) or randomized controlled trials, we are cautiously optimistic that future research with

There are some limitations to this retrospective study. First, although the response rate to the survey was generally good (Maine=47%, Vermont=41%, (32), the number of respondents that reported regular alcohol use was on the low end relative to what might be anticipated based on national data (18) which could reflect that dispensary members are notrepresentative of the general population. Second, recall of consumption of alcohol or other drugs that acutely impair memory is recognized as imperfect and future research could be conducted in a laboratory environment (55) or employ more objective measures (e.g., medical or dispensary records). Although more time consuming for participants, additional information about alcohol or cannabis use should be obtained using empirically validated procedures (56) in future efforts. Third, an assumption of this retrospective study is that cannabis use increased after becoming a dispensary member. However, further prospective research would be necessary to quantify any recreational cannabis use that preceded MC. Although the term “substitution” is in general use to describe a MC/opioid effect (17), this label may be non-optimal as what was studied was a reduction in alcohol after starting MC. Other research is needed to determine if MC potentiates the effects of alcohol and less may be needed because of synergistic psychoactive effects. Fourth, the studied dispensaries do not uniformly obtain information about THC or CBD levels in the many strains and products they offer. Fifth, although this was an exploratory study exploring a novel topic, the variables identified were not hypothesized a priori, no corrections were made for multiple statistical comparisons, and findings should be interpreted with caution. Finally, unlike pharmacotherapies that have completed standard approval processes (e.g., by the Food and Drug Administration), rigorously obtained efficacy data about the most optimal dose, dose frequency, or delivery for MC is sparse.

Conclusion This retrospective study identified some demographic and health variables that differentiated persons who substituted MC for alcohol from those who did not. Given the need for more efficacious AUD pharmacotherapies, additional prospective or controlled research of the alcohol substitution effect following MC with a sample with more advanced alcohol misuse patterns may be warranted. More broadly, there is currently little correspondence between the recommendations of the National Academies of Science, Engineering, and Medicine regarding the conditions where there is evidence supportive of MC and the conditions it is certified for. This report indicated that evidence was substantial or conclusive for cannabis or cannabinoids’ efficacy for chronic pain, as antiemetics for chemotherapy-induced nausea and vomiting and multiple sclerosis spasticity. However, there was limited evidence to conclude that cannabis was ineffective for dementia and glaucoma (1) and more research is urgently needed. Further investigations to develop safer and more efficacious interventions for AUD and opioid use disorder are ongoing.

Profile of Dispensary Patients that Substitute Cannabis for Alcohol

Acknowledgments Rebecca M. DeKeuster, MEd, and Alexander T. Abess, MD, aided in survey construction. Olapeju M. Simoyan, MD, and Stephanie D. Nichols, PharmD, provided feedback on an earlier version of this manuscript. Technical support was provided by Iris Johnston. This research was supported by the Center for Wellness Leadership.

Disclosures

for pharmaceutical agents for pain, anxiety, and sleep. J Psychopharm 2017; 31: 569–575. 11. Corroon JM, Mischley LK, Sexton M. Cannabis as a substitute for prescription drugs–a cross-sectional study. J Pain Research 2017; 10:989–998. 12. Bachhuber MA, Saloner B, Cunningham CO, Barry CL. Medical cannabis laws and opioid analgesic overdose mortality in the United States, 1999–2010. JAMA Intern Med 2014; 174: 1668–73.

AH has no conflicts of interest. In the past four years, BJP has received research supplies from the National Institute of Drug Abuse (NIDA); travel from the NIDA, the Wellness Connection of Maine, Patients out of Time, and the Hereditary Neuropathy Foundation. He is on the advisory board (pro bono) for the Center for Wellness Leadership.

13. Bradford AC, Bradford WD. Medical marijuana laws reduce prescription medication use in Medicare Part D. Health Affairs 2016; 35:1230–1236.

References

15. Caputi TL. Whether medical marijuana is ever substituted for other substances is not the full story. Drug Alc Rev 2017; 36:E3–E4.

National Academies of Sciences, Engineering, and Medicine. The health effects of cannabis and cannabinoids: The current state of evidence and recommendations for research. Washington, DC: The National Academies Press; 2017.

2. National Conference of State Legislatures. (2018). State Medical Marijuana Laws. November 8. Retrieved December 16, 2018, from http://www.ncsl.org/research/ health/state-medical-marijuana-laws.aspx 3. Karst A. Weighing the benefits and risks of medical marijuana use: A brief review. Pharmacy 2018; 6:128. doi:10.3390/pharmacy6040128 PMID: 24012796 4. National Organization for the Reform of Marijuana Laws (2018). Accessed December 16, 2018 at: http://norml.org/ legal/item/maine-medical-marijuana 5. Reiman A, Welty M, Solomon P. Cannabis as a substitute for opioid-based pain medication: Patient self-report. Cannabis Cannabinoid Research 2017; 2:160–6. 6. Piper BJ, Desrosiers CE, Lipovsky JW, Rodney MA, Baker RP, McCall KL, et al. Use and misuse of opioids in Maine: Results from pharmacists, the Prescription Monitoring, and the Diversion Alert Programs. J Stud Alcohol Drugs 2016; 77:556–65. 7.

Piper BJ, Shaw DT, Simoyan OM, McCall KL, Nichols SD. Trends in medical use of opioids in the United States: 2006–2016. Am J Prevent Med 2018; 54: 652–660.

8. Simpson KJ, Moran MT, Foster ML, Shah DT, Chung DY, Nichols SD, et al. Trends in pharmaceutical and nonpharmaceutical arrests, use, and overdoses in Maine. BMJ Open 2019; 9:e027117. 9. Wright ER, Kooreman HE, Greene MS, Chambers RA, Banerjee A, Wilson J. The iatrogenic epidemic of prescription drug abuse: County-level determinants of opioid availability and abuse. Drug Alcohol Depend 2014; 138:209–215. 10. Piper BJ, DeKeuster RM, Beals ML, Cobb CM, Burchman CA, Perkinson L, et al. Substitution of medical cannabis

14. Bradford AC, Bradford WD. Medical marijuana laws may be associated with a decline in the number of prescriptions for Medicaid enrollees. Health Affairs 2017; 36:1–7.

16. Mehta AJ. Alcoholism and critical illness: A review. World J Crit Care Med 2016; 5:27–35. 17. Grant BF, Chou SP, Saha TD, et al. Prevalence of 12-month alcohol use, high-risk drinking, and DSM-IV Alcohol Use Disorder in the United States, 2001–2002 to 2012–2013: Results from the National Epidemiologic Survey on Alcohol and Related Conditions. JAMA Psychiatry 2017; 74:911– 923. 18. Rehm J, Dawson D, Frick U, Gmel G, Roerecke M, Shield KD, et al. Burden of disease associated with Alcohol Use Disorders in the United States. Alcohol Clin Exp Res 2014; 38:1068–1077. 19. Ferri M, Amato L, Davoli M. Alcoholics Anonymous and other 12-step programmes for alcohol dependence. Cochrane Database Syst Rev 2006; 3:CD005032. 20. Alcoholics Anonymous. 2011. The AA Member-Medications & other drugs. Alcoholics Anonymous World Services, Inc. 21. Skinner MD, Lahmek P, Pham H, et al. Disulfiram efficacy in the treatment of alcohol dependence: A meta-analysis. PLoS One 2014; 9:e87366. 22. Rösner S, Hackl-Herrwerth A, Leucht S, Lehert P, Vecchi S, Soyka M. Acamprosate for alcohol dependence. Cochrane Database Syst Rev 2010: CD004332 23. Jonas DE, Amick HR, Feltner C, Bobashev G, Thomas K, Wines R, et al. Pharmacotherapy for adults with alcohol use disorder in outpatient settings: A systematic review and meta-analysis. JAMA 2014; 311:1889–1900. 24. Thompson LJ, Proctor RC. The use of pyrahexyl in the treatment of alcoholics and drug withdrawal conditions. North Carolina Medical Journal 1953; 14:520–3 25. Mikuriya T: Cannabis as a substitute for alcohol: A harm reduction approach. Journal of Cannabis Therapeutics 2008, 4:79–93 https://doi.org/10.1300/J175v04n01_04 26. Reiman A. Cannabis as a substitute for alcohol and other drugs. Harm Reduction Journal 2009; 6:35 d

Profile of Dispensary Patients that Substitute Cannabis for Alcohol

27. Lucas P, Walsh Z, Crosby K, Callaway R, Belle-Isle L, Kay R, et al. Substituting cannabis for prescription drugs, alcohol and other substances among medical cannabis patients: The impact of contextual factors. Drug Alcohol Rev 2016; 35:326–33. 28. Sloan ME, Gowin JL, Ramchandani VA, Hurd YL, Le Foll B. 2017. The endocannabinoid system as a target for addiction treatment: Trials and tribulations. Neuropharmacology 124:73–83. 29. Viudez-Martínez A, García-Gutiérrez MS, Navarrón CM, Morales-Calero MI, Navarrete F, Torres-Suárez AI, et al. Cannabidiol reduces ethanol consumption, motivation and relapse in mice. Addict Biol 2018; 23:154–164. 30. Center for Behavioral Health Statistics and Quality. (2016). Key substance use and mental health indicators in the United States: Results from the 2015 National Survey on Drug Use and Health (HHS Publication No. SMA 16-4984, NSDUH Series H-51). Retrieved from http://www.samhsa. gov/data/ Accessed 12/15/2018. 31. Piper BJ, Beals ML, Abess AT, Nichols SD, Martin MW, Cobb CM, et al. Chronic pain patients’ perspectives of medical cannabis. Pain 2017; 158: 1373–79. 32. Piper BJ. Mother of Berries, ACDC, or Chocolope?: Examination of strains used by medical cannabis patients in New England. J Psychoactive Drugs 2018; 50: 95–104.

in reversing binge ethanol-induced neurotoxicity. J Pharmacol Exp Ther 2005; 314:780–8. 42. Liput DJ, Hammell DC, Stinchcomb AL, Nixon K. Transdermal delivery of cannabidiol attenuates bingealcohol-induced neurodegeneration in a rodent model of an Alcohol Use Disorder. Pharmacol Biochem Behav 2013; 111:120–127. 43. Mechoulam R, Peters M, Murillo-Rodriguez E, Hanus LO. Cannabidiol—recent advances. Chemistry & Biodiversity 2007; 4:1678–92. 44. Russo EB, Burnett A, Hall B, Parker KK. Agonistic properties of cannabidiol at 5-HT1A receptors. Neurochem Res 2005; 30:1037–43. 45. Campos AC, Moreira FA, Gomes FV, Del Bel EA, Guimarães FS. Multiple mechanisms involved in the large-spectrum therapeutic potential of cannabidiol in psychiatric disorders. Philos Trans R Soc B Biol Sci 2012; 367:3364– 3378. 46. Gaston TE, Friedman D. Pharmacology of cannabinoids in the treatment of epilepsy. Epilepsy & Behav 2017; 70:313– 8. 47. Subbaraman MS, Metrik J, Patterson D, Swift R. Cannabis use during treatment for alcohol use disorders predicts alcohol treatment outcomes. Addiction 2017; 112:685–694.

33. Martin GW, Rehm J. The effectiveness of psychosocial modalities in the treatment of alcohol problems in adults: a review of the evidence. Can J Psychiatry 2012; 57:350–8.

48. Rueger SY, Trela CJ, Palmeri M, King AC. Self-administered web-based Timeline Followback procedure for drinking and smoking behaviors in young adults. J Stud Alcohol Drugs 2012; 73:829–833.

34. Caille S, Alvarez-Jaimes L, Polis I, Stouffer DG, Parsons LH. Specific alterations of extracellular endocannabinoid levels in the nucleus accumbens by ethanol, heroin, and cocaine self-administration. J Neuroscience 2007; 27:3695–702.

49. Mattick RP, Breen C, Kimber J, et al. Methadone maintenance therapy versus no opioid replacement therapy for opioid dependence. Cochrane Database Syst Rev 2009; 8:CD002209

35. George DT, Herion DW, Jones CL, Phillips MJ, Hersh J, Hill D, et al. Rimonabant (SR141716) has no effect on alcohol self-administration or endocrine measures in nontreatmentseeking heavy alcohol drinkers. Psychopharmacol 2010; 208:37e44.

50. Nielson S, Larance B, Degenhardt L, Gowing L, Kehler C, Lintzeris N. Opioid agonist treatment for pharmaceutical opioid dependent people. Cochrane Database Syst Rev 2016; 9:CD011117

36. Soyka M, Koller G, Schmidt P, Lesch OM, Leweke M, Fehr C, et al. Cannabinoid receptor 1 blocker rimonabant (SR 141716) for treatment of alcohol dependence. J Clin Psychopharmacol 2008; 28:317–24. 37. Seely KA, Brents LK, Franks LN, Rajasekaran M, Zimmerman SM, Fantegrossi WE, et al. AM-251 and rimonabant act as direct antagonists at mu-opioid receptors: implications for opioid/cannabinoid interaction studies. Neuropharmacology 2012; 63:905e915. 38. Metrik J, Spillane N, Leventhal AM, Kahler CW. Marijuana use and tobacco smoking cessation among heavy alcohol drinkers. Drug Alcohol Depend 119:194–200. 39. Mikuriya TH. Cannabis as a substitute for alcohol: A harmreduction approach. J Cannabis Ther 2004; 4:79–93. 40. Subbaraman MS. Can cannabis be considered a substitute medication for alcohol? Alcohol Alcohol 2014; 49:292–8. 41. Hamelink C, Hampson A, Wink DA, Eiden LE, Eskay RL. Comparison of cannabidiol, antioxidants, and diuretics

51. Boehnke KF, Litinas E, Clauw DJ. Medical cannabis use is associated with decreased opiate medication use in a retrospective cross-sectional survey of patients with chronic pain. J Pain 2016; 17:739–744. 52. Wardell JD, Shuper PA, Hendershot CS. A longitudinal investigation of the association between cannabis use and alcohol use among people living with HIV. Drug Alcohol Depend 2018; 193:7–13. 53. Sloan ME, Gowin JL, Ramchandani VA, Hurd YL, Le Foll B. The endocannabinoid system as a target for addiction treatment: Trials and tribulations. Neuropharmacology 2017; 124:73–83. 54. Penetar DM, Toto LH, Lee DY, Lukas SE. A single dose of kudzu extract reduces alcohol consumption in a binge drinking paradigm. Drug Alcohol Depend 2015; 153:194– 200. 55. Robinson SM, Sobell LC, Sobell MB, Leo GI. Reliability of the Timeline Followback for cocaine, cannabis, and cigarette use. Psych Addict Behav 2014; 28:154–162.

Scholarly Research In Progress • Vol. 3, November 2019

Proliferative Retinopathy Associated with a Case of Hemoglobin C Trait Karl M. Andersen1† and Randall R. Peairs2

Geisinger Commonwealth School of Medicine, Scranton, PA 18509 Northeastern Eye Institute, Scranton, PA 18510 †Doctor of Medicine Program Correspondence: kandersen@som.geisinger.edu 1

Abstract Background: We present a rare case of bilateral proliferative retinopathy that developed in an African-American male secondary to the heterozygous condition, hemoglobin AC. Methods: Clinical findings in the eye were obtained by slit lamp and indirect ophthalmoscopy, fundus photography and fluorescein angiography. Hemoglobinopathy was diagnosed by electrophoresis. Results: A 46-year-old man undergoing active treatment for iridocyclitis presented with visual disturbance related to unilateral vitreous hemorrhage. Proliferative retinopathy with neovascularization was diagnosed in the eye without hemorrhage and later noted in the other eye as the hemorrhage cleared. Hemoglobin electrophoresis showed hemoglobin AC. To help prevent progression of proliferative retinopathy, panretinal photocoagulation was applied to areas of non-perfusion and medical treatment of hypertension was performed. Conclusion: Although hemoglobin AC is a rare, usually asymptomatic genetic disorder, it should be considered in the differential diagnosis of unexplained proliferative neovascular retinopathy.

Introduction Hemoglobin C trait is a condition in which individuals are heterozygous for the hemoglobin C mutation, with the second allele being wild type (hemoglobin A). This carrier condition is often simply referred to as hemoglobin AC. The hemoglobin C mutation involves replacement of the normal glutamic acid residue at position six of the β-globin subunit with lysine, changing the structure and function of the hemoglobin molecule and, as a result, the red blood cell (1). However, the presence of this mutation with a mutation-free second allele most often yields no clinical manifestations. The frequency of hemoglobin AC is estimated to be 2.4% in the U.S. African-American population (2), and hemoglobin C alleles are most concentrated in West Africa (3). Although sickle cell-hemoglobin C (SC) proliferative retinopathy has been well established in the literature, proliferative (neovascular) retinopathy is extremely rare in hemoglobin AC individuals, with a very limited number of cases having been reported (4–8).

Case Presentation A non-diabetic male of African-American descent presented at the age of 46, complaining of blurry vision. Visual acuity was found to be count fingers OD (oculus dextrus; right eye) and 20/40 OS (oculus sinister; left eye). Slit lamp exam of the right 52

eye showed rare cells in the anterior chamber and fundus exam revealed dense vitreous hemorrhage without overt retinal detachment. B-scan identified vitreous hemorrhage with an attached retina. The left eye demonstrated a deep and quiet anterior chamber, and the retina exam showed old sheathed vessels (“silver wires”), lattice degeneration and a tiny dot hemorrhage superiorly. Fundus photography also showed these sheathed occluded retinal vessels (Figure 1). At the time of initial evaluation, the patient’s blood pressure was slightly elevated at 168/84. He was being actively treated for iridocyclitis OD with topical prednisolone acetate. The patient returned one week after initial presentation, at which time fluorescein angiography (FA) was performed showing early proliferative retinopathy with peripheral neovascularization and severe nonperfusion OS (Figure 2). FA imaging was limited on the right due to persistent vitreous hemorrhage (Figure 3). Diagnostic testing revealed HLA-B27 positivity. Highperformance liquid chromatography followed by hemoglobin electrophoresis showed no evidence of hemoglobin S, but did reveal the presence of AC hemoglobinopathy. Hemoglobin A composed 61.9% of the total hemoglobin, hemoglobin C composed 35.0% and hemoglobin A2 composed 3.1%. Testing also yielded a positive QuantiFERON-TB Gold test, and on subsequent questioning the patient recalled being treated for tuberculosis as a child. His chest X-ray showed no evidence of disease. Additionally, the patient reported a history of smoking, but denied any current use of cigarettes. Panretinal photocoagulation (PRP) was applied to areas of ischemia adjacent to peripheral neovascularization in the left eye. As the patient’s vitreous hemorrhage spontaneously cleared in the right eye, areas of non-perfusion and neovascularization were noted on exam (without repeating FA imaging), and PRP was subsequently performed. Intravitreal injections of anti-VEGF (vascular endothelial growth factor) were deferred, given a concern of inducing traction in areas of neovascularization in the right eye. At a follow-up visit, three months following the onset of initial symptoms, the patient had developed a spontaneous horseshoe tear of the retina without detachment OD, in an area of neovascularization. It was prophylactically repaired by laser photocoagulation. Referrals to a primary care physician and to the hematology service were initiated, and medical treatment for hypertension was initiated with lisinopril (5 mg once per day). Hematology recommended observation based on the absence of other manifestations of hemoglobin AC. Five months after initial presentation, the patient developed a new spontaneous vitreous hemorrhage without retinal detachment in the left eye. He deferred vitrectomy and the hemorrhage spontaneously improved over the next three months.

Proliferative Retinopathy Associated with a Case of Hemoglobin C Trait

Discussion Based on Dr. Morton Goldbergâ&#x20AC;&#x2122;s 1971 classification system for SC proliferative retinopathy (9), our patient demonstrated a feature of stage IV disease OD (vitreous hemorrhage) and of stage III disease OS (neovascular proliferation) at initial presentation. Although this classification was specifically created to characterize the progression of SC proliferative retinopathy, its application to our case of hemoglobin AC retinopathy illustrates that carriers of the mutated hemoglobin C gene can develop the severe retinal disturbances observed in the spectrum of proliferative retinal disease.

Figure 1. Fundus photograph of the left eye showing sheathed occluded retinal vessels (â&#x20AC;&#x153;silver wiresâ&#x20AC;?).

Figure 2. Fluorescein angiography images of the left eye showing severe nonperfusion and evidence of proliferative retinopathy in the peripheral retina, characterized by multiple microaneurysms and A-V shunts (right image), as well as small areas of neovascular changes (left arrow).

Figure 3. Fluorescein angiography image of the right eye showing a vitreous hemorrhage which obscured proper visualization of the fundus.

Descriptions of the hematologic diseases that have ocular manifestations have revealed that in certain contexts the hemoglobin C mutation exerts an effect that can potentiate significant harm in the highly vascular retina. This is especially apparent in the synergistic negative effects that coexisting hemoglobin C and hemoglobin S alleles have on the retina. Ocular disease manifestations are more common in SC disease than in sickle cell disease (SS) itself; however, systemic manifestations are more common in the latter than the former (8). The proliferative retinopathy of SC is characterized by capillary and arteriolar closure, capillary network tortuosity, microaneurysms, A-V anastomoses (see Figure 2) and neovascularization, all of which contribute to frequent progression to retinal and vitreous hemorrhage and retinal detachment (10). The proliferative retinopathy of hemoglobin AC is likely brought about by similar mechanisms. Individuals with this hemoglobinopathy have been shown to have decreased erythrocyte plasticity and increased blood viscosity (8), factors that may fundamentally contribute to the primary event of vascular closure and the culminating event of vitreous hemorrhage, observed as the presenting symptom of hemoglobin AC retinopathy in our case. Hemoglobin AC proliferative retinopathy has previously been reported in a patient with a history of active tuberculosis (TB) infection (7). Interestingly, our patient was found to test positive for TB exposure by the QuantiFERON-TB Gold test, and questioning revealed that he had been treated for active TB as a child. TB can involve the eye in many different ways (11). Eales disease is an occlusive perivasculitis limited to the retina (12) that is associated in many cases with evidence of TB and that can progress to retinal ischemia and angiogenesis mediated by inflammation (13). These observations suggest that ocular inflammation related to TB may contribute to retinal changes that culminate in vitreous hemorrhage in some patients. HLA-B27 positive iridocyclitis, although usually confined to the anterior segment of the eye, can progress to severe inflammation in the posterior segment, leading to vitreous detachment and hemorrhage (14). Smoking clearly 53

Proliferative Retinopathy Associated with a Case of Hemoglobin C Trait

causes vascular inflammation (15). Additionally, there is a strong statistically significant correlation between uveitis (especially infectious uveitis) and smoking (past or current); however, this correlation, interestingly, does not hold for the African-American subgroup (16). Perhaps the additive effect of hemoglobin AC, a history of smoking and TB, and HLA-B27positive iridocyclitis in our patient combined to bring about vascular closure and proliferative changes in the retina leading to vitreous hemorrhage. Additive vascular insults mediated in our patient by abnormal hemoglobin and inflammation from multiple sources may have led to pathologic vascular changes and symptoms in a condition that would otherwise have been asymptomatic. In conclusion, the retinopathy of our patient demonstrates a rare manifestation of hemoglobin AC, which is almost always a benign, asymptomatic carrier condition. In accord with previous reports of hemoglobin AC retinopathy, we propose that our patient’s proliferative neovascular retinal disease is most likely attributable to the rheologic and hematologic aberrancies observed in the context of a single mutated hemoglobin C allele. However, because very few individuals with hemoglobin AC ever develop proliferative retinopathy, we also propose a possible multifactorial etiology that considers both local and systemic inflammatory abnormalities.

Acknowledgments We thank Morton Goldberg, MD, for reviewing the manuscript and for his expert opinion.

Disclosures The authors declare no conflicts of interest.

References 1.

Naeim F, Rao PN, Song SX, Grody WW. Disorders of Red Blood Cells-Anemias. In: Atlas of Hematopathology. 1st ed. Waltham, MA: Academic Press; 2013. p. 675–704.

2. Schneider RG, Hightower B, Hosty TS, et al. Abnormal hemoglobins in a quarter million people. Blood. 1976;48(5):629–637. 3. Allison AC. Genetic factors in resistance to malaria. Ann N Y Acad Sci. 1961;91:710–729. 4. Welch RB, Goldberg MF. Sickle-cell hemoglobin and its relation to fundus abnormality. Arch Ophthalmol. 1966;75(3):353–362. 5. Moschandreou M, Galinos S, Valenzuela R, et al. Retinopathy in hemoglobin C trait (AC hemoglobinopathy). Am J Ophthalmol. 1974;77(4):465–471. 6. Pichard E, Resnikoff S, Serre L, Coulibaly D. [Retinovitreous hemorrhage in hemoglobinopathies]. J Fr Ophtalmol. 1991;14(6–7):377–381. French. 7.

Abrams LS, Goldberg MF. Retinopathy associated with hemoglobin AC. Arch Ophthalmol. 1994;112(11):1410–1411.

8. Hingorani M, Bentley CR, Jackson H, et al. Retinopathy in haemoglobin C trait. Eye (Lond). 1996;10 (Pt 3):338–342. 9. Goldberg MF. Natural history of untreated proliferative sickle retinopathy. Arch Ophthalmol. 1971;85(4):428–437. 54

10. Condon PI, Serjeant GR. Ocular findings in hemoglobin SC disease in Jamaica. Am J Ophthalmol. 1972;74(5):921–931. 11. Helm CJ, Holland GN. Ocular tuberculosis. Surv Ophthalmol. 1993;38(3):229–256. 12. Namperumalsamy P, Shukla D. Eales Disease. In: Retina. 5th ed. Elsevier Saunders; 2013. p. 1479–1485. 13. Papaliodis G. Eales' Disease. UpToDate. (accessed online 5/30/19, https://www.uptodate.com/contents/ealesdisease?topicRef=7999&source= see_link). 14. Kase S, Namba K, Horie Y, Kotake S, Ohno S. Repeated exacerbations of ocular inflammation with vitreous hemorrhage in a patient with HLA-B27 associated uveitis. J Med Invest. 2007;54(3–4):350–353. 15. Orosz Z, Csiszar A, Labinskyy N, et al. Cigarette smokeinduced proinflammatory alterations in the endothelial phenotype: role of NAD(P)H oxidase activation. Am J Physiol Heart Circ Physiol. 2007;292:H130–H139. 16. Lin P, Loh AR, Margolis TP, Acharya AR. Cigarette smoking as a risk factor for uveitis. Ophthalmology. 2010; 117(3): 585–590.

Scholarly Research In Progress • Vol. 3, November 2019

Transulnar Approach for Balloon Aortic Valvuloplasty for Severe Aortic Stenosis: A Novel Approach Despite Multiple Comorbidities Bradley Very1†, Stephen Long1†, Stephen Voyce2, and Yassir Nawaz2

Geisinger Commonwealth School of Medicine, Scranton, PA 18509 Geisinger Community Medical Center, Danville, PA 17821 †Doctor of Medicine Program Correspondence: bvery@som.geisinger.edu 1

Abstract The increasing utilization of transcatheter aortic valve implementation has limited the role of balloon valvuloplasty to patients with severe aortic stenosis in cardiogenic shock or as a palliative measure in patients unable to tolerate definitive surgical intervention. The transradial approach is the primary preferred approach for percutaneous interventions due to morbidity and mortality benefits. However, ulnar access is associated with a similar safety profile and can be used in cases of ipsilateral radial access failure. This case describes balloon valvuloplasty via an ulnar approach as a bridge to planned definitive treatment with transcatheter aortic valve implantation.

Introduction Balloon valvuloplasty (BAV) initially emerged as a low-cost and well-tolerated treatment for severe aortic stenosis (AS); however, due to early restenosis and limited (if any) mortality benefit, its clinical utility has been limited to patients with severe AS in cardiogenic shock or as a palliative measure in patients deemed unfit for surgical intervention (1, 2). The emergence of transcatheter aortic valve implantation (TAVI) has further limited the role of BAV in the management of AS (1). Additionally, the transradial approach has largely been adopted as the preferred vascular access site of coronary intervention due to morbidity and mortality benefits (3). Recent studies have suggested that ulnar artery access is associated with a similar safety profile to radial access and can be used after ipsilateral radial access failure. To date, a case of BAV performed via a transulnar has not been reported, and the ulnar artery is rarely selected as the first-line vascular access point for percutaneous coronary procedures (4). Here we describe a case in which a BAV via an ulnar approach was used as a bridge to planned definitive treatment with TAVI.

Case Presentation A 59-year-old Caucasian male with a past medical history significant for morbid obesity with BMI>65, obstructive sleep apnea, coronary artery disease status post stent placement in left anterior descending, left ramus, and left circumflex arteries, congestive heart failure, and aortic stenosis presented to the emergency department for evaluation of two weeks of exertional chest pain, shortness of breath, weight gain, and increasing lower extremity edema. Physical exam revealed tachycardia, tachypnea, bilateral 2+ pitting edema of the lower extremities, marked jugular venous distention, and a systolic ejection murmur over the aortic area. Initial lab studies showed

a brain natriuretic peptide (BNP) of 1654 pg/mL and Troponin I of 0.070 ng/mL. Chest X-ray showed mild pulmonary vascular congestion and cardiomegaly. The patient was admitted to telemetry for management of acute on chronic congestive heart failure. Transthoracic echocardiography on showed a left ventricular ejection fraction of 20% to 24% with severe to critical aortic stenosis. Per the patient, he had previously been refused transcatheter aortic valve replacement at an outside institution. The patient was evaluated by cardiac surgery and interventional cardiology and was determined not to be candidate for open aortic valve replacement. The patient was considered for TAVI, but at the time of evaluation, the patient was unable to tolerate supine positioning due to orthopnea. Additionally, he was considered high risk for TAVI via femoral artery access due to morbid obesity and large abdominal pannus. Thus, the decision was made to proceed with balloon aortic valvuloplasty for symptom control and as a bridge to definitive TAVI therapy. Vascular access was obtained via the right ulnar artery under direct ultrasound guidance without complication. Ulnar access was selected based on favorable ulnar artery size compared to the ipsilateral radial artery—the distal ulnar artery was measured to have a diameter of 4.0 mm while the distal radial artery dimeter was measured at 2.8 mm. Selective coronary angiography of the right and left coronary arteries was performed using a 5 Fr Tiger catheter (Terumo, Somerset, NJ). Valvuloplasty was performed on the aortic valve using twoballoon inflation using 20-mm and 22-mm Tyshak II balloons (Braun, Kronberg, Germany) inserted via an 8 Fr sheath in the right ulnar artery. Prior to valvuloplasty, a Langston catheter (Teleflex, Morrisville, NC) was used to perform simultaneous LV/AO pressure tracings showing a peak-to-peak pressure gradient of 42 mmHg and a mean pressure gradient of 34 mmHg. After BAV was performed, peak-to-peak pressure gradient improved to 34 mmHG and mean pressure gradient was reduced to 28 mmHg. The patient was also found to have very high filling pressures, with central venous pressure of 25 mmHg and pulmonary capillary wedge pressure of 30 mmHg. The patient was found to be in new onset atrial fibrillation with rapid ventricular rate (RVR) in the days following BAV. This was managed with amiodarone and heparin drip. Milrinone drip was initiated to achieve adequate urine output and diuresis. The patient’s symptoms continued to improve in the days following BAV. By post-op day eight, the patient was diuresed almost 30 L. The patient was discharged to a rehabilitation facility with plan to be evaluated for TAVR as an outpatient.

Transulnar Approach for Balloon Aortic Valvuloplasty for Severe Aortic Stenosis

Notably, this patient was discharged on a Lovenox bridge to warfarin. One week after discharge, the patient developed severe melena without hematochezia and was found to have a hemoglobin of 6.5 and INR of 1.91. His warfarin was discontinued and two units of packed red blood cells and two units of fresh frozen plasma were transfused. He was admitted to the intensive care unit, and was managed aggressively for presumed GI bleed and CHF exacerbation. After developing increasing levophed requirements, the patient’s family declared him to be a limited code and the patient subsequently expired.

Discussion Historically, due to early restenosis and failure to improve long term survival, BAV was limited to patients with severe symptomatic AS in cardiogenic shock or symptom palliation in patients too frail for surgical aortic valve replacement (1, 2). The emergence of TAVI has created new discussion on the clinical applications of BAV to include assessment of therapeutic response of a reduction in aortic gradient in patients with multiple comorbidities, assessment of symptomatic improvement prior to consideration of definitive TAVI intervention, and as a bridge therapy in patients with symptomatic severe AS in patients with prohibitive perioperative risk (1, 5). The European Society of Cardiology guidelines state that BAV may be considered as a bridge to surgery or TAVI and is a reasonable approach in hemodynamically unstable patients or in patients with symptomatic severe aortic stenosis who require urgent noncardiac surgery (1, 6), and the American Heart Association/ American College of Cardiology suggest BAV may be considered as a bridge to TAVI or AVR for patients with severe symptomatic AS (class IIb, level of evidence: C) (1, 7). Currently, 20% to 25% of BAV procedures are performed as a bridge to TAVI, and previous studies have suggested that BAV is associated with a procedural mortality rate of 1.5% and in-hospital mortality rate of 4.0% (8). Notably, the widespread adoption of TAVI has led to significantly improved outcomes in this challenging patient population (8). Additionally, while the transradial route is gaining acceptance as a first-line approach for percutaneous coronary intervention and angiography, the ulnar artery is rarely selected for coronary procedures (4). Randomized trials by Hahalis et al. have demonstrated that the ulnar approach appears noninferior to the transradial route when crossover has not been required and may serve as a second-line arterial approach (4). Patients with limitations to transradial approaches such as small artery size, anatomic variations, previous radial artery harvesting for coronary artery bypass grafting, radial artery occlusion, and radial artery spasm can be considered for safe and effective transulnar artery interventions (9). This case describes the successful use of a transulnar approach for balloon aortic valvuloplasty in a patient with multiple comorbidities. This was intended to serve as a therapeutic bridge to TAVI, but this patient unfortunately expired before this definitive treatment could be performed. This case supports the viability of transulnar access for BAV either as the first-line vascular access site or after failure of ipsilateral access transradial access. The feasibility of ulnar access for TAVI and transcatheter aortic valve replacement (TAVR) has not been assessed, and no cases of this approach 56

have been documented. This case also reflects the significant morbidity and mortality seen in this patient population. While this patient’s fatality is likely multifactorial, the anticoagulation therapy needed to reduce the stroke risk conferred by newonset atrial fibrillation was undoubtedly the cause of his severe anemia.

Conclusion The ulnar artery is emerging as a potential vascular access option in patients undergoing percutaneous coronary intervention, and has been shown to confer adverse event rates comparable to transradial access. This case supports the feasibility of ulnar artery access in patients requiring BAV as palliation or as a bridge to TAVI.

References 1.

Keeble TR, Khokhar A, Akhtar MM, Mathur A, Weerackody R, Kennon S. Percutaneous balloon aortic valvuloplasty in the era of transcatheter aortic valve implantation: a narrative review. Open Heart. 2016;3(2):e000421.

2. Kogoj P, Devjak R, Bunc M. Balloon aortic valvuloplasty (BAV) as a bridge to aortic valve replacement in cancer patients who require urgent non-cardiac surgery. Radiol Oncol. 2014;48(1):62–66. 3. Ferrante G, Rao SV, Jüni P, Costa, BR, et al. Radial Versus Femoral Access for Coronary Interventions Across the Entire Spectrum of Patients With Coronary Artery Disease. JACC: Cardiovascular Interventions, (2016);9(14):1419–34. 4. Hahalis G, Tsigkas G, Xanthopoulou I, et al. Transulnar compared with transradial artery approach as a default strategy for coronary procedures: a randomized trial. The Transulnar or Transradial Instead of Coronary Transfemoral Angiographies Study (the AURA of ARTEMIS Study). Circulation. Cardiovascular interventions. Jun 2013;6(3):252–261. 5. Sandhu K, Krishnamoorthy S, Afif A, Nolan J, Gunning MG. Balloon aortic valvuloplasty in contemporary practice. Journal of interventional cardiology. Jun 2017;30(3):212– 216. 6. Baumgartner H, Falk V, Bax JJ, et al. ESC Scientific Document Group, 2017 ESC/EACTS Guidelines for the management of valvular heart disease, European Heart Journal, 21 Sep 2017;38(36):2739–91. 7.

Nishimura RA, Otto CM, Bonow RO, et al. 2014 AHA/ACC Guideline for the Management of Patients With Valvular Heart Disease. A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. 2014;63(22):e57–e185.

8. Kumar A, Paniagua D, Hira R, et al. Balloon Aortic Valvuloplasty in the Transcatheter Aortic Valve Replacement Era. J Invasive Cardiol. 2016; 28:341–348 9. Roghani-Dehkordi F, Mansouri R, Khosravi A, Mahaki B, Akbarzadeh M, Kermani-Alghoraishi M. Transulnar versus transradial approach for coronary angiography and angioplasty: Considering their complications. ARYA Atheroscler. 2018;14(3):128–131.

Scholarly Research In Progress • Vol. 3, November 2019

Risk Factors Associated with Opioid Overdose Deaths across 30 Counties in Pennsylvania for 2018 Eric A. Ligotski1*‡, Laura M. Loeser1*‡, and Alison T. Varano1†‡

Geisinger Commonwealth School of Medicine, Scranton, PA 18509 *Master of Biomedical Sciences Program †Doctor of Medicine Program ‡Authors contributed equally Correspondence: avarano@som.geisinger.edu 1

Abstract Introduction: The opioid crisis in the United States has become an increasing public health concern, affecting millions of people across the United States. We sought to analyze which, if any, socioeconomic risk factors predisposed certain populations to a greater risk of opioid overdose deaths. We examined data on a county-level for Pennsylvania. Methods: The opioid overdose deaths from the thirty counties in Pennsylvania for 2018 reported by Overdose Free PA data were examined. The socioeconomic risk factors were obtained from United States Census Bureau and First Time Home Buyer. Socioeconomic risk factors analyzed were ethnicity, education, poverty, median income per one person, and health insurance. Data was then adjusted per 100,000 people. Pearson product moment correlations and R2 values were calculated. Results: The counties with the most opioid overdose deaths were Susquehanna and Erie. There were no significant associations between opioid overdose deaths and the socioeconomic risk factors. For example, median income per one person, had an R2 value of 0.06 and a p-value of 0.20. Discussion: This ecological study used county-level data that was previously reported. We conclude there were no statistically significant correlations between any socioeconomic risk factors and opioid overdose deaths. Future studies may consider expanding the sample of counties or examining additional risk factors. Two possible limitations from our investigation may be that the group-level data is not representative on an individual level and that the reported overdose data are inconsistently obtained and reported. Although we chose to use county-level data to address the population, future studies may further analyze the data on an individual level.

Introduction Opioid use and overdoses have been increasing across the United States in the past years as a result of a multitude of factors. It is also a possibility that the deaths attributable to opioid overdose may be greater than what has been documented (1). Over the past 20 years chronic pain and the perceived undertreatment of chronic pain have contributed to the escalation of prescriptions and abuse (2). The misuse of such drugs, along with their synthetic analogues, has created a national crisis. Starting in the late 1990s, pharmaceutical companies began developing new opioids, while reassuring health providers

of the safety of their products. This led health care providers to prescribe stronger opioids like fentanyl and oxycodone, which led to illicit distribution and misuse, indicating that these substances had a high potential for misuse (3). In 1995, Dr. James Campbell introduced the concept of utilizing pain as a fifth vital sign (4). This idea spread and became incorporated into other national organizations such as Veterans Health Administration and The Joint Commission, which advocated for pain to be assessed in all patients (4). This began the trend of increased opioid prescriptions when patients did not respond to milder recommendations such as acetaminophen or NSAIDs (4). A devastating consequence of opioid misuse is that it leads to individuals turning to opiates such as heroin, providing further opportunity for infectious diseases such as HIV or hepatitis to spread as a result of injection drug use (3). One way the opioid crisis is being addressed is the removal of pain as the fifth vital sign (5). Further, education on managing pain is also being implemented to teach health care providers about proper examination and management of pain, as this is one of the contributing factors to increased prescriptions (4). The National Institutes of Health, in conjunction with major pharmaceutical companies working to construct research and new methods to manage chronic pain, developed new medications and technologies to treat opioid misuse and provided more overdose prevention and reversal interventions (3). Socioeconomic risk factors such as ethnicity, income, poverty, health insurance, and education were examined in this study to identify if there were any correlations with respect to higher opioid overdose deaths. It has been previously established that the increased amounts of opioid overdoses are related to the decreased life expectancy of white individuals in the U.S. (6). Changing economic and marketing factors have increased the availability of these drugs to the white population, which contributes to the rising overdose rates in this population (6). Data collected in California has demonstrated specifically that low-income white populations are especially susceptible to opioid overdoses (7). The researchers proposed this could be attributed to the fact that the lowest-income quintiles had the highest proportion of white habitants (7). This provided our rationale for examining any possible correlations between ethnicity, income, and poverty. The death toll due to overdose-related deaths has increased among all populations, regardless of education level. While the overdose rate has increased since the 1990s, educational gradients in life expectancy have also been increasing (8). Between the years of 1992 and 2011, data throughout the U.S. demonstrates that the life expectancy gap between individuals who attended

Risk Factors Associated with Opioid Overdose Deaths

college and individuals who did not attend college can be explained, in part, by the higher overdose rates in individuals that did not attend college (8). Health insurance was examined as another potential risk because of the hypothesis that areas with more access to health care had the ability to obtain and subsequently illicitly distribute opioid drugs. Although there have been previous studies conducted on a national level, in the current study, we wished to further examine opioid overdose deaths specifically for the state of Pennsylvania. According to data available from 2016, 13 individuals from Pennsylvania died each day from drugrelated overdoses (9). In 85% of those drug-related deaths, opioids were identified as being present (9). However, few investigations have been conducted on a county level for trends of opioid overdose deaths. The present study investigated socioeconomic factors that may be contributing to the opioid overdose deaths in the years 2017 and 2018, such as income and poverty level, access to health care, education attainment, and ethnicity. We chose to examine opioid overdose deaths classified as accidental or suicidal among the counties. This study seeks to determine if there are specific socioeconomic risk factors for deaths related to the use of opioids and whether there were any specific trends among counties for opioid overdose deaths in Pennsylvania.

Results Thirty of the 67 counties in Pennsylvania had data available for analysis. Susquehanna and Erie counties had the highest number of deaths (166 and 102 deaths respectively, as shown in Figure 1). These were the only counties with number of populations-corrected opioid deaths over 100. Susquehanna, with 166 opioid deaths, was a statistically significant outlier (Grubbs’ p<0.05). Mercer, Cambria, Dauphin, Lawrence, Greene, York, Wayne, and Westmoreland were the remaining counties in the top 10 of the total 30 counties with the highest opioid deaths.

Methods Procedures Complete data about 2018 opioid overdose deaths as well as the previously identified demographic characteristics were obtained for the following 30 counties: Adams, Allegheny, Armstrong, Beaver, Berks, Bucks, Butler, Cambria, Centre, Chester, Clarion, Clinton, Crawford, Dauphin, Erie, Fayette, Franklin, Greene, Indiana, Lackawanna, Lancaster, Lawrence, Lycoming, Mercer, Montgomery, Susquehanna, Union, Wayne, Westmoreland, and York. The percentages of the socioeconomic risk factors were multiplied by the county population to determine the number of people with the socioeconomic risk factor and then adjusted to represent the percent per 100 000 people. The number of deaths were also adjusted to per 100 000 people. Data were collected from three databases: Overdose Free PA, Pennsylvania Home Buyer, and United States Census Bureau. The quantity of opioid overdose deaths, as well as the substances implicated in these deaths, was collected through Overdose Free PA (10). The drugs they included when looking into the deceased include opioids along with alcohol and cocaine (10). Median income per one person at a county level was obtained through Pennsylvania Home Buyer (11). The remaining risk factors were from the United States Census Bureau (12). Data analysis Figures were created using GraphPad Prism. The R2 and p-value were calculated through linear regression analysis using Prism. The p-value was deemed significant if <0.05 and given the small number of counties with complete information (30), as a trend if p<0.10. Outliers were identified on GraphPad Prism with the Grubbs’ Test.

Figure 1. Ten Pennsylvania counties with highest number of opioid deaths in 2018.

The relationship of socioeconomic risk factors and number of opioid deaths for each county analyzed are shown in Figure 2. There was no obvious trend correlating the percentage of the white population with an increase or decrease in opioid deaths, as shown in Figure 2A (R2 =0.02; p=0.44). However, once the population reaches approximately 87%, the number of opioid deaths increases slightly with two extreme scores. There are more counties with greater than 87% white population, therefore this may contribute to the increase. Similarly, when comparing the percentage of the black population with number of opioid deaths, there was no obvious trend as shown in Figure 2B (R2 =0.02; p=0.41). There was no association between the median income per one person and the number of opioid deaths, as shown in Figure 2C (R2 =0.06; p=0.20). Overall, the deaths appear to be spread equally regardless of income. Although not significant as determined by statistics, the number of opioid deaths showed a slight decrease once reaching a median income per one person of $46,000, and the greatest number of deaths were observed for those with an median income per one person under $36,000. Likewise, there was no association between

Risk Factors Associated with Opioid Overdose Deaths

Discussion The opioid epidemic throughout the United States has gained national attention. Previous studies have examined trends of opioid overdose deaths on a national and state level (10, 13); however, we chose to examine trends within individual Pennsylvania counties. The methodology employed in this study allows for an in-depth analysis of these trends on a county level to determine which counties have or report the greatest number of deaths and explore any potential socioeconomic risk factors which may be associated with opioid overdose deaths. The results of this study showed no significant associations or trends among selected socioeconomic risk factors and number of opioid deaths per county. We hypothesized that certain populations of individuals may be at greater risk for opioid deaths due to socioeconomic factors such as lack of health care and education resources. We examined ethnicity, educational attainment, median income per person, poverty, and access to health care. Our results did not support the original hypothesis. There was no significant correlation found among the white or black population and number of opioid deaths. Similar to a previous analysis of overdose deaths in Pennsylvania, our data shows that the racial breakdown of deaths corresponds approximately with the racial breakdown of the overall county population (9). Our findings suggest that ethnicity, at a county level, was not a risk factor for opioid deaths. We also hypothesized that lack of access to health insurance would result in a greater number of opioid deaths. In contrast, our results indicated no significant relationship in the percentage of the population without health insurance and the number of opioid deaths. It is difficult to determine if access to health insurance would lead to a greater Figure 2. Number of opioid deaths per county compared to: A) Percentage of white amount of deaths due to prescription opioids, population; B) Percentage of black population; C) Percentage of population in poverty; D) or if lack of access to health insurance Median household income per one person; E) Percentage of population without health insurance; F) Percentage of population with a bachelor’s degree. would result in street drugs contaminated with fentanyl. Our data suggests it may be a combination of both of these factors. We the percentage of the population in poverty and number of also predicted that individuals with a lower median income, or 2 opioid deaths, as shown in Figure 2D (R =0.02; p=0.44). the greater percentage of population living in poverty, would Additionally, there was no correlation between the percentage result in a greater number of opioid deaths. However, our of the population under 65 years of age without health results indicated no such association. Median income per one insurance with the number of opioid deaths as shown in Figure person did not vary greatly among selected counties, which 2E (R2 =0.02; p=0.55). Although not significant as determined may have influenced our results. Susquehanna county was the by statistics, it appeared counties with a greater population only statistical outlier (Grubbs’ p<0.05); however, none of the without health insurance had a slightly lower number of opioid examined socioeconomic risk factors became significant with deaths. There was also no correlation between the percentage Susquehanna removed after analysis. of population with a bachelor’s degree and number of opioid deaths, as shown in Figure 2F (R2 =0.03; p=0.33).

Risk Factors Associated with Opioid Overdose Deaths

There were several limitations to our analysis in this study. One major limitation was that Overdose Free PA did not include all the counties in Pennsylvania. Reported data on opioid overdose deaths varies from database to database (14). We reported opioid overdose deaths from one database to remain consistent. Another limitation of our study was that although we narrowed deaths down to accidental and/or suicide, it would have been more accurate to report one or the other; however, some counties combined the two. One limitation for the socioeconomic risk factors selected dealt with ethnicity. The United States Census Bureau mentioned that Hispanic and Latino numbers may not be completely accurate, because they do not always report their ethnicity as such and may report it as white (12). Finally, the procedures employed in the U.S. for death determination are not homogenous. Pennsylvania had the highest percentage (50.8%) of death certificates in the country where the drug involved was listed as unspecified (1). Inconsistency in death determination procedures across counties might introduce variability which could decrease demographic correlations.

Conclusion Although efforts have been made to address the opioid crisis, increasing rates of opioid overdose deaths remain a public health concern. The results of this study suggest that it is not a simple matter of distinguishing one population that is at a higher risk. Additionally, there is not one specific socioeconomic risk factor examined in this study that shows a statistically significant correlation to a higher number of opioid overdose deaths. We hope that these findings will provide insight when considering other risk factors in the future. Further, this data can be compared to similar studies in other counties to potentially rule out certain variables as risk factors for overdoses. In conclusion, our results indicate that the opioid epidemic affects everyone, regardless of socioeconomic factors such as ethnicity, financial stability, and educational status. Further research will be needed. We propose to expand our study to include more counties nationally as the next step. In addition, we propose to further classify socioeconomic risk factors, such as whether a death was due to accidental overdose or a suicide, removing possible confounding factors. One could also examine whether a income from two people instead of one person holds any significant value. Since our data indicates that (at least for these 30 counties in Pennsylvania) demographic and socioeconomic factors do not necessarily predispose a certain population to increased overdose risk, future research could examine genetic risk factors to overdose trends. This could aid in examining multifactorial traits that may contribute to opioid overdoses. Overall, our study indicates that the examined socioeconomic risk factors do not put one specific population at a higher risk of overdosing on opioids within the counties we examined in Pennsylvania.

Acknowledgments We would like to thank Brian Piper, PhD, and Elizabeth Kuchinski, MPH, for their feedback on earlier versions of this manuscript.

References 1.

Buchanich J, Balmert L, Williams K, Burke D. The effect of incomplete death certificates on estimates of unintentional opioid-related overdose deaths in the United States, 19992015. Public Health Rep. 2018; 133(4):423–31.

2. Manchikanti L, Helm S, Fellows B, Janata J, Pampati V, Grider J. Opioid epidemic in the United States. Pain Physician. 2012; 15(3 supplementary): S9–38. 3. The National Institution on Drug Abuse. Opioid overdose crisis. (Revised January 2019). https://www.drugabuse.gov/ drugs-abuse/opioids/opioid-overdose-crisis. Accessed: 5 April 2019. 4. Morone N, Weiner D. Pain as the fifth vital sign. Clin Ther. 2013; 35(11): 1728–32. 5. Cabrera F, Gamarra E, Garcia T, Littlejohn A, Chinga P, Pinental-Morillo L, et al. Opioid distribution trends (2006– 2017) in the US territories. PeerJ. 2019; 7: e6272 6. Hansen H, Netherland J. Is the prescription opioid epidemic a white problem? Am J Public Health. 2016; 106(12): 2127–29. 7.

Friedman J, Kim D, Schneberk T. Assessment of racial/ ethnic and income disparities in the prescription of opioids and other controlled medications in California. JAMA Intern Med. 2019; 179(4): 469–76.

8. Ho J. The contribution of drug overdose to educational gradients in life expectancy in the United States, 1992– 2011. Demography. 2017; 54(3): 1175–1202. 9. Drug Enforcement Administration, University of Pittsburgh Pharmacy. Analysis of overdose deaths in Pennsylvania, 2016. https://www.overdosefreepa.pitt.edu/wp-content/ uploads/2017/07/DEA-Analysis-of-Overdose-Deaths-inPennsylvania-2016.pd_-1.pdf. Accessed: 10 April 2019. 10. OverdoseFreePA. Opioid Threat Report. https://www. overdosefreepa.pitt.edu/know-the-facts/view-overdosedeath-data/. Accessed: 20 April 2019 11. Pennsylvania Home Buyer. 2018 Median household income for Pennsylvania counties. http://www. pafirsttimehomebuyer.net/pa-median-income.html. Accessed: 20 April 2018 12. U.S. Census Bureau. QuickFacts: Pennsylvania. https:// www.census.gov/quickfacts/fact/table/pa/PST045218. Accessed: 20 April 2019. 13. Hedegaard H, Warner M, Miniño AM. Drug overdose deaths in the United States, 1999–2016. NCHS Data Brief, no 294. https://www.cdc.gov/nchs/products/databriefs/ db294.htm. Accessed: 23 April 2019. 14. Cordant Health Solutions. Why Opioid Overdose Statistics Vary. http://cordantsolutions.com/opioid-overdosestatistics-vary/. Accessed: 23 April 2019.

Scholarly Research In Progress • Vol. 3, November 2019

Medical Cocaine Use from 2006–2017 and its Application in U.S. Medicine Kristin D. Feickert1*and Stephanie E. England1*

Geisinger Commonwealth School of Medicine, Scranton, PA 18509 *Master of Biomedical Sciences Program Correspondence: kfeickert@som.geisinger.edu 1

Abstract Background: The purpose of this study was to investigate the distribution of cocaine in 10 states from 2006 through 2017 and examine its use in medicine today. Methods: Cocaine distribution was collected over a 12-year period for 10 East Coast states via the U.S. Department of Justice’s Automation of Reports and Consolidated Orders System (ARCOS). Population data was obtained from the U.S. Census Bureau and statistical analyses were performed using Prism. Results: There was an inverse relationship between retail distribution and time. Cumulative acquisition of all 10 states by year showed that medical cocaine’s distribution peak occurred in 2006. South Carolina’s distribution declined by 79.75%, while other states exhibited a more moderate decline (22.82–72.36%). A two-tailed paired t-test of retail cocaine distribution between 2006 and 2008 yielded statistically significant results; t(9)=2.8, p≤0.05. All years post 2008 also resulted in significant reductions. This decline for the years compared helps support the notion that there is a decreasing trend of cocaine distribution occurring in the marketplace. The correlation coefficient between 2006 and 2017 was r=0.71, p≤0.001. Conclusion: The observed decline in medical cocaine use is due in part to the addictive properties, serious side effects (such as myocardial infarction), and number of safer alternatives available. Despite cocaine’s downward trend, it still has applicability as an anesthetic in various surgeries, a diagnostic tool in Horner’s syndrome, as a standard in analytical chemistry, and potentially as a drug for relieving symptoms of Parkinson’s.

Introduction Cocaine is rarely thought of as medication. It is more commonly considered an illegal and addictive drug. However, cocaine has been used medically to numb patients in ancient procedures like trephination (1). Carl Koller was one of the first people to rediscover this technique in 1884. After hearing about its tongue-numbing properties, he applied drops of cocaine into the eyes of a frog and saw that he could insert a needle without the frog reacting (1). He and a colleague then repeated this experiment on themselves and noticed that they could not feel any pain. This discovery led Koller to use cocaine as an anesthetic. He was the first to be widely recognized for this technique, where he used it in an operation for a patient with glaucoma (1). Early societal uses were mostly as a panacea and included wine made of cocaine. One example is Vin Mariani, a brew

which inspired the original formula of Coca-Cola. Furthermore, prominent figures used cocaine and promoted its uses. Of note was Sigmund Freud, who thought it could be used to wean morphine addicts off morphine and cure depression. Moreover, the founder of the first surgical residency program in the nation at Johns Hopkins Hospital was a frequent cocaine user. He designed residency expectations off the large number of hours and little sleep he needed while under the drug’s influence (1). In contemporary medicine, cocaine has been used for a variety of different purposes, including that of an anesthetic in various types of surgery (ophthalmology, otolaryngology, and even urology) (1, 2). In addition, cocaine is used as an anesthetic for intraurethral procedures (2). Labetalol is often given to counteract any complications that could arise from anesthetic cocaine administration, such as cardiovascular complications and death (3). Moreover, the advent of alternative drugs like xylocaine, lidocaine, and oxymetazoline has reduced the dependency on cocaine since they produce similar results without the adverse side effects (2, 4). Lidocaine, in particular, is more efficacious than cocaine at inducing comfort in patients undergoing painful procedures (such as insertion of a nasogastric tube) (5). It also has been utilized to diagnose specific diseases such as Horner’s syndrome and shows potential application for common diseases like Parkinson’s disease. In the latter case, it was found to help relieve “off” periods in patients (6). However, the overall utilization of cocaine has been waning relative to its peak use in U.S. medicine around the late 1800s (where it was used as a panacea) (1). There is good reason for this decline, as its addictive properties and side effects can often prove deadly. There are a variety of cardiac events related to cocaine use that can result in myocardial infarctions and/or death (7). With technology ever advancing, many synthetic substitutions have replaced medical cocaine use. An example is apranoclonidine, which has greater sensitivity than cocaine for diagnosising Horner’s syndrome (8). In this study, we aimed to determine trends of medical cocaine use in 10 states near the East Coast, from 2006 through 2017. We hoped to ascertain if a downward trend in cocaine distribution was observable, given the plethora of safer alternatives available for treatment and diagnosis.

Methods Procedures Cocaine retail drug distribution by weight (g) was extracted for 10 states on the East Coast (Delaware, Florida, Georgia, Maryland, North Carolina, Pennsylvania, South Carolina,

Medical Cocaine Use

Table 1. Population data

Table 2. Cocaine retail distribution in milligrams

Tennessee, Virginia, and West Virginia). Raw data was obtained for 2006 through 2017 from ARCOS. State-level population data for the same years was obtained from the U.S. Census Bureau; exact values can be viewed in Table 1. Procedures were deemed exempt from review by the IRB of the University of New England. Data analysis The raw data was then converted to mg (Table 2) and population corrected to achieve units of cocaine in mg/ population. Figures and statistical calculations (two-tailed paired t-test and linear regression) were completed and prepared using Prism 8. The paired t-test was evaluated at the 0.001, 0.01, and 0.05 significance level. A correlation (r) between population corrected use in 2006 and 2017 was achieved. Percent change of cocaine distribution between 2006 and 2017 was also calculated for all 10 states.

Figure 1. Total cocaine distribution of 10 states by year

Results Total cocaine distribution per year was evaluated over the 12-year time span (Figure 1). The highest cocaine distribution occurred in 2006, while the lowest was in 2017. As such, retail distribution between 2006 and 2007 was computed yielding nonsignificant results. However, a paired t-test of the year 2006 versus 2008 showed statistically significant results; t(9)=2.8, p=0.02. Analyses for 2006 versus subsequent years are shown in Table 3. Linear regression was then used to calculate the correspondence between total cocaine distribution in 2006 and 2017. The results are shown in Figure 2, with each data point representing a state. Half (50%) of the variation in cocaine distribution was due to differences between 2006 and 2017. The correlation coefficient was 0.71 for these 2 years. Cocaine distribution by time for the 10 states is shown in Figure 3. From 2006 to 2017, South Carolina declined the most, by 79.75%. Other states exhibited a more moderate decline and 62

Table 3. Paired t-test results

can be seen in Table 4. Also notable was that West Virginia experienced a 26.95% increase from 2009 to 2010 and then a 27.45% decrease from 2010 to 2011.

Discussion The results generally supported our hypothesis that there would be a decreasing trend in cocaine distribution. This was

Medical Cocaine Use

expected since there are safer alternatives available for the medical setting and provider use. However, there were a few unexpected findings. In 2010, West Virginia had a medical cocaine distribution peak that was unprecedented. Furthermore, Delaware’s distribution decreased the least from 2006 to 2017. Its cocaine acquisition over time remained relatively stable, often fluctuating in a somewhat negative pattern.

Figure 2. Distribution of cocaine with time

Figure 3. Distribution of cocaine by state and time

Table 4. Percent decrease in cocaine distribution from 2006 to 2017

Despite these unexpected findings, overall cocaine distribution decreased significantly. This decline is consistent with the phasing out of cocaine. Horner’s syndrome was commonly diagnosed with cocaine eye drops, but limitations in terms of its diagnostic utility were identified as early as 1986. One such limitation was that the cocaine eye drop test produced too many false positives in patients who did not have the disease (9). In 2005, it was discovered that cocaine was a weak dilator and therefore proved ineffective for diagnosing milder forms of Horner’s syndrome (particularly partial Horner’s syndrome) (10). Substitutions like apranoclonidine may not account for much of the observed cocaine decrease, as there is a low frequency of Horner’s syndrome in the population (10). Furthermore, apranoclonidine cannot serve as a complete cocaine substitute due to its high risk among certain patient populations. For example, infants risk central nervous system depression with use of apranoclonidine, so cocaine eye drops must be used instead (11). Although there is declining use of cocaine for Horner’s syndrome, preliminary studies show promising results for its use in Parkinson’s disease. Case studies revealed that stuttering and motor tics (which tend to be associated with Parkinson’s disease) responded to cocaine administration, disappearing completely (12). There have also been reported cases of patients with Parkinson’s that self-medicate with cocaine to relieve dyskinesias in the “off” periods when their L-Dopa medication is not effective (13). In the past, people thought cocaine increased the risk of getting Parkinson’s disease via neurodegeneration of dopamine neurons. However, ambiguity exists regarding the certainty of this mechanism and rodent trials seem to suggest otherwise (14). A study in 2012 found that there was no increased risk of Parkinson’s disease among cocaine users. In fact, methamphetamine was more predictive of disease development (15). There were a variety of limitations to this study that will require further research. One limitation was the number of states that were studied. Only 10 out of the 50 that comprise the United States were examined, so overall national trends cannot be accurately stated. Furthermore, the ARCOS data set is restricted to the United States. Finding more or other data sources that track the use of medical cocaine worldwide would be particularly useful. This would help determine medically global cocaine trends and it would be interesting to see if substitutions are occurring around the world. Would there be countries who were outliers to this trend? If so, why would that be? In addition, ARCOS does not account for mail-order pharmacies that could ship cocaine across state lines. It also does not contain a mechanism to differentiate between human and veterinary use (e.g., the basic science research of addiction). Another limitation was that medical 63

Medical Cocaine Use

cocaine acquisition by specific entities (e.g., pharmacies, hospitals, providers) was not evaluated. This makes it difficult to distinguish the purpose for medical cocaine use. Is it being used for mostly diagnostic or anesthetic purposes? Therefore, a breakdown of buyers would be useful but comes with its own set of complications. For example, cocaine use in hospitals could cause one to presume a surgical use, when this might not be accurate. Careful, delineated data (including the specific setting and/or purpose for acquisition) would be the ultimate goal for finding precise explanations behind shifting trends. Moreover, consulting experts in the field (e.g., practicing health care physicians) could be particularly fruitful and provide a better foundation for the use and/or disuse of medical cocaine. Further research is needed and would be necessary to determine the cause of unexpected results like that of West Virginia’s cocaine distribution in 2010. In March of this year, the Affordable Care Act was signed into law. The intention of this act was to increase health care access for all U.S. citizens. This new legislation could have created more funding, which in turn allowed more procedures to take place and/ or more hospitals/private practices to open. This rationale could explain the 2010 peak that occurred in West Virginia. However, more investigation is needed in order to deduce if this is an accurate explanation. As other states did not exhibit such a marked increase, more detail on the topic is warranted. Likewise, investigating medical practices and/or policies in Delaware could provide insight behind its trend. One possible explanation for its modest distribution and small decline could be that there is not a large number of practitioners in the specialties of interest that use cocaine. Delaware is a particularly small state, so looking into the number of doctors and their specialties (ophthalmology, otolaryngology, and urology) could provide the information needed to explain the state’s usage.

Conclusion The overall findings showed a decreasing trend for medical cocaine use in the 10 states observed. There was also a moderately strong relationship between cocaine distribution and time. Half (50%) of the variation in cocaine distribution was due to differences between 2006 and 2017. This indicates that the prescribing patterns were consistent half of the time. The remaining half was due to other factors. In the field of medicine, literature suggests that alternative substitutions have been made, overtaking the need for medical cocaine. However, cocaine use should not be completely disregarded or eliminated from medicine. Medicinal cocaine still holds applicability for certain patient populations and may have potential use in fields like Parkinson’s and addiction research.

Acknowledgments Thanks to feedback from Kenneth McCall, PharmD, University of New England, and our professors Elizabeth Kuchinski, MPH, and Brian Piper, PhD, at Geisinger Commonwealth School of Medicine.

References 1.

Altman AJ, Albert DM, Fournier GA. Cocaine’s use in ophthalmology: our 100-year heritage. Surv Opthalmol. 1985;29(4):300–306.

2. Gordetsky J, Bendana E, O'Brien J, Rabinowitz R. (Almost) painless surgery: a historical review of the evolution of intraurethral anesthesia in urology. Urology. 2011;77(1):12–6. 3. McGoldrick KE. Principles of opthalmic anesthesia. J Clin Anesth. 1989;1(4):297–312. 4. Long H, Greller H, Mercurio-Zappala M, Nelson LS, Hoffman RS. Medicinal use of cocaine: a shifting paradigm over 25 years. Laryngoscope. 2004;114(9):1625–9. 5. Ducharme J, Matheson K. What is the best topical anesthetic for nasogastric insertion? A comparison of lidocaine gel, lidocaine spray, and atomized cocaine. J Emerg Nurs. 2003;29(5):427–30. 6. Grybzowski A. Cocaine and the eye: a historical overview. Opthamologica. 2008;222(5):296–301. 7.

Goldstein RA, DesLauriers C, Burda AM. Cocaine: history, social implications, and toxicity—a review. Dis Mon. 2009;55(1):6–38.

8. Bremner F. Apranoclonidine is better than cocaine for detection of Horner Syndrome. Front Neurol. 2019;10:55. 9. Van der Wiel HL, Van Gijn J. The diagnosis of Horner's syndrome. Use and limitations of the cocaine test. J Neurol Sci. 1986;73(3):311–6. 10. Freedman KA, Brown SM. Topical apraclonidine in the diagnosis of suspected Horner syndrome. J Neuroophthalmol. 2005;25(2):83–5. 11. Martin GC, Aymard PA, Denier C, Seghir C, Abitbol M, Boddaert N, et al. Usefulness of cocaine drops in investigating infant anisocoria. Eur J Paediatr Neurol. 2017;21(6):852–857. 12. Linazasoro G, Van Blercom N. Severe stuttering and motor tics responsive to cocaine. Parkinsonism Relat Disord. 2007;13(1):57–8. 13. Di Rocco A, Nasser S, Werner P. Inhaled cocaine used to relieve "off" periods in patients with Parkinson disease and unpredictable motor fluctuations: a report of 2 cases. J Clin Psychopharmacol. 2006;26(6):689–90. 14. Mursaleen LR, Stamford JA. Drugs of abuse and Parkinson's disease. Prog Neuropsychopharmacol Biol Psychiatry. 2016;64:209–17. 15. Callaghan RC, Cunningham JK, Sykes J, Kish SJ. Increased risk of Parkinson's disease in individuals hospitalized with conditions related to the use of methamphetamine or other amphetamine-type drugs. Drug Alcohol Depend. 2012;120(1–3):35–40.

Scholarly Research In Progress â&#x20AC;˘ Vol. 3, November 2019

Alternative Treatments for Major Depressive Disorder Alexandra Cruz-Mullane1*, Michael A. Freeman1*, Amalie K. Kropp1*, and Shane P. Ruddy1*

Geisinger Commonwealth School of Medicine, Scranton, PA 18509 *Master of Biomedical Sciences Program Correspondence: akropp@som.geisinger.edu 1

Abstract Major depressive disorder (MDD) is a common but serious mood disorder that is not limited to but is often characterized by persistent feelings of sadness, hopelessness, anhedonia, and difficulty sleeping. Millions of people of all ages from all over the world suffer from depression on a regular basis, and many do not respond to first-line treatments. Current mainstay treatments include medications, psychotherapy, or a combination therapy. This traditional approach may not be beneficial for everyone, and therefore, these patients have treatment-resistant depression. Examining different treatment options allows for the patient to have some possible relief of symptoms during the therapeutic lag of traditional pharmacotherapies for depression.

effects of MDD (5). ECT and rTMS are used as second-line treatments due to the side effects of ECT and the limited success of rTMS (6). Despite the wide variety of therapeutic options, there are MDD patients who still show decreased quality of life and risk of suicidal ideation. In this paper, we will review alternative treatment options for MDD that have shown promise. Ketamineâ&#x20AC;&#x2122;s effect on NMDA receptors in the CNS have shown antidepressant effects. Ayahuasca functions as a 5-HT agonist, reducing depression. Similarly, cannabis reduces depressive symptoms, by improving GPCR-NMDA communication and stabilizing neurotransmitter levels. Finally, non-medical alternatives, such as improving sleep, food, and exercise habits, can help alleviate MDD symptoms and may be considered supplemental.

This review examines several alternative treatments for depression. Intranasal ketamine has a high potential to treat patients through its effects on glutamate receptors. Ayahuasca, an ancient herbal brew, has also been found to improve patientsâ&#x20AC;&#x2122; depression based on its psychedelic effects. Cannabis and the endocannabinoid system affect the ERK1/2 pathway, which mimics the effects of other antidepressant drugs, alleviating depressive symptoms. Lastly, non-medical treatments such as sleep, exercise, and diet can help depression, especially in combination with other treatments.

Methods

Introduction

Ketamine

Major depressive disorder (MDD) remains one of the most commonly diagnosed and debilitating mental health disorders in the United States. Globally, it affects more than 300 million individuals (1, 2), and research has demonstrated that up to one-third of patients do not respond to appropriate courses of at least three different antidepressants (3). MDD is characterized by the DSM-5 as a depressive disorder. DSM-5 criteria for MDD includes depressed mood, loss of interest in previously pleasurable activities, changes in weight, insomnia/ hypersomnia, worthlessness, psychomotor retardation, and suicidal ideation (2). Treatment options for MDD include pharmacotherapy, cognitive behavioral therapy (CBT), interpersonal therapy (IPT) and brain stimulation therapies (4, 5). Pharmacological options include selective serotonin reuptake inhibitors (SSRI), serotonin norepinephrine reuptake inhibitors, monoamine oxidase inhibitors (MAOI) and tricyclic antidepressants (TCA). Brain stimulation therapies include electroconvulsive therapy (ECT) and repetitive transcranial magnetic stimulation (rTMS) (5). Traditional first-line treatments for MDD include a combination of pharmacotherapy, CBT, and IPT. Antidepressant drugs are used to primarily modulate neurotransmitter levels in the brain, particularly serotonin (5-HT) and norepinephrine (NE) while CBT and IPT are effective in helping patients with the behavioral and lifestyle

One innovative treatment that has recently been developed for depression is intranasal ketamine. The science and history behind ketamine, its different formulations (especially compared to the current antidepressants), and an important clinical study showing the beneficial use to treat MDD will all be further discussed.

A literature search evaluated current alternative treatments for MDD using Medline/PubMed and Google Scholar. The search terms included major depression, MDD, major depressive disorder, ketamine, intranasal ketamine, ayahuasca, cannabis, marijuana, non-medical, alternatives, lifestyle, diet, and exercise.

Discussion

Ketamine was originally discovered in the 1960s after a failed attempt at producing phencyclidine or PCP (7). It was originally created to be used as a battlefield anesthetic and a sedative agent (8). Organically, there are several structural forms of ketamine. Those structures are the racemic, S, and R enantiomer forms. There are several ways that a patient can be given ketamine. A patient can receive ketamine orally, sublingually, intranasally, subcutaneously, intramuscularly, or intravenously (8). The primary focus will be on the intranasal formulation. Pharmacokinetics and pharmacodynamics are additional aspects to consider. Research has shown that ketamine has a short half-life of only 2 to 4 hours. It has a large volume of distribution and a high clearance rate. The S form is eliminated 4 to 7 hours after administration (8). Ketamine rapidly produces anesthetic and analgesic effects (8). The S-enantiomer form has the greatest effect on the body and depression; it plays a crucial role when interacting with the glutamate or NMDA receptors of the central nervous system

Alternative Treatments for Major Depressive Disorder

(7, 8). This review will focus on this mechanism, which is being used to treat MDD and treatment resistant depression. In March of 2019, intranasal ketamine was approved by the U.S. Food and Drug Administration and the United States Department of Veterans Affairs. Intranasal esketamine demonstrated strong benefits, especially when compared to current depression treatments (9). Current depression treatments include SSRIs, MAOIs, and TCAs. Not only has intranasal ketamine been shown to treat depression, but one study has shown it to improve anxiety, aggression, and other behavioral symptoms (10). Ketamine in a nasal form is more efficacious for the patient because it is more local and thus has better bioavailability, as it penetrates the blood-brain barrier, allowing for faster onset (8). Additionally, studies have shown that the new formulation of intranasal ketamine has numerous benefits for patients with treatment-resistant depression. Unfortunately, there are several adverse side effects associated with ketamine of all formulations. Several studies demonstrated that ketamine has caused its typical adverse effects such as hallucinations. Others have noted nausea and dizziness (8). Moving forward, the focus needs to be on preventing adverse effects, while monitoring the long-term use of intranasal ketamine. The first study to show increased benefits from intranasal ketamine was a randomized control trial for MDD (11). Researchers selected participants from referrals, advertisements and an academic outpatient psychiatric clinic. The criteria for participating was that the patients needed to have at least one failed response to an FDA approved antidepressant medication. Overall, 18 out of 36 participants successfully completed the randomized trial. On average, these patients had dealt with depression for about 27 years and had failed to respond to roughly 4 antidepressant treatments (11). During the trial, the patient was first given ketamine and then a saline placebo, or vice versa. Patients’ depression was measured by the Montgomery-Asberg Depression Rating Scale (MADRS) at 24 hours following the intranasal intervention. Additionally, they were also evaluated 2, 3, and 7 days following each administration (11). According to the MARDS score, the results showed that in 24 hours, the mean difference between ketamine and saline was 7.6±3.7 (11). One could infer that intranasal ketamine was successful in treating depression symptoms. However, there was no significant difference between day 3 and 7 (11). Other secondary outcomes included a longer duration of ketamine benefits, decreased levels of depression and anxiety in self-reports and a higher proportion of responders. Potential psychotomimetic, dissociative, hemodynamic, and general adverse effects associated with ketamine were also measured (11). Overall the results showed that intranasal ketamine experienced rapid improvement of depressive symptoms. The study’s limitations were caused by an augmentation approach between ketamine and current antidepressants, a limited sample size and an inability to correlate ketamine dosages and effects in this study. Additionally, the blindness of the study may have been compromised. The augmentation approach made it hard to identify if ketamine worked well alone or in combination with other antidepressants (11), as participants

were permitted to stay on their current antidepressant medication. The study was limited in the number of participants, so it may not accurately represent the general population. This experimental study was considered a crossover study, which is when a participant gets both treatments. The crossover may have caused the study to not be truly blind, especially if a participant was able to distinguish ketamine from the placebo. The last limitation was that the study had difficulty correlating optimal dosing and determining both the long-term safety and efficacy of ketamine (11). During the trial, only one dose of 50 mg was used, as a result, researchers could not determine the future dose of intranasal ketamine. The researchers were also unable to correlate their results with long-term safety and efficacy and long-term effects, because of its past history of abuse. Ayahuasca Ayahuasca is another option that could help treat MDD, specifically patients with treatment-resistant depression. Ayahuasca is a brew that was customarily used for spiritual and healing purposes (12); it is generally considered to be a psychedelic, originating from the indigenous people of the Amazon Basin (13). The word “ayahuasca” translated means the “vine of the spirits.” It is still used in other parts of the world. It has obtained legal status for rituals in Brazil in 1987. Ayahuasca is prepared using two different plants: Psychotria viridis and Banisteriopsis caapi. The first plant contains N, N-dimethyltryptamine (N, N-DMT), a serotonin and sigma-1 receptor agonist. Sigma-1 receptors are found in the dentate gyrus, thalamus and hypothalamus (14). The second plant comprises of reversible, selective monoamine oxidase inhibitors (such as harmine and harmaline) and tetrahydroharmine (a weak serotonin reuptake inhibitor) (15, 16). Previous animal studies have found that animals given harmine demonstrated reduced immobility time and increased climbing and swimming time in the forced swim test. All results are congruent with antidepressant effects, providing further evidence for ayahuasca’s potential for treatment-resistant depression (17). This chemical mixture produces acute psychological effects that are known to last around 4 hours. Patients may feel intense perceptual, cognitive, emotional, and affective changes. This concoction promotes moderate sympathomimetic effects (16, 18). Less objective sources have found this substance to cause profound mystical or religious experiences, or overall positive life changes. Adverse effects include nausea, vomiting, and diarrhea. Despite these adverse effects, ayahuasca has a generally safe profile. It is noteworthy to consider that MAOIs are involved in the brew, a hypertensive crisis and serotonin syndrome could result, although unlikely (19). The vomiting may or may not be considered a full adverse effect depending on the study. It is considered by some to be part of the purging process. There have been no addictive properties associated with the tonic. Ayahuasca has also been found to work faster than the traditional method of modulating monoamines within the brain (20). One source conducted a rigorous study on the antidepressant effects of ayahuasca in the first ever controlled trial. A parallel-arm, double-blind, randomized placebo-controlled

Alternative Treatments for Major Depressive Disorder

trial with 29 patients with treatment-resistant depression was conducted in Brazil. After 1, 2 and 7 days after dosing, the changes in the patient’s depression were evaluated with both the HAM-D and the MADRS scales. Both depression scales have a good track record of reliable results and are considered the standard. The MADRS is used in addition to the HAM-D scale, as the MADRS tends to be more sensitive to changes caused by antidepressants. There were significant antidepressant efforts at all time periods when compared to the placebo (17). At the day 7 mark, 64% of the patients stated they felt that their depressive symptoms seemed diminished, while only 27% within the placebo group felt that. The authors compared the effectiveness of ayahuasca to the previously stated ketamine findings. Both treatments are associated with rapid antidepressant effects, but their time courses differ, as ketamine has a much faster time course due to route of administration (12). Overall, the study did have some limitations. There were only 17 patients in the study and participation was limited to those only with treatment resistant depression. It was also difficult to maintain the double-blind aspect of the study. The participants in this study were naïve to ayahuasca. The researchers acknowledge the unique effects of psychedelics, but they actually had 5 patients misclassify the placebo for ayahuasca (12). The patients’ depression severity was assessed before, during, and after the treatment using the standard scoring of the HAM-D and the MADRS scales. Despite its limitations, this study shows ayahuasca has promise as an alternative treatment for MDD (12). Other mental health evaluators have assessed the efficacy of ayahuasca as treatment for other aspects of mental well-being. Some of the additional benefits include preserved cognitive function, increased well-being and a reduction in both anxiety and depressive symptoms (21). Within the United States, the legal status and use of ayahuasca is slightly ambiguous. The plant itself is not illegal, but some individuals have been able to create DMT, a Schedule I tryptamine hallucinogen. People have tried to fight this aspect, advocating its use for religious purposes. The U.S. Supreme Court ruled in favor of churches in a unanimous decision to allow the use of the brew as part of the Religious Freedom Act in 2005 (22). They found that there was not sufficient evidence of adverse health risks and there was very little risk that the brew would be abused recreationally. Using this indigenous brew seem may seem unusual, but in patients that have treatment-resistant depression, this treatment could be considered to improve their quality of life significantly. Cannabis Among Schedule I substances, marijuana is among the most commonly used drug in the United States (23). Marijuana contains a psychoactive substance called delta 9-tetrahydrocannabinol (THC), which targets receptors in the endocannabinoid system and can produce altered mental status and behavioral changes. The exact mechanism of action of cannabis is largely unknown. However, there are links between CB1 and CB2 receptors and classical neurotransmitter systems (23). The systems consist of 5-HT, NE, GABA, glutamate, and dopamine (DA). The behavioral effects of THC

include euphoria, relaxation, and an altered ability to learn and concentrate (24). In addition, feelings of paranoia or panic were possible. THC also has numerous physiological changes to include increased heart rate, decreased blood pressure, vasodilation, and increased respiration (25). The most widely known effect of THC usage is a rapid increase in appetite, also known in popular culture as "the munchies." In recent years, there has been growing evidence that the endocannabinoid system may be able to be used for pharmacological treatment of various mental health disorders (23). Some of these disorders include autism, schizophrenia, bipolar disorder, post-traumatic stress disorder, major depressive disorder, and even suicidal ideations (23). It has been suggested that THC may also have analgesic properties (25–27). While the focus of this paper is on alternative treatments for major depressive disorder, it should be noted that cannabis may have potential to treat or alleviate symptoms for a myriad of mental and physiological conditions. In a recent, cohort comparison study, the prevalence of mental health disorders and symptoms in millennials over a 10-year period was examined. They found an increase in major depressive disorder symptoms (9.0–14.8%) and self-harm (11.8–14.4%) (28). MDD comorbidities such as sleep deprivation and high BMI almost doubled. Adolescents that reported less than 8 hours of sleep went from 5.7% to 11.5% and obesity went from 3.8% to 7.3% (28). As cannabis can cause a relaxed feeling, there could be potential in it being used as a sleep aid and reducing anxiety, in addition to its use as an antidepressant. In trying to treat MDD using the endocannabinoid system, there needs to be both a molecular approach and a systemic approach. The reward pathways, which consist of opioid, endocannabinoid, and endovanilloid systems, can produce antidepressant effects (29). Curcumin is an endocannabinoid modulator and was tested on its effects with CB1/CB2 receptors. The drug caused an increased in endocannabinoid activity and in the ERK1/2 pathway, which mimics effects of other antidepressant drugs (29). MDD and physical pain are prevalent comorbidities wherein up to 80% of patients suffering from MDD may experience chronic pain to some degree (26). Previous clinical studies have shown that the stimulation of endocannabinoid receptors decreased physical pain, as well as depressive symptoms (26). A majority of patients with the pain-depression comorbidity do not respond well to first-line pharmacotherapies. As previously stated, the opioid system is interconnected with antidepressant effects, suggesting pain levels can play a role in depression (25,26). Given that cannabis has shown analgesic and antidepressant effects, it could show promise as an alternative therapy for this debilitating combination of diseases (26). At the molecular level, major depressive disorder is connected to varied levels of G-protein coupled receptors (GCPR) and N-Methyl-D-aspartate receptors (NMDAR) (26). As of now, there is no isolated mechanism of action due to THC’s effect on multiple targets at the cellular level to include benzodiazepine receptors, protein metabolism pathway and nucleic acid metabolism (26). The endocannabinoid system is a type of GCPR system. A dysregulation of the endocannabinoid system can produce adverse effects on the GCPR-NMDA

Alternative Treatments for Major Depressive Disorder

communication. This can in turn lead to irregular levels of serotonin, causing MDD. Low levels of endocannabinoid signaling, specifically CB1R, have also been linked to depressive symptoms (28,30,31). As a result, theoretically by stimulating the CB1R receptors, it could be possible to stabilize 5-HT levels and restore effective cross talk between GPCRs and NMDARs (30,32). Despite the positive correlations between MDD and the endocannabinoid system, marijuana remains a Schedule I substance. In addition, its abuse for non-medical purposes has caused a stigma, which has slowed the drug’s acceptance into both normative society and the medical community (29, 32). Cannabis could have adverse effects on patients that are undergoing treatment for depression. Studies reported medical usage of cannabis showed lower cognitive functioning (33). Patients that reported recreational usage had fewer psychiatry visits, increased prevalence of suicidal ideations and also decreased cognitive functioning (34). Marijuana usage, especially for non-medical/recreational usage could potentially harm treatment of MDD (35). Overall, more research is needed to fully understand the endocannabinoid system’s role in systemic depression. However, recent research has shown that endocannabinoids could prove effective as a pharmacotherapeutic target for MDD in the upcoming years (27, 29, 30). “Nonmedical” alternatives No analysis of alternative approaches to treating depression would be complete without an examination of lifestyle approaches, meditation, and exercise, as well as natural remedies and diet in their ability to alleviate depression. Lifestyle medicine, which seeks to examine the individual, his/ her lifestyle and environment as a whole, allows for a deeper understanding of various factors that interplay and influence the individual and his/her health. This approach would benefit the whole population, especially those with depression, as depression is often multifactorial. Lifestyle medicine ideally would look at the patient’s physical and emotional health and take into consideration alternative factors that need to be considered more often, such as alcohol/ substance abuse, sleep and social/recreational life. Alcohol and substance abuse often co-occur with mood disorders; there is actually a two- to three-time lifetime risk for depression and anxiety with substance dependence (36). The substance dependence may also affect treatment outcomes; additionally, drinking, smoking, and inappropriate drug use during teenage years tends to correlate positively with development of mood disorders, especially depression. Sleep is critical to functioning appropriately and disturbances to circadian rhythm have been linked to a slew of medical problems (depression, myocardial infarction, stroke, obesity, diabetes, coronary artery disease) that all possibly and probably affect and exacerbate each other, especially depression (36). Additionally, sleep dysfunction and depression are deeply intertwined—sleep dysfunction may cause depression and depression may cause sleep dysfunction. One’s social and recreational life permeates all other aspects of his/her life. Friends, family, and pets, as well as life-work balance and hobbies, are all vital to consider in depression. Alcohol/substance abuse, sleep and social/

recreational life are all factors to consider when determining the mental and physical health of the patient, but especially so when considering the treatment and cause of depression. Meditation has been examined critically to see if it may be useful for depression and the mindfulness type has been the most investigated: mindfulness-based stress reduction (MBSR) and mindfulness-based cognitive therapy (MBCT). Both are eight-week long programs developed to allow practitioners to understand their normal thought and reaction process, and how to change both. MBSR and MBCT have been shown to increase mindfulness and decrease negative repetitive thinking, by increasing reappraisal, decreasing rumination, and shifting cognitive reactivity (37–39). This also provides alternative methods of dealing with and approaching stress, as well as possibly fostering self-compassion (38, 40). Serum cortisol was also shown to decrease, clinically validating these results (41). However, there were limitations in these studies: they often had a small sample size and there may have been bias as they often used self-reported scales (38). Additionally, the patients often had mild to moderate depression and the results cannot necessarily be extrapolated to severe depression. The National Institute for Health and Clinical Excellence recommends mindfulness and meditation for people with depression (40). Exercise has been found to have beneficial effects for all individuals, including depressed individuals. Physical inactivity has been determined as a risk factor for depression, with an odds ratio of 3.15 for those physically inactive (42). Exercise physically rewires the brain. It increases brain neurogenesis as well as parasympathetic vagal tone and changes the hypothalamic pituitary adrenal axis (43). Physical activity causes the stimulation and release of B-endorphins, vascular endothelial growth factor, brain-derived neurotrophic factor, serotonin, adrenocorticotropic hormone and endocannabinoids, all the while decreasing cortisol levels (42, 43). Along with these structural brain changes, exercise may also change physical ability and health, while increasing self-esteem and self-concept. It may also have social benefits. All of these factors come into play and can reduce depressive symptoms as well as benefit mental health (for example, reducing anxiety). There are some limitations in the studies and reviews examined, some of which include casual factors and some more important cofounds (e.g., genetics, personalities, approaches to exercise). Regardless, daily physical activity benefits everyone: according to the HUNT [Health Study of Nord-Trondelay County] Cohort Study (43), up to 12% of future cases of depression could be avoided if only 1 hour of physical activity a week was completed. Additionally, several studies and a Cochrane meta-analysis determined that exercise is comparable to antidepressants as a first-line treatment for mild to moderate depression (42). Diet and natural remedies/supplements may very well also have very important roles in health. Food is fuel for the body and is associated directly with mental health (36). The Western diet and lifestyle are both gaining traction worldwide, and along the way, both encourage some unhealthy perspectives and values. The Western diet promotes rushing through meals that may lack nutrients, protein, fatty acids, and micronutrients, often

Alternative Treatments for Major Depressive Disorder

causing one to feel hungry again soon after. More attention to what one is eating and incorporating healthy foods, as well as “natural” remedies into daily habit will only bring benefits. Some supplements to consider include Camellia sinensis, resveratrol, anthocyanidins, omega 3 fatty acids, vitamin B and Hypericum perforatum. Camellia sinensis, otherwise known as tea, has many benefits; higher consumption of green tea, specifically, has been strongly and negatively correlated with incidents of depression (44), and it may be due to its inhibitory effects on MAO activity and COX. Resveratrol, found in grape skins, increases 5-HT and NE levels, and has mild analgesic effects (44). Anthocyanidins, a group of plant pigments that includes berry anthocyanins and the hibiscus flower, inhibit MOA activity and cause changes in the dopaminergic, noradrenergic, and serotonergic systems (44). Both omega-3 fatty acids and vitamin B deficiencies have been implicated to have a positive correlation with a risk of developing depression (44). Hypericum perforatum, also known as St. John’s wort, has been used for mild to moderate depression; it works mainly through its SSRI effects, but has also been found to have MAOI and COMT effects (45). According to a meta-analysis and systematic review, there were no differences in the group taking clinically prescribed medications and the group using alternative approaches (acupuncture, omega-3 fatty acids, S-adenosyl methionine, St. John’s wort, and exercise) in response and remission. The only difference found is that the discontinuation of SSRIs (for example) results in more severe side effects than discontinuing with the alternative approaches (46). Therefore, changes in diet and incorporating some of these compounds into daily routine would benefit the entire population; for depressed individuals, using this approach to supplement their current medications may be something to consider as well.

ketamine, has high potential as an antidepressant, especially for treatment-resistant depression, through its effects on glutamate receptors. Similarly, ayahuasca benefits those with treatment resistant depression—it has serotonin and sigma receptor agonist effects and acts similarly to MAOIs and SSRIs. Cannabis’ mechanism of action remains largely unknown but is related to a multitude of pathways and systems within the brain. It has many benefits and may be used as a therapeutic for various mental health disorders. It may not be as beneficial for depression; however, there is the question if those who use marijuana were already at higher risk for depression or if marijuana exacerbates depressive symptoms. “Nonmedical” alternatives, including lifestyle approaches, meditation, exercise, and natural remedies/diet, are all effective for depression and benefit all individuals. Most of these nonmedical alternatives work by increasing neurogenesis, releasing hormones and neuropeptides and allowing the brain an escape from work, anxiety, and rumination. As previously mentioned, depression is often multifactorial and the treatments should be as well. Until further evidence is provided proving these medications and approaches to be harmful or beneficial, they should be considered supplemental.

References 1.

Depression [Internet]. World Health Organization. World Health Organization; c2019 [cited 2019Apr4]. Available from: https://www.who.int/news-room/fact-sheets/detail/ depression

2. American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 5th ed. Washington DC:2013.

Health is often complex, having a cause-and-effect relationship between many factors. Future approaches in medicine should take a “multifactorial” view. Lifestyle modifications, including alcohol/substance dependence, sleep, social/recreational life, meditation, exercise, and alternative approaches benefit all individuals and should be part of treatment and prevention of disease, especially mood disorders and specifically depression. Despite the benefits found in the studies and reviews (36, 40, 42), more extensive research should be done to evaluate the broad applications of these complementary and alternative approaches toward all severities of depression, as well as to replicate the results previously obtained. Ultimately, all of the lifestyle modifications mentioned are not very expensive and would be fairly easy to incorporate into daily life. They have a multitude of advantages that would benefit the entire population; for now, they should be considered supplemental, not first-line treatments for depression.

3. Callaway JC, McKenna DJ, Grob CS, Brito GS, Raymon LP, Poland RE, et al. Pharmacokinetics of Hoasca alkaloids in healthy humans. J Ethnopharmacol. 1999; 65: 243–256.

Conclusion

8. Andrade C. Ketamine for depression, 4: In what dose, at what rate, by what route, for how long, and at what frequency? J Clin Psychiatry. 2017;78(7)

Depression remains a growing health concern for a large proportion of the population. Many current pharmacological treatments have limited efficacy; therefore, new approaches are needed. The alternative approaches mentioned in this paper may ultimately be the focus of future treatment, however, more extensive research needs to be done to evaluate these approaches in comparison with current FDA-approved treatments. Ketamine, specifically intranasal

4. Depression [Internet]. NAMI; c2019 [cited 2019Apr4]. Available from: https://www.nami.org/Learn-More/MentalHealth-Conditions/Depression/Treatment 5. Otte C, Gold SM, Penninx BW, Pariante CM, Etkin A, Fava M, et al. Major depressive disorder. Nat Rev Dis Primers. 2016;2:16065. 6. Mion G, Villevieille T. Ketamine pharmacology: An update (pharmacodynamics and molecular aspects, recent findings). CNS Neurosci Ther. 2013;19(6):370–80. 7.

Peltoniemi MA, Hagelberg NM, Olkkola KT, Saari TI. Ketamine: A review of clinical pharmacokinetics and pharmacodynamics in anesthesia and pain therapy. Clin Pharmacokinet. 2016;55(9):1059–77.

9. Andrade C. Intranasal drug delivery in neuropsychiatry: Focus on intranasal ketamine for refractory depression. J Clin Psychiatry. 2015;76(05). 10. Jankauskas V, Necyk C, Chue J, Chue P. A review of ketamine’s role in ECT and non-ECT settings. Neuropsychiatr Dis Treat. 2018;14:1437–50.

Alternative Treatments for Major Depressive Disorder

11. Lapidus KA, Levitch CF, Perez AM, et al. A randomized controlled trial of intranasal ketamine in major depressive disorder. Biol Psychiatry. 2014;76(12):970–976. doi:10.1016/j. biopsych.2014.03.026

25. Fitzgibbon M, Finn DP, Roche M. High times for painful blues: The endocannabinoid system in pain-depression comorbidity. Int J Neuropsychopharmacol. 2015;19(3): 1–20.

12. Palhano-Fontes F, Barreto D, Onias H, Andrade KC, Novaes MM, Pessoa JA, et al. Rapid antidepressant effects of the psychedelic ayahuasca in treatment-resistant depression: a randomized placebo-controlled trial. Psychol Med. 2018;49(04):655–63.

26. Sharma P, Murthy P, Bharath MM. Chemistry, metabolism, and toxicology of cannabis: clinical implications. Iran J Psychiatry. 2012;7(4):149–156.

13. Luna LE. Indigenous and mestizo use of ayahuasca: An overview. In: Guimarães dos Santos R, editor. The Ethnopharmacology of Ayahuasca. Kerala, India: Transworld Research Network; 2011. p. 1–25. 14. Cai S, Huang S, Hao W. New hypothesis and treatment targets of depression: an integrated view of key findings. Neurosci Bull. 2015;31(1):61–74. 15. Carbonaro TM, Gatch MB. Neuropharmacology of N,Ndimethyltryptamine. Brain Res Bull. 2016;126:74–88. 16. Riba J. Human pharmacology of ayahuasca: Subjective and cardiovascular effects, monoamine metabolite excretion, and pharmacokinetics. J Pharmacol Exp Ther. 2003;306(1):73–83. 17. Fortunato JJ, Réus GZ, Kirsch TR, Stringari RB, Fries GR, Kapczinski F, et al. Chronic administration of harmine elicits antidepressant-like effects and increases BDNF levels in rat hippocampus. J Neural Transm. 2010;117(10):1131–7. 18. Santos RGD, Valle M, Bouso JC, Nomdedéu JF, Rodríguez-Espinosa J, Mcilhenny EH, et al. Autonomic, neuroendocrine, and immunological effects of ayahuasca. J Clin Psychopharmacol. 2011;31(6):717–26. 19. Callaway JC, Grob CS. Ayahuasca preparations and serotonin reuptake inhibitors: A potential combination for severe adverse interactions. J Psychoactive Drugs. 1998;30:367–9. doi:10.1080/02791072.1998.10399712. 20. Otte C, Gold SM, Penninx BW, Pariante CM, Etkin A, Fava M, et al. Major depressive disorder. Nat Rev Dis Primers. 2016;2:16065. 21. Barbosa PCR, Strassman RJ, Silveira DXD, Areco K, Hoy R, Pommy J, et al. Psychological and neuropsychological assessment of regular hoasca users. Compr Psychiatry. 2016;71:95–105. 22. Rainey JG. Gonzales v. O Centro Espirita Beneficente Uniao Do Vegetal [Internet]. Gonzales v. O Centro Espirita Beneficente Uniao Do Vegetal; c2019 [cited 2019Apr4]. Available from: https://www.mtsu.edu/first-amendment/ article/745/gonzales-v-o-centro-espirita-beneficente-uniaodo-vegetal 23. Rubino T, Zamberletti E, Parolaro D. Endocannabinoids and mental disorders. Handbook of Experimental Pharmacology Endocannabinoids. 2015:261–83. 24. Koby C, Abraham W, Aviv W. Modulatory effects of cannabinoids on brain neurotransmission. Eur J Neurosci. 2019; in press, doi: 10.1111/ejn.14407.

27. Huang W-J, Chen W-W, Zhang X. Endocannabinoid system: Role in depression, reward and pain control (review). Mol Med Rep. 2016;14(4):2899–903. 28. Patalay P, Gage S. Trends in millennial adolescent mental health and health related behaviours over ten years: a population cohort comparison study. bioRxivorg. 2018; 1–22. doi:10.1101/407585. 29. Chávez-Castillo M, Núñez V, Nava M, Ortega Á, Rojas M, Bermúdez V, et al. Depression as a neuroendocrine disorder: Emerging neuropsychopharmacological approaches beyond monoamines. Adv Pharmacol Sci. 2019;2019:1–20. 30. Pacher P. The endocannabinoid system as an emerging target of pharmacotherapy. Pharmacogn Rev. 2006;58(3):389–462. 31. Smaga I, Bystrowska B, Gawlinski D, Przegalinski E, Filip M. The endocannabinoid/endovanilloid system and depression. Curr Neuropharmacol. 2014;12(5):462–74. 32. Rodríguez-Muñoz M, Sánchez-Blázquez P, Callado LF, Meana JJ, Garzón-Niño J. Schizophrenia and depression, two poles of endocannabinoid system deregulation. Transl Psychiatry. 2017; 7(12):1–12. 33. Bahorik AL, Sterling SA, Campbell CI, Weisner C, Ramo D, Satre DD. Medical and non-medical marijuana use in depression: Longitudinal associations with suicidal ideation, everyday functioning, and psychiatry service utilization. J Affect Disord. 2018;241:8–14. 34. Bassi MS, Gilio L, Maffei P, Dolcetti E, Bruno A, Buttari F, et al. Exploiting the multifaceted effects of cannabinoids on mood to boost their therapeutic use against anxiety and depression. Front Mol Neurosci. 2018;11. 35. Hathaway AD, Comeau NC, Erickson PG. Cannabis normalization and stigma: Contemporary practices of moral regulation. Criminol Crim Justice. 2011;11(5):451–69. 36. Sarris J, O’Neil A, Coulson CE, Schweitzer I, Berk M. Lifestyle medicine for depression. BMC Psychiatry. 2014;14(1). 37. Parmentier FBR, García-Toro M, García-Campayo J, Yañez AM, Andrés P, Gili M. Mindfulness and symptoms of depression and anxiety in the general population: The mediating roles of worry, rumination, reappraisal and suppression. Front Psychol. 2019;10:1–10. 38. Takahashi T, Sugiyama F, Kikai T, Kawashima I, Guan S, Oguchi M, et al. Changes in depression and anxiety through mindfulness group therapy in Japan: the role of mindfulness and self-compassion as possible mediators. Biopsychosoc Med. 2019;13(1):1–10.

Alternative Treatments for Major Depressive Disorder

39. Winnebeck E, Fissler M, Gärtner M, Chadwick P, Barnhofer T. Brief training in mindfulness meditation reduces symptoms in patients with a chronic or recurrent lifetime history of depression: A randomized controlled study. Behav Res Ther. 2017;99:124–30. doi:10.31231/osf.io/jy3gs. 40. Solem S, Hagen R, Wang CEA, Hjemdal O, Waterloo K, Eisemann M, et al. Metacognitions and mindful attention awareness in depression: A comparison of currently depressed, previously depressed and never depressed individuals. Clin Psychol Psychother. 2015;24(1):94–102. 41. Daya Z, Hearn JH. Mindfulness interventions in medical education: A systematic review of their impact on medical student stress, depression, fatigue and burnout. Med Teach. 2017;40(2):146–53. 42. Harvey SB, Øverland S, Hatch SL, Wessely S, Mykletun A, Hotopf M. Exercise and the prevention of depression: Results of the hunt cohort study. Am J Psychiatry. 2018;175(1):28–36. 43. Carek PJ, Laibstain SE, Carek SM. exercise for the treatment of depression and anxiety. Int J Psychiatry Med. 2011;41(1):15–28. 44. Nabavi SM, Daglia M, Braidy N, Nabavi SF. Natural products, micronutrients, and nutraceuticals for the treatment of depression: A short review. Nutr Neurosci. 2015;20(3):180–94. 45. Micozzi MS, Koop CE, Lundberg GD. Fundamentals of complementary and alternative medicine. St. Louis, MO: Elsevier; 2018. 46. Asher GN, Gartlehner G, Gaynes BN, Amick HR, Forneris C, Morgan LC, et al. Comparative benefits and harms of complementary and alternative medicine therapies for initial treatment of major depressive disorder: systematic review and meta-analysis. J Altern Complement Med. 2017;23(12):907–19.

Scholarly Research In Progress • Vol. 3, November 2019

Use of Venous Blood Gases for Management of Acid-Base Status in Patients with Severe Septic Shock Garrison Davis1,2†, Marc Incitti1†, Spencer Davis2, and Khaled Sorour2,3

Geisinger Commonwealth School of Medicine, Scranton, PA 18509 Signature Healthcare Brockton Hospital, Brockton, MA 02302 3 Harvard Medical School, Boston, MA 02115 †Doctor of Medicine Program Correspondence: mincitti@som.geisinger.edu 1

Abstract Arterial punctures are painful and may be difficult to obtain in many patient populations. Generally, venous punctures are significantly easier to obtain and less painful to perform. Obtaining a venous sample can be made even easier with placement and utilization of an intravenous line. The goal of this study was to determine the relationship between venous and arterial blood gases in patients with severe septic shock. Venous and arterial samples were collected simultaneously. Our results showed a poor agreement in pCO2, pH, and _ _ HCO3 , although the mean difference in HCO3 was marginal, and the clinical significance questionable. The results of this report confirm the unreliability of venous blood gases alone for evaluating acid base status in this patient population. Future studies should focus on veno-arterial CO2 gradients in estimating cardiac output. Evaluating CO2 gradients in different stages of septic shock may have prognostic value in this patient population.

Introduction Venous blood gases (VBGs) have become an invaluable tool for rapidly assessing ventilation and tissue oxygenation in critically ill patients. Although arterial blood gases (ABGs) are traditionally used to evaluate acid-base status and oxygenation in critically ill patients, prolonged complications of arterial catheterization include infection, hemorrhaging, nerve and vascular injury, and even limb loss (8–11). In many cases, VBGs have been used as a more convenient and safer alternative to ABGs. The accuracy of VBGs has been well established in a variety of patient populations (1–7). Specifically, one recent report found that adjusting the VBG pH up by 0.05 and PCO2 down by 5 mm Hg improves agreement between venous and arterial blood gases (12). Increases in veno-arterial CO2 gradient (dpCO2) have been reported in several conditions, often related to impairment in cardiac output, including cardiogenic, hypovolemic and septic shock (13–15). Few studies have evaluated the clinical usefulness and reliability of VBGs in evaluating acid-base status in patients with septic shock, and no studies, to our knowledge, have evaluated the safety or efficacy of protocols regarding the use of VBGs for evaluating acid-base status in this patient population. In this report, we evaluated the use of venous blood gases in patients with severe septic shock. We hypothesized that venous blood gas data may not be a reliable indicator of acidbase status in this patient population. We believe it is pertinent to obtain arterial blood gases early in order to effectively manage the care of these critically ill patients. The Signature 72

Healthcare Brockton Hospital Institutional Review Board approved the preparation and submission of this report.

Materials and Methods A retrospective cohort of 11 adult patients (8 male, 3 female) diagnosed with septic shock were reviewed. A diagnosis of septic shock required the following: presence of two of the four systemic inflammatory response syndrome (SIRS) criteria, a source of infection and hypotension. The four SIRS criteria were defined as tachycardia (heart rate >90 beats/ min), tachypnea (respiratory rate >20 breaths/min), fever or hypothermia (temperature >38 or <36°C), and leukocytosis, leukopenia, or bandemia (white blood cells >1,200/mm3, <4,000/mm3 or bandemia ≥10%) (16). Hypotension was defined as a systolic blood pressure (SBP) less than 90 mm Hg, a mean arterial pressure (MAP) less than 60, or an SBP decrease greater than 40 mm Hg below baseline (17) that failed to respond to 30 ml/kg bolus of crystalloids delivered over 60 minutes or treated with norepinephrine within 10 minutes to maintain a MAP of 65 mm Hg. All venous blood gases were drawn from a central line. Arterial blood gases were drawn from an arterial line or radial artery. Both arterial and venous blood gases were drawn within 10 minutes of each other and compared. No changes were made in mechanical ventilator settings, vasopressors, inotropes, or alkalizing agents during the time between when the arterial and venous samples were collected. Blood gases were analyzed using the ABL80 FLEX blood gas analyzer (Radiometer, Copenhagen, Denmark). All data was stored and analyzed using R Statistical Software version 3.3.3 (Foundation for Statistical Computing, Vienna, Austria). Statistical analysis was performed using the paired, two sample t-test or Wilcoxon signed rank test where appropriate. The adjusted venous blood gas values were compared to arterial blood gas values. Venous blood gases were altered by adjusting the VBG pH up by 0.05, and pCO2 down by 5 mm Hg to account for the standard difference. _ Veno-arterial differences in HCO3 were not adjusted. A p-value of less than 0.05 was considered statistically significant.

Results We collected data from 11 patients (6 male, 5 female), ranging in age from 64 to 86 years (mean 75.3±6.9). Of these patients, 9 were invasively mechanically ventilated, and 1 was noninvasively ventilated. There was a statistically significant difference between the venous pCO2 (43.7±9.0 mm Hg) and the arterial pCO2 (30.4±10.1 mm Hg) with a mean difference

Use of Venous Blood Gases for Management of Acid-Base Status in Patients with Severe Septic Shock

of 13.3(± 4.1) mm Hg (Figure 1). The mean values for the arterial and venous pH also differed significantly (P <0.05) with the arterial (7.32 ± 0.07) being higher than the venous (7.22 ± 0.05), and a mean difference of 0.10 (± 0.039) (Figure 2). There was also a statistically significant _ difference for the HCO3 . The venous _ HCO3 (17.2 ± 2.4 mEq/L) was slightly higher than the arterial (15.6 ± 3.4 mEq/L), with a mean difference of 1.7(± 1.5 mEq/L) (Figure 3).

Discussion Figure 1. Mean arterial and venous blood gas pCO2 levels. P<0.05 for arterial vs mathematically corrected venous level.

Figure 2. Mean arterial and venous blood gas pH. P<0.05 for arterial vs mathematically corrected venous level.

Figure 3. Mean arterial and venous blood gas bicarbonate ion levels. P<0.05 for arterial vs venous level.

Our results validate the findings from Bakker et al. demonstrating that septic shock is associated with increases in the veno-arterial carbon dioxide levels. (28) The dpCO2 in our patients (13.3 ± 4.1 mm Hg) and mean difference of 0.10 (± 0.039) is likely clinically significant in the management of acid-base status in _ these patients. The difference in HCO3 was small, yet statistically significant. This may have resulted from outliers in a small sample size. Nevertheless, it is questionable whether this difference would be clinically significant. In clinical practice, obtaining venous blood gases for patients with severe septic shock may be particularly lucrative. In these patients it may be particularly difficult to obtain arterial access. Additionally, many of these patients may be unstable, requiring frequent blood gas samples. Veno-arterial CO2 gradients result from the production and clearance of carbon dioxide. Typically, this gradient will not exceed 6 mm Hg (18, 19). In septic shock an increase in metabolic activity from the inflammatory response and anaerobic respiration with resultant lactic acidosis leads to an increase in respiratory workload (21). In this low-flow state there is an increased ventilation-toperfusion ratio. The increase in venous accumulation of CO2 and increased pCO2 elimination through ventilation may lead to the observed increased dpCO2. Several studies have reported increased dpCO2 in other low-flow states, including cardiac arrest, severe cardiac failure, and hemorrhagic shock (22–24). The apparent increase in dpCO2 can be analyzed by applying the Fick principle, which describes the mathematical relationship between carbon dioxide production, veno-arterial carbon dioxide

Use of Venous Blood Gases for Management of Acid-Base Status in Patients with Severe Septic Shock

gradient, and cardiac output. Fick’s principle states that cardiac output is equal to oxygen consumption divided by the difference between the inspired and expired O2. There is an inverse relationship between cardiac output and veno-arterial carbon dioxide differences. Specifically, the relationship has been described as being an inverse curvilinear relationship (20). Although increased dpCO2 in patients with septic shock can be associated with the decrease in cardiac output, septic shock is often considered a hyperdynamic state. In these patients, this phenomenon may still be observed due to additional factors. One particular study found that venous hypercarbia was correlated with systemic hypoperfusion in severe sepsis (21), potentially contributing to a widening dpCO2. Unsurprisingly, the dpCO2 has since drawn practitioners’ attention for its potential prognostic value. One recent study found that a high dpCO2 during the early resuscitation of septic shock was associated with more severe multi-organ dysfunction and worse outcomes at day 28 (20). Another study found that dpCO2 was a predictor of mortality only in the nonventilated patient group (27). The dpCO2 may be valuable in predicting mortality in patients with septic shock. Our study had limitations which must be addressed. Patients were not recruited randomly, but rather enrolled based on convenience and availability. Additionally, the sample size was small. Although all patients were on relatively high doses of vasoactive agents, the exact concentrations of administered agents were not recorded. We also did not discriminate based on the stage of septic shock, making it difficult to determine whether a difference in dpCO2 levels existed in the early and/or late stages based on our data. Finally, there were no recorded cardiac output measurements used in these patients; we were not able to evaluate the exact relationship between cardiac output and dpCO2. In summary, venous blood gases alone may be insufficient in the evaluation and management of acid-base status in patients with septic shock. Studies should be completed with larger sample sizes in order to further validate our findings and determine quantitative criteria to predict a widening dpCO2. Furthermore, future studies should also evaluate the different stages of septic shock for the presence of this phenomenon.

References 1.

Kelly AM, McAlpine R, Kyle E. Agreement between bicarbonate measured on arterial and venous blood gases. Emerg Med Australas 2004. 16(5–6)407–409.

2. Kelly AM, McAlpine R, Kyle E. Venous pH can safely replace arterial pH in the initial evaluation of patients in the emergency department. Emerg Med J 2001. 18340–342. 3. Ma OJ, Rush MD, Godfrey MM, et al. Arterial blood gas results rarely influence emergency physician management of patients with suspected diabetic ketoacidosis. Acad Emerg Med 2003. 10836–841. 4. Gokel Y, Paydas S, Koseoglu Z, et al. Comparison of blood gas and acid-base measurements in arterial and venous blood samples in patients with uremic acidosis and diabetic ketoacidosis in the emergency room. Am J Nephrol 2000. 20319–323. 74

5. Kelly AM, Kyle E, McAlpine R. Venous pCO2 and pH can be used to screen for significant hypercarbia in emergency patients with acute respiratory disease. J Emerg Med 2002. 2215–19. 6. Middleton P, Kelly AM, Brown J, et al. Agreement between arterial and central venous values for pH, bicarbonate, base excess, and lactate. Emerg Med J 2006. 23622–624. 7.

Chu YC, Chen CZ, Lee CH, et al. Prediction of arterial blood gas values from venous blood gas values in patients with acute respiratory failure receiving mechanical ventilation. J Formos Med Assoc 2003;102(8):539–543.

8. Frezza EE, Mezghebe H. lndications and complications of arterial catheter use in surgical or medical intensive care units: analysis of 4932 patients. Am Surg. 1998;64(2):127– 131. 9. Lorente L, Santacreau R, Martin MM, Jiménez A, Mora ML. Arterial catheter-related infection of 2,949 catheters. Crit Care. 2006; 10(3):R83. 10. Scheer B, Perel A, Pfeiffer UJ. Clinical review: complications and risk factors or peripheral arterial catheters used for hemodynamic monitoring in anaesthesia and intensive care medicine. Crit Care. 2002;6(3):199–204. 11. Valentine RJ, Modrall JG, Clagett GP. Hand ischemia after radial artery cannulation. J Am Coll Surg. 2005;201(l):18– 22. 12. Walkey AJ, Farber HW, O'Donnell C, Cabral H, Eagan JS, Philippides GJ. The accuracy of the central venous blood gas for acid-base monitoring. J Intensive Care Med. 2010;25(2)104–110. 13. Mecher CE, Rackow EC, Astiz ME, Weil MH. Venous hypercarbia associated with severe sepsis and systemic hypoperfusion. Crit Care Med 1990, 18: 585–589. 14. Adrogue HJ, Rashad MN, Gorin AB, Yacoub J, Madias NE. Assessing acid–base status in circulatory failure. Differences between arterial and central venous blood. N Engl J Med 1989, 320: 1312–1316. 15. Kazarian KK, Del Guercio LR. The use of mixed venous blood gas determinations in traumatic shock. Ann Emerg Med 1980, 9: 179–182. 16. Bone RC, Balk RA, Cerra FB, Dellinger RP, Fein AM, Knaus WA, Schein RM, Sibbald WJ. The ACCP/SCCM Consensus Conference Committee. American College of Chest Physicians/Society of Critical Care Medicine Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Chest. 1992 17. Levy MM, Fink MP, Marshall JC, Abraham E, Angus D, Cook D, Cohen J, Opal SM, Vincent JL, Ramsay G. SCCM/ ESICM/ACCP/ATS/SIS 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Crit Care Med. 2003;31:1250–6. 18. Steinberg JJ, Harken, AH. The central venous catheter in the assay of acid base status. Surg Gynecol Obstet. 1981: 152:221–22.

Use of Venous Blood Gases for Management of Acid-Base Status in Patients with Severe Septic Shock

19. Guyton AC, Transport of oxygen and carbon dioxide in the blood and bodily fluids. In: Guyton AC, ed. Textbook of medical physiology. Philadelphia: WB Saunders Co. 1981: 504–15. 20. Ospina-Tascon GA, Bautista-Rincon DF, Umaña M, Tafur JD, Gutierrez A, Garcia AF, et al. Persistently high venous-to-arterial carbon dioxide differences during early resuscitation are associated with poor outcomes in septic shock. Crit Care. 2013;17:R294. 21. Mecher CE, Backow EC, Astiz ME, Weil MH. Venous hyper-carbia associated with severe sepsis and systemic hypoperfusion. Crit Care Med 1990; 18:585–89 22. Weil MH, Rackow EC, Trevino R, Grundler W, Falk JL, Criffel MI. Difference in acid-base state between venous and arterial blood during cardiopulmonary resuscitation. N Engl J Med 1986; 315:153–55. 23. Adrogue HJ, Rashad MN, Gorin AB, Yacoub J, Madias NE. Assessing acid-base status in circulatory failure. N EngI J Med 1989; 320:1312–16 24. Williams KB, Christmas AB, Heniford BT, Sing RF, Messick J. Arterial vs venous blood gas differences during hemorrhagic shock. World J Crit Care Med. 2014;3(2):55– 60. 25. Mecher CE, Backow EC, Astiz ME, Weil MH. Venous hypercarbia associated with severe sepsis and systemic hypoperfusion. Crit Care Med 1990; 18:585–89 26. Cuschieri J, Rivers EP, Donnino MW, Katilius M, Jacobsen G, Nguyen HB, et al. Central venous-arterial carbon dioxide difference as an indicator of cardiac index. Intensive Care Med. 2005;31:818–22. 27. Troskot R, Šimurina T, Žižak M, Majstorović K, Marinac I, Šutić IM. Prognostic Value of Venoarterial Carbon Dioxide Gradient in Patients with Severe Sepsis and Septic Shock. Croat Med J. 2010;51(6):501–508. 28. Bakker JA, Vincent JL, Gris P, Leon M, Coffernils M, Kahn RJ. Veno-arterial Carbon Dioxide Gradient in Human Septic Shock. Chest. 1992;101(2):509–515.

Scholarly Research In Progress â&#x20AC;˘ Vol. 3, November 2019

A Review of the Effect of Maternal Obesity on Maternal and Fetal Health Oluwaseyi Olulana1*

Geisinger Commonwealth School of Medicine, Scranton, PA 18509 *Master of Biomedical Sciences Program Correspondence: oolulana@som.geisinger.edu 1

Abstract

Methods

Obesity is a growing epidemic, and it has been at the root of the deteriorating public health in western society. Specifically, obesity is increased in minority populations, especially those with low socioeconomic status. Studies have determined that obesity during pregnancy can lead to maternal and fetal diabetes, hypertension, heart disease, and cognitive disorders such as attention deficit hyperactive disorder (ADHD), stress, and mood disorders. Additionally, maternal obesity increases the risk of preeclampsia and stillbirth due to the function of the placenta in fetal development. Research shows that there are genetic and pathophysiological changes that explain the increase in metabolic disorders in both the mother and child during pregnancy. There are nutritional recommendations, exercise plans, and breastfeeding duration plans that can reduce the risks of these diseases for the mother and her offspring. Physicians should educate their patients accordingly before, during, and after pregnancy. Prenatal counseling can inform patients of the risks before pregnancy, so that fetal health and development is not compromised. A study of reported research on this topic will highlight the effects of maternal obesity and contribute to the discussion on potential solutions to this health care problem.

To assess the impact of maternal obesity on fetal and maternal health postpartum, a literature review was conducted using the PubMed database from 2000 through 2018. Primary research articles and literature reviews on this topic were used as sources and cited accordingly.

Introduction Thirty-four percent of adult and 20 percent of children in the United States suffer from obesity (1). Due to the complicated nature of obesity, it has become one of the most challenging public health issues in western society and the world at large (1). There is growing concern with how obesity affects maternal and fetal health. Although obesity may present as a culmination of personal choices, there are social factors that contribute to this issue, such as socioeconomic disparities. However, it remains that nutritional decisions during pregnancy can dictate the quality of life for the mother and child. Studies have reported that obesity during pregnancy can lead to maternal and fetal diabetes, hypertension, and heart disease (2). Studies have also shown that obesity in the mother can lead to pregnancy complications such as ectopic pregnancies, premature delivery, cognitive impairment, and cesarean sections (2). Research has demonstrated a correlation between maternal obesity, body mass index (BMI), gestational weight gain, and socioeconomic status to the admissions of infants to the intensive care unit, long hospital stays for the infant, and premature birth (3). This review will assess the impact of maternal obesity on fetal and maternal health postpartum. In order to accomplish this goal, an analysis of the current reported research on this topic will highlight the effects of maternal obesity and contribute to the discussion on possible solutions. 76

Discussion Epidemiology Obesity is defined by BMI. The BMI is calculated as weight (in kilograms) divided by height squared (in meters). An individual with a healthy BMI range is between 18.5 and 25; an obese individual has a BMI greater than 30 (1). There are data showing that the increase in obesity in the United States is due to multiple factors (1). Researchers have reported that people from certain ethnic groups and income levels exhibit a marked increase in obesity (1). African-American women and Mexican-American women have a higher obesity occurrence rate than Caucasian women (1). Analyses demonstrate that income status is influenced by obesity levels. Cheap food, containing high levels of fats and sugar, is more accessible to low-income populations (1). Low-income families are also more likely to encourage children to stay indoors instead of getting active in their neighborhood, and a sedentary lifestyle is more ideal within these economic groups (4). The sedentary lifestyle could potentially make it difficult for the child to learn to be healthy later in life. This cycle starts from birth and is propelled by other life circumstances. Researchers have found that maternal obesity increases the risk of offspring obesity. Moreover, pre-pregnancy obesity effects are more potent than gestational weight gain or postpartum weight retention (5). Congenital complications The prevalence of obesity among the female population of minority groups is increasing (1). Pregnancy is traditionally accepted as the time for women to gain weight. Research shows, however, that the BMI of the mother before pregnancy determines the recommended weight gain. Women are expected to gain 11 to 20 pounds while pregnant; however, research has found that greater than 50 percent of women who are obese supersede their recommended gestational weight (6). Obese mothers put their fetuses at higher risk of developing different diseases. In pregnancy, the growth of the fetus is dependent on the nutrients that are in the motherâ&#x20AC;&#x2122;s blood circulation and are transported through the placenta (7). The mother supports the fetal growth with glucose, amino acids, and fatty acids (7). There are scientific data and analyses that show that BMI, weight gain, exercise, and smoking can influence fetal growth and pregnancy outcome (7). The

Effect of Maternal Obesity on Maternal and Fetal Health

mechanism of this change is poorly understood. It is postulated, however, that high levels of lipids in the placenta increase the levels of macrophages; reduce placental blood flow, which results in hypoxia; increase placental inflammation; and deplete nutrient transport (8). The same pathway through which the mother provides nutrients to the fetus provides a passage for detrimental nutrients that can lead to developmental complications for the fetus. It has also been shown that a highfat diet during pregnancy increases placental inflammation and reduces uteroplacental perfusion (9). Maternal obesity in conjunction with a high-fat diet leads to degraded function of the placenta such as placental ischemia and decreased placental perfusion; further, high-fat diet is associated with increased proinflammatory mediators in the placenta (9). These findings confirm the possibility that high-fat diet and obesity during pregnancy increases the risk of stillbirths (9).

Maternal obesity poses increased risk for metabolic disorders. In a recent study, researchers found that adipose tissue macrophages in obese mice secreted miRNA exosomes that caused glucose intolerance and insulin resistance after the exosomes were injected into lean mice (14). This is evidence that increased adiposity is significant in the development of metabolic disorders. In another study, researchers found that diet-induced obesity caused a change in the activation states of adipose tissue macrophages from a state that protects the adipocytes from inflammation to a proinflammatory state leading to insulin resistance (15). These studies indicate that metabolic abnormalities are significantly increased by obesity. Additionally, the adipose tissue macrophage disorder can be linked to the development of diabetes.

It is reported that maternal obesity increases the risk of preeclampsia threefold (10). Preeclampsia is a pregnancyspecific disorder that presents as hypertension and increased protein levels in the urine (11). Researchers have reported the possibility that an increase in asymmetric dimethylarginine (ADMA) correlates with the increased risk of preeclampsia in pregnant women (10). ADMA inhibits the conversion of nitric oxide by nitric oxide synthase. The increased levels of ADMA lead to an increase in oxidative stress (10). ADMA concentration is regulated by the degradation enzyme dimethylarginine dimethylaminohydrolase (DDAH) (10). Although the increased levels of ADMA might be due to the inhibition of the degradation enzyme DDAH, the reasons for the increased levels of ADMA in pre-eclampsia patients are still under investigation (10).

The impact of maternal obesity is not limited to cardiovascular or metabolic disorders. The possibility of a cognitive impairment has also been associated with obesity. Researchers have found that male offspring with high-fat diet and maternal obesity develop spatial cognitive deficiency (16). They found that the cognitive deficiency could be due to the interaction of the fatty acids and metabolite pathways in the prefrontal cortex and hippocampus (16). In another study, researchers exposed rat dams, which are pregnant female rats, to a saturated fat diet before and during their pregnancy (17). The study design simulated maternal obesity before and during pregnancy. The researchers found that microglial activation markers were increased in the hippocampus of the obese offspring. They concluded that the increase in microglial activation basally in the hippocampus of the saturated fat diet group was indicative that the cells were chronically primed, giving them reason to believe these effects were programmed early in life (17). Another study concluded that severely obese mothers have a 67 to 88 percent risk of having an offspring with mild neurodevelopmental disorders, attention deficit/hyperactivity disorder, psychotic, mood and stress-related disorders when compared to mothers with a normal BMI (18).

Obesity and cardiovascular disease One of the most potent effects of maternal obesity is the risk of cardiovascular diseases. The excessive weight gain in more than 50 percent of mothers induces a 33 percent higher risk of childhood obesity (12). Obesity can induce metabolic disorders in mothers during pregnancy, and these disorders diminish cardiac function in the offspring (13). Human studies suggest that epigenetic marks or imprinting in offspring might be caused by obesity and gestational obesity in early pregnancy; such epigenetic changes might present as weakened glucose tolerance and obesity in the mother (12). The effect of epigenetic changes on adiposity, cardiovascular and metabolic diseases should be further explored in order to establish enough knowledge so that possible therapies could be developed. Another mechanism for development of cardiovascular disease might be the impaired organ development of the fetus (13). High-fat programming is a result of the maintenance of a highfat diet during fetal development that induces metabolic and physiological disorders that compromise the health of the fetus (13). High-fat programming can include hypertension during the intrauterine development of the offspring (13). Reported data shows that a maternal high-fat diet impacts the expression of blood pressure regulatory factors that can potentially lead to deformed cardiac function and increased risk of ischemic injuries (13). Not only are these injuries a risk, but even if the offspring does not end up clinically obese, there is the increased risk of cardiomyocyte hypertrophy (13).

Obesity and metabolic disease

Nutrition in pregnancy One of the recommended solutions to combat this issue starts with the nutrition of these mothers. The World Health Organization (WHO) has made many recommendations for pregnant women to stay healthy and prevent fetal health degeneration. Folate is recommended as a supplement since it is required during the neural tube formation in embryogenesis (6). Adequate levels of folate prevent neural tube defects in the children. Pregnant women can obtain dietary folate from citrus, dark-green vegetables, nuts, and liver (6). However, the foods needed for a healthy pregnancy are expensive and could be difficult for pregnant women in low-income communities to obtain (19). Due to the potential for weight gain in pregnancy, it is recommended that prospective mothers maintain a relatively normal weight range before pregnancy. Although maternal obesity is increasingly diagnosed among Latin and AfricanAmerican populations, where the economic effects of malnutrition is pronounced, research has indicated that breastfeeding is a potential protective factor against maternal and fetal obesity (20). Consequently, the infant will not rely

Effect of Maternal Obesity on Maternal and Fetal Health

on supplemental baby food as an alternative to its diet (21). Supplemental baby food is not the better alternative to the nutrients that are available from breastfeeding. Exposing an infant to chemically processed foods early in life can potentially result in health difficulties. In the 6 to 12 months after childbirth, there is significantly more milk production, a developmental benefit for the child. Additionally, increased production may help the mother return to her pre-pregnancy weight; however, if the mother was obese before pregnancy, the benefits of breastfeeding to return to a healthy weight may be minimal (21). Even so, the mother who breastfeeds her infant is at a lower risk of developing metabolic syndrome, Type 2 diabetes, postpartum depression, cardiovascular disease, and premenopausal breast and ovarian cancer (21). It has been reported that the rate of breastfeeding in the United States is low (21). Breastfeeding intervention to rectify this will promote improvement in maternal and fetal health (21). Physician responsibility Due to the increasing occurrence of maternal obesity, physicians should continue to address this issue with patients. When women come for yearly appointments, health and wellness should be a conversation that is followed by sustainable solutions. Some of these solutions should include dietary changes and physical exercise. In a recent study, overweight and obese women who reported that they exercised during their pregnancy had a lower incidence of small infants in terms of gestational weight (22). Contrary to the erroneous belief that exercise can induce labor, studies show no evidence that exercise increases the risk of preterm birth or reduces the gestational age at birth (23, 24). Thus, pregnant women should be encouraged to remain physically active during pregnancy. Obstetrician-gynecologists (OBGYNs) should also promote the decision to breastfeed. Data shows that the decision to breastfeed is beneficial for postpartum weight loss and optimal infant growth (21). Further, breastfeeding has proven to be effective for rectifying maternal obesity and promoting fetal health outcome. OBGYNs should work with mothers to create breastfeeding plans. Additionally, physicians and hospitals that are committed to resolving the issue of maternal obesity can employ nutrition specialists to be a part of their health care team. Logistically, it will be more impactful if these specialists are physically present in the physician’s office or hospital in order to improve access for patients. These specialists can work with the physicians to identify patients with increased risk and to take preventative measures before birth. Data shows that only 35 percent of physicians believe prenatal counseling is effective, and about 80 percent of OBGYNs provide advice to their patients on the importance of staying healthy (25). However, consistent counseling with physicians over time may prove to be more effective for patients, and the inclusion of nutritional counselors as part of the team can potentially improve these success rates.

Conclusion The increasing incidence of maternal obesity in the American society and the impact on fetal health is an important issue that must be addressed. OBGYNs and other health care providers can take a more active role in resolving this issue. 78

Studies show that obesity can lead to metabolic, cognitive, and cardiac disorders. Physicians should continue to advice women who are planning a pregnancy to aim for a healthy BMI before, during, and after their pregnancy in order to reduce the fetus’ risk for metabolic and cognitive disorders and increased cardiac issues. Having a better understanding of the factors that impact maternal obesity and the effects of maternal obesity on the developing fetus is key to combating this public health problem.

Acknowledgments Responsibilities were as follows: OO conducted the review, selected articles, wrote the publication, and had primary responsibility for the final content; Darina Lazarova, PhD, provided mentorship, suggested edits, recommended primary resources for the review. Sonia Lobo, PhD, provided mentorship for publication, edits, and formatting guidelines. All read and approved of the content.

References 1.

Mitchell NS, Catenacci VA, Wyatt HR, Hill, JO. (2011). Obesity: overview of an epidemic. The Psychiatric clinics of North America, 34(4):717–32.

2. Fitzsimons KJ, Modder J, Greer IA. (2009). Obesity in pregnancy: risks and management. Obstetric Medicine, 2(2):52–62. 3. Baugh N, Harris D, Aboueissa A, Sarton C, Lichter E, (2016). The Impact of Maternal Obesity and Excessive Gestational Weight Gain on Maternal and Infant Outcomes in Maine: Analysis of Pregnancy Risk Assessment Monitoring System Results from 2000 to 2010. Journal of Pregnancy. 2016:5871313. 4. Sahoo K, Sahoo B, Choudhury AK, Sofi NY, Kumar R, Bhadoria AS. (2015). Childhood obesity: causes and consequences. Journal of Family Medicine and Primary Care, 4(2):187–92. 5. Leonard SA, Rasmussen KM, King JC, Abrams B. (2017). Trajectories of maternal weight from before pregnancy through postpartum and associations with childhood obesity. The American Journal of Clinical Nutrition, 106(5):1295–1301. 6. Kominiarek MA, Rajan P. (2016). Nutrition Recommendations in Pregnancy and Lactation. The Medical Clinics of North America, 100(6):1199–1215. 7.

Brett KE, Ferraro ZM, Yockell-Lelievre J, Gruslin A, Adamo KB. (2014). Maternal-fetal nutrient transport in pregnancy pathologies: the role of the placenta. International Journal of Molecular Sciences, 15(9):16153–85.

8. Wilson RM, Messaoudi I. (2015). The impact of maternal obesity during pregnancy on offspring immunity. Molecular and Cellular Endocrinology, 418 Pt 2(2):134–42. 9. Frias AE, Morgan TK, Evans AE, Rasanen J, Oh KY, Thornburg KL, Grove KL. (2011). Maternal high-fat diet disturbs uteroplacental hemodynamics and increases the frequency of stillbirth in a nonhuman primate model of excess nutrition. Endocrinology, 152(6):2456–2464.

Effect of Maternal Obesity on Maternal and Fetal Health

10. Roberts JM, Bodnar LM, Patrick TE, Powers RW. (2011). The Role of Obesity in Preeclampsia. Pregnancy Hypertension, 1(1):6–16. 11. Gathiram P, Moodley J. (2016). Pre-eclampsia: its pathogenesis and pathophysiolgy. Cardiovascular Journal of Africa, 27(2):71–78. 12. Gaillard R. (2015). Maternal obesity during pregnancy and cardiovascular development and disease in the offspring. European Journal of Epidemiology, 30(11):1141–52.

24. Tinloy J, Chuang CH, Zhu J, Pauli J, Kraschnewski JL, Kjerulff KH. (2014). Exercise during pregnancy and risk of late preterm birth, cesarean delivery, and hospitalizations. Women's health issues: official publication of the Jacobs Institute of Women's Health, 24(1):e99–e104. 25. Leddy MA, Power ML, Schulkin J. (2008). The impact of maternal obesity on maternal and fetal health. Reviews in Obstetrics & Gynecology, 1(4):170–8.

13. Cerf ME. (2018). High-fat Programming and Cardiovascular Disease. Medicina (Kaunas, Lithuania), 54(5):86. 14. Ying W, Riopel M, Bandyopadhyay G, Dong Y, Birmingham A, Seo JB, Olefsky JM. (2017, October 05). Adipose Tissue Macrophage-Derived Exosomal miRNAs Can Modulate in Vivo and In Vitro Insulin Sensitivity. Cell, 171(2):372–384. 15. Lumeng CN, Bodzin JL, Saltiel AR. (2007). Obesity induces a phenotypic switch in adipose tissue macrophage polarization. The Journal of Clinical Investigation, 117(1):175–84. 16. Zhu C, Han T, Zhao Y, Zhou X, Mao X, Qi H, Zhang H. (2018). A mouse model of pre-pregnancy maternal obesity combined with offspring exposure to a high-fat diet resulted in cognitive impairment in male offspring. Exp Cell Res. 368(2):159–166. 17. Bilbo SD, Tsang V. (2010). Enduring consequences of maternal obesity for brain inflammation and behavior of offspring. FASEB J. 24(6):2104–15. 18. Kong L, Norstedt G, Schalling M, Gissler M, Lavebratt C. (2018). The Risk of Offspring Psychiatric Disorders in the Setting of Maternal Obesity and Diabetes. Pediatrics. 142(3). pii: e20180776. 19. Caprio S, Daniels SR, Drewnowski A, Kaufman FR, Palinkas LA, Rosenbloom AL, Schwimmer JB. (2008). Influence of race, ethnicity, and culture on childhood obesity: implications for prevention and treatment: a consensus statement of Shaping America's Health and the Obesity Society. Diabetes Care, 31(11):2211–21. 20. Yan J, Liu L, Zhu Y, Huang G, Wang PP. (2014). The association between breastfeeding and childhood obesity: a meta-analysis. BMC Public Health, (14):1267. 21. Williams CB, Mackenzie KC, Gahagan S. (2014). The effect of maternal obesity on the offspring. Clinical Obstetrics and Gynecology, 57(3):508–515. 22. Myrex P, Harper L, Gould S. (2018). An Evaluation of Birth Outcomes in Overweight and Obese Pregnant Women Who Exercised during Pregnancy. Sports (Basel, Switzerland), 6(4):138. 23. Wang C, Wei Y, Zhang X, Zhang Y, Xu Q, Sun Y, . . . Yang H. (2017, April). A randomized clinical trial of exercise during pregnancy to prevent gestational diabetes mellitus and improve pregnancy outcome in overweight and obese pregnant women. Am J Obstet Gynecol, 216(4):340–351.

Scholarly Research In Progress • Vol. 3, November 2019

Trends in Buprenorphine Prescription in the United States from 2008–2017 Amir R. Pashmineh Azar1*, Warren S.L. Lam1†, Suhail H. Kaleem1†, Laura B. Lockard1†, Mark R. Mandel1†, and Brian J. Piper1,2

Geisinger Commonwealth School of Medicine, Scranton, PA 18509 Center for Pharmacy Innovation and Outcomes, Geisinger Precision Health Center, Forty Fort, PA 18704 *Master of Biomedical Sciences Program †Doctor of Medicine Program Correspondence: skaleem@som.geisinger.edu 1

Abstract Background: The opioid epidemic in the United States continues to be a problem with no clear solution. Recent advances in the medical field have offered prescription medications such as buprenorphine as options to treat opioid dependence. The purpose of this study was to elicit trends in buprenorphine prescription across states and U.S. territories. Methods: Data was extracted from the U.S. Drug Enforcement Administration’s Automation of Reports and Consolidated Orders System (ARCOS) for 2008 through 2017. Each state was analyzed for buprenorphine prescription in milligrams per 100 000 persons. Results: There was a 248% increase in the amount of buprenorphine prescribed throughout the United States from 2008 to 2017. West Virginia (2,729.59g/100K) had the highest gross change while Iowa (180.31g/100K) had the lowest gross change. North Carolina and Arkansas saw the greatest percentage increase in buprenorphine prescription per 100 000 people at 785% and 701%, respectively. Maine and Utah had the lowest percentage increase with 105% and 116%, respectively. In general, the Midwest region had the greatest overall percentage increase in buprenorphine prescription over the decade, however, the northeast region has consistently shown the highest prescription per 100 000 people throughout the 10-year span. Conclusion: All states and territories across the United States had increased buprenorphine prescription over the last decade from 2008 through 2017. Further consideration into the causal factors for these observed trends and regional disparities is needed.

Introduction The United States is currently experiencing a public health crisis due to the growing opioid epidemic. From 1999 to 2017, the United States experienced an almost sixfold increase in opioid-related deaths (1). The Centers for Disease Control and Prevention (CDC) reported 47,600 fatal overdose cases in 2017 as a result of opioid abuse (2). In order to combat the rise in opioid use and the subsequent overdoses, it is important to take advantage of the evidence-based medical treatments that are available. Buprenorphine is a synthetic opioid that was first brought to market in 1978 in the United Kingdom as an injectable analgesic for treating acute and post-operative pain (3). Buprenorphine is a high-affinity, partial agonist of the mu-

opioid receptor with pharmacologic properties which contribute to its efficacy in treating opioid use disorder (OUD). The partial agonistic behavior of the drug at the mu-receptor produces less tolerance, less intoxicating effects, and a lower likelihood of causing respiratory depression compared to other forms of treatment (3). The high affinity of the drug for the mu-receptor allows buprenorphine to cause a blockade of the receptor, similar to the emergency reversal agent naltrexone (3). When the drug is administered sublingually, it has a very long half-life of 28 to 37 hours and therefore is a good choice for outpatient therapy. In addition, buprenorphine has been found to be a kappa-opioid receptor antagonist, which might explain its ability to aid in discomfort associated with opioid withdrawal (3). Buprenorphine has anywhere between 25 and 100 times greater potency than morphine (4). This, along with its increased dissociation time from the mu-receptor, are supportive reasons for its therapeutic use in treating patients with pain and/or OUD (4). Buprenorphine was granted FDA approval for treating OUD in 2002 (5). Although practicing physicians are aware of buprenorphine use in maintenance treatment for OUD, several barriers still prohibit the ease and speed of adopting this treatment method. The U.S. Drug Addiction Treatment Act of 2000 (DATA 2000) requires that medical practitioners wishing to prescribe the drug must undergo an 8-hour online training module (6). In addition to overcoming this hurdle, there is also a limit on the number of patients that an individual provider is permitted to prescribe buprenorphine for, although this number can be increased if the provider meets certain requirements (6). There is a vast body of evidence attesting to the effectiveness of agonist substitution treatment. Meta-analysis and various studies have demonstrated a high efficacy of the muopioid partial agonist buprenorphine as an OUD therapy (7). Alternative drug therapies such as methadone also exist, but have higher chances of misuse and diversion. Buprenorphine was introduced in France in the 1990s as a widely available prescription drug by primary care physicians to combat the effects of opioid addiction. Implementation of the drug for treatment resulted in a major decline in the number of overdose deaths (8). Since the benefits of buprenorphine treatment for OUD have already been established, the objectives of this descriptive study were to identify trends in buprenorphine distribution across the United States between 2008 and 2017, detect regional differences, and suggest possible factors that could have contributed to the observed trends.

Trends in Buprenorphine Prescription in the United States from 2008–2017

Figure 1. Change in U.S. buprenorphine distribution in grams per 100K population from 2008 to 2017.

Figure 3. Average buprenorphine distribution in grams in the United States over a 10-year span from 2008 to 2017 in different regions and United States as a whole. States were divided into four regions (Northeast, Midwest, West, South) according to U.S. Census Bureau standard.

Methods Study sample and measures Data was obtained from the U.S. Drug Enforcement Administration’s Automation of Reports and Consolidated Orders System (ARCOS) for 2008 through 2017. ARCOS is a federal program mandated by the 1970 Controlled Substances Act that reports on narcotics in Schedules I to III from pharmacies, hospitals, and Narcotic Treatment Programs. Subsequently, data for buprenorphine prescription in grams per 100 000 persons for each state was extracted and compiled into a database with Microsoft Excel. Individual states were grouped into one of four regions (Northeast, Midwest, West, South) and the values were also converted into percent change with distribution in 2008 set at baseline 0%. Data analysis Statistical analysis was completed using IBM SPSS software and included Analysis of Variance (ANOVA) among four geographical regions of distribution and independent T-tests comparing annual changes in distribution as well as comparing

Figure 2. Change in buprenorphine distribution in grams per 100K population from 2008 to 2017. Darker shades of blue indicate a greater relative increase in distribution of buprenorphine when compared to distribution values in 2008.

Figure 4. Percent increase in buprenorphine distribution per 100K population from 2008 to 2017. Darker shades of red indicate a greater relative percent increase in distribution of buprenorphine when compared to distribution values in 2008.

each region to the others was performed with significance value of p<0.05. Grubb’s test was used to identify potential outliers.

Results Review of overall buprenorphine distribution controlled for population in the United States between 2008 and 2017 (grams/100K population) showed an increase of 248% (from 376.6 to 1311.4 grams/100K) over this 10-year period, as seen in Figure 1. The overall trend of growth appears consistent with a linear progression (dD/dT=98.99 grams/100K/year, R2=0.9942). However, individual states did not demonstrate the same linear increase over time. On T-test evaluation, increases were statistically greater than expected average for a linear increase between 2008 and 2009 (p=0.03) and between 2016 and 2017 (p=0.01), and also statistically less than expected between 2009 and 2010 (p<0.01) and between 2010 and 2011 (p=0.01). In Figure 2, we calculated changes in buprenorphine distribution in individual states over the past decade, which ranged from 180.31 g/100K in Iowa to 2729.59 g/100K in West 81

Trends in Buprenorphine Prescription in the United States from 2008–2017

Virginia, with an overall average change of 934.84 g/100K. Grubb’s test analysis showed that no states were significant outliers. After categorization of states into U.S. census regions, regional distribution was greatest in the Northeast (1458.6 g/100K), followed by South (1177.3 g/100K), West (640.6 g/100K), and Midwest (517.3 g/100K), as shown in Figure 3. To confirm if regions were significantly different from each other at the beginning and end of this time frame, we performed a one-way ANOVA comparing all regions as well as independent T-tests between individual regions for 2008 and 2017 data as seen in Table 2. There were statistically significant differences between all regions in both years except when comparing West vs. South in 2008 (p=0.111) and comparing Midwest to West in 2017 (p=0.184). We also evaluated percent increase in distribution between 2008 and 2017 to account for the possibility that states which already had higher levels of buprenorphine distribution in 2007 may also be more likely to increase distribution at a greater rate for multiple reasons, as seen in Figure 4. Percent increases ranged from 104.8% in Maine to 785.3% in North Carolina, with an overall average percentile increase of 304% for the U.S. Grubb’s test review suggests that North Carolina’s percentage increase would be considered the sole outlier (F=3.22). Figure 5 shows regional comparison of percentage changes. While

Figure 5. Percent change in buprenorphine distribution per 100K population compared to 2008 in different regions and the United States as a whole. States were divided into four regions (Northeast, Midwest, West, South) according to U.S. Census Bureau standard.

states in the Northeast had the greatest gross changes of per capita buprenorphine distribution on average, they also had the lowest percent change (198%) followed by West (257%), whereas states in the Midwest (365%) and the South (353%) had the greatest percentile changes. Comparing percent changes in regional buprenorphine distribution in 2017, one-way ANOVA of the regions was statistically significant (p=0.017) and independent T-test of mean comparing regions was statistically significant when comparing Northeast vs. Midwest (p=0.002), Northeast vs. South (p=0.024), and Midwest vs. West (p=0.029), as shown in Table 2.

Discussion The data obtained in this report indicated a gross overall increase in grams of buprenorphine prescribed per 100 000 people across the United States from 2008 to 2017 that followed a linear trend, although the same trend did not apply when evaluating individual states. It should be noted that the U.S. territories were not included in data analysis due to missing data points and inconsistent data across the timespan of interest. Additionally, an exponential increase did demonstrate better fit for the change in distribution when excluding the data in 2008 to 2009. Given that data collection started in 2007, it would not be surprising if there were initial issues during the first years that could make the data less accurate, and an exponential trend would correlate better with state-level data. It would be worthwhile to investigate how the distribution of buprenorphine changes in the next several years to observe how trends change. Given that buprenorphine has been prescribed as a treatment for OUD in the United States since 2002 (3) and that efforts have focused in the last decade to reduce patient dependence on opioids, it is possible that this overall increase reflects further efforts to curtail the consequences of opioid overprescription (9). According to the CDC, overall rates of opioid prescription have decreased within the same observed timeframe (10, 11). This additional data supports the notion that increased buprenorphine availability may coincide with a reduction in the prescription of other opioids. The increase in buprenorphine distribution is also observed when looking at the trends for each individual state. Table 1 shows that West Virginia had the greatest gross increase in buprenorphine distribution (2,729.59 grams/100K) across the decade, and Iowa had the lowest increase (180.31 grams/100K).

Table 2. T-test comparison of average buprenorphine distribution between different regions in 2008 and 2017 and percent change in 2017. Bolded values indicate statistical significance.

Trends in Buprenorphine Prescription in the United States from 2008â&#x20AC;&#x201C;2017

However, these gross increases in distribution may be partially explained by variances in initial buprenorphine distributions observed in 2008. Only five states had higher initial buprenorphine distribution in 2008 than West Virginia (Maine, Massachusetts, Rhode Island, Utah, Vermont), and only South Dakota had a lower initial distribution than Iowa. This finding supports the notion that states starting out with higher levels of buprenorphine distribution in 2008 had higher gross increases over the span of the decade and states that started with lower levels of distribution had lower gross increases. This trend could simply be due to unchanging factors across the United States, as a state that had a high distribution relative to all other states in 2008 is expected to have a high distribution relative to all other states in each subsequent year up to 2017 and beyond. Table 1 also shows that when investigating individual states for their percent increase in distribution from 2008 to 2017, Maine had the lowest percent increase in distribution (104.77%), and North Carolina had the highest percent increase in distribution (785.27%). Similar to the patterns seen in the gross buprenorphine distribution, some of the differences that were observed in the percent increase between states may be explained by differences in initial buprenorphine distribution rates. Since the method in this study to detect changes in distribution relied on comparing each yearâ&#x20AC;&#x2122;s distribution per 100,000 people to the distribution rate it exhibited in 2008, it would be expected that a state with a higher initial distribution rate would have a smaller percent increase and a state with a lower initial distribution rate would have a larger percent increase if both states increased by the same gross amount. Maine started with the second-highest distribution rate in 2008, which may in part explain why it had the lowest percent increase when compared to all other states. Likewise, North Carolina had the 12th lowest distribution rate in 2008, and so its higher percent increase may partially be explained by a lower starting point. However, given that Maine and North Carolina did not have the highest and lowest distribution rates in 2008, respectively, there must be additional factors that contributed to the difference in distribution rates observed in this study. One of the factors that could also contribute to the differences observed between states is the availability and accessibility of treatment options. While the use of buprenorphine in medication-assisted treatment (MAT) for opioid addiction is regulated by the federal government, it is up to each individual state to provide the resources for treatment (12). As a result, MAT programs for opioid addiction vary greatly between states and have been drastically changing over the last several years to try to combat the opioid epidemic. In 2016, Maine recognized that it had an epidemic of opioidrelated deaths and formed the Maine Opiate Collaborative. The Collaborative then proceeded to evaluate what some of the failings in their methods of offering opioid addiction treatment and recognized that a significant contribution to the problem was a gap in publicly funded treatment options. It was not until April 2017 that Maine adopted the Opioid Health Home (OHH) program in an emergency ruling, a program that allowed more individuals with OUD to receive treatment (13, 14). This delay in a proper treatment program may explain in part why Maine

Table 1. Gross and percentage changes in state buprenorphine distribution between 2008â&#x20AC;&#x201C;2017.

Trends in Buprenorphine Prescription in the United States from 2008–2017

had the lowest percent increase of all states observed in the timeframe of this study. It would be beneficial to continue to evaluate the state’s distribution rates of buprenorphine to see if any discernible changes come about as a result of implementing this new treatment program. Additionally, an explanation for Vermont’s higher percent increase in distribution may be related to its treatment model for OUD. Vermont’s “hub and spokes” system for treating OUD is a good example. After this system’s implementation, there was a 64% increase in physicians able to prescribe buprenorphine and a 50% increase in individuals served per provider from 2013 to 2017 (17). This correlates with the finding that Vermont was the third highest state in gross buprenorphine distribution increase from 2008 to 2017, suggesting that individual state treatment programs may influence the amount of buprenorphine distributed per state. In Figures 4 and 5, the regions of the United States were evaluated for change in buprenorphine distribution from 2008 to 2017 in gross amount as well as in percent increase, respectively. The U.S. regions were determined by the CDC’s standards for the U.S. Census Bureau and were divided into the Northeast, Midwest, West, and South (15). The gross amount of increase per region from 2008 to 2017, in order of lowest increase to highest increase, was Midwest (517.3 grams/100K), West (640.6 grams/100K), South (1,177.3 grams/100K), and Northeast (1,458.6 grams/100K). Unsurprisingly, the Midwest had the lowest gross amount increase across the decade. This may be explained in part due to the lack of providers who have a DEA waiver to prescribe buprenorphine for OUD treatment in this region relative to all other regions in 2017 (16). Those able to prescribe include physicians, nurse practitioners, and physician assistants. The Midwest has relatively more counties that have either no buprenorphine providers or counties with a low diversity of providers compared to the other three regions (16). In comparing the percent increase of buprenorphine distribution in the four regions, all were found to be significantly different from one another. The Northeast had the lowest percent change over the 10-year period (198%), and was significantly lower than the Midwest (365%) and the South (353%). The Midwest had the highest percent change and was significantly higher than the West (257%) and the Northeast. These percent increase trends may similarly be due to the fact that the Northeast had many states that already started with a high distribution rate in 2008 and the Midwest had many states that started at low distribution rates in 2008. Other factors that contribute to the trends shown in Figures 4 and 5 are likely due to the variability of treatment programs of states in each of these regions. Interestingly, it is evident that each region had a sharper slope increase in gross amount (Figure 4) and percent increase (Figure 5) from the time span of 2016 to 2017 compared to any other previous one-year timespan. This indicates a higher rate of prescription after 2016 in all four regions. This trend could be due to the addendum to the DATA 2000 act that allowed providers to prescribe buprenorphine to 100 individuals at any given time after the spring of 2016 (18). Previously, providers were only limited to prescribing to 30 individuals at any given

time (3). Because providers who intend to prescribe Schedule III controlled substances like buprenorphine need to take an 8- to 20-hour training course on its use, the dissemination of buprenorphine across the U.S. is largely dependent on the willingness of individual providers to undertake the necessary steps to prescribe. The relationship between physician participation rates in buprenorphine training and buprenorphine distribution would be very difficult to measure and is likely influenced by the states and regions that have the highest opioid overdose levels and the highest need for MAT for OUDs. Further investigations should focus on whether these events are directly correlated to one another. There were some limitations in this investigation. While measuring buprenorphine prescriptions in grams per 100 000 people allows data analysis to control for relative population size, it cannot differentiate whether an observed increase is due to an increased number of individuals being prescribed or due to an increase in the amount of buprenorphine prescribed per person. All data used for analysis was collected from ARCOS, and therefore was limited to the reporting system used to record buprenorphine prescriptions. ARCOS does not differentiate whether these buprenorphine prescriptions are being used for chronic pain or for the treatment of addiction. Additionally, ARCOS does not indicate the brands included in evaluating the distribution of buprenorphine. Different brands of the drug may or may not include naloxone in treatment, and each brand has a different indication for use (19). Furthermore, ARCOS also reports on buprenorphine used for veterinary purposes; however, medications for this purpose are almost negligible compared to use in humans (20). For future investigations, it would be worthwhile to compare prescription rate data to rates of emergency room visits attributable to opioid overdose. Following the logic that increases in buprenorphine prescription may be due to greater efforts in treating OUD, one hypothesis would be that higher prescription rates coincide with decreased emergency room visits due to an expected lower incidence of overdose. Additionally, buprenorphine is not exclusively given as a single drug therapy for the treatment of OUD, but may be given in combination with naloxone. Thus, it may be beneficial to compare the trends of individual formulations of buprenorphine treatments to see if they follow the same trends that were observed in this study. This would suggest which specific buprenorphine treatment methods are more popular as well as provide insight on which methods are considered more useful. Furthermore, buprenorphine is not the only pharmacological intervention for OUD. Other treatments such as naltrexone or methadone, in addition to combinations of naloxone and buprenorphine, are all approved by the FDA for the treatment of opioid dependence (9). Given this diversity of treatment options, future research should be aimed at comparing the efficacy of these treatments (21).

Conclusion Between the years of 2008 and 2017, buprenorphine distribution rates significantly and consistently increased across the United States as a whole. This increasing trend remained

Trends in Buprenorphine Prescription in the United States from 2008–2017

consistent when evaluating individual states. Regional analysis showed that the Midwest had the lowest increase in gross amount of buprenorphine distribution which could be explained by its relative lack of providers able to prescribe compared to other regions. Although further investigation is needed, the overall increase in buprenorphine distribution rates across the United States could indicate a promising trend in the fight against opioid addiction.

12. U.S. Department of Health & Human Services. MAT Statutes, Regulations, and Guidelines | SAMHSASubstance Abuse and Mental Health Services Administration [Internet]. 2019 [cited 2019 May 13]. Available from: https://www.samhsa.gov/medicationassisted-treatment/statutes-regulations-guidelines

References

14. Opioid Health Homes [Internet]. 2017 [cited 2019 May 14]. Available from: https://www.maine.gov/legis/opla/ OTFDHHSOpioidHealthHomes.pdf

Overdose Death Rates | National Institute on Drug Abuse (NIDA) [Internet]. [cited 2019 May 16]. Available from: https:// www.drugabuse.gov/related-topics/trends-statistics/ overdose-death-rates

2. National Academies of Science Engineering and Medicine. Pain Management and the opioid epidemic: Balancing societal and individual benefits and risks of prescription opioid use. The National Academies Press. 2017. 3. Shulman M, Wai JM, Nunes EV. Buprenorphine treatment for Opioid Use Disorder: An overview. CNS Drugs [Internet]. 2019 May; Available from: https://doi.org/10.1007/ s40263-019-00637-z 4. Khanna IK, Pillarisetti S. Buprenorphine–An attractive opioid with underutilized potential in treatment of chronic pain. Journal of Pain Research. 2015; 8: 859–70. 5. Jaffe JH, O’Keeffe C. From morphine clinics to buprenorphine: Regulating opioid agonist treatment of addiction in the United States. Drug and Alcohol Dependence. 2003; 70: 3–11. 6. Mcnicholas L. Buprenorphine clinical guide clinical guidelines for the use of buprenorphine in the treatment of opioid addiction: a treatment improvement protocol TIP 40 [Internet]. [cited 2019 May 14]. Available from: http://www. kap.samhsa.gov/products/ 7.

Ling W, Casadonte P, Bigelow G, Kampman KA, Patkar A, Bailey G.L., et al. Buprenorphine implants for treatment of opioid dependence: A randomized controlled trial. JAMA-J Am Med Assoc. 2010; 304(14): 1576–83.

13. Criminal M, Academy J. Maine Opiate Collaborative, May 2016 Law Enforcement Task Force.

15. CDC-U.S. Census Regions and Divisions That Met U.S. Cancer Statistics Publication Criteria-USCS-Cancer [Internet]. 2018 [cited 2019 May 14]. Available from: https:// www.cdc.gov/cancer/uscs/technical_notes/criteria/censusregions-divs.htm 16. Andrilla CHA, Moore TE, Patterson DG, Larson EH. Geographic Distribution of Providers With a DEA Waiver to Prescribe Buprenorphine for the Treatment of Opioid Use Disorder: A 5-Year Update. J Rural Heal. 2019;35(1):108–12. 17. Brooklyn JR, Sigmon SC. Journal of Addiction Medicine. 2017;11(4):286–92. 18. Knudsen HK, Havens JR, Lofwall MR, Studts JL, Walsh SL. Buprenorphine physician supply: Relationship with statelevel prescription opioid mortality HHS Public Access. 2017 [cited 2019 May 14];173(1):55–64. 19. Velander JR. Suboxone: Rationale, Science, Misconceptions. Ochsner J [Internet]. 2018;18(1):23–9. 20. Piper BJ, Shah DT, Simoyan OM, McCall KL, Nichols SD. Trends in Medical Use of Opioids in the U.S., 2006–2016. Am J Prev Med [Internet]. 2018;54(5):652–60. 21. U.S. Food & Drug Administration. Information about Medication-Assisted Treatment (MAT) [Internet]. [cited 2019 May 14]. Available from: https://www.fda.gov/drugs/ information-drug-class/information-about-medicationassisted-treatment-mat

8. Auriacombe M, Fatséas M, Dubernet J, Daulouède JP, Tignol J. French field experience with buprenorphine. Am J Addict. 2004;13 Suppl 1:S17–28. 9. Welsh C, Valadez-Meltzer A. Buprenorphine: a (relatively) new treatment for opioid dependence. Psychiatry (Edgmont) [Internet]. 2005;2(12):29–39. 10. Alderks CE. Trends in the Use of Methadone, Buprenorphine, and Extended-Release Naltrexone At Substance Abuse Treatment Facilities: 2003–2015 (Update). CBHSQ Rep August 22, 2017 [Internet]. 2017;(October 2010):1–6. Available from: https://www. samhsa.gov/data/sites/default/files/report_3192/ ShortReport-3192.pdf 11. Hoots BE, Xu L, Kariisa M, Wilson NO, Rudd RA, Scholl L, et al. 2018 Annual surveillance report of drug-related risks and outcomes: United States [Internet]. [cited 2019 May 16]. Available from: https://www.cdc.gov/

Scholarly Research In Progress • Vol. 3, November 2019

Transcutaneous Vagus Nerve Stimulation for TreatmentResistant Depression Brianna Dade1*, Tina Giutashvili1*, and Christina Michel1*

Geisinger Commonwealth School of Medicine, Scranton, PA 18509 *Master of Biomedical Sciences Program Correspondence: tgiutashvili@som.geisinger.edu 1

Abstract

Therefore, a promising treatment for depression is necessary.

Management of depression has become an increasing concern, because studies show depression is linked to suicide. Additionally, many mildly and moderately depressed patients do not respond to traditional treatments (termed treatment-resistant depression). For this reason, vagus nerve stimulation (VNS) has been studied as an alternative therapy. The treatment usually targets the locus coeruleus and dorsal raphe nucleus to oppose sympathetic responses to stress. The hypothalamus, however, is also being investigated because it is indicated as a potential brain area associated with depression. The research presented in this paper focuses on the more novel, transcutaneous VNS (tVNS) therapy and its effectiveness in treating major depressive disorder (MDD), while investigating a distinct biomarker for the efficacy of the treatment. Participants in the study were classified as having mild or moderate depression by the 10th revision of the International Statistical Classification of Diseases and Related Health Problems (ICD). The experimental group received tVNS therapy while the control group received sham therapy. tVNS proved to continuously reduce depression HAM-D scale ratings as well as SDS scores from baseline through 12 weeks. Functional magnetic resonance imaging was used to assess the functional connectivities. The functional connectivity between the medial hypothalamus and the rostral anterior cingulate cortex was correlated with improvements in the depression scores and served as the distinct biomarker of the treatment effects of tVNS.

Beyond the fact that there is a risk for suicide, depression management has gained attention because as many as 20% to 40% of patients do not respond to a series of treatments, such as antidepressant medications, psychotherapy, TMS, and ECT (1, 4). TMS and ECT techniques are often used when other traditional treatments do not relieve symptoms (1). TMS is a noninvasive therapy in which magnetic fields stimulate nerve cells of the brain, specifically areas involving mood (5). ECT emits small electric currents through the brain, intentionally triggering a seizure. With this procedure, brain chemistry is changed to quickly reverse symptoms (6).

Introduction Studies have shown neuronal dysregulation in the brain of depressed patients, specifically within the frontal lobe. Disturbances include reduced metabolism and blood flow in the prefrontal cortex and anterior cingulate cortex and increased activity in the subgenual cingulate and amygdala (1). Depression has been a targeted focus of study because major depression has been linked to suicide. Specifically, a recent study investigated short-term suicide risk among depressed patients after psychiatric hospital discharge. Nearly 1.9 million Medicaid claims were reviewed in a longitudinal cohort study that showed short-term (up to 90 days post-hospital release) rates of suicide were highest among adults diagnosed with depression disorder, compared to other psychiatric conditions such as bipolar disorder and schizophrenia. This was significant, as the reported data showed a rate of 235.1 per 100 000 person-years versus 11.6 per 100 000 person-years among people with non-mental disorders (2). Additionally, depression is the fourth leading cause of disability worldwide and it is the number one disease-driven cause of death among people 15 to 40 years of age in the United States (1, 3). 86

Vagus nerve stimulation for treatment-resistant depression Vagus nerve stimulation (VNS) is being studied and implemented as an alternative and/or supplemental therapy for treatment-resistant (hard-to-treat) depression (i.e., when a patient does not respond to several traditional treatments). Treatment-resistant depression is often seen in patients suffering from mild or moderate depression (3). In addition to depression, VNS therapy may also be used to treat epileptic patients. Similar to depression, as many as 30% to 40% of epilepsy cases are not well controlled by drug therapy (4). VNS was approved in Europe for pharmacoresistant epilepsy in 1994 and 1997 in the United States (1). The procedure is also being studied for its potential to treat severe pain, headaches and migraines, and diseases such as multiple sclerosis and Alzheimer’s (4). VNS, similar to ECT, is a type of neuromodulation in which neuronal activity is targeted by electrical stimulation. There is a vagus nerve on each side of the body, which begins at the brainstem and travels through the neck and chest. The afferent fibers in the neck carry impulses to the brainstem, targeting the locus coeruleus and dorsal raphe nucleus, opposing sympathetic responses to stress (4). In addition, the hypothalamus is being investigated, because it is indicated as a potential brain area associated with MDD (3, 9). The focus of this paper is to evaluate transcutaneous VNS (tVNS) therapy as an alternative treatment for patients with MDD, while investigating a distinct biomarker to measure the treatment’s efficacy. This newer, noninvasive therapy is a potentially safer, more convenient therapy, as it does not require surgery. It has been approved for the treatment of epilepsy, depression, and pain in Europe. However, tVNS has not yet been approved for treatment in the United States (3). Transcutaneous VNS therapy stimulates the external afferent auricular branch of the left vagus nerve, which is located medial of the tragus (toward the inside of the ear) at the acoustic meatus (7). The device is handheld, with the terminal end of the wires connected to a generator that delivers the

Transcutaneous Vagus Nerve Stimulation for Treatment-Resistant Depression

electrical signal in a pulse-like fashion (8). Previous studies have concluded that the location of the ear is ideal for VNS, as it is the only place on the surface of the body in which the afferent vagus nerve is accessible, which when stimulated, can produce similar effects to that of traditional VNS (9).

Materials and Methods Overview of clinical trials To test the overall effectiveness of tVNS on depression and other mood disorders, a non-randomized, single-blinded clinical trial was performed with a treatment and placebo group. Specifically, in one experimental trial conducted, 120 cases of adult volunteers who were diagnosed with mild or moderate depression (treatment-resistant depression) were recruited for the study, and divided into two groups: a tVNS and a sham vagus nerve stimulation group (Figure 1).

The ICD-10, the 10th revision of the International Statistical Classification of Diseases and Related Health Problems created by the World Health Organization (WHO) (10), was first utilized by researchers to categorize participants as having one of the two stated forms of depression. The ICD-10 diagnosis standards for mild and moderate depression states consists of the appearance of two typical and two other core symptoms, and exhibiting two typical and three other core symptoms, respectively (10). Typical symptomologies of depression, as noted by this classification, are depressed mood, loss of interest in daily activities, and increased fatigue (10). Core symptomologies can range from thoughts of suicide, to sleep disturbances and loss of appetite. Excluded individuals were participants who engaged in risky behaviors, such as smoking, excessive prescription drug use and alcoholism; women who were pregnant; individuals diagnosed with bipolar disorder or postpartum depression; or people who simply refused to sign a consent form to allow researchers to proceed with the study (3, 9). VNS VNS can occur in any amount, with any frequency, in any combination, and in any order (9). However, skin resistance to the treatment is tested prior to stimulation, specifically with a high frequency current. High resistance is measured when there is a low quality of skin-to-electrode contact. The specific dosage of electrical administration may be dependent on a variety of factors, such as the severity of the case, age of the patient receiving the treatment, lifestyle, or the physician (8). Generally, participants in the tVNS group had the ear clips attached to points of the auricular concha area, where rich vagus nerve branch distribution is present, while participants in the sham tVNS group had stimulation points located at the superior scapha, where no vagus nerve distribution is noted (9). The electrophysiological stimulation parameters noted were: 20-Hz density wave of stimulation and varying intensity (amplitude) of the stimulation based on individual tolerance (within a 4 to 6 mA range).

Figure 1. The study, a non-randomized clinical trial, resting state fMRI, continuous tVNS or sham tVNS fMRI, and clinical scores were used to establish a baseline. Treatments in the experimental and control groups were self-administered, with the exception of the stimulation treatments during fMRI scans. Adapted from (3).

Inclusion and exclusion criteria Clinical inclusion criteria for tVNS studies investigating depression require patients with mild or moderate states, because these forms of depression have been scientifically proven to be treatment-resistant (3). The non-randomized format of the studies kept ethical concerns intact, as participants acted as both controls and patients in order for both groups to experience the relative effects of tVNS therapy.

When an electric stimulus is administered, a majority of the electrical current occurs during periods of inexcitability (8). That is, the current flows following the several milliseconds after a neuronal action potential has been fired, in which it is almost impossible to fire another, subsequent action potential: the absolute refractory period. Repeated brief pulses generated by VNS triggers action potential firing that parallels intrinsic firing patterns of neurons in crucial CNS regions (8). In another single-blinded clinical trial, two cohorts were utilized to compare and contrast visual differences that tVNS potentially caused on both fMRI scans and resting-state default mode networks (DMN). The first cohortâ&#x20AC;&#x2122;s tVNS duration lasted 12 weeks, and a second cohort was recruited to receive a sham tVNS for the first 4 weeks of treatment, before all converting to the tVNS treatment for the remainder of the 12 weeks (3). Personal journal entries All participants were asked to keep a detailed log of any potentially experienced side effects during the course of the study (3, 9).

Transcutaneous Vagus Nerve Stimulation for Treatment-Resistant Depression

Hamilton Depression Rating Scale (HAM-D) Clinical outcomes (change in depressive episodes) were primarily measured using the HAM-D (3, 9). This 24-item assessment was used by licensed physicians to categorize the severity of depressive states in conjunction with clinical interviews. Although the assessment itself consists of 24 questions, scoring is based only on the first 17. The first eight items are scored from zero (depression not present) to 4 (severe form of depression present), while the next nine items are scored from zero to 2. Ultimately, the study participantsâ&#x20AC;&#x2122; depressive states were scored based on the sum of the 17 items throughout the duration of the clinical trial as being normal, mild, moderate, severe, and very severe (3,9). The HAM-A, the SAS, and the SDS were used as secondary measures (3, 9). Functional Magnetic Resonance Imaging (fMRI) acquisition In both the tVNS and sham tVNS groups, fMRI for every participant was captured to measure brain activity via changes in blood oxygen levels. During data processing, the first 10 volumes were disposed to allow signal equilibration (3, 9). Seed-based functional connectivity (FC) Studies investigated the hypothalamus, a key brain area of the limbic system, using the bilateral medial and lateral hypothalamus (LH) as seeds. FCs were computed between seed time courses and every other voxel time course. Two null seeds were chosen for both hypothalamic brain areas in order to analyze their activities during stimulation (3).

Results and discussion Effectiveness of tVNS on MDD The efficacy of tVNS is being investigated for the treatment of hard-to-treat MDD. The antidepressant effect of tVNS as a standalone treatment was explored. In this treatment there were two cohorts: one that received tVNS for 12 weeks to test the efficacy of the treatment and a second cohort in which participants received sham tVNS for 8 weeks, then switched to the experimental treatment for the following 8 weeks. The clinical outcomes were measured at baseline, 4, 8, and 12 weeks. The primarily outcome of interest was the 24-item HAM-D score. The patients were divided into a mild or moderate depression subgroup based on their HAM-D scores. Both experimental subgroups had a significant treatment effect in the mild subgroup (p=0.04) and moderate subgroup (p<0.0001) (9). The results of the depression scales show a continuous reduction in depression scores from baseline through 12 weeks (Figure 2). Response rate, defined as a 50% reduction in HAM-D scores, were significantly different (p<0.00001) between the sham and the experimental groups. Throughout the study there was a continuous increase in response rate, especially during week 4 where the experimental group had 24 patients respond, and the sham group had zero. After the sham tVNS group shifted to the experimental group, there was no significant difference in response (p=0.07). Overall, there was a more significant decrease in the experimental group. Similar results were seen in a self-assessment depression

scale, where the patients had to assess their perceived levels of depression. Remission, defined as a HAM-D score <8, was also assessed (Table 1). The most significant differences between the groups were seen during weeks 0 to 4 and 8 to 12 in the tVNS group (p=0.0001), indicating that remission rates were continuously increased throughout the study. The study concluded that tVNS may be used as a first-line, standalone treatment (7). Although promising, this conclusion is premature to make at this time, since tVNS is still a novel treatment, many more studies in the future are needed to safely conclude its efficacy as a standalone, first-line treatment (9). Neural biomarker of tVNS The neural mechanism of tVNS therapy is one that has not been clearly defined previously in scientific literature. A biomarker was investigated for tVNS treatment in a singleblinded study. The goal of this investigation was not only to assess the changes in the brain connectivity but also to elucidate a viable biomarker for the tVNS state, brought on by the continuous stimulation of the vagus nerve. The first cohort of the study was used to assess the antidepressant effects of tVNS for demonstrating efficacy. After that, a second cohort was recruited to compare the differences between sham and experimental groups. Both the sham and the experimental group self-administered the treatment, except when the 6-minute treatment was administered during the MRI scan. The primary clinical outcome was measured by the 24-item HAM-D. Secondary outcome measures were the HAM-A, the SAS, and the SDS. The results, outlined in Table 2, showed that there were no significant differences between the groups at baseline (3). The results of the functional connectivity test using the medial hypothalamus as the seed, mapped the connectivity to the rostral anterior cingulate cortex (rACC), right medial frontal gyrus, and the cerebellum. The experimental group had a significantly lower functional connectivity between the medial hypothalamus (MH) and the rACC. A lower functional connectivity was also seen between the medial hypothalamus and the right medial frontal gyrus. Furthermore, a higher functional connectivity was seen between the MH and the cerebellum. Further connectivities were analyzed using the LH as a seed to the mid-cingulate cortex (MCC) and putamen, which proved to be significantly lower in the experimental group compared to the sham group. Statistical comparisons, outlined in Table 3, indicated that the most significant observations in both the sham and experimental groups were seen at the onset of the treatment. In the resting state, neither groupsâ&#x20AC;&#x2122; functional connectivities were correlated with improved HAM-D scores. The only functional connectivity that was correlated with improvements in the HAM-D score was the connection between the medial hypothalamus and the rACC (Table 4 and Figure 2). Based on this finding, the researchers suggested the MH-rACC functional connectivity may be a distinct biomarker for the treatment effects of tVNS (3). This finding was also supported by published journal research (11, 12). These studies also showed that the activity in the rACC predicted treatment responses for patients with MDD. The neurobiological mechanism of the treatment effect is still obscure. However, it is proposed that the mechanism can be attributed to a bottom-up processing. Vagal projections to the

Transcutaneous Vagus Nerve Stimulation for Treatment-Resistant Depression

Figure 2. Clinical outcome measurements at different time points. Adapted from (9).

Table 1. The number of patients who responded (50% reduction in HAM-D 24 scores) and achieved remission (HAM-D score <8). Adapted from (9).

Table 2. Clinical outcomes of different measurements in experimental and sham group. Adapted from (3).

Table 3. Statistical comparisons between hypothalamic FC in varying conditions. Adapted from (3).

Table 4. Correlations and p-values for the hypothalamic functional connectivities and the HAM-D improvements. Adapted from (3).

Transcutaneous Vagus Nerve Stimulation for Treatment-Resistant Depression

nucleus tractus solitarius (NTS) directly and indirectly project to the hypothalamus, thalamus, locus coeruleus, and insula, among other regions in the limbic system, which affect mood and emotion (13).

3. Tu Y, Fang J, Cao J, Wang Z, Park J, Jorgenson K, et al. A distinct biomarker of continuous transcutaneous vagus nerve stimulation treatment in major depressive disorder. Brain Stimul. 2018 Jan 17; 1–8.

Limitations

4. Vagus nerve stimulation [Internet]. Mayo Clinic. Mayo Foundation for Medical Education and Research; 2018. Available from: https://www.mayoclinic.org/testsprocedures/vagus-nerve-stimulation/about/pac-20384565

There were several limitations in the studies discussed. The studies assessed the effectiveness in the short term (12 weeks), which may not have shown the full benefits or risks of tVNS. To date, there has not been a long-term study conducted on human models to assess the efficacy and effects of long-term tVNS. Additionally, studies (9) did not utilize a true randomized study design—the sham group shifted to the experimental group after 4 weeks. This was due to ethical concern about withholding the potentially promising tVNS treatment from those suffering from depression. Future research should explore this further using an ethically randomized trial. The trials were solely conducted on hospitalized patients, which can result in Berkson’s bias. Future studies can build upon these findings by conducting trials in other, broader populations as well. Additionally, the participants in this study had a 2-week washout period, which potentially may not be sufficient for oral antidepressants. However, researchers and physicians may be interested in the combination of this treatment with cognitive behavioral therapy. Further research should also consider the efficacy of tVNS on suicide ideation and attempts, because there seems to be a link between suicide and depression. In addition to this, future research should expand to include a broader spectrum of patients. The study sample in these particular studies excluded potentially many individuals who may suffer from MDD, due to the selection criteria. By altering the inclusion criteria, researchers will be able to obtain results more applicable to the general population.

Conclusion The studies discussed show that tVNS can significantly reduce the symptoms of depression and potentially allow patients to achieve remission in the 12 weeks during treatment. The data indicate that symptom reductions were greater and more significant in the experimental groups compared to the sham. This indicates that tVNS may be a viable treatment for patients with mild or moderate MDD, which is categorically known to be treatment-resistant. Continuous tVNS attenuated the strength of the MH-rACC functional connectivity, which was significantly associated with tVNS treatment effects. Previous studies, mentioned earlier, showed that the dysregulation of the hypothalamus and hyperactivity of rACC are both indicated in depression. This study aligns and extends those findings by elucidating a functional biomarker (3).

References 1.

O'Reardon JP, Cristancho P, Peshek AD. Vagus nerve stimulation (VNS) and treatment of depression: To the brainstem and beyond. Psychiatry (Edgmont) [Internet]. 2006 May; 3(5):54–63.

2. Olfson M, Wall M, Wang S, Crystal S, Liu SM, Gerhard T, et al. Short-term suicide risk after psychiatric hospital discharge. JAMA Psychiatry. 2016 Nov; 73(11):1119–26.

5. Transcranial magnetic stimulation [Internet]. Mayo Clinic; 2017. Available from: https://www.mayoclinic.org/testsprocedures/transcranial-magnetic-stimulation/about/pac20384625 6. Electroconvulsive therapy (ECT) [Internet]. Mayo Foundation for Medical Education and Research; 2018. Available from: https://www.mayoclinic.org/ tests-procedures/electroconvulsive-therapy/about/pac20393894 7.

Van Leusden JWR, Sellaro R, Colzato LS. Transcutaneous vagal nerve stimulation (tVNS): A new neuromodulation tool in healthy humans? Front Psychol. 2015 Feb 10; 6:102.

8. Boveja B, Widhany A. Method and system to provide therapy for depression using electroconvulsive therapy (ECT) and pulsed electrical stimulation to vagus nerve(s) [Internet]. Google Patents. Google; 2005. Available from: https://patents.google.com/patent/US20050165458f 9. Rong P, Liu J, Wang L, Liu R, Fang J-L, Zhao J, et al. Effect of transcutaneous auricular vagus nerve stimulation on major depressive disorder: a nonrandomized controlled pilot study. J Affect Disord. 2016 May; 195:172–9. 10. World Health Organization. The ICD-10 classification of mental and behavioural disorders [Internet]. 10th ed. Bluebook. Geneva: World Health Organization; 1992. 11. Pizzagalli D, Pascual-Marqui RD, Nitschke JB, Oakes TR, Larson CL, Abercrombie HC.et. Anterior cingulate activity as a predictor of degree of treatment response in major depression: Evidence from brain electrical tomography analysis. Am J Psychiatry. 2001; 158(3):405–15. 12. Pizzagalli DA. Frontocingulate dysfunction in depression: Toward biomarkers of treatment response. Neuropsychopharmacology. 2010; 36(1):183–206. 13. Mohr P, Rodriguez M, Slavíčková A, Hanka J. The Application of Vagus Nerve Stimulation and Deep Brain Stimulation in Depression. Neuropsychobiology. 2011; 64(3):170–81.

Scholarly Research In Progress • Vol. 3, November 2019

Understanding the Opioid Crisis in the Southern Half of the United States Manraj Sahota1* and Makayla E. Boyle1*

Geisinger Commonwealth School of Medicine, Scranton, PA 18509 *Master of Biomedical Sciences Program Correspondence: msahota@som.geisinger.edu 1

Abstract Background: The opioid epidemic is a major problem within the United States, as 130 or more people die every day from opioid-related overdoses. To better understand this epidemic, 10 prescription opioids were chosen for study in four states within the southern half of the United States (Arkansas, Louisiana, New Mexico and Oklahoma). The opioids of interest included codeine, fentanyl, hydrocodone, hydromorphone, meperidine, morphine, oxycodone, oxymorphone, buprenorphine, and methadone. Methods: Data was collected from the Automated Reports and Consolidated Ordering System (ARCOS) database and National Institute of Drug Abuse on gram weight per 100 000 people of each opioid, opioids prescribed per 100 people, and opioid-related overdose deaths per 100 000 people. Analysis was done via conversion of gram weight to the oral morphine milligram equivalent (MME) and calculation of percent changes. Results: Overall, percent changes in opioids used for pain management, opioid use disorder (OUD) medications and population were calculated. Louisiana had 8.72%, -42.42%, and 10.22% changes, respectively. Arkansas had 33.22%, -22.59%, and 7.74% changes and Oklahoma showed 23.42%, 32.59%, and 10.11% changes, respectively. Finally, New Mexico showed some of the greatest changes with -4.46%, 119.02%, and 7.83%, respectively. Discussion: The prominent changes in prescription opioid use observed may be explained by laws, regulations, and guidelines on prescribing and use of the opioids in each state. Explanations for correlation between prescribed opioids and opioid-related overdose deaths are also included. Conclusion: The increasing importance of prescription drug monitoring programs (PDMP) is emphasized in that they may help to reduce the opioid epidemic seen throughout the United States and studied specifically in the southern half.

Introduction Opioids are typically prescribed for moderate to severe pain due to surgery, an injury, or from other health conditions such as cancer. Their use for chronic non-cancer pain is controversial. This study focused on eight opioids primarily used to relieve pain (codeine, fentanyl, hydrocodone, hydromorphone, meperidine, morphine, oxycodone, oxymorphone). The remaining two drugs, buprenorphine and methadone, are used for the treatment of an opioid use disorder (OUD). In the late 1990s, pharmaceutical companies reassured the medical community that opioid pain relievers are not addictive and health care providers across the country

began to prescribe them at greater rates (1). The increase in prescription of opioid medications led to widespread misuse, and it finally became clear that opioid medications were highly addictive. This realization led the U.S. Department of Health and Human Services to declare a public health emergency to address this national opioid crisis. Approximately 130 Americans die every day from an opioid overdose on average; about 46 of those deaths involve prescription opioids (2). Based on this data, it is clear that there is an opioid epidemic in the United States. This study examined trends in observed opioid use in four different states in the southern tier of the country to understand why these trends may be occurring. Though similar analyses have been done for states like Texas, this study addresses the dearth of literature about these trends in other southern states (3). As part of understanding the reasons for the trends, this report examined whether each of the four states offers a prescription drug monitoring program (PDMP). Such programs help prevent drug misuse and protect the health of the community. The Centers for Disease Control and Prevention (CDC) has committed to fighting this epidemic, supporting states to identify outbreaks, and providing care to those in communities that need it. The CDC even has guidelines for prescribing opioids for chronic pain, which helps ensure that patients have access to safer and more effective pain treatment while also helping to reduce the risk of OUD, overdose, and death (4). As of April 2018, 46 states have implemented activities to improve prescribing practices in their respective communities that align with the guidelines set by the CDC. Arkansas, Louisiana, Oklahoma, and New Mexico are among the 49 states that use a Prescription Drug Monitoring Program and among the 46 states that follow guidelines set by the CDC (5). Arkansas Senate Bill 717 allows the Department of Health to review the PDMP information to make sure that a person is not obtaining prescriptions in a manner that represents misuse. If the department discovers any indication of abuse, it notifies the practitioners, the licensing board of the prescriber, and the pharmacy that dispensed the prescription (6). Arkansas’s Emergency Departments (EDs) also have opioid prescribing guidelines, which includes 17 rules for providers working in the Emergency Department to follow related to prescribing opioids, including that “emergency medical providers should not provide replacement doses of methadone for patients in a methadone treatment program” and “EDs are encouraged to use the Arkansas Prescription Drug Monitoring Program on appropriate patients” (7). In 2017, Act 820 took effect in Arkansas and requires prescribers to check the PDMP when prescribing either schedule II (i.e., hydromorphone, methadone, meperidine, oxycodone, and fentanyl) or schedule III (i.e., buprenorphine) opioids.

Understanding the Opioid Crisis in the Southern Half of the United States

New Mexico has a prescription drug misuse and overdose prevention and pain management advisory council which meets to receive public comment (8). There is also the New Mexico Clinical Guidelines for Prescribing Opioids for Treatment of Pain that try to improve patient care and prevent misuse of opioids. The New Mexico Clinical Guidelines are categorized as “opioid treatment for acute pain” or “opioid treatment for chronic pain.” One example of a recommendation falling under the acute pain category is that “opioid medications should only be used for treatment of acute pain when the severity of the pain warrants that choice and after determining that other non-opioid pain medications or therapies will not provide adequate pain relief” (9). One example of recommendation from the chronic pain category is that the “provider should screen for risk of abuse or addiction before initiating treatment” (9). Oklahoma has implemented different rules and regulations in order to prevent opioid abuse-related deaths, including the Oklahoma Opioid Prescribing Guidelines, Oklahoma Emergency Department and Urgent Care Clinic (ED-UCC) Opioid Prescribing Guidelines, and a Prescribing and Dispensing Profile. The Prescribing and Dispensing Profile describes schedule II, III, and V prescribing limitations, miscellaneous prescribing/dispensing requirements (for all health care providers including optometrists and dentists), pain management regulations, and all the board guidelines in detail. One of the guidelines under the Oklahoma ED-UCC Guidelines states that “in patients suspected of opioid addiction, abuse, or diversion, health care providers should check the Oklahoma PDMP and perform screening, brief intervention, and referral to treatment, if indicated” (10). In 2013, Louisiana was ranked first in the nation in prescribing opioids and as such, accompanied misuse has been a major issue (11). However, throughout 2017, many policies were implemented to regulate opioid prescriptions and decrease opioid misuse. According to the Louisiana State Board of Medical Examiners, under Act 802, medical practitioners are prohibited from prescribing more than a 7-day supply when issuing a first-time opioid prescription for outpatient use to an adult patient with an acute condition. Falling under the same act are additional similar laws related to prescribing opioids for the health care providers of Louisiana. Act 76 also came into effect in 2017 and required all prescribers in Louisiana to receive 3 education credit hours from a one-time course as a prerequisite of license renewal in the first annual renewal cycle after Jan. 1, 2018, to learn about drug diversion training, best practices for the prescribing of controlled substances, and appropriate treatment for addiction.

Methods Procedures Data was collected from the publicly available Automated Reports and Consolidated Ordering System (ARCOS) database of prescription opioid use per state. ARCOS data from Veterans Affairs patients and other pharmacies treating military personnel, Indian Health Services, practitioners which dispense prescriptions, and Narcotic Treatment Programs is collected from manufacturers and distributors and then compiled by the Drug Enforcement Administration (DEA) in accordance with 92

the 1970 Controlled Substances Act. Currently, ARCOS has information publicly available from 2000 to 2017. State data was collected for New Mexico, Louisiana, Arkansas, and Oklahoma. The prescription opioids included were hydrocodone, oxycodone, fentanyl (base), morphine, hydromorphone, oxymorphone, codeine, meperidine, buprenorphine, and methadone. The data utilized for this study was quantified within ARCOS as gram weight per 100 000 people based on the 2010 U.S. census (12). Use of this data was approved by the IRB of the University of New England. Generalized data on the opioid epidemic in the United States was collected from the National Institute of Drug Abuse. This included opioid-related overdose deaths nationally and per state. Opioid-related overdose deaths are identified via underlying cause-of-death codes (X40–X44, X60–X64, X85, and Y10–14), which correlated to the cause of death due to heroin, other opioids, methadone, other synthetic narcotics, or unspecified narcotics, respectively. This data was then analyzed to compare opioids prescribed per state to the opioid-caused overdoses (13). Data analysis The weight in grams of each listed opioid was collected from ARCOS from 2006 to 2017 and subsequently converted to the milligram morphine equivalent (MME) using the following multipliers: codeine 0.15, hydrocodone 1, hydromorphone 4, morphine 1, oxycodone 1.5, oxymorphone 3, buprenorphine 10, methadone 12 and meperidine 0.1. (14, 15). The multiplier for methadone varies based on the dose per day, and so the upper multiplier of 12 was chosen since information on dose per day was not available. MME for fentanyl varies based on route of administration, as well as other factors. In general, fentanyl is considered to be about 75 to 100 times more potent than morphine, so 75 was chosen for the multiplier (16). Data on prescribed opioids was compared to the number of deaths in that state caused by opioid overdose. Data analysis and figure preparation was conducted using GraphPad Prism8.

Results and Discussion Table 1 summarizes the overall percent changes comparing 2006 to 2017 per opioid for each state in the southern half of the United States. Looking at Figure 1A, there is a sharp increase of MME per 100 000 people between 2014 and 2015 in Louisiana. According to the 2014 Louisiana Board of Pharmacy, specific laws or warning against codeine prescription was not advised. In fact, carisoprodol products were reassigned from schedule IV to schedule II except for the products that were combined with codeine; this indicates that products with codeine were seen as risky. However, as shown in Figure 1A, there was still an increase in prescription use of codeine during this time period. In addition, a major jump is observed in prescription use of codeine between 2013 and 2015, but no specific laws were passed regarding codeine use in Arkansas at this time and so the changes cannot be explained by legislation. In 2017, Act 820 passed requiring prescribers to check the Prescription Monitoring Program when prescribing schedule II or schedule III opioids. It can also be seen that Oklahoma had a major increase in prescription use of codeine between 2013 and 2015 (Figure 1A). In 2013, overdose deaths were much higher than motor vehicle crashes

Understanding the Opioid Crisis in the Southern Half of the United States

Figure 1. MME per 100K people shown for the following states: Louisiana, Arkansas, Oklahoma, and New Mexico. A) Codeine; B) Fentanyl; C) Hydrocodone; D) Hydromorphone; E) Meperidine; F) Morphine; G) Oxycodone; H) Oxymorphone; I) Buprenorphine; J) Methadone

Understanding the Opioid Crisis in the Southern Half of the United States

as the leading cause of unintentional injury in Oklahoma, with overdose being the leading cause of death for residents aged 25 to 64 (17). In order to decrease these statistics and help combat prescription drug abuse, authorities tried to limit the number of prescriptions that Medicaid recipients could receive and limit the number of pills that physicians could prescribe. “Shame tactics” were also used to help address key issues surrounding the treatment of chronic pain and stigma. Federal drug reclassification also occurred which banned prescriptions containing hydrocodone from being refilled by pharmacy. In addition, even a law was signed in 2013 that required patients to visit a physician every time they wanted such a prescription (18). Although a lot of changes were implemented in 2013 to help combat the increase in prescription drug abuse, there was still a surprising increase in use of prescription codeine in Oklahoma. Figure 1B shows that there is a stable pattern of MME per 100K people over the years for Oklahoma until 2015 where a notable drop in the use of prescription fentanyl is noticed. One reason for this major change could be that while a work group created ED-UCC Opioid Prescribing Guidelines in 2013 to help reduce the misuse, it might have taken some time to see a change. According to those guidelines, fentanyl was not allowed to be prescribed from the ED-UCC. Oklahoma also has opioid prescribing guidelines for health care providers in an officebased setting, which include guidance that fentanyl should not be used for treatment of acute pain since it is a long duration opioid. While a noticeable change in MME per 100K people occurred in Oklahoma, the other three states examined in this paper are all relatively stable over the years. Figure 1C demonstrates that New Mexico’s prescription use of hydrocodone is much lower than the other three states. According to Prescription Drug Misuse and Overdose Prevention and Pain Management Advisory Council, the state of New Mexico has been consistently changing prescription rules and regulations. The reason for their involvement and constant improvements is because New Mexico’s drug overdose death rate has been much higher than the national rate for many years; however, in 2015 New Mexico improved its national ranking due to these efforts. The low prescription use of hydrocodone could be explained by the rescheduling of hydrocodone from schedule III to schedule II, which came into effect on Oct. 6, 2014. Refills of prescriptions are allowed for schedule III drugs, but not schedule II drugs, meaning that any prescriptions written after Oct. 6, 2014 could not be refilled. Ultimately, the total amount of hydrocodone dispensed by New Mexico pharmacies declined between 2014 and 2015, as did the number of patients prescribed hydrocodone. Figure 1C shows that Louisiana experienced a prominent increase in prescription hydrocodone use from 2006 to 2007, but then the use dropped back again from 2007 to 2008. According to the Louisiana Board of Pharmacy, the PDMP was enacted in 2006, which directed the Board of Pharmacy to establish and maintain a database of prescription transactions for controlled substances dispensed to Louisiana residents, which may explain the notable jump we see in hydrocodone use. Over time, the use of prescribed hydrocodone decreases, potentially due to the 2009 establishment of a PDMP advisory council that included 24 state organizations that meet regularly to provide guidance on the program.

Louisiana experienced a decrease in the prescription use of hydromorphone (Figure 1D), which may be due to some changes that were made in the already established ACT 865. One of these changes included that prescription of schedule II drugs expires 90 days after the issue. Previously, prescription of schedule II drugs would expire 6 months after the issue. Another rule added to the ACT was that the prescriber must access the patient’s history in Louisiana PDMP prior to prescribing schedule II drugs to patients for treatment of non-cancer related chronic pain. Oklahoma experienced a similar trend to Louisiana, where the use of prescribed hydromorphone began to decrease in 2014, which may be explained by the passing of House of Bill 1948. The bill originally failed to win approval, because it required doctors to check PDMP every time they wrote or refilled a prescription for schedule II/III substances, including oxycodone. Providers did not support this law, because it was tedious work that they did not have the time, patience, or staff to complete. Oklahoma’s government decided to change some of the requirements, so the new bill only required doctors to check PDMP before prescribing for the first time and every 180 days thereafter. While all states experienced a decreasing trend in the use of prescribed meperidine (Figure 1E), only a few can be explained by changes in the law and guidelines. In 2015, changes were made to the controlled substances status, limiting opioid prescription to 10 days if the prescriber is not licensed by the state of Louisiana. In 2016, changes were made once again to the controlled substances status, stating that there would be no limit for patients with cancer or another terminal illness. Similarly, in 2015, a law began to require Arkansas hospitals with an emergency department to adopt guidelines concerning prescribing of opioids in the emergency department. In those guidelines, it was stated that “the administration of Demerol (Meperidine) in the ED is discouraged” (19). The use of prescribed morphine in the state of Oklahoma is much higher compared to the other states, as seen in Figure 1F. The prescribing guidelines in Oklahoma indicating that morphine is rarely used for acute pain, are simply meant to be an educational tool based on expert opinion (20). However, the guidelines do discuss that the risk-benefit ratio is less favorable at higher doses. It is recommended that physicians be careful when prescribing opioids at all dosage levels but especially at higher doses. According to these guidelines, no clear threshold had been established for a high dose. Although a threshold of 120 MME per day was suggested, there was no evidence to support this cap. Thus, it is possible that physicians did not have enough knowledge on how much morphine to prescribe and may have exceeded the suggested limit, explaining the higher use of morphine in Oklahoma. Some laws in Louisiana may explain the observed increasing pattern in opioid use (Figure 1G). In 2014, Act 192, which limited pharmacists to dispense a maximum of a 10-day supply for schedule II (i.e., oxycodone) and III drugs, was passed. The act also required pharmacists to inform the prescriber of the limited dispensing and cancellation of any authorized refills and prohibited the dispensing of that same medication to the same patient when prescribed by a non-Louisiana licensed practitioner for 60 days. In 2015, the same act was amended to state that if the PDMP information from the state where the prescriber is located is available to the dispensing pharmacist,

Understanding the Opioid Crisis in the Southern Half of the United States

then the limitation of 60 days does not apply. In 2016, that act was changed again to include no limitation on dispensing if the prescriber confirms a diagnosis of cancer or terminal illness for the patient. There was also a major increase in prescription oxycodone use for Oklahoma, but use began to decrease around 2015, which may be due to the passing of House Bill 1948 in 2015. As discussed previously (21, 22). No prominent changes in oxymorphone use were observed for any of the states (Figure 1H). Accordingly, no information was found regarding new guidelines or laws from 2006 to 2017. Figure 1I shows that the prescription use of buprenorphine increased over time in all four states. One possible reason for this consistent increase is that the DEA placed buprenorphine and all products containing buprenorphine into schedule III in 2002 (23). Since then, abuse of buprenorphine has become more common in the United States. Nothing under the Controlled Substances Act or the DEA regulations impose any limitations on medical staff to maintain a person with buprenorphine, which means that a patient with an opioid dependency who is admitted to a hospital for a primary medical problem other than opioid dependency may be administered opioid medications, such as buprenorphine (24). Another possibility for the continuous increase could be that strict regulations on buprenorphine are not in place throughout the country.

Figure 1J shows that the use of prescribed methadone in Louisiana dropped drastically between 2007 and 2008 and continuously thereafter. In Louisiana, Act No. 166 was passed in 2008 which extended the inhibition on methadone maintenance programs. These programs allow for long-term prescribing of methadone as an alternative to any other opioid on which a client is originally dependent (25). One major benefit of allowing the use of methadone over other opioids is that methadone is well-known to alleviate the symptoms of opioid withdrawal (26). Although the banning of methadone programs cannot explain the prominent decrease seen between 2007 and 2008, the act does help explain why methadone prescription in Louisiana continues to decrease after 2008. New Mexicoâ&#x20AC;&#x2122;s methadone prescription is relatively stable until 2010 during which an increase is observed and continues. According to the New Mexico Clinical Guidelines on Prescribing Opioids, methadone is rarely indicated for acute pain treatment and should only be prescribed by clinicians familiar with it (9). Although there is no clear information for the observed increase shown in Figure 1J, the New Mexico Department of Health reported that since the late 1990s, methadone became popular among physicians as an effective and somewhat affordable analgesic, leading to an increase in the number of prescriptions dispensed for methadone by 700% between 1998 and 2006. This data could lead one to speculate that there is a chance that New Mexico experienced the same trend within the state and this could also help explain the continuous increase in use of prescribed methadone in New Mexico. As previously stated, overall percent changes were calculated for the groupâ&#x20AC;&#x2122;s opioids for pain management and OUS medications. Along with these results, percent changes in population were also calculated for each of the states. Overall, Louisiana showed a -42.42% change in OUD medications and an 8.72% change in opioid for pain management, with an increase in the population of only 10.22%. Arkansas had a -22.59% change and a 33.22% change for OUD medications and pain management opioids, respectively, with a 7.74% population increase. Oklahoma had a 32.59% change and a 23.42% change in OUD drugs and opioids for pain management, respectively, and a population increase of 10.11%. Lastly, New Mexico had a 119.02% change in OUD medications and a -4.46% in pain management opioids, but increased in population by 7.83%. Based on these results, it can be stated that some changes in prescription opioid use may also be attributed to population, as well as the previously described laws and regulations for each state or prescription.

Figure 2. Comparison of overdose deaths and prescription opioids. Overdose deaths per 100K people in 2016 are shown in pink. Overdose deaths per 100K people in 2015 are shown in blue.

Finally, Figure 2 showed an inverse correlation between number of prescribed opioids and opioid-related overdoses, with states with higher prescriptions actually having lower opioid-related overdoses. No specific explanation was found for this correlation. It may be explained by the possibility that people may obtain opioids without a prescription. If this were the case, more overdoses may be seen without a correlated increase in prescriptions.

Understanding the Opioid Crisis in the Southern Half of the United States

Conclusion Overall, some changes seen in opioid use may be attributable to increases in the population but are unlikely to account for all of the changes. Therefore, laws and regulations, such as those previously discussed, as well as other factors, likely contribute to these changes. Furthermore, higher amounts of opioids prescribed do not correlate to a higher number of overdoses. In fact, our analyses of the selected opioids and states indicate that they are inversely related. This may be due to people obtaining opioids from sources not captured by our data. Further research is necessary to understand this inverse correlation. Another area in which further research is necessary is figuring out the success of implementation of state policies. As discussed earlier, the majority of policy influencing pain management recommendations and impacting opioid prescribing is typically enacted at state government level. Although there has been an increase in the number of states mandating some level of physician CE focused on pain management, there are still only 5 states that require it (27). PDMP policies are another way for states to promote balance between adequate pain management and preventing harm that is associated with the abuse of prescription opioids. To date, all evaluations of PDMP effects are observation and findings related to PDMP effectiveness in impacting proximal and distal outcomes are mixed and physician awareness, registration and use of PDMPs remains low. Proximal outcomes are defined by actions such as “promoting physician confidence in prescribing decisions, enhanced detection of abuse, or overdose” while distal outcomes are defined by “reducing overall prevalence of diversion and overdose” (27). While policies adopted by a number of states have shown some potential to increase registration and use of PDMPs through data such as a decrease in prescribing commonly abused opioids, research is still limited in the area. “Additional efforts are necessary to evaluate the benefit-risk balance of the wide array of PDMP policies being implemented from state to state" (27). Most states have an online prescription monitoring program and pharmacies are required to log every prescription for controlled substances within 5 minutes of filling it. This is surprising, because doctors are not even required to check the PDMP registry before prescribing to make sure the patient they are prescribing to be not already obtaining prescriptions from other physicians. This was a major takeaway from this study— the guidelines are simply there to guide. Unfortunately, even though it is highly recommended for doctors to follow such guidelines and check the PDMP, many health care providers do not adhere to these guidelines. It is crucial for health care providers to understand the seriousness of the opioid epidemic, and how efficacious a tool PDMPs can be in saving lives when used.

Acknowledgments

supported by Brian Piper, PhD, and Elizabeth Kuchinski, MPH, faculty advisors for the Community Health Research course at Geisinger Commonwealth School of Medicine’s MBS program, and Kenneth McCall, PharmD, at the University of New England.

References 1.

Opioid Overdose Crisis [Internet]. Drugabuse.gov. 2019 [cited 23 April 2019]. Available from: https://www. drugabuse.gov/drugs-abuse/opioids/opioid-overdosecrisis.

2. Opioid overdose [Internet]. Centers for Disease Control and Prevention. Centers for Disease Control and Prevention; 2018 [cited 2019Apr8]. Available from: https:// www.cdc.gov/drugoverdose/epidemic/index.html. 3. Ighodaro ET, McCall KL, Chung DY, Fraiman J, Nichols SD, Piper BJ. Dynamic changes in prescription opioids from 2006 to 2017 in Texas. Dynamic changes in prescription opioids from 2006 to 2017 in Texas. Geisinger Commonwealth School of Medicine. 4. Opioid Overdose [Internet]. Centers for Disease Control and Prevention. Centers for Disease Control and Prevention; 2019 [cited 2019May14]. Available from: https:// www.cdc.gov/drugoverdose/prescribing/guideline.html. 5. State Profiles | The PDMP Training and Technical Assistance Center. [cited 2019May14]. Available from: http:// www.pdmpassist.org/content/state-profiles. 6. State of Arkansas (2015). Senate bill 717. 7.

Arkansas Medical Society. (2019). Prescription Drug AbuseArkansas Medical Society. [online] Available at: https:// www.arkmed.org/resources/prescription-monitoring/ [Accessed 23 Apr. 2019].

8. Prod.nmhealth.org. (2019). Opioid Safety. [online] Available at: https://prod.nmhealth.org/about/erd/ibeb/pos/ [Accessed 23 Apr. 2019]. 9. New Mexico Department of Health. New Mexico clinical guidelines on prescribing opioids for treatment of pain [Internet]. 2011. Available from: https://nmhealth.org/ publication/view/general/271/. 10. Opioid Prescribing Guidelines for Oklahoma Workgroup. Oklahoma emergency department (ED) and urgent care clinic (UCC) Opioid prescribing guidelines [Internet]. 2013. Available from: https://www.ok.gov/health2/documents/ UP_Oklahoma_ED-UCC_Guidelines.pdf. 11. Knecht P, Kuy S. Frequently asked questions about opioid prescribing [Internet]. Louisiana Department of Health; 2017 [cited 17 April 2019]. Available from: http://ldh.la.gov/assets/ opioid/OpioidFAQFactSheet.pdf.

We would like to thank Charles Poli for helping us with the navigation of Prism8.

12. Automation of reports and consolidated orders system (ARCOS). [cited 2019Apr8]. Available from: https://www. deadiversion.usdoj.gov/arcos/index.html.

We would also like to acknowledge Ebuwa Ighodaro, who worked on an ARCOS-based research project in Texas. He allowed us to take his ideas and connect them with the entire southern tier of the United States. Lastly, this research was

13. National Institute on Drug Abuse. Opioid summaries by state [Internet]. NIDA. 2019 [cited 2019Apr8]. Available from: https://www.drugabuse.gov/drugs-abuse/opioids/ opioid-summaries-by-state.

Understanding the Opioid Crisis in the Southern Half of the United States

14. Centers for Disease Control and Prevention. Calculating total daily dose of opioids for safer dosage [Internet]. U.S. Department of Health and Human Services;. Available from: https://www.cdc.gov/drugoverdose/pdf/calculating_ total_daily_dose-a.pdf. 15. Opioid morphine equivalent conversion factors [Internet]. [cited 9 April 2019]. Available from: https:// www.cms.gov/medicare/prescription-drugcoverage/ prescriptiondrugcovcontra/downloads/opioid-morphineeq-conversion-factors- march-2015.pdf.

26. Overview of methadone maintenance treatment-Portico [Internet]. Porticonetwork.ca. 2019 [cited 17 April 2019]. Available from: https://www.porticonetwork.ca/web/ knowledgex- archive/amh-specialists/overview-mmt. 27. Barth KS, Guille C, Mccauley J, Brady KT. Targeting practitioners: A review of guidelines, training, and policy in pain management. Drug and Alcohol Dependence [Internet]. 2017Apr1;173. Available from: https://www.ncbi. nlm.nih.gov/pmc/articles/PMC5555357/.

16. McPherson M. Demystifying opioid conversion calculations. Bethesda, Md.: American Society of Health-System Pharmacists; 2010. 17. Hydrocodone no longer No. 1 drug prescribed to Oklahoma's Medicaid patients [Internet]. NewsOK.com. 2019 [cited 17 April 2019]. Available from: https://newsok. com/article/5396314/hydrocodone-no-longer-no-1-drugprescribed-to- oklahomas-medicaid-patients. 18. Address patient shame, stigma when treating opioid misuse [Internet]. American Medical Association. [cited 2019May14]. Available from: https://www.ama-assn.org/ delivering-care/opioids/address-patient-shame-stigmawhen-treating-opioid-misuse. 19. Arkansas emergency department opioid prescribing guidelines [Internet]. [cited 23 April 2019]. Available from: https://www.arkmed.org/wp- content/uploads/2013/07/ Arkansas_EmergencyDepartment_Opioid_Prescribing_ Guidelines-1pg.pdf. 20. Opioid prescribing guidelines for Oklahoma health care providers in the office-based setting [Internet]. 2014 [cited 17 April 2019]. Available from: https://www.ok.gov/health2/ documents/UP_Oklahoma_Office_Based_Guidelines.pdf. 21. Cox, Morrissette, and Ownbey (House) and Griffin (Senate) (2015). Enrolled house bill no. 1948. 22. Oklahoma Bureau of Narcotics and Dangerous DrugsPMP Mandatory Check. [cited 2019May14]. Available from: https://www.ok.gov/obndd/Prescription_Monitoring_ Program/PMP_Mandatory_Check.html. 23. Buprenorphine [Internet]. Drug Enforcement Administration Office of Diversion Control Drug and Chemical Evaluation Section; 2013 [cited 23 April 2019]. Available from: https://www.deadiversion.usdoj.gov/drug_chem_info/ buprenorphine.pdf. 24. Special circumstances for providing buprenorphine | SAMHSA-Substance Abuse and Mental Health Services Administration [Internet]. Samhsa.gov. 2019 [cited 17 April 2019]. Available from: https://www.samhsa.gov/medicationassisted-treatment/legislation- regulations-guidelines/ special. 25. State of Louisiana-department of health and hospitals office of behavioral health request for applications (RFA) community canvassers [Internet]. Louisiana Department of Health and Hospitals Office of Behavioral Health; 2008 [cited 17 April 2019]. Available from: http://ldh.la.gov/assets/ docs/BehavioralHealth/OTPRFA82712.pdf.

Scholarly Research In Progress • Vol. 3, November 2019

Using Implementation Science to Address the Opioid Crisis: Deciphering Vocabulary, Adopting Sustainable Practices, and Overcoming Challenges Niraj J. Vyas1*

Geisinger Commonwealth School of Medicine, Scranton, PA 18509 *Master of Biomedical Sciences Program Correspondence: nvyas@som.geisinger.edu 1

Abstract

Methods

Overdose deaths and opioid abuse continue to increase in the United States, posing a challenge to the health care system. As an effort to address the issue in the past decade, interventions such as prescription drug monitoring programs, drug take-back programs, changes to opioid prescription guidelines and distribution of short-acting opioid antagonists have been implemented. The extent of their impact, however, is not clearly measured. In this review, basic components of implementation science, an emerging ﬁeld that seeks to maximize effectiveness and efﬁciency of evidence-based interventions (EBIs) using feedback-based and adaptive methods, were examined. Further, the numerous components involved in implementation science and their relationship to each other were defined. Promoting Action on Research Implementation in Health Services (PARiHS) and the Consolidated Framework for Implementation Research (CFIR) are presented as example frameworks with an analysis of their strengths and weaknesses followed by recent examples of opioid interventions that utilize implementation science. In summary, while many interventions are being considered and integrated, the use of implementation science while constructing interventions from evidence could increase the effectiveness of new practices, allow for clearer data collection, and even indicate necessary augmentation to interventions.

Google Scholar was used to identify articles from credible databases, such as MEDLINE/PubMed, for this review. Articles investigating opioid trends in the United States were selected based on inclusion of various opioids and length of study. Papers investigating these trends included those published before and after the revision of the CDC opioid prescribing guidelines in 2016. The types of current opioid interventions were identified by volume of previous research in the last 10 years. The example frameworks in this paper are among those which have been consistently studied and incorporated in health care. Example interventions using implementation science were selected if the study was published in the last two years and a specific framework was identified in the creation of an opioid intervention.

Introduction Although implementation science has started to gain some traction among health care providers and researchers in recent years, it still remains a convoluted field of jargon and meta-analysis. Implementation science aims to improve the quality and effectiveness of health care by studying the method by which research findings and evidence-based interventions (EBIs) are assimilated into routine practice (1). We witness new research findings and EBIs on a daily basis; however, most take years to make their way to patients, if they ever do. Implementation science investigates why or how the process of integrating these findings was successful or did not work as planned. Identifying the determinants of successful and failed clinical implementation allows for increased efficiency in changing health care practices to deliver effective patient care (2). In this review, current trends in the opioid crisis and practices already implemented to address opioid abuse were examined. Additionally, we aimed to simplify the complex vocabulary associated with implementation science and address challenges in using implementation science to increase the efficiency of opioid abuse interventions.

Results State of the opioid crisis Over the past decade, opioid abuse and overdose deaths have increased but specific statistics within this trend continue to be very fluid. In 2015, over 33,000 drug overdoses involved an opioid (3). In 2016, there was a total of 63,632 drug overdose deaths, of which nearly two thirds involved prescription opioids, illicit opioids, or both—an increase of 27.7% from 2015 (4). The peak age for overdose deaths was between 45 and 54 years (3), an unsurprising association because older adults account for a significant portion of opioid consumption. Older adults, due to higher rates of illness and medical complexities, are likely to be prescribed opioids with a low daily dose. It should be noted that prescribed opioids in older adults are an iatrogenic component of the opioid epidemic and that prescription opioids could lead to heroin and/or fentanyl use in some patients (5). Despite an 18% decrease from 2010 to 2015 in morphine milligram equivalent (MME) per capita, there was an increase in daily supply (6). It is estimated that 1,124 metric tons of opioids were prescribed in 2016, the equivalent of a sevenyear supply of 5-mg tablets of hydrocodone per capita in the US (5). Between 2004 and 2011, there was a 20% decrease in the volume of codeine, but an increase in the percentage of fentanyl, methadone, morphine, hydrocodone, oxycodone, hydromorphone, and buprenorphine (7). However, there was a pronounced reduction of all of these opioids between 2015 and 2016 with the exception of buprenorphine (5). From 2011 to 2016, there was an increase in drug overdose deaths mentioning at least one specific drug or substance (73% to 85%), a decrease in drug overdose deaths that mentioned

Using Implementation Science to Address the Opioid Crisis

only a drug class but not a specific drug or substance (5.1% to 2.5%), and a decrease in the percentage of deaths that did not mention a specific drug or substance (22% to 13%). Although rankings may have shifted every year, the top 10 drugs in overdose deaths remained consistent from 2011 to 2016 across three categories: opioids (fentanyl, heroin, hydrocodone, methadone, morphine, and oxycodone), benzodiazepines (alprazolam, diazepam) and stimulants (cocaine, methamphetamine) (8). While oral use has remained the primary nonmedical route of prescription opioid abuse, 41.1% of overdoses are via inhalation, smoking and injection (9). The data mentioned here along with numerous other studies demonstrate the necessity of implementing interventions, especially since the opioid crisis has adversely impacted the U.S. life span (10). Current interventions Recognizing the growing severity of opioid abuse in the U.S., health care providers, researchers, organizations, and policymakers have taken various courses of action to curb the crisis. The recent reduction of opioids has been attributed to a number of interventions: increased public awareness of prescription opioid overdoses and the opioid crisis (3), development of new prescription guidelines for chronic pain (11), adjusting daily MME doses to prevent overdose (12), and creation of legislation to address pill mills (13). It should be noted that the quality of data from these interventions tend to be low and that the empirical evaluations are unsatisfactory (14). Prescription drug monitoring programs (PDMPs) are very common and easily implemented. Unfortunately, these programs tend to have low study quality and limitations in experimental design and capacity for longitudinal observation. There is often a lack of a comparison group or even baseline data due to small sample sizes, which combined with selfreported outcomes and only short-term follow-ups, result in inadequate statistical testing and significance. Additionally, these programs tend to measure prescribing practices rather than health outcomes (15). Maine has implemented a Diversion Alert (DA) Program—an innovative tool that distributes prescription drug arrest registry information to prescribers and pharmacists. The database includes, among other information, patient name, town of residence, drug-related charge and implicated drugs. This resource allows health care professionals to identify prescription drug abuse or illicit drug use (14). Additionally, the DA serves an educational source for new registrants on how to respond to patients charged with drug crimes and as tool to provide pharmacoepidemiologic information (16). However, there are limitations to the DA program. It is a voluntary service; therefore, not all entities report to DA and it is highly underutilized. It only reports drugs detected by field tests and drug arrests (excluding minors), not convictions. It has been suggested that DA, with a refined data entry system for illicit and prescribed drugs, be merged with PDMP in order to increase the effectiveness of prescription drug monitoring (14). Drug take-back programs have been studied and implemented. Gray and Hagemeier investigated drug take-back programs in rural Appalachia (17) and Stewart examined a similar program in Maine (18). Both studies found that controlled medications totaled approximately 9% of their collection, of which

hydrocodone combinations were the largest proportion of returns (32% and 21.2%, respectively), followed by oxycodone and oxycodone combinations (11% and 21.2%, respectively). It should be noted that Maine had more individual participants in the drug take-back program. Maine also, however, had fewer individual medications returned. These studies reported the medications returned but did not assess reasons for returning the medications. Speculated reasons include over prescription, adverse drug events, ineffectiveness of the drug, or lack of adherence (18). Short-acting opioid antagonists can be used to treat individuals who have overdosed. Antagonists, such as naloxone, have a long and safe history of use in effectively treating opioid overdoses by organized health care systems and providers. Although the safety of naloxone administration by non-healthcare professionals has not been formally established, it will likely parallel medically supervised experiences in the future (19). Many organizations have already started to distribute to non-health-care professionals, such as the free distribution of naloxone to Pennsylvania residents in December 2018. It should be noted that naloxone and opioid overdose training programs should accompany the distribution of antagonists. Many drug users do not know the appropriate overdose prevention strategies or have the necessary skills to respond to an overdose. Studies indicate individuals are reluctant to call emergency services or seek medical attention due to fear of police arrest. However, 20% of people who receive training felt it was unnecessary to call emergency medical services because they believed naloxone administration was enough to address overdoses (20). Research has experimented with developing more difficult to abuse opioids. For example, buprenorphine, an opioid rarely preferred for its inherent euphorigenic properties, has seen a steep increase in use in the last five years, particularly in those who use heroin. It appears that buprenorphine serves as a substitute for other drugs, including heroin, and is used as an alternative to methadone in self-medicating withdrawal symptoms (21). There is concern that difficult to abuse opioids facilitate a transition from oral and nasal routes to injectable routes of abuse (14). A study in Indiana found that a network of drug users began to inject extended-release oxymorphone which led to the introduction and rapid transmission of HIV between 2015 and 2016 (22). Educating health care professionals to be more judicial and cautious when prescribing opioids has become an important intervention in the opioid crisis. To spearhead the effort, the CDC released new guidelines for prescribing opioids in 2016 in order to improve communication between clinicians and patients about opioid therapy for chronic pain, improve effective pain treatment, and reduce risks associated with long-term opioid use (11). The guidelines come as a response to primary care physicians’ concerns regarding opioid pain medication misuse and patient addiction, specifically expressing increased stress when managing patients with chronic pain. Additionally, they report insufficient training in prescribing opioids (23). The CDC reports the use of the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) framework in order to implement evidence-based interventions and researching findings into the guidelines. Recommendations were developed and placed into 99

Using Implementation Science to Address the Opioid Crisis

three recommendations in determining when to initiate or continue opioids for chronic pain. Clinicians should consider opioid therapy only if expected benefits for both pain and function outweigh risks to the patient or outweigh the benefits of nonpharmacologic therapy or nonopioid pharmacologic therapy. Clinicians are advised to also use nonpharmacologic therapy and nonopioid pharmacologic therapy in combination if opioids are prescribed. Additionally, clinicians need to be able to set realistic goals and discuss the treatment with their The CDC published a total of 12 recommendations in the patients initially and periodically during time of treatment, 2016 guidelines across three categories. The report includes emphasizing that opioid therapy will be discontinued if the benefits do not outweigh the risks. Four recommendations were suggested for opioid selection, dosage, duration, follow-up, and discontinuation. For chronic pain, clinicians should prescribe immediate-release opioids instead of extended-release/longacting (ER/LA) opioids with lowest effective dosage (carefully assessing dosage at 50 MME, with no more than 90 MME per day without proper justification). The report makes note that a supply of three days or less should be sufficient for acute pain. Lastly, the CDC offers five suggestions for addressing risk and harm of opioid use. It is recommended that Table 1. Interpretation of recommendation categories and evidence type, adapted from CDC clinicians review the patient’s history guidelines (11) of controlled substance prescriptions, such as benzodiazepines, (via PDMP, DA, or similar programs) when starting opioid therapy and periodically during opioid therapy. Additionally, urine drug testing should be used initially and periodically to assess the use of other controlled prescription drugs or illicit drugs. Lastly, health care professionals should offer evidence-based treatment for patients with opioid use disorder (11). categories (A or B) based on quality of evidence, net benefit versus harm, values and preferences, and resource allocation (Table 1). Three recommendation categories were developed: no evidence indicating a long-term benefit of opioids (with outcomes examined for at least one year), extensive evidence indicating possible harms of opioids, and extensive evidence indicating some benefits of nonpharmacologic and nonopioid pharmacologic therapy (11).

Frameworks, theories, and models

Figure 1. Adapted outline of Ostrom’s framework for institutional analysis (30) with common terms and definitions in implementation science (29) 100

Implementation science addresses the need for a consistent process across multiple disciplinary languages, multiple levels of analysis, relevant aspects of the world, and cultural phenomena by three levels that often confused with one another: frameworks, theories, and models (Figure 1). A particular problem is analyzed at different degrees of speciﬁcity at each of these levels. Identiﬁcation of elements and relationships between them occurs through the development and use of a general framework. In addition to providing a general list of variables, frameworks use metatheoretical language to compare theories, organize diagnostic and prescriptive inquiry, and help generate

Using Implementation Science to Address the Opioid Crisis

the questions that need to be addressed for analysis (30). A framework can provide an extensive paradigm, which places emphasis on professional consensus within a particular scientific community. It does not need to indicate specifics of the relationships, such as direction, or identify critical hypotheses (31). Theories make general working assumptions about specific and relevant elements of a framework. Analysts use theories to diagnose a phenomenon, explain its processes, and predict outcomes. It should be noted that several theories can be compatible within a single framework. Models, such as logic, mathematics, game theory, experimentation, and simulation, are used to systematically assess the consequences of assumptions made about a limited set of parameters and variables (30). Models offer a way of testing conceptual coherence between the different levels within implementation science (31). When systematic, comparative institutional assessments are not implemented, recommendations and interventions may be based on superficial ideas instead of performance analysis. It is essential to use a common framework with an appropriate family of theories to address the questions of reform. Welltailored models are then used to predict the outcomes of these highly simplified structures. Errors can occur if inappropriate models are used to study problematic situations that do not fit the assumptions of the model. As a result, institutions are not able to properly respond to fluid environments and the sustainability of resources and investments is negatively affected. These outcomes are affected by actions taken and the cumulative effects of three levels of rules that govern them. Operational rules directly affect daily decisions made by participants in any environment. Operational activities are managed by collective-choice rules, which determine eligible participants and the rules involved in changing operational rules. Constitutional-choice rules determine who is eligible to craft collective-choice rules and how those changes are to be made (30). Implementation science works to provide insight on both the processes related to practice change and the outcomes from those changes. Effective research designs combine clinical effectiveness with appropriate implementation research. These hybrid research designs evaluate the change process and measure patient outcomes in the same study (Figure 1). The method by which implementation is accomplished can vary. A hybrid design can include gathering process measures when a new innovation is first placed into practice. Secondly, an intervention and an implementation can be tested simultaneously. Lastly, the implementation strategy can be tested while gathering information on clinical outcomes. Regardless of the exact methodology, all hybrid research designs make a deliberate effort to systematically document patient and process outcomes, which are ideally identified a priori as part of the experimental design (2). There is some discussion that the classification of all strategies under a single term may constrain efforts to synthesize findings across tests of strategy effectiveness and limit advancement of the full range of strategies of importance

to implementation science and practice. In addition, the term “implementation strategy” may result in essential strategies to implementation science being overlooked. As a result, it has been recommended to classify strategies according to who enacts them. These actors include delivery system actors, who adopt and integrate EBIs into their practice settings, and support system actors, who promote and support EBI adoption and implementation with a focus on building the general capacity of the delivery system’s ability to implement EBIs. Synthesis and translation system actors are responsible for identifying, translating, and disseminating EBIs (29).

Discussion One approach to opioid interventions is the use of the PARiHS framework, which represents successful implementation (SI) as a function of nature and type of evidence (E), the qualities of the context (C) in which the evidence is being introduced, and the method of facilitation of the entire process (F); SI=f (E, C, F). Like many other conceptual frameworks, the PARiHS framework also includes the identification of elements and relationships, explicit or implicit embedded theoretical positions, and a methodology to explain a complex set of phenomena which allows for action (Figure 2) The distinguishing factor of the PARiHS framework is the ability to map the interrelationships and the potential to be used as a practical and pragmatic tool by practitioners and researchers at the local level (31). The PARiHS framework, like all others, has assumptions behind its main features. The framework includes research evidence, clinical experience, professional craft knowledge, patient preferences and experiences, local information, and other codified and non-codified sources of knowledge. It should be noted that a shared understanding of benefits, drawbacks, risks and advantages of the new over the old is required in melding and implementing evidence into practice. In other words, this dialectical process involves many components requiring careful management and choreography. It cannot be completed in isolation and requires a team effort. Contexts involving “transformational leaders,” components of learning institutions, and proper evaluation have historically been more conducive to the successful implementation of evidence into practice. The likelihood of successful implementation improves with appropriate facilitation. Specifically, the “state of preparedness” of an individual or team, regarding their acceptance and understanding of evidence, receptivity of their place of work or context in terms of resources, culture and values, leadership

Figure 2. Interrelationships between elements of PARiHS framework and linked theories and models outlined using Ostrom’s typology (31)

101

Using Implementation Science to Address the Opioid Crisis

style, and evaluation strategy, determines the type of facilitation and the role and skill of the facilitator (31). One suggestion to test the state of preparedness is to aggregate responses to the questions and translate them onto a grid that plots the positions the team judges themselves to be in before the initialization of the implementation process (Figure 3). F1 indicates a facilitation method for transforming weak context and strong evidence into a highly receptive context. On the other hand, a method to manage strong context and weak evidence situations are indicated by F3, which often involves issues of routine and power. The most challenging facilitation method is F2, which manages both weak context and weak evidence, often involving issues of safety and basic competence. Many sources of information will be required to decide on an appropriate course of action; therefore, the diagnostic score can serve as an indication of the starting point. As the intervention, facilitation can hybridize approaches ranging from task-focused to more enveloping processes, such as personal development and action learning.

Figure 3. Adapted PARiHS diagnostics and evaluative grid (31)

The role of the facilitator is to assess the situation, individuals, team, and workplace readiness to develop change and evaluation strategies, to support the implementation process, and to mentor the team through changes. Kitson suggests an analogy of a chess game to try to understand the task at hand. Metaphorically, there is a chessboard in front of us with game pieces (PARiHS elements and sub-elements) whose moves need to be tested. Once the movements of these pieces are recognized, games can be set up (i.e., particular interventions) to see what happens. Research endeavors will produce guiding principles for the moves that practitioners can use in order to successfully implement research into practice.

102

It should be noted that practitioners do a significant portion of the work to transform principles into effective actions in their workplaces (31). Echoing the PARiHS framework, CFIR is composed of five major domains: the intervention, inner and outer settings, individuals involved, and the process by which implementation is accomplished. Adaption is a necessity for interventions to be a good fit. Poor fit is usually a result of resistance by individuals who will be affected by the intervention, which requires an active methodology to engage individuals in order to properly accomplish implementation. Interventions have core components, the essential and indispensable elements of the intervention, and adaptable periphery elements (32). It is not unusual for changes in the “inner setting, containing structural, political and cultural contexts,” mediate changes in “the outer setting consisting of economic, political, and social contexts,” which ultimately influence implementation. However, there is a dynamic and precarious relationship between the inner and outer setting; therefore, it is not always easy to distinguish the two. Additionally, individuals, carriers of cultural, organizational, and professional mindsets, also can influence others “with predictable or unpredictable consequences for implementation.” Successful implementation requires an active change process that is aimed at both individual and organizational level use of the intervention. However, the implementation process does not need to be a single intervention; it can be an interrelated series of sub-processes that do not necessarily occur sequentially (32). Becker et al., recognizing the critical need to promote implementation of evidence-based practice, investigated training opioid addiction treatment providers to adopt contingency management. Their approach includes evaluation of the effectiveness of implementation models prospectively, comprehensive measurement of implementation effects via consideration of interplay between multiple domains that have been shown to influence implementation effectiveness, and use of theory to identify and test potential mechanisms of implementation effectiveness. Their results have two key implications for researchers and clinicians: 1) applying science to service laboratory (SSL) model in a large multi-state, multisite agency is a feasible option and indicates evidence of effectiveness; and 2) SSL was associated with increased contingency management adoption despite no effects on putative mediators, indicating a need for more comprehensive examination of mechanisms. Further evaluation of mediators could assist in implementation of SSL in community agencies by demonstrating how it promotes adoption of evidence-based practice and highlighting the most essential components of the

Using Implementation Science to Address the Opioid Crisis

model. Additionally, a stepped wedge cluster randomized trial to assign clinics to conditions was noted as a promising design (33). Evaluation of the magnitude of within-facility changes in an addiction treatment quality measure regarding pharmacotherapy for opioid use disorder over a period of two years was the primary aim of a study by Finlay et al. They investigated the proportion of patients that received pharmacotherapy for opioid use disorder at Veterans Health Administration (VHA) facilities to determine whether the rate of receipt increased or decreased over time and its magnitude of change. This information could provide insight into the mechanisms of change in performance over time and could assist in designing strategies to improve access and use of pharmacotherapy for opioid use disorder. Additionally, the study aimed to serve as an example of how health care quality surveillance can be aided by examining changes in a quality measure. Although only pharmacotherapy for opioid use disorder was the only quality measure this study was interested in, any measure could be examined over longer time periods to provide insight about the health care system. The study found that 53 of 129 facilities had a 6% or greater positive or negative change in receipt of pharmacotherapy for opioid use disorder. Additionally, it was difficult for facilities with average or high performance to maintain their performance over time. The results suggest that using multi-year snapshots of quality measures could be a more effective method to select facilities for comprehensive quality work. On the other hand, facilities with low performance that improve substantially provide the most useful information considering the mechanisms of change while locations indicating consistent low performance provide insight into how to protect high performance rates (34). Systems consultation, a novel implementation strategy designed to promote clinical guideline implementation for opioid prescribing in primary care, was the center of a report by Quanbeck et al. While blended approaches have demonstrated they increase the speed of implementation of evidence-based practices and clinical guidelines, the blend of strategies, each with its own set of evidence, in the systems consultation model had not been tested before. The study identifies potential improvements to allow for broader, more effective use. Additionally, a subsequent randomized trial could deliver an adaptive implementation strategy and identify the most essential elements of implementation strategies in various clinical settings. These findings can then be used to determine the most efficient methods of promoting clinical guideline adoption for opioid prescribing in primary care (35).

Conclusion The dynamics and growing state of opioid use disorder and overdose deaths in the United States requires effective, longitudinal, and self-evaluating evidence-based interventions. PDMPs, drug take-back programs, short-acting antagonists, and opioid prescription guidelines among other interventions have already been implemented; however, these approaches have no procedure in place to evaluate effectiveness, do not serve as a long-term solution, cannot be changed in response to evaluation, or some combination of these issues. Using

implementation science when developing or modifying opioid interventions could prove useful in ensuring the intervention works as intended and to its maximum potential. Through this review, it was demonstrated that frameworks such as PARiHS, CFIR, and potentially hybridized approaches like systems consultation, are essential in delivering consistent, dynamic, and effective care and longitudinal prevention.

Acknowledgments The author expresses sincere gratitude to mentor Brian J. Piper, PhD, for his continuous support and guidance in the research and writing of this paper.

Disclosures The author has no relevant disclosures.

References 1.

Eccles MP, Mittman BS. Welcome to implementation science. Implement Sci. 2006;1:1.

2. Williams JK, Feero WG, Leonard DG, Coleman B. Implementation science, genomic precision medicine, and improved health: A new path forward? Nurs Outlook. 2017;65:36–40. 3. Rudd RA, Seth P, David F, Scholl L. Increases in drug and opioid-involved overdose deaths–United States, 2010– 2015. MMWR Morb Mortal Wkly Rep. 2016;65:1445–1452. 4. Vivolo-Kantor AM, Seth P, Gladden RM, Mattson CL, Baldwin GT, Kite-Powell A, et al. Vital signs: trends in emergency department visits for suspected opioid overdoses–United States, July 2016-September 2017. MMWR Morb Mortal Wkly Rep. 2018;67:279–285. 5. Piper BJ, Shah DT, Simoyan OM, McCall KL, Nichols SD. Trends in medical use of opioids in the U.S., 2006–2016. Am J Prev Med. 2018;54:652–660. 6. Guy GP, Zhang K, Bohm MK, Losby J, Lewis B, Young R, et al. Vital signs: changes in opioid prescribing in the United States, 2006–2015. MMWR Morb Mortal Wkly Rep. 2017;66:697–704. 7.

Atluri S, Sudarshan G, Manchikanti L. Assessment of the trends in medical use and misuse of opioid analgesics from 2004 to 2011. Pain Physician. 2014;17:E119–128.

8. Hedegaard H, Bastian BA, Trinidad JP, Spencer M, Warner M. Drugs most frequently involved in drug overdose deaths: United States, 2011–2016. Natl Vital Stat Rep. 2018;67. 9. Jones CM, Christensen A, Gladden RM. Increases in prescription opioid injection abuse among treatment admissions in the United States, 2004–2013. Drug Alcohol Depend. 2017;176:89–95. 10. Dowell D, Arias E, Kochanek K, Anderson R, Guy Jr GP, Losby JL, et al. Contribution of opioid-involved poisoning to the change in life expectancy in the United States, 2000–2015. JAMA. 2017;318:1065–1067.

103

Using Implementation Science to Address the Opioid Crisis

11. Dowell D, Haegerich TM, Chou R. CDC guideline for prescribing opioids for chronic pain-United States, 2016. JAMA. 2016;315:1624–1645. 12. Bohnert AS, Valenstein M, Bair MJ, Ganoczy D, McCarthy JF, Ilgen MA, et al. Association between opioid prescribing patterns and opioid overdose-related deaths. JAMA. 2011;305:1315–1321. 13. Kennedy-Hendricks A, Richey M, McGinty EE, Stuart EA, Barry CL, Webster DW. Opioid overdose deaths and Florida’s crackdown on pill mills. Am J Public Health. 2016;106:291–297. 14. Piper BJ, Desrosiers CE, Fisher HC, McCall KL, Nichols SD. A new tool to tackle the opioid epidemic: description, utility, and results from the Maine Diversion Alert program. Pharmacotherapy. 2017;37:791–798. 15. Haegerich TM, Paulozzi LJ, Manns BJ, Jones CM. What we know, and don’t know, about the impact of state policy and systems-level interventions on prescription drug overdose. Drug Alcohol Depend. 2014;145:34–47. 16. McCall K, Nichols SD, Holt C, Ochs L, Cattabriga G, Tu C. Prescription monitoring program trends among individuals arrested in Maine for trafﬁcking prescription drugs in 2014. Pharmacotherapy. 2016;36:585–589. 17. Gray JA, Hagemeier NE. Prescription drug abuse and DEA-sanctioned drug take-back events: characteristics and outcomes in rural Appalachia. Arch Intern Med. 2012;172:1186–1187. 18. Stewart H, Malinowski A, Ochs L, Jaramillo J, McCall K, Sullivan M. Inside Maine’s medicine cabinet: ﬁndings from the Drug Enforcement Administration’s medication takeback events. Am J Public Health. 2015;105:e65–e71. 19. Wermeling DP. Review of naloxone safety for opioid overdose: practical considerations for new technology and expanded public access. Ther Adv Drug Saf. 2015;6:20–31. 20. Bennett AS, Bell A, Tomedi L, Hulsey EG, Kral AH. Characteristics of an overdose prevention, response, and naloxone distribution program in Pittsburgh and Allegheny County, Pennsylvania. J Urban Health. 2011;88:1020–1030. 21. Cicero TJ, Ellis MS, Surratt HL, Kurtz SP. Factors contributing to the rise of buprenorphine misuse: 2008– 2013. Drug Alcohol Depend. 2014;142:98–104. 22. Peters PJ, Pontones P, Hoover KW, Patel MR, Galang RR, Shields J, et al. HIV infection linked to injection use of oxymorphone in Indiana, 2014–2015. N Engl J Med. 2016;375:229–239. 23. Jamison R, Sheehan K, Scanlan E, Matthews M, Ross E. Beliefs and attitudes about opioid prescribing and chronic pain management: survey of primary care providers. J Opioid Manag. 2014;10:375–382. 24. Proctor EK, Powell BJ, McMillen JC. Implementation strategies: recommendations for specifying and reporting. Implement Sci. 2013;8:139.

104

25. Wandersman A, Duffy J, Flaspohler P, Noonan R, Lubell K, Stillman L, et al. Bridging the gap between prevention research and practice: the Interactive Systems Framework for Dissemination and Implementation. Am J Community Psychol. 2008;41:171–181. 26. Powell BJ, Waltz TJ, Chinman MJ, Damschroder LJ, Smith JL, Matthieu MM, et al. A reﬁned compilation of implementation strategies: results from the Expert Recommendations for Implementing Change (ERIC) project. Implement Sci. 2015;10:21. 27. Flottorp SA, Oxman AD, Krause J, Musila NR, Wensing M, Godycki-Cwirko M, et al. A checklist for identifying determinants of practice: a systematic review and synthesis of frameworks and taxonomies of factors that prevent or enable improvements in health care professional practice. Implement Sci. 2013;8:35. 28. Brown CH, Curran G, Palinkas LA, Aarons GA, Wells KB, Jones L, et al. An overview of research and evaluation designs for dissemination and implementation. Annu Rev Public Health. 2017;38:1–22. 29. Leeman J, Birken SA, Powell BJ, Rohweder C, Shea CM. Beyond “implementation strategies”: classifying the full range of strategies used in implementation science and practice. Implement Sci. 2017;12:125. 30. Ostrom E. Institutional rational choice. Theories of the policy process. 1999:35–72. 31. Kitson AL, Rycroft-Malone J, Harvey G, McCormack B, Seers K, Titchen A. Evaluating the successful implementation of evidence into practice using the PARiHS framework: theoretical and practical challenges. Implement Sci. 2008;3:1. 32. Damschroder LJ, Aron DC, Keith RE, Kirsh SR, Alexander JA, Lowery JC. Fostering implementation of health services research ﬁndings into practice: a consolidated framework for advancing implementation science. Implement Sci. 2009;4:50. 33. Becker SJ, Squires DD, Strong DR, Barnett NP, Monti PM, Petry NM. Training opioid addiction treatment providers to adopt contingency management: a prospective pilot trial of a comprehensive implementation science approach. Subst Abus. 2016;37:134–140. 34. Finlay AK, Binswanger IA, Timko C, Smelson D, Stimmel MA, Yu M, et al. Facility-level changes in receipt of pharmacotherapy for opioid use disorder: implications for implementation science. J Subst Abuse Treat. 2018;95:43– 47. 35. Quanbeck A, Brown RT, Zgierska AE, Jacobson N, Robinson JM, Johnson RA, et al. A randomized matchedpairs study of feasibility, acceptability, and effectiveness of systems consultation: a novel implementation strategy for adopting clinical guidelines for Opioid prescribing in primary care. Implement Sci. 2018;13:21.

Scholarly Research In Progress • Vol. 3, November 2019

Trichostasis Spinulosa Masquerading as Hypertrichosis and Presenting at an Unusual Site in a 13-year-old Female Kendall Shifflett1†, Howard Pride2, Matthew Palmer2, and Eric Hossler2

Geisinger Commonwealth School of Medicine, Scranton, PA 18509 Geisinger Medical Center, Danville, PA 17821 †Doctor of Medicine Program Correspondence: kshifflett@som.geisinger.edu 1

Abstract Trichostasis spinulosa is a frequently unrecognized and underdiagnosed disorder of the pilosebaceous unit, characterized by many retained vellus hairs embedded in keratinaceous debris. There are two types of trichostasis spinulosa, a classic nonpruritic type seen in elderly sundamaged patients, and a nonclassic “pruritic” form seen in darker-skinned young women. We report a case of trichostasis spinulosa that has features of both types which presented in a fair-skinned teenage girl on her cheeks and nose.

Introduction Trichostasis spinulosa (TS) is a frequently unrecognized disorder of the pilosebaceous unit (1). It appears as slightly raised, small, dark, follicular spines that resemble open comedones (blackheads). When examined closely, these “spines” can be plucked by tweezing and examined under a microscope to reveal many fine vellus hairs embedded in keratinaceous debris (2). Two different types of TS are described, differing by their location, age at presentation, and skin color. TS can be found anywhere on the body, but the face, upper chest, and back are common, with one case found in the axilla (3). We present a case that mimicked facial hypertrichosis in a teenage girl.

Case presentation

Figure 1. Many black colored macules resembling open comedones vs. hairs on the nose and small fine black hairs on the cheeks

A 13-year-old female with a past medical history of asthma and gastric esophageal reflux disease was referred to the pediatric dermatology clinic for abnormal hair growth on the upper cheeks and nose. As an infant, her parents noted excessive facial hair that extended to her ears, which

slowly resolved. Then, 2 years prior to presentation, hair was again noticed on the upper cheeks and nose without any precipitating event. The hair was not located in androgensensitive areas. She had not tried any treatment at the time. She was diagnosed with hypertrichosis and referred to pediatric dermatology. Her review of systems was negative for other issues or skin concerns. Her medications included fluticasone nasal spray, albuterol inhaler, and ranitidine 75 mg. She had no known allergies, and her family history was negative for any skin conditions. On physical exam she had distinctly no hair in the sideburns or mustache area. There was very small, densely packed stubble of black hairs over the medial upper cheeks and nose (Figure 1A-C). Several of the dark facial hairs were plucked and examined under light microscopy, revealing a hyperkeratotic plug with 4 to 6 hairs coming from a single node (Figure 2A–C). A diagnosis of TS was made, and she was treated with a trial of 0.01% tretinoin gel applied over the affected areas at bedtime. At her 2-month follow-up visit, she had been tolerating the 0.01% tretinoin gel well and noted some improvement of the lesions. Overall, she is happy with the improvement. Although the hairs are now much harder to see, it was decided to increase the tretinoin gel to 0.025% (Figure 3A–C) and follow up in 2 months.

Discussion TS is a common disorder but often goes unrecognized and undiagnosed. It is a disorder of the pilosebaceous unit and is often confused for open comedones (blackheads). In TS, the pilosebaceous unit contains upwards of 60 fine, vellus hairs within one single follicle. TS was first described and named by Felix Franke, in 1901 who termed it “pinselhaar” (paintbrush hair) (1). The etiology of this condition is not yet certain, but two theories exist. The first is that there is an abnormal angulation between the isthmus and the infundibulum that may lead to entrapment of the vellus hairs within the follicle. The infundibulum starts at the surface and extends to the sebaceous gland, whereas the isthmus extends from the sebaceous gland to the pili erector muscle. The second theory is that the dilated vellus hair follicle has hyperkeratosis, leading to retention of additional telogen hairs. The result of these two theories is that the hair follicle could contain between 5 and 60 retained hairs (1, 2). Two variants of TS exist. The first is the “nonpruritic” or

105

Trichostasis Spinulosa Masquerading as Hypertrichosis

Conclusion This case brings awareness to a common but underdiagnosed dermatologic condition and describes an unusual presentation of TS on the upper cheeks and nose in a fair-skinned teenage female; further, it highlights the importance of trichograms in diagnosing this condition. By plucking a few hairs and examining them microscopically, a diagnosis of TS can readily be established.

References 1.

Naveen KN, Shetty SR. Trichostasis spinulosa: An overlooked entity. Indian Dermatol Online J. 2014;5(Suppl 2):S132–S133.

2. White SW, Rodman OG. Trichostasis spinulosa. J Natl Med Assoc. 1982;74:31–3. 3. Ramteke MN, Bhide AA. Trichostasis Spinulosa at an Unusual Site. Int J Trichology. 2016 Apr-Jun;8(2):78–80. 4. Deplewski D, Rosenfield RL. Role of hormones in pilosebaceous unit development. Endocr Rev. 2000 Aug;21(4):363–92. Figure 2. A bundle of more than 5 vellus hairs in a keratinous sheath A) 20X magnification B) 100X magnification C) 200X magnification.

Figure 3. Trichostasis spinulosa after a 2-month treatment of 0.01% tretinoin gel

classical type, typically seen in light-skinned, elderly citizens with long-term sun damage, presenting as blackhead-like lesions on the face and nose. The “pruritic” or nonclassical type is normally found in darker-skinned (Fitzpatrick Type III or greater), young women, characterized by multiple small papules on the trunk and upper extremities (1, 2). A case was reported that highlighted TS in the interscapular region in a 27-year-old male, because TS in this location is often mistaken for keratosis pilaris (5). Our case showcases a mixture between the classical, nonpruritic type and nonclassical pruritic type, because it occurred in a young teenage girl with Fitzpatrick type <III skin on the cheeks and nose. It is interesting because this case was masquerading as hypertrichosis. Hypertrichosis refers to generalized excessive growth of vellus body hair that thus appears in both nonsexual and sexual areas (4). Another term that is often used for hypertrichosis is “werewolf syndrome” because of excessive body hair growth (6). TS is idiopathic, but there are cases that suggest various triggers that include topical steroids, minoxidil (7), clobetasol cream (8), chronic renal failure (9), oils, dust, UV light, and other irritants (1). A study also suggested that certain pityrosporum and bacteria, such as Propionibacterium acnes, may be a possible etiologic factor of TS (10). Treatments for TS include emollients, hydroactive adhesive tape, keratolytics, local and oral retinoids (1), repeated peeling with capryloyl salicylic acid (11), and long-pulsed alexandrite lasers (12). Topical treatments are thought to be temporary, whereas laser hair removal might serve as a permanent treatment option (12).

106

5. Gutte RM. Itchy black hair bristles on back. Int J Trichology. 2012;4(4):285–286. 6. Balaji P, Balaji SM. Gingival fibromatosis with hypertrichosis syndrome: Case series of rare syndrome. Indian J Dent Res. 2017 Jul-Aug;28(4):457–460. 7.

Navarini AA, Ziegler M, Kolm I, Weibel L, Huber C, Trüeb RM. Minoxidil-induced trichostasis spinulosa of terminal hair. Arch Dermatol. 2010;146:1434–5.

8. Janjua SA, McKoy KC, Iftikhar N. Trichostasis spinulosa: Possible association with prolonged topical application of clobetasol propionate 0.05% cream. Int J Dermatol. 2007;46:982–5. 9. Sidwell RU, Francis N, Bunker CB. Diffuse trichostasis spinulosa in chronic renal failure. Clin Exp Dermatol. 2006;31:86–8. 10. Chung TA, Lee JB, Jang HS, Kwon KS, Oh CK. A clinical, microbiological, and histopathologic study of trichostasis spinulosa. J Dermatol. 1998 Nov;25(11):697–702. 11. Wollina U. Trichostasis spinulosa-successful treatment by repeated peeling with capryloyl salicylic acid. J Clin Exp Dermatol Res. 2012;3:2. 12. Toosi S, Ehsani AH, Noormohammadpoor P, Esmaili N, Mirshams-Shahshahani M, Moineddin F. Treatment of trichostasis spinulosa with a 755-nm long-pulsed alexandrite laser. J Eur Acad Dermatol Venereol. 2010 Apr;24(4):470–3.

Scholarly Research In Progress • Vol. 3, November 2019

Controlled Substance Distribution in West Virginia from 2006 to 2017 Julius A. Hatcher IV1*, Sneha Vaddadi1†, and Bradley D. Nafziger1†

Geisinger Commonwealth School of Medicine, Scranton, PA 18509 *Master of Biomedical Sciences Program †Doctor of Medicine Program Correspondence: jhatcher@som.geisinger.edu 1

Abstract Introduction: The United States is undergoing an opioid epidemic, which is exacerbated in rural populations by economic stressors that affect prescription administration and illicit use. This study provides a statewide examination of opioid distribution patterns from 2006 to 2017 in West Virginia. Other controlled substances are reported for comparative purposes. Methods: Data were obtained from the U.S. Drug Enforcement Administration’s Automation of Reports and Consolidated Orders System for 2006 through 2017. Quantitative analyses include weight of six opioids (buprenorphine, codeine, hydrocodone, morphine, methadone, and oxycodone,) and four non-opioids (methylphenidate, pentobarbital, lisdexamfetamine, and amphetamine), with opioids converted to oral morphine milligram equivalents (MME). Results: Opioid utilization for pain increased until peaking in 2012, after which followed a steep decline until 2017. Methadone’s decline was matched by an increase in the distribution of buprenorphine. Lisdexamfetamine had pronounced growth with no decline in other attention deficit/ hyperactivity disorder (ADHD) medications. Conclusions: Although this study suggests positive results from recent efforts to control the opioid epidemic, there is still progress to be made in opioid access for opioid use disorder (OUD) treatment and analyzing how raw opioid distribution corresponds to opioid misuse. Additionally, a rise in amphetamine distribution may warrant further studies in order to prevent future epidemics in this area.

Introduction Misuse of prescription opioids remains an ongoing problem in the U.S. and especially in areas with large rural populations, such as West Virginia (1). Fatal drug overdoses have increased threefold in the U.S. since 1991 (2), and opioid analgesics are now known to be responsible for more deaths than suicide and motor vehicle crashes (3). This increase is primarily due to prescription opioid medications (2). More than one-sixth of those who use opioids non-therapeutically receive their medication from one doctor only (3). This poses a problem, because most opioid users are able to get opioid prescriptions from more than one doctor with no medical crosstalk. Furthermore, risk of opioid misuse and related deaths may be greater in rural populations due to a higher opioid prescription rate per capita in rural areas, easier access to opioids, and economic stressors which contribute to drug use (1). According

to the 2010 National Survey on Drug Use and Health (NSDUH), an estimated 8.9% of the American population older than 12 years of age were either current or past users of illicit drugs (3). The Automation of Reports and Consolidated Orders System (ARCOS) is a federal, computerized public data set which monitors the distribution of controlled substances that are reported to the U.S. Drug Enforcement Administration (DEA) (4, 5). The ARCOS dataset has many advantages in that it provides information about controlled substances on both a national and a state level. Also, information on each drug is provided via drug weight in grams and classified as being dispensed by pharmacies, hospitals, practitioners, or narcotic treatment programs (4). Thus, an analysis of data from ARCOS would allow for a recent and comprehensive source to observe drug distribution changes from pharmacies in West Virginia to help improve public knowledge about drug use and potential misuse. Prior research determined that West Virginia ranked ninth in the United States in 2016 for opioids used for pain management (5). Knowing that rural populations are at an increased risk for opioid misuse, it stands to reason that licit drugs in general may also be a chief area of concern, and West Virginia could potentially have high per capita use of certain non-opioid drugs (6). Therefore, the objectives of this study were to provide an overview of prescription opioids and non-opioids in an area with a large rural population and to examine the prescribed retail drug weight distribution of six opioids and four non-opioids in West Virginia from 2006 to 2017.

Materials and Methods Data collection and measures In this assessment, six opioids (buprenorphine, codeine, hydrocodone, morphine, methadone, and oxycodone,) and four non-opioids (amphetamine, lisdexamfetamine, methylphenidate, and pentobarbital) were extracted by drug weight per year. Data were obtained from the U.S. DEA’s ARCOS for 2006 to 2017 biannually. These drugs were chosen due to their common utility and availability within ARCOS (5). IRB approval was obtained from the University of New England. Statistical analysis Four analyses were completed: the total raw weight for each drug from 2006 to 2017 biannually; percent change in total weight for each drug from 2006 to 2017 biannually;

107

Controlled Substance Distribution in West Virginia from 2006 to 2017

total MME for all six opioids biannually; and total controlled drug weight (in kg) for pain from 2006 to 2017 biannually. These analyses were completed to identify recent changes in commonly utilized pharmacotherapies and to investigate the overall trends of opioid retail distribution in a large rural population. To correct for the relative potency of each opioid, oral MME conversions were completed using the following multipliers: buprenorphine 30, oxycodone 1.5, codeine 0.15, morphine 1, methadone 3, and hydrocodone 1 (7). Data analysis was completed using Excel, version 16.23. All figures and graphs were prepared with GraphPad Prism, version 8.0.2.

Results Figure 1A demonstrates MME normalized opioid weights that were distributed by pharmacies in West Virginia from 2006 to 2017. Methadone comprised the majority of the MME sold in 2006 at 1,355 kg and a linear decrease is evident until 2017 where distribution reaches a low of 740 kg, just above buprenorphine at 621 kg. The distribution of buprenorphine, unlike methadone, markedly increased in retail MME sales from 23 kg in 2006 to being second largest in 2017. Oxycodone and hydrocodone exhibited parabolic patterns, peaking in 2014 and 2012 at 929.13 kg and 558.28 kg, respectively, yet they were no longer the highest weight after normalization to MME. Morphine and codeine were relatively low in comparison with morphine peaking at 159 kg and codeine at 18 kg.

Figure 1. West Virginia opioid drug distribution normalized to morphine milligram equivalents and non-opioids from 2006 to 2017

Figure 1B illustrates the non-opioid controlled drug weights distributed by pharmacies in West Virginia from 2006 to 2017. Methylphenidate dominated the state consistently and increased gradually over the course of the study. Pentobarbital was the only drug that trended downward, while amphetamine had a gradual increase and lisdexamfetamine had a substantial increase from 4 kg in 2007 to 74 kg in 2012, when it leveled off. Figure 2 shows the percent change associated with each pharmaceutical. Buprenorphine and lisdexamfetamine exhibited the greatest percent change from 2006 to 2017, followed by amphetamine and methylphenidate. Pentobarbital decreased the most, followed closely by methadone. Morphine and hydrocodone also showed a marked decrease over the course of the study. Oxycodone and codeine exhibited minimal change compared to the other drugs.

108

Figure 2. Percent change in total drug weight in West Virginia from 2006 to 2017

Controlled Substance Distribution in West Virginia from 2006 to 2017

gain traction and bring national attention to the opioid crisis. Additionally, this was during the end of the Iraq War in the Middle East, in which the U.S. Armed Forces had been involved for almost a decade. Opioid use for chronic pain management by veterans would have severely declined after the Department of Veteran Affairs released new guidelines to attempt to curtail opioid distribution. Additionally, the steep decline in total controlled opioid weights from 2014 to 2016 may be in part due to the rescheduling of hydrocodone from schedule III to schedule II (9). Making hydrocodone more difficult to obtain undoubtedly played a role in the observed decrease in opioid distribution (9).

Figure 3. Total controlled drug weights for pain from 2006 to 2017

Figure 3 illustrates the trend of total controlled drug weight distribution for pain from 2006 to 2017. Oxycodone, hydrocodone, methadone, codeine, and morphine drug weights were combined to constitute total drug weight. Combined drug weight distribution increased linearly, reaching a distinct peak of 2,779.74 kg in 2012. After 2012, there was a notably steeper decline which dropped below the starting value of 775.78 kg in 2006. The steepest decline in total controlled drug weight occurred between the years 2014 and 2016; however, that may change depending on future 2018 data.

Discussion This study examines West Virginia pharmacy controlled substance distribution between 2006 and 2017. Trends in the sale and use of the top 10 drugs, as measured by total weight per year, were analyzed to determine changes in their distribution and to connect these changes to health care policy developments that may have been implemented around the same time. The most striking finding was the steady increase in raw opioid distribution by pharmacies in West Virginia until 2012, at which point it sharply declined until 2017. However, when looking at non-opioids, there was no such pattern. This could be due to several different factors but may be primarily due to the heightened awareness of the rising opioid epidemic. Both the U.S. Department of Veteran Affairs and the U.S. Department of Defense released clinical practice guidelines in May of 2010, titled “Management of Opioid Therapy for Chronic Pain” (8, 9). It presumably took at least a year for the new guidelines to

Many other national studies identified similar trends when analyzing databases of opioid sales and distribution throughout the United States (10). A 22% decrease in opioid prescriptions occurred from 2013 and 2017 in the United States because of increased thoughtfulness on the part of physicians and legislative action to curb the opioid epidemic (9, 11). However, this data suggests that although the opioid epidemic may not be resolved, effective steps were taken to stop the upward trend in opioid prescriptions. The most prescribed opioids ended up at lower levels of distribution in 2017 than they were in 2012, showing signs of hope for continued decrease in prescriptions and effectiveness of guidelines and other policies. Opioid pharmaceuticals still have the highest distribution by weight by a wide margin. However, the data was not analyzed to normalize for the recommended single dose of each drug. A few specific findings are of particular interest. It was surprising that the net weight of buprenorphine climbed substantially from 2006 to 2017. Buprenorphine was the only controlled drug analyzed for opioid addiction treatment (12) (note that this study did not report on methadone from narcotic treatment programs). Although the trend of pharmaceutical sales of opioids to treat pain decreased greatly after 2012, there were no such alterations observed for buprenorphine. If more people were being transferred off opioid pain therapy and were put on an addiction management plan, this would subsequently decrease the amount of pain drugs prescribed and increase the treatment drugs prescribed. We see this reflected in the trends of buprenorphine distribution and offer this as a possible reason for the increase in buprenorphine. Therefore, although controversial, it may be a future step to discover effective methods for decreasing the number of opioids prescribed for addiction therapy in order to end the opioid crisis effectively. Another possibility for the increase in buprenorphine is the drug’s several advantages over methadone, an alternative opioid addiction therapy drug. Although the two therapeutic opioids both act as µ opioid agonists, they have several differences. First, buprenorphine is less potent and therefore more suitable for easing the patient off opioids with long-term utilization (12). It is also less subject to abuse, because it offers a

109

Controlled Substance Distribution in West Virginia from 2006 to 2017

soft cap on the levels of receptor activation that can occur (12). Finally, buprenorphine is able to be distributed by pharmacies, whereas methadone can only be distributed by an authorized addiction treatment center due to its potency (12). Being able to self-administer buprenorphine would make it appealing to those taking it and possibly contribute to the success it has seen in pharmaceutical sales. However, methadone is more effective at treating severe addiction because of the increased subjective effects it provides when compared to buprenorphine (12). This may be an additional concern with utilizing methadone and may warrant a change in prescription amounts to make sure that it is not used for maintaining an addicted status instead of remediating. An increase in lisdexamfetamine (+1,617%) and, to a lesser extent, amphetamine in West Virginia from 2006 to 2017 may be attributed to a rise in ADHD diagnoses and associated treatment. Based on data from the National Survey of Children’s Health, the national average of percent medicated ADHD in children age 4 to 17 has risen from 4.8% to 6.1% from 2007 to 2011 (14). Particularly in the southern states, medicated ADHD percentage rose from 5.8% to 7.3% (14). In 2007–2008, West Virginia was one of five states with the highest rates of medicated ADHD, with 7.8% compared to the national average of 4.8% (15). There was a particularly appreciable rise in lisdexamfetamine prescription compared to amphetamine, which can be attributed to its clinical use as a longer-acting agent with a smaller potential for abuse (16). ARCOS does not account for drugs or medications procured from other countries such as Canada (5). This may not be as critical a point, as it would be for Canadian border states, since this study focused on West Virginia. However, in the age of technology, more Americans are searching the internet to obtain prescription drugs (17). A small and presumably negligible amount of controlled substances are used for veterinary purposes, and ARCOS does not contain a mechanism to exclude these. While this research is novel in that it investigated West Virginia (a state very impacted by opioid overdoses), opioid and non-opioid trends as a whole, and the effect of major treatment opioids on pain opioid distribution in the area, we only included six opioids and four non-opioids. Additionally, this research only looked at pharmaceutical distribution and could be expanded to include addiction treatment centers and hospitals. Oral MME values should be further considered as these might be underestimates (5). There is controversy regarding the MME conversion factor for methadone. ARCOS reports do not differentiate between opioid forms (e.g., buprenorphine tablet versus patch versus film). Some weight comparisons may not be directly relatable due to differences in individual dosing of the medication.

Conclusion ARCOS tracks the distribution of controlled substances which can be utilized to report trends in pharmaceutical data, providing quantitative analysis of relevant societal issues like the opioid epidemic. Our results suggest that the national attention on opioid control may be producing beneficial changes towards ending the opioid crisis evidenced by the reduction of opioid distribution in West Virginia since

110

2012. However, there has not been a subsequent decrease in therapeutic opioid medications that combat addiction, suggesting further health care policy improvements are required to end the opioid crisis. Another possible area of concern is the growing number of stimulant pharmaceuticals being distributed in West Virginia, as demonstrated by the steady increase in ADHD medication distribution observed from 2006 to 2017.

Acknowledgments Special thanks to Brian Piper, PhD, and Elizabeth Kuchinski, MPH, for their guidance and that of Kenneth McCall, PharmD, at the University of New England, as well as the Department of Medical Education of the Geisinger Commonwealth School of Medicine for the utilization of their facilities and software throughout the research process.

Disclosures The authors have no conflicts of interest to disclose.

References 1.

Keyes K, Cerdá M, Brady J, Havens J, Galea S. Understanding the rural–urban differences in nonmedical prescription opioid use and abuse in the United States. Am J Public Health. 2014;104(2):52–59.

2. Modarai F, Mack K, Hicks P, Benoit S, Park S, Jones C, et al. Relationship of opioid prescription sales and overdoses, North Carolina. Drug Alcohol Depend. 2013;132(1–2):81–86. 3. Manchikanti L, Helm S, Fellows B, Janata JW, Pampati V, Grider JS, et al. Opioid epidemic in the United States. Pain Physician. 2012;15. 4. Atluri S, Sudarshan G, Manchikanti L. Assessment of the trends in medical use and misuse of opioid analgesics from 2004 to 2011. Pain Physician. 2014; 17. 5. Piper B, Shah D, Simoyan O, McCall K, Nichols S. Trends in medical use of opioids in the U.S., 2006–2016. Am J Prev Med. 2018;54(5):652–660. 6. Fighting Substance Abuse [Internet]. Ago.wv.gov. 2014 [cited 24 April 2019]. Available from: https://ago.wv.gov/ consumerprotection/Fighting%20Substance%20Abuse/ Pages/default.aspx 7.

[Internet]. Cms.gov. 2019 [cited 19 April 2019]. Available from: https://www.cms.gov/Medicare/Prescription-DrugCoverage/PrescriptionDrugCovContra/Downloads/OpioidMorphine-EQ-Conversion-Factors-Aug-2017.pdf

8. U.S Department of Veterans Affairs. VA/DoD CLINICAL PRACTICE GUIDELINE FOR MANAGEMENT OF OPIOID THERAPY FOR CHRONIC PAIN. Department of Veterans Affairs Department of Defense; 2017. 9. Jones CM, Lurie PG, Throckmorton DC. Effect of US Drug Enforcement Administration’s rescheduling of hydrocodone combination analgesic products on opioid analgesic prescribing. JAMA internal medicine. 2016 Mar 1;176(3):399–402.

Controlled Substance Distribution in West Virginia from 2006 to 2017

10. U.S. Opioid Prescribing Rate Maps [Internet]. Cdc.gov. 2018 [cited 11 April 2019]. Available from: https://www.cdc.gov/ drugoverdose/maps/rxrate-maps.html 11. Harris P. AMA sees progress in declining opioid prescriptions [Internet]. American Medical Association. 2019 [cited 11 April 2019]. Available from: https://www. ama-assn.org/press-center/ama-statements/ama-seesprogress-declining-opioid-prescriptions 12. Whelan PJ, Remski K. Buprenorphine vs methadone treatment: A review of evidence in both developed and developing worlds. J Neurosci Rural Pract. 2012;3(1):45. 13. ADHD Throughout the Years [Internet]. Centers for Disease Control and Prevention. 2018 [cited 11 April 2019]. Available from: https://www.cdc.gov/ncbddd/adhd/timeline.html 14. Visser S, Danielson M, Bitsko R, Holbrook J, Kogan M, Ghandour R, et al. Trends in the parent-report of health care provider-diagnosed and medicated attention-deficit/ hyperactivity disorder: United States, 2003–2011. J Am Acad Child Adolesc Psychiatry. 2014;53(1):34–46. 15. State-based Prevalence Data of Parent Reported ADHD Medicated Treatment [Internet]. Centers for Disease Control and Prevention. 2018 [cited 15 May 2019]. Available from: https://www.cdc.gov/ncbddd/adhd/medicated.html 16. Goodman D. Lisdexamfetamine dimesylate (vyvanse), a prodrug stimulant for attention-deficit/hyperactivity disorder. P T. 2010; 35:5, 273–276, 282–287. 17. Martin J. United States prescription drug crisis. J Legal Med. 2006; 27:4,477–492.

111

Scholarly Research In Progress • Vol. 3, November 2019

Tea as Treatment: Applications of the Kava Plant (Piper methysticum) in Modern Medicine Everett M. Blough1*

Geisinger Commonwealth School of Medicine, Scranton, PA 18509 *Master of Biomedical Sciences Program Correspondence: eblough@som.geisinger.edu 1

Abstract The kava plant (Piper methysticum) has been an integral part of Pacific Island cultures for thousands of years. Ceremonially, kava’s roots are prepared as a sort of tea used to elicit sedative, relaxing, and sometimes euphoric effects. Recently, kava has become a topic of medical research for its anxiolytic, anti-cancer, and immunomodulatory properties. Studies involving both mice and humans indicate that kava extract is responsible for numerous effects in the body. These effects may vary depending on the relative concentrations of kava-based molecules like kavalactones and flavokavains. Research on kava’s relaxing properties show that it increases GABAA receptor activity, though through a somewhat unknown mechanism. Through various pathways such as Skp2 degradation and androgen receptor downregulation, kava-derived chemicals play a role in inhibiting cancer growth. Lastly, these compounds can alter the body’s immune response, typically through mediation of TNF-α pathways. Despite kava’s numerous potential benefits, there are risks which must also be considered and studied further. For example, following kava’s initial export from the Pacific Islands, there were many cases of hepatotoxicity that were called into question. Additionally, kava-induced skin inflammation is a rare condition still under investigation. Regardless, kava’s permeation into other cultures is apparent. It is sold as a pill-based supplement and a prepackaged tea and enjoyed in “kava lounges” in the United States. As kava research and consumer use continues, more light must be shed on this curious plant’s medicinal properties.

Introduction Since the dawn of humanity, medicine has quite literally taken its roots in local plants. Traditional remedies often involved the local flora of tribes and villages, each prepared in a particular way to promote healing. For cultures native to the Pacific Islands, this backstory describes their use of the kava plant (Piper methysticum), alternatively named “kava kava.” Kava, a member of the pepper family, has an extensive history of use in Pacific civilizations. Most notably, its roots are ground and prepared into a tea-like drink consumed ceremonially by the chiefs and priests of some tribes (1). Primarily, the drink is valued for its non-addictive, sedative qualities. Also, while kava preparations in the modern world are not typically concentrated enough to induce euphoria (as in these ceremonies), kava’s properties as a relaxant have brought it into the spotlight. As word of kava spread, its export to other countries was inevitable and swift. However, cases of kava-induced hepatotoxicity brought the safety of the plant-based drink into question. Kava is essentially a collection of various 112

compounds, primarily composed of kavalactones (KL) and flavokavains (a variety of aromatic ketones). Research into these molecules indicates that the kavalactones are responsible for the “relaxing” effects of kava tea and the flavokavain concentration is what ultimately leads to liver damage (2). Not all kava plants are created equal. There are two strains of kava—dubbed “noble” and “non-noble”—which differ in their chemotypes. In short, noble kava is described as safe to use as a social beverage, whereas non-noble (sometimes “Tudei”) kava is a highly concentrated strain similar to that used in aforementioned rituals of Pacific cultures (3). Lack of knowledge regarding the differences in strains likely exposed individuals to the unsafe kava cultivars, leading to both liver damage (in the form of acute hepatitis) and widespread controversy. In response to these health risks, some countries began to shut down kava imports. In some European countries, the use of kava products was banned altogether (4). Although most nations have repealed these bans in light of the “noble versus non-noble” information, kava still remains a bit of a mystery in the field of medicine. Because of its anxiolytic properties, kava was first considered to be a potential non-addictive alternative to benzodiazepines. However, recent explorations into particular kavalactones and flavokavains show potential for kava compounds to treat cancer and autoimmune diseases. Plant-based medicine is an important field of study for discovering new and alternative therapeutics. In this review, the rodent and human investigations of kava treatments will be discussed, as well as the potential future for kava use. (The methods for conducting this literature review are discussed in detail in the “Appendix: Research Methods” section at the end of the paper.)

Materials and Methods For this review paper, the majority of sources were retrieved from the PubMed Central (PMC) digital repository, where fulllength articles are freely available. Other databases used for this review include SpringerLink, ClinicalKey via the Geisinger Commonwealth School of Medicine online library proxy service, TheFreeLibrary and the PubMed online database. Searches were conducted with engines such as the Mansfield University North Hall Library site, Geisinger Commonwealth School of Medicine online library and Google Scholar. Search terms included keywords such as “kava,” “kava and cancer,” “kava culture,” “kava hepatotoxicity,” “kavalactones,” “flavokavains,” “kava case study,” “kava tea,” and “kava extract.” Primary research articles were typically excluded if only the abstract was available and only recent (within the past two decades) scientific research articles and clinical studies were used.

Tea as Treatment: Applications of the Kava Plant (Piper methysticum) in Modern Medicine

Discussion Rodent studies Most research into kava started with identifying the various kavalactone and flavokavain compounds within the kava extract. Then, exploration into the mechanisms and physiological effects of these molecules were further examined. As a relaxing agent, the anti-anxiety effects of kava were explored in mice in vivo. Kava extract was administered to mice, which were then subjected to environments they typically would avoid. “Anxiolysis” was defined as the mice spending time in a normally avoided area or as reduced hesitance to enter said area. While statistically significant, dose-dependent anxiolytic effects were seen when mice were administered flumazenil, this competitive benzodiazepine receptor antagonist had no impact on kava’s behavioral effects (5). This indicates that kavalactones do not mediate anxiety through the GABAA receptor complex. Therefore, their mechanism may be beneficial for those with anxiety who cannot be prescribed benzodiazepines. Further rodent studies explored the potential anti-cancer effects of kava compounds. Firstly, prostate cancer was explored as a remarkable case of kava’s potential anti-cancer properties. It had been noted that kava-drinking men in Fiji have a particularly low incidence of prostate cancer. Upon moving to Australia, however, prostate cancer incidence in Fijian men increased fivefold (6). As a dietary or cultural factor must have protected these men from prostate tumorigenesis, kava use was explored as the “common link.” Sure enough, a study (6) involving mouse cell lines (in vitro) indicates that tumor growth was suppressed when exposed to kava extract (most notably, flavokavain B). The research suggests that flavokavain B down-regulates expression of androgen receptor (AR) and its splice variants via Sp1 inhibition. Sp1 is a transcriptional factor leading to AR expression. For AR-dependent prostate cancer cells, this would inevitably lead to cell death. Perhaps more notably, flavokavain B also suppressed growth of androgenindependent cell lines in vitro (6). It is apparent that we are only scratching the surface for the novel applications of these anti-tumor compounds. An additional anti-cancer investigation focused on the effects of kava extract on lung cancer in mice (7). This in vivo experiment examined whether kava ingestion had an impact on reducing lung tumorigenesis and/or tumor multiplicity. The strain of mice used (dubbed “A/J”) are bred for high susceptibility to carcinogen-induced tumors. Mice were exposed to nicotinederived nitrosamine ketone (NNK), a carcinogen present in tobacco smoke (7). The mice fed kava supplements, which are rich in kavalactones, prior to NNK exposure expressed a 67– 87% reduction in adenoma incidence—and a staggering 99% reduction in tumor multiplicity. However, mice that were fed the kava supplement following NNK exposure did not experience these inhibitory effects. Therefore, some evidence exists in support of kavalactones for prevention of lung tumorigenesis. It is worth nothing some kava-based research in the realm of immune system modulation. Chalcones are aromatic ketones recently discovered to play some modulatory role in the immune system, be it stimulation or inhibition (8). Flavokavains, being chalcones, are speculated to have some degree of anti-inflammatory, antiviral, or antifungal effects. One report

examined the effect of Kava-241, a kava-derived compound, on mice with infective arthritis (9). This in vivo research uncovered Kava-241’s ability to significantly reduce tumor necrosis factor TNF-α secretion via reducing toll-like receptors 2 and 4 (TLR-4/ TLR-2) expression. By reducing toll-like receptor expression and shutting down these signaling cascades, inflammatory cytokines like TNF-α are unable to cause inflammation in the synovial tissue of individuals with rheumatoid arthritis. Interestingly, flavokavain A modulates this same TLR-4/TLR2 pathway (9). A second immunomodulation study, however, showed that kava does not always decrease the immune response. For murine mast cells exposed in vitro to an aqueous kava solution (a preparation based traditional methods) calcium influx and subsequent degranulation was observed (1). Mast cell degranulation is a cause of inflammation and is perhaps responsible for the kava-induced skin inflammation rarely occurring in humans. These two seemingly contrasting studies (1, 9) highlight a key facet of kava use; its immune system modulation is complex and the specific effects may be dependent on the particular kavalactones/flavokavains present in the extract. Human studies and potential therapeutic targets Because of kava’s history with hepatotoxic controversy, human trials must be carried out with caution. One study instead utilized human recombinant GABAAR expressed in frog oocytes. Results indicate that kavalactones bind GABAA receptors with no subtype selectivity, leading to the same inhibitory CNS effects that GABA would (10). This report was valuable in supporting murine findings (5) that kava-derived compounds do not bind the high-affinity benzodiazepine site on GABAAR. However, human trials do provide some greater insight into the mechanism of kava-based compounds. Studies on anxiolysis make up the bulk of kava research in humans. One investigation found that kava use is not linked to a reduction of problem-solving and cognitive function in adults (as measured by the RCat, a computerized test) (11). The importance of this finding is that while being an anxiolytic, kava does not appear to lend to the drowsiness or mental “fog” associated with some anti-anxiety medications. This study is also important as an in vivo measure of kava’s effects on humans, unlike the GABAAR experiment done in frog oocytes. In the same vein as the rodent studies, kava’s anti-cancer effects have been a subject of interest. Flavokavain B—the same compound from the mouse study (6)—in human prostate cancer cell lines (in vitro) bound the APP-BP1 subunit of NAE1 and caused Skp2 degradation (12). By binding APP-BP1, flavokavain B prevents the neddylation event associated with cell cycle progression (and, consequently, cell division); furthermore, Skp2 degradation releases Skp2’s inhibition on p27 which in turn leads to cell cycle arrest. Growth inhibition through these mechanisms is a treatment option for castrateresistant prostate cancer, further lending to this finding’s importance (12). However, flavokavain B only represents one of the several flavokavains with potential immunotherapeutic properties. Another study evaluating flavokavain A, found similar Skp2 targeting and degradation—this time, using in vitro human cells with osteosarcoma (13) By preventing osteosarcoma progression and metastasis, kava extract compounds like the flavokavains may play pivotal roles in anticancer treatments. 113

Tea as Treatment: Applications of the Kava Plant (Piper methysticum) in Modern Medicine

Research on kavalactones have also found potential antineoplastic effects. One of the primary kavalactones, yangonin, reduced viability of human bladder cancer cell lines in vitro (14). Yangonin seems to act in a dose-dependent manner of regulating mTOR, a pathway allowing cancer cells to survive in stressful conditions. Additionally, yangonin exposure increased the sensitivity of bladder carcinomas to flavokavain A, ultimately inhibiting cell growth (14). This finding may be of further consequence, since kava consumption typically involves (noble) strains with a higher kavalactone-to-flavokavain ratio. Therefore, for the average kava tea drinker, there may be some relative health benefit in regard to cancer prevention. Lastly, the role of kava-based compounds in immune system modulation has also been explored in human cells. For cells exposed in vitro to a TNF-α transcription factor (thus, with high levels of the TNF-α cytokine), the major kavalactone kavain significantly inhibited further TNF-α release. (15) While the mechanism of kavain is not inherently understood, some of its absorption, distribution, metabolism, and excretion (ADME) properties described in the article highlight its importance as a potential therapeutic. Kavain (and kavalactones in general) boast an impressive oral bioavailability and plasma stability—99% in aqueous buffer as well as mouse plasma (15). The half-life of kavalactones appears to be roughly 40 minutes for oral absorption. Therefore, some molecular changes might need to be made if kavalactones’ effects are mediated in a long-term manner. Fortunately, this same study (15) also discovered by editing kavalactone structure that an open-ringed analog of kavain also mediated TNF-α levels, but through a unique mechanism (via suppressing levels of the TNF-α transcription factor). This finding suggests that modifications may be made to some of the many kavalactones, allowing them to function through different mechanisms of action. Risks and considerations While many of the findings in the aforementioned rodent and human studies seem promising, there are still concerns surrounding kava use which must be addressed. Although plant-based medicine may provide non-synthetic alternatives to drug and treatment options, this does not necessarily reduce their risks. The primary example of this in kava is hepatotoxicity. While this concern was somewhat reduced upon the revelation of noble and non-noble kava strains, it cannot be entirely dismissed. The hepatic damage associated with flavokavains may still be a drawback to its anti-cancer and immunomodulatory benefits (2). As with any drug, it will be important to consider both the benefits and the associated adverse effects if kava-based treatments are approved for such use. Additionally, while aqueous kava preparations serve as short-term anxiolytics, a meta-analysis of seven placebo-controlled clinical trials also cites liver damage as a result of long-term kava use (16). These findings indicate that hepatotoxic risks are a function of kava use duration (i.e., every day for 8 or more weeks) rather than a function of kava concentration. While strains of kava with more dangerous chemotypes may present a greater risk of developing acute hepatitis, the findings of the aforementioned clinical trial indicate that liver damage is more strongly related with the long-term use of kava supplements.

114

As a counterargument, however, some scholars (and kava enthusiasts) still argue that hepatotoxicity is not a direct risk of kava consumption. While hepatotoxins have been found in various strains of kava plants, there is evidence that these toxins are caused by molds—not kavalactones or flavokavains (17). The growth conditions of kava in the South Pacific are characterized by high temperature and humidity. It is, therefore, a valuable consideration that molds producing aflatoxins and hepatotoxins contaminated the plants prior to their harvest and exportation. Before quality control measures for kava growth and imports were established, these fungi may have caused a number of the early cases of alleged kava-induced hepatitis. Furthermore, it is worth considering a point raised in the aforementioned meta-analysis (16); six of the seven kavaversus-placebo clinical trials did not report any heightened lab values indicating hepatitis. Therefore, the rarity of hepatotoxicity is something to consider, especially in cases of short-term use. Lastly, there is another adverse effect associated with kava that must be considered. There are some case studies (18, 19) from the past few years detailing an unusual dermatologic condition following kava consumption. This condition is described in both of the aforementioned case studies as a “sebotropic eruption,” meaning that the sebaceous glands are stimulated in response to kava tea consumption. Both cases identified elevated neutrophil counts and edematous papules/plaques on the face, chest, back, and upper/lower extremities. Overall, both of these reports indicate a sort of systemic inflammatory response. As stated previously, kava and its associated molecules have very complex interactions with the human immune system. It is possible for up- or downregulation of various modulatory pathways. In this type of rare condition, it is proposed that kavapyrones (a variety of kavalactones) concentrate in the sebaceous glands, resulting in a lymphocytic response targeting these glands (19). Notably, while one patient benefited solely from the use of oral prednisone and topical hydrocortisone (18), the other patient responded ineffectually to prednisone and only improved following cyclosporine administration (19). Due to the recency of kavalactone studies and the rarity of these cases, it is difficult to determine the precise factors which cause kava-induced dermopathy. Until a collection of case reports and laboratory values are gathered for this condition, it may be difficult to determine the underlying mechanism. Areas for future research While the progression of kava-based medical research has made some remarkable findings—particularly in its transition from murine to human models—additional research will be critical in further characterizing kava’s therapeutic potential. Perhaps the most pressing topic of investigation is the safety of kava consumption. Individuals who are partial toward homeopathic medicine and plant-based supplements will certainly look to kava as it rises in popularity. Therefore, protecting these populations which may be consuming kava on a daily basis by regulating kava production should be prioritized. Research into this topic might take similar steps as the TNF-α mediation study (15), which first analyzed the pharmacokinetic (ADME) properties of the kavalactone used. By analyzing such properties of individual kavalactones,

Tea as Treatment: Applications of the Kava Plant (Piper methysticum) in Modern Medicine

perhaps they can be further categorized or considered for their interactions in the body. Pharmacokinetic properties still require further research and identification, both in mice and humans, to understand how they impact the body at various concentrations; currently, not many pharmacokinetic factors have been investigated for specific kavalactones and flavokavains. Moreover, these factors may provide some insight into which molecules are metabolized most heavily by the liver, potentially indicating which ones are of highest hepatotoxic risk. Once risk factors for certain kavalactones and flavokavains have been clearly established, it may be reasonable to perform more in vivo studies of kava in humans—an area of research which is currently outnumbered by its in vitro counterparts. A second topic would be to find a more efficient method to identify which kavalactones/flavokavains exist in a particular kava strain. Only recently, in 2018, was a validated method developed which can determine the chemical composition of kava products (20). This practice, dubbed high-performance liquid chromatography with ultraviolet detection (HPLC-UV), was described as being able to identify six major kavalactones and three flavokavains with a minimum of 92% accuracy. Additionally, it may be applied to a variety of kava-based products: dry-filled capsules, liquid phytocaps, tinctures and raw kava material itself. Especially once individual kavalactone and flavokavain properties are identified, it would be highly beneficial to know which of these molecules are present in a supplement or tea. In the case of the potentially dangerous kavapyrones, it would be important for individuals who have experienced kava-related AEs and other susceptible persons to know which kava strains to avoid. While this HPLC-UV method has demonstrated efficacy, other practices such as RP-HPLC, rapid HPLC, and UPLC-MS/MS have seen use in multiple kava studies (2). Each of these methods have their own particular strengths and weaknesses; as kava studies continue developing and refining extraction methods, the advantages of each method will be better understood. There are also some “offshoots” of kava-based research which—while not clinical in nature—may provide some use. For example, there are potent plant growth inhibitors in the roots of the kava plant (21). Different chemicals, some kavalactones and some flavokavains, inhibit the growth in different regions of plants (i.e., shoot versus root). Therefore, some potential exists for kava to become a natural herbicide. Also, there is a need for surveying the knowledge of people who either use kava regularly or avoid it for prior hepatotoxic claims. In a culture where the use of supplements is a sometimesdivisive discussion, there may be some value in examining the conceptions and misconceptions of kava products. Education might be an important next step for addressing kava’s controversial history; while there may be some evidence of liver risk associated with long-term use, the “non-noble” kava strains which gave the plant its bad reputation are not being sold to consumers at large. Education may also be important for reintroducing kava products into the consumer market as new health benefits are discovered.

Conclusion Kava is a unique plant with a rich cultural and economic history. From its early use in Pacific Island rituals, subsequent exportation and legal restrictions and current subject of medicinal research, kava’s story is far from over. Currently, kava is widely available for consumer use in the United States; supermarket shelves are stocked with prepackaged kava tea bags, and “kava lounges” are opening from coast to coast as a means to socially enjoy the beverage. While the disparity between noble and non-noble cultivars has been established, it is important to further uncover kava’s long-term health effects— both therapeutic and adverse—as kava use continues to rise. Culture and consumer aspects aside, there are reasonable prospects for kavalactones and flavokavains to be used in the realm of medicine. Anxiolytic properties via modulation of GABAA may provide a non-addictive alternative to benzodiazepines. Furthermore, the multitude of pathways that flavokavains may influence regarding cell signaling, growth, and proliferation make these chemicals possible anti-cancer therapeutics. Lastly, while immunomodulation through kavabased compounds is complex, disrupting TNF-α signaling may be a possible therapeutic for those with autoimmune disorders. Still, these studies are mostly done in vitro, and the complexity of signaling pathways in vivo may lead to unknown adverse effects. Therefore, it is vital that research into kava proceeds in the direction of adverse effects and pharmacokinetics. Controversy surrounding hepatotoxicity and rare cases of kava-induced dermopathy are subjects requiring further investigation. Additionally, until properties such as kavalactone/flavokavain distribution and maximum serum concentration (Cmax) are identified, it is difficult to know what levels of these compounds are efficacious versus what levels are dangerous. Ultimately, it is vital for the field of medicine to continue exploring plantbased remedies in search of novel therapeutics. The unique properties of kava extract and its specific components may provide the roots for improved disease treatment.

Acknowledgments I would like to specifically thank Brian J. Piper, PhD, assistant professor of neuroscience, for his diligent advisement and support on this paper. I would also like to thank Darina Lazarova, PhD, associate professor of molecular biology, for the opportunity to write this review.

References 1.

Shimoda LMN, Park C, Stokes AJ, et al. Pacific island ‘Awa (kava) extracts, but not isolated kavalactones, promote pro-inflammatory responses in model mast cells. Phytother Res. 2012;26(12):1934–1941.

2. Tang Y, Fields C. A UHPLC-UV method development and validation for determining kavalactones and flavokavains in Piper methysticum (kava). Molecules. 2019;24(7):1245. 3. Teschke R, Sarris J, Glass X, et al. Kava, the anxiolytic herb: back to basics to prevent liver injury? Br J Clin Pharmacol. 2011;71(3):445–448.

115

Tea as Treatment: Applications of the Kava Plant (Piper methysticum) in Modern Medicine

4. Kuchta K, Schmidt M, Nahrstedt A. German kava ban lifted by court: the alleged hepatotoxicity of kava (Piper methysticum) as a case of ill-defined herbal drug identity, lacking quality control, and misguided regulatory politics. Planta Med. 2015;81(18):1647–1653. 5. Garrett KM, Basmadjian G, Khan IA, et al. Effects of kava (Piper methysticum) induce acute anxiolytic-like behavioral changes in mice. Psychopharmacology. 2003;170(1):33–41. 6. Li X, Liu Z, Xu X, et al. Kava components down-regulate expression of AR and AR splice variants and reduce growth in patient-derived prostate cancer xenografts in mice. PLoS One. 2012;7(2). 7.

Leitzman P, Narayanapillai SC, Balbo S, et al. Kava blocks 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone-induced lung tumorigenesis in association with reducing O6methylguanine DNA adduct in A/J mice. Cancer Prev Res. 2014;7(1):86–96.

8. Lee JS, Bukhari SNA, Fauzi NM. Effects of chalcone derivatives on players of the immune system. Drug Des Devel Ther. 2015;9:4761–4778. 9. Huck O, You J, Han X, et al. Reduction of articular and systematic inflammation by Kava-241 in a Porphyromonas gingivalis-induced arthritis murine model. Infect Immun. 2018;86(9):e00356–18. 10. Chua HC, Christensen ETH, Hoestgaard-Jensen K, et al. Kavain, the major constituent of the anxiolytic kava extract, potentiates GABAA receptors: functional characteristics and molecular mechanism. PLoS One. 2016;11(6):e0157700. 11. Gendle MH, Stroman AK, Mullin DP. Effect of an acute dose of crude kava root extract on problem solving in healthy young adults. AJMH. 2011;23(4):160–163. 12. Li X, Pharm V, Tippin M, et al. Flavokawain B targets protein neddylation for enhancing the anti-prostate cancer effect of Bortezomib via Skp2 degradation. Cell Commun Signal. 2019;17:25. 13. Zhang Y, Zvi YS, Batko B, et al. Down-regulation of Skp2 expression inhibits invasion and lung metastasis in osteosarcoma. Sci Rep. 2018;8:14294. 14. Liu Z, Ha US, Yu K, et al. Kavalactone yangonin induces autophagy and sensitizes bladder cancer cells to flavokawain A and docetaxel via inhibition of the mTOR pathway. J Biomed Res. 2017;31(5):408–418. 15. Pollastri MP, Whitty A, Merrill JC, et al. Identification and characterization of kava-derived compounds mediating TNF-α suppression. Chem Biol Drug Des. 2009;74(2):121– 128. 16. Smith K and Leiras C. The effectiveness and safety of kava kava for treating anxiety symptoms: a systematic review and analysis of randomized clinical trials. Complement Ther Clin Pract. 2018;33:107–117. 17. Teschke R, Qui SX, Lebot V. Herbal hepatotoxicity by kava: update on pipermethystine, flavokavain B, and mould hepatotoxins as primarily assumed culprits. Dig Liver Dis. 2011;43(9):676–681.

116

18. Rzepecki A, Wald J, Amin B, et al. Kava-induced acute cutaneous toxicity: an increasingly recognized characteristic clinicohistologic pattern. JAAD Case Rep. 2018;4(10):1037–1038. 19. Brown-Joel ZO, Colleran ES, and Stone MS. Inflammatory sebotropic reaction associated with kava kava ingestion. JAAD Case Rep. 2018;4(5):437–439. 20. Liu Y, Lund JA, Murch SJ, et al. Single-lab validation for determination of kavalactone and flavokavains in Piper methysticum (kava). Planta Med. 2018;84(16):1213–1218. 21. Van TM, Xuan TD, Minh TN, et al. Isolation and purification of potent growth inhibitors from Piper methysticum root. Molecules. 2018;23(8): 1907.

Scholarly Research In Progress • Vol. 3, November 2019

Assessing the Effectiveness of the Murine Model in Advancing X-Linked Agammaglobulinemia Research Shijo Benjamin1† and Mary Taglieri1†*

Geisinger Commonwealth School of Medicine, Scranton, PA 18509 †Doctor of Medicine Program *Master of Biomedical Sciences Program Correspondence: sbenjamin@som.geisinger.edu 1

Abstract The life-threatening yet rarely inherited condition X-linked agammaglobulinemia (XLA) is an immunodeficiency disease that depletes B cells within the human body. B cells are antibody-producing lymphocytes that are antigen-specific and critical for the adaptive immune response. Thus, a patient’s immune system is compromised and exposed to environmental pathogens without B cells. In fact, Streptococcus pneumoniae and Haemophilus influenzae arise as the most common organisms causing recurrent bacterial infections in XLA patients (1). Some of these infections include sinusitis, pneumonia, otitis, and bronchitis, which are all due to an absent humoral response causing compromise of the neutralization, opsonization, complement activation and receptor-mediated phagocytosis of microbial cells (2). Physically and characteristically, decreased CD19+ B cells will cause the tonsils and lymph nodes to become reduced in size (1). A decreased white blood cell count, the reduction of serum immunoglobulins and a mutation in the Bruton agammaglobulinemia tyrosine kinase (Btk) gene confirms the diagnosis of this genetic abnormality. In order to manage the disease complications with treatment, intravenous (IV) infusion of gamma globulin and aggressive antibiotics are suggested (1). This partially restores the human body’s ability to fight off environmental pathogens. However, a model organism is required as a genetic platform for investigation to create more effective treatments. A mouse model is an ideal genetic platform for investigation due to its low cost, high reproductive rate, and, most importantly, its immune system’s similarity to the human immune system with B cell lineage and pathophysiology in mutations (3). Therefore, understanding the molecular mechanisms of the XLA disease through the murine model organism may be the key to future long-term treatment.

Introduction Genetics of XLA XLA is a recessive condition directly affecting males that has a live birth frequency of 1 in 200 000 and genetically displays full penetrance (2). As a single gene disorder, mutations in Btk, a 37.5 kb gene containing 19 exons located at Xq22.1, alone cause XLA. The gene product is the 659 amino acid, cytoplasmic non-receptor Bruton’s tyrosine kinase or Btk (1). This protein and its domains are displayed in Figure 1. Of importance, there are three serine and two tyrosine residues that serve as regulatory phosphorylation sites for the Btk metalloprotease. Zn2+ specifically binds to the Tec homology (TH) domain of the protein to provide stability and optimization (4).

Figure 1. Btk domains with regulatory phosphorylation sites and conserved domains. PH: pleckstrin homology, TH: Tec homology, SH3/SH2: Src homology, TK: tyrosine kinase. Image modified without permission (4).

Figure 2. Btk signaling pathways within the cytoplasm and nucleus facilitate negative regulation and feed-forward loop autoregulation (red). Image modified without permission (4).

Btk belongs to the Tec family kinases (Tfk) and is essential in the signaling, development, differentiation, proliferation and survival of B cells (4). Lack of the protein specifically results in the inhibition of B cell development from the pro- to pre-stage, eliminating circulating B cells in a loss-of-function mutation (4). The Btk pathway begins with phosphorylation by the Src family kinase Lyn after B cell receptor activation from an antigen, followed by Btk protein phosphorylation of Cγ2 phospholipase, triggering an array of downstream signaling proteins. However, the mechanisms are not fully understood (4). Figure 2 depicts the role of Btk in its own autoregulation and transcription of NFAT (nuclear factor of activated T cells) and NFκΒ (nuclear factor kappa-light-chain-enhancer of activated

117

Assessing the Effectiveness of the Murine Model in Advancing X-Linked Agammaglobulinemia Research

B cells). NFAT is a transcription factor involved in T cell activation and thymocyte development, facilitating a stronger immune response (5). The negative regulators of Btk include protein kinase C (PKC), which interferes with the targeting of Btk to the plasma membrane and activates calcineurin (CaN) to generate NFAT; Sab (synonymous SH3BP5), which inhibits autophosphorylation; and Pin-1 (protein interacting with NIMA1), which destabilizes and decreases the kinase activity. Btk is additionally involved in the NFÎşB signaling pathway by regulating transcription of Btk and NFÎşB (4). Btk has a feedforward loop, to potentially maintain normal B cell maturation. Furthermore, the Btk protein itself is critical in the B cell mediated humoral immune system response through the B cellsâ&#x20AC;&#x2122; role in secreting antibodies. It is necessary to understand the genetics and role of the Btk mutation at the organismal level by assessing the development of immature B cells. As shown in Figure 3, a normal bone marrow stromal cell grows from pro-B to pre-B to an immature B cell in bone marrow, capable of forming the mature B cell in the spleen that is vital to mounting an immune response (6). IgM is the first immunoglobulin-antibody to develop on the B cell in response to initial exposure to an antigen (6). However, the Btk mutation on the X chromosome (xid) results in abnormal expression of critical B cell surface proteins. In addition, mutations in regulatory factors and other Btk pathway factors in the figure have also been shown to arrest development and fail to produce pre-B cells and mature B-cells (7). The proposed model necessitates an in vivo method of analysis in translating the mechanisms of therapy to human XLA patients.

Figure 3. B cell development and phenotypic outcomes dependent on XLA (6).

118

The murine model to study XLA As indicated, the murine model provides a valuable vertebrate mammalian model in which to study human diseases due to their small size and short lifespan. Although similar in immune system structure, the lymphocyte and neutrophil balance differs (8). In particular, murine X-linked immunodeficiency (xid) develops from a missense mutation at a conserved arginine residue within the Btk PH domain as a comparable model to human XLA (7). While XLA patients have fewer than 1% of normal B cells, xid mice have at least 50% of normal B cells (9). Thus, the limitation is that the mutant does not reach the severity of human immunodeficiency. Khan et al. constructed the first Btk-mutant mouse model by introducing null mutations, Btk-Kin- and Btk-PH-, into the Btk gene with the goal of better elucidating the functions of Btk. Compared to the previously mentioned xid model that still produced Btk gene product, these null mutations completely blocked the expression of the Btk protein. This is evidenced by the protein-ablation found in the western blot in Figure 4 (10). Pin-1, a negative regulator of Btk and a mitotic regulator depicted in Figure 2, has also been investigated in mice to understand regulation of Btk rather than a mutation of Btk. Pin-1 is a peptidyl-prolyl cis-trans isomerase and specifically dephosphorylates Btk, inducing the instability and degradation of the critical Btk protein (4). This is demonstrated by the missing bands and thus decreased or absent Btk in the presence of Pin-1 in the immunoprecipitation analysis of phosphorylated Btk in the in vitro COS-7 cell line in Figure 5. In the absence of Pin-1, the Btk protein is phosphorylated and stable, but in the presence of Pin-1, this stability is lost. Pin-1 may facilitate the destabilization of the Btk protein by recruiting a specific phosphatase that dephosphorylates Btk (4). The essential Btk protein is directly affected by conformational change, which in turn inhibits the previously

Figure 4. Western blot analysis of Btk-PH- and Btk-Kin- gene product demonstrating the protein ablation. From left to right: lysates from a normal mouse as wild-type control (Cont), a Btk-PH- mutant mouse (PH-), a Btk-Kin- mutant mouse (Kin-) and a molecular mobility ladder (T-cell). Image modified without permission (10).

Assessing the Effectiveness of the Murine Model in Advancing X-Linked Agammaglobulinemia Research

described molecular pathway that leads to B cell development complications (4). Another comparable mutant to the xid mice is the bone marrow-liver-thymus (BLT)-humanized mice that exhibit primary antibody deficiency with low gamma globulin levels (hypogammaglobulinemia), as well as a poor antigen-specific IgG response (11). This can phenotypically present as the XLA condition. The aforementioned Btk-deficient mice display signaling defects and are remarkably identical in phenotype to xid mice (10). It is interesting that the Btk point mutation in xid mice presents as the same loss-of-function phenotype as complete Btk deficiency, even though xid expresses the Btk mRNA and the Btk-/- mice do not express Btk gene product (10). Therefore, regulators of the protein can be identified with comparable model organisms and these studies utilize mice as a means to understand the mechanisms and clinical manifestations of XLA. Uncovering treatment possibilities of XLA with the mutant mice model The resemblance of the adaptive immune system being compromised makes the mutant model essential to XLA

Figure 6. Murine B cell differentiation and developmental stages identified by surface markers. CD43, IgM, IgD, HSA, and B220 serve as indicators of development (7).

Figure 5. Pin-1 negative regulator alters the Btk protein. COS-7 cell line in vitro was stained with anti-phosphotyrosine antibody in the top image and stained with anti-Btk antibody in the bottom image. White arrow points to post-translationally modified Btk and black arrow represents the normal Btk MW (4).

studies. According to researchers, the xid phenotype results in cell loss of early pre-B cells, which is the same developmental stage for XLA in humans (7). Figure 6 provides the appropriate progression from pro- to pre- to immature B cell formation and development in bone marrow and from immature to mature B cell in the spleen. In the same way that developmental stages can be observed by structural changes in the cell, specific markers can be used to identify the particular stage of a B cell. Maturation of the B cell is required for the ability to be activated by antigens to produce an adaptive immune response. Xid inhibits two points of this pathway, but the ectopic rescue Bcl-2 (B cell lymphoma 2) human transgene blocks the function of xid in the spleen. To illustrate this, xid/bcl-2 mice were generated, in which xid was deleted and ectopic Bcl-2 was expressed (7). Cancro et al. (7) validated previous findings concluding that the Btk-/- mice generated a reduced proliferation of mature B cells due to disrupted developmental transition. Although the restoration of pre-B cells does not reverse the xid lesion, ectopic Bcl-2 reverses the peripheral xid lesion since Btk functions differently at the checkpoints (7). Thus, the human Bcl-2 transgene enables some restoration of mature B cell number and presents a possible method of treatment in prolonging survival of pre-B cells and immature and mature B cells. One of the treatments for xid is a bone marrow transplant, yet it can be risky due to toxicity. A study immunized transplanted xid mice, wild type and untreated xid controls with NP-Ficoll, a hapten conjugated to a polysaccharide prior to irritating and activating the immune response (9). Since this HaptenFicoll mimics a T cell independent antigen, the wild-type mice produced an appropriate immune response with elevated levels of antibody titer after immunization, but the xid mice did not, as depicted in Figure 7. From the irradiated xid mice, none that received less than 2.5 x 106 transplanted cells made antibodies against NP, showing poor post-immunization reaction and suggesting the importance of the quantity of bone marrow transplant cells (9). Furthermore, the ability of increased transplant cells to improve the immunization response suggests a â&#x20AC;&#x153;transformingâ&#x20AC;? factor that can be therapeutic. Another possible treatment option is the transduction of a retroviral vector, with a murine stem cell virus promoter driving the expression of human Btk in bone marrow. In a study by Yu et al. (12) lethally irradiated mice deficient for the Btk homology and Tec homology domains (BtkTec-/-) were engrafted with hematopoietic cells, previously transduced with a vector expressing human Btk (12). This adoptive cell transfer experiment slightly restores Btk protein from 1% in BtkTec-/- animals to 11% for mice treated with the MBS (retroviral Btk-vector) in Btk bone marrow expression as depicted in Figure 8 and seems to be a way to improve immune system function. This rescue increases Btk gene expression, potentially reducing the severity of B cell loss in the manifestation of XLA. Each of these interventions are promising, as restoration of Btk function has been observed in mice by transgenic rescue in Figure 6, bone marrow transplantation in Figure 7, or viral vector gene transfer in Figure 8. The effects of murine xid and human Btk deficiency differ quantitatively but possess the same disease processes (13). Therefore, these options are promising to the future of human XLA treatment. However, the side effects such as cell ablation due to toxicity of each of these treatments 119

Assessing the Effectiveness of the Murine Model in Advancing X-Linked Agammaglobulinemia Research

assist in elucidating the molecular mechanisms fundamental to therapeutics in patients suffering from XLA.

References 1.

Figure 7. Assessment of bone marrow transplant in xid mice. Indication of pre (grey) and post (black) immunization measured by ELISA. Groups represented on the X-axis: Group 1 was radiated and received xid bone marrow transplant cells with 2.5 x 104 cells. The remaining xid were irradiated and transplanted with 2.5 x 106 cells (Group 2), 2.5 x 105 cells (Group 3), and 2.5 x 104 cells (Group 4). Xid was not treated and wildtype served as a control (9).

Conley, M, Howard V. X-Linked Agammaglobulinemia. GeneReviews: University of Washington, Seattle; 2001.

2. Vihinen M, Kwan S, Lester T, Ochs H, Resnick I, Valiaho J, et al. Mutations of the human BTK gene coding for bruton tyrosine kinase in X-linked agammaglobulinemia. Human Mutation 1999. p. 280–5. 3. Spencer G. Background on Mouse as a Model Organism. NIH National Human Genome Research Institute; 2012. 4. Mohamed, A, Yu L, Backesjo C, Vargus L, Faryal R, et al. Bruton’s tyrosine kinase (Btk): Function, regulation, and transformation with special emphasis on the PH domain. Immunological Reviews 2009. p. 58–73. 5. Macian F. NFAT proteins: Key regulators of T-cell development and function. Nat Rev Immunol Nature Reviews Immunology 2005. p. 472–84. 6. Bates K. X-linked agammaglobulinemia (XLA). 2005. 7.

Cancro M, Sah A, Levy S, Allman D, Schmidt D, et al. Xid mice reveal the interplay of homeostasis and Bruton's tyrosine kinase- mediated selection at multiple stages of B cell development. International Immunology 2001. p. 1501–14.

8. Mestas J, Hughes C. Of Mice and Not Men: Differences between Mouse and Human. Immunology. 2004. p. 2731–8.

Figure 8. Retroviral transduction. A) The murine stem cell virus drives human Btk expression in murine bone marrow. B) Transduction with viral vector rescues Btk deficiency (12).

must be evaluated to prevent unintended consequences. Further studies can elucidate the role of the Btk pathway in these options and how to restore B cells to a sufficient level in providing the humoral response necessary for normal body functioning without life-threatening infections.

Discussion X-linked agammaglobulinemia has been an incurable and severe immunodeficiency disease known for its array of infections and health risks. The essential tyrosine kinase protein Btk is positively and negatively regulated by cellular pathways in B lymphocyte development to produce the phenotypic outcome of the XLA condition. To investigate the effects of mutation on the Btk protein and to propose treatment options in restoring humoral immune function, the murine mouse model has been widely used in medical research with more advantages than disadvantages. Studies reveal that the cellular processes of B cell deficiency can be replicated in this model and several interventions have succeeded in restoring Btk and overall immune functionality. In future studies with the murine model, exploring the signal transduction of Btk, assessing the safety of gene transfer and understanding why different mice strains can vary in transplantation response may 120

9. Porpiglia A, Rohrer J, Conley M. Reconstitution of B cell function in murine models of immunodeficiency. Clinical Immunology 2003. p. 90–7. 10. Khan W, Alt F, Gerstein R, Malynn B, Larsson I, Rathbun G, et al. Defective B cell development and function in Btkdeficient mice. Immunity 1995. p. 283–99. 11. Martinez-Torres, F, Nochi T, Wahl A, Garcia J, Denton P. Hypogammaglobulinemia in BLT Humanized Mice–An Animal Model of Primary Antibody Deficiency. 10 ed. PLoS ONE 2014. 12. Yu, P, Tabuchi R, Kato R, Astrakhan A, Humblet-Baron S, et al. Sustained Correction of B-cell Development and Function in a Murine Model of X-linked Agammaglobulinemia (XLA) Using Retroviral-mediated Gene Transfer. Blood 2004. p. 1281–90. 13. Kerner J, Appleby M, Mohr R, Chien S, Rawlings D, Maliszewski C, et al. Impaired expansion of mouse B cell progenitors lacking Btk. Immunity 1995. p. 301–12.

Scholarly Research In Progress • Vol. 3, November 2019

Buprenorphine Distribution between 2007 and 2017 in the United States Alexandra Cruz-Mullane1*, Jaclyn C. Podd1*, Stephanie D. Nichols2,3, Kenneth L. McCall3, and Brian J. Piper1,4

Geisinger Commonwealth School of Medicine, Scranton, PA 18509 Husson University School of Pharmacy, Bangor, ME 04401 ³University of New England, Biddeford, ME 04005 ⁴Center for Pharmacy Innovation and Outcomes, Geisinger Precision Health Center, Forty Fort, PA 18704 *Master of Biomedical Sciences Program Correspondence: jpodd@som.geisinger.edu 1

Abstract Buprenorphine is a partial opioid agonist that suppresses cravings and opioid withdrawal, making it an effective treatment for an opioid use disorder (OUD). This paper analyzed its overall distribution from 2007 to 2017, primarily focusing on its retail distribution in 2016 and 2017 across the United States including the territories, as reported to the Drug Enforcement Administration (DEA). Buprenorphine data was obtained through the Automated Reports and Consolidated Ordering System (ARCOS). The results showed pharmacies as the (>90%) predominant source of buprenorphine distribution over the past 10 years. Examining the data, figures, and trends leads to the conclusion that buprenorphine distribution has increased in the U.S. (both in the states and territories) over the past 10 years, but there are pronounced differences in per-capita use of this OUD treatment. More research into buprenorphine’s effectiveness as an OUD pharmacotherapy and potential for nonmedical use is needed.

Introduction The U.S. opioid epidemic has developed into one of the country’s largest public health concerns, with prescription opioid distribution peaking in 2011 (1). The opioid epidemic has large and far reaching effects on the U.S. Chronic pain affects 116 million Americans, more than those affected by heart disease, cancer, and diabetes, combined (2). Opioid dependency and misuse only further augment the distribution of opioids. The quantity of opioid prescriptions has quadrupled in the last 10 years, with opioid-related deaths tripling in the last 10 years (2). Within this period, there has also been an increase in buprenorphine retail distribution over the past 5 years (1). For example, buprenorphine (combined with methadone) made up approximately 89% of Puerto Rico’s total morphine mg equivalent (MME) in 2017 (3). Buprenorphine was originally produced in 1978 in the United Kingdom (5). It is a mu receptor partial-agonist, a kappa receptor antagonist, and a delta receptor antagonist. Buprenorphine functions to suppress cravings, withdrawal symptoms, and common euphoric effects associated with opioids (6). Patients are not in withdrawal, nor do they experience a “high” from buprenorphine or other opioid drugs when maintenance with therapeutic buprenorphine doses are employed. However, buprenorphine will displace any circulating opioids, which could be problematic in precipitating withdrawal if it is started in a patient already dependent on a full agonist such as methadone or heroin (7). Buprenorphine is also associated with a ceiling effect,

meaning after a certain dose, taking more will not increase the effects. Buprenorphine’s most common adverse effects include sedation, dizziness and nausea. Additional, but rare, adverse effects might involve CNS depression and anticholinergic effects (8). It has a high first pass effect, with a high volume of distribution and highly variable half-life. It is highly protein bound and metabolized by P450 CYP 3A4 to norbuprenorphine (5,9). Buprenorphine comes in several formulations: a parenteral formulation for moderate to severe pain, a transdermal patch for continuous pain management, a sublingual tablet primarily for OUD and off-label for chronic pain, and a transmucosal film with naloxone for opioid dependence (5). Buprenorphine is a schedule III-controlled substance. Buprenorphine is often compared to methadone. Data remains inconclusive concerning which is better for OUD (10). However, both methadone and buprenorphine are more cost-effective than no drug therapy, and both have been shown to reduce the mortality rate when compared to non-pharmacotherapy. A combination buprenorphine/naloxone generic product has recently become available in the U.S. Buprenorphine is newer than methadone, and some physicians are less familiar with prescribing it (11) or have not completed X-waiver training. Buprenorphine is similarly effective to methadone at equivalent doses of 16 mg/day of buprenorphine and >84mg/ day of methadone. Additionally, buprenorphine is safer than methadone due to the decreased likelihood of overdoses owing to the ceiling effect (4). It is also less socially stigmatizing in that it is not detectable by standard urine drug tests, and it can be prescribed by waivered physicians—meaning that the patient does not have to go to a government clinic every day to obtain it—unlike methadone (6). The main objective of this study was to examine buprenorphine distribution from 2007 to 2017 in the U.S. states and territories through the lens of different sources of distribution (e.g., pharmacies, hospitals, practitioners, Narcotic Treatment Programs [NTPs]). A closer examination of retail distribution in 2016 and 2017 per state/ territory was also performed in order to determine if any discernable patterns of changes in distribution existed.

Materials and Methods Procedures This is a secondary analysis of data obtained from the Automated Reports and Consolidated Ordering System (ARCOS) Retail Drug Summary Reports (12) found on the Drug Enforcement Administration’s (DEA) website. According

121

Buprenorphine Distribution between 2007 and 2017 in the United States

to Report 7: U.S. Summary of Retail Drug Purchases, 2007 to 2017 buprenorphine-related data was gathered from four distribution sources: pharmacies, hospitals, practitioners, and NTPs. Additional data was obtained from Report 2: Retail Drug Distribution by Drug Code for Total U.S. for 2016 and 2017 buprenorphine distribution. Population data for 2007 through 2017 was obtained from the U.S. Census Bureau for U.S. mainland states, the District of Columbia and Puerto Rico. Finally, 2007 to 2017 population data was obtained from the World Bank for American Samoa, Guam, and the Virgin Islands. This study was deemed exempt by the Institutional Review Board of the University of New England. Data analysis The 2007 to 2017 buprenorphine distribution data from Report 7 of ARCOS was converted from gram to kilogram from four different distribution sources: pharmacies, hospitals, practitioners, and NTPs. Mid-level providers (i.e., advanced practice providers), and teaching institutions were not included in the analysis because of the sparse distribution of buprenorphine from these sources. The 2007 to 2017 total buprenorphine distribution data from Report 7 was converted from gram to kilogram and corrected for total US state and territory per 100â&#x20AC;&#x2030;000 people. The raw weight of buprenorphine distributed within each US state and territory in 2016 and 2017 was obtained from ARCOS Report 2. A percent change in distribution between 2016 and 2017 data was calculated. The results were converted from gram to kilogram and plotted on geographic heat maps. Buprenorphine data for 2017 for each U.S. state and territory was further examined and expressed as kilogram of buprenorphine and kilogram/100â&#x20AC;&#x2030;000 people. Each state or territory was then ranked 1 to 55; for example, 1 represents the highest amount of buprenorphine distribution. All figures and data analysis were completed with GraphPad Prism, version 8.1.0 and Microsoft Excel version 16.23.

Figure 1. Total 2007 through 2017 retail buprenorphine distribution in the U.S. states and territories

Results Figure 1 demonstrates total buprenorphine distribution per capita from 2007 through 2017, and the overall distribution showed an upward trend. Figure 2 shows buprenorphine by distributor from 2007 through 2017. The distribution of buprenorphine has more than tripled in 2017 relative to 2007. Pharmacies made up approximately 95% of all retail purchases. Hospitals showed a gradual increase in distribution, although the amount was relatively modest compared to pharmacies. Buprenorphine distribution by practitioners peaked at 29,218 kg in 2009, followed by decreases thereafter. Distribution from NTPs remained consistently low until 2011 to 2012, followed by a continuous increase. Figure 3 highlights distribution of buprenorphine by state. Buprenorphine distribution was lowest in American Samoa (rank = 55), Guam (54), the U.S. Virgin Islands (53), South Dakota (52), and North Dakota (51), while it was highest in Massachusetts (5), Florida (4), Ohio (3), New York (2), and Pennsylvania (1). The population-corrected data (buprenorphine mg/person) in Figure 4 showed that buprenorphine distribution was lowest in American Samoa (55), Guam (54), the U.S. Virgin Islands (53), South Dakota (52), and Iowa (51), and highest in Rhode Island (5), Maine (4), Kentucky (3), West Virginia (2), and 122

Figure 2. Total 2007 through 2017 retail buprenorphine distribution in the U.S. states and territories by distributor

Vermont (1). State rankings 52 through 55 remained the same after the population correction. State rankings 1 through 5 changed noticeably, as there were no common states between Figures 3 and 4. In Figure 3, North Dakota had a ranking of 5, but jumped to a ranking of 13 in Figure 4. Some states had extreme ranking changes in percent of buprenorphine distribution when correcting for population: Vermont (+97.297%), Maine (+85.714%), Rhode Island (+85.294%), Florida (-675%), New York (-1200%), and Pennsylvania (-9,000%). Additionally, the U.S. territories all had no changes in ranking with the exception of Puerto Rico (which went from 43 to 49, with a percent change of -13.953). The percent change between 2016 and 2017 buprenorphine distribution data was examined critically by region as seen in Figure 5. Adjusting for population, the Northeast was distributed evenly among increases and decreases of

Buprenorphine Distribution between 2007 and 2017 in the United States

buprenorphine retail rankings (5 states went up and 4 states went down). The Midwest rankings dropped greatly with the population correction (10 states went down, 1 state went up, and 1 state had no change). The South was dispersed evenly, much like the North (9 states went up and 8 went down), and the West exhibited a fair distribution as well (7 states went up and 6 went down). Again, the territories had no changes, except for Puerto Rico (which went down).

Discussion

Figure 3. Total 2017 buprenorphine retail distribution in the U.S. states and territories

There are several key findings from this report. Figure 1 indicated that retail buprenorphine distribution per capita in the U.S. states and territories has steadily increased from 2007 through 2017. Specifically, Figure 2 shows that pharmacies have been the predominant distributor of buprenorphine from 2007 through 2017, and they have made up approximately 95% of all retail purchases of buprenorphine in 2017. This is consistent with previous data showing that pharmacies were the main source responsible for buprenorphine distribution in the U.S. states and territories (1, 3). The exception is Puerto Rico, which had the highest average MME per capita of buprenorphine and methadone distribution coming from NTP (3).

Figures 3 and 4 compared the uncorrected buprenorphine distribution data and the population-corrected data for each state/ territory in 2017. Generally, the states with larger populations had higher retail of buprenorphine, and states with smaller buprenorphine retail had higher percent change in growth. The territories (Guam, Figure 4. Total 2017 buprenorphine retail distribution in the U.S. states and territories, Puerto Rico, and the Virgin Islands), on the population corrected other hand, do not follow this expectation. This is quite possibly due to lack of access and population sizes that are not as large as the mainland states (3).

Figure 5. Percent change in retail buprenorphine between 2016 and 2017

As previously mentioned, Figure 5 shows the regional differences in percent change of buprenorphine distribution. Two regional areas (the Midwest and the South) did not follow expected trends. In the Midwest, 10 states went down, 1 state went up, and 1 state had no change. This may be partially explained by the fact that when looking geographically in the United States, the Midwest is largely lacking providers who can prescribe buprenorphine (15). In the South, 9 states went up and 8 went down; trends in some of these states more closely resemble those in the Northeast region (e.g., Delaware, Maryland, and DC). As mentioned, both the Northeast and West regions exhibited even buprenorphine distributions, while the territories had no changes, except for Puerto Rico. The results from Figures 3 and 4 may imply that state/territory population size consideration is an important 123

Buprenorphine Distribution between 2007 and 2017 in the United States

factor in buprenorphine distribution discussions. Disparities in buprenorphine use may also be associated with socioeconomic and racial differences (13,14). Specific state trends may be explained by their approaches to buprenorphine and the opioid crisis. In 2016, California had one of the highest numbers of waivered physicians in the US, but it was also found to have the lowest number of patients receiving buprenorphine treatment per physician (16). Possible reasons for this include novice prescribers’ inefficient access to more experienced buprenorphine-prescribing physicians as well as a need for new strategies to help current prescribers treat patients safely and effectively (16). Additionally, there seems to be a lower need for them as exhibited by OD rates and scarce fentanyl-cut heroin. New Jersey has adopted methods in an attempt to combat the opioid crisis, including a much more stringent set of protocols for prescribing opioids for first-time opioid users, more strict controls and parameters for opioid prescriptions, and pushing policies that would diminish doctor shopping (17). Similarly, two Pennsylvania health plans (Independence Blue Cross/IBC Foundation and UPMC Health Plan) are working to decrease and prevent OUD. Their plans include promoting guidelines and enforcing certain regulations, increasing warm handoffs and exposure to behavioral health care. IBC Foundation covers methadone and UPMC Health Plan covers buprenorphine. With some of these policies, the IBC Foundation has already managed to decrease the number of members taking opioids by 30% and the number of opioid prescriptions by more than a third (18). Vermont has exponentially increased its capacity to treat patients with OUD within approximately the last 10 years, making it currently the state with the highest capacity to treat OUD (19,20). Credit to this expansion has been given to the enactment of various state-level legislation, leading to a “hub-and-spokes” model of health care for the treatment of OUD. It should also be recognized that buprenorphine may have some potential for nonmedical use. There were more arrests reported to the Maine Diversion Alert Program for buprenorphine than for oxycodone, hydrocodone, methadone, tramadol, and morphine combined (21). The street value of Suboxone (buprenorphine/naloxone) was listed at $10 to $20 for the 8 mg/2 mg or 12 mg/3 mg formulations (22). Suboxone was ranked 120th in the U.S. in 2007 ($70.6 million/quarter) but increased to 28th with $415 million in sales by the first quarter of 2013 (23). It may also be that people are self-treating for OUD due to limited access via the health care system. Ultimately, a deeper examination into certain state policies will allow for a deeper understanding of the patterns of buprenorphine use. The main limitation of this report was lack of access to the 2018 ARCOS data at the time of data collection. An unknown (but presumably negligible) volume of buprenorphine is used for feline anesthesia (24). ARCOS does not contain a mechanism to filter out buprenorphine when used for acute veterinary pain relief (25) or to differentiate between the formulations used for opioid withdrawal or maintenance versus for pain. As buprenorphine continues to increase in U.S. distribution, additional examinations of other state-level specific approaches or state-level policies regarding this opioid’s effectiveness in the treatment of OUD would be warranted. Moving forward, it is important that more information is obtained on the efficacy of these models and how they can be replicated in other states. 124

Future research should also include alternative treatments for OUD. By doing so, alternative and more cost-effective methods to help various people and populations may be found. Reducing the social stigmas of OUD and its management are also important for the future treatment of OUD. Buprenorphine could be considered a safe alternative, and it could potentially be prescribed more often; additionally, more health care providers need to undergo training to be able to prescribe buprenorphine for OUD (26).

Conclusion This study focused on buprenorphine distribution in the U.S. over the last 10 years, which has steadily increased. The predominance of buprenorphine retail comes from pharmacies. Accounting for state/territory population size may be important for a fuller understanding of buprenorphine distribution in the U.S. states and territories. With the 2018 ARCOS data, it will be interesting to determine whether the elevations continue. Moving forward, future research should further investigate the cause(s) of buprenorphine trends in order to influence future policy on opioid treatment(s), as well as to better determine the efficacy and the patient outcomes of buprenorphine when used for OUD.

Acknowledgments We would like to thank Elizabeth Kuchinski, MPH, and our classmates in the GCSOM Community Health Research class of 2019 for their helpful feedback on an earlier version of the manuscript. Software to support this research was generously provided by the graduate school. Additionally, we thank Iris Johnston in the library for technical support.

Disclosures ACM, JCP, SDN, and KLM have no disclosures. BJP has a grant in review with Pfizer and is supported by the Fahs-Beck Fund for Research & Experimentation and the Health Resources Services Administration.

References 1.

Piper BJ, Shah DT, Simoyan OM, McCall KL, Nichols SD. Trends in medical use of opioids in the U.S., 2006–2016. Am J Prev Med. 2018;54(5):652–660.

2. Garcia AM. State laws regulating prescribing of controlled substances: balancing the public health problems of chronic pain and prescription painkiller abuse and overdose. J Law Med Ethics. 2013;41 Suppl 1:42–5 3. Cabrera FF, Gamarra ER, Garcia TE, Littlejohn AD, Chinga PA, Pinentel-Morillo LD, et al. Opioid distribution trends (2006–2017) in the US Territories. PeerJ. 2019;7:e6272. 4. Hefei W, Jason MH, Harold AP. Association of buprenorphine-waivered physician supply with buprenorphine treatment use and prescription opioid use in Medicaid enrollees. JAMA Netw Open. 2018; 1(5): e182943. 5. Cleary JP, Gottwald J. A brief review of buprenorphine products [Internet]. Pharmacy Times. 2016.

Buprenorphine Distribution between 2007 and 2017 in the United States

6. Strain E. Pharmacotherapy for opioid use disorder. UpToDate. 7.

Whelan PJ, Remski K. Buprenorphine vs methadone treatment: A review of evidence in both developed and developing worlds. J Neurosci Rural Pract. 2012; 3(1): 45–50.

21. Simpson KJ, Moran MT, Foster ML, Shah DT, Chung DY, Nichols SD, et al. Descriptive, observational study of pharmaceutical and non-pharmaceutical arrests, use, and overdoses in Maine. BMJ Open 2019; 9:e027–117. 22. StreetRx: latest street prices for prescriptions drugs. Accessed 5/5/2019 at: streetrx.com

8. Kumar R, Saadabadi A. Buprenorphine. StatPearls [Internet] Bottom of Form 2019.

23. Top 200 drugs for 2007. Accessed 5/5/2019 at: https:// www.drugs.com/top200_2007.html

9. Elkader A, Sproule B. Buprenorphine: clinical pharmacokinetics in the treatment of opioid dependence. Clin Pharmacokinet. 2005;44(7):661–80.

24. Sramek MK, Haas MC, Coleman GD, Atterson PR, Hamlin RL. The safety of high-dose buprenorphine administered subcutaneously in cats. J Vet Pharmacol Ther. 2015; 38(5):434–42. doi: 10.1111/jvp.12203.

10. Gowing L, Ali R, White JM, Mbewe D. Buprenorphine for managing opioid withdrawal. Cochrane Database Syst Rev. 2017, Issue 2. Art. No.: CD002025. DOI: 10.1002/14651858.CD002025.pub5

25. Flecknell PA. The relief of pain in laboratory animals. Lab Anim 1984; 18:147–160. 26. Buprenorphine [Internet]. SAMHSA. 2019.

11. Connock M, Juarez-Garcia A, Jowett S, Frew E, Liu Z, Taylor RJ, et al. Methadone and buprenorphine for the management of opioid dependence: a systematic review and economic evaluation. Health Technol Assess. 2007; 11:1–171. 12. Diversion Control Division. Automation of reports and consolidated orders systems (ARCOS) [Internet]. US Department of Justice Drug Enforcement Agency. 13. Duncan LG, Mendoza S, Hansen H. Buprenorphine maintenance for opioid dependence in public sector health care: Benefits and barriers. Journal of Addiction Medicine and Therapeutic Science. 2015;1(2):31–36. 14. Hansen H, Siegel C, Wanderling J, DiRocco D. Buprenorphine and methadone treatment for opioid dependence by income, ethnicity and race of neighborhoods in New York City. Drug and Alcohol Dependence. 2016;164:14–21. 15. Andrilla CHA, Moore TE, Patterson DG, Larson EH. geographic distribution of providers with a DEA waiver to prescribe buprenorphine for the treatment of opioid use disorder: a 5-year update. J Rural Health. 2019;35(1):108– 112. 16. Stein BD, Sorbero M, Dick AW, Pacula RL, Burns RM, Gordon AJ. Physician capacity to treat opioid use disorder with buprenorphine-assisted treatment. JAMA. 2016; 316(11):1211–121. 17. Stainton LH. Rutgers med first in U.S. to offer certified medication-assisted treatment [Internet]. NJ Spotlight. 2018. 18. Lewis D. Protecting patients from opioid abuse [Internet]. AHIP. 2017. 19. Traynor K. Opioid control law in Vermont addresses provider status for pharmacists. Am J Health Syst Pharm. 2016;73(14):1024–1025. 20. Brooklyn JR, Sigmon SC. Vermont hub-and-spoke model of Care for opioid use disorder: development, implementation, and impact. J Addict Med. 2017;11(4):286– 292.

125

Scholarly Research In Progress • Vol. 3, November 2019

The Cellular Hallmarks of Multiple Sclerosis Ashley A. Bross1*‡, Laura M. Loeser1*‡, Elizabeth J. Pavis1*‡ and Alison T. Varano1†‡

Abstract Multiple sclerosis (MS) is a debilitating autoimmune and neurodegenerative disease. The etiology of this chronic disease is yet to be discovered. The disease consists of three different forms: relapsing and remitting, secondary progressive, and primary progressive. The severity and progression of this disease varies among the stages. Disease modifying treatments have been identified to slow disease progression, but no cure has been found. There are several hallmarks involved in the manifestation of this lifealtering condition. The most namely hallmarks, which this review will focus on, are leukocyte infiltration, astrogliosis, neurodegeneration, axonal damage, and intermediate activation of macrophages. Understanding the characterization of these clinical symptoms and their role in MS is essential for identifying improved treatments and prevention of the disease.

Introduction This review examines MS, an autoimmune disease of the central nervous system involving inflammation and demyelination. One of the most indicative findings of MS is the pervasive leukocyte infiltrates along demyelinated axons (1). However, there are other factors that are indicative of this neurodegenerative disorder as well, such as astrogliosis, axonal loss, neurodegeneration, and remyelination (1). A previous study has demonstrated the presence of other immune cells such as activated microglia in demyelinating MS lesions (1). This review will further explore the characteristics and stages of this autoimmune disease, including the bloodbrain barrier and leukocyte infiltration, the role of astrocytes, neurodegeneration, and axonal damage as the disease worsens and further review on the effects of macrophages and microglia in MS. MS generally presents with neurological dysfunction at 30 years old followed by periodic remission (2). Symptoms of this neurological disorder include slurred speech, fatigue, numbness and tingling, vision problems, pain, and emotional changes (2). During periods of remission, these symptoms may alleviate but then return during periods of neurological dysfunction. The exact etiology of this disease is unclear but it has been proposed that there is involvement of both genetic and environmental factors (3). The onset of MS typically occurs between 20 and 40 years of age, which indicates this is the most common neurological disability among this age group (3). More than 2.5 million individuals are affected, and more women are affected than men. A previous study has demonstrated an increased likelihood in disease acquisition when siblings are also affected, with the strongest correlation

126

between likelihoods in monozygotic twins (3). However, the researchers conclude the risk does not reach 100% in monozygotic twin studies, which suggests environmental components play a role (3). Such environmental factors include smoking, a history of chronic viral infections, vitamin deficiencies, and diet/exercise, as well as geographic location and ethnicity. For example, individuals living in northern Europe and North America exhibit a higher disease incidence (3). MS can be defined in three stages: relapsing-remitting multiple sclerosis (RRMS), secondary progressive multiple sclerosis (SPMS), and primary progressive multiple sclerosis (PPMS). Disease progression among individuals of MS may vary, however, the most common form of disease is RRMS. Patients with RRMS develop symptoms and neurological dysfunction for a varying duration of time, which may then resolve back to an asymptomatic state (4). Autoreactive T cells may cause inflammation and demyelination in the CNS, thus resulting in the return of symptoms (5). Magnetic resonance imagining (MRI) may be useful in MS diagnosis by the appearance of lesions occurring from damage to the CNS (5). Inflammation may resolve during the remission phase and remyelination may occur to damaged demyelinated axons. Remyelination may restore neurological function in lesions of MS; however, over time the process of remyelination begins to fail (6). It is hypothesized that failure of remyelination leads to the progression of the disease (6). The mechanisms behind failure of remyelination are not fully understood, but may be a therapeutic target for future studies. Unfortunately, a large number of patients with RRMS will then develop SPMS, which is characterized as the continual progression of the disease without symptoms subsiding (4). As the disease progresses, axonal damage accumulates and chronic inflammation results in scar formation (5). Symptoms and disability steadily worsen in SPMS. In a smaller percentage of patients, the disease progression is rapid without symptoms lessening and is diagnosed as PPMS (4). Although prognosis of PPMS is poor, diagnosis of the disease still remains challenging as symptoms may vary among patients. In comparison to SPMS, PPMS is characterized by less inflammation and corresponding brain lesions and instead, more prominent spinal cord lesions (5). Over the past couple decades, life expectancy of patients with MS has increased substantially, with a difference of only 6 to 7 years between the life expectancy of an MS patient and a healthy individual (7). There is currently no cure for this neurodegenerative disorder. However, treatments are available to help manage MS symptoms in order to make the patient more comfortable. Examples of such treatments include corticosteroids and plasmapheresis, which are

The Cellular Hallmarks of Multiple Sclerosis

used for attacks of MS and result in substantial improvement in neurologic function (8). Additionally, there are diseasemodifying treatments that aim to hinder disease progression. One of the FDA-approved therapies to modify progression is the drug ocrelizumab. Ocrelizumab-treated patients have shown slightly decreased disease progression in primary progressive patients when compared to the control (9). Further disease-modifying therapies will be discussed subsequently in this paper.

Discussion Neurodegeneration and axonal damage Neurodegeneration is characterized as the gradual loss or damage to neurons and axons over time in the central nervous system (10). Further understanding of the causes of neurodegeneration can be beneficial for future therapies to treat MS. Researchers have focused on the relationship between inflammation and neurodegeneration in individuals with MS. Neurodegeneration has been believed to occur as a result of inflammation. However, there have been previous studies that have proposed neuronal degeneration may be occurring independently from inflammation in progressive MS (11, 12). An investigation analyzed if neurodegeneration— and thus progression of the disease—can occur without the presence of inflammation (13). If the investigation found that neuronal loss is present in the absence of inflammation, then the researchers can conclude there is an independent relationship between inflammation and neurodegeneration (13). Researchers found that active demyelination and neurodegeneration was present only in patients with inflammation in the brain, and therefore reported a close association between neurodegeneration and inflammation (13). Axonal loss or injury specifically, a characteristic of MS, has previously been reported may be due to demyelination (4). Axons are myelinated by oligodendrocytes in the central nervous system. Oligodendrocytes produce myelin that surrounds the axon for protection and supports survival. When axonal damage is present, there may be diminished transport of proteins along the axons. The precise mechanisms behind axonal injury are still not fully understood. Throughout MS, as the disease progresses, damage to the axons may result from erosion of the protective myelin sheath layer, which is referred to as demyelination. The amyloid precursor protein (APP), is present in lesions with demyelination and has been used previously to quantify axonal injury. However, a report analyzes axonal damage as a result of alternative mechanisms from demyelination and concluded axonal injury may occur independently of demyelination. Axonal damage was found to have a possible mechanism by deregulation of glial-neuronal crosstalk and exposure to glutamate and cytokines (4). It has been seen that in neurodegenerative disorders, cytokines are released due to the inflammatory response. Also, glutamate, when present in high amounts, can be associated with neuronal death (10). Thus, researchers propose it is possible that axonal damage can occur independently of demyelination (4).

An additional study also investigated the relationship between axonal damage and demyelination (14). Similar to the previous investigation, this study used APP expression to quantify axonal injury and proposed that axonal injury and demyelination may have slightly differing mechanisms. Demyelination was found to be reversible by remyelination; however, axon transection was suggested to be irreversible. Increasing axonal damage correlated with increasing macrophages and CD8-positive T cells, which are both killer cells of the immune system, in contrast to tumor necrosis factor-alpha and inducible nitric oxide synthase, which are cell signaling proteins involved in inflammation and were used as mediators for demyelination. These findings indicate the importance for future therapies targeting both demyelination and axonal injury. Deficits in neuromuscular transmission have been examined when considering determinants of MS. Individuals with MS lack acetylcholine, which is also seen in patients with myasthenia gravis. This indicates that agents used in myasthenia may be clinically useful in patients suffering from MS. This was proposed because a neuromuscular concern of MS is resultant upon the fact that the action potentials do not propagate to the axon terminal appropriately, releasing adequate acetylcholine (15). The overlap between MS and myasthenia gravis may aid future research in developing diagnostic and therapeutic technologies. Since these preliminary findings, though, there has not been much research regarding anticholinesterase therapy and repetitive electrical stimulation published in literature. In future studies, the hope is that these types of treatments can be implemented in MS. Astrocyte involvement in MS Astrocytes play a multifaceted role in MS and the disease’s progression. Astrocytes’ many roles include its part in the innate immune system, its cytotoxic factors, its inhibition of remyelination and axonal regeneration by glial scar formation, its role in axonal mitochondrial dysfunction, and the role age has on astrocyte regulatory mechanisms. In this section, there will be a brief synopsis of each of these astrocyte factors and how each contributes to the pathology of MS. Firstly, astrocytes take part in the innate immune system, which is the body’s mechanism for non-specific response to foreign pathogens. The innate immune system consists of cellular barriers, such as the blood-brain barrier (BBB), cells of myeloid origin and cells of non-myeloid origins, like astrocytes. Astrocytes have a wide variety of pattern recognition receptors (PRRs) and therefore can direct the immune system in many ways. In particular, astrocytes PRRs are involved in influencing the BBB’s permeability and the subsequent leukocyte infiltration. Astrocytes also release cytokines that attract peripheral immune cells and resident central nervous cells to active lesion sites. This action may be the primary reason for astrocyte immune-mediated demyelination and neurodegeneration. Additionally, astrocytes affect the outcome of T lymphocytes' phenotypes. Cytokines, which can be a number of substances released by astrocytes, can determine whether a T cell develops into a Th1 cell, a Th17 cell, or a noninflammatory T cell phenotype. Astrocytes also can contribute to the activation of macrophages and microglia (16).

127

The Cellular Hallmarks of Multiple Sclerosis

Another role that astrocytes play with regard to MS is that they are a source of cytotoxic factors. Astrocytes release compounds with damaging effects on neurons, axons, oligodendrocytes, and myelin. These toxic compounds include reactive oxygen and nitrogen species, glutamate and ATP. Glial scar formation by astrocytes is another contributing factor to MS progression. A glial scar consists mainly of astrocytes, but may also contain oligodendrocyte progenitor cells and fibromeningeal cells depending on the severity of the lesion. Glial scars are prominent in tissues obtained from MS patients. While the initial function of glial scarring is to protect the remaining brain tissue from damage, glial scars also inhibit the remyelination process and the axonal regeneration process (16). Both of these processes are inhibited by separate mechanisms. Astrocytes also contribute to mitochondrial dysfunction which may lead to neurodegeneration and axonal damage, both key characteristics of MS. It is believed that astrocytes deplete mitochondrial metabolism through multiple pathways. First, astrocytes increase the release of nitric oxide. Additionally, the increased release of glutamate also may contribute to mitochondrial dysfunction. The third plausible route is due to defective glycogenolysis and lactate formation due to a beta-2 adrenergic receptor deficiency in astrocytes. This mechanism ultimately leads to a reduction in N-acetyl aspartate (NAA) synthesis. This decrease in NAA can contribute to the alteration of myelin membrane turnover and lead to myelin loss, a hallmark sign of MS. Lastly, aging affects astrocytes and their role in regulating synaptic plasticity, metabolic balance, and BBB permeability. As individuals age, there is an increase in TGFβ1 signaling in the brain. TGFβ1 prevents astrocyte proliferation and increases GFAP (glial fibrillary acidic protein) expression, which functions to help maintain astrocyte integrity. Additionally, TGFβ1 is known to induce senescence in cells. Senescent astrocytes possess a decreased ability to promote neuronal survival and neurite outgrowth. Additionally, senescent astrocytes release proinflammatory cytokines. Ultimately, cellular senescence in astrocytes results in a low level of chronic inflammation, which enhances acute pathological conditions, while also worsening age-related neurodegenerative processes like MS (16). Immune response in MS: macrophages and microglia As MS is a chronic inflammatory disease, the immune system, and specifically macrophages, play a critical role in its pathology. Macrophages are typically known as the classical white blood cells responsible for phagocytosing foreign bodies and waste. However, in the pathology of MS, macrophages are seen to have a dual function. Macrophages are involved in lesion formation and axonal damage, while also producing anti-inflammatory cytokines which contribute to repair mechanisms. There are many different known subpopulations of macrophages which can be categorized according to their phenotype. Their phenotypes are determined based on their surface markers and cell function. The macrophages can then differentiate into these subpopulations via different signals such as cytokines (1). Macrophages can be broken down into two main categories: the classically activated M1 and the alternatively activated M2

128

categories. The classically activated M1 were described as having both cytotoxic and pro-inflammatory properties, while the alternatively activated M2 are seen to be involved in growth promotion and tissue repair (1). M1 macrophages function to protect against pathogenic microbes by means of producing nitric oxide and reactive oxygen species. M2 pathway macrophages confer advantages in tissue repair by producing anti-inflammatory cytokines while also building extracellular matrix material (17). As M1 and M2 macrophages seem to have opposing functions, it makes sense that different signals will stimulate their formation. While there are many signals that can trigger the formation of an M1 versus and M2 macrophage, the main signals for M1 are interferon-γ (IFN-γ) or lipopolysaccharide (LPS), whereas for M2 it is mainly interleukin-4 (IL-4). There are methods of determining the stage of a patient’s MS by examining immune cell status. HLA-DR (human leukocyte antigen–DR isotype) is a protein receptor found on professional antigen-presenting cells. HLA-DR is specifically a MHC Class II receptor and its presence is critical in determining the staging of an individual’s MS. An abundance of CD68-positive macrophages is also useful for staging the lesions. Investigating the markers of M1 vs. M2 macrophages in chronic MS lesions demonstrates that M2 markers were not present in chronic active lesions, indicating that the M2 markers decrease as the active lesions further develop into chronic active lesions (1). The M1 and M2 markers that were indicative of each type of macrophage were CD40 and MR, respectively; however, the researchers also found that MS lesions associated with macrophages possess an intermediate activation status due to the fact that the majority of macrophages in active demyelinating lesions expressed both types of markers. This opens the possibility for future research to determine factors that could lead to this intermediate phenotype. Microglia, like the macrophages looked at in the previous study, have been shown to have two opposing activities. They are seen to be involved in both the destruction of myelin and the remyelination process in MS patients. In the past, scientists have described macrophages and microglia as one in the same due to their similarities in functions and morphology. However, recent studies have shown that they are actually quite different. It was discovered that the main difference lies in the work they do that contributes to the demyelination process involved in MS (18). Macrophages appear to initiate the demyelination process and are associated with nodes of Ranvier, whereas microglia were found to clear the debris generated from the breakdown in the mouse model of MS. Like macrophages, microglia are also seen to play a dual role in MS pathology. Activated microglia were found to contribute to demyelination and neurodegeneration, by producing ROS and NO radicals in MS lesions. Microglia are seen to also contribute to the remyelination process by releasing cytokines such as TNF-a, IGF-1 and FGF-2 which are active in the proliferation of oligodendrocytes (18). The novel research done on intermediately activated macrophages could possibly help develop new therapies for patients with MS. As discussed, the M2 macrophages are the phenotype that are known to support repair functions (1). They are stimulated by IL-4. A current therapy approved by the FDA,

The Cellular Hallmarks of Multiple Sclerosis

glatiramer acetate is a first line agent that induces Th2 cells, which secrete anti-inflammatory cytokines including IL-4 (19). This therapy has been found to promote the M2 phenotype through the secretion of the IL-4. This drug inhibited nitric oxide, which is known to stimulate the M1 pro-inflammatory phenotype. Developing new therapies for MS can prove to be a difficult task, namely because of the role macrophages and microglia play in demyelination and remyelination at different time points of the disease. Although a simple response may be to suggest therapies should be targeted at the M1 proinflammatory phenotypes due to their cytotoxic nature, these macrophages are also seen to play an important role in parts of the remyelination process. Further research into the time points where the activity of M1 and M2 macrophages are beneficial or detrimental could lead to novel therapies that target these phenotypes at the optimum time for enhancement of remyelination and suppression of demyelination (19). Leukocyte infiltration of the blood brain barrier: possible therapeutic target The central nervous system has a distinct and unique relationship with the immune system. This distinction is primarily due to the BBB. The BBB prevents the infiltration of leukocytes into the CNS along with other soluble mediators. In MS, one of the many early signs of the disease is the increased permeability of the barrier as well as the decreased integrity of the BBB (20). This breakdown of the barrier can occur in two ways. The first possibility is the increased perivascular leakage of soluble mediators into the CNS via the destruction of tight junctions. This degradation of tight junctions is due to the decrease in claudin-5 proteins, which are integral in maintaining tight junction strength (21). The second is the increased transcellular entry of inflammatory T cells across the brain endothelial cells via the upregulation of adhesion molecules. Specifically, the upregulation of VCAM-1 in endothelial cells was found in the autopsies of many MS patients. VCAM-1 is a cellular adhesion molecule that is important in the migration of T cells from the periphery into the CNS. The VCAM-1 and its respective ligand interaction is essential for the rolling, tethering, activation, arrest-crawling, and transmigration of leukocytes into the CNS (21). The CAMs (i.e., VCAM-1) are located on the BBBendothelial cells while the ligands are found on the leukocytes. In addition to the role of CAMs and their respective ligands, it has been found that extracellular matrix components play a critical role in the migration of leukocytes across the BBB (20). The leukocytes migrating across the BBB increase the barrier permeability and therefore allow for further leukocyte infiltration. It should also be noted that active T cells can migrate into the CNS regardless of their antigen specificity (20). One of the many therapeutic targets of MS is improving the integrity of the BBB. Some studies have shown that glucocorticoids have increased claudin-5 expression and therefore improved the strength of the barrier. It also has been shown that active forms of Vitamin D have enhanced the upregulation of claudin-5 while decreasing VCAM-1 expression. Additionally, the most common drug used to treat RRMS is natalizumab. This drug is a humanized monoclonal antibody for VLA-4 (21). VLA-4 is the integrin for VCAM-1 ligand and is found on leukocytes. By binding to this specific ligand, natalizumab

inhibits the interaction with the ligand on the endothelial cell and prevents leukocyte infiltration. Roughly 184â&#x20AC;&#x2030;900 MS patients have been treated with natalizumab worldwide and it has shown the greatest efficacy in treating relapsing and remitting MS amongst drug treatments. A randomized, double-blinded placebo-controlled Phase III trial was conducted to confirm the Food Drug Administration's approval of natalizumab for RRMS treatment. In this study, the natalizumab treatment group compared to the placebo group showed a 68% decrease in annualized relapse rate at 1 year and a 42% decrease of the risk of confirmed disability worsening at 2 years (22). The proportion of relapse-free patients was significantly higher in the natalizumab group compared to the control group at 1 year. Additionally, natalizumab decreased the mean number of new or enlarging T2 hyperintense MRI lesions over 2 years by 83%. The mean number of T1- hypointense MRI lesions over 2 years reduced by 76% in the treatment group compared to the placebo group (22).

Conclusion MS is a debilitating autoimmune disease resulting in neurological dysfunction that can rob a person of their independence. Three different stages of disease progression have been identified: relapsing-remitting MS, secondary progressive MS, and primary progressive MS, with the latter recognized as the most severe. The defining features associated with MS as discussed in this review, although not mutually exclusive, are leukocyte infiltration in the MS lesions, astrocytes and their role in the pathology, neurodegeneration and axonal damage, and the intermediately activated macrophages otherwise known as foamy macrophages. Although the exact etiology behind MS may not be fully understood, modification of the complex biological mechanisms involved in this disease could lead to new therapeutic possibilities. Future studies will be needed to continue addressing the mechanisms behind why individuals are developing MS.

Acknowledgments We would like to thank Brian Piper, PhD, for his feedback on earlier versions of this manuscript.

Disclosures We have no disclosure to address.

References 1.

Vogel D, Vereyken E, Glim J, Heijnen P, Moeton M, van der Valk P et al. Macrophages in inflammatory multiple sclerosis lesions have an intermediate activation status. J Neuroinflammation. 2013;10(35).

2. Confavreux C, Vukusic S Moreau T, Adeleine P. Relapses and progression of disability in multiple sclerosis. N Engl J Med. 2000; 343(20):1430â&#x20AC;&#x201C;1438. 3. Didonna A, Oksenberg J. Multiple sclerosis: perspectives in treating and pathogenesis. Brisbane, Australia: Codon Publications. 2017.

129

The Cellular Hallmarks of Multiple Sclerosis

4. Haines JD, Inglese M, Casaccia P. Axonal damage in multiple sclerosis. Mt Sinai J Med. 2011; 78(2):231–243. 5. Harris VK, Tuddenham JF, Sadiq SA. Biomarkers of multiple sclerosis: current findings. Degener Neurol Neuromuscul Dis. 2017;7:19–29 6. Foote AK, Blakemore WF. Inflammation stimulates remyelination in areas of chronic demyelination. Brain. 2005; 128(3):528–539. 7.

Sadovnick A, Ebers G, Wilson R, Paty D. Life expectancy in patients attending multiple sclerosis clinics. Neurology. 1992; 42(4):991–4.

8. Dau P, Petajan J, Johnson K, Panitch H, Bornstein M. Plasmapheresis in multiple sclerosis: preliminary findings. Neurology. 1980; 30(10):1023–8. 9. Montalban X, Hauser S, Kappos L, Arnold D, Bar-Or A, Comi G. Ocrelizumab versus placebo in primary progressive multiple sclerosis. N Engl J Med. 2017; 376(3): 209–220. 10. Amor S, Puentes F, Baker D, Van Der Valk P. Inflammation in neurodegenerative diseases. Immunol. 2010; 129(2):154– 169. 11. Trapp BD, Peterson J, Ransohoff RM, Rudick R, Mork S, Bo L. Axonal transection in the lesions of multiple sclerosis. N Engl J Med. 1998; 338(5):278–85. 12. Bielekova B, Kadom N, Fisher E, Jeffries N, Ohayon J, Richert N, et al. MRI as a marker for disease heterogeneity in multiple sclerosis. Neurology. 2005;65:1071–6 13. Frischer J, Bramow S, Dal-Bianco A, Lucchinetti C, Rauschka H, Schmidbauer M et al. The relation between inflammation and neurodegeneration in multiple sclerosis brains. Brain. 2009; 132(5):1175–1189. 14. Bitsch A, Schuchardt J, Bunkowski S, Kuhlmann T, Brück W. Acute axonal injury in multiple sclerosis: correlation with demyelination and inflammation. Brain. 2000; 123(6):1174– 1183. 15. Patten B, Hart A, Lovelace R. Multiple sclerosis associated with defects in neuromuscular transmission. J Neurol Neurosurg Psychiatry. 1972; 35(3):385–94. 16. Correale J, Farez M. The role of astrocytes in multiple sclerosis progression. Front Neurol. 2015; 6(180):1–12. 17. Martinez F, Gordon S. The M1 and M2 paradigm of macrophage activation: time for reassessment. F1000 Prime Reports. 2014; 6(13). 18. Luo C, Jian C, Liao Y, Huang Q, Wu Y, Liu X, et al. The role of microglia in multiple sclerosis. Neuropsychiatr Dis Treat. 2017; 13:1661–1667. 19. Rawji K, Yong V. The benefits and detriments of macrophages/microglia in models of multiple sclerosis. Clin Dev Immunol. 2013; 2013:1–13. 20. Larochelle C, Alavarez JI, Prat A. How do immune cells overcome the blood–brain barrier in multiple sclerosis. FEBS Letters. 2011; 585(23):3770–3780.

130

21. Shimizu F, Nishihara H, Kanda T. Blood-brain barrier dysfunction in immuno-mediated neurological diseases. Immuno Med. 2018; 41(3):120–128. 22. Brandstadter R, Katz Sand I. The use of natalizumab for multiple sclerosis. Neuropsychiatr Dis Treat. 2017;13:1691– 1702.

Scholarly Research In Progress • Vol. 3, November 2019

Community-Based Survey to Assess Prescription Affordability in Lackawanna County, Pennsylvania Laina Gagliardi1†, Amalie Kropp1*, Alice Thompson1†, Danielle LaPointe1†, Terrence Habiyaremye1†, Sneha Vaddadi1†, Katarina Smigoc1†, Elizabeth Stackhouse1†, Michael Belko1†, Jasmine Santos1†, and Jennifer Joyce, MD1

1 Geisinger Commonwealth School of Medicine, Scranton PA, 18509 *Master of Biomedical Sciences Program †Doctor of Medicine Program Correspondence: lgagliardi@som.geisinger.edu

Abstract Background: The United States has the highest cost of health care globally. Many individuals nationwide struggle to afford prescription medications, which significantly impacts patient health outcomes. The objective of this study was to understand community-level challenges and experiences with medication affordability in Lackawanna County, Pennsylvania. Methods: Medical students in the Primary Care Progress Chapter at Geisinger Commonwealth School of Medicine conducted a survey to assess prescription affordability in Lackawanna County. Self-administered surveys were collected at several community-based sites in Lackawanna County. Data were analyzed using descriptive statistics and qualitative responses were thematically coded. Results: Eighty-two participants completed surveys at six community locations. More than two-thirds of respondents reported any difficulty affording medication copays or getting prescriptions covered by insurance. Of these individuals, 68% reported not filling a prescription due to financial burden. Participants reported few instances of discussing finances with their health care provider, but a strong desire for their physician to consider medication costs before prescribing. Forty-two participants provided qualitative responses to survey questions. Respondents reported common barriers to affording prescription medication including: 1. Insufficient income; 2. Issues with insurance coverage; and 3. High cost of copays. Conclusion: Results show that prescription affordability is a significant issue in northeast Pennsylvania (NEPA) and highlight various challenges to affording lifesaving medications. Findings underscore the need for ongoing research in this field and revisions to the national health care system to address this issue.

Introduction The United States spends more on health care than any developed country globally. Despite health system reforms and restructuring over the past two decades, these sizeable difference in health care expenditures are primarily due to the high price of receiving clinical care (1). Prices of medical devices and brand-name prescription drugs in the United States are 2 to 3 times greater than in other countries (2). Patients in the United States can expect to pay anywhere from 5% to 198% more for prescription medication in comparison to other countries, including Australia, Canada, France, and the United Kingdom (3).

The high cost of prescription drugs in the United States has led to fewer Americans being able to afford their medications. A February 2019 poll from the Kaiser Family Foundation found that a quarter of U.S. adults report difficulty affording prescription drugs and 29% did not take medications as prescribed due to cost (4). Patients often employ strategies to offset the rising cost of prescriptions such as requesting a lower-cost drug from providers and not taking the medication as prescribed. The latter includes detrimental practices such as skipping doses, taking less than the prescribed dose, or delaying prescription filling (5). The rising cost of prescription medications in the United States has substantial implications for patient health and quality of life. Research has shown that cost-prohibitive medication adherence results in higher rates of hospitalization and poorer health outcomes, as patients are less likely to take lifesaving medications such as antihyperlipidemic agents, antidepressants, and antibiotics (6, 7). Low-income individuals are more likely to not take medication as prescribed to save money for basic needs such as food, housing, and transportation (8). Lackawanna County, Pennsylvania, is a medically underserved area in northeast Pennsylvania (NEPA) with a high rate of chronic disease burden. Almost half of adults in NEPA experienced health care affordability issues in the past year (9). A recent Community Health Needs Assessment reported that more than one fifth of Lackawanna County residents did not receive regular health care because of cost (copays, deductibles, or prescriptions). Although the report acknowledges that patients in the community struggle to afford medications, there is still little data on prescription affordability in this area (10). The aim of this exploratory study was to describe prescription affordability in Lackawanna County and highlight individual experiences with medication costs.

Materials and Methods Lackawanna County is located in northeast Pennsylvania (NEPA) and has a population of 211,960. The community is primarily white and has a poverty rate above than the national average. Lackawanna County has higher rates of diabetes, liver disease, mental health disorders, and some cancers in comparison to both state and national averages (11). Medical students in the Primary Care Progress (PCP) chapter at Geisinger Commonwealth School of Medicine (GCSOM) in Scranton, Pennsylvania, began the “Catalyst Project” to

131

Community-Based Survey to Assess Prescription Affordability in Lackawanna County

Table 2. Study sites and total surveys completed at each community location.

Table 1. Survey items

describe prescription affordability in Lackawanna County. The project is part of a national PCP initiative to address local community health needs through student leadership and stakeholder engagement (12). The PCP Leadership Team at GCSOM met with local stakeholders in the spring and fall of 2018 to learn clinical and local perspectives on prescription affordability to guide survey development. Stakeholder input highlighted three components of prescription affordability: 1. Health insurance status; 2. Prescription affordability experiences; and 3. Provider consideration of patient financial status. All three areas were incorporated into a brief eight-item survey (Table 1). Participants were also encouraged to provide open-ended responses to more broadly capture their personal experiences with prescription affordability. A convenience sampling strategy was employed at multiple locations in the greater Scranton area. Locations of survey administration and the quantity of surveys collected at each site are presented in Table 2. Data were collected between February 14 and April 26, 2019. Catalyst team members approached study participants at each site and asked them to participate in a short survey about their experiences with prescription affordability. Surveys were self-administered and completed on paper. Participants did not need assistance completing the survey. Data were entered into a central database and analyzed using Microsoft Excel. Central tendency and variability in participant experiences with prescription affordability were examined using descriptive statistics. Study procedures were approved by the Institutional Review Board of the Geisinger Commonwealth School of Medicine.

Results A total of 82 surveys were collected at the 6 community-based sites. Table 2 lists both the number and percent of surveys completed per site. As this survey was exploratory in nature, participant demographic characteristic data were not collected. Quantitative findings A total of 72 participants (91%) reported having health insurance. Seven (9%) did not have health insurance and 3 declined to answer. Of those who reported having health insurance, 32% (n=23) were enrolled in Geisinger Health Plan 132

(GHP). Fifty one percent of participants (n=37) had an alternate form of insurance coverage. Twelve participants declined to report their type of health insurance. In comparison to non-GHP respondents, participants enrolled in the GHP reported less overall difficulty affording copays and fewer instances of failing to fill a prescription due to financial burden. More than two-thirds of respondents (n=28, 36%) reported that either they or a family member had ever had difficulty paying for copays or getting prescriptions covered by insurance. Of those that experienced any challenges with prescription affordability, 68% (n=19) reported not filling a prescription because it was too expensive. Additionally, 43% (n=15) of participants that reported any difficulties with prescription affordability indicated ever having to choose between paying for medication and basic needs, such as groceries or utility bills. This was significantly greater than the 19% of total respondents reporting ever choosing between medication and basic needs. Less than one-quarter of participants indicated that their physician had ever discussed payment issues during a clinical exam; however, 75% of respondents wanted their doctor to consider their financial situation before ordering prescriptions, laboratory tests, or procedures. Qualitative findings Forty-two participants (51%) provided a qualitative response to one or more survey questions. Responses were coded and analyzed for themes. Three primary themes emerged as barriers to affording prescription medications: 1. Insufficient income; 2. Issues with insurance coverage; and 3. High cost of copays. Insufficient Income Twenty participants (48%) indicated that income was a barrier to affording medications. Participants noted that at times they were unable to afford the high cost of lifesaving prescriptions such as inhalers, insulin, and antidepressants. Despite having a job and health insurance, some participants did not have enough money to afford both basic needs (e.g., food, transportation, housing, utilities) and medication costs. One participant that expressed difficulty paying for insurance indicated that at times he did not have an extra $5 for prescription medications. Other participants echoed similar issues.

Community-Based Survey to Assess Prescription Affordability in Lackawanna County

Last month I had to buy food and didn’t have the $20 for prescriptions. This month is the same. My husband and I are working professionals. Our insurance is $195 per week and we are in collections for hospital bills that we pay monthly, but the hospital still wants more. To cope with the high cost of prescription medications, some participants resorted to pill rationing or not filling prescriptions. For example, one patient with trigeminal neuralgia had to break pills in half or go without medication because it was too costly. Even when participants did not themselves face issues with prescription affordability, they knew of others who had trouble paying for medications. My neighbor rations his insulin. A coworker diagnosed her own UTI online and treated it with cat antibiotics. Issues with Insurance Coverage Fourteen participants providing qualitative responses (33%) expressed that insurance coverage was a barrier to affording prescription medications. Participants noted a disconnect between medications prescribed by their physician and those covered by their insurance. After I was hospitalized for a suicide attempt in 2012, I was prescribed an antidepressant not covered by my insurance, which cost $350 per month. I paid for it once, but then couldn't afford it again. When my wife was pregnant, she had terrible morning sickness. The insurance company denied coverage of medicine prescribed by her doctor. Participants also noted that this disconnect contributes to “surprise” medical bills and unanticipated costs. One participant said, “Anytime I use health care services, I'm afraid I will end up with a surprise bill.” Another commented on the “mystery” of costs for prescription medication or clinical tests. High Cost of Copays One fifth of participants providing qualitative data (n=9) indicated that the high cost of copays was an additional barrier to affording prescription medications. Patients noted that even with insurance coverage they could not afford copays for essential, lifesaving medications. For example, one participant had a $400 copay for eyedrops and another said that copays for insulin were continuously increasing.

Discussion

that 36% of participants experienced health care affordability issues. This is higher than national data reporting that 24% of adults have challenges affording prescription medication (13). These findings, however, are lower than the reported figure that 48% of adults in NEPA experience challenges affording health care (8). This can be partly explained by our study’s small sample size and that we only included one NEPA county. Qualitative findings support national research that patients cannot afford prescription medication due to rising costs and insurance issues (4). A strength of this study is that it utilized stakeholder engagement throughout the survey development process to ensure that local priorities were addressed. An additional strength is the wealth of qualitative data gathered that highlights participant stories and emphasizes the specific challenges NEPA residents have in affording prescription medication. To our knowledge, previous research in NEPA has focused more on identifying the problem of prescription affordability. This study helps fill this knowledge gap by highlighting multiple patient experiences and adding additional depth to the numerical data. This study is also strengthened by its assessment of the physician-patient relationship around prescription affordability. Research has shown that physicians are often unaware of the costs associated with prescription drugs (14). Given that only a quarter of participants in our sample discussed medication pricing with their physician, this represents an area where health care providers can intervene directly in the exam room to improve prescription affordability. Study limitations included a small sample size using a convivence sampling methodology, which both limit study generalizability to the overall NEPA community. As this study was exploratory in nature, the research team did not collect information on participant demographics, number of participants approached, location of survey completion and specific type of health insurance (e.g., Medicare, Medicaid, commercial). This decision further limits study generalizability and the ability to assess participant response bias. Future research should include these items to improve methodological rigor and conduct subgroup analyses to determine whether prescription affordability in NEPA differs by factors such as race, income, and insurance type. Despite the many strengths of the survey, future research should also seek out validated instruments on prescription affordability such as the Living with Medicines Questionnaire or the Treatment Burden Questionnaire to supplement stakeholder input (15, 16). Additionally, the survey was not pilot tested with community members; therefore, respondents could have potentially misinterpreted questions or had difficulty comprehending survey items. For example, given the current insurance environment, it is possible that patients did not understand the difference between copays and deductibles in Question 2, which could have impacted survey responses. Future research should include a pilot test phase into the survey development and administration process to ensure data fidelity.

This exploratory, community-based study found that a convenience sample of NEPA residents reported difficulty affording prescription medications, despite a high rate of insured participants. Participants noted issues both with insurance coverage and balancing basic needs with the rising cost of prescription drugs. Qualitative data showed the various challenges participants faced to afford prescription medications including income barriers, insufficient insurance coverage, and the high cost of copays. However, these challenges were rarely discussed with health care providers.

Conclusion

Results of this study were somewhat aligned with national and local data on prescription affordability. Our study found

This study underscores the importance of addressing the complex issue of prescription affordability at the local, state,

133

Community-Based Survey to Assess Prescription Affordability in Lackawanna County

and national levels. Some participants in our study mentioned U.S. health care reform to a single-payer or universal system modeled after Canada and European countries. For example, one respondent said, “The markup on meds is borderline with robbery… like in so many other countries everyone should be able to receive health care and medication, regardless of their ability to pay.” Results of this study can be used to supplement and build on the growing body of literature describing prescription affordability in the U.S. health care system. Future health policies should support the majority of Americans in favor of expanded public health coverage and lowered prescription costs (17). As the United States continues to debate and revise its health care system, we must ensure that patients can afford prescription medications to both reduce health care expenditures and improve individual quality of life.

Acknowledgments Authors contributed in the following roles: LG, AK, TH, DL, KS, and SV participated in the completion and support of the manuscript production. LG and AK edited and revised the final version of the manuscript. The overall Catalyst project planning was done by AT and JS. Coordination, research development, and management were done by AT. Materials and methods development was finalized by JS. MB and ES accomplished the IRB approval and completion. Data and survey collection was done by AK, AT, TH, DL, KS, SV, ES, and MB. Data analysis was completed by LG, MB, ES, and TH. JJ was our faculty advisor for the project. We gratefully acknowledge Primary Care Progress for their guidance in implementing the Catalyst Project. We also thank the many community organizations that partnered with us on this project, including The Wright Center, United Neighborhood Center, Hope Clinic, Albright Public Library, Library Express at the Steamtown Mall, and the Southside Farmers Market. Additionally, we thank Grace McGrath and Shubhra Shetty, MD, for their partnership and tireless efforts to forward the goals and vision of Primary Care Progress.

References 1.

Anderson P, Hussey P, Petrosyvan V. It's still the prices, stupid: why the US spends so much on health care and a tribute to UWE Reinhardt. 2019. Health Affairs. 38(1):87–95.

2. Papanicolas I, Woskie LR, Jha AK. Health care spending in the United States and other high-income countries. 2018. JAMA. 319(10):1024–1039. 3. Kanavos P, Ferrario A, Vandoros S, Anderson GF. Higher US branded drug prices and spending compared to other countries may stem partly from quick uptake of new drugs. 2013. Health Affairs. 32(4):753–61. 4. Kirzinger A, Lopes L, Wu B, Brodie M. KFF Health Tracking Poll–February 2019: Prescription drugs [Internet]. Kaiser Family Foundation; 2019 [cited 2019 July 1]. Available from: https://www.kff.org/health-costs/poll-finding/kff-healthtracking-poll-february-2019-prescription-drugs/ 5. Cohen RA and Villaroel MA. Strategies used by adults to reduce their prescription drug costs: United States, 2013 [Internet]. Centers for Disease Control; 2015 [cited 2019 July 1]. Available from https://www.cdc.gov/nchs/products/ databriefs/db184.htm 134

6. Kennedy J, Tuleu I, Mackay K. Unfilled prescriptions of Medicare beneficiaries: prevalence, reasons and types of medicines prescribed. 2008. Journal of Managed Care Pharmacy. 14(6): 553–60. 7.

Mojtabai R, Olfson M. Medication costs, adherence and health outcomes among Medicare beneficiaries. 2003. Health Affairs. 22(4):220–9.

8. Berkowitz SA, Seligman HK, Choudhry. Treat or eat: food insecurity, cost-related medication underuse and unmet needs. 2014. The American Journal of Medicine. 127(4): 303–310. 9. Healthcare Value Hub. Northeast/North Central Pennsylvania: 48% of adults experienced health care affordability burdens in the past year [Internet]. Washington, DC: Altarum; 2018 [cited 2019 May 16]. 2 p. Report No.: 16. Available from: https://www.healthcarevaluehub.org/ files/5015/3936/0008/Hub-Altarum_Data_Brief_No._16_-_ PA_Northeast_and_North_Central_Region.pdf 10. Geisinger Community Medical Center. Community health needs assessment July 1, 2018–June 30, 2021 [Internet]. Scranton, PA: Geisinger Health; 2018 [cited 2019 May 16]. 115 p. Available from: https://www.geisinger.org/-/media/ OneGeisinger/pdfs/ghs/about-geisinger/chna/2018-reports/ gcmc-chna-2018.pdf?la=en 11. Institute for Health Metrics and Evaluation (IHME), US County Profile: Lackawanna County, Pennsylvania [Internet]. Seattle, WA: IHME, 2016 [cited July 1, 2019] Available from: http://www.healthdata.org/sites/default/files/files/county_ profiles/US/2015/County_Report_Lackawanna_County_ Pennsylvania.pdf 12. Primary Care Progress. Our programs [Internet]. Cambridge, MA: Primary Care Progress; [cited 2019 May 16]. Available from: https://www.primarycareprogress.org/ our-programs/ 13. Kamal R, Cox Cynthia, McDermott D. What are the recent and forecasted trends in prescription drug spending? [Internet]. Peterson-Kaiser; 2019 [cited 2019 May 16]. Available from: https://www.healthsystemtracker.org/chartcollection/recent-forecasted-trends-prescription-drugspending/#item-start 14. Schutte T, Tichelaar J, Nanayakkara P, Richir M, vab Agtmael M. Students and doctors are unaware of the cost of drugs they frequently prescribe. 2017. Basic Clinical Pharmacological Toxicology. 120(3): 278–283. 15. Krska J, Katusiime B, Cortlett SA. Validation of an instrument to measure patients’ experiences of medicine use: the living with medicines questionnaire. 2017. Patient Preference and Adherence. 11:671–679. 16. Tran V, Harrington M, Montori VM, Barnes A, Wicks P, Ravaud P. Adaptation and validation of the Treatment Burden Questionnaire (TBQ) in English using an internet platform. 2014. BMC Medicine. 12:109. 17. Kirzinger A, Munana C, Brodie M. KFF health tracking poll–January 2019: the public on next steps for the aca and proposals to expand coverage [Internet]. Kaiser Family Foundation; 2019 [cited 2019 July 1]. Available from: https:// www.kff.org/health-reform/poll-finding/kff-health-trackingpoll-january-2019/

Scholarly Research In Progress • Vol. 3, November 2019

Medication Prescribing Patterns in Patients with Hidradenitis Suppurativa: A Population-Based Study Andrea J. Borba1†‡ and Pierce H. Deng2‡

Geisinger Commonwealth School of Medicine, Scranton, PA 18509 Keck School of Medicine of University of Southern California, Los Angeles, CA 90033 †Doctor of Medicine Program ‡Authors contributed equally Correspondence: andieborba@gmail.com 1

Abstract Background: Hidradenitis suppurativa (HS) is a chronic inflammatory skin disease affecting up to 4 percent of the population. Despite the historical lack of treatment options and the recent availability of new systemic treatments for patients with HS, little is known about the recent treatment prescribing patterns for this population. Our objective was to determine the recent medication prescribing patterns for patients with HS. Methods: We conducted a retrospective, population-based analysis using the National Ambulatory Medical Care Survey and the National Hospital Ambulatory Care Survey from the years of 2012 to 2016. Results: An estimated total of 1 080 942 visits were made by HS patients between 2012 and 2016 in the United States. Visits to physicians in the United States for HS are increasing by approximately 100 000 visits per year. Antibiotics were the most commonly prescribed medication class at visits for HS, followed by pain relievers and topical corticosteroids. The most commonly prescribed antibiotics included clindamycin, sulfamethoxazole/trimethoprim, and doxycycline. No prescriptions for biologics or systemic therapies were identified in the top medications prescribed at visits for HS. Conclusions: Antibiotics remain the most commonly prescribed medications at visits for HS despite the recent availability of more targeted treatment. Future research is needed to determine whether clinicians utilize updated therapies in order to improve patient outcomes.

Introduction Hidradenitis suppurativa (HS) is a chronic inflammatory skin disease that involves the occlusion and rupture of hair follicles in the apocrine gland-bearing areas of the body, including the axillary, inframammary, and inguinal regions. Clinically, HS presents with recurrent abscesses, painful nodules, draining sinus tracts with mucopurulent discharge, and subsequent scarring (1, 2). The prevalence of HS has been previously estimated to range from less than 1 percent to 4 percent (3–5). However, due to the involvement of sensitive regions in the body and potentially embarrassing manifestations of the disease, it is likely that HS is underreported among patients. In any case, HS is not an uncommon disease and it disproportionally affects young adults, females, and African American patients (3). Furthermore, HS is known to be associated with many comorbidities and complications, including recurrent infections

and development of squamous cell carcinoma (2,5). The condition also has a profoundly negative impact on quality of life and psychosocial well-being (6,7). While the overall burden of HS is significant, the treatment options are currently limited and often only partially effective. Topical clindamycin has been proposed as a first-line option in mild localized disease. For patients with moderate to severe HS, systemic antibiotics such as oral doxycycline is often recommended as initial therapy. Alternatively, oral retinoids (such as acitretin) and anti-androgen therapy (such as spironolactone) may be helpful when patients are not responsive to oral tetracyclines. More recently, biologic treatments such as tumor necrosis factor-alpha (TNF-α) inhibitors have been recommended as next-line for severe refractory HS (5). Finally, surgical intervention is another option, which may include punch debridement, deroofing, or wide excision (2, 5). Since there have been considerable scientific developments in HS therapies over the past several years, a large gap exists in understanding the current treatment patterns that physicians and patients are actually using to manage this complex condition on a nationwide level. Davis et al. had previously looked at how physicians in the U.S. managed HS using data from the 1990–2009 National Ambulatory Medical Care Survey (NAMCS), an annual survey conducted by the Centers for Disease Control and Prevention (CDC) (8). We aimed to provide an update to how HS is currently managed in the U.S. from the years 2012 through 2016.

Methods We conducted our analysis of visit trends and clinician prescribing patterns for HS using the National Ambulatory Care Survey (NAMCS) and the National Hospital Ambulatory Care Survey (NHAMCS) from 2012 through 2016. The NAMCS and NHAMCS are organized annually by the National Center for Health Statistics (NCHS) at the Centers for Disease Control and Prevention (CDC). These surveys use data from the medical records of non-federally employed, office-based physician visits (9). The NAMCS is administered to physicians in 25 distinct sampling regions: 16 of the most populous states in the U.S., varied by year, with the remainder of the states divided into 9 regions according to their U.S. Census Bureau divisions (10). The basic sampling unit is the patient visit. A stratified two-stage sampling method is used to collect information from the medical records of individual patient visits (Figure 1). The NHAMCS uses a similar approach, but collects data from visits to emergency departments. Both surveys

135

Medication Prescribing Patterns in Patients with Hidradenitis Suppurativa

Figure 2. Time trend of estimated total number of visits per year with a diagnosis of HS in the U.S. Error bars indicate 95% CI.

Table 1. Top medications prescribed at visits with a diagnosis of HS.*

Our population of interest was children and adults with HS in the U.S. We used the International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) code 705.83, defined as “hidradenitis,” and the International Classification of Diseases, 10th Revision, Clinical Modification (ICD-10) code L73.2, defined as “hidradenitis suppurativa,” to isolate HS. As the NAMCS and NHAMCS allow clinicians to record multiple diagnoses for a single visit, we also identified the population with a sole diagnosis of HS during this five-year period. Medications were identified using the various drug mention codes within the NAMCS and NHAMCS as well as the Multum drug therapeutic class (12). All analyses were carried out using Stata Version 13 (Stata Corp, College Station, Texas).

Results Figure 1. Sampling process used by NAMCS.

allow options to report multiple diagnoses and prescriptions within a single visit. It is important to note that the NAMCS and NHAMCS record characteristics of individual patient visits, and therefore should not be interpreted as individual patients. The NAMCS and NHAMCS contain weighted variables that allow for extrapolation of results to a population-wide estimate (11).

136

Using the application of weighted variables, our study showed that an estimated total of 1 080 942 visits were made by children and adults with a diagnosis of HS between 2012 and 2016 in the U.S. Of these visits, 486 317 (CI: 293 371–679 262) were made by children and adults with a sole diagnosis of HS. The number of visits for HS increased by an estimated approximation of 100 000 visits per year during this time period (Figure 2). The leading individual medications prescribed at visits without a sole diagnosis of HS are shown in Table 1. Antibiotics

Medication Prescribing Patterns in Patients with Hidradenitis Suppurativa

composed the majority of these medications, with clindamycin and sulfamethoxazole/trimethoprim prescribed at over 50% of these visits. Antibiotics were the most commonly prescribed individual medications at visits made by patients with a sole diagnosis of HS (Table 2). Clindamycin and doxycycline composed the top two medications prescribed at these visits. Similarly, antibiotics composed the highest percentage of total medications by class prescribed at visits with a sole diagnosis of HS (Figure 3) and at visits without a sole diagnosis of HS (Figure 4). By medication class, more topical steroids but fewer pain relievers were prescribed at visits for HS with a sole diagnosis compared to HS visits without a sole diagnosis. Notably, systemic therapies with the exception of antibiotics and pain relievers were not observed in the top prescribed medications at visits with or without a sole diagnosis of HS.

Table 2. Top medications prescribed at visits with a sole diagnosis of HS.*

Discussion The results from our population-wide study suggest that antibiotics are the most commonly prescribed medication class at visits for HS, followed by pain relievers and topical corticosteroids. Clindamycin and doxycycline are the top two most commonly prescribed medications for visits with a diagnosis of HS. HS is a complex disease that can be difficult to treat adequately. The number of available interventions is increasing, but there still is no uniform consensus among physicians for how to best manage HS. Previously, from the years of 1990 to 2009, Davis et al. found that HS was treated with a medication in 74% of ambulatory visits, while a procedure was performed in 11% (8). Antibiotics were by far the most commonly prescribed medicine, followed by pain medications and topical steroids. There were no records of patients treated with acitretin or TNF-α inhibitors such as adalimumab or infliximab in the 1990 to 2009 NAMCS. Isotretinoin was prescribed in 3.8% of HS visits, but current evidence suggests that this medication is ineffective for HS (13). Aside from antibiotics and pain medications, our results indicate that there continues to be a lack of systemic therapies being prescribed for HS patients in the United States. This is despite notable progress in our understanding of HS therapies since 2009. Perhaps most importantly, in 2015, adalimumab received the first and only U.S. Food and Drug Administration (FDA) approval for moderate-to-severe HS following the pivotal PIONEER trials (14). Studies from this decade have also shown promising results for the use of acitretin, as well as antiandrogen therapy with spironolactone (15–18). A recent randomized controlled trial in the Netherlands showed promising results for the use of apremilast in treating HS (19). Currently, there are additional biologic therapies, including IL-1 and IL-17 inhibitors, that are being investigated that have potential to be an effective treatment for HS (20–22). Additionally, we found that visits for HS are increasing in the United States. This phenomenon could be explained by increased public awareness of the condition due to widespread availability of health information through social media and health sectors, in addition to an improvement in access to care (23). It is also possible that there is a true worsening of the HS disease epidemic in the United States (23).

Figure 3. Percentage of leading medication classes from all medications prescribed at visits with a diagnosis of HS.*

Figure 4. Percentage of leading medication classes from all medications prescribed at visits with a sole diagnosis of HS.

It is possible that the NAMCS datasets from 2012 through 2016 do not yet reflect the newer developments in clinical practice. Physicians may also not be aware of the advancements in HS research, and might not be familiar with prescribing systemic drugs to treat HS. Therefore, it could be beneficial for physicians to refer more complex HS patients to a dermatologist with experience in HS in order to optimize these patients’ health outcomes. Increased use of systemic treatments and biologics in HS management may lead to better patient outcomes and improved patient satisfaction.

137

Medication Prescribing Patterns in Patients with Hidradenitis Suppurativa

Our study needs to be interpreted in the context of the study design. The notable strength of this study lies in its utilization of large validated databases that contain nationally representative samples of U.S. health care visits. These CDC surveys have markedly high response rates, which further help to minimize selection bias when compared to other dermatologic surveys (24). A limitation of the study is that 2016 is the most recent year that the NAMCS dataset has been made publicly available by the CDC. For example, since adalimumab was FDA approved in 2015, it may be that the usage of the drug has been increasing but we are not yet able to observe the prescription rates in 2017 and beyond. Additionally, as in any large database study, misclassification of ICD diagnostic codes is a potential limitation.

11. NAMCS/NHAMCS-Estimation Procedures [Internet]. 2019 [cited 2019 Apr 24]. Available from: https://www.cdc.gov/ nchs/ahcd/ahcd_estimation_procedures.htm

Altogether, there continues to be high demand for further research on HS treatment efficacy through randomized controlled trials. HS is often a particularly debilitating condition and there is a vital need for more data and understanding of new treatments to improve patient outcomes.

15. Golbari NM, Porter ML, Kimball AB. Antiandrogen therapy with spironolactone for the treatment of hidradenitis suppurativa. J Am Acad Dermatol. 2019 Jan;80(1):114–9.

References 1.

Jemec GBE. Hidradenitis Suppurativa. New England Journal of Medicine. 2012 Jan 12;366(2):158–64.

12. NAMCS/NHAMCS-Ambulatory Care Drug Database System-Search [Internet]. [cited 2019 May 17]. Available from: https://www2.cdc.gov/drugs/applicationnav1.asp 13. Boer J, van Gemert MJ. Long-term results of isotretinoin in the treatment of 68 patients with hidradenitis suppurativa. J Am Acad Dermatol. 1999 Jan;40(1):73–6. 14. Kimball AB, Okun MM, Williams DA, Gottlieb AB, Papp KA, Zouboulis CC, et al. Two Phase 3 Trials of Adalimumab for Hidradenitis Suppurativa. New England Journal of Medicine. 2016 Aug 4;375(5):422–34.

16. Lee A, Fischer G. A case series of 20 women with hidradenitis suppurativa treated with spironolactone. Australas J Dermatol. 2015 Aug;56(3):192–6. 17. Boer J, Nazary M. Long-term results of acitretin therapy for hidradenitis suppurativa. Is acne inversa also a misnomer? Br J Dermatol. 2011 Jan;164(1):170–5.

2. Alikhan A, Lynch PJ, Eisen DB. Hidradenitis suppurativa: A comprehensive review. Journal of the American Academy of Dermatology. 2009 Apr;60(4):539–61.

18. Matusiak L, Bieniek A, Szepietowski JC. Acitretin treatment for hidradenitis suppurativa: a prospective series of 17 patients. Br J Dermatol. 2014 Jul;171(1):170–4.

3. Garg A, Kirby JS, Lavian J, Lin G, Strunk A. Sex- and AgeAdjusted Population Analysis of Prevalence Estimates for Hidradenitis Suppurativa in the United States. JAMA Dermatol. 2017 Aug 1;153(8):760–4.

19. Vossen ARJV, van Doorn MBA, van der Zee HH, Prens EP. Apremilast for moderate hidradenitis suppurativa: Results of a randomized controlled trial. Journal of the American Academy of Dermatology. 2019 Jan 1;80(1):80–8.

4. Jemec GB, Heidenheim M, Nielsen NH. The prevalence of hidradenitis suppurativa and its potential precursor lesions. J Am Acad Dermatol. 1996 Aug;35(2 Pt 1):191–4.

20. Riis PT, Thorlacius LR, Jemec GB. Investigational drugs in clinical trials for Hidradenitis Suppurativa. Expert Opinion on Investigational Drugs. 2018 Jan 2;27(1):43–53.

5. Zouboulis CC, Desai N, Emtestam L, Hunger RE, Ioannides D, Juhász I, et al. European S1 guideline for the treatment of hidradenitis suppurativa/acne inversa. J Eur Acad Dermatol Venereol. 2015 Apr;29(4):619–44.

21. Tzanetakou V, Kanni T, Giatrakou S, Katoulis A, Papadavid E, Netea MG, et al. Safety and Efficacy of Anakinra in Severe Hidradenitis Suppurativa: A Randomized Clinical Trial. JAMA Dermatol. 2016 Jan;152(1):52–9.

6. Onderdijk AJ, van der Zee HH, Esmann S, Lophaven S, Dufour DN, Jemec GBE, et al. Depression in patients with hidradenitis suppurativa. J Eur Acad Dermatol Venereol. 2013 Apr;27(4):473–8.

22. Matusiak Ł, Jemec GB, Szepietowski JC. Pharmacological development in hidradenitis suppurativa. Current Opinion in Pharmacology. 2019 Jun 1;46:65–72.

Kouris A, Platsidaki E, Christodoulou C, Efstathiou V, Dessinioti C, Tzanetakou V, et al. Quality of Life and Psychosocial Implications in Patients with Hidradenitis Suppurativa. Dermatology (Basel). 2016;232(6):687–91.

8. Davis SA, Lin H-C, Balkrishnan R, Feldman SR. Hidradenitis Suppurativa Management in the United States: An Analysis of the National Ambulatory Medical Care Survey and MarketScan Medicaid Databases. Skin Appendage Disorders. 2015;1(2):65–73. 9. NAMCS/NHAMCS-Scope and Sample Design [Internet]. 2019 [cited 2019 Apr 5]. Available from: https://www.cdc. gov/nchs/ahcd/ahcd_scope.htm 10. 2014 National Ambulatory Medical Care Survey Public Use Micro-Data File Documentation. Available from: http://www. nber.org/namcs/docs/namcs2014.pdf

138

23. Garg A, Lavian J, Lin G, Strunk A, Alloo A. Incidence of hidradenitis suppurativa in the United States: A sex- and age-adjusted population analysis. J Am Acad Dermatol. 2017 Jul;77(1):118–22. 24. Ahn CS, Allen M-M, Davis SA, Huang KE, Jr ABF, Feldman SR. The National Ambulatory Medical Care Survey: A resource for understanding the outpatient dermatology treatment. Journal of Dermatological Treatment. 2014 Dec 1;25(6):453–8.

Scholarly Research In Progress • Vol. 3, November 2019

Assessment Using the Sorour Airway Visualization Evaluation (SAVE) Score Predicts Difficulty of Tracheal Intubation via an Inter-Observer Reliability Study Garrison Davis1,2†, Connor Magura1†, Conor Lynch1†, Stephanie Tilberry1†, Sahil Pandya1†, Julia Shamis1†, Sharmeen Mian1†, and Khaled Sorour2,3 Geisinger Commonwealth School of Medicine, Scranton, PA 18509 Signature Healthcare Brockton Hospital, Brockton, MA 02302 3 Harvard Medical School, Boston, MA 02115 †Doctor of Medicine Program Correspondence: gdavis01@som.geisinger.edu 1

Abstract Background: The Sorour Airway Visualization Evaluation (SAVE) score is a novel, objective method to efficiently and accurately predict the difficulty of intubating a patient. The score is an integrated calculation that combines various factors that influence the difficulty of intubation and can be calculated in fewer than 90 seconds. Having previously shown the value and efficacy of this tool for airway assessment, this inter-observer reliability study was conducted to assess both the external validity and user bias of previous studies. Methods: During the course of this study, attendings, residents, respiratory therapists and medical students in the ICU (intensive care unit) at Signature Healthcare Brockton Hospital calculated SAVE scores on 14 patients presenting with various conditions. The score was not used to change the clinical decision-making for the method of intubation. Statistical analyses were retrospectively completed to compare scores for group variance. Results: SAVE scores calculated by respiratory therapists had 77.8% congruence with scores calculated by the attending physician while residents had a 71.4% congruence (p<0.001). Medical students, however, only had a 66.7% agreement with scores calculated by the attending physician (p<0.001). Phi coefficient was 1.0 for all three groups. Conclusion: This data shows promise regarding the utility of the SAVE score to predict the difficulty of intubation when assessed by different health care providers. Further training may be implemented to ensure that all members of the health care team are able to accurately predict scores with limited heterogeneity. Despite variation in scores, there was complete agreement (Phi coefficient=1) regarding whether a patient was easy or difficult. With adequate training, the SAVE score has the potential to be a useful tool for all members of the health care team to assist in patient intubation preparations and improve outcomes.

Introduction Obtaining a difficult airway can be a critical challenge for even the most accomplished anesthetist. Serious negative outcomes including brain damage or death can occur if there is failure to maintain airway patency for more than a few minutes. Difficult airways may be defined as those requiring three attempts or more by a trained anesthetist or more than 10 minutes to complete using conventional laryngoscope

tracheal intubation. Difficult airways are relatively rare, possibly occurring in only 1.8% of intubations (1). Despite their low prevalence, unanticipated and poorly managed situations may lead to life-threatening complications or death (2). Traditional teaching enlists a group of 10 to 12 signs and historic data points supported by separate studies with varying degrees of statistical and clinical rigor. Numerous patient characteristics have been implicated with difficult intubation, including: obesity (1, 3, 4), limited head and neck mobility (5–8), presence of long upper incisors (2, 3, 8), high Mallampati score (1, 3, 5–9), a short neck (4), thyromental distance (1, 5–7, 9), neck circumference (10) and retrognathia (overbite) (11). The Mallampati score is a measure of endotracheal intubation difficulty determined by visually assessing the space between the base of the tongue and the roof of the mouth while the patient’s tongue is protruded. The score is calculated based on how many of certain anatomical structures can be visualized. The Mallampati score has a range of 1 to 4, with a higher score indicating less space between the tongue base and mouth roof. Any combinations of these assessments are possible, yet there are no strict guidelines regarding how to put these assessments together or interpret any one combination. Furthermore, the complexity of integrating various scores and implementing them with critical patient presentations in a timely manner can lead to even more complications. Airway predictive tools need to be near-universal, simple to implement, and should demonstrate ease of use for all medical team personnel. Prior attempts to create conglomerate scores include the Wilson risk sum score (2) and the LEMON method (12), which suffered from poor specificity and positive predictive value (13–15). These methods have not been widely adopted due to fatal issues. The SAVE score was developed as a method for predicting difficulty of intubation. This assessment system incorporates many of the previously described characteristics into an effective, easy-to-use model. Previous data has shown the effectiveness of the SAVE score in both the ICU and the operating room (OR). This study aimed to assess the ability of various medical personnel to evaluate a patient's SAVE score in order to evaluate its user bias and external validity as an airway assessment tool. The capability of medical team members with varying degrees of education to appropriately utilize the SAVE score further displays its usefulness and unique contribution to the current lack of difficult intubation prediction guidelines. 139

Assessment Using the Sorour Airway Visualization Evaluation (SAVE) Score

Materials and Methods Procedures During the course of this study, medical professionals from several different backgrounds were recruited to assess the SAVE score of a variety of patients in the Intensive Care Unit (ICU) at Signature Healthcare Brockton Hospital. These participants included attendings, residents, respiratory therapists, and medical students. During each patient encounter assessed in this study, each of the participants calculated the SAVE score to the best of their ability using the score calculator shown in Figure 1. These scores were recorded and analyzed retrospectively at a later time. The SAVE scores were not used to influence clinical decisionmaking. A total of 14 patients with various clinical presentations had their SAVE scores calculated and recorded.

Figure 1. SAVE pre-intubation airway assessment method

Data analysis Statistical analyses were retrospectively completed to compare the scores for group variance. Group data was further stratified by the classification of an intubation as “easy” or “difficult.” These evaluations were based on airway visualization, as shown in Figure 2. An “easy” intubation was considered grade 1 or 2, whereas a “difficult” intubation was considered grade 3 or 4. All data was stored using Microsoft Excel (2016) and analyzed using R Statistical Software version 3.3.3 (Foundation for Statistical Computing, Vienna, Austria). A phi coefficient was calculated to determine if clinicians agreed that patients were difficult or easy intubations. A p-value of less than 0.05 was considered statistically significant.

Figure 2. SAVE post-intubation airway assessment method

Results Analysis of inter-observer reliability of SAVE scores between medical professionals was conducted for 14 total patients. Three of those patients were classified as difficult intubations and 11 were considered easy intubations. Analyzed scores were calculated by one attending physician, seven residents, three respiratory therapists and two medical students (Table 1). SAVE scores calculated by residents had 71.4% agreement with scores calculated by the attending physician (φ=1.0, p<0.001, n=14). SAVE scores calculated by respiratory therapists had 77.8% agreement with scores calculated by the attending physician (φ= 1.0, p<0.001, n=9). SAVE scores calculated by medical students had 66.7% agreement with scores calculated by the attending physician (φ= 1.0, p<0.001, n=9) (Table 2).

Table 1. Summary of patient and clinician participants included in interobserver reliability analysis

Discussion This is the first study attempting to validate the SAVE score airway assessment as a homogenous means to predict difficult intubation across different groups of health care providers, an area in which other prediction tools have fallen short (13). Utilization of such a tool may lead to significant improvement in patient outcomes through effective, rapid dissemination of clinical information to all medical personnel (2). Further, application of this tool may help to develop a standardized approach for patients who have a greater potential to present with a difficult airway. Although such presentations are relatively rare (1), their high potential for morbidity and mortality

140

Table 2. Reliability analysis for SAVE score between clinicians

if poorly managed necessitate a well-prepared and rapid response (2). In this study, the primary goal was to assess and compare the perceived difficulty of intubation across various health care providers. Consistency in evaluations made by different health care providers is important for predictive tools such as the SAVE score for use in collection and dissemination

Assessment Using the Sorour Airway Visualization Evaluation (SAVE) Score

of important clinical information (18). Assessments made using other airway prediction tools have had difficulty in terms of such interprofessional consistency (3, 13, 18). To remain consistent across various members of the care team while retaining clinical validity, evaluation criteria should be based on factors that can be described simply but can also be measured accurately (18.) Preliminary assessment of the data collected shows promise regarding the utility of the SAVE score to predict intubation difficulty. With physician reported difficulty as the gold standard, both residents and respiratory therapists were in concordance 71.4% and 77.8% of the time, respectively. These data provide invaluable insight regarding the reliability and utility of the score, irrespective of the health care professional performing the intubation. This shows that respiratory therapists and residents may, with further training, be able to match the scores of attending physicians. Additionally, given a larger sample size, the data may display much closer SAVE score values assessed by these different groups.

Acknowledgments We would like to thank the staff at Signature Healthcare Brockton Hospital Critical Care Unit for their assistance with data acquisition and protocol adaptation.

References 1.

Rose DK, Cohen MM. The airway: problems and predictions in 18,500 patients. Can J Anaesth. 1994;41:372–83.

2. Wilson ME, Spiegelhalter D, Robertson JA, Lesser P. Predicting difficult intubation. Br J Anaesth. 1988;61:211–6. 3. Oates JD, Macleod AD, Oates PD, Pearsall FJ, Howie JC, Murray GD. Comparison of two methods for predicting difficult intubation. Br J Anaesth. 1991 Dec;66:305–9. 4. Rocke DA, Murray WB, Rout CC, Gouws E. Relative risk analysis of factors associated with difficult intubation in obstetric anesthesia. Anesthesiology. 1992;77:67–73.

Of additional note was the medical student congruence score of 66.7%, despite complete agreement regarding the intubation difficulty. Upon adopting SAVE as a clinical decisionmaking tool, medical students may need significant training to be able to accurately assess a patient's SAVE score. It may even be considered that medical students not use their SAVE calculation as a clinical decision-making tool, but rather as a learning tool to increase their individual understanding of airway management.

5. Tse JC, Rimm EB, Hussain A. Predicting difficult endotracheal intubation in surgical patients scheduled for general anesthesia: a prospective blind study. Anesth Analg. 1995;81:254–8.

The clinical significance of this tool extends beyond the OR and the ICU. For instance, in acute events in the Emergency Department requiring intubation, the SAVE score allows health care providers to obtain a quick, reliable indication of intubation difficulty and disseminate this information to the care team. Moreover, in calculating the SAVE score as part of routine analysis prior to intubation, clinicians can determine whether additional medical personnel, such as anesthesiologists or respiratory therapists, will be required to assist with bedside intubation.

Although the SAVE score has shown validity and reliability in this study, this must be confirmed in future studies with a more robust sample size. Moreover, the data includes patients who were intubated in the relatively controlled environment of the ICU. Further data would need to be collected to determine whether the SAVE score can be applied in other emergent settings such as the Emergency Department or field intubation. Additionally, the data collected in this study were sampled from one hospital. Collecting data from other hospitals would allow for more applicable generalizations regarding the utility of the SAVE score as well. While there is certainly some variance among the different health care providers assessed in this study, data shows promise that these groups will be capable of assessing patients with the same accuracy as attending physicians. With adequate education and training, the SAVE score has the potential to serve as a tool that can be used by all members of the health care team to assist in intubation preparations and ultimately improve patient outcomes.

6. el-Ganzouri AR, McCarthy RJ, Tuman KJ, Tanck EN, Ivankovich AD. Preoperative airway assessment: predictive value of a multivariate risk index. Anesth Analg. 1996;82:1197–204. Arné J, Descoins P, Fusciardi J, Ingrand P, Ferrier B, Boudigues D, et al. Preoperative assessment for difficult intubation in general and ENT surgery: predictive value of a clinical multivariate risk index. Br J Anaesth. 1998;80:140–6.

8. Saghaei M, Safavi MR. Prediction of prolonged laryngoscopy. Anaesthesia. 2001;56:1198–201. 9. Frerk CM. Predicting difficult intubation. Anaesthesia. 1991 Mar;46:1005–8. 10. Riad W, Vaez MN, Raveendran R, Tam AD, Quereshy FA, Chung F, et al. Neck circumference as a predictor of difficult intubation and difficult mask ventilation in morbidly obese patients: A prospective observational study. Eur J Anaesthesiol. 2016; 33: 244–9. 11. Kheterpal S, Han R, Tremper KK, Shanks A, Tait AR, O'Reilly M, et al. Incidence and predictors of difficult and impossible mask ventilation. Anesthesiology. 2006;105:885–91. 12. Reed MJ, Rennie LM, Dunn MJG, Gray AJ, Robertson CE, McKeown DW. Is the ‘LEMON’ method an easily applied emergency airway assessment tool? Eur J Emerg Med. 2004;11:154–7. 13. Yentis SM. Predicting difficult intubation-worthwhile exercise or pointless ritual? Anaesthesia. 2002;57:105–9. 14. Ezri T, Medalion B, Weisenberg M, Szmuk P, Warters RD, Charuzi I. Increased body mass index per se is not a predictor of difficult laryngoscopy. Can J Anaesth. 2003;50:179–83.

141

Assessment Using the Sorour Airway Visualization Evaluation (SAVE) Score

15. Kim Si-Oh. Airway Management. J Korean Med Assoc. 2007 Dec;50(12):1048–1056. 16. Mallampati SR, Gatt SP, Gugino LD, Desai SP, Waraksa B, Freiberger D, et al. A clinical sign to predict difficult tracheal intubation: A prospective study. Can Anaesth Soc J. 1985;32:429–34. 17. Kim Si-Oh. Airway Management. J Korean Med Assoc. 2007 Dec;50(12):1048–1056. 18. Adamus M, Jor O, Vavreckova T, Hrabalek L, Zapletalova J, Gabrhelik T, et al. Inter-observer reproducibility of 15 tests used for predicting difficult intubation. Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub. 2011 Sep;155(3):275–81.

142

Scholarly Research In Progress • Vol. 3, November 2019

Assessing Adverse Childhood Experience Education for an Interprofessional Audience Jacob Arnold1†‡, Alayna Craig-Lucas1†‡, Mark White1, and F. Dennis Dawgert1

Geisinger Commonwealth School of Medicine, Scranton, PA 18509 †Doctor of Medicine Program ‡Authors contributed equally Correspondence: jarnold@som.geisinger.edu 1

Abstract Adverse childhood experiences (ACEs) have been linked to myriad health problems later in life, the relationships of which are only beginning to be understood (1, 2). An ACE is defined by the Centers for Disease Control and Prevention (CDC) as a term used to “describe all types of abuse, neglect, and other potentially traumatic experiences that occur to people under the age of 18” (3). Attendees (n=120) of a recent symposium about ACEs and resiliency were asked to participate in pre (n=81) and post (n=66) surveys assessing their self-reported awareness and understanding of ACEs and trauma-informed care, the results of which were analyzed using two-tailed T-tests. Participants demonstrated statistically significant increases across all categories assessed. We report those data and briefly discuss their implications for future educational programming.

Introduction Between 1995 and 1997, Kaiser Permanente health care system completed the original ACE study (1). Nearly 10,000 individuals completed confidential surveys in the first wave that sought to study seven categories of adverse child experiences. The seven categories were psychological, physical, or sexual abuse; violence against mother; or living with household members who were substance abusers, mentally ill or suicidal, or ever imprisoned. What was learned from this study changed the understanding of how our experiences impact us for the duration of our lives. Researchers determined there was a strong relationship between several leading causes of death in U.S. adults and the amount of abuse/household dysfunction that was experienced during childhood. A recent study found that more than 60% of people surveyed had experienced at least one ACE, and that such experiences are common across socioeconomic groups (4). However, despite a growing acceptance and understanding that ACEs represent an enormous public health burden, there is a gap in training and education for health care providers and community members at large. A recent survey of family medicine residents found that while 80 percent believed it was their job to screen for ACEs, fewer than half reported any formal training, and only 2 percent screened patients for ACEs in first encounters (5). It is therefore incumbent upon the medical community to design and implement educational programming to increase confidence and competence in caring for people who have experienced ACEs. Unfortunately, relatively few studies have examined design of such trainings, and none found in our literature review targeted heterogenous or multidisciplinary audiences (6, 7, 8).

On Friday, Nov. 16, 2018, Geisinger Commonwealth School of Medicine hosted a Keystone Symposium titled “Thinking Ahead to the Future Directions of Adverse Child and Adult Experiences and Their Relationship to Building Communities of Resilience.” The symposium was targeted at informing a wide range of community members including physicians, psychologists, social workers, counselors, nurses, physician assistants, medical students, parents, educators, first responders, and social service providers (see appendix for more information about the symposium). We aimed to build awareness of and competence with ACEs and resiliency, and trauma-informed care across the community in northeastern Pennsylvania.

Materials and Methods Participants were asked to complete pre- and post-surveys (see appendix), providing demographic information and rating their knowledge of and familiarity with adverse childhood experiences (ACEs) and resiliency scores using a sevenchoice Likert scale. Of the 120 attendees (n=120), 81 (n=81) participants completed pre-surveys and 66 (n=66) postsurveys; only the 66 paired pre-post surveys were included for evaluation. Answers to the Likert-scaled questions were assigned a numeric value from 1 (strongly disagree) to 7 (strongly agree). Blank responses were left blank and the corresponding question from the same participant was voided. Means were ascertained for all questions that used a Likert scale. Eight of the questions on the pre and post surveys were identical. Using Microsoft Excel’s Data Analysis package, results of the eight paired sets were subjected to two-tailed T-tests assuming unequal variance, a hypothesized mean difference of zero, and an alpha of 0.05. The data were then binned by profession and the tests were repeated within the subgroups following the same procedure. Due to the limited sample size, we grouped “social services, social work, counseling, or psychology” with “legal services, law enforcement or criminal justice,” and “educator, guidance counselor, teacher, or educational administrator” with “other.” We also grouped participants who indicated more than one primary profession into a separate “multi-professional” category.

Results Of the analyzed surveys (n=66), participants self-reported their primary fields to be student (n=27); Medicine, Clinical Care, or Health Care (n=24); Social Services, Social Work, Counseling, or Psychology (n=13); Educator, Guidance Counselor, Teacher, or Educational Administrator (n=9); Other (n=5); Legal Services, Law Enforcement, or Criminal Justice (n=2) (Figure 1). Several 143

Assessing Adverse Childhood Experience Education for an Interprofessional Audience

Figure 1. A bar graph showing the self-reported current occupations of participants

Figure 2. A pie chart showing the years of experience of participants in their respective fields

participants selected more than one option. Of those who selected â&#x20AC;&#x153;other,â&#x20AC;? (n=5) the free responses were respectively nonprofit management, speech pathology, ministry, fast food, and mother. The plurality of participants (n=25) reported 16 or more years of experience in their respective field, 6 reported 11 to 15 years of experience, 22 reported 6 to 10 years of experience, 5 reported 1 to 5 years of experience, 7 reported less than 1 year of experience in their respective fields, and 1 person declined to respond (Figure 2). The majority of participants (n=40) reported working with both adults and children, while 15 reported working only with adults and 6 only with children. Participants overwhelmingly agreed on the post survey that ACEs, resiliency, and trauma-informed care is important (mean:

144

Table 1. A table showing the means of the pre and post survey responses

6.8), that they felt ACEs, resiliency, and trauma-informed care will allow them to change their work (mean: 6.3), and largely agreed that they planned to use ACEs, resiliency, and traumainformed care in their work (mean: 5.8). The results of the paired pre and post survey questions can be found in Table 1. All eight questions showed statistically significant increases following the symposium and variation among responses markedly decreased for all questions except those asking whether the participant regularly uses ACE or resiliency scores respectively. We believe the increase without a convergence of responses suggests that some participants may have been

Assessing Adverse Childhood Experience Education for an Interprofessional Audience

incorporating ACEs, resiliency, and trauma-informed care into their work without knowing the terms for it. Additionally, the greatest increases were seen in the questions asking whether participants had heard of and were familiar with ACE and resiliency scores. When viewed categorically by profession, these data become less clear, likely due to the small sample sizes. However, for most questions, there were significant increases among participants who indicated medicine was their primary field and those who chose more than one primary field.

Discussion There are several key limitations of this study. First, the post survey was administered immediately following the symposium, so it is possible the effects, if evaluated longitudinally, would be more modest than the data suggest. Second, 15 participants completed pre-surveys but did not complete post surveys, so it is possible that participants who did not feel that they were benefiting from the symposium self-selected out of the participant pool. Third, data were not controlled for demographics, so the results may not be broadly applicable to all populations. Given these data, it appears that the symposium achieved its goals of increasing awareness and knowledge of ACE and resiliency scoring, and also increased participants’ belief in the importance of ACE and resiliency scores and trauma-informed care. We believe that the loss of statistical significance when viewed categorically by profession is likely due to the small sample size; future interventions with larger samples may consider reexamining this. Of note, medical professionals and people who selected more than one profession seemed to have the most consistent increase, which could indicate that programming was more effective for these groups. Indeed, several participants from other professions wrote in their comment section asking for greater community involvement in future interventions, with suggested participation ranging from local nonprofits to parents and foster parents. We wholeheartedly agree that ACEs cannot be effectively discussed solely within the context of a single medical specialty, or even medicine as a whole. Nevertheless, given the overall trends among all participants, we are confident that most participants, regardless of profession, gleaned some benefit through their attendance. Interestingly, there was also an increase in participants’ reporting that they regularly used ACE and resiliency scores, though variance remained relatively consistent. This could reflect a portion of participants who use similar techniques but did not call them ACE or resiliency scores or could simply have been a misunderstanding of the survey question. We would suggest that future interventions could attempt to incorporate a longitudinal component to the survey to better understand whether such educational programming can achieve lasting benefit. The importance of educating medical providers and community members about adverse childhood experiences cannot be understated. Such adverse experiences have repeatedly been linked to poor health outcomes, including mental health and substance abuse disorders, obesity, heart disease, asthma, and stroke (1, 2). It is incumbent upon all members of the community to understand adverse childhood experiences, and perhaps

more importantly, to strive to reduce their occurrence. Even people from medical specialties and parts of the community not typically included in the conversation around ACEs (such as vascular surgery and speech pathology) felt that they benefited from the symposium. Although previous works have found a high degree of enthusiasm among providers and students for learning more about ACEs (6, 7), training has not yet been widely adopted by the medical community (9). We have shown that educational programs can effectively address a heterogenous audience and reaffirmed that medical professionals at all levels and members of the community have a strong desire to learn more about ACEs, resiliency, and trauma-informed care. Educational programs such as the Keystone Symposium are an important first step upon which we hope future interventions may be built.

Acknowledgments The authors wish to thank Sierra Hall and Shawn Siroka for their assistance with our data analysis.

Appendix The objectives of the symposium are outlined below: 1. Define adverse childhood and adult experiences (ACAE) and their impact on an individual, a family and community. 2. Identify strategies to build community resilience. 3. Discuss an understanding/awareness of how local community programs can shift to enhance community resilience. 4. Define the common elements (guiding principles) in an effective clinical program that addresses trauma and its comorbidities. 5. Examine the future direction of cutting-edge research on trauma and the enhancement of resilience. Presentations at the symposium were provided by Leighton Huey, MD, associate dean of behavioral health integration and community care transformation; Julian Ford, PhD, professor of Psychiatry University of Connecticut School of Medicine and President International Society for Traumatic Stress Studies; and Brooks Keeshin, MD, assistant professor of Pediatrics and University of Utah board member for the Academy on Violence and Abuse. Following these seminars, a panel discussion with community leaders was moderated by Francis Dennis Dawgert, M.D. Pre-Survey Questions: 1.

I currently work in or have primarily worked in: (Student; Medicine, Clinical Care, or Health Care; Social Services, Social Work, Counseling, or Psychology; Educator, Guidance Counselor, Teacher, or Educational Administrator; Legal Services, Law Enforcement, or Criminal Justice; Other)

2. I have worked in the above field or fields approximately (<1, 1–5, 6–10, 11–15, 16+) 3. I work with (adults, children, both) 4. I have heard of ACEs (1–7)

145

Assessing Adverse Childhood Experience Education for an Interprofessional Audience

5. I have heard of Resiliency Scores (1–7) 6. I have heard of trauma informed care (1–7) 7.

I am familiar with ACE scoring (1–7)

5. Tink W, Tink JC, Turin TC, Kelly M. Adverse Childhood Experiences: Survey of Resident Practice, Knowledge, and Attitude. Family Medicine. 2017 Jan;49(1):7–13.

9. I currently often use ACE scores (1–7)

6. Stefanski K, Mason K. Acing Education: Pilot Curriculum on Adverse Childhood Experiences. Medical Education. 2017 Sep;51(11) 1167–8.

10. I currently often use resiliency scores (1–7)

8. I am familiar with resiliency scoring (1–7)

11. I believe that the utilizaition of ACE scores, resiliency scores, and or trauma informed care would benefit my clients, patients or constituents (1–7) 12. Comments Post-Survey Questions: 1.

I feel that ACE, Resiliency, and Trauma informed care are important (1–7)

2. I feel that ACE, Resiliency, and Trauma informed care are important (1–7) 3. I plan to utilize ACE scores, resiliency scores, and Trauma Informed Care in my work (1–7) 4. I have heard of ACE (adverse childhood experiences) scores (1–7) 5. I have heard of resiliency scores (1–7) 6. I have heard of trauma informed care (1–7) 7.

I am familiar with ACE scoring (1–7)

8. I am familiar with Resiliency scoring (1–7) 9. I currently often use ace scores (1–7) 10. I currently often use resiliency scores (1–7) 11. I believe that the utilization of ACE scores, resiliency scores, and or trauma informed care would benefit my clients, patients or constituents (1–7) 12. Comments

References 1.

Felitti VJ, Anda RF, Nordenberg D, Williamson DF, Spitz AM, Edwards B, et al. Relationship of Childhood Abuse and Household Dysfunction to Many of the Leading Causes of Death in Adults. American Journal of Preventive Medicine. 1998 May;14(4):245–58.

2. Gilbert LK, Breiding MJ, Merrick MT, Thompson WW, Ford DC, Dhingra SS, Parks SE. Childhood Adversity and Adult Chronic Disease: An Update from Ten States and the District of Columbia. American Journal of Preventive Medicine. 2015 Mar;48(3)345–9. 3. Centers for Disease Control and Prevention. Violence Prevention [Internet]. Atlanta, GA; National Center for Injury Prevention and Control; updated 2019 Apr. Available from https://www.cdc.gov/violenceprevention/ childabuseandneglect/acestudy/aboutace.html 4. Merrick MT, Ford DC, Ports KA, Guinn AS. Prevalence of Adverse Childhood Experiences From the 2011–2014 Behavioral Risk Factor Surveillance System in 23 States. JAMA Pediatrics. 2018 Nov;172(11):1038–44.

146

Goldstein E, Murray-Garcìa J, Sciolla AF, Topitzes J. Medical Students’ Perspectives on Trauma-Informed Care Training. The Permanente Journal. 2018 Jan;22:17–126.

8. Gill ME, Zhan L, Rosenberg J, Breckenridge LA. Integration of Adverse Childhood Experiences Across Nursing Curriculum. Journal of Professional Nursing. 2019 MarApr;35(2):105–11. 9. Ford DE. The Community and Public Well-being Model: A New Framework and Graduate Curriculum for Addressing Adverse Childhood Experiences. Academic Pediatrics. 2017 Sep-Oct;17(7): S9–11.

Scholarly Research In Progress • Vol. 3, November 2019

Opioid Mortality Following Implementation of Medical Marijuana Programs in the United States Daniel E. Kaufman1*, Asawer M. Nihal1*, and Janan D. Leppo1*

Geisinger Commonwealth School of Medicine, Scranton, PA 18509 *Master of Biomedical Sciences Program Correspondence: dk811162@gmail.com 1

Abstract Background: Cannabis is a plant with origins tracing back to the ancient world. Cannabis was used medicinally during the 19th and 20th centuries in the U.S. California was the first state to legalize cannabis for medical purposes in 1996. Until recently, cannabis was purportedly a “gateway drug” whose use led to the use of other, more harmful drugs. With America struggling to fight the opioid epidemic, communities have started to turn to cannabis. Recent observational and epidemiological studies have found that medical marijuana programs are associated with a reduction in the use of opioids and associated morbidity and mortality (1). Since opioids and marijuana are both utilized to treat chronic pain, it is reasonable to hypothesize that medical marijuana may serve as a viable long-term alternative to the highly addictive opioid pharmaceuticals. Methods: This study examined the effects of medical marijuana on opioid mortality as reported by the Centers for Disease Control and Prevention’s WONDER database. We calculated slopes for both three years pre and post medical marijuana implementation. States without medical marijuana programs (Alabama, Georgia, Idaho, Indiana, Iowa, Kansas, Kentucky, Louisiana, Mississippi, Nebraska, North Carolina, South Carolina, South Dakota, Tennessee, Texas, Virginia, Wisconsin, and Wyoming) as controls. The start dates were determined to be when each state initiated medical marijuana sales. Start date of controls was defined by the average start date of medical marijuana programs. Results: Overall, opioid overdose deaths increased in each of the states studied. Additionally, the factor of medical marijuana was found to be statistically significant between the two groups of opioid overdose death data. Opioid overdose deaths increased significantly after implementation in states without medical marijuana (p<0.05). Overdose deaths showed a non-significant elevation in states with medical marijuana (p>0.05). Conclusion: This study showed that states with medical marijuana had an attenuated increase in opioid mortality. Further research is needed to determine if this pattern generalizes to state-level recreational marijuana policies.

Introduction The opioid receptor family consists of the mu, kappa, and delta G protein-coupled receptors and mediates the descending pain pathway. The opioid receptors are distributed throughout the brain, with the mu receptor being the most widely distributed out of the three (1). When bound by exogenous peptides like morphine or endogenous peptides including endorphins, these receptors mediate responses

of analgesia, reward, and addiction (2). They have thus been targets of drug therapy for decades due to these effects (2). When an individual has been taking opioids for analgesia and pain management, tolerance to opioids can occur. Larger quantities are needed on subsequent dosages (3). This can lead to opioid overdose which results from respiratory depression that a larger dose of opioids induces (3). More than two-thirds (68%) of 70,200 drug overdose deaths in 2017 were opioid-related (4). Opioid-related overdose deaths were six times higher in 2017 than in 1999 (4). About 130 Americans die every day from an opioid overdose, many of which are accidental and preventable (4). It is also estimated that the total “economic burden” of opioid misuse is around $78.5 billion a year (5). This total includes the cost of health care, treatment for opioid addiction and criminal involvement. According to the National Institutes of Health (NIH), about 21% to 29% of patients are given opioid prescriptions for pain and reported to misuse them. Of the patients that do use them, more than 12% develop opioid use disorder while opioid overdoses have increased by more than 30% from July of 2016 to September of 2017 in 45 states (5). Endocannabinoids (eCB) and their receptors are found in the human body. They play a role in eating, sleeping, relaxing, forgetting, and protecting. Deficiency in eCB signaling may contribute to the pathogenesis of depression. Recent studies have shown endocannabinoid deficiency has been linked to schizophrenia, MS, Huntington’s disease, and Parkinson’s disease (6). A variety of physiological effects occur when cannabinoid receptors are stimulated. Cannabinoid receptor type 1 (CB1) is most abundant G protein coupled receptor. Cannabinoid receptor 2 (CB2) is involved in immune function and in central nervous system executive functioning (6). The pharmacokinetics of an active ingredient in cannabis, tetrahydrocannabinol (THC), differ depending on the route of their administration. The three most common ways to administer cannabis are through inhalation, smoking, and consuming edibles containing the drug. The path of administration affects the onset and intensity of the effects of the drug. Compared to the administration route of ingestion, smoking the drug results in the rapid transfer of cannabinols from the lungs to the blood. It was also revealed that THC was detected in plasma immediately after the first inhalation of marijuana smoke (6). THC is lipophilic. It therefore distributes to tissues and fat. Inhaling THC causes maximum plasma concentration within a few minutes of taking it (6). Effects reach their peak around 15 to 30 minutes and wear off within 2 to 3 hours when inhaled. Plasma levels are seen to be lower after administration by the oral route. If taken orally, the effects are witnessed within 30 to 90 minutes and they reach their peak effect after 2 to 3 hours.

147

Opioid Mortality Following Implementation of Medical Marijuana Programs in the United States

Depending on the dose, the effects can last anywhere from 4 to 12 hours (6). Cannabis has legal, ethical, and societal controversies surrounding its use. According to the U.S. Drug Enforcement Agency’s Comprehensive Drug Abuse Prevention and Control Act of 1970, marijuana is recognized as a schedule I controlled substance, where it is listed as having a high potential for abuse. This scheduling also signifies no currently federally accepted medicinal purpose and a lack of safety data for the use of treatment under medical supervision (6). According to the World Health Organization, cannabis is the most illicitly abused drug in the world. It has an annual prevalence of being consumed by 147 million people worldwide. In the year 2014, 22.2 million Americans age 12 years and older admitted to current cannabis use. More recently, cannabis has gained acceptance across the states as reported by legislative actions and public opinion polls. In 2016, a Gallup poll reported that more than 60% of the American population believed that cannabis should be legalized, with 81% agreeing legalization of cannabis should be made for medical reasons. Cannabis has been legalized for medicinal use in 28 states by the end of the year 2017 (6). Recent research indicates that legalizing access to marijuana has had a beneficial effect on public health and drug abuse. Studies support that medical cannabis administration is linked to a reduced use of opioids (7, 8). States with medical cannabis laws from 1999 to 2010 had 24.8% lower mean annual opioid overdose mortality rate compared to states without medical cannabis law (7). In states where medical cannabis was legal, the annual number of doses prescribed per physician fell by 1,826 doses (8). Research has also found that cannabis use can positively impact chronic pain treatment success rate (8). It is hypothesized that since cannabis has been reported to show relief in the treatment of pain, it may contribute to a reduction in opioid use and misuse. It is hypothesized that opioid mortality will have a greater increase over time in states without medical marijuana programs.

Materials and Methods Procedures The opioid overdose mortality rate from 1999 to 2017 in each state was extracted from the Centers for Disease Control and Prevention’s (CDC's) Wide-ranging Online Data for Epidemiologic Research (WONDER) database (9). The data are based on death certificates for U.S. residents. Opioid overdose deaths were defined using the International Statistical Classification of Diseases, 10th revision codes: X40–44, X60– 64, X85, Y10–Y14 (10). Only data that was coded for opioids (T40.0-T40.4) was used (11). This included opium, heroin, other opioids, methadone, and other synthetic narcotic overdose deaths. This definition and range of data includes all street and pharmaceutical opioid drugs, providing an effective source for the interrupted time series examining any trends around the time of states’ medical marijuana program implementation. Microsoft Excel was used for data management. The data provided was based on deaths per 100 000 (9). We defined the start dates of medical marijuana programs

148

as the year the state began medical marijuana sales. This data was found on either state’s medical marijuana program website, or from local newspaper articles detailing the start of medical marijuana sales. Only Arizona (implemented in 2012), Connecticut (2014), Delaware (2015), District of Columbia (2013), Illinois (2015), Maine (2011), Massachusetts (2013), Minnesota (2015), New Jersey (2012), New Mexico (2007), Rhode Island (2013), and Vermont (2004) had start dates within the year range of 1999 to 2017, with three years of data available both pre- and post-medical marijuana program implementation. States that have no medical marijuana laws served as controls (Alabama, Georgia, Idaho, Indiana, Iowa, Kansas, Kentucky, Louisiana, Mississippi, Nebraska, North Carolina, South Carolina, South Dakota, Tennessee, Texas, Virginia, Wisconsin, and Wyoming). To determine a mean year to analyze for the control states, the average start date was calculated by finding the mean of all the medical marijuana program implementation dates (2012). Procedures were deemed exempt by the IRB of the University of New England. Data analysis Opioid death data points from three years before the implementation of medical marijuana, the year of and three years after, along with data points from the control group were transferred to GraphPad Prism 8. Slopes were then calculated for both pre- and post-implementation in each state. Pre-and post-implementation slopes included three pre data points. Slopes for each state were transferred back to Microsoft Excel for data analysis. Slopes were indicative of the change before and after the implementation in each of the states examined. Statistical analysis was conducted using SPSS software, v25.0 (IBM Corp.). Analyses were conducted to determine any significant differences between states with medical marijuana programs and those that do not have these programs. A p value less than 0.05 was considered statistically significant. A mixed (time x implementation) ANOVA was completed. Outliers were then tested for using GraphPad Prism 8 Grubbs Test (13). Any outliers in either the pre- or post- slope category were examined. Only one was found, being Kentucky’s post-implementation slope. The data from Kentucky was removed before data analysis was completed. T-tests were completed using the SPSS software to determine any statistical significance between the two groups of opioid overdose death data. Heat maps were created using Microsoft Excel.

Results Slopes were averaged for each group. A mixed ANOVA was completed. There was a main effect of medical cannabis (F (1, #)=9.256, p =0.005). Additionally, the repeated measure was significant (F (3, #)=12.350, p=0.002). The interaction was nonsignificant (F (2, #)=2.099, p=0.158). The means and SEMs are reported in Figure 1. Both pre and post implementation slopes in medical marijuana states were greater than slopes without medical marijuana programs. Also, post slopes in states without medical marijuana were significantly (p<0.05) greater than their respective pre slopes. The increase from pre to post slopes in states with medical marijuana was non-significant (p<0.10).

Opioid Mortality Following Implementation of Medical Marijuana Programs in the United States

Heat maps consisting of the difference between post-slope and pre-slope to visualize opioid mortality were created (Figure 2). The difference between slopes is indicative of the increase of opioid overdose deaths between the range of three years pre- and postimplementation of medical marijuana. One can see the largest increase in Massachusetts and a slight east coast Figure 1. Averages between pre and post trend, with states implementation slopes of states with and (South Carolina and without medical marijuana (*significant Delaware) on the p<0.05, #non-significant p<0.10) east coast slightly darker than the rest of the country. Darker colors represent a greater increase in opioid mortality in their respective defined dates, while lighter colors are indicative of a greater decrease in opioid mortality. After illustrating the states we examined in this study, it was decided to show the increase of opioid mortality in every state (Figure 3). This was completed by taking the latest data point of opioid mortality from CDC’s WONDER for each state (2017) and subtracting the earliest possible data point (2). Some states did not have data available early in the data set (1999–2001). The uptick in both West Virginia and Ohio is apparent, as is a slight uptick in the Northeast.

Discussion This study assessed whether state level medical marijuana policies impacted opioid mortality and is an extension of prior research (4). Opioid mortality significantly increased (p<0.05) in states without medical marijuana programs. However, this was not the case in states with medical marijuana programs (p>0.05), which provides some support for the hypothesis. Figures 2 and 3 visualize the opioid epidemic by illustrating the increase in opioid mortality in the U.S. Figure 2 shows an increase of opioid mortality between post and pre slopes in each of the states examined, with Massachusetts having the greatest increase. Figure 3 provides a broader visualization of the opioid epidemic in the United States. West Virginia and Ohio showed the greatest increase in opioid mortality since 1999 with a noticeable uptick in the rest of the Northeastern U.S. Figure 1 provides the average slopes between the controls and experimental groups, showing the findings of a significant increase in the controls and a non-significant increase in the experimental group.

Figure 2. Heat map of the United States indicative of the difference between the post and pre slope in overdoses as reported by the CDC’s WONDER database

Figure 3. Heat amp of the United States with darker colors indicating a larger difference between defined opioid deaths in 2017 and the earliest available data point in each state

Regarding limitations for this quasi-experimental report, only a limited number of states (12) met the criteria of having three years before and three years after implementation of medical marijuana. With some states having implemented their programs within the past two to three years, data was not available for examination in this study. In subsequent years, more data may become available from the CDC WONDER database for analysis. This would enable more states to be studied using this methodology, enabling future researchers to examine the trends of opioid mortality data in these additional states. We cannot exclude that there was some unmeasured confound which differed among medical marijuana (largely in the northern and eastern U.S.) and non-medical marijuana states. It should be noted that within the CDC WONDER database, each state’s classification of an opioid death occurs differently according to the coroner’s determination. This lack of homogeneity in the quality of death determinations may influence data in the database depending the pronounced variations in reporting.

149

Opioid Mortality Following Implementation of Medical Marijuana Programs in the United States

Conclusion Opioid mortality has been on the rise over the past decade, leading to the designation of an opioid epidemic or opioid crisis (7). Although medical marijuana’s beneficial effects for chronic pain have been sufficiently documented and reported, it has not been recognized as a viable remedy for the current opioid epidemic until recently (8). Our study observed and analyzed trends in opioid mortality three years before and three years after the implementation of medical marijuana programs. Opioid overdose deaths increased in each of the states examined. Furthermore, deaths increased significantly in states without legalized medical marijuana. Deaths did not significantly increase in states that legalized medical marijuana which is broadly concordant with past research (7). Further study is warranted to determine if there are additional benefits or risks of legalizing marijuana for adults for recreational purposes.

Acknowledgments The authors would like to thank Brian Piper, PhD, and Elizabeth Kuchinski, MPH, for their help and guidance throughout the research process, as well as Kenneth McCall, PharmD, of the University of New England.

References 1.

Piper BJ. Somatosensory I. Geisinger Commonwealth School of Medicine. 2019.

2. European College of Neuropsychopharmacology. How does the opioid system control pain, reward and addictive behavior? ScienceDaily. ScienceDaily; 2007 3. Miller KA. Pain Pharmacology. Geisinger Commonwealth School of Medicine. 2019. 4. Ozluk P. The effects of medical marijuana laws on utilization of prescribed opioids and other prescription drugs. SSRN [internet]; Accessed from https://papers.ssrn. com/sol3/papers.cfm?abstract_id=3056791 5. National Institute on Drug Abuse. Opioid overdose crisis [internet]. Accessed from https://www.drugabuse.gov/ drugs-abuse/opioids/opioid-overdose-crisis 6. Bridgeman M, Abazia D. Medicinal cannabis: history, pharmacology and implications for the acute care setting. P T. 2017. 42(3):180–188. 7.

Bachhuber MA, Saloner B, et al. Medical cannabis laws and opioid analgesic overdose mortality in the United States, 1999–2010. JAMA Intern Med. 2014. 174(11):1875.

8. Lucas P, Rationale for cannabis-based interventions in the opioid overdose crisis, Harm Reduct J. 2017. 14(1):58. 9. Centers for Disease Control and Prevention. CDC Wonder. http://wonder.cdc.gov/. February-March 2019. 10. Mack A, Joy J. Marijuana and pain. National Academies Press [internet]. 11. Powell D, Pacula R. Do medical marijuana laws reduce addictions and deaths related to pain killers? J Health Econ. 2018. 58(1): 29–42. Accessed from; https://www.sciencedirect.com/science/article/abs/pii/ S0167629617311852

150

12. Livingston M, Barnett T, Delcher C, Wagenaar AC. Recreational cannabis legalization and opioid-related deaths in Colorado, 2000–2015. Am J Public Health. 2017. 11(1): 1827–1829. Accessed from https://ajph. aphapublications.org/doi/abs/10.2105/AJPH.2017.304059 13. GraphPad. Outlier calculator. https://www.graphpad.com/ quickcalcs/Grubbs1.cfm. March 2019. 14. Baker DW. History of the Joint Commission’s pain standards. JAMA. 2017. 317(11):1117–1118. 15. Wide-ranging online data for epidemiologic research (WONDER). Atlanta, GA: CDC, National Center for Health Statistics; 2017. Accessed from http://wonder.cdc.gov.

Scholarly Research In Progress • Vol. 3, November 2019

Prescription Opioid Distribution Before and After Legalization of Recreational Marijuana in Colorado, 2007–2017 Amalie K. Kropp1*, Stephanie D. Nichols2, Daniel Y. Chung1*, Kenneth L. McCall2, and Brian J. Piper1

Geisinger Commonwealth School of Medicine, Scranton, PA 18509 University of New England, Portland, ME 04005 *Master of Biomedical Sciences Program Correspondence: akropp@som.geisinger.edu 1 2

Abstract Background: Opioid-related overdoses and overprescribing continue to be an ongoing issue in the United States. Further consideration of nonopioid alternatives as a substitute to treat chronic noncancer pain and in the treatment of opioid-use disorders (OUD) is warranted. Objective: To examine the association between the legalization of Colorado’s recreational marijuana and prescription opioid distribution trends. Two states that have not legalized recreational cannabis were selected for comparison. Methods: The United States Drug Enforcement Administration’s Automation of Report and Consolidated Orders System (ARCOS) was used to examine 9 pain medications: oxycodone, fentanyl, morphine, hydrocodone, hydromorphone, oxymorphone, tapentadol, codeine, and meperidine, and two OUD medications: methadone and buprenorphine, from 2007–2017 in Colorado, Utah, and Maryland. The drug weights were extracted, examined, and graphed. Medications were converted to their oral morphine milligram equivalents (MME) using standard conversion factors.

an OUD. Utah distributed 61.00% for pain and 39.00% for OUD. However, Maryland was one-third (37.89%) for pain but more than three-fifths (62.11%) for an OUD. Conclusion: This study found that there has been a significant decrease in the prescription opioid distribution in Colorado after the legalization of marijuana. This finding was particularly notable for opioids indicated predominantly for analgesia such as hydrocodone, morphine, and fentanyl. Colorado had a larger decrease in opioid distribution after 2012 than Utah or Maryland. Therefore, marijuana should be considered as an alternative treatment for chronic pain and reducing use of opioids. Also, when combined with other research reviewed in this study, marijuana use may reduce the overdose death rate. Additional research with more comparison states is ongoing.

Introduction An epidemic is plaguing the United States regarding the misuse of prescription opioids over the last 15 years. The opioid epidemic stems from the early 1990s when the medical community started to recognize pain as a fifth vital sign (1). This philosophical shift caused the implementation of national initiatives to improve pain-related care in 2001. Specifically, The Joint Commission released guidelines that affected physician prescribing behavior on a state and national level. The Drug Enforcement Administration (DEA) found a marked increase in the total number of opioids prescribed every year

Results: Colorado reached a peak of pain MME weight in 2012 and had an -11.66% reduction from 2007 to 2017. During the same interval, Utah had a +9.64% increase in pain medication distribution and Maryland a -6.02% reduction. As for medications used for OUD, Colorado, Utah and Maryland had +19.42% increase, -31.45% reduction, and +66.56% increase, respectively. Analysis of the interval pre (2007–2009) versus post (2013–2017) marijuana legalization was completed. Statistically significant changes were observed for Colorado (P=0.033) and Maryland (P=0.007), but not Utah (P=0.659) for pain medications. Analysis of the OUD medications identified significant changes for Colorado (P=0.0003) and Maryland (P=0.0001), but not Utah (P=0.0935). Over the decade, Colorado’s opioid distribution was predominantly (72.49%) for Table 1. Table shows the relative comparison demographics between Maryland, Colorado, and Utah. pain with one-quarter (27.51%) for Data was obtained from the US Census 2017–2018.

151

Prescription Opioid Distribution Before and After Legalization of Recreational Marijuana

since the mid-1990s (2), leading up to the national peak in opioid distribution in 2012. Opioid prescribing increased from 148 million prescriptions in 2005 to over 206 million by the end of 2011 (3). Although the intent of these guidelines was to improve pain care, there was no significant increase in the quality of pain management (1). This has had long-lasting and devasting effects that have rippled throughout the entire country as more people became dependent and overdosed on opioids. While there have been downward trends in the prescribing of opioids after 2012, patients are still dying. Ninety Americans die each day due to opioid overdoses (4). There has been a 320% increase in opioid-related mortality between 2000 and 2015, and it is still 3 times higher than in 1999 (5). Physicians and other health care providers have the responsibility to treat their patients' noncancer pain, while also considering nonopioid alternatives.

Controlled Substances Act that collects data on narcotics in Schedules II to III from hospitals, narcotic treatment programs (NTPs, also known as methadone programs) and pharmacies (2). The program is overseen by the Drug Enforcement Administration (DEA) with the goal of preventing, detecting, and investigating controlled pharmaceuticals (2). Nine opioid pain medications (oxycodone, fentanyl, morphine, hydrocodone, hydromorphone, oxymorphone, tapentadol, codeine, and meperidine) and two primarily opioid-use disorder (OUD) medications (methadone and buprenorphine) were examined from 2007 through 2017 in Colorado, Utah, and Maryland. Colorado and Maryland had similar demographics in terms of population size, percent of house owners, percentage of bachelor’s degrees or higher education

Since California first legalized medical marijuana in 1996, 33 states and the District of Columbia have passed laws broadly legalizing marijuana, either medically or recreationally, as of November 2018. Washington, DC, and 10 other states have expanded their laws to include recreational marijuana use (6). With the endorsement of the states, more objective, clinical evidence is surfacing that marijuana can be used to manage chronic pain (7), reduce overdose mortality rates (8, 9), treat opioid withdrawal (10), and decrease opioid prescribing rates (5, 11, 12). Marijuana has a much lower risk of addiction and virtually no overdose danger, which is in stark contrast to opioids (13, 14, 15, 16). In January 2017, the National Academies of Sciences, Engineering and Medicine released a peer-reviewed, comprehensive review showing “conclusive evidence” that cannabis can be used safely and effectively to treat chronic pain (17). According to a Pew research center poll conducted in 2018, 62% of Americans are supportive of legalized marijuana for medical purposes, which has doubled in over a decade from 31% in 2000 (18). The popularity of marijuana is quickly rising in the 50 and older age group (19). This age group may be most likely to experience chronic-pain-related conditions and are open to the analgesic effects of marijuana (7). Overall, marijuana is gaining strong support politically. This rise in acceptance also helps dispel myths of cannabis being a gateway drug, decreasing the stigma of this alternative treatment. To date, there has been little to no research conducted on the effects of adult use marijuana laws on opioid distribution. In November 2000, 54% of Colorado voters approved Amendment 20, implementing the legalization of medical marijuana (20). Twelve years later, Colorado approved Amendment 64, legalizing adult-use or recreational marijuana (20). By January of 2014, dispensaries were opened to the public. This study compares medical opioid distribution in Colorado with two states that have not legalized recreational marijuana.

Materials and Methods Procedures The Automation of Report and Consolidated Orders System (ARCOS) is a federal program created because of the 1970

152

Figure 1. Medications for pain (oxycodone, fentanyl, morphine, hydrocodone, hydromorphone, oxymorphone, tapentadol, codeine, meperidine), opioid use disorder (OUD) (methadone and buprenorphine), and all 11 drugs in their morphine milligram equivalents in kilograms per year in Colorado, Maryland, and Utah from 2007 to 2017.

Prescription Opioid Distribution Before and After Legalization of Recreational Marijuana

holders, and uninsured rates gathered from the U.S. Census 2018 data. Utah was chosen as the most geographically similar state with similar body mass index (BMI) and median household income (Table 1) (21). These social determinants of health are important to consider when looking at opioid distribution at a state level. Social, regional, and economic factors play a substantial part in opioid use and misuse when examining the multifactorial realm of pain and treatment options. Institutional Review Board approval was provided by the University of New England.

Statistical analysis All 11 medications were adjusted to their oral morphine milligram equivalents (MME). This enabled the agents to be compared despite their differences in relative potency. Oral MME conversions were completed using the following multipliers: oxycodone 1.5, fentanyl 75, morphine 1, hydrocodone 1, hydromorphone 4, oxymorphone 3, tapentadol 0.4, codeine 0.15, meperidine 0.1, methadone 8, and buprenorphine 10 (22). Heat maps of three-digit ZIP codes were prepared using QGIS; other figures were created with

Figure 3. Heat maps showing morphine milligram equivalents per ZIP code, per population in kilogram weight. Capital and major cities are shown. Population data from Statista. Maps created with QGIS.

Figure 2. Morphine milligram equivalents of 11 drugs in Colorado, Utah, and Maryland from 2007 to 2017.

153

Prescription Opioid Distribution Before and After Legalization of Recreational Marijuana

Figure 4. Morphine milligram equivalents of 9 pain opioid in Colorado, Utah, and Maryland per person from 2007 to 2017. Colorado had an -11.66% reduction from 2007 to 2017. Utah had a +9.64% increase in pain medication distribution and Maryland, -6.02% reduction (top). Morphine milligram equivalents of 2 OUD drugs in Colorado, Utah, and Maryland per person from 2007 to 2017. Colorado, Utah, and Maryland had +19.42% increase, -31.45% reduction, and +66.56% increase, respectively (bottom).

Figure 5. Percent of opioid use for pain (oxycodone, fentanyl, morphine, hydrocodone, hydromorphone, oxymorphone, tapentadol, codeine, meperidine) and OUD (methadone and buprenorphine), by their oral morphine milligram equivalents in Colorado, Maryland, and Utah from 2007 to 2017.

Figure 6. T-test analysis of the 9 pain medications found statistically significant reductions for Colorado (P=0.033) and Maryland (P=0.007), but not Utah (P=0.659). T-test analysis of the 2 OUD medications found statistically significant changes for Colorado (P=0.0003) and Maryland (P=0.0001), but not Utah (P=0.0935). 154

Prescription Opioid Distribution Before and After Legalization of Recreational Marijuana

GraphPad Prism, version 8.1. Population data was taking from Statista. T-tests compared pre-marijuana legalization (2007– 2012) and post-marijuana legalization (2013–2017) for the 9 pain medications and 2 OUD medications. T-tests were calculated with GraphPad QuickCalcs.

Results Figure 1 shows the total MMEs of all 11 opioids, opioids used for OUD and opioids used for pain from 2007 to 2017. Maryland had the highest weight and peaked in 2011 at 12,167 kg MME for all 11 opioids, over twice the weight than the comparison states. At their peaks, Colorado and Utah only reached 5,029 kg MME in 2012 and 3,429 kg in 2015, respectively. Over the decade, Maryland had an increase in methadone and buprenorphine and about the same amount of distribution of pain medications. Utah remained relatively constant in all categories over the decade. Colorado, similar to Maryland, also experienced an increase distribution of OUD medications, an increase in all opioids, and small elevations in pain medications. Figure 2 shows the MME of each of the 11 opioids. Oxycodone and methadone were the most distributed in all states. In Maryland, methadone peaked in 2011 at 7,237 kg. In contrast, methadone in Colorado peaked at 1,384 kg in 2017 and, in Utah, 1,054 kg in 2013. Tapentadol, codeine, and meperidine were the least distributed. Buprenorphine increased over the decade in all states. Figure 3 shows heat maps of the relative MME per ZIP code, per population in kg weight in 2017. Colorado’s most populated ZIP codes, 802-prefix, surrounding Denver, had more than 1.24 metric tons of opioids in MME distributed, reflecting the ZIP code with the highest distribution amount. Baltimore, Maryland, had the densest distribution of opioids, with more than 4.00 metric tons distributed throughout the 212-prefix ZIP codes. Interestingly, Utah had two ZIP codes around Salt Lake with similar MME weights—840 and 841 each had MME weights of 1.15 metric tons. The MMEs of the 9 pain medications per person were graphed over the decade within each state in Figure 4. Colorado had an -11.66% reduction of these 9 pain medications from 2007 to 2017. Utah had a +9.64% increase and Maryland, a -6.02% reduction in distribution of the 9 pain medications. The distribution of two OUD medications, methadone and buprenorphine, were also reviewed from 2007 to 2017. Colorado, Utah and Maryland had +19.42% increase, -31.45% reduction, and +66.56% increase, respectively. This sizable increase in Maryland was a unique finding, as neither of the other states showed this substantial a change over time. Although, Colorado and Maryland were trending up over the last decade in regard to OUD medication distribution, Utah had a significant decrease in OUD distribution within the same timeframe. The percent of total opioid distribution for pain and OUD, by their oral MME in Colorado, Maryland, and Utah from 2007 to 2017 is highlighted in Figure 5. Colorado and Utah showed a similar trend, with a total of 79.49% and 61.00% of pain medications, respectively, being dispersed into the states. Maryland revealed an opposite trend, with only 37.89% pain

medication distributed and 62.11% OUD medications distributed over the decade. Colorado and Utah only had 27.51% and 39.00% of OUD medications dispersed. Overall, more medications being used for OUD were being dispensed into Maryland than Colorado and Utah. The T-tests performed between the pre-marijuana legalization (2007–2012) and post-marijuana legalization (2013–2017) for the 9 pain medications and 2 OUD medications are shown in Figure 6. These years referenced the legalization of recreational marijuana in Colorado in late November 2012. The T-test analysis was looking to show statistical significance of the effects of the legalization of marijuana in Colorado during the proposed timeframes. There were statistically significant reductions for Colorado (P=0.033) and Maryland (P=0.007), but not Utah (P=0.659). Analysis of the OUD medications found statistically significant differences for Colorado (P=0.0003) and Maryland (P=0.0001), but not Utah (P=0.0935). These findings show that the legalization of marijuana in Colorado could have reduced the amount of pain and OUD medication distribution into Colorado, but not a reduction based on chance alone. The legalization showed a real difference in the two timeframes.

Discussion The results of this novel pharmacoepidemiology and public policy research study support recreational marijuana as a substitute for prescribing opioids. In research on overprescribing of opioids in states with medical and recreational marijuana laws, they found a 5.88% reduction in states with medical marijuana laws and a 6.38% decrease in states with adult use laws. This included a decrease in Schedule II drugs of 7.79% and Schedule III to V by 10.40% (12). Their study found that Maryland did not have discernable changes in opioid prescribing, and Colorado had significantly lower prescribing rates (12). While in our research, Maryland did have a statistically difference in the pre/post marijuana years, highlighting a difference in opioid distribution due to unknown or multiple confounding factors. But Colorado still experienced a larger decrease in opioid distribution than Maryland or Utah. Maryland actually had larger weights in every year, in every drug, when compared to Colorado and Utah. While Maryland did legalize medical marijuana, the law went into effect in June of 2014 with no real operations until late 2016. This rollout could have been less effective due to bureaucratic and legal issues within the state, leading to no significant changes in opioid prescribing (12). Olfson and colleagues found an association between illicit cannabis use and an increased incidence of nonmedical prescription opioid use (odds ratio, 5.78; 95% CI 4.23–7.90) using the National Epidemiological Survey on Alcohol and Related Conditions (NESARC) data set (23). However, the NESARC findings may not apply to our study, which evaluated an association between legalized cannabis and prescription opioids rather than between illicit cannabis use and nonmedical opioid use. Regardless, further research is needed to establish the relationship between cannabis and opioid use. Marijuana may also be considered to help keep people from overdosing on and dying from opioids. In 2014, a study found a 24.8% reduction in deaths among states that had medical

155

Prescription Opioid Distribution Before and After Legalization of Recreational Marijuana

cannabis laws (MCLs) (8) and fewer daily doses filled by patients within the Medicare Part D population (5). Another report found a 6.5% reduction in overdose deaths after the legalization of recreational marijuana in Colorado (9). This represents an important reversal in the upward trend that Colorado was experiencing in opioid-related deaths. With fewer opioids being distributed to the states, and concrete evidence of a decrease in opioid-related deaths, Colorado’s legalization of recreational marijuana shows an encouraging tread to treat people’s pain and keep them safe. Opioid prescribing practices were beginning to decrease before any additional guidelines were released. When the CDC released the Guideline for Prescribing Opioids for Chronic Pain in March 2016, there was a substantial decrease in opioid prescribing practices (24). Research from this study supports this finding. Colorado and Maryland experienced an overall decrease in opioid distribution, but Colorado’s decrease was larger. While the nation as a whole experienced a decrease in opioid distribution, Colorado’s greater decrease shows promise to the potential impact of recreational marijuana. This study did have some limitations. There was a limited data range within ARCOS that was available. For enhanced data, it would have been advantageous to look at another decade. ACROS did not start posting data until 2000. Getting closer to the beginning of the opioid epidemic may produce some important information about long-term trends within the United States. Also, our data set only looked at distribution data, not individual or pharmacy level data. Improved data could come from tracking the drugs at the ZIP code level, but that is confounded by mail-order pharmacies and internet pharmacies. Confounding variables such as changing public opinion on cannabis use and related public health policies were not evaluated. During the timeframe of this study, neither Maryland nor Utah legalized recreational marijuana. However, Maryland passed a medical marijuana law in 2013 with sales of medical cannabis beginning in late 2017. Also, Maryland decriminalized possession of up to 10 grams of marijuana in 2014. Additionally, public opinion about cannabis use has been changing as evidenced in 2018 when Utah voters passed a medical cannabis law. This law will become operational in Utah in 2020 (25). These related cannabis policy changes could have had a confounding effect on medical opioid use. It is unknown if patients actually reduce opioid use directly because of increased access to marijuana or if this association results from other confounding factors. The intricacies of pain control and state level bureaucracies is tremendously multifactorial, and it makes it difficult to control every factor to isolate direct causation of marijuana on opioid distribution changes. The impact of guideline changes on the opioid distribution in 2016 may have confounded the results, not just the impact of marijuana within the three states. Further studies should focus more selectively on the subset of populations that are using recreational marijuana.

156

Conclusion In this study, we observed the differences in opioid distribution of 11 medications used for pain and OUD within Colorado, Utah and Maryland from 2007 to 2017. Colorado, having legalized recreational marijuana, had a significant decrease in pain opioid distribution from 2007 to 2017. These findings further strengthen the need for additional research into the association of cannabis laws and prescription opioid use. Combined with other recent marijuana research, marijuana may be useful for pain relief without dependency on opioids. Future guidelines may support decreasing prescribing practices for physicians and the acceptance of marijuana as a treatment option for chronic pain. Physicians and policymakers should consider other options for this opioid epidemic, as it affects people across the country. Further, there should be a national level push for legalization of marijuana under United States federal law to at least allow for proper, concrete research on the overall health effects. If there is an initial reduction in opioid distributions in states with recreational marijuana laws, it is conceivable that opioid misuse, addiction, and overdose deaths could decrease. Further research on this topic should be considered and expanded to include more states and to observe the long-term effects of recreational marijuana legalization.

Acknowledgments Data for this paper was obtained from the DEA as reported to ARCOS. The public availability of this data is valued. We could like to thank Iris Johnston at Geisinger for the article collection. No external funding was received for this research.

References 1.

Mularski RA, White-Chu F, Overbay D, Miller L, Asch SM, Ganzini L. Measuring pain as the 5th vital sign does not improve quality of pain management. J Gen Intern Med. 2006;21(6):607–612.

2. Automation of Reports and Consolidated Orders System (ARCOS). DEA Diversion Control Division. Available at: http://www.deadiversion.usdoj.gov/arcos/. Accessed April 21, 2019. 3. Council NR. Relieving pain in America: a blueprint for transforming prevention, care, education, and research. Washington, D.C.: National Academies Press; 2011. 4. Rudd RA, Aleshire N, Zibbell JE, Gladden RM. Increases in drug and opioid overdose deaths-United States, 2000– 2014. Am J Transplan. 2016;16(4):1323–1327. 5. Bradford AC, Bradford WD, Abraham A, Adams GB. Association between US state medical cannabis laws and opioid prescribing in the Medicare Part D population. JAMA Intern Med. 2018;178:667. 6. Marijuana Overview: Legalization. National Conference of State Legislatures. Available at: http://www.ncsl.org/ research/civil-and-criminal-justice/marijuana-overview.aspx. Accessed April 21, 2019.

Prescription Opioid Distribution Before and After Legalization of Recreational Marijuana

Hill KP, Palastro MD. Medical cannabis for the treatment of chronic pain and other disorders: misconceptions and facts. Pol Arch Intern Med. 2017 Nov 30;127(11):785–789.

8. Bachhuber MA, Saloner B, Cunningham CO, Barry CL. Medical cannabis laws and opioid analgesic overdose mortality in the United States, 1999–2010. JAMA Intern Med. 2014;174:1668.

20. Amendments and Propositions on the 2012 Ballot. Available at: https://www.sos.state.co.us/pubs/elections/ Initiatives/ballot/contacts/2018.html. Accessed April 21, 2019. 21. U.S. Census Bureau QuickFacts: United States. Census Bureau QuickFacts. Available at: https://www.census.gov/ quickfacts/fact/map/US/INC110217. Accessed April 21, 2019.

9. Livingston MD, Barnett TE, Delcher C, Wagenaar AC. Recreational cannabis legalization and opioid-related deaths in Colorado, 2000–2015. Am J Public Health. 2017;107 (11):1827–1829.

22. Collins LK, Pande LJ, Chung DY, Nichols SD, Mccall KL, Piper BJ. Trends in the medical supply of fentanyl and fentanyl analogues: United States, 2006 to 2017. Prev Med. 2019;123:95–100.

10. Ren Y, Whittard J, Higuera-Matas A, Morris CV, Hurd YL. Cannabidiol, a nonpsychotropic component of cannabis, inhibits cue-induced heroin seeking and normalizes discrete mesolimbic neuronal disturbances. J Neurosci. 2009;29(47):14764–14769.

23. Olfson M, Wall MM, Liu SM, Blanco C. Cannabis use and risk of prescription opioid use disorder in the United States. Am J Psychiatry. 2018;175(1):47–53.

11. Bradford AC, Bradford WD. Medical marijuana laws may be associated with a decline in the number of prescriptions for Medicaid enrollees. Health Aff. 2017;36 (5):945–951. 12. Wen H, Hockenberry JM. Association of medical and adultuse marijuana laws with opioid prescribing for Medicaid enrollees. JAMA Intern Med. 2018;178 (5):673–679.

24. Bohnert AS, Guy GP, Losby JL. Opioid prescribing in the United States before and after the Centers for Disease Control and Prevention's 2016 opioid guideline. Ann Intern Med. 2018;169:367–375. 25. National Conference of State Legislatures. State Medical Marijuana Laws. Available at: http://www.ncsl.org/research/ health/state-medical-marijuana-laws.aspx. Accessed May 15, 2019.

13. Anthony JC, Warner LA, Kessler RC. Comparative epidemiology of dependence on tobacco, alcohol, controlled substances, and inhalants: basic findings from the National Comorbidity Survey. Exp Clin Psychopharmacol. 1994;2 (3):244–268. 14. Lopez-Quintero C, Pérez de los Cobos J, Hasin DS, Okuda M, Wang S, Grant BF, Blanco C. Probability and predictors of transition from first use to dependence on nicotine, alcohol, cannabis, and cocaine: results of the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC). Drug Alcohol Depend. 2011;115 (1–2):120–130. 15. Anthony JC. The epidemiology of cannabis dependence: cannabis dependence: its nature, consequences and treatment. Cambridge, UK: Cambridge University Press; 2006:58–105. 16. Hall WD, Pacula RL. Cannabis use and dependence: public health and public policy. Cambridge, UK: Cambridge University Press; 2003. 17. National Academies of Sciences, Engineering, and Medicine. The health effects of cannabis and cannabinoids: The current state of evidence and recommendations for research. Washington, DC: The National Academies Press; 2017. 18. Hartig H, Geiger A, Hartig H, Geiger A. 62% of Americans favor legalizing marijuana. Pew Research Center 2018. Available at: https://www.pewresearch.org/facttank/2018/10/08/americans-support-marijuana-legalization/. Accessed April 21, 2019. 19. Kaskie B, Ayyagari P, Milavetz G, Shane D, Arora K. The increasing use of cannabis among older Americans: A public health crisis or viable policy alternative? Gerontologist. 2017;57(6): 1166–1172.

157

Scholarly Research In Progress • Vol. 3, November 2019

2020 Summer Research Immersion Program Each year the Geisinger Commonwealth School of Medicine Summer Research Immersion Program (SRIP) brings together first-year medical students for an opportunity to gain research experience in basic science, clinical science, public/ community health, behavioral health, or medical education under the guidance of a research mentor. The summer research experience includes a $2,500 educational stipend. At the end of the program, students present their research in a poster session. In addition to research, SRIP students participate in a variety of complementary enrichment activities: •

Geisinger Commonwealth and Geisinger faculty research seminars

•

Geisinger Commonwealth Grand Rounds and clinical seminars at our hospital partners

•

Special events or conferences related to your research topic

•

Clinical exposure

•

Scientific writing & communication workshops

SRIP program goals: •

To provide Geisinger Commonwealth medical students with an in-depth research experience under the guidance of a mentor

•

To enhance students’ knowledge of the scope and types of research relevant to improving health in the region, nationally and globally

•

To provide research opportunities that span the translational continuum, from laboratory-based biomedical studies to clinical and public health research conducted with community partners

•

To provide opportunities for students to engage in peer learning and networking

•

To enhance students’ skills in oral and written scholarship

Program dates: SRIP 2020 will be an eight-week program held June 1 through July 24, 2020.

Program deadlines: Application release date: Dec. 2, 2019 Application submission deadline: Feb. 3, 2020 For more information, contact the SRIP director, Elizabeth Kuchinski, MPH (ekuchinski@som. geisinger.edu), or Sonia Lobo, PhD, associate dean for research & scholarship, (slobo01@ som.geisinger.edu).

158