2 Essential Tremor: A Focused Literature Review on its Pathophysiology, Neurophysiology, Etiology, and Management
Carrie A. Bradley, Nyah M. Garrison, Amber S. Washington, and Brandon Wood
10 Dysbiosis of the Gut Microbiota Being a Contributing Factor to Parkinson's Disease: A Literature Review
Maria Gikoska, Thomas S. K. Gilbert, Lucyna A. Wierzbicki, Arabella R. Thompson, and Brian J. Piper
24 The Effects of Antipsychotics on Brain Structure
Emily R. Cranmer and Craig W. Reichard Jr.
32 The Role of Statins in the Prevention and Management of Alzheimer's Disease: A Focused Review
Taylor A. Copelin, Madison L. Kobzeff, Rachelle A. Franco, and Songul M. Gafurova
40 A Narrative Review of the Bifrontal Decompressive Craniotomy in Clinical Practice
Julian Michael Burwell
48 Molecular Insights into Glioblastoma Resistance and Implications for Treatment Advancements: A Systematic Review
Kaustov Chakrabarti, Amanda M. Mohabir, Mahesh Nukala, Sara M. Safiullah, and Tierney R. Woitas
55 Moyamoya Disease: Physiological Mechanisms and Treatment Approaches
Kaustov Chakrabarti
63 Rural EmPATH Unit Decreases Subjective Distress Levels in Patients with Psychiatric Complaints Presenting to the Geisinger Medical Center Emergency Department in Danville, Pa.
Kamil Falkowski, Tanner Thompson, Hunter Yarnell, Mia Gianello, and Jennifer Margaret Yarnell
71 Dopamine and Serotonin Interactions in Schizophrenia: A Focused Review of Mechanisms and Therapeutic Implications
Sara Abdo, Joseph P. Osborne, Tawn A. Tomasi, and Ryder P. Lathrop
84 Group-Based Telehealth Music Therapy Intervention for Patients with Dementia: A Pilot Study
Khevna P. Joshi, Alysha D. Suley, Nicole M. Hooper, Brian J. Piper, Mark V. White, and Maya L. Lichtenstein
93 CRISPR-Cas9 Gene Therapy Effects on Inherited Eye Disorders
Chloe E. Pisack, Krishna S. Patel, and Tyler S. Batista-Nieto
101 Investigating Racial and Ethnic Variation in High Sensitivity C-Reactive Protein Levels Among Individuals with Prediabetes
Jacob George, Savanna Shaw, Kevin Sajan, Alicia Johns, Amanda Young, and Bobbie Johannes
108 Pause and Recharge: Wellness for Medical Students
Cameron Jones, Christopher D. Manko, Lakshmi Ilango, Kevin Xu, Dylan Quinn, and John Pamula
113 Modification of Dialysis for Management of Elevated Intraocular Pressure: A Case Report
Giovanni Battistini and Daniel Upton, MD
116 Glioblastoma’s Effect on Brain Circuitry
Iletou B. Ehinnou and Anthony J. Bragoli
123 Mitral Regurgitation Severity and Echocardiographic Changes at 1 Year Postoperative: A Comparative Study of Surgical Aortic Valve Replacement Versus Transcatheter Aortic Valve Replacement
My N. Nguyen, Tariq Ahmad, and Tyler J. Wallen
133 Predictive Scoring Analysis for Identifying Head and Neck Squamous Cell Cancer
Christopher Lloyd, My Nguyen, and Christian Kauffman
141 Thoracic Kyphosis Correction Using Pre-Bent PatientSpecific Rods in Adolescent Idiopathic Scoliosis: A Retrospective Analysis
Steven J. Grampp, Mark Seeley, and Meagan Fernandez
147 An Evaluation of Child Maltreatment Coding within a Large Health Care System in Rural Pennsylvania
Emily Harasym, Michelle Pistner Nixon, Adam Cook, H. Lester Kirchner, Paul Bellino, and Lisa Bailey-Davis
160 A Serial Cross-Sectional Study of Older Adult Caregivers in the United States from 1997 to 2019
SooYoung HS VanDeMark
171 Afferent Baroreflex Failure Following Neck Radiation: A Case Report
Morgan Glasser and Blake Garmon
175 Paradoxical Acute Angle Closure Following Topiramate Discontinuation: A Case Report
Carolyn Young and Tatiana Franco
183 2026 Summer Research Immersion Program
184 Medical Research Honors Program
185 Finding Your Way: Opportunities for student funding
A message from the editor-in-chief
Welcome to the ninth edition of the Journal of Scholarly Research in Progress (SCRIP)
I continue to be inspired by the resilience and growth our students demonstrate in their pursuit of knowledge and innovation. The plant on the cover is a fitting metaphor, rooted yet reaching, it thrives in conditions that demand adaptability and perseverance. Likewise, our students flourish in an environment that fosters curiosity and supports research. Their success is rooted in the guidance of our dedicated mentors who nurture ideas, provide direction, and help transform questions into meaningful contributions.
This year’s edition reflects the breadth of inquiry at Geisinger College of Health Sciences (GCHS), from understanding the gut-brain axis and neurodegenerative diseases to advancing surgical techniques and improving diagnostic safety. Public health, health equity, and wellness also take center stage with studies addressing rural health care and the well-being of medical students, patients, and providers. Each student’s scholarly work adds to the advancement of knowledge at GCHS and within the broader canopy of medical science.
SCRIP embodies our shared mission to advance health through education, research, and community engagement. Our students are not only developing as compassionate health care professionals and scholars, but also as contributors to a high reliability organization that values safety, continuous learning, and accountability. Thank you to the students, mentors, reviewers, editorial board, and production team for bringing this edition of SCRIP to life. 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 slobo1@geisinger.edu.
Sincerely,
Sonia Lobo, PhD, E-RYT Editor-in-Chief
Editorial Board
John Arnott, PhD
Christian Carbe, PhD
Kathryn Lafferty-Danner, PhD
Elizabeth Kuchinski, MPH
Sireesha Mamillapalli, PhD
Mary M. Pelkowski, MD
Saishravan Shyamsundar, MBS
Student Editors
Co-editor-in-chief
Elyse Guziewicz, MD Class of 2028
Morgan Glasser, MD Class of 2026
Ahmad Higazy, MD Class of 2027
John Finnie-Maloney, MD Class of 2028
Roshni Raj, MD Class of 2028
Marketing, Communications, and Design
Jessica L. Martin, Managing Editor
Geisinger Marketing & Communications
Heather M. Davis, MFA
Director, Marketing & Communications
Elizabeth Zygmunt
Director of Public Relations and Media
Shannon Lesniak
Graphic Designer
Acknowledgments
The SCRIP journal would not be possible without the contributions of the 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.
Tanja Adonizio, MD
Thomas Churilla, MD
Thomas Cronin, MD
Brian J. Piper, PhD
Mathangi Rajaram-Gilkes, MBBS, MSc, MS, MEd
Margrit Shoemaker, MD, FACP
Ying Sung, PhD
Divvy Upadhyay, MD, MPH
Jason Woloski, MD
Office of Research & Scholarship
MSB, Suite 2024, 2nd Floor West
570-504-9662
Sonia Lobo, PhD, E-RYT
Associate Dean for Research & Scholarship
Professor of Biochemistry
Michele Lemoncelli
Program Coordinator, Research & Scholarship
Tracey Pratt, MPH
Manager, Research Education Resources
Laura E. Mayeski MT(ASCP), MHA
Manager, Research Compliance
Essential Tremor: A Focused Literature Review on its Pathophysiology, Neurophysiology, Etiology, and Management
Carrie A. Bradley1*‡, Nyah M. Garrison1*‡, Amber S. Washington1*‡, and Brandon Wood1*‡
1Geisinger
College of Health Sciences, Scranton, PA 18509
*Master of Biomedical Sciences Program
‡Authors contributed equally
Correspondence: cabradley@som.geisinger.edu
Abstract
Essential tremor (ET) is among the most common neurological movement disorders, with incidence rates increasing with age. The disorder is characterized by a bilateral action tremor of the upper limbs, enduring for a minimum of 3 years in duration, and shows early signs of onset with progression over time. Recent terminology has categorized a subcategory of ET, known as essential tremor-plus (ET-plus), which consists of a combination of symptoms that include tremors at rest and additional neurological signs. This includes memory impairment, gait disturbances, or various other neurological symptoms that do not have full clinical significance for diagnosis and suggest a more complex underlying neurological condition. Abnormalities located primarily in the cerebellum, including the loss and damage to Purkinje cells (PC), were core pathological features of cerebellar damage in postmortem ET. Given the complexity and uncertainty of the pathophysiological aspects of tremor generation, treatment options are still limited, and current therapeutic measures lack the capacity to fully alleviate symptoms. In consideration, there are current pharmaceutical, stimulation, and genetic studies being conducted to characterize the etiology and pathophysiology of ET and aid in the reduction of progression and symptoms. In this review, various aspects of ET including neurophysiology, pathophysiology, genetic factors and etiology, treatment options and future directions of advancements for ET management are discussed.
Introduction
Essential tremor (ET) is a prevalent but obscure neurological disorder of motion, increasingly taken into consideration to be something more than a simple
tremor disorder, with excessive costs of misdiagnosis because of symptom overlap, highlighting the mission of decoding its mechanisms (1). Despite limitations and complexities, ET was first described by Pietro Burresi in 1874 and is characterized as an involuntary, rhythmic shaking that can occur in the arms, legs, head, and torso (1-5). ET is often recognized as a hereditary condition, inherited as an autosomal dominant trait, and typically occurs without any other associated neurological condition (1,6). According to the Tremor Task Force of the International Parkinson’s and Movement Disorders Society, ET is defined as an isolated tremor syndrome involving tremors of both upper limbs that persists within 3 years, independent of movement in lower limbs (7–11). ET can be classified into various subtypes, such as action, postural, and kinetic tremor based on the affected body region, and the type of tremor being displayed, which could be constant or periodic overtime (3, 12, 13, 14). The region of the brain affected by ET is the cerebellum, responsible for movement and coordination (3). Research, although limited to clinical studies, neuroimaging and postmortem analysis (15), has revealed cerebellar abnormalities associated with ET that includes changes to Purkinje cell (PC) axons and dendrites, displacement and loss of PCs, changes to basket cell axonal processes, abnormal distribution of climbing fiber connections to PCs and changes in gamma-aminobutyric acid receptors in the dentate nucleus (15, 16). In addition, ET has been linked to causing subtle morphologic changes also to the brain stem, basal ganglia, frontal lobes, inferior olivary nucleus and thalamus (9, 17). Despite these detailed findings, crucial in understanding the structural-functional changes of the cerebellum region associated with ET, the exact cause of ET remains unknown (2, 6, 18–20). In the United States, ET has
affected approximately 10 million individuals, slightly influencing men more than women (2, 4, 8, 19, 21). This benign, progressive disorder affects both men and women of all ages. The average onset age is between 10 to 19 in younger individuals and 50 to 59 for older adults but is more frequently observed in the older population (10, 12, 14, 18, 41). There have been clinical manifestations in the elderly, but ET also appears in children and adolescents (17). Currently, ET diagnosis is based on a physician’s physical examination. The hallmark symptom observed is an involuntary, kinetic tremor, which manifests when the patient moves his or her hands or other upper extremities. Investigating the patient’s family history also helps in accurately diagnosing ET (17). Although there is no evidence of life-threatening effects of ET, it can impact individuals’ quality of life and increase the risk of developing Parkinson‘s disease (PD) (22).
The understanding of ET is still evolving due to limited studies (16), substantial uncertainties and ongoing debates concerning its genetic makeup, neurophysiology, pathophysiology, and etiology. One area of ongoing discussion is the concept of the term essential tremor-plus (ET-plus) (15). According to the Consensus Classification of Tremor 2018, ET-plus is deemed a subcategory of ET. It is characterized as individuals with ET, but they also present other neurological signs, not just action and postural tremors. These other neurological conditions include parkinsonism, ataxia, cognitive changes, and dystonia (15, 22–25). The proposal of this term was brought about to better categorize patients and to distinguish ET cases from those with additional neurological signs and determine whether there are currently any underlying pathological differences (15). The controversy raises questions as to whether ET is its own separate entity, or if there could be a broad spectrum of subcategories related to the disorder. The uncertainty surrounding ET, along with the importance of each study, demonstrates the need for further research to create a better understanding of the disorder’s complexity; with the aim to explore its causes, clarify diagnostic criteria, propose more future studies, investigate more treatment options/therapies and possible related conditions (26).
Methods
Peer reviewed literature research on ET was identified using key terms including neurophysiology, treatment, pathophysiology, parkinsonism, and genetics. Articles
used in this literature review were obtained from PubMed, Google Scholar, and Science Direct and compiled into a detailed summary describing the etiology and pathophysiology of ET, as well as potential connections between PD and ET. Our literature research was limited to only the English language. There were no limitations on article publication dates. To distinguish between ET and PD, we excluded PD from our search criteria. MeSH terms were used in the PubMed search.
Discussion
Etiology
Although the exact origin of ET remains unknown, it is suggested that a combination of genetic, nonpenetrance, and non-autosomal dominant factors play a role in ET development (6,7). While studies on monozygotic twins have provided insight (7, 27–29), and other researchers have observed an autosomaldominant pattern of inheritance (6, 7, 29, 26), it is still unknown how individuals can be affected since the variability in how ET manifests between generations and person is still not fully recognized (6, 29). The cardinal symptom of ET is defined as kinetic tremor, which manifests when moving one’s hands but is less noticeable when resting (6–7). While kinetic tremor worsens as patients age, medicine or stress can exacerbate symptoms (7). Genetic or environmental factors contribute to the etiology of the disease. First degree relatives have a higher likelihood of developing ET than the general population (6–7). Typically, symptoms increase appreciably if the disease starts at an early age. A cohort study sought to identify the gene that could cause ET (30). Results revealed that out of 16 family members, 10 had the disease while 6 were unaffected (30). A TUB p.V431I variant (rs75594955) was consistent with autosomal-dominant inheritance (30). Exome resequencing of TUB in 820 unrelated patients with sporadic ET and a control group of 620 patients demonstrated nonsynonymous TUB variants (e.g. rs75594955: p.V431I, rs1241709665: p.Ile20Phe, rs55648406: p.Arg49Gln.) were directly involved in the pathogenesis of this disease (30). TUB, a member of the Tubby family of transcription factors, are primarily expressed in neurons in the cerebellum and, via thyroid signaling and G-protein coupled receptor signaling, regulates the pathways responsible for neuron function during development and postdifferentiation (30–31). Moreover, TUB protein
modulates dopaminergic and cholinergic pathways, 2 pathways associated with the tremor phenotype (30).
Pathophysiology and Neurophysiology
ET is believed to originate in the brainstem or cerebellum and is known to worsen under stressful situations. ET is also considered a risk factor of PD with physical symptoms becoming more prominent when patients extend both arms in front of themselves (15, 18). Recent postmortem studies have confirmed the involvement of several brain structures in the development of ET: olive nuclei, cerebellum, red nucleus, thalamus, and cerebral cortex (7, 9, 15, 18, 32). Furthermore, 3 main hypotheses are proposed as key contributors in the pathogenesis of ET: 1) Neurodegenerative hypothesis; 2) Central oscillatory network hypothesis; and 3) The GABAergic hypothesis (33). Underlying mechanisms that may relate to oscillatory activity in the cortico-olivo-cerebellothalamic circuit and its relationship with ET, may be a result of disruption in GABAergic neurotransmission causing dysfunction and the neurodegeneration of cerebellar structures (34). Additionally, neurological damage and dysfunction in the cerebellum, a critical region in movement control, ultimately results in involuntary shaking and tremors.
There are 2 major pathologies of ET: one involving brainstem Lewy bodies that are restricted to the locus coeruleus, and the other, more frequent pathology consisting of either the loss of PCs, or a presence of ovoid enlargement of the proximal part of PC axons and heterotopias (33). As more research is published, due to the observed projection of locus coeruleus noradrenergic neurons into the cerebellum and synapsing with PCs along with the apparent outflow from the cerebellum being the final common pathway of the disease, it is likely that the neurodegeneration of the cerebellum is the cause of ET (32, 33).
The tremor stability index (TSI) is a quantitative measure taken by accelerometry (35). Accelerometry differentiates ET from other tremor syndromes with a 95% sensitivity and a 90% accuracy. The criteria include: 1) Tremor frequency of 5-15 Hz; 2. A peak dispersion equal to or below 2.5 Hz; 3. Spectral coherence higher than 80%; 4. No unilateral tremor; and 5. Action amplitude greater than resting amplitude. Another way to differentiate ET from PD is that the TSI index is higher in patients with ET while the PD-tremor index is lower (36).
Biomarkers
Uric acid, a metabolite known to be a natural antioxidant believed to have a protective role in neurodegenerative disease, was lower in patients with ET in comparison to control patients (28, 37–38). Studies have demonstrated low levels of uric acid in patients with PD, Alzheimer’s disease, and amyotrophic lateral sclerosis (28, 37–38). However, the levels found were not high enough to serve a protective role in neurodegeneration. Interestingly, sporadic cases showed a correlation between lower levels of uric acid and higher age (37–38). Thus, uric acid can serve as a potential indicator of neurodegeneration.
While cutting-edge technologies such as long-read sequencing (LRS), next-generation sequencing (NGS), and gene editing technologies have improved our understanding of genetic mutations, precise targeting is still necessary to identify disease-causing variants in ET and similar conditions (39). In addition to its potential to correct genetic defects related to ET, CRISPR-Cas9, a powerful DNA editing tool, is currently in the preclinical phase. It has not yet been used clinically in treatment with amplification-free approaches (No-Amp) or next generation sequencing platforms (40–41). CRISPR has been extensively researched for genetic disorders with clearer causes, such as Huntington’s disease (HD) and spinocerebellar ataxias (SCAs), but further investigations are needed before it can be used in ET (42). As opposed to HD or SCA, ET involves multiple genes and environmental factors, making it more difficult to target.
Brain Imaging
Magnetic resonance imaging (MRI) has been proposed as a means of identifying the neuromelanin inside the neuron of the substantia nigra in patients with ET (37). While this signal remains the same in ET patients, the neuromelanin-sensitive MRI does identify a decrease in PD patients. Other sources state that with a specificity of 96.2% and a sensitivity of 84.4%, dopamine transporter (DAT) has been shown to be a useful protein in differentiating ET from tremor dominant PD (37, 42). However, position emission tomography is expensive and uses radiation. Therefore, MRI is viewed as the preferred neuroimaging procedure for distinguishing the 2 motor diseases (37). Neuromelanin has noticeable paramagnetic properties on the MRI called neuromelanin-sensitive MRI (NM-MRI) (11, 37).
This has demonstrated a decrease in PD patients while remaining the same in ET patients (11, 37).
Misdiagnosis
There is an elevated risk of misdiagnosis between ET and PD due to overlapping symptoms similarities in the display of tremors that may appear in the initial stages; however, there were found to be some noteworthy differences (43–44). A 2006 study that examined misdiagnosis of PD and ET found that out of 71 patients, 26 (37%) were initially misdiagnosed (false ET). ET typically involves bilateral action tremors unlike PD which typically features unilateral resting tremors with bradykinesia and rigidity (10, 15, 45–46). With some cases, ET and PD may coexist within a single case, thus adding to the complexity (45). Specific tremor features (e.g., frequency, pattern) may help distinguish patients with these two conditions if they are compared to associated neurological findings in the laboratory. Misdiagnosis can lead to inappropriate treatment, as antiparkinsonian medications may be prescribed incorrectly (44). Differentiation is crucial to achieving the best possible outcomes.
Treatments: Deep Brain Stimulation, Pharmaceutical, and Therapeutic Measures
Current treatments of ET include pharmaceutical interventions, physical and occupational therapy, and deep brain stimulation (DBS) alleviate the severity of tremor amplitudes and frequency (9, 18, 35–36, 44, 47–54). Beta-adrenergic antagonists such as propranolol have evidence of weakening tremors through peripheral sites; however, beta-2 affinity agents may show higher benefits in reducing symptoms (49). Commonly used medications such as propranolol, primidone, and topiramate are useful oral medications to reduce tremors; although, given their mechanism of action, these medications do not fully attenuate ET symptoms (49). Propranolol is a cardiac medication used to slow heart rate and reduce high blood pressure, yet it has been found to reduce unrelated symptoms such as shaking and sweating, making it an accepted treatment option for mild tremors (50–51). While propranolol reduces shaking, more effective and successful options for tremor reduction include GABA inhibitors and calcium channel blockers. Drugs such as clonazepam (a benzodiazepine) and flunarizine (slow channel calcium blocker) could be far more beneficial (50–51). Benzodiazepines increase inhibitory GABA effects,
resulting in muscle relaxation which, with further evaluation and research, can be a more targeted medication option for ET symptoms (50). Additionally, calcium channel blockers have been suggested as alternative treatment as they work by blocking calcium channels through the control of calcium ions into cells (initiating and regulating biological processes), a key contributor in the role of muscle control and nerve function (50–51). Calcium channel blockers can mitigate overactive signaling that is a contributor to tremors (50–51). Flunarizine, as previously mentioned, is a calcium channel blocker that can cross the bloodbrain barrier, affecting the neurons in the brain directly, making for an efficient tremor medication (51). This suggests that the Flunarizine, by inhibiting both calcium dependent and independent release of glutamate, could modulate the cortico-olivo-cerebellothalamic circuit excitably for conditions like ET and PD (50–51). Furthermore, GABA inhibitors and calcium channel blockers are currently only used in research and additional studies need to be done to approve the drugs for clinical use (51).
DBS is one of the most-used treatments for movement disorders that is approved by the U.S. as a treatment for PD, dystonia and tremors (18, 35, 52–55). DBS uses electrical stimulation through surgical implantation of electrodes. Electrodes are implanted bilaterally into deep structures of the brain, most commonly in the thalamus, which are connected to a pulse generator through wires. The pulse generator is placed in the chest wall, sending continuous electrical impulses to the electrodes implanted in the ventral intermediate nucleus of the thalamus (35, 52–53). DBS works to deliver targeted electrical impulses to motor control regions, modulating abnormal neural activity and reducing tremors. DBS is more suitable for patients with more severe tremors who typically do not respond well to oral medications, with ventral intermediate thalamic DBS relieving tremors within seconds (53). While DBS is considered a safe procedure (53), side effects such as intracranial hemorrhage, infections, and problems with implanted hardware devices should be considered prior to treating the patient. DBS does not cure movement disorders, but can alleviate symptoms. Recent studies have shown a 50%–90% decrease in tremors and improvement in mobility and functionality in patients who have undergone DBS (18, 55).
A more non-invasive route to treating ET involves therapeutic exercises used to moderate symptoms
involving occupational therapy, speech therapy, or psychotherapy (26). Various therapy options allow for the assessment needed to determine specific daily tasks that are typically difficult for patients on a case-by-case basis. By determining an individualized care plan, occupational therapy can be incorporated to provide the skills needed to manage tremors and learn to function more comfortably. Occupational therapy utilizes interventions such as weighted utensils, hands-free electronic devices, and dressing options such as Velcro sneakers (26, 47). Psychotherapy aids in the overall improvement of an individual’s mental health with individuals impacted by the disorder (26). Overall, while there are no clinically approved cures for ET, there is a wide variety of treatment and symptom management options available to reduce the struggles of living with tremors. Furthermore, ET is an active research area which will ensure further advancements in pharmaceutical and therapeutic treatments for ET (26).
Conclusion
ET is a chronic, autosomal dominant progressive neurological disorder that primarily impacts the basal ganglia (1, 2, 6). However, recent research indicates involvement of the cerebellum and inferior olivary nucleus (3–4, 7, 9, 14–15). The condition is characterized by involuntary, rhythmic tremors, which often worsen with age and are exacerbated by actions like lifting against gravity. This disorder is most observed in older adults (10, 12, 18, 44). Recent advancements in neurophysiology, genetics, and imaging have illuminated ET’s potential relationship to cerebellar degeneration and its connections to other neurodegenerative diseases, such as PD (11, 14, 30, 53). The broader implications of ET and its various presentations have spurred ongoing debate and research. The broader implications of ET and its different forms continue to be debated and studied. The “ET-plus” classification, which categorizes ET patients with extra neurological symptoms, is controversial (15, 22, 32). A better understanding of ET’s pathogenesis and links to other neurodegenerative conditions, like PD, could be gained by identifying biomarkers, like uric acid, and genetic variants associated with them (4, 28, 30, 38). To enhance diagnostic accuracy and differentiate ET from other tremor-related disorders, including PD, advances in imaging techniques, such
as neuromelanin-sensitive MRI, should be further explored (40). Furthermore, it is imperative that future research explores the efficacy of treatments such as DBS and new pharmacological approaches, especially in patients with ET and overlapping PD features (16, 18, 29, 53–55). Developing more effective and personalized treatment strategies will depend on understanding how ET progresses, especially when it evolves into PD or other neurological disorders (17–18, 42, 47). Interdisciplinary collaborations are essential to understanding ET’s complex relationship with other neurodegenerative processes.
The development of biomarkers for the early detection of ET needs to be a priority area of further research. Further studies are required to better understand ET’s genetic basis, refine diagnostics, and develop more effective treatments beyond current options like DBS and symptom management (26, 54). For accurate diagnosis of genetic disorders, targeted sequencing remains essential despite advancements in longread sequencing (LRS), next-generation sequencing (NGS), and gene editing technologies. In addition, a better understanding of how environmental factors impact ET outside of genetics could allow preventative measures to be developed to reduce its risk.
Disclosures
The authors have no conflicts of interest.
Acknowledgments
We thank Dr. Brian Piper for his assistance in editing and revising this manuscript. His expertise and continuous support enhanced the quality of our work. We appreciate the time and effort he dedicated to this manuscript.
References
1. Elble, R.J. What is Essential Tremor? Curr Neurol Neurosci. 2013; 13:353.
14. Cleveland Clinic Health Library. Essential Tremor: Symptoms, Causes, Treatment and Prevention. Cleveland Clinic. 2022.
15. Gionco JT, Hartstone WG, Martuscello RT, Kuo S-H, Faust PL, Louis ED. Essential tremor versus “et-plus”: A detailed postmortem study of cerebellar pathology. Cerebellum. 2021. 25;20(6):904–12.
16. Louis, E.D., Faust, P.L. Essential tremor pathology: Neurodegeneration and reorganization of neuronal connections. Nat Rev Neurol. 2020;16:69–83.
17. Klaming R, Annese J. Functional anatomy of essential tremor: Lessons from neuroimaging. AJNR Am J Neuroradiol. 2014;35(8):1450-7.
18. Van den Berg KRE, Helmich RC. The role of the cerebellum in tremor – evidence from neuroimaging. Tremor Other Hyperkinet Mov (N Y). 2021;11:49.
19. Baltuch G. Essential Tremor Affects Millions, But Few Know the Treatment Options. Columbia University Irving Medical Center. 2023.
20. Reich SG. Essential Tremor. Med Clin North Am 2019;103(2):351-356.
21. Pan MK, Li YS, Wong SB, Ni CL, Wang YM, Liu WC, et al. Cerebellar oscillations driven by synaptic pruning deficits of cerebellar climbing fibers
contribute to tremor pathophysiology. Sci Translat Med. 2020;12(526): eaay1769.
22. National Tremor Foundation. Information and advice for patients Essential Tremor. Sandwell and West Birmingham Hospital NHS. 2012.
23. Rajalingam R, Breen DP, Lang AE, Fasano A. Essential tremor plus is more common than essential tremor: Insights from the reclassification of a cohort of patients with lower limb tremor. PD Relat Disord. 2018; 56:109-110.
24. Deuschl G, Bain P, Brin M. Consensus statement of the movement disorder society on tremor. Ad Hoc scientific committee. Mov Disord.1998; 13:2-23.
25. Bhatia KP, Bain P, Bajaj N, Elble RJ, Hallett M, Louis ED, et al. Consensus Statement on the classification of tremors. From the Task Force on Tremor of the International Parkinson and Movement Disorder Society. Mov Disord 2018;33(1):75-87.
26. Hopfner F, Deuschl G. Managing essential tremor. Neurotherapeutics. 2020;17(4):1603-21.
27. Santagata S, Boggon TJ, Baird CL, Gomez CA, Zhao J, Shan WS, et al. G-protein signaling through tubby proteins. Science. 2001;292(5524):2041-50.
28. Çelık RGG, Köksal A, Şahın B, Şen A, Sakalli NK, Nalbantoğlu M. The relationship between serum uric acid levels and clinical features in essential tremor. Noro Psikiyatr Ars. 2020;57(1):33-6.
29. Pastor P, Deuschl G, Blawundreat C. Genetics of essential tremor. Int PD Mov Disord Soc. 2022;1:1.
30. Sailani MR, Jahanbani F, Abbott CW, Lee H, Zia A, Rego S, et al. Candidate variants in TUB are associated with familial tremor. PLOS Genet. 2020;4(2):1-3.
31. Lorenz D, Frederiksen H, Moises H, Kopper F, Deuschl G, Christensen K. High concordance for essential tremor in monozygotic twins of old age. Neurology. 2004;62(2):208-11.
32. Bove F, Di Lazzaro G, Mulas D, Cocciolillo F, Di Giuda D, Bentivoglio AR. A role for accelerometry in the differential diagnosis of tremor syndromes. Funct Neurol. 2018; 33(1):45-9.
33. Louis ED, Faust PL, Vonsattel JP, Honig LS, Rajput A, Robinson CA, et al. Neuropathological changes in essential tremor: 33 cases compared with 21 controls. Brain. 2007;130(12): 3297-307.
34. Zhang X, Santaniello S. Role of cerebellar GABAergic dysfunctions in the origins of essential tremor. Proc Natl Acad Sci U S A. 2019;116(27):13592–601.
35. Lawson McLean A, Schwarz F. Directional deep brain stimulation in Parkinson’s disease and essential tremor: A retrospective analysis. Neuromodulation. 2024 ;27(3): 589-50.
36. Tarakad A, Jankovic J. Essential tremor and Parkinson’s disease: Exploring the relationship. Tremor Other Hyperkinet Mov. 2019;8:589.
37. Sepulveda MSC, Fasano A. Essential tremor: New advances. Clin PD Relat Disord.2020;3: dev100031.
38. Lee CY, Yoon JH, Park HJ, Hong YJ, Sunwoo KM, Kang SY. Uric acid level may not be reduced in essential tremor. Int J Neuros. 2018; 128(12): 1163-7.
39. Steffen JK, Jergas H, Petry-Schmelzer JN, Dembek TA, Thies T, Jost ST, et al. Thalamic deep brain stimulation in essential tremor plus is as effective as in essential tremor. Brain Sci. 2020;10(12):970.
40. Jiang H, Tang M, Xu Z, Wang Y, Li M, Zheng S, Zhu J, Lin Z, et al. CRISPR/Cas9 system and its applications in nervous system diseases. Genes & Diseases. 2024;11(2):675-86.
41. Marsili L, Duque KR, Bode RL, Kauffman MA, Espay AJ. Uncovering essential tremor genetics: The promise of long-read sequencing. Front Neurol. 2022;13:821189.
42. Jin L, Wang J, Wang C, Lian D, Zhou Y, Zhang Y, et al. Combined visualization of nigrosome-1 and neuromelanin in the substantia nigra sing 3T MRI for the differential diagnosis of essential tremor and de novo Parkinson’s disease. Front Neurol 2019;10:100.
43. Wang J, Huang Z, Li Y, Ye F, Wang C, Zhang Y, et al. Neuromelanin-sensitive MRI of the substantia nigra: An imaging biomarker to differentiate essential tremor from tremor-dominant Parkinson’s disease. Parkinsonism Relat Disord 2019;58:3-8.
44. Montgomery EB Jr, Baker KB, Lyons K, Koller WC. Motor initiation and execution in essential tremor and Parkinson’s disease. Mov Disord. 2000;15(3):511-5.
45. Thenganatt MA, Louis ED. Distinguishing essential tremor from Parkinson’s disease: bedside tests and laboratory evaluations. Expert Rev Neurother. 2012;12(6):687-96
46. Meara J, Bhowmick BK, Hobson P. Accuracy of diagnosis in patients with presumed Parkinson’s disease. Age Ageing. 1999;28(2):99-102.
47. Shahed J, Jankovic J. Motor symptoms in Parkinson’s disease. Handb Clin Neurol. 2007;83:329-42.
48. Baizabal-Carvallo JF, Morgan JC. Drug-induced tremor: clinical features, diagnostic approach, and management. J Neurol Sci. 2022;435:120192.
49. Haubenberger D, Hallett M. Essential tremor: clinical perspectives and pathophysiology. N Engl J Med. 2018; 378(19), 1802:1810.
50. Spagnam, S., Ferng, A., Chou, K. Evaluation and management of tremor. Primary Care. 2024;51(2), 253:267.
51. Pan M-K, Kuo S-H. Essential tremor: Clinical perspectives and pathophysiology. J Neurol Sci 2022;435:120198.
52. Cousin MA, Nicholls DG, Pocock JM. Flunarizine inhibits both calcium-dependent and -independent release of glutamate from synaptosomes and cultured neurones. Brain Research. 1993;606(2):227–36.
53. Neudorfer C, Kultas-Ilinsky K, Ilinsky I, Paschen S, Helmers A-K, Cosgrove GR, et al. The role of the motor thalamus in deep brain stimulation for essential tremor. Neurotherapeutics. 2024 ;21(3): e00313.
54. Herrington TM, Cheng JJ, Eskandar EN. Mechanisms of deep brain stimulation. J Neurophysiol. 2016;115(1):19–38.
55. Hopfner F, Haubenberger D, Galpern WR, Gwinn K, Van’t Veer A, White S, et al. Knowledge gaps and research recommendations for essential tremor. Parkinsonism Relat Disord. 2016;33:27-35.
56. Sturchio A, Marsili L, Mahajan A, Grimberg MB, Kauffman MA, Espay AJ. How have advances in genetic technology modified movement disorder nosology? Euro J Neurol. 2020;27(8):1461-70.
57. Tanner CM, Goldman SM, Lyons KE, Aston DA, Tetrud JW, Welsh MD, et al. Essential tremor in twins: an assessment of genetic vs environmental determinants of etiology. Neurology 2001;57(8):1389-91.
58. Louis ED. The primary type of tremor in essential tremor is kinetic rather than postural: Crosssectional observation of tremor phenomenology in 369 cases. Euro J Neurol. 2013;20 (4):725-727.
59. Fernandez HH, Galvez JN, Cooper S, Bowes Surgical considerations for tremor and dystonia. Cle Clin J Med. 2012;79 (7):1-5.
60. Siegal A, Sapru H. Essential Neuroscience. 4th ed. Philadelphia (PA): Walters Kluwer. 2019.
61. Nazzaro JM, Lyons KE, Pahwa R. Deep brain stimulation for essential tremor. Handb Clin Neurol. 2013;116:155-166.
62. Tamas G, Pálvölgyi L, Takats A, Szirmai I, Kamondi A. Movement-related beta responses in essential tremor and Parkinson’s disease. Ideggyogy Sz 2006;59(11-2):417-424.
63. Saad M, Hefner Sofia, Donovan S, Bernhard D, Tripathi R, Factor SA, et al. Development of a tremor detention algorithm for use in an academic movement disorders center. Sensors (Basel). 2024;24(15):4960.
64. Jain S, Lo SE, Louis ED. Common misdiagnosis of a common neurological disorder: how are we misdiagnosing essential tremor? Arch Neurol 2006;63(8):1100-4.
65. Kumar S, Goyal L, Singh S. Tremor and rigidity in patients with Parkinson’s disease: Emphasis on epidemiology, pathophysiology and contributing factors. CNS Neurol Disord Drug Targets. 2022;21(7):596-609.
66. Lohmann K, Klein C. Next generation sequencing and the future of genetic diagnosis. Neurotherapeutics. 2014;11(4):699-707.
Dysbiosis of the Gut Microbiota Being a Contributing Factor to Parkinson's Disease: A Literature Review
Maria Gikoska1*‡, Thomas S. K. Gilbert1*‡, Lucyna A. Wierzbicki1*‡, Arabella R. Thompson1*‡, and Brian J. Piper1,2
1Geisinger College of Health Sciences, Scranton, PA 18509
²Center for Pharmacy Innovation and Outcomes, Geisinger Precision Health Center, Forty Fort, PA 18704
*Master of Biomedical Science Program
‡Authors contributed equally
Correspondence: mgikoska@som.geisinger.edu
Abstract
Parkinson’s disease (PD) is a prevalent neurodegenerative disorder characterized by symptoms such as tremors, dementia, and bradykinesia. Early non-motor signs, including constipation and hyposmia, suggest a potential connection between PD and the gut. The loss of dopaminergic neurons in the substantia nigra is a hallmark of PD, with unknown causes. While current treatments provide symptom relief, there are no therapies to slow the progression of the disease. The gut microbiota plays a role in inflammatory and immune responses and has been implicated in neurological disorders. The connection between gut microbiota and neurodegenerative disease remains unclear. In this review, we explored the potential link between gut microbiome dysbiosis in PD, focusing on constipation, an early disease symptom. Our review examined the evidence suggesting that chronic intestinal inflammation may contribute to the development of PD and implicate further connections to other neurodegenerative diseases.
Introduction
Parkinson's disease (PD) comes second to Alzheimer's disease as one of the most prevalent neurodegenerative disorders in the United States (U.S.). PD is characterized by notable symptoms such as tremors, dementia, and bradykinesia. However, the early unique non-motor signs are hyposmia and gastrointestinal disorders, such as constipation, which could indicate a potential for this disease to arise outside the central nervous system (CNS) (1). The pathology of PD is characterized by the loss of dopaminergic neurons in the substantia nigra, and the precise causes are not yet known. Current treatments alleviate PD symptoms; the most effective
is carbidopa-levodopa, but no currently available therapies slow the progression of PD-related neurodegeneration (2). Levodopa (L-Dopa) is a natural chemical converted to dopamine in the brain. The drug must be combined with carbidopa to ensure that levodopa would not convert to dopamine outside of the CNS to ensure proper efficacy.
Our literature review investigated the potential contribution of gut microbiome dysbiosis to PD, with constipation being one of the early-onset symptoms. The dysregulation in the composition of the gut microbiota and their metabolites can lead to intestinal inflammation and disrupt the communication and integrity of the blood-brain barrier (BBB). Specifically, gut dysbiosis is believed to be a potential cause of developing a "leaky gut," leading to alterations in the BBB and subsequent neuroinflammation (3). IFNγ and TNFα have shown a correlation with the altered gut microbiome among the various cytokines. The elevated levels of proinflammatory cytokines and chemokines, such as IL-6, TNF, IL-1B, and IL-2, may indicate peripheral manifestations of PD, leading to the deterioration of the intestinal barrier.
Consequently, this allows for systemic exposure to bacterial products like lipopolysaccharides (LPSs) found in the outer membrane (3, 4). This can induce an acute inflammatory response and trigger the production of several inflammatory cytokines in various cell types. In this context, Zonulin and alpha1-antitrypsin are 2 factors that contribute to the promotion of increased gut permeability. α-synuclein (αSyn) is a protein in PD that affects neuronal function and communication by damaging dopaminergic neurons responsible for movement. The propagation of αSyn from the gut to the brain is thought to contribute to the development of PD. Due to intestinal inflammation and increased gut permeability, the
accumulated αSyn moves from the enteric nervous system (ENS) to the CNS through the vagus nerve, leading to synucleinopathies impacting motor function (2, 5, 6, 7, 8).
Methods
Electronic searches were conducted using the following databases: PubMed, ScienceDirect, and the National Institute of Health (NIH) online library. Additionally, reference lists of selected studies were used as a source. The search strategy included keywords such as “Parkinson’s disease,” “Immunology,” “gut microbiome dysbiosis,” and “gut microbiome.”
Searches were conducted from April to May of 2023. All identified articles were cross-referenced among group members to ensure comprehensive coverage and minimize redundancy. Some articles were excluded from our review due to redundancy, limited accessibility, outdated information, or lack of scientific rigor. Redundant studies presented findings that closely mirrored those of the papers we analyzed, offering no additional insight. Articles behind paywalls were also excluded if no alternative access was available, as this restricted our ability to thoroughly evaluate their content. Additionally, studies with outdated data or conclusions that had been refuted by more recent research were not included to ensure our review reflected the most current understanding of the topic. Furthermore, articles with small sample sizes, weak methodologies, or insufficient statistical analyses were unfavorable, as they lacked the robustness necessary to contribute meaningful evidence. Studies that did not focus specifically on gut microbiota’s role in PD or that examined unrelated neurodegenerative disorders without clear relevance were also excluded. This selection process ensured that only the most relevant, high-quality, and evidencebased articles were considered in our review.
Discussion
The gut-brain axis represents a reciprocal communication pathway linking the gastrointestinal (GI) tract and the CNS. Notably, the vagus nerve plays an essential role in establishing this connection. Ongoing investigations have proposed that the gut microbiota influences the axis, potentially contributing to the pathogenesis of PD. Experimental studies utilizing rodent models have elucidated the ability of αSyn to disseminate from the gut to the brain via the vagus nerve (8, 9, 10). Furthermore, cohort studies
have revealed a reduced incidence of PD among individuals who have undergone vagotomy. These findings underscore the alleged involvement of the gut-brain axis in the etiology and progression of PD.
The gut microbiota, comprising various microorganisms, is crucial in maintaining normal physiological function and impacting human health. Age, diet, genetics, antibiotics, probiotics, and stool transplants can alter the gut microbiota composition (11, 12). A typical early symptom of PD, constipation, contributes to gut dysbiosis. Pivotal bacteria such as Prevotella, Akkermansia, short-chain fatty acid (SCFA)-producing bacteria, and Bacteroides in the gut microbiota support the pathogenesis of PD. Decreased SCFA levels can increase endotoxin and neurotoxicity, affecting microglial activation and neuronal function (9, 10, 11, 12, 13, 14).
Bacteroides species may contribute to inflammation in PD, while Akkermansia's increased levels can disrupt intestinal integrity and promote αSyn formation (10, 14). Reduced levels of Prevotella may lead to gut inflammation and oxidative stress, triggering neuroinflammation and neurodegeneration (15). As manifested by their associations with disease duration, cognitive impairment, and potential implications for therapeutic advancements, these findings highlight the gut microbiota dysbiosis as a potential biomarker and therapeutic target for PD.
Lastly, researchers are investigating the potential of diets, such as the Mediterranean diet (MEDI) and ketogenic diets, as an intervention for individuals with PD (16, 17). Probiotic dietary supplementation and fecal microbial transplantation (FMT) have shown initial promise in clinical trials, particularly in alleviating PD-related symptoms (18, 19). Overall, these interventions can potentially manage PD symptoms and warrant further investigation in clinical settings.
Dopamine in PD
PD is caused by the loss of dopaminergic neurons in the substantia nigra, a midbrain dopaminergic nucleus, resulting in decreased dopamine production (20). The substantia nigra is involved in the control of movement, cognition, and the limbic system (20). The substantia nigra is divided into the pars compacta and pars reticulata (20). The pars compacta is the region of the substantia nigra that contains dopaminergic neurons (20). The pars reticulata
consists of GABAergic neurons (20). The substantia nigra pars compacta appears dark due to the elevated levels of neuromelanin pigment within the cell bodies of the dopaminergic neurons (20). The nigrostriatal pathway, extending from the substantia nigra to the putamen, contributes to the motor deficits associated with PD (20).
Dopamine synthesis begins with the precursor L-dihydroxyphenylalanine or L-Dopa, which can be synthesized via a direct or indirect pathway (21). Following a direct synthesis pathway, L-Dopa can be synthesized from tyrosine, a non-essential amino acid (21). The indirect synthesis of L-Dopa utilizes phenylalanine, an essential amino acid, which will be the pathway discussed for synthesizing dopamine (21). L-phenylalanine is converted to L-tyrosine by the enzyme phenylalanine hydroxylase in the liver. Phenylalanine hydroxylase utilizes oxygen, iron, and tetrahydrobiopterin as cofactors, essential for catalysis (21). L-tyrosine is then transported to the dopaminergic neurons in the brain via an active transport mechanism (21). Once in the brain, L-tyrosine is converted into L-Dopa by hydroxylation of the phenol by tyrosine hydroxylase. A further decarboxylation reaction then converts L-Dopa into 3,4-dihydroxy phenethylamine (dopamine) via the enzyme L-3,4-dihydroxyphenylalanine decarboxylase at the pre-synaptic terminal (21). Dopamine metabolism follows a few various pathways. One pathway resulting in dopamine metabolism is the conversion of dopamine to 3-methoxytryramine catalyzed by catechol-O-methyl transferase or COMT (21). Monoamine oxidase (MAO) then converts 3-methoxytryramine to 3-methoxy-4hydroxyacetaldehyde. Aldehyde dehydrogenase then oxidizes the aldehyde in 3-methyoxy-4hydroxyacetaldehyde into a carboxylic acid yielding homovanillic acid (21).
Deficits in dopamine release involve several mechanisms that include defective synaptic vesicle exocytosis, impaired dopamine synthesis, and defective dopamine release (22). The release of dopamine is facilitated by calcium-dependent exocytosis. In a normally functioning dopaminergic axon, the SNARE complex is formed on the plasma membrane (22). The formation of the SNARE complex on the plasma membrane allows for direct interaction with PD proteins such as αSyn and LRRK2, resulting in changes in the expression of the SNARE complex (22). LRRK2-dependent phosphorylation of the Snapin
protein coding gene inhibits SNAP25, which results in defective exocytosis of dopamine (22). Mutations in these complexes lead to mitochondrial dysfunction, leading to insufficient axonal energy to produce an action potential and resulting in synaptic loss. This synaptic loss disrupts the release of dopamine in dopaminergic neurons (22). In models with PD mutations in the SNCA gene, which encodes for αSyn inhibit the release of dopamine. The aggregation of αSyn in the synapse results in impaired dopamine release by blocking protein function (22).
αSyn in Parkinson’s
αSyn is a neuronal protein encoded by the SNCA gene that plays a physiological role in the pathogenesis of PD (23, 24, 25). Abnormal accumulation and aggregation of αSyn results in the degeneration of dopaminergic neurons associated with PD (26). Although the exact function of αSyn remains unknown, convincing evidence supports the involvement of a-synuclein in synaptic plasticity and neurotransmitter release (26). To better understand the connection between neuronal activity and αSyn, researchers used cultured neurons and living mice (27). Studies conducted by both Prashun et al. and Yamada et al., each in separate papers, hypothesized that neuronal activity regulates the release of αSyn from neurons (8, 27). Researchers concluded that depolarization of neurons led to increased exocytosis of pre-synaptic vesicles, which is important in a-synuclein release (27).
The underlying genetics contribute to the pathogenic role of αSyn in PD which include point mutations, duplication of SNCA, and polymorphic variants (26). Under native conditions, αSyn exists in a state between an unfolded monomer and folded tetramer, which prevents aggregation (26). A decrease in the tetramerto-monomer ratio leads to an increased level of αSyn in the unfolded monomer state, favoring aggregation (25, 26). As a result of conformational change, αSyn adopts a ß-sheet-rich structure facilitating aggregation of the protein and accumulation in Lewy Bodies (25, 26, 27). αSyn undergoes post-translational modification with phosphorylation being a significant pathological marker of PD. Evidence revealed a relationship between phosphorylation and αSyn degradation. Increased phosphorylation of αSyn due to inhibition of the ubiquitin-proteasome system suggests regulation of αSyn degradation (26). Failure to inhibit the ubiquitin-proteasome system can prevent phosphorylation of αSyn and dysregulation of αSyn
degradation allowing for the accumulation of αSyn in Lewy bodies (26, 27).
αSyn propagation from the gut to the brain contributes to the etiology of PD (28). The connection between pathological αSyn and the gut-brain axis follows Braak’s hypothesis. Braak’s hypothesis suggests progressive PD results from pathogens entering the nasopharyngeal cavity and migrating from the gastrointestinal tract to the brain resulting in loss of dopaminergic neurons (5, 28, 29). The gut microbiota imbalance promotes intestinal inflammation and impairment in the mucosal barrier (6). The endotoxins from the entering pathogens cause misfolding and accumulation of αSyn resulting in inflammation (6). The accumulated αSyn spreads from the enteric nervous system and migrates to the central nervous system via the vagus nerve spreading synucleinopathies from the brainstem to the substantia nigra causing motor dysfunction that contributes to the clinical manifestations of PD such as delayed gastric emptying and severe constipation (5, 6, 29).
In recent research conducted, researchers aimed to investigate three different aSyn species including oligomer αSyn (OSyn) and phosphorylate αSyn (PSyn) (6). OSyn was found to display toxic effects on the nervous system and contribute to the pathogenesis of PD (6). Researchers found that αSyn resided in both PD and control groups with variations in distribution within intestinal mucosa, while OSyn excretion occurred from the mucosal layer (6, 7). Enteroendocrine cells lining the intestinal tract aid in the propagation of defective αSyn (7). OSyn is an abnormal aggregate of αSyn (6). Increased excretion of Osyn from the mucous layer into the gut causes toxic protein accumulation and exhibits a correlation between the gut-brain axis retrograde theory (30). In PD patients, the increased production of Osyn in the gut and propagation to the brain via the vagus nerve leading to motor dysfunction and severe constipation (6, 31).
Gut-Brain Axis
The gut-brain axis has a reciprocating relationship with the CNS, sympathetic nervous system (SNS), ENS, hypothalamic-pituitary-adrenal (HPA) axis, and gut microbiota (12, 32). These systems combined contribute to the altered gut, production of neuromodulatory components, vagus nerve activation, and immune pathways. They are probiotic strains
that induce changes in neurotransmitter pathways, including serotonin, dopamine, and norepinephrine (32). A well-maintained and consistent composition of the gut microbiota is vital for preserving intestinal barrier integrity and reducing inflammation in healthy individuals (3). This, in turn, positively influences brain development and behavior through the intricate network known as the microbiota-gut-brain axis (3). Multiple pathways exist connecting the brain and the intestine, with the vagus nerve facilitating the most direct route (2). The vagus nerve originates from the dorsal motor nucleus in the medulla oblongata and extends throughout the abdominal region, innervating the visceral organs (2). When stimulation within the intestine occurs, the stimulus can initiate afferent signaling via the vagus nerve. This signaling process is crucial in neuroimmune inflammatory reflex circuits, vital for maintaining peripheral immune regulation (2). Additionally, emerging evidence suggests that the vagus nerve may serve as a direct pathway through which substances and signals from the intestine can be transmitted to the brain (2). This highlights the potential bidirectional communication between the gut and the CNS mediated by the vagus nerve (2).
Given the bidirectional communication between the CNS and ENS, there is a postulation of whether the onset of PD is due to "body-first" or "brain-first" (2, 14). One would automatically think of the "brainfirst" aligning with the pathology of PD since it would indicate that the misfolded αSyn would arise in the CNS and inevitably impact the dopaminergic neurons present in the substantia nigra, and eventually descend towards the rest of the body, causing subsequent symptoms such as motor impairments and constipation. On the other hand, the body-first phenomenon would indicate that an external pathogen would enter the gastrointestinal system and initiate the αSyn misfolding, suggesting that the progressive neurodegeneration could be peripheral (14). Then, this misfolding would continue to occur until it would progress into the dorsal motor nucleus of the vagus nerve to the medulla oblongata, eventually reaching the cerebral cortex (14). Another way an external pathogen may enter the host is through the olfactory bulb, which suggests a 2-hit hypothesis that would trigger the onset of PD more substantially (14). Returning to the gastrointestinal tract obtaining a pathogen, the chronic inflammation that could be associated with it will eventually cause dysbiosis of the gut and lead to increased intestinal permeability. Since
intestinal permeability will increase, it increases the chances of other bacteria and toxins crossing the ENS, potentially causing αSyn misfolding (14).
Notably, abnormal accumulation of αSyn, an essential constituent of Lewy bodies, has been identified within the ENS (10). A study on healthy rodents involved injecting human αSyn fibrils into the gut tissue, which induced αSyn to spread to the vagus nerve and brainstem (10). Furthermore, some studies involved the targeted expression of human αSyn in the vagus nerve by injecting adeno-associated viral vectors to which αSyn aggregates spread progressively from the medulla oblongata to other brain regions (8). In addition, a rodent study revealed that dysbiosis of the gut has effects preceding the formation of αSyn in mouse brains (9). These findings support that the vagus nerve is a conduit for transmission from the GI tract to the brain (8). Additionally, cohort studies performed in Northern Europe yielded compelling outcomes revealing that individuals who underwent truncal vagotomy had a lower risk of developing PD compared to the control group that was age and gender-matched (8). Furthermore, experimental manipulations in mice showed that cervical vagotomy performed before the injection of αSyn into the gastric wall prevented αSyn formation in the vagus's dorsal motor nucleus (8). As supported by the experimental evidence, these findings substantiate that αSyn can propagate from the gut to the brain, providing clarity on crucial aspects of this pathological process (8, 10).
Studies have highlighted that the gut microbiota can regulate the gut-brain axis through endocrine, immunological, and direct neuronal processes, contributing to the idea that PD is a pathological condition that extends from the gut to the brain (8,10). Aligning with the occurrence of gastrointestinal dysfunctions that precede the typical motor symptoms in individuals with PD, it becomes progressively apparent that the impact of microbiota community changes on the interactions within the gut-brain axis is also implicated in PD development (3). During the pathogenesis of PD, gut microbiota dysbiosis can elicit persistent inflammation within the intestinal epithelium, which could inevitably induce neuroinflammation through the microbiota-gut-brain axis (3). These inflammatory responses and fluctuations between the population levels of some bacteria contribute to microbiota dysbiosis, which will disrupt the integrity of the intestinal barrier, referred to as the "leaky gut" (3). Since there is increased
permeability of the intestinal barrier, this will promote the entry of pro-inflammatory byproducts such as lipopolysaccharides (LPSs) and cytokines into the systemic circulation (3). Consequently, these molecules breach the BBB, accessing the substantia nigra (SN), contributing the neuroinflammation, and potentially leading to the demise of dopaminergic neurons, which is a hallmark feature of PD (3).
Gut Microbiome and PD
The gut microbiota contains a collection of microorganisms, such as bacteria, viruses, fungi, and archaea, that live in a symbiotic relationship with the host. These microorganisms have many crucial functions, such as promoting normal motility, developing the gastrointestinal epithelium and ENS, and contributing to the development of priming the immune system to maintain a barrier towards pathogenic bacteria and digestion of nutrients. Its composition can be altered throughout one's lifetime from various factors such as diet, genetics, use of antibiotics or probiotics, energy intake, and stool transplants (11, 12). Normal physiology and the host's vulnerability to illnesses are impacted by the metabolic processes and interactions of the gut microbiota (33). At the bacterial level, unique differences are caused by changes in pH, immunological factors, and digestive enzymes. The microbiota directly impacts human health through the production and release of diverse components, including vitamins, essential amino acids, and lipids (33). This ecosystem is susceptible to changes in one's external environment, such as diet, sleep, chronic noise, and sedentary behaviors. These behaviors could contribute to the onset of several neurodegenerative diseases, such as PD. One of the predominant contributing factors that is also correlated with the inevitable dysbiosis of the gut is aging (32). Throughout aging, there is a potential for a decrease in bowel motility, a sign of constipation (32). Constipation is one of the early symptoms presented in PD. Chronic constipation could result in low-grade inflammation of the intestines, causing a leaky gut, ultimately leading to changes in the blood-brain barrier and neuroinflammation. A study conducted on PD and control groups utilized the Wexner Constipation Scoring System to evaluate constipation severity (15). The results revealed a significant association between the two groups, indicating a statistically significant positive correlation (15). This study confirmed an increased prevalence of constipation among PD patients, particularly in severe grades, thereby
establishing a link between constipation and the severity of PD symptoms (15). These findings suggest that individuals experiencing severe constipation are more susceptible to developing PD, accentuating the potential utility of constipation as a preclinical biomarker (15). Several bacterium levels, persistently noted across studies, have been associated with PD, highlighting the critical role of the gut microbiota in the pathogenesis of the disease (4, 10, 14, 15). Such bacterium includes decreased Prevotella and short-chain fatty acid (SCFA)-producing bacteria and increased Bacteroides and the Verrucomicrobiaceae genus, Akkermansia (1) (see Table 1).
Prevotella, a gram-negative bacterium, is thought to maintain mucosal integrity and produce neuroprotective short-chain fatty acids (SCFAs) that have the potential to exert a protective influence on dopaminergic neurons, shielding them from degradation (15). Additionally, Prevotella is involved in the protection of dopaminergic neurons via the secretion of hydrogen sulfide (15). Hydrogen sulfide is a gaseous neurotransmitter in the gut (15). One consistent finding across studies is a decreased abundance of the bacterium Prevotella in PD patients across various populations (1, 9, 10, 11, 13, 14, 15, 33). The mean abundance of Prevotella was reduced by 46.6% in PD patients compared to healthy controls (4). Decreased levels of Prevotella could increase inflammation and oxidative stress in the gut, which would inevitably trigger neuroinflammation and neurodegeneration in the brain. Moreover, Prevotella has been found to correlate with the occurrence of REM sleep behavior disorder (RBD) along with the progression of motor symptoms in PD over two years (10). In a separate study, individuals with an increased abundance of Prevotella exhibited lower rates
of constipation, indicating a decreased probability of experiencing subthreshold parkinsonism (10) (see Table 1). Subthreshold parkinsonism refers to the presence of mild Parkinsonian features that do not meet the criteria for a proper PD diagnosis (10). These significant findings could suggest that a reduction in Prevotella abundance could serve as a biomarker for PD diagnosis and a target for disease-modifying interventions (15) (see Table 1).
Studies have consistently demonstrated reduced short-chain fatty acid (SCFA)-producing bacteria, including Faecalibacterium prausnitzii and Roseburia from the Lachnospiraceae family in PD patients (9, 13, 14). Acetate, propionate, and butyrate are the three principal SCFAs produced by gut bacteria (12). They typically cause activities such as antitumorigenic, anti-inflammatory, and anti-microbial effects, change gut integrity, induce reactive oxygen species, and change cell proliferation and function.
Gut Microbiota Function
Prevotella Maintains the integrity of the mucosal lining and produces neuroprotective substances, including shortchain fatty acids (SCFAs) and hydrogen sulfide, which is a gasotransmitter.
SCFA-producing bacterium that maintains epithelial integrity, and has antiinflammatory properties.
SCFA-producing bacterium that maintains gut integrity and promotes gastrointestinal motility.
Bacteroides genus Produces pro-inflammatory neurotoxins such as lipopolysaccharides (LPS) and toxic proteolytic peptides. Stimulates immune cells.
Verrucomicrobiaceae family ( Akkermansia ) Maintains intestinal integrity, aids in mucus digestion, and plays a role in immune function.
Potential Contribution/Impact on PD
Reduced levels are linked to greater inflammation, oxidative stress, neuroinflammation, and neurodegeneration.
These lower levels are associated with constipation and the progression of motor symptoms.
Decreased levels may act as a potential biomarker for the diagnosis of PD.
Reduced levels in PD patients may contribute to impaired gut motility and increased inflammation, potentially triggering neurodegeneration and gastrointestinal issues like constipation.
Increased abundance in PD, particularly in the non-tremor subtype, is linked to more rapid disease progression.
May contribute to neuroinflammation through LPSmediated immune activation, exacerbating PD pathology and promoting αSyn aggregation.
Increased abundance in PD, particularly in the non-tremor subtype, is linked to more rapid disease progression.
May contribute to neuroinflammation through LPSmediated immune activation, exacerbating PD pathology and promoting αSyn aggregation.
Increased levels of Akkermansia in PD lead to increased intestinal permeability and inflammation, exposing the intestinal neural plexus to toxins like LPS. This exposure may promote αSyn aggregation and Lewy body formation, contributing to neuroinflammation and PD progression. It is also linked to IFN γ production.
Table 1: Table 1. Functional Roles of Gut Microbiota and Their Associations with Parkinson’s Disease. This table summarizes key gut microbiota species, their functional roles within the body, and their potential associations with PD pathology. Abbreviations: SCFA, short chain fatty-acids; LPS, Lipopolysaccharides. (Adapted from 1, 4, 9, 10, 11, 13, 14, 15, 33, 35).
SCFA-producing bacteria function to maintain epithelial integrity, possess anti-inflammatory properties, and promote gastrointestinal motility while regulating ENS function (10). Several ways could lead to the depletion of this bacteria, including diet, physical activity, sleep, antibiotics, and exposure to environmental toxins such as metals. The decrease in SCFA levels may lead to an increase in endotoxin and neurotoxic occurrence, which have been linked to the onset of PD (10). Recent studies implicated that SCFAs can promote microglia's maturation and inflammatory capabilities (34). Microglia does not express the SCFA receptors. However, they express the associated responsive genes, such as histone deacetylases, which could modulate the gene expression (34). Increased microglia activation and the production of pro-inflammatory cytokines could alter neuronal function and increase cell death in PD patients (34). Moreover, decreased SCFA levels may contribute to the emergence of gastrointestinal motility disorders, including constipation in PD, since one of these bacterium's functions is to promote gastrointestinal motility (10). There is a significant association in PD patients that low SCFAs are substantially correlated with poor cognition and lower BMI, and these individuals were found to have poorer postural instability-gait disorder scores as well (10) (see Table 1).
The Bacteroides genus, which falls within the Bacteroidetes phylum, exhibited an exceptional abundance among patients diagnosed with the nontremor subtype of PD (4). It is worth emphasizing that PD is a clinically heterogeneous disorder. Individuals with the non-tremor subtype experience more rapid disease progression and display heightened αSyn pathology in the ENS neurons compared to those with the tremor subtype (4). Specific genera within the Bacteroidetes phylum can produce diverse pro-inflammatory neurotoxins, including lipopolysaccharides (LPSs) and toxic proteolytic peptides (4). Consequently, this suggests a potential role for Bacteroides species in contributing to the inflammatory processes observed in PD. Furthermore, an intriguing positive correlation was observed between the abundance of Bacteroides and the plasma levels of the pro-inflammatory cytokine TNFα in PD patients (4). Notably, Bacteroides have been demonstrated to stimulate immune cells, such as macrophages and monocytes, to release TNFα through LPS-mediated pathways (4) (Table 1).
Akkermansia, which is a bacterium that is a member of the Verrucomicrobiaceae family, predominantly gramnegative, primarily inhabits the mucus layer of the large intestine, where it plays a vital role in maintaining intestinal integrity and facilitating the digestion of the intestinal mucus (35, 10). Akkermansia has been linked to certain health benefits such as improved immune function, accelerated wound healing, and protection against obesity (10). The degrading feature must accompany the pro-inflammatory pathways since the gut barrier's breakdown would increase resident immune cells' exposure to pathogens, causing abnormal aggregation of αSyn formation in the ENS. Increased levels of Akkermansia increase intestinal permeability and intestinal inflammation, which would expose the intestinal neural plexus to toxins such as lipopolysaccharide (LPS), which would lead to the abnormal aggregation of αSyn and generation of Lewy bodies (10). Furthermore, a study revealed a modest correlation between the abundance of Verrucomicrobia and IFNγ. IFNγ is a pro-inflammatory cytokine derived from type I helper T cells (4).
The dysregulation of the gut microbiota has been associated with abnormal immune responses within the gut and throughout the body, which would lead to the excess production of inflammatory cytokines in the bloodstream, and that would trigger motor impairments and neurodegeneration by inducing neuroinflammation in a PD mouse model with the overexpression of αSyn (4). Therefore, since increased levels of Akkermansia are pronounced in PD, this could indicate the contribution to the disorder's progression.
These crucial bacteria strongly suggest the active involvement of gut microbiota in the pathogenesis of PD, as evidenced by their correlations with disease duration, cognitive impairment, and potential implications for treatment development. It is imperative to conduct further research to investigate the specific microbiota differences and their functional roles that contribute to the progression of PD. To gain comprehensive insights, the design of a longitudinal study aimed at monitoring disease progression and characterizing alterations in the taxonomic composition of the gut microbiome would prove highly advantageous. Such an approach would provide valuable information regarding the dynamic relationship between the gut microbiota and PD, enabling a deeper understanding of the disease mechanisms and potentially paving the way for targeted therapeutic interventions.
Immune System and the Gut Microbiota
The gut microbiome is a dynamic environment that balances the immune system. Regulating immune homeostasis is one critical benefit gut microbiota provides to the organism (36). When changes occur within the body, homeostatic balance becomes disrupted, and similar results occur with alterations to gut microbiota (36). Various gut microbiota cytokine interaction patterns have been discovered and are specific to a stimulus or cytokine specificity (37). Microbial metabolic pathways link together specific microbiota species and various cytokine levels. Differences in cytokine levels are associated when more significant or lesser percentages of bacteria are present (37). Such taxonomic associations include the typical gut microbiota, Dorea formicigenerans, where higher levels correlate to greater levels of IFNγ. The opposite is seen for Dorea species which negatively correlates to IFNγ response. Taxonomic associations are generally associated with bacterialinduced inflammatory cytokine response (38). Such microbiota changes and cytokine immune responses are correlated with symptoms of PD (39). Changing gut microbiota in PD patients is associated with varying concentrations of IFNγ and TNFα (39). Such changes in cytokine levels were thought to contribute to irregular immune responses and enhance inflammatory processes in PD patients (39).
Gut microbiotas produce metabolites that play critical roles in inflammatory signaling. When such changes occur, disrupting the balance of the gut, alterations to the immune response follow (40). The reciprocal host-gut microbiota axis describes the relationship between gut microbiota and immune responses (41). The axis’ primary defense is innate immunity, which correlates to gut microbiota taxis (41). Gut-associated Lymphoid Tissue (GALTs) is critical in establishing a connection between gut microbiota and innate immune homeostasis (42). When gut microbiota is disturbed, GALTs structure is altered, interfering with the local immune responses. Although the mechanism is not entirely understood, one theory is based on the pro-inflammatory cytokines produced by GALTS derived from different gut microbiota. Such cytokines can lead to greater susceptibility to autoimmune disorders (42). The microbiota-gut-brain axis connects gut, immune, and brain functions. Gut dysbiosis and increased pro-inflammatory cytokines are said to improve intestinal and BBB permeability leading to an accumulation of misfolded proteins and axonal damage
(43–45). It is through damaged protein buildup that the pathogenesis of neurodegenerative diseases occurs, including PD (43). PD patients with high levels of Enterobacteriaceae displayed increased levels of lipopolysaccharides within the blood, which acts as a neurotoxin (43–45). Serum LPS may induce the pro-inflammatory cytokine response, which disrupts the blood-brain barrier leading to the degradation of dopaminergic neurons and substantia nigra (43–45).
Studies show that changes in an individual’s gut microbiota result in many symptoms. How these changes to the gut microbiome occur varies by patient. Alterations to the gut microbiota composition may be due to environmental factors, stress, and diet. Such changes result in changes in epileptiform activity (46).
Modifications within the gut microbiota associated with neuroinflammation are also common symptoms of aging. As one ages, gut microbiota decreases, affecting the gut-brain axis (47). Such environmental factors would trigger T-cell infiltration and immunemediated neural damages that coincide with symptoms of PD (48). Aging supports a pro-inflammatory microenvironment and, when coupled with a decrease in gut microbiota diversity, can result in increased progression of neurodegeneration (49). Both increase and decrease in specific genera of bacteria, including enriching Lactobacillus, Akkermansia, and Bifidobacterium, as well as depleting Lachnospiraceae family, are some of the most displayed alterations in PD patients (50). The GI tract is the primary communication site between the host immune system and the gut microbiota, highlighting the importance of the relationship (51). Dysregulation of the relationship results in disorders like that of PD. Understanding the variations of gut microbiota and its relationship with the inflammatory immune response will help to advance the treatment of patients with PD.
PD Treatments
Most clinically approved treatment modalities for PD center around treating PD symptoms and do not address underlying pathophysiology in the brain or gut. Current first-line pharmacological treatment for PD symptoms includes dopamine replacement therapies such as levodopa (L-Dopa) and combination levodopacarbidopa (52–54). L-Dopa is regarded as the standard gold treatment for early-stage PD, administered to replenish the loss of striatal dopamine (54–56). It has a long history of use in alleviating the primary motor effects of PD, namely Parkinsonian tremors.
However, these effects can be limited (57). Prolonged high-dose L-Dopa treatment is complicated by the onset of characteristic involuntary motor movements referred to as levodopa-induced dyskinesia (LID) (54, 58). LID symptom onset occurs in approximately 40% of PD patients on L-Dopa therapy lasting four or more years (58, 59). Clinical data has established a positive correlation between L-Dopa dose and LID manifestation (60, 61). Alternatives to L-Dopa seek to achieve similar efficacy without LID but are still targeted toward symptom management. Preliminary clinical data suggest levodopa-carbidopa administered to the jejunum as a long-acting gel can increase LID-free periods associated with off-peak L-Dopa blood levels by approximately two h (95% CI -3.05 to -0.76]; p=0.0015) but requires surgical placement of jejunostomy, an effective and invasive procedure (62). Research continues to seek methods to improve long-term L-Dopa efficacy, reduce LID-associated complications, or serve as less invasive, effective alternative interventions.
As interest in the gut-brain axis's involvement in PD pathophysiology has grown, so has the concurrent interest in gut-brain axis-mediated prevention and intervention. Investigations that have reached clinical trials include nutritional therapy via diet modification and direct gut microbiome modulation via pre-, post-, and probiotic means (16, 63, 64). Thus far, these treatment avenues remain experimental and have not been widely adopted in the clinic.
A growing body of research has identified diet as a factor that can be protective against PD (64,65). Specifically, the study finds the Mediterranean (MEDI) diet, rich in olive oil, fresh fruit and vegetables, fish, poultry, beans, and pulses, is protective (66–69). Data from a 2019 Greek cohort study suggest adherence to the MEDI diet is associated with a modest 2% lower probability of developing prodromal PD symptoms, defined as gut motility issues, depression, and cognitive decline (p < 0.001) (66). A 2021 analysis from the Rotterdam Cohort Study (n = 9414) finds adherence to a MEDI-like diet has a protective effect against PD with a hazard ratio of 0.89 (95% CI 0.74–1.07) (67). Conversely, a 2009 study from Japan concludes diets high in consumption of animal fats, specifically cholesterol intake, were positively associated with PD risk finding an odds ratio (OR) of 2.09 (95% CI 1.21-3.64) —further suggesting a connection between diet and PD (70). These data suggest dietary modifications, exceptionally minimal
consumption of red meat, animal fats, and alcohol, and increased adherence to a MEDI-like diet can protect against PD.
Beyond preventive benefits, researchers have begun investigating diet as an intervention for those already diagnosed with PD, focusing on the MEDI diet and the related Mediterranean-DASH Intervention for Neurodegenerative Delay (MIND) diet (16, 66, 67). The MIND diet differs from the MEDI diet in food prioritization, with the MIND diet emphasizing more leafy green vegetable consumption instead of MEDI’s broader emphasis on fresh vegetable consumption (16, 66). Regardless, data suggests both diets may serve as nutritional therapy for PD patients. 2022 data from the ongoing Modifiable Variables in Parkinsonism study found patients adhering to a MIND diet reported 52.9 points lower on the patient-reported outcomes in PD (PRO-PD) tool — a self-report survey consisting of 32 sliding scale questions to assess PD symptom severity and frequency (17, 68). On average, patients adhering to a MEDI diet reported 29.6 points lower in the same metric (68). Similarly, a 2017 crosssectional analysis using PRO-PD metrics found patients adhering to diets like MEDI and MIND were associated with slower progression of PD symptoms (p < 0.05), while patients reported adhering to diets high in refined sugars, fried foods, and full-fat dairy products were associated with more rapid progression of PD (p < 0.05) (17). Further differentiation in dietary factors also shows diet modification as a promising intervention. A small 2020 randomized control trial (RCT) of 47 PD patients concludes low fat and specifically ketogenic diets improve nonmotor and motor daily living experiences by 41% (p < 0.001) in patients already prescribed L-Dopa as reported using the Movement Disorder Society’s Unified Parkinson’s Disease Rating Scale MDS-UPDRS (71).
Recent research has sought to move beyond associative measures and establish a causal relationship between diet interventions, gut microbiome changes, microbial metabolite changes, and disease states. This has led to experimental treatment avenues that directly modulate microbiome and microbial metabolites, such as short-chain fatty acid SCFA levels in PD patients (72, 73). An 86-subject 2020 case-control study not only confirmed gut microbial dysbiosis between healthy subjects and PD-diagnosed subjects, but further concluded adherence to an ovo-lacto vegetarian diet and the addition of a bowel cleansing regimen in PD subjects
currently treated with L-Dopa significantly improves MDS-UPDRS part III motor evaluation scores. It is associated with elevated levels of beneficial Ruminococcaceae microbial taxa known to metabolize dietary components to SFCAs (pooled p < 0.01) (72). The benefit of such intervention may be ascribed to the dual effect introduction of SCFA precursors via an ovo-lacto vegetarian diet and a beneficial change in gut microbiota that generates more SCFAs via a bowel cleansing regimen (72). Further, the 2021 open-label trial (RESIST_PD) of 87 subjects concluded that altering bowel SCFA levels via prebiotic dietary supplementation of resistant starches, i.e., starches that undergo digestion only by gut microbiota, results in less gut dysbiosis, increased colonic SCFA levels as measured by fecal butyrate levels (p = 0.029), and improvement of non-motor related PD symptoms on the Non-Motor Symptom Questionnaire (NMSQ) (p = 0.001) (73).
Future Research
Dietary intervention trials and RESIST-PD prebiotic trials have shown promise in the clinic as stand-alone and supplementary interventions to prevent the onset of PD and reduce symptoms associated with it (73). Research is ongoing to understand better how gutbrain axis-mediated interventions can treat or reverse PD symptoms.
Research into probiotic dietary supplementation has shown preliminary promise in clinical trials for PD-associated symptoms, mostly constipation (74, 75). Data from a 2016 120-subject RCT suggests probiotic dietary supplementation with fermented milk containing probiotic microbial strains and prebiotic fiber significantly increases reported complete bowel movements in PD suffering from PD-associated constipation with a mean difference of mean difference in movements of 1.1 (95% CI 0.41.8; p = 0.002) (74). A smaller 46-subject 2022 RCT also concludes probiotic supplementation, including defined probiotic microbial strains, improves PDassociated constipation symptoms (75). The study further concludes probiotics can partially restore beneficial microbial taxa that are often observed to be reduced or absent from PD patient gut microbial populations (75). More research will be needed to connect probiotics and their modulatory effect on the gut microbiome to PD symptoms.
Another growing area of interest is using fecal microbial transplantation (FMT) to mitigate microbial
dysbiosis and alleviate PD symptoms. Recent studies demonstrate that restoration of gut microbial diversity is linked to reducing pro-inflammatory signals and protective effects against PD symptoms (18, 19.) A 2018 study concludes FMT reduces gut dysbiosis, suppresses pro-inflammatory TLR4 and TNFα signaling, and significantly PD PD-associated motor symptoms in murine models of PD (18). A similar 2021 study supports and expands on this model, concluding that FMT restores gut microbial diversity and barrier integrity, reduces pro-inflammatory TLR4 and NFkB signaling, and restores motor function in rotenoneinduced PD mouse models (19). This pre-clinical data shows promise and a roadmap for future human trials.
Conclusion
The significance of the gut-brain axis in neurodegenerative disorders is clear and warrants further investigation to understand the underlying pathophysiology of these diseases better. Dysregulation of the gut microbiome contributes to neuroinflammation by modulating immune responses and altering gut permeability, ultimately promoting αSyn aggregation and neurodegeneration. Understanding the connection between the gut microbiome and immune response, we can develop targeted treatments to restore gut microbiome diversity and inflammatory cytokines, thereby preventing neurodegeneration.
A consistent pattern of gut microbiota shifts has been observed in PD patients, with notable reductions in Prevotella, Faecalibacterium prausnitzii, and Roseburia, all of which play key roles in maintaining gut homeostasis and anti-inflammatory signaling. Notably, Prevotella levels are significantly reduced in PD patients, potentially contributing to neuroinflammation and neurodegeneration. This suggests that Prevotella could serve as a potential biomarker for PD. Similarly, Faecalibacterium prausnitzii and Roseburia, both members of the Lachnospiraceae family, exhibit lower abundance in PD, which may lead to decreased anti-inflammatory SCFA production, resulting in neuroinflammation. Their depletion may contribute to increased neuroinflammation through decreased SCFA production.
In contrast, an increase in Bacteroides has been observed, potentially exacerbating neuroinflammation through increased lipopolysaccharide (LPS) production, leading to more rapid disease production.
Additionally, the Verrucomicrobiaceae family is often elevated in PD patients, which may be linked to altered gut barrier function exposing the gut neural plexus to toxins leading to an accumulation of αSyn aggregation and Lewy body formation, increasing the progression of PD. These microbial alterations are strongly associated with increased circulating cytokines, disrupting the BBB and accelerating neurodegenerative processes. Homeostatic gut microbiome disruptions and innate immune response are closely associated with the motor and non-motor symptoms present in PD. Numerous studies have identified a consistent pattern of gut microbiota dysbiosis in PD, with specific bacterial taxa playing a key role. These shifts in key gut microbial populations may contribute to PD pathophysiology through gut permeability, neuroinflammation, and microbial metabolite production mechanisms.
Given these findings, targeted modulation of gut microbiota presents a promising therapeutic approach. Restoring microbial diversity through prebiotics, probiotics, or dietary interventions could reduce inflammatory cytokines, enhance gut barrier function, and potentially slow PD progression. Further research is essential to establish transparent causal relationships and identify specific microbiota that could serve as diagnostic biomarkers or therapeutic targets. Understanding the direct communication mechanisms between the gut microbiome, immune responses, and the brain could pave the way for novel microbiomebased interventions, potentially transforming the treatment of neurodegenerative diseases.
Disclosures
The authors have declared no conflicts of interest relevant to this study. Ethical approval was not required for this study, as it is a systematic review of existing literature. The data supporting this work were obtained from multiple published sources, as referenced in the manuscript. MG, TG, LW, and AT were responsible for the study concept, primary literature search, and manuscript drafting. They also contributed equally to interpreting the primary literature data and revising the manuscript. All authors reviewed and approved the initial draft. BP supervised the project, reviewed the manuscript for further modifications, and approved the final version for submission.
Acknowledgments
This research was conducted with the editorial and publication assistance of Geisinger Commonwealth School of Medicine. The authors thank Dr. Piper and Dr. Lobo for their guidance and feedback.
References
1. Aho VTE, Pereira PAB, Voutilainen S, Paulin L, Pekkonen E, Auvinen P, et al. Gut microbiota in Parkinson’s disease: Temporal stability and relations to disease progression. EBioMedicine. 2019;(44):691–707.
2. Houser MC, Tansey MG. The gut-brain axis: Is intestinal inflammation a silent driver of Parkinson’s disease pathogenesis? Npj Parkinson’s Disease. 2017;(3):3.
3. Zhao Z, Ning J, Bao X, Shang M, Ma J, Li G, et al. Fecal microbiota transplantation protects rotenone-induced Parkinson’s disease mice via suppressing inflammation mediated by the lipopolysaccharide-TLR4 signaling pathway through the microbiota-gut-brain axis. Microbiome 2021;(9):226.
4. Lin CH, Chen CC, Chiang HL, Liou JM, Chang CM, Lu TP, et al. Altered gut microbiota and inflammatory cytokine responses in patients with Parkinson's disease. J Neuroinflammation 2019;16(1):129.
5. Challis C, Hori A, Sampson TR, Yoo BB, Challis RC, Hamilton AM, et al. Gut-seeded alpha-synuclein fibrils promote gut dysfunction and brain pathology specifically in aged mice. Nat Neurosci. 2020;23(3):327-36.
6. Shi J, Wang Y, Chen D, Xu X, Li W, Li K, et al. The alteration of intestinal mucosal α-synuclein expression and mucosal microbiota in Parkinson’s disease. Appl Microbiol Biotechnol. 2023;107(5–6):1917–29.
7. Chandra R, Hiniker A, Kuo Y-M, Nussbaum RL, Liddle RA. α-Synuclein in gut endocrine cells and its implications for Parkinson’s disease. JCI Insight 2017;(2):92295.
8. Prasuhn J, Strautz R, Lemmer F, Dreischmeier S, Kasten M, Hanssen H, et al. Neuroimaging correlates of substantia nigra hyperechogenicity in Parkinson's disease. J Parkinsons Dis. 2022;12(4):1191-200.
9. Jin M, Li J, Liu F, Lyu N, Wang K, Wang L, et al. Analysis of the gut microflora in patients with Parkinson’s disease. Front Neurosci. 2019;13:1184.
10. Zhu M, Liu X, Ye Y, Yan X, Cheng Y, Zhao L, et al. Gut microbiota: A novel therapeutic target for Parkinson’s disease. Front Immunol. 2022; 13: 93755.
11. Li Z, Liang H, Hu Y, Lu L, Zheng C, Fan Y, et al. Gut bacterial profiles in Parkinson’s disease: A systematic review. CNS Neurosci Ther 2023;29(1):140–57.
12. Młynarska E, Gadzinowska J, Tokarek J, Forycka J, Szuman A, Franczyk B, et al. The role of the microbiome-brain-gut axis in the pathogenesis of depressive disorder. Nutrients. 2022;14(9):1921.
13. Aho VTE, Houser MC, Pereira PAB, Chang J, Rudi K, Paulin L, et al. Relationships of gut microbiota, short-chain fatty acids, inflammation, and the gut barrier in Parkinson’s disease. Mol Neurodegeneration. 2021;16(1):6.
14. Nuzum ND, Loughman A, Szymlek-Gay EA, Teo W-P, Hendy AM, Macpherson H. To the gut microbiome and beyond: The brain-first or body-first hypothesis in Parkinson’s disease. Front Microbiol. 2022;13:791213.
15. Mekky J, Ahmad S, Metwali M, Farouk S, Monir S, ElSayed A, et al. Clinical phenotypes and constipation severity in Parkinson’s disease: Relation to Prevotella species. Microbes and Infectious Diseases. 2022;0(0):0–0.
16. Fox DJ, Park SJ, Mischley LK. Comparison of associations between MIND and Mediterranean Diet scores with patient-reported outcomes in Parkinson's disease. Nutrients. 2022;14(23).
17. Mischley LK, Lau RC, Bennett RD. Role of diet and nutritional supplements in Parkinson's disease progression. Oxid Med Cell Longev. 2017;2017:6405278.
18. Sun MF, Zhu YL, Zhou ZL, Jia XB, Xu YD, Yang Q, et al. Neuroprotective effects of fecal microbiota transplantation on MPTP-induced Parkinson's disease mice: Gut microbiota, glial reaction and TLR4/TNF-alpha signaling pathway. Brain Behav Immun. 2018;70:48-60.
19. Zhao Z, Ning J, Bao XQ, Shang M, Ma J, Li G, et al. Fecal microbiota transplantation protects rotenone-induced Parkinson's disease mice
via suppressing inflammation mediated by the lipopolysaccharide-TLR4 signaling pathway through the microbiota-gut-brain axis. Microbiome 2021;9(1):226.
20. Sonne J, Reddy V, Beato MR. Neuroanatomy, substantia nigra. In: StatPearls. Treasure Island (FL): StatPearls Publishing; 2023.
21. Zahoor I, Shafi A, Haq E. Pharmacological treatment of parkinson’s disease. In: Stoker TB, Greenland JC, editors. Parkinson’s Disease: Pathogenesis and Clinical Aspects. Brisbane (AU): Codon Publications; 2018.
22. Cramb KML, Beccano-Kelly D, Cragg SJ, WadeMartins R. Impaired dopamine release in Parkinson’s disease. Brain. 2023; awad064.
23. Stefanis L. α-synuclein in Parkinson’s disease. Cold Spring Harb Perspect Med. 2012;2(2):a009399.
24. Booth HDE, Hirst WD, Wade-Martins R. The role of astrocyte dysfunction in Parkinson’s disease pathogenesis. Trends in Neurosciences 2017;40:358–70.
25. Sulzer D, Edwards RH. The physiological role of α-synuclein and its relationship to Parkinson’s disease. J Neurochem. 2019;150(5):475–86.
26. Gómez-Benito M, Granado N, García-Sanz P, Michel A, Dumoulin M, Moratalla R. Modeling Parkinson’s disease with the alpha-synuclein protein. Front Pharmacol. 2020;11:356.
27. Yamada K, Iwatsubo T. Extracellular α-synuclein levels are regulated by neuronal activity. Mol Neurodegeneration. 2018;13(1):9.
28. Kim S, Kwon SH, Kam TI, Panicker N, Karuppagounder SS, Lee S, et al. Transneuronal propagation of pathologic alpha-synuclein from the gut to the brain models Parkinson's disease. Neuron. 2019;103(4):627-41 e7.
29. Rietdijk CD, Perez-Pardo P, Garssen J, Van Wezel RJA, Kraneveld AD. Exploring braak’s hypothesis of parkinson’s disease. Front Neurol. 2017;8.
30. Travagli RA, Browning KN, Camilleri M. Parkinson disease and the gut: new insights into pathogenesis and clinical relevance. Nat Rev Gastroenterol Hepatol 2020;17:673–85.
31. Böttner M, Zorenkov D, Hellwig I, Barrenschee M, Harde J, Fricke T, et al. Expression pattern and localization of alpha-synuclein in the human
enteric nervous system. Neurobiology of Disease 2012;48:474–80.
32. Ong J-S, Lew L-C, Hor Y-Y, Liong M-T. Probiotics: The next dietary strategy against brain aging. Prev Nutr Food Sci. 2022;27(1):1–13.
33. Zhang H, Chen Y, Wang Z, Xie G, Liu M, Yuan B, et al. Implications of gut microbiota in neurodegenerative diseases. Front. Immunol. 2022; 13:785644.
34. Sampson TR, Debelius JW, Thron T, Janssen S, Shastri GG, Ilhan ZE, et al. Gut microbiota regulate motor deficits and neuroinflammation in a model of Parkinson’s disease. Cell. 2016;167(6):14691480.e12.
35. Bonnechère B, Amin N, van Duijn C. What are the key gut microbiota involved in neurological diseases? A systematic review. Int J Mol Sci. 2022;23(22):13665.
36. Wu HJ, Wu E. The role of gut microbiota in immune homeostasis and autoimmunity. Gut Microbes. 2012; 3(1) 4-14.
37. Schrimer M, Smeekens SP, Vlamakis H, Jaeger M, Oosting M, Franzosa EA, et al. Linking the human gut microbiome to inflammatory cytokine production capacity. Cell. 2016; 167(4):11251136.
38. Lin CH, Chen CC, Chiang CL, Liou JM, Chang CM, Lu TP, et al. Altered gut microbiota and inflammatory cytokine response in patients with Parkinson’s disease. J Neuroinflammation 2019;16(1):129.
39. Cantanzaro R, Anzalone M, Calabrese F, Milazzo M, Capuana M, Itaia A, et al. The gut microbiota and its correlations with central nervous system disorders. Panminerva Med. 2015; 57(3): 127-43.
40. Yoo JY, Groer M, Dutra SVO, Sarkar A, McSkimming DI. Gut microbiota and immune system interactions. Microorganisms.2020;8(10):1587.
41. Jiao Y, Huntington ND, Zhang X. Crosstalk between gut microbiota and innate immunity and its implication in autoimmune diseases. Front Immunol.2020; (11).
42. Sarkar SR, Banerjee S. Gut microbiota in neurodegenerative disorders. J Neuroimmunol 2019;(328):98-104.
43. Block M, Zecca L, Hong JS. Microglia-mediated neurotoxicity: uncovering the molecular mechanisms. Nat Rev Neurosci. 2007;8(1):57-69
44. Chen H, O’Reilly E, Schwarzschild MA, Ascherio A. Peripheral inflammatory biomarkers and risk of Parkinson’s disease. AM J Epidemiol 2008;167(1):90-5.
45. Dobbs RJ, Charlett A, Purkiss AG, Dobbs SM, Weller C, Peterson W. Association of circulating TNF-alpha and IL-6 with ageing and parkinsonism. Acta Neurol Scand. 1999:100(1):34-41.
46. Reale M, Larlori C, Thomas A, Gambi D, Perfetti B, Nicola MD, Onofrj M. Peripheral cytokines profile in Parkinson’s Disease. Brain Behav Immun. 2009;23(1):55-63.
47. Henion C, Khaw PT. The interplay between inflammation, immunity and commensal microflora in glaucomatous neurodegeneration. AES.2019;(4).
48. Russo R, Varano GP, Adornetto A, et al. Retinal ganglion cell death in glaucoma: exploring the role of neuroinflammation. Eur J Pharmacol 2016;(787):134-42.
49. Frank MG, Barrientos RM, Biedenkapp JC, Rudy JW, Watkings LR, Maier SF. mRNA upregulation of MHC II and pivotal pro-inflammatory genes in normal brain aging. Neurobiol Aging 2006;27(5):717-722.
50. Romano S, Savva GM, Bedarf J, Charles IG, Hildebrand F, Narbad A. Meta-analysis of the Parkinson’s disease gut microbiome suggests alterations linked to intestinal inflammation. NPJ Parkinsons Dis. 2021:(7).
51. Round JL, Mazmanian SK. The gut microbiota shapes intestinal immune responses during health and disease. Nat Rev Immunol. 2009:(9);313-323.
52. Armstrong MJ, Okun MS. Diagnosis and treatment of Parkinson disease: a review. JAMA. 2020;323(6):548-60.
53. Church FC. Treatment options for motor and non-motor symptoms of Parkinson's disease. Biomolecules. 2021;11(4):612
54. Haddad F, Sawalha M, Khawaja Y, Najjar A, Karaman R. Dopamine and levodopa prodrugs for the treatment of Parkinson's disease. Molecules 2017;23(1):40
55. Tanaka H, Kannari K, Maeda T, Tomiyama M, Suda T, Matsunaga M. Role of serotonergic neurons in L-DOPA-derived extracellular dopamine in the striatum of 6-OHDA-lesioned rats. Neuroreport. 1999;10(3):631-4.
56. LeWitt PA. Levodopa therapy for Parkinson's disease: Pharmacokinetics and pharmacodynamics. Mov Disord. 2015;30(1):64-72.
57. Negrotti A, Secchi C, Gentilucci M. Effects of disease progression and L-dopa therapy on the control of reaching-grasping in Parkinson's disease. Neuropsychologia. 2005;43(3):450-9.
59. Ahlskog JE, Muenter MD. Frequency of levodoparelated dyskinesias and motor fluctuations as estimated from the cumulative literature. Mov Disord. 2001;16(3):448-58.
60. Schrag A, Quinn N. Dyskinesias and motor fluctuations in Parkinson's disease. A communitybased study. Brain. 2000;123 ( Pt 11):2297-305.
61. Warren Olanow C, Kieburtz K, Rascol O, Poewe W, Schapira AH, Emre M, et al. Factors predictive of the development of Levodopa-induced dyskinesia and wearing-off in Parkinson's disease. Mov Disord. 2013;28(8):1064-71.
62. Olanow CW, Kieburtz K, Odin P, Espay AJ, Standaert DG, Fernandez HH, et al. Continuous intrajejunal infusion of levodopa-carbidopa intestinal gel for patients with advanced Parkinson's disease: a randomised, controlled, double-blind, double-dummy study. Lancet Neurol 2014;13(2):141-9.
63. Tan AH, Lim SY, Lang AE. The microbiomegut-brain axis in Parkinson disease - from basic research to the clinic. Nat Rev Neurol. 2022;18(8):476-95.
64. Vascellari S, Palmas V, Melis M, Pisanu S, Cusano R, Uva P, et al. Gut Microbiota and Metabolome Alterations Associated with Parkinson's Disease. mSystems. 2020;5(5):e00561-20
65. Jackson A, Forsyth CB, Shaikh M, Voigt RM, Engen PA, Ramirez V, et al. Diet in Parkinson's disease: Critical role for the microbiome. Front Neurol 2019;10:1245.
66. Maraki MI, Yannakoulia M, Stamelou M, Stefanis L, Xiromerisiou G, Kosmidis MH, et al. Mediterranean diet adherence is related to reduced probability of prodromal Parkinson's disease. Mov Disord. 2019;34(1):48-57.
67. Strikwerda AJ, Dommershuijsen LJ, Ikram MK, Voortman T. Diet quality and risk of Parkinson's disease: The Rotterdam Study. Nutrients. 2021;13(11):3970
68. Mischley LK, Lau RC, Weiss NS. Use of a self-rating scale of the nature and severity of symptoms in Parkinson's Disease (PRO-PD): Correlation with quality of life and existing scales of disease severity. NPJ Parkinsons Dis. 2017;3:20.
69. Solch RJ, Aigbogun JO, Voyiadjis AG, Talkington GM, Darensbourg RM, O'Connell S, et al. Mediterranean diet adherence, gut microbiota, and Alzheimer's or Parkinson's disease risk: A systematic review. J Neurol Sci. 2022;(434):120166.
70. Miyake Y, Sasaki S, Tanaka K, Fukushima W, Kiyohara C, Tsuboi Y, et al. Dietary fat intake and risk of Parkinson's disease: a case-control study in Japan. J Neurol Sci. 2010;288(1-2):117-22.
71. Phillips MCL, Murtagh DKJ, Gilbertson LJ, Asztely FJS, Lynch CDP. Low-fat versus ketogenic diet in Parkinson's disease: A pilot randomized controlled trial. Mov Disord. 2018;33(8):1306-14.
72. Hegelmaier T, Lebbing M, Duscha A, Tomaske L, Tonges L, Holm JB, et al. Interventional influence of the intestinal microbiome through dietary intervention and bowel cleansing might improve motor symptoms in Parkinson's disease. Cells. 2020;9(2).
73. Becker A, Schmartz GP, Groger L, Grammes N, Galata V, Philippeit H, et al. Effects of resistant starch on symptoms, fecal markers, and gut microbiota in Parkinson's disease - The RESISTAPD Trial. GPB. 2022;20(2):274-87.
74. Barichella M, Pacchetti C, Bolliri C, Cassani E, Iorio L, Pusani C, et al. Probiotics and prebiotic fiber for constipation associated with Parkinson disease: An RCT. Neurology. 2016;87(12):1274-80.
75. Du Y, Li Y, Xu X, Li R, Zhang M, Cui Y, et al. Probiotics for constipation and gut microbiota in Parkinson's disease. Parkinsonism Relat Disord 2022;(103):92-7.
The Effects of Antipsychotics on Brain Structure
Emily R. Cranmer1*and Craig W. Reichard Jr.1*
¹Geisinger College of Health Sciences, Scranton, PA 18509
*Master of Biomedical Sciences Program
Correspondence: ertrocchi@som.geisinger.edu
Abstract
Schizophrenia, defined by the presence of delusions, hallucinations, and flat affect, is more commonly seen alongside depression, anxiety, and substance misuse in patients. Antipsychotics, including second-generation antipsychotics olanzapine, clozapine, and risperidone, are prescribed to control symptoms of schizophrenia, and function through the antagonism and agonism of various receptors including D2, 5-HT2C, 5-HT2A, and 5-HT1A. The effects of antipsychotics are widespread and often include varying impacts to multiple cerebral areas and functions such as measured cortical thickness, regional homogeneity, and volume of numerous brain components like the caudate, putamen, and gray matter. This study aimed to analyze the widespread effects of antipsychotic use on various brain structures based on a review of current scientific literature. Our review revealed that impacts to brain function were widely varied, showing a need for more robust clinical research. A comprehensive understanding of the effects on brain structures of antipsychotics will help to guide medical professionals in the treatment and management of psychosis-related syndromes.
Introduction
Antipsychotics were first used in anesthesia in the 1950s but are now utilized for their antipsychotic properties (1). These drugs are classified as either typical or atypical based on differences in their adverse effects and mechanism of action (1). The mechanism of action of antipsychotics are similar overall, but with some key differences. Typical antipsychotics work by inhibiting D2 dopamine, cholinergic, and histaminergic receptors while atypical antipsychotics work by also inhibiting D2 dopamine receptors as well as sub-type 5-HT2A serotonin receptors (2). Some research shows that antipsychotics block the D1, D2, and D3 dopamine receptors in the forebrain, affecting the basal ganglia as well as other areas of the brain (3, 4). Prescribed more frequently than typical antipsychotics, atypical antipsychotics work on a wider range of symptoms in
mood disorders (5). Typical antipsychotics work mainly on reducing the positive symptoms of mood disorders, while atypical antipsychotics are effective on negative and cognitive symptoms (2). Positive symptoms are usually defined as symptoms that distort daily life through a surplus of symptoms such as delusions or hallucinations, while negative symptoms are usually defined as an absence of certain behaviors that are deemed normal (6). Another way to define typical and atypical antipsychotics is using the nomenclature of first-generation (FGA) or second-generation antipsychotics (SGA), respectively (7).
Antipsychotics affect the brain in a regional manner, rather than a global manner (8). There is an association of greater changes in brain activity and gray matter volumes (GMV) with typical antipsychotics when compared to atypical antipsychotics (8). The areas of the brain that these medications work on were shown to be centered around the frontal and temporal lobes, limbic system, and basal ganglia (9). However, the changes in brain activity and volume are dependent on the type of antipsychotic administered (9). The frontal lobe is essential for memory and speech production (10). The temporal lobe is divided into the superior temporal sulcus and lateral temporal sulcus, with the main functioning areas being the superior temporal gyrus (STG) and the middle temporal gyrus (MTG). These areas are involved in memory, but it also has a defining role in speech comprehension (10). Lesions in the frontal lobe can lead to adverse effects such as personality disorders, aphasia, and apraxia while lesions in the temporal lobe can lead to auditory or visual hallucinations (10). The basal ganglia are responsible for the initiation of motor movement as well as sending signals to the limbic system, cortex, and thalamus. This allows the basal ganglia to also play a role in emotion and executive function (11). Although there is no universal consensus on the identity of the structures that compose the limbic system, the structures that are normally thought of as being part of the system are the limbic cortex, hippocampus, hypothalamus, amygdala, and septal area. This system
is thought to be responsible for many functions but is mainly known for being responsible for emotions (12).
Psychosis is a very common symptom of many different psychiatric and neurological disorders, especially those on the schizophrenia spectrum (13). Psychosis is defined as any symptoms that cause a detachment from reality as well as a loss of functioning in daily life (13). Psychosis is usually diagnosed through a collection of symptoms, rather than a single symptom (14). However, there is no precise definition of the mechanisms in place that cause psychosis or clinical symptoms of psychotic disorders (14). Although the use of SGA is prevalent in treating psychosis, there are also many researched adverse effects of SGAs that are metabolic, such as obesity and Type 2 diabetes (2, 15). Some of the medications relevant to the suppression of negative symptoms and cognitive deficits are clozapine, risperidone, and olanzapine (15). The focus of this review is to examine the changes to brain structure associated with the use of antipsychotic medications.
Methods
To examine the role of antipsychotics and their effects on brain structure, we conducted a literature search using PubMed, Google Scholar, and Science Direct databases. Screened keywords included: amplitude of low-frequency fluctuations, antipsychotics, aripiprazole, brain structure, caudate, changes in brain structure, clozapine, gray matter volume, olanzapine, putamen, regional homogeneity, risperidone. We obtained 62 total articles. References used were peer-reviewed and determined to be acceptable and reliable. References were evaluated by the number of citations used within the paper, the impact on the scientific community as determined by the number of times the paper was cited, and the impact factor of the journals in which the paper was published. Papers from journals with an impact factor of 2 or less were excluded from this review. There was no date range applied. Articles from journals not written in English were excluded from the analysis.
Discussion
Effects of Antipsychotic Use on Cortical Thickness
Cortical thickness is defined as the width of gray matter of the cortex (16) and is positively correlated with intellectual ability (17). In patients with schizophrenia, cortical thinning is demonstrated in
the fronto-temporo-parietal region (18). Without treatment, schizophrenia patients exhibit a gradual cortical thinning and flattening, which is associated with negative clinical outcomes (19). Two antipsychotics, olanzapine and clozapine, have been shown to impact cortical thickness.
Olanzapine, an antagonist of dopamine D2 receptors, has been shown to decrease cortical thickness, with the greatest decreases seen in patients who had sustained remission of schizophrenia compared to placebo patients (20). A key limitation of this finding was that in this investigation, patients were also using sertraline which has been previously proven to protect brain structure (21), complicating the results and not allowing for a complete understanding of olanzapine’s singular effects on cortical thickness. Animal studies using rats have demonstrated that antipsychotics themselves may directly cause a decrease in volume and thickness of the anterior cingulate cortex and cortex after chronic use of antipsychotics, illustrating that the changes seen in these studies may be attributable to the antipsychotics and not the underlying mechanisms of schizophrenia (22). Moreover, additional research showed that mid- to long-term use of atypical antipsychotics such as olanzapine led to significant decreases in cortical thickness in the frontal, temporal, and parietal areas (23). Decreases in cortical thickness were also correlated to worsened symptoms and cognitive deficits in these patients (23). A possible mechanism in humans involves the loss of synapses, which contributes to cortical thinning in patients with longterm use of atypical antipsychotics (24).
Lastly, clozapine, an antipsychotic used for treatmentresistant schizophrenia, has also been linked to cortical thinning in patients. Significant cortical thinning was noted after just 12 weeks of treatment with clozapine (25), while longer periods of clozapine administration resulted in further cortical thinning (26), indicating a progression of cortical thinning related to the duration spent using clozapine. Although the cortical thinning exhibited by antipsychotics has been correlated to decreases in cognitive function and overall worse outcomes for patients, the mechanism behind cortical thinning may be the brain’s attempt at self-preservation, possibly as a response to increased metabolic stress or with the pruning of malfunctioning neurons (27). All in all, atypical antipsychotics have been shown to cause accelerated cortical thinning
in schizophrenia patients with a noted correlation between cortical thinning and the time spent using these medications. Research involving animals demonstrates that chronic use of antipsychotic drugs causes decreases in volume and thickness of the cortex (22), but more research must be done to confirm the mechanisms behind antipsychotic medication-induced cortical thinning in human models. An important limitation of these studies is the inherent difficulty of formulating in vivo studies that account for the reality of schizophrenia. For example, the disease process of schizophrenia, including age of diagnosis, severity of disease, and effectiveness of treatment plan, have been shown to cause varying effects on brain structures in human patients (18), leading to an inconsistent baseline between patients with schizophrenia for study. Moreover, the high rates of comorbidity of other mental illnesses including depression and anxiety (21) may confound the results and make studying solely schizophrenia and the impact of these medications difficult.
Effects of Antipsychotic Use on Gray Matter Volume, Regional Homogeneity, and Amplitude of LowFrequency Fluctuations
Frequently prescribed antipsychotics, such as olanzapine, aripiprazole, and risperidone, share similar MOA. Defined as atypical antipsychotics (28), these drugs are SGAs which produce significantly fewer extrapyramidal symptoms than first-generation antipsychotics (FGA) (29), including restlessness, muscle contractions, tremors, and involuntary movements (30). However, SGAs promote the development of metabolic conditions such as insulin resistance, clinically significant weight gain, Type 2 diabetes mellitus, and dyslipidemia at much higher rates than FGAs (31). Mechanistically, SGAs target monoaminergic GPCRs and interfere with hypothalamic center activity. In doing so, SGAs modulate hypothalamic arcuate and paraventricular nuclei activity, leading to interactions at histamine (H1) receptors that favor weight gain and metabolic dysregulation (31). Additionally, SGAs blockade of serotonin (5-HT2C) receptors may be key in causing substantial weight gain (32). As SGAs, olanzapine, aripiprazole, and risperidone interact non-specifically with serotonin-dopamine receptors such as dopamine D2 and serotonin 5-HT2A/5-HT1A/5-HT2C receptors (33). Specifically, olanzapine is an antagonist of dopamine D2 receptors and works to decrease positive
symptoms of schizophrenia (34). Aripiprazole can be defined as a partial agonist of dopamine D2 and 5-HT2A receptors in the mesolimbic and mesocortical pathways (35). Lastly, risperidone functions again at the dopamine D2 but acts strongly on the 5-HT2A and 5-HT1A receptors as an antagonist (36).
When used as part of psychosis treatment, these antipsychotics have been shown to cause changes in brain structure, including gray matter volume (GMV), regional homogeneity (ReHo), and amplitude of lowfrequency fluctuation (ALFF) in humans (37), measured with structural magnetic resonance imaging. GMV can be used as a measure of an individual’s gray matter, which is involved in movement control, memory, and emotions (38). Studies have shown patients with schizophrenia have lower measures of cortical gray matter volume at presentation compared to patients with no diagnosed schizophrenia (39, 40), contributing to the lower quality of life and neurocognitive deficits (39). ALFF measures fluctuations in the blood-oxygenlevel-dependent signal. This shows spontaneous neural activity of specific regions in the brain (41). The reduction in measurement of ALFF is correlated to cognitive impairment in schizophrenia patients (41). Lastly, ReHo measures brain activity and evaluates the similarity of a given voxel and its nearest neighbors (42) to show the degree of brain damage in patients (43) and the ability of the brain to coordinate neuron activity (37).
After a course of treatment with antipsychotics such as olanzapine, aripiprazole, and risperidone, GMV is shown to increase across various brain areas including the right cerebellum, right inferior temporal gyrus, and left middle frontal gyrus (37). Importantly, these increases in GMV after antipsychotic treatment were correlated with abating clinical symptoms, indicating that these drugs can reduce the strength of schizophrenia symptoms (37). Contrarily, these antipsychotics also caused a reduced GMV after treatment in the left occipital lobe, gyrus rectus, and right orbital frontal cortex (37). The gyrus rectus and right orbital frontal cortex are found within the prefrontal cortex and contribute to working memory and social cognition (44). The noted decreases of gray matter within these areas contribute to the positive and negative symptoms exhibited by patients with schizophrenia. ALFF was shown to increase in the medial superior frontal gyrus, bilateral prefrontal and parietal cortex and right caudate nucleus with
respective p-values of 0.042, 0.043, and 0.002 (45). Additionally, increases were shown after short-term use of these antipsychotics in the anterior cerebellum, although no correlation was found between ALFF volumes and psychiatric symptoms, indicating a lack of clinical significance (46). Lastly, ReHo was shown to increase in the caudate with a p-value of 0.003 (47), showing an increase in coordinated activity after treatment with risperidone. The lack of clinical significance found could be attributed to small sample size (37, 40, 43, 44), the thickness of examined brain slices (39), not separating the data for specific subgroups (40), participants not being in the ideal resting-state (41), or to the previously mentioned limitations of human, in-vivo schizophrenia studies. Overall, treatment with SGAs such as olanzapine, aripiprazole, or risperidone has been shown to result in changes to GMV, ALFF, and ReHo, although future research is required to assess the clinical significance of the changes seen in ALFF and ReHo.
Effects of Antipsychotic Use on Caudate and Putamen Volume
The caudate and putamen, also known as the corpus striatum, are one of the main components of the basal ganglia and integrate information for motor control, cognition, and emotion (48). Schizophrenia causes abnormalities within these structures (49). Enlarged volumes of the caudate, +9.5%, and putamen, +15.9%, were seen across 15 human male patients with diagnosed schizophrenia (49). Similarly, in a study comparing 37 patients with schizophrenia and 37 patients without schizophrenia, MRI and traces on axial slices showed marked enlargements on the dorsal putamen (50). Although larger caudate and putamen volumes were observed, postmortem studies showed that there was nearly a 40% decrease in the number of mitochondria per synapse within these structures despite their greater volumes (51).
Abnormal function in the hyperconnectivity between the putamen and prefrontal cortex may cause auditory verbal hallucinations, as well as putamen regional and network functional deficits may be associated with the imbalance in neuromodulation of auditory verbal hallucinations in schizophrenia (52). Individuals with schizophrenia that are antipsychotic-naive showed larger bilateral putamen size overall, but the ventrally located putamen was associated with less severe symptoms (53). A mutation in the gene SLC39A8 increases GMV in the putamen, but this association
is greatly reduced in those with schizophrenia (54). Putamen volume is seen to increase when neurological and psychiatric disorders are present, but it is seen to decrease in males and females who are not affected by a disorder as they age (55). Overall, putamen volumes may be a longitudinal marker of treatment responsiveness and outcomes in individuals with schizophrenia (56).
Schizophrenia patients have altered spatial activity patterns and decreased intra-network functional connectivity in the caudate (57). The differences seen between gene expressions based on antipsychotic toxicology are different between different brain regions, which could be because of the different cell types affected (58). Antipsychotic medication was seen to influence caudate gene expression, and the caudate gene expression networks highlight interactions involving schizophrenia risk (59). Typical antipsychotics induced striatal enlargement in schizophrenia patients (60). Schizophrenia patients who have switched from typical antipsychotics to clozapine have shown decreased caudate volume (60). Clozapine displays a lower affinity to dopamine receptors compared to other antipsychotics (61), which may contribute to the neural changes seen when a patient switches from a typical antipsychotic to clozapine. All in all, antipsychotic use related to enlarged striata, while switching from typical antipsychotics to clozapine led to a decrease in caudate volume, showing the effects that different antipsychotics have on brain structures.
Conclusion
The changes seen in brain structure and volume after the use of antipsychotics are of particular concern to medication-prescribing providers and the patients using these medications. This is because the decreases in a patient’s measured cortical thickness after the use of antipsychotics are correlated to poor patient outcomes and must be better understood to help mitigate these changes in patients. Antipsychotic use was shown to decrease cortical thickness and the number of mitochondria per synapse despite the larger volume seen overall in the caudate and putamen. Antipsychotic use also caused decreased GMV and ALFF values while contributing to an increase in ReHo. Overall, the increase and decrease in GMV, ALFF, and ReHo across the putamen and other brain structures are of key importance and must be further understood,
as these changes impact patients in ways yet to be realized. A key limitation to the data presented are the issues associated with functional MRI data, as the results may be gratuitous and show false positive activations (62). Moreover, the studies referred to in this review are limited by small sample sizes, which limits the generalizability of the results. Utilizing larger sample sizes, with patients taking only the designated antipsychotic, would help fill the knowledge gaps and aid medical professionals in understanding more precisely how antipsychotics will impact their patients’ brain structure, leading to better patient outcomes.
Disclosures
The authors declare no relevant or material financial interests that relate to the research described in this paper.
Acknowledgments
We would like to express gratitude to our faculty member, Brian Piper, PhD. We would also like to express our gratitude to a course teaching assistant, Maria Tian, MBS, for her assistance. Lastly, we would like to convey our appreciation to Youssef Soliman, MD, PhD, for his time and support in reviewing this paper.
References
1. Shen WW. A history of antipsychotic drug development. Compr Psychiatry. 1999;40(6):407414.
2. Grinchii D, Dremencov E. Mechanism of action of atypical antipsychotic drugs in mood disorders. Int J Mol Sci. 2020;21:9532.
3. Howes O, Egerton A, Allan V, McGuire P, Stokes P, Kapur S. Mechanisms underlying psychosis and antipsychotic treatment response in schizophrenia: Insights from PET and SPECT imaging. Curr Pharm Des. 2009;15:2550–2559.
4. Larson ER, Ariano MA. D3 and D2 dopamine receptors: Visualization of cellular expression patterns in motor and limbic structures. Synapse 1995;20:325–337.
5. Abou-Setta AM. First-generation versus secondgeneration antipsychotics in adults: Comparative effectiveness. NCBI Bookshelf. 2012; Report No.:12-EHC054-EF.
6. Correll CU, Schooler NR. Negative symptoms in schizophrenia: A review and clinical guide for recognition, assessment, and treatment. Neuropsychiatr Dis Treat. 2020;16:519–534.
7. Chokhawala K. Antipsychotic medications. StatPearls Publishing. 2023.
8. Navari S, Dazzan P. Do antipsychotic drugs affect brain structure? A systematic and critical review of MRI findings. Psychol Med. 2009;39:1763–1777.
9. Yang C, Tang J, Liu N, Yao L, Xu M, Sun H, et al. The effects of antipsychotic treatment on the brain of patients with first-episode schizophrenia: A selective review of longitudinal MRI studies. Front Psychiatry. 2021;12.
14. Gaebel W, Zielasek J. Focus on psychosis. Dialogues Clin Neurosci. 2015;17:9–18.
15. MacKenzie NE, Kowalchuk C, Agarwal SM, Costa-Dookhan KA, Caravaggio F, Gerretsen P, et al. Antipsychotics, metabolic adverse effects, and cognitive function in schizophrenia. Front Psychiatry. 2018;9:180-187.
16. Tahedl, M. Towards individualized cortical thickness assessment for clinical routine. J Transl Med. 2020;18:151.
17. Menary K, Collins PF, Porter JN, Muetzel R, Olson EA, Kumar V, et al. Associations between cortical thickness and general intelligence in children, adolescents and young adults. Intelligence. 2013;41(5):597-606.
18. Yan J, Cui Y, Li Q, Tian L, Liu B, Jiang T, et al. Cortical thinning and flattening in schizophrenia and their unaffected parents. Neuropsychiatr Dis Treat. 2019;15:935-946.
19. Haren EM, Schnack HG, Cahn W, Heuvel MP, Lepage C, Evans AC, et al. Changes in cortical thickness during the course of
illness in schizophrenia. Arch Gen Psychiatry 2011;68(9):871–880.
20. Voineskos AN, Mulsant BH, Dickie EW, Neufeld NH, Rothschild AJ, Whyte EM, et al. Effects of antipsychotic medication on brain structure in patients with major depressive disorder and psychotic features: Neuroimaging findings in the context of a randomized placebo-controlled clinical trial. JAMA Psychiatry. 2020;77(7):674683.
21. Manji HK, Duman RS. Impairments of neuroplasticity and cellular resilience in severe mood disorders: Implications for the development of novel therapeutics. Psychopharmacol Bull 2001;35(2):5-49.
22. Vernon AC, Crum WR, Lerch JP, Chege W, Natesan S, Modo M, et al. Reduced cortical volume and elevated astrocyte density in rats chronically treated with antipsychotic drugs-linking magnetic resonance imaging findings to cellular pathology. Biol Psychiatry. 2014;75(12):982–990.
23. Feng R, Womer FY, Edmiston EK, Chen Y, Wang Y, Chang M, et al. Antipsychotic effects on cortical morphology in schizophrenia and bipolar disorders. Front Neurosci. 2020;14:579139.
24. Huang XF, Song X. Effects of antipsychotic drugs on neurites relevant to schizophrenia treatment. Med Res Rev. 2019;39:386–403.
25. Grant MQ, John L, Diederen K, Howes OD, Kempton MJ, Krajner F, et al. Subcortical volume reduction and cortical thinning 3 months after switching to clozapine in treatment resistant schizophrenia. Schizophrenia. 2022;14:1605.
26. Ahmed, M, Cannon DM, Scanlon C, Holleran L, Schmidt H, McFarland J, et al. Progressive brain atrophy and cortical thinning in schizophrenia after commencing clozapine treatment. Neuropsychopharmacol. 2015;40:2409–2417.
27. Templeton N, Kivell B, McCaughey-Chapman A, Connor B La Flamme AC. Clozapine administration enhanced functional recovery after cuprizone demyelination. PLoS One. 2019:14:12-16.
28. Chokhawala K, Stevens L. Antipsychotic medications. StatPearls Publishing. 2023.
31. Makary S, Abd El Moez K, Elsayed M, Hassan H. Second-generation antipsychotic medications and metabolic disturbance in children and adolescents. Egypt J Neurol. 2023;59:14-18.
32. Lord CC, Wyler SC, Wan R, Castorena CM, Ahmed N, Mathew D, Lee S, Liu C, Elmquist JK. The atypical antipsychotic olanzapine causes weight gain by targeting serotonin receptor 2C. J Clin Invest. 2017;127(9):3402-3406.
33. Chokhawala K, Stevens L. Antipsychotic medications. StatPearls Publishing. 2023.
34. Thomas K, Saadabadi A. Olanzapine. Statpearls Publishing. 2023.
35. Shapiro DA, Renock S, Arrington E, Chiodo LA, Liu LX, Sibley DR, et al. Aripiprazole, a novel atypical antipsychotic drug with a unique and robust pharmacology. Neuropsychopharmacol 2003;28:1400–1411.
36. McNeil SE, Gibbons JR, Cogburn M. Risperidone. StatPearls Publishing. 2023.
37. Yin PG, Zhao CH, Li YA, Liu XI, Chen L, Hong N. Changes in brain structure, function, and network properties in patients with first-episode schizophrenia treated with antipsychotics. Front Psychiatry. 2021;10:3389.
39. Gur RE, Turetsky BI, Bilker WB, Gur RC. Reduced gray matter volume in schizophrenia. Arch Gen Psychiatry. 1999;56(10):905–911.
40. Vita A, De Peri L, Deste G, Sacchetti E. Progressive loss of cortical gray matter in schizophrenia: A meta-analysis and meta-regression of longitudinal MRI studies. Transl Psychiatry. 2012;2(11):e190.
41. Wang P, Yang J, Yin Z, Zhang R, Sun J, Xu Y, et al. Amplitude of low-frequency fluctuation (ALFF) may be associated with cognitive impairment in schizophrenia: A correlation study. BMC Psychiatry. 2019;19:30-37.
42. Zang Y, Jiang T, Lu Y, He Y, Tian L. Regional homogeneity approach to fMRI data analysis. NeuroImage. 2004; 22(1):394–400.
43. Qiu X, Xu W, Zhang R, Yan W, Ma W, Xie S, Zhou M. Regional homogeneity brain alterations in schizophrenia: An activation likelihood estimation meta-analysis. Psychiatry Investig. 2021;18(8):709717.
44. Nestor PG, Nakamura M, Niznikiewicz M, Thompson E, Levitt JJ, Choate V, Shenton ME, McCarley RW. In search of the functional neuroanatomy of sociality: MRI subdivisions of orbital frontal cortex and social cognition. Soc Cogn Affect Neurosci. 2013;8(4):460-467.
45. Lui S, Li T, Deng W, Jiang L, Wu Q, Tang H, et al. Short-term effects of antipsychotic treatment on cerebral function in drug-naive first-episode schizophrenia revealed by "resting state" functional magnetic resonance imaging. Arch Gen Psychiatry. 2010;67(8):783–792.
46. Chen Y, Womer FY, Feng R, Zhang X, Zhang Y, Duan J, et al. A real-world observation of antipsychotic effects on brain volumes and intrinsic brain activity in schizophrenia. Front Neurosci. 2022;15(1):460-1035.
47. Hu ML, Zong XF, Zheng JJ, Pantazatos SP, Miller JM, Li ZC, et al. Short-term effects of risperidone monotherapy on spontaneous brain activity in first-episode treatment-naïve schizophrenia patients: A longitudinal fMRI study. Sci Rep. 2016;6:342-387.
48. Schröder H, Moser N, Huggenberger S. The mouse caudate putamen, motor system, and nucleus accumbens. Neuroanatomy of the mouse: Springer Cham. 2020;305–318.
49. Hokama H, Shenton ME, Nestor PG, Kikinis R, Levitt JJ, Metcalf D, et al. Caudate, putamen, and globus pallidus volume in schizophrenia: A quantitative MRI study. Psychiatry Res. 1995;61(4):209-229.
50. Buchsbaum MS, Shihabuddin L, Brickman AM, Miozzo R, Prikryl R, Shaw R, et al. Caudate and putamen volumes in good and poor outcome patients with schizophrenia. Schizophr Res. 2003;64(1):53-62.
51. Roberts RC. Postmortem studies on mitochondria in schizophrenia. Schizophr Res. 2017;187:17-25.
52. Cui LB, Liu K, Li C, Wang LX, Guo F, Tian P, et al. Putamen-related regional and network functional deficits in first-episode schizophrenia with
auditory verbal hallucinations. Schizophr Res 2016;173(1-2):13-22.
53. Chemerinski E, Byne W, Kolaitis JC, Glanton CF, Canfield EL, Newmark RE, et al. Larger putamen size in antipsychotic-naïve individuals with schizotypal personality disorder. Schizophr Res 2013;143(1):158-164.
54. Luo Q, Chen Q, Wang W, Desrivières S, Quinlan EB, Jia T, et al. Association of a schizophreniarisk nonsynonymous variant with putamen volume in adolescents: A voxelwise and genomewide association study. JAMA Psychiatry 2019;76(4):435-445.
55. Halkur Shankar S, Ballal S, Shubha R. Study of normal volumetric variation in the putamen with age and sex using magnetic resonance imaging. Clin Anat. 2017;30(4):461-466.
56. Mitelman SA, Canfield EL, Chu KW, Brickman AM, Shihabuddin L, Hazlett EA, et al. Poor outcome in chronic schizophrenia is associated with progressive loss of volume of the putamen. Schizophr Res. 2009;113(2-3):241-245.
57. Zheng Y, Wu C, Li J, Li R, Peng H, She S, et al. Schizophrenia alters intra-network functional connectivity in the caudate for detecting speech under informational speech masking conditions. BMC Psychiatry. 2018;18(1):90.
58. Perzel Mandell KA, Eagles NJ, Deep-Soboslay A, Tao R, Han S, Wilton R, et al. Molecular phenotypes associated with antipsychotic drugs in the human caudate nucleus. Mol Psychiatry. 2022;27(4):2061-2067.
59. Benjamin KJM, Chen Q, Jaffe AE, Stolz JM, Collado-Torres L, Huuki-Myers LA, et al. Analysis of the caudate nucleus transcriptome in individuals with schizophrenia highlights effects of antipsychotics and new risk genes. Nat Neurosci. 2022;25(11):1559-1568.
60. Lang DJ, Kopala LC, Vandorpe RA, Rui Q, Smith GN, Goghari VM, et al. Reduced basal ganglia volumes after switching to olanzapine in chronically treated patients with schizophrenia. Am J Psychiatry. 2004;161(10):1829–1836.
61. Khokhar JY, Henricks AM, Sullivan EDK, Green AI. Unique effects of clozapine: A pharmacological perspective. Adv Pharmacol. 2018;82:137-162.
62. Bennett CM, Baird AA, Miller MB, Wolford GL. Neural correlates of interspecies perspective taking in the post-mortem Atlantic Salmon: An argument for multiple comparisons correction.
J of Serendip and Unexpec Res. 2009;1(1):1-5.
The Role of Statins in the Prevention and Management of Alzheimer's Disease: A Focused Review
Taylor A. Copelin1*, Madison L. Kobzeff1*, Rachelle A. Franco1*, and Songul M. Gafurova1*
¹Geisinger College of Health Sciences, Scranton, PA 18509
*Master of Biomedical Sciences Program
Correspondence: tacopelin@som.geisinger.edu
Abstract
Alzheimer’s disease (AD) is a neurodegenerative disease that is characterized by a progressive decline in cognitive functions that primarily affects older adults. Alzheimer’s is pathophysiologically characterized by the accumulation of amyloid-beta (Aß) plaques, neuroinflammation, accumulation of tau proteins in neurofibrillary tangles (NFTs), and tissue loss. Statins, commonly prescribed to lower cholesterol in the blood, and their anti-inflammatory effects have gained attention for their potential impact on influencing AD progression. This literature review explores the relationship between the use of statins and AD, examining the effect of statins on cholesterol metabolism, neuroinflammation, apolipoprotein E (ApoE), and Aß production. Research in both human and rodent models demonstrate statins abilities to reduce inflammatory markers, oxidative stress, and Aß levels while modulating ApoE function. However, the precise mechanism for which statins slow the progression of neurodegeneration in AD is still unclear and warrants further research before clinical application.
Introduction
Alzheimer's disease (AD) has become increasingly prevalent in the United States, with over 6 million Americans living with the disorder, and research suggests this number will continue to rise (1). AD remains the most common type of dementia and is prevalent among 60–70% of dementia cases. AD is characterized by abnormal effects on cognition and behavior, with the underlying pathophysiology involving the manifestation of extracellular amyloid plaques, synaptic deterioration, and neuronal death (2, 3). A retrospective study conducted on 1.8 million individuals found that increased levels of low-density lipoprotein cholesterol (LDL-C) are a possible risk factor for AD (4). Amid these developments, statins,
such as atorvastatin, fluvastatin, lovastatin, and others are often prescribed for the management of hypercholesterolemia and the prevention of cardiovascular diseases (5). Statins have also gained recent attention for their potential neuroprotective effects due to their anti-inflammatory properties. The relationship between statin use and AD remains complex, with multiple studies suggesting that statin use may reduce the risk of developing the disease, while others have found no association. This review aims to examine the utility of statins in the management and prevention of AD, focusing on their effects on neuroinflammation, Aß production, and ApoE modulation, while also exploring the mechanisms through which statins may exert neuroprotective effects.
Methods
For this literature review, we used the Google Scholar and PubMed databases to search for articles. Neurodegeneration, statins, pleiotropic, and neuroprotective were keywords used when obtaining applicable articles starting in 1994 to present. Articles focused on the statins and AD relationship and were limited to articles published after 2004. Chemical structures were drawn using KingDraw.
Discussion
Mechanisms of Neurodegeneration and Cognitive Decline in AD
Amyloid-ß Plaques and Neurofibrillary Tangles
Typically occurring together, amyloid-ß plaques (Aß) and neurofibrillary tangles (NFTs) are the hallmarks of neurodegeneration in AD. Plaques result from proteolytic cleavage of the amyloid precursor protein (APP) into 40-43 amino acid-long peptides (6). The Aß peptide forms amyloid fibrils that aggregate, with Aß [1-40] being the most recognized for its role in AD-
affected brains (7). NFTs are aggregates of misfolded and hyperphosphorylated tau proteins (8). These intracellular aggregates disrupt the axoplasmic flow of neuronal communication and directly correlate to the degree of cognitive decline (8, 9). Multiple studies using both human and mouse models (transgenic mouse expressing human APP and tau transgenes (10)) have investigated imaging options to identify Aß and NFTs in the brain (11–13). The main objective of these studies was to identify a relationship between the two and neurodegeneration (10, 14). The transgenic mice expressing human APP and tau trans genes generated Aß and NFTs and subsequently suffered from neurodegeneration (10). During the period of 11 to 13 months, the mice developed significant tissue atrophy and neuronal loss associated with the increases in plaques and NFTs (10). Multitracer positron emission tomography (PET) imaging using Pittsburgh Compound B (PIB) and FDDNP probes allow images of the plaques and tangles to be produced for brain monitoring (11–13). PIB selectively binds to Aß, allowing for PET imaging to measure the binding in AD individuals using a standardized uptake value ratio (SUVR) for gray matter (11). FDDNP probes bind to Aß and NFTs, measuring the binding in AD brains using a SUVR for gray matter as well (11). Multitracer PET imaging using a combination of these probes can be useful in accurately tracking disease progression (11).
Oxidative Stress
The formation of Aß plaques and NFTs can be both initiated and exacerbated by the presence of free radicals (15, 16). Oxidative stress occurs when antioxidants are out of balance with the oxidants in the body, and these oxidants are formed when oxygen molecules have an unpaired electron. The superoxide radical (O2·−) is generated in the brain and removed by conversion to hydrogen peroxide (H2O2) by superoxide dismutases (17). Under normal circumstances, the protective antioxidant mechanisms are in balance with the formation of free radicals, keeping possible tissue damage and buildup under control (18). Brain tissue is composed of oxidation-sensitive lipids and is an organ with high O2 consumption, making it extremely
vulnerable to damage and high reactive oxygen species (ROS) (15, 16). The products of peroxidation are colocalized with Aß plaques, with two of the primary metabolites occurring from white and gray matter (16). There have been investigations looking into whether these lipid peroxidation products could be potential AD biomarkers. Isoprostanes are oxidized from adrenic acid and in the white matter of the brain (Figure 1) (16). Compared to non-AD controls, patients with AD have increased levels of 8-isoprostane in their blood (19, 20). Oxysterols are the oxidized product of normal cholesterols in the brain. In tissue samples from the frontal and occipital cortex, oxysterols are found in excess in AD-affected individuals (21). This links the oxidation of sterols to AD neurodegeneration. However, when observed in cerebrospinal fluid (CSF), isoprostanes were not found to be a viable biomarker (22). In CSF, isoprostanes increased with age independently from established AD CSF markers (Aß and NFTs) (22). Isoprostanes are a promising biomarker for the diagnosis of AD, but further research is needed to substantiate this conclusion and find other suitable biomarkers as well.
Neuroinflammation
Neuroinflammation is a hallmark of many other neurodegenerative diseases, with its main purpose lying in CNS protection. In AD, glial activation and proinflammatory factors influence disease progression (23). The blood-brain barrier (BBB) is a semi-permeable membrane that monitors what can cross the barrier to the brain or what cannot. Surface transporters on the BBB move Aß from the brain to the blood, but Aß can affect the expression
Figure 1. Chemical structures of adrenic acid and 8-isoprostane. Adrenic acid is oxidized in the brain (white matter) into 8-isoprostane due to oxidative stress. This isoprostane is a possible AD biomarker.
of tight junctions in the barrier (24). A major cause of disruption in the BBB is increased inflammatory cytokines. This disruption can then cause the accumulation of Aß, leading to barrier dysfunction and increased monocyte adhesion (24). Astrocytes and microglia produce cytokines and are the main inflammatory cells in the CNS (23, 24). Microglia use danger-associated molecular pathways (DAMPs) and pathogen-associated molecular pathways (PAMPs) to recognize, activate, and phagocytize Aß fibrils (23, 25). They then send chemokines to the site, and contrary to the normal mechanism, diseased brains show an upregulation of CCL2, -3, and -5 (25). These chemokines exacerbate the inflammatory response. In astrocytes near Aß plaques in diseased brains, CCL4 is upregulated (25).
Multiple rodent trials using transgenic mice have demonstrated a link between neuroinflammation and AD pathology. Both interleukins and inflammasomes, known for their role in inflammatory responses, are produced by the microglial cells in AD mouse models (25). However, the transgenic APP mouse model is not a completely accurate representation of human AD (26). APP knock-in mice showing the appropriate amyloid deposition were used, and it was found that these mice share common neuroinflammatory genes with humans (26). In these mice, the triggering receptors expressed on myeloid cells 2 (TREM2) were upregulated and used as Aß sensors to activate microglia, which in turn activate proinflammatory signals in the brain (26). In vivo, magnetic resonance spectroscopy (MRS) is the gold standard for early imaging of neuroinflammation and prevention of further damage (27, 28). MRS has shown that neuroinflammation is an early indication of AD and mainly precedes neurodegeneration (27, 28).
Overall, AD is a multifactorial disease, and there is no single, primary cause. It is a combination of factors like the ones discussed above and in different grades. Treatment mainly involves a regimen of medications like donepezil and memantine. Donepezil is an acetylcholinesterase inhibitor that enhances cholinergic transmission at the synapse (29). It also has an effect in downregulating microglial activation in the brain, reducing inflammation (29). If a patient is unable to take an acetylcholinesterase inhibitor, memantine can be used as a supplement. Memantine blocks the overactivation of the excitatory NMDA receptor, preventing neuronal damage (30).
Statins Lipid-Lowering and Pleiotropic Effects
Low-density lipoproteins are fat molecules that circulate in the blood carrying cholesterol for cellular processes and are deposited on arterial walls (31). Elevated LDL levels can cause a buildup of plaque on the arterial walls. This is the main cause of coronary heart disease (32). LDLs are prone to oxidation, and oxLDL is pro-inflammatory, leading to endothelial dysfunction that exacerbates the deposition of plaques (32). High-density lipoprotein (HDL) is a major component of reverse cholesterol transport. HDL transports cholesterol from peripheral tissues into the liver while also counteracting the oxidation of LDL (32). When LDL outweighs HDL, the counteractive effects of the HDL are not enough to prevent oxidation and deposition of plaques, in turn increasing the risk of heart disease. This also gives LDL the moniker “bad cholesterol.” Very low-density lipoprotein triglycerides are produced in the liver and contain a large protein called apoprotein B (33). Through interaction with a lipase and apoprotein E, VLDL converts to LDL unless apoprotein B is targeted for degradation (33).
Statins, like atorvastatin and rosuvastatin, are the main therapeutic drugs used to lower high cholesterol levels in the blood, and they do so by reducing cholesterol biosynthesis in their target organ, the liver (hepatoselective) (34–36). Mechanistically, this occurs through the inhibition of hydroxymethylglutaryl-CoA (HMG-CoA) reductase (35-39). HMG-CoA reductase converts HMG-CoA to mevalonate, a precursor to cholesterol (38). HMG-CoA is a liver enzyme that is regulated by AMPK and activated by a phosphoprotein phosphatase (38). Inhibition of this enzyme results in an increase in LDL clearance from the blood by the liver in patients with high cholesterol, while in patients with hyperlipidemia, the inhibition results in a decrease in both cholesterol and triglycerides by decreasing the production of the transport apoprotein B100 (37). However, with growing evidence of statins exhibiting neuroprotective effects, there may be another mechanism of action.
One such mechanism is an anti-inflammatory downregulation of neuroinflammation. Neuroinflammation, as discussed above, is one of the major mechanisms of neurodegeneration in AD. Statins, except pravastatin which cannot cross the BBB, can bind to leukocyte function antigen-1(LFA-1) and prevent adhesion to intracellular adhesion molecule-1 (ICAM-1) (35, 40). As a result, ICAM-1
signals are reduced, downregulating the inflammatory response (35, 41, 42). In an investigation using Sprague-Dawley rats, fluvastatin reduced both cellular adhesion molecules and oxidative stress, independent of its lipid-lowering effects (42). Another study using atorvastatin and lovastatin in mice and guinea pigs demonstrated reduced substance P and calcitonin gene-related peptide (CGRP) in the dorsal root ganglia (43). These are two main peptides that can induce a proinflammatory response.
AD Responses to Statin Effects
Cholesterol Metabolism in the Brain
In a trial treating individuals with AD with 80 mg of atorvastatin, the statin was shown to reduce cholesterol levels and positively impact the individual's Alzheimer's Disease Assessment Scale-Cognitive (ADAS-Cog) score (44). Compared to the placebo group, atorvastatin use showed a significant increase in the ADAS-Cog score after 6 months, while also showing mild-to-moderate AD cases may incur the largest benefits from statin therapy (44). Cholesterol plays a vital role in neuronal function and maintenance. The highest concentration of cholesterol is within the brain at approximately 20% (45). Of this 20%, cholesterol biosynthesis occurs mainly by glial cells; however, it is also partially synthesized by neurons. Glial cells produce roughly two times more cholesterol than neurons. (45, 46). However, the cholesterol supply is observed at a greater value in astrocytes than in neurons (45). Cholesterol is transferred in the brain with the use of a protein called apolipoprotein E (ApoE), which metabolizes lipoprotein-bound cholesterol (45). ApoE is produced mainly through astrocytes and functions through shuttling cholesterol from glial cells to neurons; this shuttling system plays a large role in the formation of efficient synapses (47).
As mentioned, the average lifespan is expanding; therefore, it is important to understand the impact of cholesterol as we age. The aging process has different effects on different regions of the brain, with the greatest decrease in cholesterol being approximately 40% (46). A study was conducted on 118 brains, ranging from age 20 to 100 years, focused on the lipid composition of the frontal and temporal lobes found that after age 80 is when significant lipid loss occurs (48). The cholesterol and phospholipid concentrations decreased in both the frontal and temporal lobes, with the temporal lobe having about 20% more reduction
than the frontal lobe (48). The evidence also showed that phospholipids decreased twice as much as cholesterol as the brain aged, which affects astrocytes due to their cholesterol/phospholipid content (48). The reduced supply of cholesterol through astrocytes was mutually exclusive with the reduced cholesterol within myelin membranes (48). As the brain ages, the density of the synapse declines while the possibility of chronic neuroinflammation becomes more apparent (49). These factors in turn damage the repair processes that astrocytes possess (49).
Statin Therapy and ApoE
Genotype
ApoE is the major lipid and cholesterol carrier in the central nervous system (CNS) and one of the most important risk factors for AD (50). The isoform ApoE4 allele is one of the most important genetic risk factors for AD, with individuals who possess the allele performing lower on cognitive tests than those without (44). Statin use has shown the most promising outcomes on individuals with the expression of the ApoE4 allele demonstrating its effectiveness through the modulation of that allele (51). Furthermore, statin use has been seen to be effective in the treatment of AD through the GTPase isoprenylation pathway, resulting in the reduction of Aß formation (51). Although it does have its limitations, it is important to mention the possible differences in outcome in the interaction of an individual's sex and ApoE4 genotype. An analysis of these measures noticed that males with the ApoE4 genotype could benefit more so than their female counterparts. Moreover, these authors also reported that carriers of the ApoE4 genotype have a decreased risk of AD (52).
Statin Therapy Effects on AD Characteristics
The exact mechanism by which statins may slow the development of AD characteristics is unclear, as the evidence is conflicting. This disagreement among studies could pertain to the groups in which statins are categorized: fungal-derived, synthetic-derived, lipophilic, and hydrophilic. Lovastatin and simvastatin are among the most popular when studying statin effects on the brain due to their ability to cross the BBB (53, 54). Simvastatin has been shown to have neuroprotective effects by preventing Aß-induced production of interferon-γ (causing neuronal damage), while lovastatin enhances Wnt signaling that protects against Aß neurotoxicity (55). Statin users also had a lower incidence of neurodegenerative disorders compared to non-users, with pitavastatin showing the
strongest reduction of incidence among the 8 statin types (56).
Statin use has neuroprotective effects and has the possibility to decrease the progression of AD by lowering cholesterol levels and reducing plaque formation (57). An analysis conducted on 40 AD male mice treated with atorvastatin demonstrated a reduction in Aß production and tau hyperphosphorylation, leading to increased learning and memory. The authors found increased levels of phosphorylated AKT and glycogen synthase kinase 3ß (GSK3ß) in the hippocampus (58). The atorvastatin effectively reversed the learning deficit in AD mice when compared to control treated mice (saline) (58). This finding is significant as evidence suggests that these two signaling pathways are impacted by Aß exposure (58, 59). Research has shown the impact that GSK3ß has on the role of AD progression as the hyperfunction GSK3ß is marked by common AD characteristics such as memory impairment and inflammatory responses (59, 60). This information could be used as a basis for future studies examining mechanisms by which we can combat the progression of AD.
Conclusion
AD is a devastating neurodegenerative disease characterized by progressive cognitive decline, primarily driven by the accumulation of Aß plaques, NFTs, synaptic dysfunction, neuroinflammation, and tissue atrophy. As the global population ages, finding effective treatments to slow or halt the progression of this disease is paramount. While statins offer a promising future for mitigating the underlying mechanisms of AD, the evidence of their effectiveness remains partially inconclusive. Limitations in the literature lie in the common reliance on animal studies and the unknown side effects of statin use. Currently, investigations primarily exist in animal model studies which show positive results with reduced inflammation and Aß production but may not replicate the complexity of AD in humans. The mechanism of protection that would benefit humans still requires evidence. The benefits of statin use for AD treatment must be weighed against the risks of side effects and uncertain future implications. As future research occurs, a more nuanced understanding of which AD patients may benefit from statin treatment is required. Future studies should prioritize larger scale,
longitudinal studies in order to understand the effects of statins in more diverse populations and the true mechanism of action for more targeted prevention. At this time, statin treatments should be used with caution and as a more individualized approach for AD prevention and management.
Disclosures
There is no financial relationship between this paper’s authors and any institution mentioned.
Acknowledgments
We would like to thank Brian Piper, PhD, and Gia Fevrier, MBS, at Geisinger Commonwealth School of Medicine for their support on this project.
4. Iwagami M, Qizilbash N, Gregson J, Douglas I, Johnson M, Pearce N, et al. Blood cholesterol and risk of dementia in more than 1·8 million people over two decades: a retrospective cohort study. Lancet Healthy Longev. 2021;2(8):e498-e506.
5. Sizar O, Khare S, Patel P, Talati R. Statin Medications. Treasure Island, FL: StatPearls Publishing; 2024.
6. Villemagne VL, Rowe CC, Macfarlane S, Novakovic KE, Masters CL. Imaginem oblivionis: The prospects of neuroimaging for early detection of Alzheimer's disease. J Clin Neurosci 2005;12(3):221-30.
7. Friedrich RP, Tepper K, Rönicke R, Soom M, Westermann M, Reymann K, et al. Mechanism of amyloid plaque formation suggests an intracellular basis of Aß pathogenicity. Proc Natl Acad Sci U S A. 2010;107(5):1942-7.
8. Braak H, Del Tredici K. Encyclopedia of Movement Disorders. Amsterdam: Elsevier Academic Press;2010
10. Paulson JB, Ramsden M, Forster C, Sherman MA, McGowan E, Ashe KH. Amyloid plaque and neurofibrillary tangle pathology in a regulatable mouse model of Alzheimer's disease. Am J Pathol 2008;173(3):762-72.
11. Shin J, Lee S-Y, Kim S-H, Kim Y-B, Cho S-J. Multitracer PET imaging of amyloid plaques and neurofibrillary tangles in Alzheimer's disease. Neuroimage. 2008;43(2):236-44.
12. Shoghi-Jadid K, Small GW, Agdeppa ED, Kepe V, Ercoli LM, Siddarth P, et al. Localization of neurofibrillary tangles and beta-amyloid plaques in the brains of living patients with Alzheimer disease. Am J Geriatr Psychiatry. 2002;10(1):24-35.
13. Mathis CA, Wang Y, Klunk WE. Imaging ß-amyloid plaques and neurofibrillary tangles in the aging human brain. Curr Pharm Des. 2004;10(13):146992.
14. Kuznetsov IA, Kuznetsov AV. How the formation of amyloid plaques and neurofibrillary tangles may be related: a mathematical modelling study. Proc Math Phys Eng Sci. 2018;474(2210).
15. Huang W-J, Zhang X, Chen W-W. Role of oxidative stress in Alzheimer's disease. Biomed Rep. 2016;4(5):519-522.
16. Peña-Bautista C, Baquero M, Vento M, CháferPericás C. Free radicals in Alzheimer's disease: Lipid peroxidation biomarkers. Clin Chim Acta 2019;491:85-90.
17. Aruoma OI. Free radicals, oxidative stress, and antioxidants in human health and disease. J Am Oil Chem Soc. 1998;75(2):199-212.
18. Smith MA, Rottkamp CA, Nunomura A, Raina AK, Perry G. Oxidative stress in Alzheimer’s disease. Biochim Biophys Acta. 2000;1502(1):139-144.
19. Dias HKI, Brown CLR, Polidori MC, Lip GYH, Griffiths HR. LDL-lipids from patients with hypercholesterolaemia and Alzheimer's disease are inflammatory to microvascular endothelial cells: Mitigation by statin intervention. Clin Sci (Lond). 2015;129(12):1195-206.
20. Ademowo OS, Dias HKI, Milic I, Devitt A, Moran R, Mulcahy R, et al. Phospholipid oxidation and carotenoid supplementation in Alzheimer’s disease patients. Free Radic Biol Med. 2017;108:77-85.
21. Testa G, Staurenghi E, Zerbinati C, Gargiulo S, Iuliano L, Giaccone G, et al. Changes in brain oxysterols at different stages of Alzheimer's disease: Their involvement in neuroinflammation. Redox Biol. 2016;10:24-33.
22. Peskind ER, Li G, Shofer JB, Millard SP, Leverenz JB, Yu C-E, et al. Influence of lifestyle modifications on age-related free radical injury to brain. JAMA Neurol. 2014;71(9):1150-4.
23. Calsolaro V, Edison P. Neuroinflammation in Alzheimer's disease: Current evidence and future directions. Alzheimers Dement. 2016;12(6):719-32.
24. Zhang F, Jiang L. Neuroinflammation in Alzheimer’s disease. Neuropsychiatr Dis Treat. 2015;11:243-256.
25. Heneka MT, Carson MJ, Khoury JE, Landreth GE, Brosseron F, Feinstein DL, et al. Neuroinflammation in Alzheimer's disease. Lancet Neurol. 2015;14(4):388-405.
26. Saito T, Saido TC. Neuroinflammation in mouse models of Alzheimer’s disease. Clin Exp Neuroimmunol. 2018;9(4):211-218.
27. Chaney A, Williams SR, Boutin H. In vivo molecular imaging of neuroinflammation in Alzheimer's disease. J Neurochem. 2019;149(4):438-451.
28. Song T, Song X, Zhu C, Patrick R, Skurla M, Santangelo I, et al. Mitochondrial dysfunction, oxidative stress, neuroinflammation, and metabolic alterations in the progression of Alzheimer’s disease: A meta-analysis of in vivo magnetic resonance spectroscopy studies. Ageing Res Rev. 2021;72.
29. Kumar A, Gupta V, Sharma S. Donepezil. Treasure Island, FL: StatPearls Publishing; 2024.
30. Kuns B, Rosani A, Patel P, Varghese D. Memantine. Treasure Island, FL: StatPearls Publishing; 2024.
31. Pirahanchi Y, Sinawe H, Dimri M. Biochemistry, LDL Cholesterol. Treasure Island, FL: StatPearls Publishing; 2024.
32. Rosenson RS. Statins in atherosclerosis: Lipidlowering agents with antioxidant capabilities. Atherosclerosis. 2004;173(1):1-12.
33. Ginsberg HN. Effects of Statins on Triglyceride Metabolism. Am J Cardiol. 1998;81(4A):32B-35B.
34. Stancu C, Sima A. Statins: mechanism of action and effects. J Cell Mol Med. 2001;5(4):378-87.
35. van der Most PJ, Dolga AM, Nijholt IM, Luiten PGM, Eisel ULM. Statins: Mechanisms of neuroprotection. Prog Neurobiol. 2009;88(1):6475.
36. Feingold KR. Cholesterol Lowering Drugs. Endotext. South Dartmouth, MA;2024.
37. Sirtori CR. The pharmacology of statins. Pharmacol Res. 2014;88:3-11.
38. Craig M, Yarrarapu SNS, Dimri M. Biochemistry, Cholesterol. Treasure Island, FL: StatPearls Publishing; 2024.
39. Tandon V, Bano G, Khajuria V, Parihar A, Gupta S. Pleiotropic effects of statins. Indian J Pharmacol. 2005;37(2):77-85.
40. Rezaie-Majd A, Prager GW, Bucek RA, Schernthaner GH, Maca T, Kress H-G, et al. Simvastatin reduces the expression of adhesion molecules in circulating monocytes from hypercholesterolemic patients. Arterioscler Thromb Vasc Biol. 2003;23(3):397-403
41. Romano M, Mezzetti A, Marulli C, Ciabattoni G, Febo F, Di lenno S, et al. Fluvastatin reduces soluble P-selectin and ICAM-1 levels in hypercholesterolemic patients: Role of nitric oxide. J Investig Med. 2000;48(3):183-9.
42. Katoh M, Kurosawa Y, Tanaka K, Watanabe A, Doi H, Narita H. Fluvastatin inhibits O2− and ICAM1 levels in a rat model with aortic remodeling induced by pressure overload. Am J Physiol Heart Circ Physiol. 2001;281(2):H655-60.
43. Wood WG, Eckert GP, Igbavboa U, Müller WE. Statins and neuroprotection: A prescription to move the field forward. Ann N Y Acad Sci 2010;1199:69-76.
44. Sparks DL, Connor DJ, Sabbagh MN, Petersen RB, Lopez J, Browne P. Circulating cholesterol levels, apolipoprotein E genotype and dementia severity influence the benefit of atorvastatin treatment in Alzheimer's disease: Results of the Alzheimer's Disease Cholesterol-Lowering Treatment (ADCLT) trial. Acta Neurol Scand Suppl. 2006;185:3-7.
45. Jin U, Park SJ, Park SM. Cholesterol metabolism in the brain and its association with Parkinson's disease. Exp Neurobiol. 2019;28(5):554-567.
46. Posse de Chaves E, Narayanaswami V. Apolipoprotein E and cholesterol in aging and disease in the brain. Future Lipidol 2008;3(5):505-530.
47. Lütjohann D. Cholesterol metabolism in the brain: Importance of 24S-hydroxylation. Acta Neurol Scand Suppl. 2006;185:33-42.
48. Svennerholm L, Bostrom K, Jungbjer B, Olsson L. Membrane lipids of adult human brain: lipid composition of frontal and temporal lobe in subjects of age 20 to 100 years. J Neurochem. 1994;63(5):1802-11.
49. Saher G. Cholesterol metabolism in aging and age-related disorders. Annu Rev Neurosci. 2023;46:59-78.
50. Fernández-Calle R, Konings SC, Frontiñán-Rubio J, García-Revilla J, Camprubí-Ferrer L, Svensson M, et al. APOE in the bullseye of neurodegenerative diseases: Impact of the APOE genotype in Alzheimer’s disease pathology and brain diseases. Mol Neurodegener. 2022;17(1):62.
51. Alsubaie N, Al-Kuraishy HM, Al-Gareeb AI, Alharbi B, De Waard M, Sabatier J-M. Statins use in Alzheimer disease: Bane or boon from frantic search and narrative review. Brain Sci. 2022;12(10):1290.
52. Dagliati A, Peek N, Brinton RD, Geifman N. Sex and APOE genotype differences related to statin use in the aging population. Alzheimers Dement (N Y). 2021;7(1):e12156.
53. Geifman N, Brinton RD, Kennedy RE, Schneider LS, Butte AJ. Evidence for benefit of statins to modify cognitive decline and risk in Alzheimer’s disease. Alzheimers Res Ther. 2017;9(1):10.
54. Rezakhani L, Salimi Z, Zarei F, Moradpour F, Khazaei MR, Khazaei M, et al. Protective effects of statins against Alzheimer disease. Ewha Med J. 2023;46(4):e17.
55. Ling L, Zhang W, Cheng S, Cao D, Parent M. Isoprenoids and related pharmacological interventions: Potential application in Alzheimer's disease. Mol Neurobiol. 2012;46(1):64-77.
56. Torrandell-Haro G, Branigan GL, Vitali F, Geifman N, Zissimopoulos JM, Brinton RD. Statin therapy and risk of Alzheimer’s and age-related neurodegenerative diseases. Alzheimers Dement (N Y). 2020;6(1):e12108.
57. Xuan K, Zhao T, Qu G, Liu H, Chen X, Sun Y. The efficacy of statins in the treatment of Alzheimer's disease: A meta-analysis of randomized controlled trial. Neurol Sci. 2020;41(6):1391-1404.
58. Zhou D, Liu H, Li C, Wang F, Shi Y, Liu L, et al. Atorvastatin ameliorates cognitive impairment, Aβ1-42 production and Tau hyperphosphorylation in APP/PS1 transgenic mice. Metab Brain Dis. 2016;31(3):693-703.
59. Reddy PH. Amyloid beta-induced glycogen synthase kinase 3ß phosphorylated VDAC1 in Alzheimer's disease: Implications for synaptic dysfunction and neuronal damage. Biochim Biophys Acta. 2013;1832(12):1913-21.
60. Hooper C, Killick R, Lovestone S. The GSK3 hypothesis of Alzheimer's disease. J Neurochem. 2008;104(6):1433-9.
A Narrative Review of the Bifrontal Decompressive Craniotomy in Clinical Practice
Julian Michael Burwell1†
¹Geisinger College of Health Sciences, Scranton, PA 18509
†Doctor of Medicine Program
Correspondence: jburwell1@som.geisinger.edu
Abstract
The bifrontal approach, pioneered by Kjellberg in 1971, remains one of the few anterior approaches to the skull, yet no unified review of its historical and modern use currently exists. A literature search was conducted using PubMed, focusing on adult studies involving bifrontal decompressive craniotomy (DC). Bifrontal DC is indicated for severe traumatic brain injury with frontal lesions, elevated intracranial pressure refractory to medical x management, and certain cases of cerebrovascular accidents. Randomized trials, including DECRA and RESCUEicp, have demonstrated the procedure's efficacy in reducing mortality and intracranial pressure, though long-term neurological outcomes remain mixed. Military neurosurgeons have reported success with early bifrontal craniectomy in combat trauma, and civilian studies highlight its utility in managing traumatic bifrontal contusions. Surgeon preference and individual skill were frequently cited as deciding factors in the approach selected. Despite its versatility, poor rates of upper recovery have been reported, along with high rates of cerebrospinal fluid (CSF) leaks and wound complications. Despite its drawbacks, the bifrontal approach remains a viable option in the neurosurgeon's toolbox, allowing wide exposure and straightforward surgical planning, thus minimizing retraction injury. Future research and clarification could be directed towards the management of the superior sagittal sinus laceration and optimized placement of the posterior burr holes.
Introduction
Dr. Raymond Kjellberg published his bifrontal approach to DC in 1971 (1). While decompressive cranial openings have been traced back to ancient human history (2, 3), his bi-hemispheric decompression technique was argued to be a more physiological approach and improve upon the
unilateral decompression due to its reduced retraction injury and ability to simultaneously decompress both hemispheres. The bifrontal craniotomy positions the patient in a neutral supine position, with minimal head rotation, which increases the visibility of midline structures and can be ideal for frontal midline masses. Kjellberg's initial investigation had high mortality (82%), and since then, there have only been a few studies on the use and implementation of the Kjellberg procedure (1).
Methods
To evaluate the use of bifrontal craniotomy in clinical practice, PubMed was searched using the terms ("Bifrontal craniectom*" OR "Bifrontal Craniotom*") on Dec. 15, 2024, yielding 364 results. Search results were screened to exclude case reports, abstracts, and gray literature. Adult investigations were included in cases where patients received bifrontal DC or craniotomy, including retrospective, prospective, clinical trials, and case-controlled analyses. Reviews were included in the discussion if they presented novel shifts in thinking on the application of the bifrontal approach. Snowballing was utilized to identify seminal works within the field of neurosurgery that were not otherwise captured by the primary literature search.
Discussion
Clinical Indications
Current clinical guidelines endorse a limited therapeutic application of the bifrontal approach. The Brain Trauma Foundation (BTF) suggests DC for severe traumatic brain injury (TBI) cases (Glasgow Coma Scale [GCS] <9) with acute subdural hematomas (aSDH) and/or diffuse hematoma burden (4). Bifrontal DC is reserved for bilateral frontal lesions. Primary prophylactic DC is considered when elevated ICP is expected postoperatively. Secondary therapeutic DC use can be for patients with cerebrovascular accident
(CVA) or who have refractory post-traumatic cerebral edema after initial medical management has failed. The bifrontal DC is also indicated for cases of epidural hematoma (EDH) and penetrating head injuries affecting the frontal lobes.
The understanding of TBI severity is rapidly changing, including the recent new CBI-M criteria, consisting of 4 pillars: clinical, biomarkers, imaging, and modifiers (5). Emerging evidence suggests patient-specific factors also influence DC outcomes. Apolipoprotein ε4 genotype has been associated with increased risk of poor neurological recovery following TBI, though specific data for its implications on DC remain limited (6). Additionally, emerging research on inflammatory biomarkers such as S100 and GFAP levels may predict which patients are most likely to benefit from early surgical intervention versus medical management (7). Age-related differences in brain elasticity and cerebrospinal fluid dynamics also influence the effectiveness of decompressive procedures, with younger patients typically demonstrating better tolerance for the procedure (8).
A 2008 review identified 3 chief areas of clinical use for DC: TBI, malignant middle cerebral artery (MCA) infarction, and traumatic subarachnoid hemorrhage (SAH) (9). The investigation found DC anecdotally valuable for treating increased intracranial pressure unresponsive to conventional treatment modalities. Notably, the investigators favored an individualized approach depending on the surgeon's expertise, given the lack of clear evidence at the time supporting the use of DC in TBI pathologies.
Emergency surgical evacuation is recommended for patients with a GCS of ≤8 and abnormal imaging findings of a SDH greater than 10 mm or midline shift greater than 5 mm on non-contrast Computed Tomography (CT), who also exhibit 1 of 3 other indicators: 1) Neuroworsening with a decrease of 2 points in GCS after admission; 2) Asymmetric or dilated pupils; and 3) Intracranial pressure (ICP) exceeding 20 mmHg (10–13). Military neurosurgeons, drawing from high-volume combat trauma experience, have demonstrated a lower threshold for early DC in ASDH cases (14–16).
Clinical Trials
Since 2000, there have been 3 randomized clinical trials (17–19) on the use and efficacy of decompressive craniectomies (17–19). In all 3 cases, DC was used
as a treatment profile for severe TBI (sTBI). These include the DECRA Trial (2011) (17), the RESCUEicp Trial (2016) (18), and the RESCUE-ASDH Trial (2023) (19). Both the DECRA and RESCUEicp trials named the bifrontal craniotomy as one of their therapeutic approaches.
In the DECRA trial (n=155) (17), the comparison between decompressive craniectomy and standard care revealed that patients undergoing craniectomy experienced less time in the intensive care unit (median 11 days DC group vs. 15 days standard care, p-value< 0.001). Furthermore, the DC patients had lower average intracranial pressures (14.4 vs. 19.1, p-value < 0.001) and spent less with intracranial pressures above the 20-mmHg treatment threshold (average 9.2 hrs DC group, vs. 30 hrs standard care, p-value < 0.001). Unfortunately, patients with DC exhibited worse scores on the Glasgow Outcome Scale Extended (GOSE) and had a higher risk of unfavorable outcomes. Mortality was similar in both groups at 6 months (patients with craniectomy (19%) and patients with standard-care group (18%)). The DECRA trial cited inconsistent findings with nonrandomized trials, such as those from 1997 (20), 2001 (21), and the continued use of bifrontal DC for post-traumatic refractory intracranial hypertension. Notably, while these earlier investigations reported increased levels of upper recovery, they did not have a comparison group and had relatively low numbers of subjects (n=35 and n=20).
In 2016, the RESCUEicp trial (n=408) (18) showed significant differences in the GOSE distribution between the surgical and medical intervention groups. At 6 and 12 months, the surgical group exhibited that the decompressive craniectomy group had lower mortality rates, lower severe disability, and upper severe disability when compared to receiving medical care. However, surgical intervention resulted in a higher rate of vegetative states, and the rates of moderate disability and good recovery were similar between the two groups. Surgical treatment for ICP management was randomized between unilateral hemicraniectomy and bifrontal craniectomy. The investigation was pivotal in validating surgical intervention as an option for trauma cases due to the applicability of DC as first-line therapy for TBI and generalizability due to the sample size not demonstrated in previous analyses. In this trial, mortality was reduced by 45% over the medical treatment-only group.
A Cochrane Review of DC for the treatment of refractory high ICP in TBI published in 2019 examined 3 investigations (22) (n=590) to evaluate the utility of DC combined with standard care compared to standard care alone in treating TBI. These were a 2001 report (23) (n=27), the previously discussed DECRA 2011, and the RESCUEicp trial 2016. When metaanalysis techniques were applied, the researchers concluded that DC can reduce mortality in TBI cases. However, the results were inconclusive about the long-term neurological outcomes associated with the use of craniectomies. Furthermore, due to TBI heterogeneity, the investigators were unable to identify clear causes and effects from adverse events along the TBI care trajectories.
Bifrontal Craniectomy in Current Practice: Military Perspectives
Much of the data compiled on bifrontal craniotomy comes from military neurosurgeons who have utilized the bifrontal technique more often than their civilian counterparts (14–16).
As reported in a 2010 analysis (14), of the 28 DCs performed between 2007 and 2009, 7 were bifrontal DCs, as recorded in neurosurgical logs from Afghanistan. Of the cases highlighted, one involved a 14-year-old civilian who was struck by shrapnel and had an associated penetrating head wound. The team reported that the surgical indication was severe bilateral frontal lobe swelling (as seen on CT). They discussed the necessity of making the craniotomy as wide as possible to prevent acute laceration to the herniated portions of the brain on the cut bone edges. After surgery, the patient was neurologically intact but emotionally labile at the 3-month followup. The investigators also discussed the advantage of an alternative approach that leaves a strip of bone over the dural sinus to preserve venous flow from the perfused skull. In this case, the surgeons additionally performed a temporal lobectomy to allow for additional brain expansion to prevent secondary brain injury. The clinicians described that bifrontal craniectomy was preferred in various cases with bilateral swelling of either the frontal lobes or the anterior portions of the temporal lobes. Critically, they discussed the value of CT angiograms in diagnosis when MRI is not available, citing upwards of 30% vascular injury in cases of penetrating head injury (24). Notably, the researchers stress that hemodynamic stability was the priority regardless of the early or late decision for DC.
In a 2010 investigation (15) of the use of DC for patients transferred to Walter Reed Army Medical Center and National Naval Medical Center in Bethesda, Maryland, records of combined Armed Forces cases from 2003 to 2008 were assessed. In this review of cases, transferred patients were referred for bifrontal DCs in the case of increased ICP during transportation. The team also mentions the challenges associated with an inferior size of cranial opening, in which case the craniotomy should be as wide as possible. Insufficient bone removal resulted in the need for follow-up osteotomies and the initiation of moderate hypothermia to control refractory bouts of intracranial hypertension.
Bifrontal Craniectomy
in Current Practice: Civilian Perspectives
More recently, a 2023 report discusses the utility of the bifrontal DC in cases of traumatic bifrontal contusions (TBC) with significant edema (25). In their investigation (n=53), of the 26 patients with TBCs who qualified for early surgical decisions based on current protocols, 24 had bifrontal DCs. Of the 13 who qualified for observation followed by late surgery, 12 had bifrontal DC. The researchers had mixed results due to the unpredictable clinical course of these patients and advocated for a tailored individualized approach to the use of the bifrontal DC. As opposed to the operations used in predominately sTBI described by military neurosurgeons, this team describes the primary subject of their analysis as a victim of lowvelocity impact in motor vehicle accidents (MVA) with clinically mild (GCS 15-13) and moderate (GCS 12-9) TBI. For the 13 patients who qualified for late surgery, the most consistent predictor was clinical deterioration (GCS drop of 1–2) and radiographic progression an average of 1.53 days after the initial trauma. The investigators note that the temporality associated with these findings was not statistically significant.
Overall, these findings agreed with a 2011 investigation (26) that earlier surgical decision was associated with better outcomes for the patients who underwent DC for severe post-traumatic cerebral contusions, a phenomenon they dubbed "blossoming of contusions," which was associated with least in part with rapidly deteriorating cognitive status associated with contusions surrounded by rapidly expanding edema. This report also asserts that a low threshold should be set for surgical intervention for patients with
TBC on repeat CT once the overall burden of contusion is greater than 30 ml. For these patients, the average volume of bifrontal contusion expanded from 24 to 43 ml 2 days after surgery. This analysis lacked specific outcome measures other than intrahospital mortality.
As of 2023, the Brain Trauma Foundation (BTF) found only Level II A (moderate-level) evidence (12) for the use of bifrontal DC in severe TBI with diffuse injury without mass lesions, citing the lack of improved outcomes at 6 months as measured by GOSE. The BTF finds DCs valuable only in cases in which ICP elevation values were >20 mmHg for more than 15 minutes within a 1-hour period that are refractory to first-tier therapies. As described in the DECRA and RESCUEicp trials (17, 18), BTF states that the bifrontal DC was found to reduce ICP and minimize the number of days patients spend in the ICU (12).
As of the update to the 4th edition of the BTF guidelines (12), the researchers were aware of the ongoing RESCUE-ASDH trial but not its findings. It is the opinion of this review that the findings of the RESCUE-ASDH trial do not conflict with the 4th edition of the BTF guidelines for management of severe TBI (12). Furthermore, the RESCUE-ASDH trial does not elevate the use of DC from Level IIA evidence.
The Bifrontal Approach in Neuro-oncology
A unique modification of the bifrontal approach can be found in neuro-oncology, which has long been held to be a suitable option for tumor excision dating back to 1958 (27) and later modified in 1971 (28) and then refined as the transbasal approach in 1995 (29). These approaches include surgical resection of frontal mass lesions, most commonly olfactory groove meningiomas (30–38). In these investigations the drawbacks of the bifrontal approach include high rates of cerebrospinal fluid (CSF) leaks, wound infections, and superior sagittal sinus perforation, themes common to the decompressive applications. While the researchers suggest that the bifrontal approach is symmetrical and applicable for sellar masses, the anterior osteotomy with the transbasal approach may be inferior to fully prevent retraction injury (31). In these analyses, bifrontal approaches have a high rate of total excision and low recurrence, which is attributed to the added benefits of a wide exposure (31–33).
Future Directions
The technique continues to be refined, as evidenced by the one-piece transbasal approach described in
2014 (39) and its continued use in neuro-oncology. It has also been proposed that the bifrontal approach could work in educational applications in neurosurgery (40), though the investigations on this application are limited in scope.
Emerging evidence supports DC’s feasibility and clinical utility in resource-limited settings. A prospective study from Brazil examining 125 patients undergoing DC showed that 51% of survivors achieved favorable outcomes at 6 months, with lower initial GCS scores and older age being the primary predictors of unfavorable outcomes rather than resource limitations (41). Similarly, a 36-month prospective study from Niger examining 74 DC cases found that despite resource constraints, including >24 hour delays between initial TBI and imaging, the impact of limited resources on patient care was moderate, with 66% survival rates and the conclusion that DC remains a viable intervention in low-resource settings (42). It should be noted the authors primarily use hemicraniectomy with bifrontal approaches accounting for only 4% of cases. A recent global survey of 208 neurosurgeons from 60 countries, including 40 low- and middle-income countries (LMIC), revealed high usage of DC techniques in resource-limited settings, especially in the case of ASDH (43) and in 25% of all patients. From this survey, respondents from LMIC indicated that simplified protocols and alternative surgical techniques, including hingecraniectomy, may be more advantageous in resource limited settings. The wide exposure offered by bifrontal DC provides distinct advantages in settings where advanced monitoring capabilities are unavailable, allowing for direct visualization of brain swelling and mass effect without relying solely on ICP-monitoring devices. These practical considerations, combined with demonstrated favorable outcomes in resource-constrained environments and the documented adaptation of decompressive techniques to local capabilities, support the continued use of bifrontal DC techniques, even in in resource-limited settings.
We have identified 2 areas that could benefit from additional commentary:
First, an area lacking much commentary is the management of superior sagittal sinus perforation (44). A recent report describes the technique of compartmentalization as a possible alternative to cranialization or obliteration. Notably, the use of the well-vascularized periosteal flap in 6 cases of frontal
trauma was not associated with any wound infections or CSF leaks. However, this case series suggests that further investigations are needed to identify the most ideal management for superior sinus laceration. It is vital to discuss optimal intraoperative management techniques for superior sagittal sinus complications, which researchers spanning 50+ years have cited as a chief concern with this approach.
Second, there is no consensus on the optimal placement of the posterior temporal burr holes. Kjellberg (1) himself favored an approach that today resembles the transbasal approach, while military surgeons in acute trauma settings have suggested, in some cases, even 2 temporo-sphenoidal burr holes (15). While novel techniques like 3D brain mapping allow for exceptional visualization of vascular structures when surgical planning is possible (45, 46), neurotrauma frequently requires quick imaging modalities to get the patient to the operating room (24). The development of standardized approaches in DC could lead to improved outcomes and minimize complications when a wide surgical approach is considered.
Several research priorities have been identified from our review of the relevant literature. First, standardized protocols for superior sagittal sinus management require prospective validation, as current techniques remain largely anecdotal and compiled from single-surgeon series. Second, comparative effectiveness research is needed to determine optimal patient selection criteria, particularly regarding timing of intervention and identification of patients most likely to benefit from bifrontal versus unilateral approaches. Third, long-term quality of life analyses beyond 12 months are essential, as recent evidence from the Hutchinson research group suggests that the full impact of TBI outcomes is usually not known for at least 2 years following DC (47). Finally, development of intraoperative guidance systems for optimal bone removal boundaries could standardize the procedure and reduce technique-dependent variability in outcomes.
Conclusion
The utility of the bifrontal DC has been demonstrated by RESCUEicp and DECRA trials for the evacuation of frontal hematoma, recent investigations for traumatic bilateral contusions, certain ruptured cerebral aneurysms resulting in multi-compartment
hemorrhage, penetrating head-injury resulting in lobectomy as demonstrated by military reports, and finally as a broad approach for frontal mass lesions in neuro-oncology.
Despite well-documented randomized controlled investigations, our review of the relevant literature demonstrates that the bifrontal approach remains a questionable but versatile approach to frontal trauma. The best example is the RESCUEicp trial, which demonstrated that surgical intervention reduced patient mortality by 45% compared to medical treatment alone. These results are promising and affirm future possibilities in the treatment of cranial pathologies.
Our findings also highlight shifting perspectives on DC over the years, from an emergent rescue operation for patients who otherwise had no chance of survival to a viable alternative to medical management. A common thread across many generations of neurosurgeons is that individual skill and preference for approaches may, in some cases, have clinical preference over demonstrated effectiveness. While there are drawbacks to the bifrontal decompressive craniotomy, the constraints of austere environments continue to force surgeons to keep an open mind to niche surgical approaches.
Disclosures
None.
Acknowledgments
The author would like to acknowledge Dr. Mathangi Rajaram-Gilkes, Dr. Ying-Ju Sung, and Dr. Alejandro Bugarini, who sparked the idea for the article with their passion of clinical anatomy. Without their influence, this paper could not have been written.
References
1. Kjellberg RN, Prieto A. Bifrontal decompressive craniotomy for massive cerebral edema. Journal of Neurosurgery. 1971 Apr;34(4):488–93.
2. Moon JW, Hyun DK. Decompressive Craniectomy in Traumatic Brain Injury: A Review Article. Korean J Neurotrauma. 2017;13(1):1.
3. Rossini Z, Nicolosi F, Kolias AG, Hutchinson PJ, De Sanctis P, Servadei F. The History of Decompressive Craniectomy in Traumatic Brain Injury. Front Neurol. 2019 May 8;10:458.
4. Bullock MR, Chesnut R, Ghajar J, Gordon D, Hartl R, Newell DW, et al. Surgical management of acute epidural hematomas. Neurosurgery. 2006 Mar;58(3 Suppl):S7-15; discussion Si-iv.
5. Manley GT, Dams-O’Connor K, Alosco ML, Awwad HO, Bazarian JJ, Bragge P, et al. A new characterisation of acute traumatic brain injury: the NIH-NINDS TBI Classification and Nomenclature Initiative. The Lancet Neurology 2025 Jun 1;24(6):512–23.
6. Bennett ER, Reuter-Rice K, Laskowitz DT. Genetic Influences in Traumatic Brain Injury. In: Laskowitz D, Grant G, editors. Translational Research in Traumatic Brain Injury [Internet]. Boca Raton (FL): CRC Press/Taylor and Francis Group; 2016 [cited 2025 Jun 20]. (Frontiers in Neuroscience). Available from: http://www.ncbi.nlm.nih.gov/ books/NBK326717/
7. Janigro D, Mondello S, Posti JP, Unden J. GFAP and S100B: What You Always Wanted to Know and Never Dared to Ask. Front Neurol. 2022 Mar 21;13:835597.
8. Paldor I, Peso D, Sviri GE. Decompressive craniectomy in elderly patients with traumatic brain injury. Journal of Clinical Neuroscience. 2020 Aug 1;78:269–72.
10. Vitali M, Marasco S, Romenskaya T, Elia A, Longhitano Y, Zanza C, et al. Decompressive Craniectomy in Severe Traumatic Brain Injury: The Intensivist’s Point of View. Diseases. 2023 Jan 30;11(1):22.
11. Monsour MA, Wolfson DI, Jacob JO, Terry DP, Zuckerman SL. Is contact sport participation associated with chronic traumatic encephalopathy or neurodegenerative decline? A systematic review and meta-analysis. J Neurosurg Sci. 2024;68(1):117–27.
12. Carney N, Totten AM, O’Reilly C, Ullman JS, Hawryluk GWJ, Bell MJ, et al. Guidelines for the Management of Severe Traumatic Brain Injury, Fourth Edition. Neurosurgery. 2017 Jan;80(1):6–15.
DL, Kao LS, Hunter JG, et al., editors. Schwartz’s Principles of Surgery, 11e [Internet]. New York, NY: McGraw-Hill Education; 2019 [cited 2024 Apr 4]. Available from: accesssurgery.mhmedical.com/ content.aspx?aid=1164315755
14. Ragel BT, Klimo P, Martin JE, Teff RJ, Bakken HE, Armonda RA. Wartime decompressive craniectomy: technique and lessons learned. Neurosurg Focus. 2010 May;28(5):E2.
15. Ragel BT, Klimo P, Kowalski RJ, McCafferty RR, Liu JM, Taggard DA, et al. Neurosurgery in Afghanistan during “Operation Enduring Freedom”: a 24-month experience. FOC. 2010 May;28(5):E8.
16. Bell RS, Mossop CM, Dirks MS, Stephens FL, Mulligan L, Ecker R, et al. Early decompressive craniectomy for severe penetrating and closed head injury during wartime. FOC. 2010 May;28(5):E1.
17. Cooper DJ, Rosenfeld JV, Murray L, Arabi YM, Davies AR, D’Urso P, et al. Decompressive Craniectomy in Diffuse Traumatic Brain Injury. N Engl J Med. 2011 Apr 21;364(16):1493–502.
18. Hutchinson PJ, Kolias AG, Timofeev IS, Corteen EA, Czosnyka M, Timothy J, et al. Trial of Decompressive Craniectomy for Traumatic Intracranial Hypertension. N Engl J Med. 2016 Sep 22;375(12):1119–30.
19. Hutchinson PJ, Adams H, Mohan M, Devi BI, Uff C, Hasan S, et al. Decompressive Craniectomy versus Craniotomy for Acute Subdural Hematoma. N Engl J Med. 2023 Jun 15;388(24):2219–29.
20. Polin RS, Shaffrey ME, Bogaev CA, Tisdale N, Germanson T, Bocchicchio B, et al. Decompressive Bifrontal Craniectomy in the Treatment of Severe Refractory Posttraumatic Cerebral Edema. Neurosurgery. 1997 Jul 1;41(1):84–94.
21. Whitfield PC, Patel H, Hutchinson PJA, Czosnyka M, Parry D, Menon D, et al. Bifrontal decompressive craniectomy in the management of posttraumatic intracranial hypertension. British Journal of Neurosurgery. 2001 Jan;15(6):500–7.
22. Sahuquillo J, Dennis JA. Decompressive craniectomy for the treatment of high intracranial pressure in closed traumatic brain injury. Cochrane Injuries Group, editor. Cochrane Database of Systematic Reviews [Internet]. 2019
Dec 31 [cited 2025 Apr 26];2019(12). Available from: http://doi.wiley.com/10.1002/14651858. CD003983.pub3
23. Taylor A, Butt W, Rosenfeld J, Shann F, Ditchfield M, Lewis E, et al. A randomized trial of very early decompressive craniectomy in children with traumatic brain injury and sustained intracranial hypertension. Child’s Nervous System. 2001 Feb 23;17(3):154–62.
24. Bell RS, Vo AH, Neal CJ, Tigno J, Roberts R, Mossop C, et al. Military Traumatic Brain and Spinal Column Injury: A 5-Year Study of the Impact Blast and Other Military Grade Weaponry on the Central Nervous System. Journal of Trauma: Injury, Infection & Critical Care. 2009 Apr;66(4):S104–11.
25. Ratan R, Mohindra S, Tripathi M, Singla R, Chauhan R. Factors determining the requirement of surgical intervention and prognosis in cases of traumatic bifrontal contusions: A prospective observational study. Surgical Neurology International. 2023 Dec 22;14:438.
26. Peterson EC, Chesnut RM. Talk and die revisited: Bifrontal contusions and late deterioration. J Trauma Inj Infect Crit Care. 2011;71(6):1588–92.
27. Unterberger S. Zur versorgung fronto-basaler verletzungen. Arch Ohren Nasen Kehlkopfheilkd 1958;(172):463–84.
28. Tessier P, Guiot G, Derome P. Orbital hypertelorism. II. Definite treatment of orbital hypertelorism (OR.H.) by craniofacial or by extracranial osteotomies. Scandinavian journal of plastic and reconstructive surgery [Internet]. 1973 [cited 2025 Jun 20];7(1). Available from: https:// pubmed.ncbi.nlm.nih.gov/4752569/
29. Derome P. Transbasal approach to tumors invading the skull base. In: Schmidek HH, Sweet WH, eds Operative neurosurgical techniques 3rd ed Philadelphia: WB Saunders, 1995. 1995. p. 427-441.
30. Kumon Y, Zenke K, Ohta S, Hatakeyama T, Sakaki S, Yanagihara N, et al. Operative results in fourteen cases of paranasal sinus and anterior cranial fossa lesions surgically treated by an extended transbasal approach. No Shinkei Geka Neurological Surgery. 1995 Jan 1;23(10):889–95.
31. Spektor S, Valarezo J, Fliss DM, Gil Z, Cohen J, Goldman J, et al. Olfactory Groove Meningiomas
from Neurosurgical and Ear, Nose, and Throat Perspectives: Approaches, Techniques, and Outcomes. Operative Neurosurgery. 2005 Oct 1;57(suppl_4):ONS-268-ONS-280.
32. Feiz-Erfan I, Han PP, Spetzler RF, Horn EM, Klopfenstein JD, Porter RW, et al. The radical transbasal approach for resection of anterior and midline skull base lesions. Journal of Neurosurgery. 2005 Sep;103(3):485–90.
33. Nakamura M, Roser F, Struck M, Vorkapic P, Samii M. Tuberculum sellae meningiomas: clinical outcome considering different surgical approaches. Neurosurgery. 2006 Nov;59(5):1019–29.
34. Goel A, Bhaganagare A, Shah A, Kaswa A, Rai S, Dharurkar P, et al. Olfactory groove meningiomas: An analysis based on surgical experience with 129 cases. Neurol India. 2018;66(4):1081.
35. Nanda A, Maiti TK, Bir SC, Konar SK, Guthikonda B. Olfactory Groove Meningiomas: Comparison of Extent of Frontal Lobe Changes After Lateral and Bifrontal Approaches. World Neurosurgery. 2016 Oct;94:211–21.
36. Ottenhausen M, Rumalla K, Alalade AF, Nair P, La Corte E, Younus I, et al. Decision-making algorithm for minimally invasive approaches to anterior skull base meningiomas. Neurosurgical Focus. 2018 Apr;44(4):E7.
37. Seaman SC, Ali MS, Marincovich A, Osorno-Cruz C, Greenlee JDW. Single surgeon experience with minimally invasive supraorbital craniotomy versus bifrontal craniotomy for anterior skull base meningiomas. Surgical Neurology International. 2020 Dec 22;11:458.
38. Kofoed Lauridsen E, Ciochon UM, Tolver A, Bech Knudsen M, Giraldi L, Springborg JB, et al. Longterm postoperative health-related quality of life in patients with subfrontal meningiomas. Journal of Neurosurgery. 2023 Jun 1;138(6):1542–51.
40. Burwell JM, Bugarini A, Rajaram-Gilkes M. Bifrontal Craniectomy: A High-Yield Surgical Training Tool. Cureus [Internet]. 2024 Dec 11
[cited 2025 Apr 26]; Available from: https:// www.cureus.com/articles/325536-bifrontalcraniectomy-a-high-yield-surgical-training-tool
41. Rynkowski CB, Basso LS, Kolias AG, Bianchin MM. Long-Term Outcome After Decompressive Craniectomy in a Developing Country. In: Depreitere B, Meyfroidt G, Güiza F, editors. Intracranial Pressure and Neuromonitoring XVII [Internet]. Cham: Springer International Publishing; 2021 [cited 2025 Jun 20]. p. 87–90. (Acta Neurochirurgica Supplement; vol. 131). Available from: https://link.springer. com/10.1007/978-3-030-59436-7_19
42. Hamma OI, Kelani AB, Ango SM, Hissene TM, Mobarak SM, Tounga YB, et al. Decompressive craniectomy at the National Hospital of Niamey: Prospective study of the epidemioclinical profile, indications, surgical techniques, and results in a context of limited resources. Surgical Neurology International. 2024 Jul 5;15:235.
43. Mohan M, Layard Horsfall H, Solla DJF, Robertson FC, Adeleye AO, Teklemariam TL, et al. Decompressive craniotomy: an international survey of practice. Acta Neurochir. 2021 May 1;163(5):1415–22.
44. Miller C, Weisbrod L, Beahm D, Chamoun R. Compartmentalization: An Open Technique for Frontal Sinus Repair and Preservation— Description of Technique and Early Case Series. J Neurol Surg B Skull Base. 2021 Apr;82(02):189–95.
45. Vezirska D, Milev M, Laleva L, Nakov V, Spiriev T. Three-Dimensional Printing in Neurosurgery: A Review of Current Indications and Applications and a Basic Methodology for Creating a ThreeDimensional Printed Model for the Neurosurgical Practice. Cureus [Internet]. 2022 Dec 30 [cited 2025 Apr 26]; Available from: https://www.cureus. com/articles/127901-three-dimensional-printingin-neurosurgery-a-review-of-current-indicationsand-applications-and-a-basic-methodology-forcreating-a-three-dimensional-printed-model-forthe-neurosurgical-practice
46. Gurses ME, Hanalioglu S, Mignucci-Jiménez G, Gökalp E, Gonzalez-Romo NI, Gungor A, et al. Three-Dimensional Modeling and Extended Reality Simulations of the Cross-Sectional Anatomy of the Cerebrum, Cerebellum, and Brainstem. Operative Neurosurgery [Internet].
47. Kolias AG, Adams H, Timofeev IS, Corteen EA, Hossain I, Czosnyka M, et al. Evaluation of Outcomes Among Patients With Traumatic Intracranial Hypertension Treated With Decompressive Craniectomy vs Standard Medical Care at 24 Months: A Secondary Analysis of the RESCUEicp Randomized Clinical Trial. JAMA Neurol. 2022 Jul 1;79(7):664–71.
Molecular Insights into Glioblastoma Resistance and Implications for Treatment Advancements: A Systematic Review
Kaustov Chakrabarti1*‡, Amanda M. Mohabir1*‡, Mahesh Nukala1*‡, Sara M. Safiullah1*‡, and Tierney R. Woitas1*‡
¹Geisinger College of Health Sciences, Scranton, PA 18509
Glioblastoma (GBM) is a highly resistant brain tumor with a dismal prognosis despite standard treatments like temozolomide (TMZ) and radiation therapy (RT). This review delineates the molecular mechanisms driving this resistance to inform future therapeutic strategies. A narrative review was conducted using PubMed, synthesizing studies from the year 2000 up to March 2025 on DNA repair enzymes, TMZ and RT resistance pathways, anti-apoptosis mechanisms, genetic heterogeneity, and cell cycle dysregulation in GBM. Five key resistance mechanisms were identified: (1) DNA repair enzymes (e.g., MGMT, APNG) counteract TMZ-induced damage, regulated by NF-κB and CHD4; (2) TMZ and RT resistance depends on MGMT methylation and MMR functionality; (3) anti-apoptotic pathways (e.g., Bcl-2, PI3K/Akt) protect GBM cells, mitigated by TRAIL sensitization; (4) genetic mutations (EGFR, PTEN, TP53) confer adaptability; and (5) quiescent and proliferative GBM stem cells (GSCs) sustain recurrence. Novel therapies targeting HK2 and neural stem cells show promise. GBM resistance is multifaceted, requiring personalized, multi-targeted approaches. Future research should prioritize epigenetic modulation and combination therapies to enhance patient outcomes.
Introduction
Glioblastoma (GBM) is the most prevalent and lethal primary brain tumor in adults, accounting for approximately 50% of all malignant brain tumors (1). Arising from astrocytes, GBM exhibits aggressive growth and infiltration into surrounding brain tissue, rendering complete surgical resection nearly impossible (1). The current standard of care involves
maximal safe surgical resection followed by 6 weeks of concurrent radiotherapy (RT) and temozolomide (TMZ), an oral alkylating chemotherapeutic agent, followed by 6 monthly cycles of adjuvant TMZ (1). RT employs high-energy beams to induce DNA singleand double-strand breaks, aiming to halt tumor cell proliferation, while TMZ methylates DNA bases — most critically at the O6 position of guanine — to trigger cytotoxicity and apoptosis (2, 3). Despite this multimodal approach, GBM remains universally fatal, with a median survival of 12–15 months and a 5-year survival rate below 5% (1).
This poor prognosis is primarily due to GBM’s remarkable resistance to both TMZ and RT, driven by a complex array of molecular mechanisms. These include upregulated DNA repair pathways, evasion of programmed cell death, extensive genetic and cellular heterogeneity, and dysregulated cell cycle dynamics. Previous studies have explored individual aspects of resistance, such as DNA repair enzymes or tumor heterogeneity, but a comprehensive synthesis integrating all major mechanisms with emerging therapeutic insights is lacking (4, 5). The cIMPACT-NOW initiative has emphasized molecular classification to refine GBM diagnosis and treatment, highlighting the need to understand resistance at a molecular level (2).
This review aims to elucidate 5 critical mechanisms of GBM resistance: 1. DNA repair enzymes, 2. TMZ and RT resistance pathways, 3. anti-apoptosis mechanisms, 4. genetic heterogeneity, and 5. cell cycle effects. By incorporating all 60 references from recent literature, we provide a detailed analysis of these mechanisms, their interplay, and potential therapeutic targets, such as HK2 inhibition (6) and neural stem cell therapies
(7). Our goal is to bridge current knowledge with actionable strategies to overcome GBM’s therapeutic barriers and improve patient survival.
Methods
This narrative review was conducted to consolidate current understanding of GBM resistance mechanisms. Literature was sourced from PubMed, with searches performed up to March 12, 2025, using terms such as "glioblastoma," "temozolomide resistance," "radiation resistance," "DNA repair," "apoptosis," "genetic heterogeneity," and "cell cycle." The review included an equal mixture of preclinical studies (e.g., cell lines, xenografts) and clinical research relevant to the 5 predefined resistance categories. No formal exclusion criteria beyond relevance were applied, though priority was given to studies offering mechanistic insights, experimental evidence, or clinical implications.
Data extraction focused on molecular pathways, experimental models, and clinical outcomes (e.g., survival metrics, resistance biomarkers). All 60 references provided were incorporated, with citations numbered sequentially based on their first appearance in the text. Findings were synthesized narratively under each resistance mechanism, emphasizing integration of molecular details with therapeutic potential.
Discussion
DNA Repair Enzymes
GBM’s resistance to TMZ and RT is significantly mediated by an enhanced DNA damage response (DDR), a network of enzymes that repair treatmentinduced DNA lesions (8). Key players include O6methylguanine-DNA methyltransferase (MGMT) and alkylpurine-DNA N-glycosylase (APNG), which counteract TMZ’s alkylating effects (9). MGMT directly removes O6-methylguanine lesions by transferring the methyl group to itself in a stoichiometric reaction, neutralizing TMZ’s cytotoxicity (10). APNG initiates base excision repair, addressing N3-methyladenine and N7-methylguanine damage, as demonstrated in photodynamic therapy studies showing increased resistance with elevated repair capacity (9).
MGMT expression is tightly regulated by transcription factors like nuclear factor-kappa B (NF-κB) (11) and chromatin remodeling via the NuRD complex,
specifically its subunit CHD4 (12). NF-κB upregulation in glioma-initiating cells (GICs) enhances MGMT levels, while CHD4 modulates chromatin accessibility, facilitating repair machinery access to damaged DNA (12). Promoter methylation of MGMT, observed in 40–50% of GBM patients, silences its expression, improving TMZ sensitivity and extending survival in promotor methylation cases from 14 to 21 months (13). Conversely, unmethylated MGMT correlates with poor response, a finding validated across glioblastoma and anaplastic glioma cohorts (13).
Preclinical studies have identified additional DDR targets, such as Rad51 and BRCA2, involved in homologous recombination (14). Inhibiting these proteins sensitizes GBM cells to alkylating agents by impairing double-strand break repair, as shown in glioma cell lines (14). Novel therapeutic approaches exploit these vulnerabilities: anticancer neural stem cells (NSCs) engineered to deliver sTRAIL with lanatoside C (which bind to and activate death receptors) enhance apoptosis in GBM stem cells (7), while antifungal drugs like ketoconazole and posaconazole target hexokinase 2 (HK2), disrupting energy metabolism and DNA repair (6). Clinical trials, such as those testing peposertib (an ATM inhibitor) with RT and TMZ, aim to overcome resistance in MGMT-unmethylated GBM by enhancing DNA damage (15). These strategies highlight the DDR’s role as a therapeutic target, with broader implications for overcoming resistance (16).
Temozolomide and Radiation Mechanisms
TMZ and RT resistance in GBM arises from both intrinsic cellular properties and adaptive responses. TMZ, a lipophilic alkylating agent, crosses the bloodbrain barrier (BBB) to methylate DNA, but its efficacy depends on MGMT status and mismatch repair (MMR) functionality (3). Hypermethylated MGMT (MGMT-M) tumors exhibit reduced MGMT expression, improving TMZ response and survival (21 months), whereas hypomethylated (MGMT-UM) tumors resist due to robust repair (14 months) (13). MMR deficiencies, such as MSH2/MSH6 mutations, fail to process O6-methylguanine abnormal base pairing, leading to persistent DNA damage and tumor recurrence (17).
RT induces DNA strand breaks, but GBM cells resist via checkpoint kinases (CHK1/CHK2), which arrest the cell cycle in GICs to allow repair, reducing RTinduced cell death (17). Systematic reviews indicate
that extending adjuvant TMZ beyond 6 cycles reduces progression (hazard ratio 0.72), though data on myelotoxicity remain limited (18). Immunotherapy, including checkpoint inhibitors, seeks to enhance RT’s immune activation, but clinical efficacy is modest due to GBM’s immunosuppressive microenvironment (19). Chronotherapy — administering TMZ in the morning — aligns with circadian rhythms, potentially improving outcomes by optimizing DNA repair inhibition (20).
Combination therapies show promise: integrating apoptosis inducers (TMZ, methotrexate, cytarabine) with inhibitors of XPO1, Bcl-2, and Mcl-1 (eltanexor, venetoclax, A1210477) enhances cytotoxicity in GBM models (21). Metabolic targeting with HK2 inhibitors sensitizes cells to RT by disrupting glycolysis (6), while DDR inhibition mitigates TMZ resistance by overwhelming repair capacity (22). These approaches underscore the need for multimodal strategies to address GBM’s adaptive resistance to TMZ and RT (17).
Anti-Apoptosis Mechanisms
GBM evades programmed cell death through intricate anti-apoptotic pathways, a hallmark of its aggressive behavior (23). Upregulation of Bcl-2 and Mcl-1 inhibits mitochondrial apoptosis, with high expression in patient tumors linked to reduced survival (24). Combining TMZ with Bcl-2 inhibitors reduces tumor volume by 40% in vivo, extending survival from 20 to 35 days (21). The PI3K/Akt/mTOR pathway, activated in 88% of GBM cases, promotes survival and proliferation; its inhibition reduces cell growth by 50% in vitro (25).
GBM stem cells resist Fas-induced apoptosis due to low Fas receptor expression (26), but TMZ upregulates death receptor 5 (DR5), enhancing TRAIL sensitivity and extending survival by 30% in rodent models (27). Interleukin-24 further amplifies apoptosis via P38 MAPK and TRAIL, with additional autophagy induction via LC3-II activation (28). Autophagy supports tumor survival under stress by recycling cellular components, but its inhibition increases chemotherapeutic efficacy by 25% (29). The interplay of senescence, apoptosis, and autophagy complicates GBM’s response to therapy, with senescence potentially driving recurrence (30).
PTEN-deficient GBM exhibits proteasome dependency, and proteasome inhibitors selectively kill these cells, achieving 40% tumor regression
without harming normal astrocytes (31). Targeting apoptosis pathways with TRAIL agonists or immune checkpoint inhibitors enhances cell death and immunosurveillance, though clinical response rates remain modest (20–35%) (32). Survival signaling via PI3K/Akt/mTOR offers additional therapeutic opportunities, with inhibitors showing preclinical promise (33). These findings highlight apoptosis evasion as a critical resistance mechanism amenable to targeted intervention (34).
Genetic Heterogeneity of GBM Cells
GBM’s genetic heterogeneity drives its adaptability and resistance, with mutations in epidermal growth factor receptor (EGFR), phosphatase and tensin homolog (PTEN), and tumor protein p53 (TP53) being prevalent (35). The ΔEGFR variant, with an exon 2–7 deletion, is constitutively active, promoting proliferation via an autocrine loop with wild-type EGFR ligands (e.g., TGFα) (36). EGFR signaling also contributes to radioresistance by enhancing DNA repair and survival pathways (37). Targeted therapies against EGFR, such as tyrosine kinase inhibitors, face challenges due to intratumoral heterogeneity (38).
PTEN loss, observed in 40% of GBMs, upregulates PI3K/Akt, increasing invasiveness and reducing DNA repair sensitivity (39). The PI3K/Akt pathway’s role in cancer is well-documented, with inhibitors showing potential to restore treatment sensitivity (40). NEDD4-1-mediated PTEN attenuation exacerbates TMZ resistance by disrupting redox balance via the Nrf2–HO-1 axis (41). Tau protein, under PTEN deficiency, enhances migration via PI3K/ Akt, contributing to GBM’s 3D organization and progression (42).
TP53 mutations include loss-of-function (LOF), impairing apoptosis and DNA repair (43), and gainof-function (GOF) variants like TP53R248L, which increase inflammation via NF-κB, suppressing antitumor immunity (44). Inflammation in GBM further complicates therapy, with novel anti-inflammatory approaches under exploration (45). Mut-p53 accumulation, due to defective Mdm2 regulation (46), forms amyloid oligomers, enhancing chemoresistance (47). Proteomic studies using LC-MS reveal mut-p53’s role in TMZ resistance, suggesting new biomarkers (48). These genetic alterations underscore GBM’s adaptability, necessitating personalized therapeutic strategies (35).
Cell Cycle Effects
GBM disrupts normal cell cycle regulation, with GSCs exhibiting quiescent (qGSCs) and proliferative (pGSCs) states that drive tumor initiation and recurrence (49). qGSCs, enriched in tumor cores, enter a dormant G0 phase via p27 accumulation and CDKN1A/cyclin B1 inhibition, enabling survival post-treatment (50). This quiescence is a dynamic state, balancing restraint and readiness to re-enter the cell cycle (51). Cyclin B1 and CDK1 interactions, critical for G2/M progression, are disrupted in qGSCs, as shown by fluorescence spectroscopy (52). MYC activation further stimulates proliferation by phosphorylating p27 and activating CDK1 (53).
pGSCs, at tumor peripheries, drive invasion with high Ki67 and CDC2 expression (54). Single-cell analyses reveal pGSCs’ dual proliferative and invasive capabilities, contributing to tumor progression (54). Extensive brainstem infiltration in end-stage GBM reflects pGSC activity, not merely mass effect (55). Ki67 positivity indicates active proliferation, correlating with aggressiveness and poor prognosis (56). Therapies targeting these dynamics include azoles like geldanamycin, which induce G2 arrest by downregulating CDC2 and cyclin B1 (57), and CDK inhibitors like AT7519, which trigger apoptosis, pyroptosis, and cell cycle arrest in GBM cells (58). These approaches aim to disrupt GSC plasticity, addressing GBM’s recurrence challenge (49).
Conclusion
GBM’s resistance to TMZ and RT emerges from a sophisticated network of molecular defenses, each offering both obstacles and opportunities for intervention. DNA repair enzymes like MGMT and APNG counteract TMZ’s alkylating effects, with MGMT methylation status serving as a key prognostic biomarker (13). NF-κB and CHD4 regulate repair pathways, presenting novel targets, though their inhibition risks off-target effects in healthy cells (11, 12). Strategies to mitigate TMZ resistance, such as DDR inhibition, enhance DNA damage but require careful optimization to avoid genomic instability (22). HK2 inhibition with antifungals disrupts energy metabolism, sensitizing GBM to RT and offering a repurposing opportunity (6).
Anti-apoptotic mechanisms, driven by Bcl-2 and PI3K/ Akt/mTOR, protect GBM cells from treatment-induced
death (24, 25). GSC resistance to Fas highlights the need for TRAIL-based therapies, which synergize with TMZ to enhance apoptosis (27). Autophagy’s dual role — supporting survival under stress yet potentiating cell death when inhibited — complicates therapeutic design (29). Genetic heterogeneity via EGFR, PTEN, and TP53 mutations fuels adaptability, with mut-p53’s inflammatory effects challenging immunotherapy (44). EGFR’s role in radioresistance and PTEN’s impact on PI3K/Akt signaling underscore the need for targeted inhibitors, though tumor heterogeneity limits efficacy (37, 39).
Cell cycle dysregulation in qGSCs and pGSCs drives recurrence, with quiescence enabling tumor regrowth post-therapy (50). CDK inhibitors and azoles target proliferative states, but addressing quiescent cells remains a hurdle (58). The cIMPACT-NOW framework emphasizes molecular profiling to guide therapy, integrating EGFR amplification and TP53 status into treatment plans (2). Emerging therapies, such as NSC-delivered sTRAIL and peposertib with RT, exploit spatial and molecular vulnerabilities, though BBB penetration and off-target effects pose challenges (7, 15).
Future directions should prioritize personalized medicine, leveraging proteomic insights (e.g., LC-MS for TMZ resistance) and circadian-timed dosing to optimize efficacy (20, 48). Combination therapies — integrating DDR inhibitors, apoptosis inducers, and immunotherapy — offer a multi-pronged approach to dismantle GBM’s defenses (21). While risks like secondary cancers and neurotoxicity persist, these strategies could shift GBM from an intractable disease to a manageable condition, significantly extending survival. The interplay of senescence, apoptosis, and autophagy further warrants exploration to prevent recurrence (30).
Disclosures
The authors declare no competing interests. The authors received no financial support for the review.
Acknowledgments
This paper received contributions from Brian J. Piper, PhD.
References
1. Lan Z, Li X, Zhang X. Glioblastoma: An update in pathology, molecular mechanisms and biomarkers. Int J Mol Sci. 2024; 25(5):3040.
2. Louis DN, Wesseling P, Aldape K, Brat DJ, Capper D, Cree IA, et al. cIMPACT-NOW update 6: New entity and diagnostic principle recommendations of the cIMPACT-Utrecht meeting on future CNS tumor classification and grading. Brain Pathol 2020; 30(4):844-856.
3. Erasmus H, Gobin M, Niclou S, Van Dyck E. DNA repair mechanisms and their clinical impact in glioblastoma. Mutat Res Rev Mutat Res. 2016; 769:19-35.
4. Rominiyi O, Collis SJ. DDRugging glioblastoma: Understanding and targeting the DNA damage response to improve future therapies. Mol Oncol. 2022; 16(1):11-41.
5. Ghahe SS, Kosicki K, Wojewódzka M, Majchrzak BA, Fogtman A, Iwanicka-Nowicka R, et al. Increased DNA repair capacity augments the resistance of glioblastoma cells to photodynamic therapy. DNA Repair (Amst). 2021; 104:103123.
6. Sorribes IC, Handelman SK, Jain HV. Mitigating temozolomide resistance in glioblastoma via DNA damage-repair inhibition. J R Soc Interface. 2020; (162); 17,162.
7. O'Shaughnessy A, Hendrich B. CHD4 in the DNA-damage response and cell cycle progression: Not so NuRDy now. Biochem Soc Trans. 2013; 41(3):777-82.
8. Happold C, Stojcheva N, Silginer M, Weiss T, Roth P, Reifenberger G, et al. Transcriptional control of O6-methylguanine DNA methyltransferase expression and temozolomide resistance in glioblastoma. J Neurochem. 2018; 144(6):780-790.
9. Bobola MS, Alnoor M, Chen JY, Kolstoe DD, Silbergeld DL, Rostomily RC, et al. O6methylguanine-DNA methyltransferase activity is associated with response to alkylating agent therapy and with MGMT promoter methylation in glioblastoma and anaplastic glioma. BBA Clin. 2015; 3:1-10.
10. Quiros S, Roos WP, Kaina B. Rad51 and BRCA2-New molecular targets for sensitizing glioma cells to alkylating anticancer drugs. PLoS One. 2011; 6(11):e27183.
11. Teng J, Hejazi S, Badr CE, Tannous BA. Systemic anticancer neural stem cells in combination with a cardiac glycoside for glioblastoma therapy. Stem Cells. 2014; 32(8):2021-32.
12. Agnihotri S, Mansouri S, Burrell K, Li M, Mamatjan Y, Liu J, et al. Ketoconazole and posaconazole selectively target HK2-expressing glioblastoma cells. Clin Cancer Res. 2019; 25(2):844-855.
13. Majd N. Peposertib and Radiation Therapy, Followed by Temozolomide for the Treatment of Patients With Newly Diagnosed MGMT Unmethylated Glioblastoma or Gliosarcoma [Internet]. Houston, Texas: ClinicalTrials.gov; 2020 [updated 2024; cited 2024 Nov 21]. Available from: https://clinicaltrials.gov/study/ NCT04555577#more-information
14. Ferri A, Stagni V, Barilà D. Targeting the DNA damage response to overcome cancer drug resistance in glioblastoma. Int J Mol Sci. 2020; 21(14):4910.
16. Gupta T, Selvarajan JMP, Kannan S, Menon N, Dasgupta A, Chatterjee A, et al. Updated systematic review and meta-analysis of extended adjuvant temozolomide in patients with newly diagnosed glioblastoma. Neuro-Oncol Adv. 2023; 5(1):vdad086.
17. Maccari M, Baek C, Caccese M, Mandruzzato S, Fiorentino A, Internò V, et al. Present and future of immunotherapy in patients with glioblastoma: Limitations and opportunities. Oncologist. 2024; 29(4):289–302.
18. Yalamarty SSK, Filipczak N, Li X, Subhan MA, Parveen F, Ataide JA, et al. Mechanisms of resistance and current treatment options for glioblastoma multiforme (GBM). Cancers (Basel) 2023; 15(7):2116.
19. Jia JL, Alshamsan B, Ng TL. Temozolomide chronotherapy in glioma: A systematic review. Curr Oncol. 2023; 30(2):1893–902.
20. Zhao K, Braun M, Meyer L, Otte K, Raifer H, Helmprobst F, et al. A novel approach for glioblastoma treatment by combining apoptosis inducers (TMZ, MTX, and Cytarabine) with E.V.A. (eltanexor, Venetoclax, and A1210477) inhibiting XPO1, Bcl-2, and Mcl-1. Cells. 2024; 13(7):632.
21. Pawlowska E, Szczepanska J, Szatkowska M, Blasiak J. An interplay between senescence, apoptosis and autophagy in glioblastoma multiforme—role in pathogenesis and therapeutic perspective. Int J Mol Sci. 2018;19(3):889.
22. Valdés-Rives SA, Casique-Aguirre D, GermánCastelán L, Velasco-Velázquez MA, GonzálezArenas A. Apoptotic signaling pathways in glioblastoma and therapeutic implications. Biomed Res Int. 2017; 2017:7403747.
23. Bertrand J, Beguad-Grimaud G, Bessette B, Verdier M, Battu S, Jauberteau M. Cancer stem cells from human glioma cell line are resistant to Fas-induced apoptosis. Int J Oncol. 2009; 34(3):717-27.
24. Fulda S. Cell death-based treatment of glioblastoma. Cell Death Dis. 2018; 9(2):121.
25. Khan I, Baig MH, Mahfooz S, Rahim M, Karacam B, Elbasan EB, et al. Deciphering the role of autophagy in treatment of resistance mechanisms in glioblastoma. Int J Mol Sci. 2021; 22(3):1318.
26. Nagane M, Pan G, Weddle JJ, Dixit VM, Cavenee WK, Huang HJ. Increased death receptor 5 expression by chemotherapeutic agents in human gliomas causes synergistic cytotoxicity with tumor necrosis factor-related apoptosis-inducing ligand in vitro and in vivo. Cancer Res. 2000; 60(4):847–53.
27. Babazadeh SM, Zolfaghari MR, Zargar M, Baesi K, Hosseini SY, Ghaemi A. Interleukin-24-mediated antitumor effects against human glioblastoma via upregulation of P38 MAPK and endogenous TRAIL-induced apoptosis and LC3-II activationdependent autophagy. BMC Cancer. 2023; 23(1):519.
28. Gousias K, Theocharous T, Simon M. Mechanisms of cell cycle arrest and apoptosis in glioblastoma. Biomedicines. 2022;10(3):564.
29. Ou A, Yung WKA, Majd N. Molecular mechanisms of treatment resistance in glioblastoma. Int J Mol Sci. 2020; 22(1):351.
30. Li X, Wu C, Chen N, Gu H, Yen A, Cao L, et al. PI3K/Akt/mTOR signaling pathway and targeted therapy for glioblastoma. Oncotarget. 2016; 7(22):33440–50.
31. Benitez JA, Finlay D, Castanza A, Parisian AD, Ma J, Longobardi C, et al. PTEN deficiency leads to proteasome addiction: A novel vulnerability in glioblastoma. Neuro Oncol. 2021; 23(7):1072–1086.
32. Eisele G, Weller M. Targeting apoptosis pathways in glioblastoma. Cancer Lett. 2013; 332(2):335–45.
33. Pimentel JM, Zhou JY, Wu GS. The role of TRAIL in apoptosis and immunosurveillance in cancer. Cancers (Basel). 2023; 15(10):2752.
34. Krakstad C, Chekenya M. Survival signaling and apoptosis resistance in glioblastomas: Opportunities for targeted therapeutics. Mol Cancer. 2010; 9:135.
35. Huang YF, Chiao MT, Hsiao TH, Zhan YX, Chen TY, Lee CH, et al. Genetic mutation patterns among glioblastoma patients in the Taiwanese population – insights from a single institution retrospective study. Cancer Gene Ther. 2024; 31(6):894–903.
36. Rayego-Mateos S, Rodrigues-Diez R, MorgadoPascual JL, Valentijn F, Valdivielso JM, Goldschmeding R, et al. Role of epidermal growth factor receptor (EGFR) and its ligands in kidney inflammation and damage. Mediators Inflamm 2018; 2018:8739473.
37. Hatanpaa KJ, Burma S, Zhao D, Habib AA. Epidermal growth factor receptor in glioma: Signal transduction, neuropathology, imaging, and radioresistance. Neoplasia. 2010; 12(9):675–84.
38. Rodriguez SMB, Kamel A, Ciubotaru GV, Onose G, Sevastre AS, Sfredel V, et al. An overview of EGFR mechanisms and their implications in targeted therapies for glioblastoma. Int J Mol Sci. 2023; 24(13):11110.
39. Hashemi M, Etemad S, Rezaei S, Ziaolhagh S, Rajabi R, Rahmanian P, et al. Progress in targeting PTEN/PI3K/Akt axis in glioblastoma therapy: Revisiting molecular interactions. Biomed Pharmacother. 2023; 158:114204.
40. Khezri MR, Jafari R, Yousefi K, Zolbanin NM. The PI3K/AKT signaling pathway in cancer: Molecular mechanisms and possible therapeutic interventions. Exp Mol Pathol. 2022; 127:104787.
41. Chuang HY, Hsu LY, Pan CM, Pikatan NW, Yadav VK, Fong IH, et al. The E3 ubiquitin ligase NEDD41 mediates temozolomide-resistant glioblastoma through PTEN attenuation and redox imbalance in Nrf2–HO-1 axis. Int J Mol Sci. 2021; 22(19):10247.
42. Pagano A, Breuzard G, Parat F, Tchoghandjian A, Figarella-Branger D, De Bessa TC, et al. Tau regulates glioblastoma progression, 3D cell organization, growth and migration via the PI3KAKT Axis. Cancers (Basel). 2021; 13(22):5818.
43. Aubrey BJ, Strasser A, Kelly GL. Tumor-suppressor functions of the tp53 pathway. Cold Spring Harb Perspect Med. 2016; 6(5):a026062.
44. Zhang Y, Dube C, Gibert M, Cruickshanks N, Wang B, Coughlan M, et al. The p53 pathway in glioblastoma. Cancers (Basel). 2018; 10(9):297.
45. Ham SW, Jeon HY, Jin X, Kim EJ, Kim JK, Shin YJ, et al. TP53 gain-of-function mutation promotes inflammation in glioblastoma. Cell Death Differ. 2019; 26(3):409–25.
46. Alorfi NM, Ashour AM, Alharbi AS, Alshehri FS. Targeting inflammation in glioblastoma: An updated review from pathophysiology to novel therapeutic approaches. Medicine (Baltimore) 2024; 103(21):e38245.
47. Honda R, Tanaka H, Yasuda H. Oncoprotein MDM2 is a ubiquitin ligase E3 for tumor suppressor p53. FEBS Lett. 1997; 420(1):25–7.
48. Pedrote MM, Motta MF, Ferretti GDS, Norberto DR, Spohr TCLS, Lima FRS, et al. Oncogenic gain of function in glioblastoma is linked to mutant p53 amyloid oligomers. iScience. 2020; 23(2):100820.
49. Teraiya M, Perreault H, Chen VC. An overview of glioblastoma multiforme and temozolomide resistance: Can LC-MS-based proteomics reveal the fundamental mechanism of temozolomide resistance? Front Oncol. 2023; 13:1166207.
50. Liu J, Peng Y, Wei W. Cell cycle on the crossroad of tumorigenesis and cancer therapy. Trends Cell Biol 2022; 32(1):30–44.
51. Wang Z, Zhang H, Xu S, Liu Z, Cheng Q. The adaptive transition of glioblastoma stem cells and its implications on treatments. Signal Transduct Target Ther. 2021; 6(1):124.
52. Xie XP, Laks DR, Sun D, Ganbold M, Wang Z, Pedraza AM, et al. Quiescent human glioblastoma cancer stem cells drive tumor initiation, expansion, and recurrence following chemotherapy. Dev Cell. 2022; 57(1):32-46.
53. Van Velthoven CTJ, Rando TA. Stem cell quiescence: Dynamism, restraint, and cellular idling. Cell Stem Cell. 2019; 24(2):213–25.
54. Barbiero M, Cirillo L, Veerapathiran S, Coates C, Ruffilli C, Pines J. Cell cycle-dependent binding between cyclin B1 and Cdk1 revealed by timeresolved fluorescence correlation spectroscopy. Open Biol. 2022; 12(6):220057.
55. García-Gutiérrez L, Bretones G, Molina E, Arechaga I, Symonds C, Acosta JC, et al. MYC stimulates cell cycle progression through the activation of Cdk1 and phosphorylation of p27. Sci Rep. 2019; 9(1):18693.
56. Ratliff M, Karimian-Jazi K, Hoffmann DC, Rauschenbach L, Simon M, Hai L, et al. Individual glioblastoma cells harbor both proliferative and invasive capabilities during tumor progression. Neuro Oncol. 2023; 25(12):2150–62.
57. Drumm MR, Dixit KS, Grimm S, Kumthekar P, Lukas RV, Raizer JJ, et al. Extensive brainstem infiltration, not mass effect, is a common feature of end-stage cerebral glioblastomas. Neuro Oncol. 2019; 22(4):470–9.
58. Miller I, Min M, Yang C, Tian C, Gookin S, Carter D, et al. KI67 is a graded rather than a binary marker of proliferation versus quiescence. Cell Rep. 2018; 24(5):1105-1112.
59. Nomura N, Nomura M, Newcomb EW, Zagzag D. Geldanamycin induces G2 arrest in U87MG glioblastoma cells through downregulation of CDC2 and cyclin B1. Biochem Pharmacol. 2007; 73(10):1528–36.
60. Zhao W, Zhang L, Zhang Y, Jiang Z, Lu H, Xie Y, et al. The CDK inhibitor AT7519 inhibits human glioblastoma cell growth by inducing apoptosis, pyroptosis and cell cycle arrest. Cell Death Dis 2023; 14(1):11.
Moyamoya Disease: Physiological Mechanisms and Treatment Approaches
Kaustov Chakrabarti1*
¹Geisinger College of Health Sciences, Scranton, PA 18509
*Master of Biomedical Sciences Program
Correspondence: kchakrabarti@geisinger.edu
Abstract
Moyamoya disease (MMD) is a rare, progressive cerebrovascular disorder marked by stenosis of the terminal internal carotid arteries and the formation of fragile collateral vessels, leading to reduced cerebral perfusion and heightened risks of ischemic and hemorrhagic strokes. Its pathophysiology involves endothelial dysfunction, aberrant angiogenesis, and genetic mutations, notably in the RNF213 gene. This paper examines normal cerebrovascular physiology, the pathological changes driving MMD, its clinical features, and current therapeutic approaches. Treatments, including medical management and surgical revascularization (e.g., direct and indirect bypass), aim to restore cerebral blood flow and mitigate stroke risk, with revascularization proving most effective. Elucidating the physiological underpinnings of MMD is vital for developing targeted therapies and enhancing patient outcomes.
Introduction
Moyamoya disease (MMD) is a rare and progressive cerebrovascular disorder bimodally distributed peaking in childhood and young adults around age 40, although its reach spans all ages (1). Named from the Japanese term for "puff of smoke," MMD is characterized by the stenosis of terminal internal carotid arteries and the emergence of fragile collateral vessels that resemble wispy clouds on angiography. This compensatory network often fails, leading to chronic cerebral hypoperfusion and a heightened risk of ischemic and hemorrhagic strokes — devastating outcomes that can rob patients of motor function, cognition, or life. Moyamoya disease symptoms are primarily driven by inadequate blood flow to the brain and can be devastating. Patients often experience neurological deficits such as sudden weakness or paralysis, typically affecting one side of the body. They may also suffer from aphasia, vision problems, or
involuntary movements. Without effective treatment, the chronic hypoperfusion and heightened risk of ischemic or hemorrhagic strokes can lead to severe cognitive impairment, permanent disability, and, in some cases, death (1). Though its pathogenesis remains elusive, genetic mutations like RNF213, prevalent in East Asian populations (2), hint at a complex interplay of hereditary and environmental triggers. Despite its rarity, MMD’s impact is profound, necessitating urgent advances in diagnosis and care. Treatment strategies, from medical management to sophisticated surgical revascularization, aim to restore blood flow and avert neurological decline, yet challenges persist. Medical therapies cannot halt the progressive vessel narrowing, and surgery, despite its efficacy, carries inherent risks such as perioperative stroke and the need for highly specialized neurosurgical expertise. Moreover, predicting long-term neurological outcomes for all patients remains difficult, and the disease's progressive nature necessitates lifelong clinical and imaging surveillance. This paper explores MMD comprehensively: from the normal physiology of cerebral blood flow and its disruption in MMD, through its genetic and hemodynamic underpinnings, to its clinical presentation, diagnostic approaches, and evolving therapies. By illuminating these facets, we seek to enhance understanding and guide the development of targeted interventions for improved patient outcomes.
Methods
A comprehensive literature search was conducted to identify relevant studies, case reports, and review articles on MMD focusing on diagnosis, treatment, and patient outcomes globally. The primary search engine employed was Google Scholar.
To maximize the relevance of the retrieved literature, the search strategy incorporated keywords and phrases such as "Moyamoya disease," "Moyamoya
diagnosis," "Moyamoya treatment," "Moyamoya outcome," "Moyamoya surgery," "Moyamoya revascularization," "Moyamoya complications," and "Moyamoya case report." Boolean operators (AND, OR) were implicitly used by Google Scholar's algorithm, which prioritizes results containing multiple terms.
While our focus was on recent evidence, with roughly 90% of included studies published in the last 15 years, the search criteria were occasionally broadened. This was necessary due to MMD rarity and the limited experimental data available. Including some older, foundational studies allowed us to gather critical insights into diagnostic approaches, long-term outcomes, and complex case presentations that have significantly shaped our understanding of the disease.
The set of studies on MMD is mostly made up of more recent studies, with approximately 90.9% of citations being published in the last 15 years. This is purposeful to prioritize recent evidence and to also recognize that evidence surrounding this rare condition is rapidly evolving. Most of the selection criteria were driven by relevance to MMD with consideration of diagnostic strategies, long-term outcomes, and complex cases. The studies are drawn from reputable medical journals.
In addition, artificial intelligence was used to structurally organize this paper and ensure grammatical quality. Specifically, each section of this manuscript was grammatically fully checked (3).
Discussion
Normal Physiology of Cerebral Blood Flow Regulation
The brain's vascular system is a specialized network that ensures a continuous supply of oxygen and nutrients for neuronal function. The internal carotid and vertebral arteries, key conduits of this system, converge at the brain base to form the Circle of Willis. This structure enables collateral blood flow, allowing blood to bypass blockages or narrowing in major arteries, thus protecting against ischemic events like strokes and maintaining stable perfusion across both hemispheres (2). Beyond nutrient delivery, the Circle of Willis may also regulate cerebral pressure and maintain intracranial homeostasis. In MMD, however, narrowing or occlusion of the internal carotid arteries triggers the formation of fragile collateral vessels, which often fail to compensate adequately, leading
to ischemic or hemorrhagic complications. This vulnerability highlights the critical role of the cerebral arterial network in sustaining circulation.
Cerebral blood flow (CBF) is tightly regulated to deliver a consistent supply of oxygen and nutrients despite fluctuations in systemic blood pressure. This autoregulation relies on multiple mechanisms. The myogenic response enables cerebral arteries to constrict or dilate in response to changes in intraluminal pressure, stabilizing CBF (4). Metabolic regulation, meanwhile, adjusts blood flow to meet local neuronal demands; increased activity releases vasodilators like adenosine and nitric oxide (NO), enhancing CBF in active regions. Though less dominant, the autonomic nervous system, particularly sympathetic innervation, also modulates CBF (5).
Additionally, CBF responds to changes in arterial carbon dioxide (PaCO2) and pH: hypercapnia (elevated PaCO2) and acidosis induce vasodilation, while hypocapnia (reduced PaCO2) and alkalosis cause vasoconstriction (4). Endothelium further regulates CBF by releasing vasoactive substances, such as NO, a potent vasodilator, and endothelin-1 (ET-1), a vasoconstrictor. The balance between these factors is essential for maintaining a vascular tone and optimal CBF (5).
Pathophysiology of Moyamoya Disease
The defining feature of MMD is the relentless, progressive stenosis and eventual occlusion of the distal internal carotid arteries and the proximal segments of the anterior and middle cerebral arteries, a process that severely compromises (CBF) to critical brain regions. The brain compensates for ischemia by forming a frail, unstable network of collateral vessels, often called a "puff of smoke" on angiography, which are prone to rupture or thrombosis. The etiology of MMD remains incompletely understood, shrouded in a complex interplay of genetic predisposition and environmental influences. Central to its pathogenesis are genetic mutations, with the RNF213 gene emerging as a pivotal player, particularly in East Asian populations where its prevalence reaches up to 90% in familial cases (6). The RNF213 mutation disrupts vascular stability by impairing endothelial cell function and promoting aberrant angiogenesis, potentially through heightened inflammation or oxidative stress. Other implicated genes, such as ACTA2 and GUCY1A3, further underscore the hereditary basis
of this disorder, altering smooth muscle integrity and guanylate cyclase signaling, which compromise vascular resilience (6). Environmental triggers, such as chronic inflammation or autoimmune processes, may amplify these genetic vulnerabilities, accelerating disease progression (6).
The stenotic occlusion of major cerebral arteries precipitates a marked reduction in CBF, often dropping below 30 mL/100g/min in affected regions, while crippling cerebrovascular reserve, the brain’s capacity to dilate vessels in response to heightened metabolic demand or stress (7). This impaired autoregulation, compounded by endothelial dysfunction, leaves the brain susceptible to ischemic insults. Endothelial cells in MMD exhibit reduced NO production, coupled with elevated endothelin-1 (ET-1), tilting the balance toward vasoconstriction and exacerbating hypoperfusion (5). The fragile collaterals, meanwhile, pose a dual threat: thrombosis can precipitate ischemic strokes, abruptly starving brain tissue of oxygen, while their rupture, driven by hemodynamic stress on weakened vessel walls, can unleash devastating hemorrhagic strokes, flooding the intracranial space with blood. Although MMD ravages the cerebral vasculature, its ripple effects extend systemically; chronic hypoperfusion erodes cognitive faculties, manifesting as memory deficits, executive dysfunction, and slowed processing speed, and undermines overall neurological function, profoundly impacting quality of life. These multifaceted consequences highlight the urgent need to unravel MMD’s physiological and molecular underpinnings to devise more effective therapeutic strategies.
Clinical Signs and Symptoms
MMD manifests through a spectrum of neurological symptoms driven by cerebral hypoperfusion and collateral vessel instability. Brief episodes of neurological dysfunction, such as unilateral weakness (hemiparesis), numbness, aphasia (language disorder caused by damage to the parts of the brain that control language expression and comprehension), or visual disturbances — termed transient ischemic attacks (TIAs) — typically resolve within minutes to hours and may be triggered by hyperventilation or dehydration, which exacerbate vasoconstriction. Sudden ischemic strokes, occurring in 50–75% of untreated patients (2), can cause permanent deficits like hemiplegia or speech impairment, reflecting occlusion of the middle cerebral artery. Seizures, arising from ischemia or cortical
scarring, affect a subset of patients (7), while chronic hypoperfusion leads to cognitive deficits, memory loss, executive dysfunction, and slowed processing speed, sometimes accompanied by irritability or depression due to frontal lobe involvement. Hemorrhagic events are more common in adults and can be up to 40% in East Asian cohorts (7), and result from fragile collateral vessel rupture, causing intracerebral or subarachnoid hemorrhage with severe headaches, neck stiffness, or unconsciousness. Chronic headaches, from vascular anomalies, are frequent across all ages. In children, developmental delays or rare choreiform movements signal chronic ischemia, contrasting with adults’ hemorrhagic predominance. Chorea refers to involuntary and irregular movement that appears to flow from one body part to another (8). These movements are typically spontaneous and purposeless, affecting various muscle groups, including the face, limbs, and trunk (8). Choreiform movements are a descriptive term for these specific types of involuntary, dance-like movements. In children with Moyamoya disease, these movements signal chronic brain ischemia, indicating insufficient blood supply to certain brain regions (8). These symptoms stem from progressive arterial stenosis, reducing (CBF), and the fragility of compensatory collaterals. Ischemic events arise from thrombosis or critically low perfusion, while hemorrhages reflect hemodynamic stress on weakened vessels. Symptom severity and type depend on the location and extent of affected brain tissue, with untreated cases progressing from transient to permanent deficits.
Diagnosis
Diagnosing MMD relies on a multimodal imaging approach to confirm progressive stenosis of the terminal internal carotid arteries (ICAs), visualize compensatory collateral networks, and assess cerebral perfusion deficits (9). Magnetic resonance imaging (MRI) and magnetic resonance angiography (MRA) serve as initial, non-invasive cornerstones. MRI detects ischemic infarcts, white matter lesions, or hemorrhagic sequelae in brain parenchyma, often revealing silent strokes in asymptomatic patients. MRA delineates the characteristic narrowing or occlusion of the distal ICAs and proximal anterior/ middle cerebral arteries, alongside the hazy, cloudlike "Moyamoya" collaterals — lenticulostriate or thalamo-perforating vessels — that define the disease’s angiographic signature. Time-of-flight MRA,
commonly used, offers high sensitivity for detecting these vascular anomalies without contrast, though its resolution may miss subtle collateral networks (10).
MRI and MRA are the preferred initial, non-invasive diagnostic tools for Moyamoya disease due to their ability to clearly visualize the characteristic narrowing of the internal carotid arteries and the compensatory "moyamoya" collateral vessels, as well as assess brain tissue for signs of ischemia or hemorrhage, all without exposing the patient to radiation, which is particularly beneficial for pediatric cases (11). While CT can detect acute issues, MRI and MRA offer superior soft tissue contrast and detailed vascular imaging, making them more effective for comprehensive diagnosis and monitoring of this progressive condition (12). Cerebral angiography remains the gold standard, providing unparalleled detail of cerebral vasculature. This invasive technique maps the extent of ICA stenosis, quantifies collateral vessel proliferation, and enables staging via the Suzuki classification (stages I–VI), which tracks disease progression from mild narrowing to extensive collateral dependence and ICA occlusion (3). The "puff of smoke" appearance — most striking in advanced stages — arises from dilated perforating arteries compensating for reduced flow, a finding critical for distinguishing MMD from mimics like atherosclerosis or vasculitis. Despite its precision, angiography carries risks (e.g., stroke from catheter manipulation), prompting its use primarily for confirmation or preoperative planning (9).
Functional imaging complements structural assessments. Single-photon emission computed tomography (SPECT), often paired with acetazolamide challenge, quantifies cerebral blood flow (CBF) and cerebrovascular reserve, identifying ischemic regions at risk of infarction. Hypoperfused areas (CBF <30 mL/100 g/min) correlate with clinical symptoms like TIAs and guide surgical candidacy (7). Transcranial Doppler (TCD) ultrasonography, a non-invasive tool, measures CBF velocity in major arteries and assesses autoregulatory capacity; diminished velocity or reactivity to CO2 changes signals impaired hemodynamics. While less specific than SPECT, TCD’s portability suits serial monitoring (8). Diagnosis hinges on integrating these findings with clinical presentation — TIAs, strokes, or seizures in children; hemorrhages in adults — excluding secondary causes (e.g., sickle cell disease, radiation injury). Advanced cases may show “ivy” signs on MRI (leptomeningeal collaterals) or flow voids in the basal ganglia, further supporting MMD.
Collectively, these tools not only confirm the diagnosis but also stratify disease severity, informing prognosis and therapeutic strategies like revascularization.
Treatment Strategies
The management of MMD encompasses both medical and surgical approaches, tailored to disease stage, patient age, and clinical presentation (ischemic vs. hemorrhagic). These strategies aim to mitigate cerebral hypoperfusion, prevent stroke, and preserve neurological function by addressing the dual threats of thromboembolism and fragile collateral vessel rupture, guided by advanced imaging to assess revascularization needs (11).
Medical therapy serves as an adjunct to surgical intervention or a temporizing measure for patients unsuitable for surgery. Antiplatelet therapy, such as aspirin (typically 81–325 mg daily), reduces the risk of thromboembolic events by inhibiting platelet aggregation in stenosed vessels and fragile collaterals, a critical consideration given the high incidence of ischemic strokes (50–75% in untreated cases) (6). Evidence suggests modest efficacy in preventing TIA recurrence, though it does not halt disease progression (3). Calcium channel blockers (e.g., nimodipine or nicardipine) are employed to prevent vasospasm and enhance cerebral blood flow (CBF) by relaxing vascular smooth muscle, particularly in patients with symptomatic vasoconstriction triggered by hyperventilation or dehydration (10). For patients with pre-existing heart failure (HF), particularly those with reduced ejection fraction, calcium channel blockers require careful consideration. While dihydropyridine CCBs like nimodipine and nicardipine primarily cause vasodilation and are less likely to directly depress myocardial contractility than non-dihydropyridine CCBs (e.g., verapamil, diltiazem), they can induce systemic hypotension. This hypotension may compromise cardiac output and worsen HF symptoms or end-organ perfusion. In such cases, management involves close hemodynamic monitoring, careful titration of the CCB dose, and optimizing volume status. Alternative strategies for improving CBF or managing vasoconstriction would focus on non-pharmacological approaches like maintaining euvolemia and avoiding triggers, as direct pharmacological substitutes for vasospasm prevention are limited (10). These agents may also alleviate chronic headaches, a frequent MMD symptom. Control of cardiovascular risk factors — hypertension,
diabetes, and hyperlipidemia — is essential to minimize additional vascular stress, with blood pressure targets often set below 130/80 mmHg to balance perfusion needs and hemorrhage risk (3). Despite these measures, medical management alone fails to address the underlying progressive stenosis, underscoring the primacy of surgical intervention.
Surgical revascularization is the cornerstone of MMD treatment, proven to restore CBF, reduce stroke risk, and improve long-term outcomes. Imaging modalities like MRI and SPECT scans guide patient selection and postoperative assessment (11). Techniques fall into 3 categories: direct, indirect, and combined bypass, each leveraging distinct mechanisms to bypass occluded arteries and bolster cerebral perfusion.
Direct bypass, exemplified by superficial temporal artery to middle cerebral artery (STA-MCA) anastomosis, establishes an immediate extracranialto-intracranial conduit (10). A branch of the external carotid artery (e.g., STA) is meticulously sutured to a cortical branch of the middle cerebral artery, bypassing the stenosed internal carotid artery (ICA) (10). The procedure delivers instantaneous CBF augmentation — often increasing from <30 mL/100 g/min to >50 mL/100 g/min in affected regions — making it particularly effective for patients with acute ischemic symptoms or poor cerebrovascular reserve (7). Success hinges on meticulous microsurgical technique and donor-recipient vessel patency, with patency rates exceeding 90% in experienced centers (1). However, it carries perioperative risks, including hyperperfusion syndrome (5–10% incidence), which can precipitate transient neurological deficits or hemorrhage (7).
Indirect bypass: Indirect methods, such as encephaloduroarteriosynangiosis (EDAS), encephalomyosynangiosis (EMS), or dural inversion, promote angiogenesis by placing vascularized tissue (e.g., temporalis muscle or pericranium) onto the brain surface (9, 11). Over weeks to months, this stimulates collateral vessel formation from extracranial sources, gradually increasing CBF (10). EDAS, the most widely used, leverages the STA without direct anastomosis, making it technically more straightforward and safer for pediatric patients or those with small-caliber vessels (10). Efficacy varies, with angiogenesis success rates ranging from 60% to 80%, influenced by factors like patient age (higher in children) and RNF213 mutation status (8). Specifically, the homozygous RNF213 mutation is associated with lower
angiogenesis success rates. While slower to effect, indirect bypass reduces perioperative risks and suits early-stage or asymptomatic cases.
Combined bypass: Integrating direct and indirect techniques, combined bypass maximizes revascularization by pairing immediate flow (via STAMCA) with long-term collateral growth (via EDAS or similar) (11). This approach is increasingly favored for advanced MMD (Suzuki stages IV–VI), where extensive collateral dependence and profound hypoperfusion demand robust intervention (12). Studies report superior stroke-free survival rates — up to 95% at 5 years — compared to single-technique strategies, though it requires excellent surgical expertise and longer operative time (13).
Surgical candidacy hinges on imaging evidence of compromised CBF (e.g., SPECT showing <30 mL/100 g/ min) and clinical factors like recurrent TIAs or strokes (10). Direct bypass excels in adults with acute ischemia, while indirect methods dominate in children due to their robust angiogenic potential (14). Combined approaches bridge these benefits, though comparative trials remain limited (15).
Beyond established methods, experimental approaches are gaining traction. Gene therapy targeting RNF213 or angiogenesis pathways (e.g., VEGF upregulation) aims to correct endothelial dysfunction and enhance collateral stability, though human trials are nascent (16). Stem cell therapy, using mesenchymal stem cells to promote vascular repair, shows promise in preclinical models by boosting NO production and reducing inflammation (14). Pharmacological agents like cilostazol, a phosphodiesterase inhibitor, are under investigation for their dual antiplatelet and vasodilatory effects, potentially augmenting CBF more effectively than aspirin alone (13). These innovations, while preliminary, signal a shift toward personalized, molecularly targeted MMD management.
Treatment selection balances efficacy, risk, and patient factors, with imaging playing a pivotal role in planning and follow-up (17). Direct bypass offers rapid relief but demands surgical precision; indirect bypass prioritizes safety and long-term gains; combined bypass optimizes outcomes at higher complexity. Postoperative care includes antiplatelet continuation, CBF monitoring via TCD or SPECT, and managing hyperperfusion risks with strict blood pressure control (e.g., systolic 100–140 mmHg) (16). Multidisciplinary teams —
neurosurgeons, neurologists, and rehabilitation specialists — ensure holistic care, addressing both vascular and neurological sequelae.
Prognosis and Long-Term Management
Physiologically, MMD’s prognosis hinges on the brain’s ability to adapt to chronic hypoperfusion and the success of interventions in re-establishing hemodynamic stability. Pre-treatment, the stenotic ICAs impair autoregulation — the myogenic and metabolic mechanisms that usually stabilize CBF despite systemic pressure fluctuations. This forces reliance on fragile collateral vessels, prone to thrombosis or rupture with their thinwalled, disorganized structure (18). Successful revascularization, whether direct (e.g., STA-MCA bypass) or indirect (e.g., EDAS), reintroduces extracranial blood flow, elevating regional CBF and partially restoring autoregulatory capacity (19). For instance, direct bypass can boost CBF from <30 mL/100 g/min to >50 mL/100 g/min acutely, alleviating ischemic stress (18). However, this sudden hemodynamic shift risks hyperperfusion syndrome (5–10% incidence), where excessive flow overwhelms previously hypoperfused tissue, triggering edema or hemorrhage due to disrupted endothelial integrity and elevated ET-1 levels (18). In the long term, the balance between NO)-mediated vasodilation and ET-1-induced vasoconstriction must stabilize to sustain vascular tone and prevent such complications (15).
Effective long-term management integrates regular physiological monitoring with tailored interventions. MRI and MRA track disease progression and bypass patients, assessing whether CBF remains above ischemic thresholds (18). SPECT with acetazolamide challenge quantifies cerebrovascular reserve, revealing regions at risk if autoregulation falters (17). Transcranial Doppler (TCD) provides realtime CBF velocity data, guiding adjustments in antiplatelet therapy (e.g., aspirin) or blood pressure control (target systolic 100–140 mmHg) to prevent thromboembolism or hyperperfusion (19). For hemorrhagic-prone patients, managing collateral vessel stability is critical; excessive hemodynamic stress on these vessels, exacerbated by hypertension or PaCO2 fluctuations can precipitate rupture, necessitating strict PaCO2 homeostasis via controlled ventilation or lifestyle adjustments (20). Emerging therapies, like cilostazol, aim to enhance NO production and vasodilation, potentially offering a
dual protective effect against ischemia and vascular instability (20).
Conclusion
MMD is a formidable challenge in cerebrovascular medicine, defined by its progressive stenosis of the internal carotid arteries and the precarious formation of fragile collateral vessels. This paper has elucidated the intricate interplay between normal (CBF) regulation — reliant on autoregulatory mechanisms like myogenic responses and NOmediated vasodilation — and its disruption in MMD, where endothelial dysfunction, genetic mutations (notably RNF213), and aberrant angiogenesis precipitate chronic hypoperfusion and vascular instability. These physiological disruptions manifest clinically as transient ischemic attacks, debilitating strokes, cognitive decline, and hemorrhagic events, underscoring the disease’s profound impact on patients across age groups. Current diagnostic tools, from MRI/MRA to cerebral angiography and SPECT, enable precise visualization of these vascular anomalies and guide therapeutic decisionmaking, while treatment strategies — spanning medical management with antiplatelets and surgical revascularization via direct, indirect, or combined bypass — offer critical avenues to restore CBF and mitigate stroke risk.
Surgical revascularization emerges as a cornerstone of MMD management, with techniques like STAMCA bypass and EDAS demonstrably improving hemodynamic stability and long-term outcomes, evidenced by stroke-free survival rates reaching up to 95% at 5 years in advanced cases (21). Yet the disease’s complexity demands more than current solutions can fully address. The persistent risks of hyperperfusion syndrome, the variable efficacy of indirect bypass, and the incomplete understanding of MMD’s genetic and environmental triggers highlight significant gaps in knowledge and practice. Moreover, the systemic ripple effects of chronic hypoperfusion cognitive impairment, developmental delays in children, and reduced quality of life emphasize the need for holistic, multidisciplinary care beyond vascular correction (22).
Unraveling the molecular underpinnings of MMD, such as the role of RNF213 in endothelial dysfunction or the potential of genes like ACTA2 and GUCY1A3 as therapeutic targets, is imperative for shifting from symptomatic management to disease-modifying
interventions. Emerging approaches, including gene therapy to enhance vascular stability, stem cell therapy to promote repair, and pharmacological agents like cilostazol to bolster CBF, hold promise but require rigorous clinical validation. Enhanced genetic screening, particularly in high-prevalence populations like East Asians, could refine risk stratification and personalize treatment (23). Ultimately, advancing MMD care hinges on bridging these scientific and clinical frontiers through collaborative research, innovative trials, and a deeper grasp of its physiological roots. By doing so, we can transform the prognosis of this enigmatic disorder, offering patients not just survival, but sustained neurological health and improved quality of life.
Disclosures
The authors declare no competing financial or nonfinancial interests in relation to the work described.
Acknowledgments
We gratefully acknowledge Dr. Gregory Shanower, Dr. Mary Pelkowski, and Dr. Michael Hardisky for their insightful discussions and valuable advice throughout this research. Their feedback significantly contributed to the clarity and direction of this work. AI was used to assist with language editing and grammar, with the final text reviewed and edited by the author and reviewers.
References
1. Cho WS, Kim JE, Kim CH, et al. Long-term outcomes after combined revascularization surgery in adult moyamoya disease. Stroke 2014;45(10):3025-3031.
2. Ihara M, Yamamoto Y, Hattori Y, et al. Moyamoya disease: diagnosis and interventions. Lancet Neurol. 2022;21(8):747-758. doi:10.1016/S14744422(22)00165-X
3. Grok 3, xAI. Artificial intelligence model for data analysis and information synthesis. xAI; 2025.
4. Zou X, Liao Y, Jiang C, et al. Brain perfusion, cognition, and plasma Alzheimer's biomarkers in moyamoya disease. Alzheimer's Dement 2023;19(8):3316-3326. doi:10.1002/alz.12958
6. Mineharu Y, Miyamoto S. RNF213 and GUCY1A3 in moyamoya disease: key regulators of metabolism, inflammation, and vascular stability. Front Neurol. (2021): 687088.
7. Pettersson SD, Olofsson HK, Ali S, Szarek D, Miękisiak G, Ogilvy CS. (2023). Risk factors for ischemic stroke after revascularization surgery in patients with moyamoya disease: an age-stratified comparative meta-analysis. Wor. Neur, 173, 146157.
8. Kraemer M, Trakolis L, Platzen J, Schwitalla JC, Bersano A, Albrecht P, et al. Movement symptoms in European Moyamoya angiopathy—First systematic questionnaire study. Clin Neurol Neurosurg. 2017;152:52-56. doi: 10.1016/j. clineuro.2016.11.017
9. Funaki T, Takahashi JC, Houkin K, Kuroda S, Takeuchi S, et al. (2018). High rebleeding risk associated with choroidal collateral vessels in hemorrhagic moyamoya disease: analysis of a nonsurgical cohort in the Japan Adult Moyamoya Trial. Jour. Neuro, 130(2), 525-530.
10. Gao G, et al. Factors influencing collateral circulation formation after Indirect Revascularization for Moyamoya Disease: A Narrative Review. Translational Stroke Research 15.6 (2024): 1005-1014.
11. Kuroda S, Houkin K. Bypass Surgery for Moyamoya Disease—Concept and Essence of Surgical Techniques. Neurologia Medico-Chirurgica 52.5 (2012): 287-294.
12. Du L, et al. Imaging methods for surgical revascularization in patients with moyamoya disease: an updated review. Neurosurgical Review (2022): 1-14.
13. Yamada I, Nakagawa T, Matsushima Y, Shibuya H. High-resolution turbo magnetic resonance angiography for diagnosis of moyamoya disease. Stroke. 2001;32(8):1825-1829. doi: 10.1161/01. STR.32.8.1825.
14. Mertens R, et al. The genetic basis of moyamoya disease. Translational Stroke Research (2022): 1-21.
15. Demartini Jr. Z, et al. Moyamoya disease and
syndrome: a review. Radiologia Brasileira 55 (2022): 31-37.
16. Canavero I, et al. Clinical management of moyamoya patients. Journal of Clinical Medicine 10.16 (2021): 3628.
17. Gonzalez NR, et al. Adult moyamoya disease and syndrome: current perspectives and future directions: a scientific statement from the American Heart Association/American Stroke Association. Stroke 54.10 (2023): e465-e479.
18. Asselman C, et al. Moyamoya disease emerging as an immune-related angiopathy. Trends in Molecular Medicine 28.11 (2022): 939-950.
19. Velo M, et al. Moyamoya vasculopathy: cause, clinical manifestations, neuroradiologic features, and surgical management. World Neurosurgery 159 (2022): 409-425.
20. Fox BM, et al. Pathophysiology of vascular stenosis and remodeling in moyamoya disease. Frontiers in Neurology 12 (2021): 661578.
21. Teo M, et al. Short- and long-term outcomes of moyamoya patient post-revascularization. Journal of Neurosurgery 138.5 (2022): 1374-1384.
22. Gatti JR, Gonzalez Torriente A, Sun LR. Clinical presentation and stroke incidence differ by moyamoya etiology. Journal of Child Neurology 36.4 (2021): 272-280.
23. Singh R, et al. Research advances in the diagnosis and treatment of moyamoya disease: a bibliometric analysis. Neurosurgical Review 45.3 (2022): 1977-1985.
Rural EmPATH Unit Decreases Subjective Distress Levels in Patients with Psychiatric Complaints
Presenting to the Geisinger Medical Center Emergency Department in Danville, Pa.
Kamil Falkowski1†, Tanner Thompson1†, Hunter Yarnell2, Mia Gianello1†, and Jennifer Margaret Yarnell3
1Geisinger College of Health Sciences, Scranton, PA 18509
2Misericordia University, Dallas, PA 18612
3Geisinger Behavioral Health, Danville, PA 17822
†Doctor of Medicine Program
Correspondence: kfalkowski@som.geisinger.edu
Abstract
Background: Mental health issues, including substance abuse, are among the leading causes of emergency department (ED) visits and account for an increasing share of overall presentations. Limited psychiatric bed availability contributes to high reliance on ED boarding, resulting in extended wait times, decreased efficiency, and worsening distress. This overcrowding strains ED resources, leading to lower-quality care and increased mortality, especially in rural areas where access to acute mental health care is limited. The Emergency Psychiatry Assessment, Treatment, and Healing (EmPATH) model addresses this gap through creating a hospital-based crisis stabilization unit separate from the general ED, available to all medically cleared patients except those with violent behavior or medical instability. Staffed by psychiatrists, nurses, social workers, and psychologists, EmPATH units provide rapid evaluation, observation, and treatment. Geisinger Medical Center (GMC) ED, like many nationwide, experiences high psychiatric boarding times. In response, a GMC EmPATH unit was established to relieve ED pressure, emphasizing patient safety, emotional regulation, and social engagement. However, data on patient experiences in rural EmPATH units remains limited.
Objective: The study aims to assess changes in patient-reported distress levels from admission to discharge in the EmPATH unit and compare perceived safety and calmness to experiences in the general ED/ hospital, along with analyzing qualitative feedback from free-text responses.
Methods: This mixed-methods study integrates quantitative data (survey scores and ratings) and qualitative data (subjective experiences), collected before patient discharge by EmPATH staff. Participation was voluntary with no exclusion criteria. Descriptive and inferential statistics were used for quantitative analysis, and thematic analysis was applied to qualitative responses.
Results: Patients reported a decrease in subjective distress scores: p-value 1.78E-14, 95% C.I. (-4.58, -3.25). They also rated the EmPATH unit as providing greater safety: 9.18, and calmness: 8.60, compared to the ED/hospital (using a 1–10 scale with 1 indicating less safe/calm and 10 indicating more safe/calm). Thematic analysis of open-ended responses highlighted themes such as a therapeutic environment, feeling heard and cared for, and identifying the unit as a space for reflection.
Conclusion: Our findings suggest that the EmPATH unit offers a more therapeutic and less traumatizing alternative for patients presenting with psychiatric emergencies. Patients reported significant reductions in distress and greater perceived safety and calmness. Additionally, qualitative findings revealed key themes supporting positive patients’ experiences at the EmPATH unit. However, limitations of the study include data collection at a singular point in time, voluntary participation, and limited participant diversity. Future studies should involve larger, more diverse samples and use standardized assessment tools such as the Columbia Suicide Severity Rating Scale to provide more objective measures of patient experiences at the EmPATH unit.
Introduction
In 2014, mental health issues, including substance abuse, were the most common reasons for ED visits nationwide (1). Between 2006 and 2011, ED visits by patients with mental health disorders and comorbidities increased by 20.5% and 53.5%, respectively (2). Mental health-related ED visits now account for approximately 12–15% of all emergency visits (3).
This trend is particularly pronounced in rural areas, where patients face significant challenges accessing timely acute mental health care (4). A literature review highlights the urgent need to address these challenges. Mental health and associated disorders are the fourth most frequently identified rural health priority. Rural areas are disproportionately affected by male suicides, yet nearly 75% of rural counties lack a single psychiatrist, and nearly 95% lack a child psychiatrist (5). Additionally, rural youths present to the ED for self-harm at higher rates than their urban counterparts (6). It is evident that emergency psychiatric care in rural areas is severely lacking, necessitating innovative solutions to improve access.
The surge in mental health ED visits presents significant operational and financial challenges for emergency departments. Limited inpatient psychiatric beds often result in “boarding,” where patients are held in the ED for days without treatment or a hospital room, frequently in hallways or corridors (7, 8). Psychiatric patients awaiting inpatient placement remain in the ED 3.2 times longer than non-psychiatric patients, preventing 2.2 bed turnovers per psychiatric patient (9).
This overcrowding places an immense pressure on EDs, leading to decreased treatment quality and increased mortality (10). Additionally, the noisy, overstimulating ED environment exacerbates selfreported distress levels (11). Financially, prolonged psychiatric patient boarding and reduced turnover result in a loss of $2,264 per patient (9). Of note, increased boarding times in the ED results in a higher likelihood of symptom escalation leading to greater demands on nursing staff (1:1s) and security needs.
Various programs, such as short-stay mental health crisis units and community crisis centers, aim to reduce ED use by psychiatric patients (12). However, many of these programs exclude more acute patients, ultimately directing them to EDs.
To address this gap, the Emergency Psychiatric Assessment, Treatment, and Healing (EmPATH) model was developed as a hospital-based crisis stabilization unit. Unlike other programs, EmPATH units are designed as destinations for all medically clear patients in crisis, with few exclusion criteria such as violent behavior and medical instability. They feature multidisciplinary staff, including psychiatrists, RNs, social workers, and psychologists, who provide rapid evaluation, observation, and treatment. Notably, restraint use in EmPATH units is typically under 1%. Additionally, these units are distinct from medical EDs, offering open, calming, and comfortable spaces that promote healing.
A review of various EmPATH units demonstrated substantial operational and financial benefits for healthcare systems and emergency departments. These units effectively reduced ED recidivism rates, psychiatric inpatient admissions, restraint use, and boarding times, thereby limiting ED overcrowding. Additionally, they were shown to increase outpatient follow-up care established at discharge. Reports also indicated increased cost savings and ED revenue (13, 14, 15, 16). However, research on patient experiences with the EmPATH model remains limited. One example from an EmPATH unit at the Mercy San Juan Medical Center reported patient satisfaction score of 85% (17).
Overview of Geisinger Medical Center EmPATH Unit
Geisinger Medical Center, located in Danville, Pennsylvania, is part of the Geisinger health system and serves as a Level I Trauma Center for the central PA community, which consists largely of a rural population. As a result, the hospital faces challenges similar to those of other rural healthcare providers in addressing acute psychiatric care.
In January 2021 alone, approximately 100 psychiatric patients at Geisinger Medical Center experienced boarding times exceeding 4 hours. To improve acute psychiatric care and access, an adult EmPATH unit was established in late 2024, featuring a flexible clinical model that emphasizes safety, emotional regulation, and social engagement.
The Geisinger Medical Center EmPATH unit is designed to stabilize psychiatric emergencies and serves as a designated destination for medically cleared patients in crisis before determining the need for inpatient admission. Upon arrival, patients are quickly evaluated by the psychiatry consult
service, and treatment is initiated if they have been determined to be appropriate for the unit. Patients are then closely monitored by medicine and psychiatry with dispositional reassessment every 24 hours. The unit itself is physically separate from the general Emergency Department. It is designed to foster a relaxing and calming environment. It includes an open milieu with murals of nature from the community. There is a quiet room for privacy as well as multiple materials, therapeutic aids for distraction. The unit follows a multidisciplinary staffing model, including a psychiatrist/psychologist, RNs (including one psych RN), a licensed social worker, a peer support specialist, and other assistants/techs.
The Geisinger EmPATH unit also employs a unique nonlinear clinical model to accommodate complex needs, crisis situations, and symptom presentations. This model consists of 3 main principles:
Safety
- Identify target symptoms and triggers
- Conduct risk assessments
- Establish physical and psychological safety in the EmPATH unit
Emotional Regulation
- Implement interventions to modulate affect
- Use cognitive behavioral strategies focused on relaxation and problem-solving
Social Engagement Module
- Identify social supports (e.g., home, work, school)
- Facilitate communication with support systems
- Assist in transitioning to home or a higher level of care
- Social enrichment and other supplemental activities
Overall, the Geisinger Medical Center EmPATH unit aims to provide timely acute psychiatric care while alleviating pressure on the Emergency Department.
Objective
Research on subjective patient experiences in EmPATH units, particularly in rural settings, remains limited. This study aims to assess changes in distress levels in patients with psychiatric complaints from admission to discharge at the Geisinger Medical Center EmPATH unit using statistical analysis.
Additionally, it will evaluate patient comparisons of care at the EmPATH unit versus the Emergency Department/general hospital in 2 key domains: perceived safety and calmness, measured on a 1–10 rating scale, along with qualitative patient experiences gathered through freehand text responses.
Methods
Design
This study employs a mixed-methods approach, integrating quantitative (survey scores and ratings) and qualitative (subjective experiences) data obtained before patient disposition. It is a survey-based observational study conducted at the Geisinger Medical Center EmPATH unit over a 4-month period. In total, 41 surveys were conducted.
Participants
All patients arriving at the EmPATH unit for acute psychiatric care from the Emergency Department are included. Of note, participation in the study survey is voluntary. While there are no study-specific exclusion criteria, admission to the EmPATH unit is restricted to adult patients and does not accept patients with severe medical instability or violent behavior. Recruitment follows consecutive sampling, with EmPATH staff enrolling and obtaining consent from patients.
Ethical Considerations
Nursing staff secured consent before administering anonymous deidentified surveys. Given the vulnerability of the population, participation was voluntary, and patients could decline at any time. IRB exemption was obtained.
Data Collection Procedure
Survey data was collected on a physical copy using a standardized questionnaire upon the patient leaving the EmPATH unit, including:
- Retrospective reported distress score at arrival
- Reported distress score at disposition
- Perceived safety compared to the Emergency Department/hospital
- Perceived calmness compared to the Emergency Department/hospital
- Open-ended response to the following question: “How would you describe the environment and care provided in the EmPATH area of the ED?”
After the surveys were complete, the responses were recorded electronically, and the physical copies were disposed.
Variables and Outcomes
Our study’s variables include perceived distress (at admission and disposition), perceived safety, perceived calmness, and subjective experience. Distress scores refer to patient’s self-reported feeling of distress in the EmPATH unit measured on a 1–10 scale, with 1 indicating less distress, and 10 indicating more distress. Perceived safety refers to patients' self-reported sense of safety in the EmPATH unit compared to the Emergency Department or hospital, measured on a 1–10 scale, with 1 indicating less safe and 10 indicating more safe. Perceived calmness is the patients’ self-reported level of calmness in the EmPATH unit compared to the Emergency Department or hospital, also measured on a 1–10 scale, with 1 indicating less calm and 10 indicating more calm. Subjective experience consists of openended responses analyzed using qualitative thematic analysis. Meanwhile, the outcome measured by this study includes the change in reported patient distress from admission to disposition from the EmPATH unit.
Statistical Analysis
Descriptive statistics were used to analyze perceived safety and calmness by identifying average scores across all participants over the study duration. Inferential statistics, specifically a two-tailed paired t-test, was used to assess the change in reported patient distress, determining whether there was a statistically significant change in distress at the EmPATH unit.
Thematic Analysis
Open-ended responses to the survey question “How would you describe the environment and care provided in the EmPATH area of the ED?” were analyzed using thematic analysis to identify recurring themes and patterns in participant responses.
Results
A total of 41 participants provided paired responses for distress scores (Table 1). The retrospective reported distress score at arrival had a mean of 7.55 with a standard deviation of 2.26. The reported distress score at disposition had a mean of 3.63 with a standard deviation of 2.35. The mean change in
reported distress score was −3.91 with a standard deviation of 2.12. The 95% confidence interval for the mean change was −4.58 to −3.25. A two-tailed paired t-test yielded a p-value of 1.78 × 10-14.
A total of 37 participants provided responses for reported perceived safety, and 36 participants provided responses for reported perceived calmness (Table 2). The reported safety score had a mean of 9.18 with a standard deviation of 1.26. The reported calmness score had a mean of 8.60 with a standard deviation of 2.02.
Thematic analysis was conducted on 35 freetext responses (Table 3). Four major themes were identified: therapeutic environment, feeling heard and cared for, space for reflection, and constructive feedback. Thematic analysis identified 4 major themes of the patient experience at the EmPATH unit (Figure 1). A total of 15 responses referenced the therapeutic environment, followed by 14 responses indicating feeling heard and cared for. Space for reflection was mentioned in 6 responses, and constructive feedback was noted in 4 responses.
Table 1. Reported Distress Scores and Two-Tailed Paired T-Test Results
Table 2. Reported Perceived Safety and Calmness
Discussion
Our results indicate a statistically significant decrease in subjective patient distress scores. Furthermore, participants in the EmPATH unit reported subjective feelings of increased safety and calmness compared to their experience in the Emergency Department or general hospital setting. Thematic analysis of free-text responses revealed additional insights, with participants noting a therapeutic environment, feeling heard and cared for, and identifying the unit as a space for reflection. The analysis also uncovered constructive feedback that could be used to improve the EmPATH unit.
While our survey results show a highly statistically significant decrease in stress levels, it is important to recognize that statistical significance does not always imply clinical significance. Many outcomes may be statistically significant yet lack clinical relevance from a practical standpoint (18). From a clinical perspective, the average reported distress score at admission was 7.55, and at discharge, it was 3.63, on a scale from 1 (least distress) to 10 (most distress). These values reflect a meaningful shift: participants initially reported distress in the upper range of the scale, which dropped to the lower range by discharge. This notable change supports the argument that the statistically significant drop is also clinically significant.
Additionally, participants' reported increases in perceived safety and calmness compared to the ED/ hospital further support clinical relevance, particularly when considered alongside the themes of therapeutic environment and feeling heard and cared for. This aligns with previous studies demonstrating that calming environments can reduce restraint use in other EmPATH units (19, 20). It is not surprising that the peaceful and therapeutic setting of the EmPATH unit resulted in positive subjective experiences.
Literature has shown that environmental noise can elevate cortisol levels, disrupt sleep, increase the need for analgesia and sedation, and alter circadian rhythms (21). In our study, participants reported feeling calmer, sleeping better, and experiencing “space for reflection.” One participant noted that the unit offered reassurance and time to reflect without the fear of immediate psychiatric admission. This is important, as prior research suggests that inpatient psychiatric hospitalization can sometimes induce fear or be perceived as traumatic (22). Our findings suggest that the EmPATH unit may function as a safe
Therapeutic Environment
Key Words: Calm, Quiet, Peaceful, Heavenly, Safety, Comfort, Soothing, Relaxing Survey responses consistently described a therapeutic environment that provided physical and emotional safety and comfort. This space helped promote relaxation and rest.
Feeling Heard and Cared for Key words: Caring, Helpful, Kind, Professional, Respectful, Understanding Survey responses highlighted the EmPATH unit staff and their professionalism and empathy. Feeling heard and cared for contributed to participant’s wellbeing.
Space for Reflection
Key Words: Time to Reflect, Time to Think, Time to Unravel
Some survey responses noted the importance of having dedicated and ample amount of time to reflect, suggesting that the EmPATH unit supported self-awareness, growth, and contemplation.
Constructive Feedback
Some survey responses provided thoughtful feedback on areas that the EmPATH unit could be improved.
“Very calming, quiet, laid back.”
“Peaceful, soothing, calming.”
“It is quiet and calm, nice.”
“Made me feel safe and comforted.”
“I like the quiet so I could be in my own thoughts, good ones and bad ones.”
“Calm, more welcoming than traditional ED.”
“The caretakers were understanding, kind.”
“Therapists, nurses, aides, and doctors are all professional and caring towards my anxieties and other problems.”
“I feel like the staff was very caring and listened to my needs.”
“Very kind and compassionate. I was treated with dignity.”
“I felt I could unravel here and sleep better.”
“Gave me time to think and I know I have to make changes.”
“Quiet and a lot of time to think and reflect on life.”
“Gave me time to wind down and think without the fear of being admitted right away.”
“Great idea! Needs kinks worked out – more organized plan.”
“Some people were more helpful and talkative than others but were nice.”
“The chair are not comfortable to sleep on though.”
“The nurse and staff was nice. I felt trapped inside but that’s normal.”
35 free-text responses.
Therapeutic environment Space for reflection Feeling heard and cared for Constructive feedback 15 14 6 4
Table 3. Thematic Analysis: Conducted on
Figure 1. Distribution of Themes
and supportive alternative to inpatient hospitalization, providing patients with a temporary refuge to regroup and reflect.
The Geisinger EmPATH model focuses on 3 domains: safety, emotional regulation, and social engagement. Our results indicate that the unit is effective in promoting safety (both physical and psychological) and emotional regulation, as demonstrated by participant reports of having time to reflect, think, and consider changes. However, our study did not produce adequate data to evaluate the social engagement component of the EmPATH model.
Participants also offered constructive feedback. While the EmPATH unit follows a structured clinical model, its flexibility allows for varied applications depending on the patient and provider. As a result, some participants felt the structure was insufficient for their needs. Another commonly reported challenge was difficulty sleeping in the unit. EmPATH units do not include beds, only chairs that can be converted into a flat position. This limitation, due to space constraints, is not easily remedied. Some patients even returned to the ED to sleep in a bed before rejoining the EmPATH unit the next day. While this workaround is not ideal, it reflects the current physical limitations of the unit. Nevertheless, constructive feedback remains critical in guiding future improvements and enhancing the unit's overall effectiveness.
In our surveys, the term “distress” was intentionally chosen for its ability to broadly capture the emotional and psychological states of patients presenting with a wide range of psychiatric symptoms. Unlike narrower terms like anxiety or depression, “distress” encompasses diverse experiences and symptom clusters. Similarly, “safety” and “calmness” were selected for their inclusive, accessible nature. These familiar terms also helped keep the survey brief and easy to complete, minimizing burden on patients and staff and allowing smoother integration into the discharge process.
However, this choice presents a limitation. Patients may interpret terms like “distress,” “safety,” or “calmness” differently depending on their psychiatric presentation, for instance, a suicidal patient versus one experiencing mania. As such, the study reflects broad, subjective experiences rather than conditionspecific insights. Additionally, the anonymous nature of the survey and lack of identifiers like MRNs or
diagnoses prevent linking results to specific psychiatric emergencies.
Survey data were collected at a single point, during patient disposition, to preserve anonymity. Collecting data at both arrival and discharge would have required identifiers on the physical surveys, compromising anonymity. This presents a limitation, as the initial distress score was reported retrospectively rather than at the actual time of admission. Additionally, the study did not account for length of stay, though most patients remained in the EmPATH unit for under 24 hours.
The survey was completed on a voluntary basis, which introduces the possibility of voluntary response bias. Patients who had more positive experiences may have been more likely to complete the survey, potentially leading to skewed results and an underrepresentation of neutral or moderate experiences.
Finally, it is important to acknowledge the limitations related to sample size. The Geisinger Medical Center (GMC) EmPATH unit is relatively new, and the GMC ED serves a predominantly rural population. As such, during the 4-month study period, the number of participants was limited. Consequently, the results are less generalizable to broader populations and may be more reflective of the local rural community in and around Danville, Pennsylvania.
Future studies should aim to include a larger and more diverse sample size to improve generalizability. They should also consider categorizing patients’ subjective experiences by specific psychiatric conditions and exploring the use of more precise terminology in survey design. Additionally, incorporating standardized assessment tools, such as the Columbia Suicide Severity Rating Scale, could provide more objective measures of patient responses and enhance the rigor of evaluating the EmPATH unit model. Furthermore, as social engagement is a core domain of the EmPATH model, future research should also examine how structured social engagement impacts patient outcomes and experience, potentially through dedicated survey questions or structured observational measures.
Conclusion
In conclusion, we found that the GMC EmPATH unit significantly reduced patient-reported distress scores over the course of their stay at the unit. Patients
also reported a greater perceived sense of safety and calmness in the EmPATH unit compared to the traditional Emergency Department or general hospital setting. Additionally, thematic analysis of open-ended text responses revealed positive key themes reflecting patients’ experiences at the GMC EmPATH unit. These findings have important clinical relevance. Clinicians strive to minimize harm, yet treating psychiatric emergencies can be particularly challenging in this regard. Involuntary commitment or traditional ED environments can be distressing or even traumatic for patients in crisis. Our findings suggest that EmPATH units may offer a more therapeutic, nontraumatizing alternative for patients presenting to the ED with psychiatric emergencies. By prioritizing safety, calmness, and emotional well-being, the EmPATH model may represent a more patient-centered approach to crisis care.
Disclosures
The authors have no conflicts of interest to declare.
Acknowledgments
We would like to acknowledge the assistance from the GMC Behavioral Health Team
References
1. Weiss AJ, Wier LM, Stocks C, Blanchard J. Characteristics and trends of emergency department visits in the United States (2010–2014). Healthcare Cost and Utilization Project (HCUP) Statistical Briefs. Available from: https:// www.sciencedirect.com
2. Larkin GL, Claassen CA, Emond JA, Pelletier AJ, Camargo CA Jr. National trends in emergency department visits by adults with mental health disorders. Psychiatric Services. 2005;56(6):671–677.
3. Nordstrom K, Berlin JS, Nash SS, Shah SB, Schmelzer NA, Worley LL. Patient perspectives of emergency mental health care in a rural state. Western Journal of Emergency Medicine 2019;20(5):690–695.
4. Nicks BA, Manthey DM. Increasing emergency department visits for mental health conditions in the United States. Emergency Medicine International. 2012;2012: Article ID 360308.
5. Gamm LD, Stone S, Pittman S. Mental health and mental disorders—a rural challenge: A literature review. Journal of Rural Health. 2010;26(2):115–124.
6. Hoffmann JA, Hall M, Lorenz D, Berry JG. Emergency department visits for suicidal ideation and self-harm in rural and urban youths. The Journal of Pediatrics. 2021; 238:282–289.e1.
7. Zeller SL, Calma NM, Stone A. Boarding of mentally ill patients in emergency departments: American Psychiatric Association resource document. Western Journal of Emergency Medicine 2014;15(1):1–6.
8. Zeller SL. "Boarding" psychiatric patients in emergency rooms: One court says "no more." Primary Psychiatry. 2013;15(8):60–64.
9. Nicks BA, Manthey DM. The impact of psychiatric patient boarding in emergency departments. Emergency Medicine International. 2012;2012: Article ID 360308.
10. Moskop JC, Sklar DP, Geiderman JM, Schears RM, Bookman KJ. Overcrowding in the emergency department: Causes, consequences, and solutions—a narrative review. Journal of Emergency Medicine. 2009;42(1):124–132.
11. Basner M, Babisch W, Davis A, Brink M, Clark C, Janssen S, Stansfeld S. Emergency department noise: Mental activation or mental stress? Emergency Medicine Journal. 2014;31(6):435–440.
12. Woods HC, Skoog V, Salo M. Short-stay crisis units for mental health patients on crisis care pathways: Systematic review and meta-analysis. Journal of Psychiatric Research. 2016; 81:1–9.
13. MacKenzie A, Bilbrey C, Mullennix S. Comparison of emergency department 14-day recidivism rates in emergency behavioral health patients: EmPath versus standard ED care. Western Journal of Emergency Medicine: Integrating Emergency Care with Population Health. 2023;24(2.1). Available from: https://escholarship.org/content/ qt6zs8z6fz/qt6zs8z6fz.pdf
14. Zeller SL, Calma NM, Stone A. Emergency psychiatric assessment, treatment, and healing (EmPATH) unit: A novel model of emergency psychiatric care. Academic Emergency Medicine. 2018;25(4):399–406.
15. Zeller SL, Orrock T. Update on EmPATH: Expanding an effective ED boarding solution. Military Medicine. 2022.
16. Jeunesse DG, McElroy ML, Salmond LL, Taylor DA. Review of EmPATH units for behavioral health casualties in prolonged field care environments. Military Medicine. 2025;190(3-4):e510–e514.
17. Thomas S. My why and the EmPATH model [PowerPoint presentation]. Vituity Emergency Medicine, Mercy San Juan Medical Center.
18. Sharma H. Statistical significance or clinical significance? A researcher's dilemma for appropriate interpretation of research results. Saudi Journal of Anaesthesia. 2021;15(4):431–434.
19. Helmsley Charitable Trust. Helmsley Charitable Trust awards $784,000 to Billings Clinic psychiatric stabilization unit. 2020. Available from: https://helmsleytrust.org
20. Zeller S, Orrock T. Update on EmPATH: Expanding an effective ED boarding solution [PowerPoint presentation]. Behavioral Health Care Symposium Riverside, California Hospital Association.
21. Pal J, Taywade M, Pal R, Sethi D. Noise pollution in intensive care unit: A hidden enemy affecting the physical and mental health of patients and caregivers. Noise and Health. 2022;24(114):130–136.
22. Jina-Pettersen N. Fear, neglect, coercion, and dehumanization: Is inpatient psychiatric trauma contributing to a public health crisis? Journal of Patient Experience. 2022; 9:23743735221079138.
Dopamine and Serotonin Interactions in Schizophrenia: A Focused Review of Mechanisms and Therapeutic Implications
Sara Abdo1*, Joseph P. Osborne1*, Tawn A. Tomasi1*, and Ryder P. Lathrop1*
¹Geisinger
College of Health Sciences, Scranton, PA 18509
*Master
of Biomedical Sciences Program
Correspondence: sabdo1@som.geisinger.edu
Abstract
Schizophrenia is a chronic and complex mental illness that substantially impacts a person’s thoughts, emotions, and daily life. It is characterized by a combination of positive symptoms such as hallucinations and delusions, and negative symptoms such as emotional withdrawal, lack of motivation, and cognitive difficulties. Historically, the dopamine system has been the forefront of research on schizophrenia, but serotonin’s role has recently come into sharper focus. While serotonin is necessary for regulating emotions and mental processes, dopamine imbalances are known to result in psychotic symptoms and cognitive impairments. Interactions between these neurotransmitters affect brain functions. Many patients have seen better results due to advancements in treatment, especially secondgeneration antipsychotics that target both the dopamine and serotonin systems. However, challenges still exist, especially in effectively managing negative and cognitive symptoms. This review highlights key findings that support a more integrative model involving serotonindopamine interactions, emphasizing the role of receptors such as dopamine receptor D1, dopamine receptor D2, 5-hydroxytryptamine 1A (5-HT1A), and 5-hydroxytryptamine 2A (5-HT2A). It also explores the therapeutic potential of newer agents like brexpiprazole, cariprazine, and pimavanserin, which offer more targeted symptoms control with fewer adverse effects. Additionally, emerging research on the gut-brain axis and glutamatergic modulation presents promising directions for future treatment strategies. In this review, we looked at the research on the roles of dopamine and serotonin in schizophrenia and discussed the therapeutic approaches that could lead to more comprehensive and effective treatment plans.
Introduction
Schizophrenia is a chronic mental illness that may affect the thoughts, behaviors, and feelings of an affected person. The diagnosis of schizophrenia has been around for many years despite changes in its definition and categorization (1). According to the Diagnostic and Statistical Manual of Mental Disorders, Fifth Version (DSM-5), schizophrenia is a chronic mental illness characterized by the presence of positive and negative symptoms as well as cognitive impairment (2). The positive symptoms associated with schizophrenia refer to the presence of abnormal symptoms, such as hallucinations, thought disorders, delusions, and repetitive movements. These symptoms can cause “significant difficulties in daily functioning, and these difficulties have been associated with impaired executive functions” (3). The negative symptoms of schizophrenia are often described as the absence of normal behavior or function. This may refer to a person’s lack of motivation, and interest or verbal and emotional expression (4). Negative symptoms domain in schizophrenia typically includes 5 key constructs: blunted affect, alogia, avolition, asociality, and anhedonia, with up to 60% of patients experiencing prominent, clinically relevant symptoms that require treatment (4). Today, according to the National Alliance on Mental Illness (NAMI), prevalence of schizophrenia ranges from 0.25% to 0.64% of U.S. adults (5). while both males and females are affected by schizophrenia, studies have shown that males tend to experience an earlier onset, more severe negative symptoms, and poorer functional outcomes. In contrast, females are more likely to have a later onset, better treatment response, and relatively improved social function (6). There are standard treatments that exist today, but medicine is always evolving and searching for better treatments. The focus of research in schizophrenia now focuses on the neurotransmitters
underlying brain function, like serotonin and dopamine (7, 8).
Dopamine (DA) and serotonin (also known as 5-hydroxytryptamine or 5-HT) are two of the many neurotransmitters currently under consideration for their potential roles in the pathophysiology of schizophrenia. Dopamine plays major roles in movement, reward, cognition, and sleep. The dopamine hypothesis of schizophrenia was developed in the 1960s as an explanation for the cause of this disease (9). The original hypothesis proposed that dopamine is an inhibitory neurotransmitter involved in the pathology of schizophrenia (10). This led to the advent of many schizophrenia therapies, all targeting the dopaminergic pathways thought to be the sole cause in the etiology of this illness. The original dopamine hypothesis has since been revised, now stating that “dopamine abnormalities in the mesolimbic and prefrontal brain regions exist in schizophrenia” (10). Current schizophrenia medications block the dopamine receptor D2 in an effort to reduce positive symptoms (9). This mode of treatment has been successful, but in recent years researchers have begun to investigate alternative targets to produce more effective treatments and therapies for schizophrenia (7, 8).
Serotonin is often targeted in pharmaceutical therapies, especially for the treatment of depression and anxiety. Class of medications like selective serotonin re-uptake inhibitors (SSRI) have been proven safe and effective and are widely prescribed. These medications increase serotonin availability in the synaptic cleft by blocking the reuptake of this neurotransmitter into the presynaptic neuron (11, 12). As a neurotransmitter, serotonin functions primarily in the mood and memory areas of the brain, and it also has a role in sleep, pain, and movement (11). Treatments that decrease the availability of serotonin have been shown to generally enhance motor activity, particularly spontaneous movement in familiar environments, and may also increase behaviors related to sexuality or aggression (11). In the pathophysiology of schizophrenia, serotonin is thought to play a role, as seen in hallucinogens as well as antipsychotic medications (12, 13). Some researchers have suggested that serotonin abnormalities may be linked to certain characteristic of schizophrenia, such as negative symptoms, progressive brain changes, and chronicity. This perspective proposes that serotonin dysfunction could underlie negative
syndrome schizophrenia and may involve increased sensitivity of postsynaptic serotonin receptors. As a result, it has been proposed that treatments for schizophrenia should target both dopamine and serotonin systems by including agents that block 5-HT receptors. Further research is encouraged to better understand the impact of this selective serotonin receptor hypersensitivity in schizophrenia (12). Building on this, recent studies have explored how serotonin interacts with glutamate, furthering our understanding of schizophrenia’s biological complexity (13). These interactions involve functional crosstalk between serotonin and glutamate systems, particularly through 5-hydroxytryptamine receptor 2A (5-HT2A) and metabotropic glutamate receptors such as metabotropic glutamate receptor 2 (mGlu2). This receptor-level interaction has been shown to regulate glutamate release and cortisol signaling in brain regions implicated in schizophrenia, including the prefrontal cortex (13). Alterations in this pathway may contribute to disrupted perceptions, cognition, and behavior observed in preclinical models of the disorder (13). Early intervention has also been emphasized as crucial in improving outcomes for patients, highlighting the importance of addressing psychosis promptly (14). On a broader level, genetic studies have revealed the intricate hereditary nature of schizophrenia, which may explain its diverse symptoms (15). Additionally, disruptions in brain development during key periods, such as infancy and childhood, may play a significant role in the disorder’s onset. The role of serotonin in shaping psychiatric treatment strategies has been critical, as has the evolution of diagnostic tools and therapeutic approaches in managing schizophrenia (16–19).
The Role of Dopamine in Schizophrenia
Evidence Supporting the Role of Dopamine
The dopamine hypothesis has been a central framework for understanding schizophrenia for decades. The original hypothesis proposed that excessive dopamine activity in the mesolimbic pathway influences the positive symptoms of schizophrenia, such as hallucinations and delusions (21). The dopamine hypothesis has evolved to incorporate mesocortical hypodopaminergia as well. Hypodopaminergic activity in the mesocortical pathway accounts for cognitive deficits and negative symptoms, including reduced motivation (20, 21). Neuroimaging evidence also supports this hypothesis.
Positron emission tomography (PET) studies reveal increased dopamine synthesis and release in the striatum of individuals with schizophrenia, correlating with psychotic symptom severity (20–22).
Genetic studies further substantiate dopamine’s central role. Polymorphisms in the dopamine receptor D2 gene (DRD2), which encodes the D2 receptor, are strongly associated with schizophrenia and influence receptor density and sensitivity. Variations in the catechol-O-methyltransferase gene (COMT), which affects dopamine degradation, influence dopamine availability in the prefrontal cortex and decisionmaking (23, 24). Postmortem studies also exposed changes in dopamine receptor density in the striatum and prefrontal cortex that relate to the positive and negative symptoms of schizophrenia (25).
Another potential interaction worth noting is environmental factors such as prenatal adversity, early-life trauma, and urban upbringing work alongside genetic predispositions to compound dopaminergic dysregulation. These stressors increase the sensitivity of the dopamine system, enhancing its responsiveness and contributing to the abnormal salience attribution often observed in psychosis (23).
Dopamine Receptors and Their Implications
The dysfunction of different dopamine receptors is key to the variety of schizophrenia symptoms. D2 receptors, primarily expressed in the striatum, are commonly linked to positive symptoms and are often found to be hyperactive in schizophrenia. Antipsychotics mainly act to reduce dopamine activity in this pathway (D2 receptor antagonists), alleviating symptoms of psychosis (21, 26). However, excessive D2 blockade can result in motor problems such as rigidity and tremors, also known as extrapyramidal symptoms, and impairments in reward processing and cognitive function (27).
Conversely, dopamine receptors D1, predominantly expressed in the prefrontal cortex, are critical for cognitive functions such as working memory and executive function. Hypoactivity of D1 receptors correlates with the cognitive deficits observed in schizophrenia, which are among the most disabling aspects of the disorder. These impairments remain inadequately addressed by current treatments (28). Experimental therapies that selectively enhance D1 receptor activity have shown promise in preclinical studies but have yet to achieve widespread clinical application (29).
Emerging evidence suggests that additional receptor subtypes may have key roles in schizophrenia. These additional subtypes include dopamine receptors D3 and D4 receptors. D3 receptors, expressed in limbic regions, influence motivation and reward processing, making them potential targets for negative symptom treatment (6). D4 receptors are related to cortical excitability and may contribute to cognitive impairments (30). These findings highlight the need for a nuanced understanding of receptor-specific contributions to schizophrenia symptomatology, allowing for advancements in targeted therapeutics (26, 30).
Cognitive Deficits, Positive and Negative Symptoms
The hallmark symptoms of schizophrenia can be classified as positive symptoms, negative symptoms, and cognitive deficits. The positive symptoms of schizophrenia are linked to hyperdopaminergic activity in the mesolimbic pathway. Some of the positive symptoms associated with this disorder are hallucinations and delusions (20). Negative symptoms, on the other hand, can arise from hypodopaminergic activity in the mesocortical pathway, which disrupts motivation and goal-directed behavior. Some negative symptoms seen in schizophrenia are diminished emotional expression and social withdrawal (22). Cognitive deficits, encompassing impairments in working memory, attention, and executive function, are attributed to prefrontal D1 receptor hypoactivity (28,29).
Preclinical models have demonstrated that striatal D2 hyperactivity induces psychosis-like behaviors, while prefrontal D1 hypoactivity impairs decision-making and working memory (31). These studies bridge molecular dysfunctions with clinical manifestations, providing a translational framework for understanding the disorder (22).
Despite these pathways operating primarily independently, their interplay is critical for understanding schizophrenia’s complex symptomatology. For example, dopaminergic disruptions in one region can indirectly influence activity in others, creating a cascade of dysfunctions that collectively contribute to the disorder (20).
Antipsychotic Treatments and Emerging Therapies
Antipsychotics have been the cornerstone of schizophrenia treatment for decades, primarily targeting dopamine pathways. First-generation
antipsychotics (FGAs), such as haloperidol, achieve their effects through robust D2 receptor antagonism. While effective for positive symptoms, these medications are associated with significant side effects, including extrapyramidal symptoms and tardive dyskinesia, limiting their long-term utility (27, 32, 33). Second-generation antipsychotics (SGAs), such as clozapine and aripiprazole, incorporate serotonin receptor modulation to demonstrate enhanced efficacy. Clozapine, in particular, is regarded as the gold standard for treatment-resistant schizophrenia (20). However, its mechanism of action remains incompletely understood and may involve glutamatergic or immune pathways alongside dopamine antagonism (22–24).
Emerging treatments seek to refine the therapeutic approach to dopamine dysregulation. Partial D2 agonists aim to balance receptor activity without inducing complete antagonism, potentially reducing side effects (33). Similarly, D1 receptor modulators are being explored for their ability to address cognitive deficits, a domain poorly served by existing therapies (28, 29). A shift toward precision medicine can be seen as treatments are tailored to address the full spectrum of schizophrenia symptoms while minimizing adverse effects.
Limitations of the Dopamine Hypothesis
The dopamine hypothesis explains many core features of schizophrenia but fails to account for the disorder’s full complexity. For instance, cognitive deficits and negative symptoms often persist despite dopamine activity being normalized (21). This suggests the involvement of more or potentially different neurotransmitter systems, such as glutamate and gamma-aminobutyric acid (GABA). The relationship between dopamine and serotonin is also particularly relevant, forming the basis for the efficacy of atypical antipsychotics. This interaction underscores the necessity of integrated models that encompass multiple neurotransmitter systems. Future research must expand out of dopamine-centric paradigms to fully capture schizophrenia’s neurobiology and develop more comprehensive treatment approaches (21).
The Role of Serotonin in Schizophrenia
Understanding Serotonin’s Role in Schizophrenia
Although dopamine has long been at the center of schizophrenia research, serotonin is
increasingly recognized as a key factor, especially in understanding negative symptoms and cognitive challenges (10, 12, 35).
Studies have shown that serotonin imbalances in areas like the prefrontal cortex and hippocampus can lead to symptoms such as emotional withdrawal, apathy, and cognitive difficulties, issues that are particularly difficult to treat (13, 28, 36). Research also points to variations in serotonin receptors (e.g., 5-HT2A) and transporters (e.g., solute carrier family 6 member 4, also known as SLC6A4) as influential factors in the development of schizophrenia and individual responses to treatment (13,28,36).
Additionally, environmental factors like prenatal adversity or early-life trauma may disrupt serotonin pathways during critical stages of brain development, increasing vulnerability to the disorder (20, 36). The importance of serotonin became more apparent with the development of SGAs which antagonize both dopamine and serotonin receptors. This dual action has provided broader symptom relief and highlighted serotonin’s important role in the disorder (10, 12, 35).
Serotonin Receptors: Key Players in Schizophrenia
Serotonin plays a vital role in the brain, influencing mood, cognition, and emotional balance. In schizophrenia, serotonin’s effects are largely mediated by specific receptors that affect how the brain regulates critical neurotransmitters like dopamine and glutamate. Understanding these mechanisms has led to advancements in treating the disorder, particularly with medications targeting serotonin receptors. The 5-HT2A receptor is perhaps the most well-studied in schizophrenia. Found abundantly in the prefrontal cortex, this receptor modulates dopamine and glutamate activity, which are crucial for cognitive and emotional balance. When overactivated, it can contribute to hallucinations, delusions, and cognitive disturbances. Medications like risperidone and olanzapine, commonly used second-generation antipsychotics, work by blocking 5-HT2A receptors, reducing these symptoms while enhancing dopamine function in key brain regions (37). Clozapine, given for treatment-resistant schizophrenia, also blocks 5-HT2A receptors but has the added benefit of partially activating 5-Hydroxytryptamine Receptor 1A (5HT1A), which are known to regulate mood, anxiety, and emotional processing. This unique combination makes clozapine especially effective for severe cases and patients with suicidal ideation (38, 39).
Another key serotonin receptor is 5-HT1A, which influences the release of serotonin itself and plays a significant role in mood and emotional regulation. Activating this receptor has shown promise in alleviating negative symptoms of schizophrenia, such as apathy and emotional withdrawal. Aripiprazole, a third-generation antipsychotic, combines partial activation of 5HT1A receptors with 5HT2A receptor antagonism. This dual action helps manage symptoms effectively while minimizing side effects like motor rigidity (39).
The 5-hydroxytryptamine receptor 2C (5-HT2C) also contributes to the broader understanding of serotonin’s role in schizophrenia. Located in the limbic system, it helps regulate dopamine in reward pathways. However, its activity is linked to metabolic side effects, such as weight gain, often seen with medications like olanzapine (40). Meanwhile, emerging research on 5-hydroxytryptamine receptor 6 (5HT6) and 5-hydroxtryptamine receptor 7 (5-HT7) highlight their involvement in cognition and memory. These receptors have become promising targets for experimental therapies aimed at improving cognitive deficits in schizophrenia, with drugs like pimavanserin showing potential in early studies (41, 42). Lastly, serotonin’s role extends beyond the brain. Most serotonin in the body exists in the peripheral system, particularly in the gut. Receptors like 5-hydroxytryptamine receptor 3 (5-HT3) and 5-hydroxytryptamine receptor 4 (5-HT4), while not directly related to schizophrenia symptoms, play a critical role in managing the side effects of antipsychotic medications, such as nausea and constipation. Additionally, emerging research into the gut-brain axis suggests that peripheral serotonin may indirectly influence mental health, opening new avenues for holistic treatment approaches (40, 41).
By targeting serotonin receptors, modern antipsychotics have significantly improved the lives of patients with schizophrenia, addressing both core symptoms and the side effects of treatment. Ongoing research continues to expand our understanding of serotonin’s complex role, paving the way for more effective and personalized therapies.
Medications Targeting Serotonin Receptors
Antipsychotic medications play a critical role in managing schizophrenia by targeting serotonin receptors to improve symptoms and minimize side effects. Medications like risperidone and olanzapine
block 5-HT2A receptors, which not only help reduce hallucinations and delusions but also enhance dopamine activity in the prefrontal cortex, improving motivation and emotional withdrawal (37, 40). For individuals with treatment-resistant schizophrenia, clozapine is often the go-to medication. Its success lies in its ability to block 5-HT2A receptors while also partially activating 5-HT1A receptors, making it particularly effective for severe cases, especially those involving suicidal thoughts (38, 39, 44).
Aripiprazole, a newer antipsychotic, takes a slightly different approach. It works by partially activating both D2 and 5-HT1A receptors, while simultaneously blocking 5-HT2A receptors. This unique mechanism allows it to treat symptoms effectively while reducing side effects like muscle stiffness, which can occur with older antipsychotics (39, 45).
Similarly, newer medications like brexpiprazole and cariprazine offer a balanced approach by targeting both dopamine and serotonin systems, allowing them to manage both the positive symptoms (like hallucinations) and the negative symptoms (like apathy and social withdrawal) of schizophrenia (40).
Another medication, pimavanserin, was initially developed for treating psychosis in Parkinson’s disease, however, it is now being explored for its ability to reduce hallucinations in schizophrenia via an antagonist effect on 5-HT2A receptors. Early research shows promise for its use in this area (41).
Schizophrenia represents a significant challenge both for individuals living with it and for healthcare systems globally. The burden of the disease highlights the need for innovative, effective treatment strategies that address core symptoms while improving overall quality of life (46). As we continue to learn more about how serotonin receptors influence the brain, the development of targeted therapies offers hope for improving outcomes and restoring balance for those affected by this complex condition.
Beyond the Brain: Serotonin’s Broader Impact
Interestingly, the majority of the body’s serotonin is not in the brain, it is in the gut and blood platelets. Among the serotonin receptors found outside the brain, 5-HT3 and 5-HT4 stand out for their unique roles in schizophrenia treatment. While these receptors do not directly affect core symptoms like hallucinations or delusions, they play a significant role in managing side effects of antipsychotic medications. For example,
5-HT3 receptors help regulate gut motility and control nausea, making them key players in addressing nausea caused by antipsychotic drugs. Similarly, 5-HT4 receptors are involved in improving gut movement, which can help relieve constipation, a common and uncomfortable side effect of many schizophrenia treatments (47). But their role may go beyond just easing side effects. Emerging research suggests that these receptors could indirectly contribute to managing schizophrenia itself. For example, 5-HT4 receptors have been linked to better cognitive flexibility, which means they could help address cognitive difficulties often experienced by people with schizophrenia. Meanwhile, blocking 5-HT3 receptors may influence dopamine pathways in a way that complements existing treatments (47).
Even more intriguing is the growing research on the gut-brain axis, which highlights how the gut and brain are deeply connected. Serotonin produced in the gut and metabolites from the gut microbiome, like short-chain fatty acids (SCFAs), might influence brain function and behavior. This connection has sparked interest in exploring how targeting gut serotonin, especially through receptors like 5-HT3 and 5-HT4, could help not only with side effects but also with improving the overall mental health of people living with schizophrenia (47).
These findings open the door to a more holistic approach to treating schizophrenia. By addressing both the brain and the gut, we are moving toward therapies that tackle core symptoms while improving overall quality of life. It is a promising step forward for those living with this complex condition (40,46).
Challenges and Opportunities in Targeting Serotonin Receptors for Schizophrenia
Serotonin receptors have a lot of potential for treating schizophrenia, but there are also particular challenges. One major obstacle is the widespread presence of serotonin receptors throughout various parts of the brain. This makes it difficult to precisely target them without affecting other systems, leading to unintended side effects such as weight gain or metabolic issues. Additionally, serotonin does not work in isolation, it interacts closely with dopamine and glutamate systems, further complicating treatment strategies. These complexities highlight the need for a more integrated approach to managing schizophrenia (47, 48).
Looking to the future, research is shifting toward the development of more selective treatments, such as serotonin-dopamine modulators. These therapies aim to address the full range of schizophrenia symptoms while minimizing side effects (48).
Challenges in Serotonin Modulation
One challenge in serotonin-focused therapies is finding the right balance between treating symptoms and avoiding unwanted side effects. For example, a recent case report highlighted a rare side effect in a patient taking cariprazine, a medication that works as a partial agonist at 5-HT1A receptors and modulates 5-HT7 receptors. While cariprazine helped improve the patient’s cognitive and psychotic symptoms, it also caused hypersexual behavior, a rare and disruptive side effect. This behavior was likely tied to the medication’s complex interactions with serotonin receptors and its influence on dopamine pathways related to reward processing. Once the medication was discontinued, the hypersexuality resolved. This case underscores the importance of monitoring patients carefully and highlights the delicate balance needed in developing serotonin-targeted treatments (49).
Opportunities in Serotonin-Dopamine Modulation
Another promising direction involves combining serotonin and dopamine modulation. A case report involving a patient with treatment-resistant schizophrenia and dopamine supersensitivity psychosis (DSP) sheds light on this approach. After prolonged use of high-dose antipsychotics, the patient was switched to asenapine, a medication with strong 5-HT2A receptor antagonism (50). This switch led to significant improvements in psychotic symptoms such as hallucinations and delusions, while also enhancing cognitive functioning. Unlike other antipsychotics, asenapine’s serotonergic properties helped restore dopamine balance without worsening DSP. This case demonstrates the potential of serotonin-dopamine activity modulators to address even the most difficultto-treat forms of schizophrenia while reducing dopamine-related side effects (50).
Adjunctive Serotonin Therapies for Negative Symptoms
Treating the negative symptoms of schizophrenia, such as apathy, emotional withdrawal, and lack of motivation, is still a challenge. These symptoms often do not respond well to standard dopamine-focused treatments. A narrative review discussed the use of serotonergic strategies to target these symptoms,
emphasizing the potential of combining serotoninfocused therapies with antipsychotics for better outcomes. For instance, SSRIs like escitalopram have shown modest but meaningful improvements in negative symptoms. These treatments offer clinicians a safe and effective way to enhance serotonergic tone while avoiding other side effects like sedation or metabolic issues (51).
Interactions Between Dopamine and Serotonin Systems
The interaction between DA and 5-HT plays a critical role in regulating mood, cognition, and behavior, making it highly relevant to understanding and treating schizophrenia. These two neuromodulators influence each other through intricate pathways. DA, originating from the ventral tegmental area (VTA) and substantia nigra pars compacta (SNc), interacts with 5-HT, which arises from the dorsal and median raphe nuclei (DRN and MRN). These systems maintain a bidirectional relationship: dopamine regulates serotonin signaling in the raphe nuclei, while serotonin projections to dopamine-rich regions like the SNc and VTA modulate dopamine activity (52,53,55,59).
Dopamine exerts its effects primarily through D1 and D2 receptor families, which are crucial for regulating motor, emotional, and behavioral functions. For instance, D2 receptors in the striatum suppress neuronal excitability to regulate motor control, and their dysfunction can lead to motor impairments and emotional disturbances (57,58). Similarly, serotonin receptors, organized into families like 5-HT1, 5-HT2, and 5-HT3, are key to signaling processes that influence mood and cognition. Among them, the 5-HT2A receptor plays a pivotal role in modulating dopamine activity, underscoring the intricate relationship between these neurotransmitters in schizophrenia (52,54,55,58).
Impact on Treatment Strategies
Understanding the interplay between dopamine and serotonin systems is essential for developing effective treatments for schizophrenia. Dysregulation in these systems contributes to both positive symptoms (like hallucinations, often linked to mesolimbic dopamine hyperactivity) and negative symptoms (such as social withdrawal, tied to mesocortical dopamine hypofunction) (60, 65). Serotonin, particularly through 5-HT2A receptors, helps modulate dopamine release, making it a valuable target for therapies that address the wide range of schizophrenia symptoms (52, 56).
Clinical Implications: Treatments Focusing on DopamineSerotonin Interactions
Modern antipsychotic treatments rely heavily on the dual modulation of dopamine and serotonin systems. SGAs like clozapine and olanzapine effectively manage both positive and negative symptoms by targeting both neurotransmitter systems. These medications partially block dopamine receptors while antagonizing 5-HT2A receptors, which enhances mood and cognitive function (56, 61, 62). For example, clozapine’s low affinity for D2 receptors reduces the risk of motor side effects, while its potent antagonism of 5-HT2A receptors improves negative symptoms and cognitive deficits by increasing dopamine release in specific pathways (63, 64).
Olanzapine and other atypical antipsychotics also target both systems, addressing the limitations of earlier, dopamine-centric therapies. By balancing dopamine and serotonin modulation, these treatments are particularly effective for individuals with treatment-resistant schizophrenia, providing relief for a population that often struggles to find effective options (56, 62).
Future Direction in Schizophrenia Research
Future Research Related to Serotonin and Dopamine in Schizophrenia
Research into the serotonin and dopamine systems continues to be pivotal for understanding and treating schizophrenia. A major challenge with current treatments targeting serotonin (e.g., 5-HT2A) and dopamine (e.g., D2) pathways lies in their limited efficacy for negative and cognitive symptoms, alongside notable off-target effects such as metabolic disturbances and movement disorders (60, 66, 67). Future research aims to refine selective serotonindopamine modulators (SDAMs) like brexpiprazole and cariprazine. These drugs leverage partial agonism and antagonism across receptors such as 5-HT1A, 5-HT2A, and D3, potentially enhancing efficacy for negative and cognitive symptoms while minimizing side effects (68, 69). Additionally, emerging studies are exploring the role of 5-HT7 receptors in regulating circadian rhythms and cognitive processes, with a focus on dual-action drugs that target both 5-HT7 and dopamine receptors (70, 71).
Negative symptoms such as apathy, anhedonia, and emotional blunting remain inadequately addressed by existing antipsychotics. These symptoms are linked to
dopamine hypofunction in the prefrontal cortex and serotonin dysregulation in limbic regions. Research is therefore directed toward therapies combining D1 receptor agonism to enhance prefrontal dopamine signaling with 5-HT2A antagonism to modulate limbic serotonin. Such an approach may restore motivation and improve reward processing (72, 73). Meanwhile, the gut-brain axis and its influence on peripheral serotonin signaling represent an untapped frontier in schizophrenia research. With 90% of serotonin located in the periphery, particularly in the gut, understanding its impact on inflammation, neuroplasticity, and neurotransmitter balance could open pathways for gut- targeted treatments like probiotics or peripheral serotonin inhibitors (48, 74).
Schizophrenia’s complexity also lies in its intricate interplay among dopamine, serotonin, and glutamate systems. Glutamate hypofunction in N-methyl-Daspartate (NMDA) receptor pathways exacerbates serotonin and dopamine imbalances. Future therapeutic strategies may include multi-modal drugs targeting glutamate (via NMDA receptor modulators), serotonin (e.g., 5-HT1A/5-HT2A), and dopamine (e.g., D1/D3) to address a broader spectrum of symptoms (75, 76). Additionally, the heterogeneity of receptor sensitivity in patients poses challenges for antipsychotic efficacy. Research into genetic and epigenetic biomarkers, such as polymorphisms in 5-HT2A and DRD2 genes, could inform personalized treatment strategies. Pharmacogenomic advancements may refine drug selection and dosing, reducing adverse effects while improving outcomes (77–79).
Future Research Beyond Serotonin and Dopamine
While 5-HT and DA remain central to schizophrenia research, scientists are increasingly exploring other pathways that may play a crucial roles in the disorder. One major area of focus is the glutamate system, particularly the NMDA receptor, which has been linked to cognitive deficits and negative symptoms. Since traditional antipsychotics do not target this pathway, researchers are developing drugs that enhance NMDA receptor activity. Examples include glycine reuptake inhibitors and D-serine analogs, which act as coagonists to boost cognitive processing (80–82).
Another promising avenue is the immune system, as evidence suggests that schizophrenia may involve chronic low-grade inflammation and immune dysregulation. Anti-inflammatory therapies and
immune-modulating agents, such as monoclonal antibodies targeting cytokines like interleukin-6 (IL-6) or tumor necrosis factor alpha (TNF-α), could help reduce neuroinflammation and improve both cognitive and negative symptoms (83, 84).
Dysfunction in the GABAergic system, which disrupts cortical inhibition, is also believed to contribute to cognitive deficits and sensory gating abnormalities in schizophrenia. Future treatments may focus on drugs that target GABAA receptors, such as positive allosteric modulators, to strengthen inhibitory signaling without causing excessive sedation (87, 88). Social dysfunction, a hallmark of schizophrenia, is still an area of unmet need. Therapies based on oxytocin, a hormone that influences empathy and trust, are being studied for their potential to enhance social bonding and reduce social anxiety. Intranasal oxytocin, in particular, is already being tested in clinical trials and shows promise for improving social interactions (87, 88).
The endocannabinoid system is another emerging target, as its dysregulation has been associated with psychosis and cognitive impairments. Research into CB1 receptor antagonists and CB2 receptor agonists aims to reduce psychotic symptoms and inflammation while preserving cognitive function. Cannabidiol (CBD) has drawn particular interest for its neuroprotective and anti-inflammatory properties (89, 90).
Disrupted neural oscillations, especially in gamma and theta frequencies, are linked to impaired cognition and sensory integration in schizophrenia. Neuromodulation therapies, such as transcranial magnetic stimulation (TMS) and direct current stimulation, are being explored to restore normal oscillatory activity. These treatments may also be combined with pharmacological agents to enhance their effects (91, 92).
Finally, environmental risk factors like prenatal stress and childhood trauma may increase susceptibility to schizophrenia through epigenetic modifications. Addressing these changes with treatments such as histone deacetylase inhibitors (HDACi) or microRNA modulators offers the potential to reverse abnormal gene expression and restore neural plasticity (93, 94).
Conclusions
DA and 5-HT are major factors in the onset and management of schizophrenia. Alterations in dopamine activity in various brain regions contribute to the
explanation of the disorder's diverse symptoms, which include motivational and decision-making problems as well as hallucinations. Serotonin, on the other hand, contributes to the emotional and cognitive aspects of schizophrenia, often overlapping with dopamine in its effects. Treatments that target both systems, such as second-generation antipsychotics, have improved symptom management for many patients, but negative symptoms like emotional withdrawal and cognitive deficits are still a major challenge.
The growing focus on emerging therapies, including drugs that modulate both serotonin and dopamine, NMDA receptor-targeting agents, and approaches that explore the gut-brain connection, offers exciting opportunities for more effective treatments. Moving forward, a personalized approach that tailor’s treatment to individual needs, based on both biological and genetic insights, could pave the way for improvements in care. By broadening the focus beyond dopamine alone and toward a more holistic understanding of schizophrenia, we can improve not just the clinical outcomes but also the quality of life of individuals living with schizophrenia.
Disclosures
The authors declare no conflicts of interest related to this work. This review was conducted as part of the academic requirements for the Master of Biomedical Sciences program at Geisinger College of Health Sciences. No external funding was received.
Acknowledgments
The authors would like to thank Brian J. Piper, Ph.D. for his mentorship, guidance, and continued support throughout the development of this work. We also acknowledge the Master of Biomedical Sciences program and the Geisinger College of Health Sciences for providing the academic resources and environment that supported the completion of this project.
References
1. Rahman T, Lauriello J. Schizophrenia: An Overview. Focus. 2016;14(3):300–7.
2. Tandon R, Gaebel W, Barch DM, Bustillo J, Gur RE, Heckers S, et al. Definition and description of schizophrenia in the DSM-5. Schizophr Res 2013;150(1):3-10.
3. Ruiz-Castaneda P, Molina ES, Loaiza HA, Gonzales MT. Positive symptoms of schizophrenia and their relationship with cognitive and emotional executive functions. Cogn Res Princ Implic. 2022;7:78.
4. Correll CU, Schooler NR. Negative symptoms in schizophrenia: A review and clinical guide for recognition, assessment, and treatment. Neuropsychiatr Dis Treat. 2020;16:519-534.
5. National Alliance on Mental Illness. Schizophrenia [Internet]. NAMI;2024 [cited 2024 Nov 20]. Available from: https://www.nami.org/aboutmental-illness/mental- health-conditions/ schizophrenia/.
6. Liu R, Fang X, Yu L, Wang D, Wu Z, Guo C, et al. Gender differences of schizophrenia patients with and without depressive symptoms in clinical charecteristics. Front Psychiatry. 2022; 12:792019.
7. Maric NP, Jovicic MJ, Mihaljevic M, Miljevic C. Improving current treatments for schizophrenia. Drug Dev Res. 2016;77:357-367.
8. U.S. Food and Drug Administration. FDA approves drug with new mechanism of action for treatment of schizophrenia [Internet]. Silver Spring (MD): FDA; 2024 Sep 26 [cited 2025 Jun 1]. Available from: https://www.fda.gov/news-events/ press-announcements/fda-approves-drug-newmechanism-action-treatment-schizophrenia
9. Yang AC, Tsai SJ. New targets for schizophrenia treatment beyond the dopamine hypothesis. Int J Mol Sci. 2017;18(8).
10. Brisch R, Saniotis A, Wolf R, Bielau H, Bernstein HG, Steiner J, et al. The role of dopamine in schizophrenia from a neurobiological and evolutionary perspective: Old fashioned, but still in vogue. Front Psych. 2014;5:47.
11. Gerson SC, Baldessarini RJ. Motor effects of serotonin in the central nervous system. Life Sci 1980;27(16):1435-1451.
12. Bleich A, Brown SL, Kahn R, van Praag HM. The role of serotonin in schizophrenia. Schizophr Bull 1988;14(2):297-315.
13. Shah UH, Gonzales-Maeso J. Serotonin and glutamate interactions in preclinical schizophrenia models. ACS Chem Neurosci. 2019;10(7):30683077.
14. Linszen D, Lenior M, De Haan L, Dingemans P, Person’s B. Early intervention, untreated psychosis and the course of early schizophrenia. Br J Psychiatry Suppl. 1998;172(33):84-9.
15. Van de Leemput J, Hess JL, Glatt SJ, Tsuang MT. Chapter three – genetics of schizophrenia: Historical insights and prevailing evidence. Adv in Gen. 2016;96:99- 141.
16. Lally J, Maloudi S, Krivoy A, Murphy K. A forgotten diagnosis in psychiatry. J Nerv Men Dis 2019;207(9):721-725.
17. Sjoerdsma A. Serotonin. N Engl J Med. 1959;261:231-237.
18. Ban TA. Neuropsychopharmacology and the genetics of schizophrenia: A history of the diagnosis of schizophrenia. Prog Neuropsychopharmacol Biol Psych. 2004;28(5):75362
19. Jablensky A, McNeil TF, Morgan VA. Barbara Fish and a short history of the neurodevelopmental hypothesis of schizophrenia. Schizophr Bull. 2017;43(6):1158- 1163.
20. McCutcheon RA, Krystal JH, Howes OD. Dopamine and glutamate in schizophrenia: Biology, symptoms and treatment. World Psychiatry. 2020;19(1):15– 33.
21. Howes OD, Kapur S. The dopamine hypothesis of schizophrenia: Version III - The final common pathway. Schizophr Bull. 2009;35(3):549–62.
22. Howes O, Egerton A, Allan V, McGuire P, Stokes P, Kapur S. Mechanisms underlying psychosis and antipsychotic treatment response in schizophrenia: Insights from PET and SPECT imaging. Curr Pharm Des.2009;15(22):2550–9.
23. Howes OD, McCutcheon R, Owen MJ, Murray RM. The role of genes, stress, and dopamine in the development of schizophrenia. Biol Psychiatry 2017;81(1):9–20.
24. Kegeles LS, Abi-Dargham A, Frankle WG, Gil R, Cooper TB, Slifstein M, et al. Increased synaptic dopamine function in associative regions of the striatum in schizophrenia. Arch Gen Psychiatry 2010;67(3):231.
25. Harrison PJ. Postmortem studies in schizophrenia. Dialogues Clin Neurosci. 2000;2(4):349–57.
26. Martel JC, Gatti McArthur S. Dopamine receptor subtypes, physiology and pharmacology: New ligands and concepts in schizophrenia. Front Pharmacol. 2020;11:1003.
27. Sakurai H, Bies RR, Stroup ST, Keefe RSE, Rajji TK, Suzuki T, et al. Dopamine D2 receptor occupancy and cognition in schizophrenia: Analysis of the CATIE data. Schizophr Bull. 2013;39(3):564–74.
28. Goldman-Rakic P. The relevance of the dopamine-D1 receptor in the cognitive symptoms of schizophrenia. Neuropsychopharmacol. 1999;21(6):S170–80.
29. Abi-Dargham A, Javitch JA, Slifstein M, Anticevic A, Calkins ME, Cho YT, et al. Dopamine D1R receptor stimulation as a mechanistic procognitive target for schizophrenia. Schizophr Bull. 2022;48(1):199–210.
30. Kesby J, Eyles D, McGrath J, Scott J. Dopamine, psychosis, and schizophrenia: the widening gap between basic and clinical neuroscience. Transl Psychiatry. 2018;8(1):30.
31. Grace AA. Dysregulation of the dopamine system in the pathophysiology of schizophrenia and depression. Nat Rev Neurosci. 2016;17(8):524–32.
32. Girgis RR, Phillips MR, Li X, Li K, Jiang H, Wu C, et al. Clozapine v. chlorpromazine in treatmentnaive, first-episode schizophrenia: 9-year outcomes of a randomised clinical trial. Br J Psychiatry. 2011;199(4):281–8.
33. Jones HM, Pilowsky LS. Dopamine and antipsychotic drug action revisited. Br J Psychiatry. 2002;181(4):271–5.
34. Remington G, Addington D, Honer W, Ismail Z, Raedler T, Teehan M. Guidelines for the pharmacotherapy of schizophrenia in adults. Can J Psychiatry. 2017;62(9):604- 16.
35. Meltzer HY. Serotonin and schizophrenia: From a hypothesis to the development of new drugs. CNS Spectr. 1999;4(4):12–22.
36. Abi-Dargham A, Laruelle M. The role of serotonin in schizophrenia. Eur Psychiatry. 2005;20(5–6):369–75.
37. Hieronymus F, Correll CU, Østergaard SD. Initial severity of the Positive and Negative Syndrome Scale (PANSS)-30, its main subscales plus the PANSS-6, and the relationship to subsequent
improvement and trial dropout: A pooled participant-level analysis of 18 placebo-controlled risperidone and paliperidone trials. Transl Psychiatry. 2023;13(1):191.
38. Celada P, Bortolozzi A, Artigas F. Serotonin 5-HT1A receptors as targets for agents to treat psychiatric disorders: Rationale and current status of research. CNS Drugs. 2013;27(9):703–716.
39. Haleem DJ. 5-HT1A receptor-dependent control of nigrostriatal dopamine neurotransmission in the pharmacotherapy of Parkinson's disease and schizophrenia. Behav Pharmacol. 2015;26(12):45–58.
40. Monahan C, McCoy L, Powell J, Gums JG. Olanzapine/Samidorphan: New drug approved for treating bipolar I disorder and schizophrenia. Ann Pharmacother. 2022;56(9):1049–1057.
41. Halberstadt AL, Sindhunata IS, Scheffers K, Flynn AD, Sharp RF, Geyer MA, et al. Effect of 5-HT2A and 5-HT2C receptors on temporal discrimination by mice. Neuropharmacol. 2016;107:364–375.
42. Meneses A. Memory formation and memory alterations: 5-HT6 and 5-HT7 receptors, novel alternative. Rev Neurosci. 2014;25(3):325–356
43. Miranda L. Antidepressant and anxiolytic effects of activating 5HT2A receptors in the anterior cingulate cortex and the theoretical mechanisms underlying them - a scoping review of available literature. Brain Res.2024;1846:149226.
44. Tandon R, Gaebel W, Barch DM, Bustillo J, Gur RE, Heckers S, et al. Definition and description of schizophrenia in the DSM-5. Schizophr Res. 2013;150(1):3–10.
45. Correll CU, Kim E, Sliwa JK, Hamm W, Gopal S, Mathews M, et al. Pharmacokinetic characteristics of long-acting injectable antipsychotics for schizophrenia: An overview. CNS Drugs. 2021;35(1):39–59.
46. Solmi M, Seitidis G, Mavridis D, Correll CU, Dragioti E, Guimond S, et al. Incidence, prevalence, and global burden of schizophrenia - data, with critical appraisal, from the Global Burden of Disease (GBD) 2019. Mol Psychiatry. 2023;28(12):5319–5327.
47. Shimizu S, Mizuguchi Y, Ohno Y. Improving the treatment of schizophrenia: Role of
5-HT receptors in modulating cognitive and extrapyramidal motor functions. CNS Neurol Disord Drug Targets. 2013;12(6):861–869.
48. Ju S, Shin Y, Han S, Kwon J, Choi TG, Kang I, et al. The gut-brain axis in schizophrenia: The implications of the gut microbiome and SCFA production. Nutrients. 2023;15(20):4391.
49. Prelog PR, Palandacic AK. Hypersexuality during treatment with cariprazine in a patient with schizophrenia? BMC Psychiatry. 2023; 23:935.
50. Takao N, Murai T, Fujiwara H. Treatmentresistant schizophrenia characterised by dopamine supersensitivity psychosis and efficacy of asenapine. BMJ Case Reports CP. 2021 1;14(4):e242495.
51. Căpățînă OO, Micluția IV, Fadgyas-Stănculete M. Current perspectives in treating negative symptoms of schizophrenia: A narrative review. Exp Ther Med. 2021;21(3):1– 1.
52. De Deurwaerdère P, Chagraoui A, Di Giovanni G. Serotonin/dopamine interaction: Electrophysiological and neurochemical evidence. Prog Brain Res. 2021;261:161– 264.
53. Ogawa SK, Watabe-Uchida M. Organization of dopamine and serotonin system: Anatomical and functional mapping of monosynaptic inputs using rabies virus. Pharmacol Biochem Behav 2018;174:9–22.
54. Lyon KA, Rood BD, Wu L, Senft RA, Goodrich LV, Dymecki SM. Sex-specific role for dopamine receptor D2 in dorsal raphe serotonergic neuron modulation of defensive acoustic startle and dominance behavior. ENeuro. 2020;1;7(6).
55. Ogawa SK, Cohen JY, Hwang D, Uchida N, Watabe-Uchida M. Organization of monosynaptic inputs to the serotonin and dopamine neuromodulatory systems. Cell Rep. 2014;8(4):1105–18.
56. Kapur S, Remington G. Serotonin-dopamine interaction and its relevance to schizophrenia. Am J Psychiatry.1996;153(4):466–76.
57. Chen R, Ferris MJ, Wang S. Dopamine D2 autoreceptor interactome: Targeting the receptor complex as a strategy for treatment of substance use disorder. Pharmacol Ther. 2020;213:107583.
58. Wong PH, Feng H, Teo WL. Interaction of the dopaminergic and serotonergic systems in the rat striatum: Effects of selective antagonists and uptake inhibitors. Neurosci Res. 1995;23(1):115–9.
59. Niederkofler V , Asher TE, Dymecki SM. Functional interplay between dopaminergic and serotonergic neuronal systems during development and adulthood. ACS Chem Neurosci. 2015;6(7):1055-70.
60. Weinstein JJ, Chohan MO, Slifstein M, Kegeles LS, Moore H, Abi-Dargham A. Pathway-specific dopamine abnormalities in schizophrenia. Biol Psychiatry. 2017;81(1):31-42.
61. Daniels MV, Ramlall S. Clozapine monitoring at a specialised psychiatric hospital: A retrospective chart review. S Afr J Psychiatry. 2023;29(1).
62. Meltzer HY, Matsubara S, Lee JC. Classification of typical and atypical antipsychotic drugs on the basis ofdopamine D-1, D-2 and serotonin2 pKi values. JPET. 1989;251(1):238-46.
63. Tibrewal P, Nair PC, Gregory KJ, Langmead CJ, Chan SK, Bastiampillai T. Does clozapine treat antipsychotic-induced behavioural supersensitivity through glutamate modulation within the striatum?. Mol Psychiatry.2023;28(5):1839-42.
64. Bueno-Carrasco MT, Cuéllar J, Flydal MI, Santiago C, Kråkenes TA, Kleppe R, et al. Structural mechanism for tyrosine hydroxylase inhibition by dopamine and reactivation by Ser40 phosphorylation. Nat Commun. 2022;13(1):74.
65. Sahu A, Gopalakrishnan L, Gaur N, Chatterjee O, Mol P, Modi PK, et al. The 5-Hydroxytryptamine signaling map: An overview of serotonin-serotonin receptor mediated signaling network. JCCS 2018;12:731-5.
66. Kapur S, Remington G. Serotonin-dopamine interaction and its relevance to schizophrenia. Am J Psychiatry.1996;153(4):466–76.
67. Meltzer HY, Matsubara S, Lee JC. Classification of typical and atypical antipsychotic drugs on the basis of dopamine D-1, D-2 and serotonin2 pKi values. J Pharmacol Exp Ther. 1989;251(1):238–46.
68. Chen R, Ferris MJ, Wang S. Dopamine D2 autoreceptor interactome: Targeting the receptor complex as a strategy for treatment of substance use disorder. Pharmacol Ther. 2020;213:107583.
69. Halberstadt AL, Sindhunata IS, Scheffers K, Flynn AD, Sharp RF, Geyer MA, et al. Effect of 5-HT2A and 5-HT2C receptors on temporal discrimination by mice. Neuropharmacology. 2016;107:364– 375.
70. Meneses A. Memory formation and memory alterations: 5-HT6 and 5-HT7 receptors, novel alternative. Rev Neurosci. 2014;25(3):325–356.
71. Ogawa SK, Cohen JY, Hwang D, Uchida N, Watabe-Uchida M. Organization of monosynaptic inputs to the serotonin and dopamine neuromodulatory systems. Cell Rep. 2014;8(4):1105–18.
72. De Deurwaerdère P, Chagraoui A, Di Giovanni G. Serotonin/dopamine interaction: Electrophysiological and neurochemical evidence. Prog Brain Res. 2021;261:161– 264.
73. Sahu A, Gopalakrishnan L, Gaur N, Chatterjee O, Mol P, Modi PK, et al. The 5-Hydroxytryptamine signaling map: an overview of serotonin-serotonin receptor mediated signaling network. J Cell Commun Signal. 2018;12:731–5.
74. Shimizu S, Mizuguchi Y, Ohno Y. Improving the treatment of schizophrenia: Role of 5-HT receptors in modulating cognitive and extrapyramidal motor functions. CNS Neurol Disord Drug Targets. 2013;12(6):861–869.
75. Niederkofler V , Asher TE, Dymecki SM. Functional interplay between dopaminergic and serotonergic neuronal systems during development and adulthood. ACS Chem Neurosci 2015;6(7):1055–70.
76. Daniels MV, Ramlall S. Clozapine monitoring at a specialised psychiatric hospital: A retrospective chart review. S Afr J Psychiatry. 2023;29(1).
77. Tibrewal P, Nair PC, Gregory KJ, Langmead CJ, Wa Chan SK, Bastiampillai T. Does clozapine treat antipsychotic-induced behavioural supersensitivity through glutamate modulation within the striatum? Mol Psychiatry. 2023;28(5):1839–42.
78. Bueno-Carrasco MT, Cuéllar J, Flydal MI, Santiago C, Krakenes TA, Kleppe R, Lopez-Blanco JR, et al. Structural mechanism for tyrosine hydroxylase inhibition by dopamine and reactivation by Ser40 phosphorylation. Nat Commun. 2022;13(1):74
79. Lyon KA, Rood BD, Wu L, Senft RA, Goodrich LV, Dymecki SM. Sex-specific role for dopamine
receptor D2 in dorsal raphe serotonergic neuron modulation of defensive acoustic startle and dominance behavior. ENeuro. 2020;7(6).
80. Javitt DC, Zukin SR. Recent advances in the phencyclidine model of schizophrenia. Am J Psychiatry.1991;148(10):1301-1308.
81. Coyle JT, Balu DT. NMDA receptor function in schizophrenia: Integrating excitatory, inhibitory, and modulatory dimensions. Nat Rev Neurosci. 2012;14(1):43- 58.
82. Howes OD, McCutcheon RA, Stone JM. Glutamate and dopamine in schizophrenia: An update for the 21st century. J Psychopharmacol. 2015;29(2):97115.
83. Khandaker GM, Dantzer R, Jones PB. Immunopsychiatry: The importance of the immune system in psychiatry. World Psychiatry. 2015;14(1):77-89.
84. Müller N, Weidinger E, Leitner B, Schwarz MJ. The role of inflammation in schizophrenia. Front Neurosci.2015;9:372.
85. Lewis DA, Hashimoto T, Volk DW. Cortical inhibitory neurons and schizophrenia. Nat Rev Neurosci.2012;13(2):20-27.
86. Chung DW, Fish KN, Lewis DA. Pathological basis for deficient excitatory drive to cortical parvalbumin interneurons in schizophrenia. Am J Psychiatry. 2016;173(11):1131-1139.
87. Feifel D, Shilling PD, MacDonald K. A review of oxytocin's effects on the positive, negative, and cognitive domains of schizophrenia. Biol Psychiatry 2016;79(3):222- 233.
88. Meyer-Lindenberg A, Domes G, Kirsch P, Heinrichs M. Oxytocin and the social brain: Neural mechanisms and perspectives in human research. Nat Rev Neurosci. 2011;12(6):524-535.
89. Leweke FM, Mueller JK, Lange B, Rohleder C. Therapeutic potential of cannabinoids in psychosis. Biol Psychiatry. 2012;71(8):703-704.
90. Bossong MG, Niesink RJ. Adolescent brain maturation, the endocannabinoid system and the neurobiology of cannabis-induced schizophrenia. Prog Neurobiol. 2010;92(3):370-385.
91. Uhlhaas PJ, Singer W. Abnormal neural oscillations and synchrony in schizophrenia. Nat Rev Neurosci.2010;11(2):100-113.
92. Fitzgerald PB, Daskalakis ZJ. The use of repetitive transcranial magnetic stimulation and transcranial direct current stimulation in the treatment of schizophrenia. Curr Opin Psychiatry. 2019;32(2):163-171.
93. Tsankova NM, Renthal W, Kumar A, Nestler EJ. Epigenetic regulation in psychiatric disorders. Nat Rev Neurosci.2007;8(5):355-367.
94. Girdhar K, Hoffman GE, Jiang Y, Brown L, Kundakovic M, Hauberg ME, et al. Cell-specific histone modification maps in the human frontal lobe link schizophrenia risk to the neuronal epigenome. Nat Neurosci. 2018;21(8):1126-1136.
Group-Based Telehealth Music Therapy Intervention for Patients with Dementia: A Pilot Study
Khevna P. Joshi1†, Alysha D. Suley2, Nicole M. Hooper3, Brian J. Piper1,5, Mark V. White1, and Maya L. Lichtenstein4
1Geisinger College of Health Sciences, Scranton, PA 18509
2Allied Services Wilkes-Barre Hospice Center Geisinger, Wilkes-Barre, PA 18702
3Geisinger Medical Center, Danville, PA 17822
4Geisinger Memory and Cognition Program, Wilkes-Barre, PA 18702
5Center for Pharmacy Innovation and Outcomes, Danville, PA 17822
†Doctor of Medicine Program
Correspondence: khevnajoshi5@gmail.com
Abstract
Background: The behavioral and psychological symptoms in dementia (BPSD) make caregiving particularly challenging. In-person music therapy sessions with a professional have been shown to be helpful for BPSD and reducing caregiver burden. There are few music therapists practicing in rural settings, specifically for this population, so telehealth offers an alternative way to access their care. However, there have been limited studies demonstrating the efficacy of telehealth music therapy, and the authors are not aware of studies investigating group-based telehealth music therapy in dementia. This investigation seeks to explore the feasibility of group-based music therapy via telehealth for people living at home with dementia.
Methods: Our participants were community-based outpatients recruited from a memory clinic with mild to moderate dementia and had a willing caregiver to participate. We ran a single blinded block-randomized control pilot study. The patients were randomized one-to-one to receive either 6 weekly telehealth music therapy group sessions with a board-certified music therapist or a personalized music playlist on a platform of their choice. There were 12 participants total in each group, for a total of 24 participants, along with their caregivers. The participants were divided into 3 cohorts that took part in the study sequentially, with 8 participants per cohort and 4 participants per group (Music Therapy vs Music Playlist). Each cohort’s session lasted for 6 weeks. Our initial outcome measures were to compare scores at baseline and at 6 weeks on measures of the patient’s cognition (Mini Mental State Examination, or MMSE) and mood (Geriatric Depression Scale, or GDS); caregiver report of BPSD (Neuropsychiatric Inventory-Questionnaire:
Severity/Distress scales, or NPI-Q/S/D); caregiver strain (Modified Caregiver Strain Index, or MCSI); and caregiver depression (Patient Health Questionnaire-9, or PHQ-9). As the investigation progressed, we measured study attrition and qualified reasons for dropout.
Results: Throughout the investigation, there was a 67% dropout rate, equal between groups. Some of the most common reasons for dropout included unavailability of the participant due to a busy schedule with other appointments, loss to follow-up with nonresponse to calls, and deteriorating health/ increasing hospitalizations that impacted participant’s ability to actively participate in sessions. There was no appreciable trend from the pre-scores to the postscores in both the interventional and control groups.
Conclusion: Despite both interventions being relatively non-burdensome, dropout rate was high among both groups, which limits the ability to draw meaningful conclusions about potential benefits of group telehealth music therapy intervention vs music playlist. Our results may help researchers to design and implement a larger group-based telehealth music therapy study with lessons we are learning from this investigation, which will hopefully better guide treatment and management of dementia in the primary care setting.
Introduction
Dementia (including Alzheimer’s disease, Lewy body disease, frontotemporal dementia, and other related disorders) can lead to premature disability and death (1). It is costly financially and emotionally, for families affected as well as for society as a whole:
afflicted patients can no longer be in the workforce; their unpaid caregivers often have to reduce their working hours or leave their own jobs in order to provide care; patients with dementia have increased hospitalizations and their stays tend to be more protracted than those without dementia; and, finally, dementia leads to earlier placement in nursing home care. The most prevalent contributing factors to emergency room visits and earlier placement in nursing homes are the behavioral and psychological symptoms of dementia (BPSD) (2). BPSD can include apathy, aggression, delusions, hallucinations, agitation, aberrant motor behaviors, and other dysregulated emotional expressions. BPSD have also been shown to be associated with increased caregiver strain, although it is unclear exactly which symptoms are most linked or if a certain threshold number is required for maximal strain (3).
There are few approved pharmaceutical treatments for BPSD, and those that are available often carry undesirable side effects and have limited efficacy (4). A systematic review and meta-analysis showed that cognitive enhancement, assessment and management of BPSD, and caregiver education and support are strategies for delaying nursing home placement (2). Music therapy has also been shown to be an effective tool in ameliorating BPSD including agitation (5–9), depression and anxiety (10–12), and stabilizing cognitive function (12). For patients with dementia, cognition and BPSD are the most cited contributing factors to nursing home placement (2), which is more expensive than care at home and may lead to decreased quality of life for the individual with dementia. By reducing BPSD, patients may enjoy a better quality of life, and caregivers may experience reduction in stress, thus delaying nursing home placement.
With limited numbers of qualified music therapists in the country, access to music therapy can be challenging, especially for patients with dementia who live in more rural areas. With the recent COVID pandemic necessitating the expansion of telehealth, music therapists could potentially reach patients who reside in more isolated regions. There is a paucity of studies for group-based telehealth music therapy for patients in this population. However, a few studies have examined the efficacy of music therapy telemedicine interventions in other populations (such as veterans (13, 14)) and children with neurodevelopmental disorders (15)). Those
studies have concluded that there is potential for music therapy conducted via a remote platform to have positive impacts on these populations (13–15). These studies that were conducted in other patient populations involved a mixture of both individual and group telehealth interventions, depending on patient preference and music therapist.
This investigation seeks to explore the feasibility of group-based music therapy via telehealth for people living at home with mild to moderate dementia. The COVID-19 pandemic has provided a unique urgency to this question, but it is relevant even outside of the recent pandemic. Given the number of rural settings in the United States, telehealth is an attractive treatment option for those patients who would otherwise not be able to access care, and a group-based modality may allow for more access to care for those with limited resources and provide additional socialization which may also provide benefits for people living with dementia and their caregivers. A meta-analysis of studies that analyzed the impact of in-person individual and group music therapy sessions suggested that music may be a promising component of the antidote to mental illness (16). Furthermore, several studies have suggested that there is an immense advantage to group-based music therapy sessions over individual for those with mental illness, as the benefits of social connections and strengthened bonds between people may improve the impact that music therapy can have for this population (17, 18).
There is evidence that suggests that in addition to stabilizing electroencephalogram (EEG) rhythms for patients with neurological disorders, music has the capacity to revive networks in our brain that have become deactivated because it is built into our evolutionary blueprint (19). Specifically, studies have shown that music can “actively facilitate the recovery of movement” in patients with various movement disorders, that those with memory disorders (such as Alzheimer’s disease) are “more resilient to neurodegenerative influences when their neuronal memory traces are built through music,” and that music can “decrease seizure frequency, stop refractory status epilepticus, and decrease EEG spike frequency.” (19) With these specific neurological correlates of music reactivating brain networks, we believe that organized music therapy, delivered by a trained music therapist through telehealth in a group setting, can create real change.
We hypothesized that group-based music therapy intervention via a telehealth platform for patients with mild to moderate dementia would demonstrate improvement in participants' BPSD and reduction in caregiver burden as compared to passive listening of a personalized music playlist. We designed a pilot study to see if delivering this intervention would be feasible and if there was signal for improved outcomes in the music therapy arm. Our first specific aim was to assess the feasibility of a telehealth group-based music therapy intervention and determine the practicality of carrying out this project over a longer period with a larger participant sample size. For this aim, we calculated study attrition and qualified reasons for dropout. Our second specific aim was to compare pre- and post-scores on measures of cognition, BPSD, caregiver strain, and depression for participant and caregiver for those undergoing the telehealth music therapy intervention vs. those who receive a personalized music playlist.
Methods
Participants
We recruited a total of 24 participant/caregiver dyads for the investigation from a memory clinic. The memory clinic was Geisinger Memory and Cognition Therapy Services Wilkes-Barre, located in WilkesBarre, PA. This clinic is in a relatively rural area, where elderly patients would have limited access to in-person
music therapy. Many of the patients travel from afar to even receive in-person care at this memory clinic.
The participants recruited had mild to moderate dementia of any cause (as diagnosed by board-certified behavioral neurologist MLL) and lived at home with their caregivers, who were also consented into the study. Participants and caregivers were specifically asked if they thought that they would be able to commit to the full period of the study (6 weeks), after being provided with a complete explanation of the protocol. An inclusion criterion was that the participants necessarily needed to have access to high-speed internet to engage in the telehealth music therapy sessions. Participants were randomized by 3 blocks of 8: one-to-one telehealth music therapy vs music playlist (Figure 1).
This protocol was approved and overseen by the Institutional Review Board. Our IRB approval number is 2021-0409. The investigation was conducted in accordance with relevant guidelines and regulations of the IRB, including the Helsinki Declaration.
Interventions
The investigation was carried out with the 24 participant/caregiver dyads divided into 3 sequential cohorts with 8 participants in each cohort, 4 in the Music Therapy group and 4 in the Music Playlist group. Each cohort’s session lasted for 6 weeks.
Reason 1 (n = 2) –
2 (n = 3) –
Reason 3 (n = 3) –
Reason 1 (n = 1) –health deteriorating too much to participate, hospitalized Reason 2 (n = 3) –schedule too busy/ overwhelming ("not right now")
Reason 3 (n = 4) –lost to follow-up, did not respond to calls for music therapy or post-study
Figure 1. Flowchart of attrition rate for all 3 cohorts
A total of 12 participants were randomized for the music therapy intervention. The music therapy took place once a week for 6 weeks via telemedicine, with the first session reserved for an informal music-based assessment by the board-certified music therapist (AS) to determine whether the patient was an appropriate candidate for music therapy as indicated by response to various interventions. There were 5 additional visits, once per week. The format of the sessions included: a greeting song to orient the participant to the start of the session; singing of 1 to 3 preferred/chosen songs to address cognition and communication; 2 movement songs with instrument playing interventions (either makeshift instruments played by the participants/ caregivers, or actual instruments played by the music therapist) to stimulate cognition and movement; songwriting for self-expression, cognitive, and emotional support; relaxation/mindfulness; and a closing song to help the participant transition at the completion of the session. The music therapist also provided training to caregivers in techniques to utilize music for behavioral support. Each group telemedicine session lasted for 1 hour every week. Additionally, each group telemedicine session had 4 participant/ caregiver dyads assigned to it (the entire cohort that was assigned to the Music Therapy group for that 6-week session).
The other 12 participants were randomized to receive a personalized music playlist that they kept and listened to as they wished — there was no particular guideline that identified how often they should be listening to the music and for how long. Specifically, each participant was called and asked a detailed set of questions about what genre and decade of music they preferred and which artists were their favorites. They were then asked how they wanted to access the playlist (Apple Music, Spotify, CD, memory stick, etc.) and how they wanted it to be sent to them — all participants/caregivers in the Music Playlist group wanted their playlists to be emailed to them on a Spotify playlist for ease and convenience. There has been no evidence of prior studies using a music playlist as a comparison. However, we wanted to use a music playlist as a comparison so that participant/caregiver dyads could get at least some form of tangible benefit from participation in the study for their dementia, instead of no benefit that would be received if the comparison group was no music intervention.
At baseline, we obtained demographic information from chart review as well as interview, which included
participants’ sex, age, dementia type, severity (per most recent behavioral neurology office visit notes), and relationship to caregiver. At baseline and at Week 6 (study completion), participants underwent a Mini Mental State Examination (MMSE) (20) – in this case, a telephonic MMSE (21) since the appointments were conducted over telemedicine. This was done to ascertain a general sense of cognition. Participants also completed the Geriatric Depression Scale (GDS) (22), which measures any symptoms related to depression with higher scores indicating more depressive symptoms. The caregivers completed the Neuropsychiatric Inventory-Questionnaire Severity/ Distress scales (NPI-Q/S/D) (23) to determine the severity of the participants' neuropsychiatric symptoms and the distress these caused the caregiver with higher scores indicating higher severity of symptoms and distress; the Modified Caregiver Strain Index (MCSI) (24) for assessing their strain related to caregiving with higher scores indicating greater strain; and the Patient Health Questionnaire-9 (PHQ-9) (25) to assess their own symptoms of depression. All data (initial demographics and survey results) were tracked in a Microsoft Excel spreadsheet dedicated to data collection. As a single-blinded intervention, the person making the pre-study and follow-up calls for survey completion did not know which participants were in the music therapy vs. music playlist group to prevent bias in data collection.
Data Analysis
In quantifying our data, descriptive statistics were provided for all variables. We primarily used the mean and range to compare age and scores on the various scales for the two groups at baseline (Table 1). We also used the mean and range to compare pre- and post-scores on various measures for both the Music Therapy and Music Playlist groups. Given the small sample size, high dropout rate, and exploratory nature of the analysis, it was not feasible to investigate statistical differences between the 2 groups. Thus, we embraced a qualitative approach in analyzing our data. This type of approach yielded more valuable conclusions about our data and its feasibility.
Results
A total of 24 participants were successfully recruited and randomized into the study and baseline measurements were completed for all participants and their caregivers (Table 1); groups were similar.
During the investigation, there was a 67% dropout rate, equal between the Music Therapy and Music Playlist groups (Figure 1). The reasons for dropout included unavailability of the patient due to a busy schedule with other appointments (6 participants), loss to follow-up with nonresponse to calls (7 participants), and deteriorating health/increasing hospitalizations that impacted the patient’s ability to actively participate in sessions (3 participants).
Given the attrition rate, there were 4 participants in each group who completed the post-intervention follow-up. We compared those participants’ pre- and postintervention scores (Tables 2 and 3).
Table 2 demonstrates the pre- and postscores on the measures of BPSD and cognitive function for the Music Therapy group that were part of the second specific aim. The difference between the post- and pre-scores for each measure has also been calculated for ease of interpretation. Note that for the MMSE, a positive difference score indicates improvement in overall cognitive function. However, in all other measures, a positive difference score would demonstrate deterioration due to a higher count of symptoms that correlate with depression, BPSD, or caregiver strain. A score of zero always indicates no change.
Based on this understanding, the scores for the Music Therapy group are split overall between improvement on some measures and deterioration on other measures, with little change in others. Thus, results were overall similar between pre- and postintervention for the Music Therapy group.
Table 3 demonstrates the pre- and postscores on the measures of BPSD and cognitive function for the Music Playlist group. The difference score for each measure has also been calculated and displayed here and should be interpreted in the same way that was previously described for Table 2.
Results average (range); MMSE=mini mental status examination, GDS=geriatric depression scale, PHQ-9 (Patient Health Questionnaire-9), NPI-Q,S,D (Neuropsychiatric Inventory Questionnaire, Severity, Distress)
MCSI (Modified Caregiver Strain Index)
Results average (range); MMSE=mini mental status examination, GDS=geriatric depression scale, PHQ-9 (Patient Health Questionnaire-9), NPI-Q,S,D (Neuropsychiatric Inventory Questionnaire, Severity, Distress), MCSI (Modified Caregiver Strain Index)
Table 2. Pre- And Post-Score Comparison of BPSD Measures and Cognitive Function for Music Therapy Group
Results average (range); MMSE=mini mental status examination, GDS=geriatric depression scale, PHQ-9 (Patient Health Questionnaire-9), NPI-Q,S,D (Neuropsychiatric Inventory Questionnaire, Severity, Distress), MCSI (Modified Caregiver Strain Index)
Table 3. Pre- and Post-Score Comparison of BPSD Measures and Cognitive Function for Music Playlist Group
Table 1. Demographics and Baseline Measures of Music Therapy and Music Playlist Groups Music Therapy (n=12) Music Playlist (n=12)
For the Music Playlist group, while the MMSE score improved by more than one-and-a-half points, the other scores either did not change, deteriorated, or improved only slightly (e.g., less than one point). Thus, there is no appreciable trend, and results were again overall similar between pre- and post-intervention for this group.
Discussion
As a pilot (proof-of-concept) study, this project presented many challenges along the way to successful completion. As the investigation progressed, there was a higher-than-expected dropout rate of participants. This two-thirds dropout led to a lower sample size, meaning that we could not use statistical testing to yield statistically significant conclusions about our data.
While there was a high attrition rate, this investigation had numerous strengths to offer. For one, the use of a single board-certified music therapist to provide the music therapy to all participants ensured that the intervention was standardized between intervention groups. In addition, the participant/caregiver dyads were also randomly assigned to the intervention and control groups. Furthermore, the pre- and poststudy assessor was blinded to the group assignments. Overall, this is a low-risk intervention that may help alleviate BPSD without causing harm.
We were interested in delving into the qualitative side of data analysis to explore the underlying reasons for the high attrition rate by following up with the participants who dropped out and the music therapist’s impressions of the intervention. Reasons for attrition included that the intervention was too time consuming, it was not interesting, there were too many house distractions, worsening course of dementia, loss to follow-up (nonresponse to followup calls), anxiety of the dyads (both members), low engagement with the therapy for some participants, scheduling difficulties, inconsistent attendance, interference with retirement, and health of the caregiver which impacted ability to participate.
There were several unanticipated challenges to successfully performing group telehealth music therapy. Limitations to this investigation include moderately high attrition rate for the reasons enumerated above leading to lower statistical significance and power; different severities of dementia in the participants (which may have led to
differential engagement during therapy); and no way to know how much those in the Music Playlist group were listening to the music on their own.
With the Music Therapy and Music Playlist pre- and post-scores for BPSD and cognitive function, with special attention given to the difference scores, it does not appear as though there is any clear pattern with the trend of scores. While overall there are fewer downward trends in the Music Playlist group, the scores appear to be relatively stable before and after the intervention. The Music Therapy group has larger upward and downward trends when they occur. With such sizable variability of trends in the pre- and postscores and a small sample size due to a high attrition rate, the data can only be qualitatively analyzed with no comment on statistical significance. With only four participants who completed the intervention in each group, the data is quite mixed, and there can be no firm conclusions drawn from it.
One question that we had regarding the high attrition rate was whether those who had dropped out had more advanced/severe stages of dementia as compared to those participants who completed the study. Because we only recruited participants with mild to moderate dementia, we wanted to explore whether those who did not complete the investigation tended to have more advanced stages of dementia while those who did complete tended to have milder dementia. This theory, however, did not hold. For the Music Playlist group, one out of the two participants with moderate dementia dropped out and 7 out of the 10 participants with mild dementia dropped out. In the Music Therapy group, 5 out of the 8 participants with moderate dementia dropped out and 3 out of the 4 participants with mild dementia dropped out. Thus, the data are again mixed, and nothing can definitively be concluded about the dementia severity of participants who tended to leave the study.
Due to the nature of this project as a pilot study, there is sparse literature exploring the feasibility of delivering group-based telehealth music therapy for persons living with dementia, while using a control group to compare effects on measures of BPSD and cognition. However, as briefly discussed in the introduction, veterans and children with neurodevelopmental disorders have been exposed to music therapy over telemedicine to relieve negative psychological symptoms experienced by these populations (13–15). Another group of studies had
looked at in-person individual and group music therapy sessions and assessed if these could be helpful for those with mental health disorders (16).
While these studies were promising, they posed several limitations, as well. One telehealth music study for veterans mentioned many challenges, including a lack of specificity in reporting and low evidence levels that supported statistical significance in music therapy making a real difference. (13) These limitations were like that in our investigation — where self-reports from participants about the quality of the music therapy varied, dropout rate was appreciable, dementia severity of the participants was uncontrolled for (i.e., proportion of mild versus moderate dementia in each group), and there was not enough power to confirm any specific trend of improvement with the intervention. However, another telehealth music study stated that while technological challenges were notable drawbacks in the study, major benefits of the neurologic music therapy were the ability to continue providing telehealth music therapy when in-person sessions were not possible, increased accessibility for clients living in rural areas, and positive outcomes due to increased engagement with the caregivers. (15) These benefits were also appreciated in our report — flexibility for participants who did not have to come in person for music therapy, increased accessibility for participants living in remote areas, and more involvement/ enjoyment for the caregiver; these were all comments directly made by participant/caregiver dyads.
Another investigation emphasized limitations of music-based therapy more generally, without the telehealth component: “synthesis continues to be precluded by a wide variety of outcomes and outcome measures, a lack of quality reporting, and the inherent complexity of music-based interventions” (16). In our report, we did have quality reporting and used real, robust outcome measures — however, like the aforementioned investigation, we found that studying music-based interventions is complex. Another interesting investigation interestingly performed group-based telehealth music therapy on caregivers themselves, rather than those with dementia. Much like our study, telehealth music therapy was found to be a promising mode of care for caregivers, but that “more research is needed to develop clinical guidelines and best practices; to determine when, where, and why inperson or telehealth services might be indicated” (26).
There are numerous next steps that can be taken to establish future directions for this feasibility study. These may involve incorporating a formal protocol (27) that could improve generalizability and replicability (e.g., standardizing how much those in the Music Playlist group should listen to the music on their own, making sure that participants have similar severities of dementia) and integrating a structured open-ended discussion with patients and their caregivers which could provide more solid qualitative feedback about the investigation. Due to the high dropout rate, we did not believe that simply increasing the sample size or improving advertising would help to enhance feasibility for this study.
The board-certified music therapist offered a few more suggestions for next steps based on her direct experience with the participants. These include ensuring that the caregivers understand the time commitment and will be actively involved in the sessions, attempting a protocol in which the music therapist trains the caregiver in the use of music (the person with dementia is not expected to attend the session) to create a more caregiver-facilitated intervention, and grouping dyads based on assessment instead of random assignment to ensure higher group cohesion.
While there is a possibility that this investigation could be reproduced in the future while addressing the above challenges, it should be cautioned that the limitations may objectively outweigh any tangible benefit the participant/caregiver dyad can obtain in the long term. In other words, feasibility of carrying out such a group-based telehealth music therapy intervention may be low. Given the differential in dementia severity, music therapy engagement, and music preferences among the participants, music therapy with a single individual (or participant/caregiver dyad) over video call may be more advantageous due to the individualized attention that can be given in that setting. Another alternative would be to continue with in-person group or individualized music therapies as being superior to telehealth options due to better ability to build therapeutic rapport in that modality. A third interesting option would be carrying out a therapist-trained caregiver study. As caregivers are the individuals that the patients spend the most time with, there may be benefit if the caregivers were directly trained by therapists to provide music therapy.
The therapy may be better received by the patients in a more flexible setting with greater opportunities for growth, for both patient and caregiver.
Conclusion
While this was truly a novel study to investigate the feasibility of a group-based telehealth music therapy intervention for patients with dementia, it is our opinion that this specific therapy modality may not be feasible for this population of patients given the numerous limitations and challenges outlined above. Despite both interventions being relatively nonburdensome, dropout rate was higher than expected for a variety of reasons, resulting in mixed data from which no firm conclusion could be drawn. We had hypothesized a trend toward some benefit for participants in the Music Therapy group as compared to the Music Playlist group, but the sample size was too small to establish enough power to support this hypothesis. However, we were able to draw some qualitative conclusions about the investigation as described above, which may be helpful as teaching points or in other practical variations of the study in which feasibility may be improved for participant/ caregiver dyads. We hope that these alternatives will better guide treatment and management of dementia in the primary care setting.
Disclosures
Khevna P. Joshi’s research was supported by the Geisinger Clinical Research Fund. Brian J. Piper’s research is supported by the Pennsylvania Academic Clinical Research Center and the Health Resources and Services Administration (D34HP31025), and his prior (2019 – 2021) osteoarthritis research was supported by Pfizer and Eli Lilly. Maya L. Lichtenstein has received funding as co-investigator from NIH/ CORI, Eisai. Alysha D. Suley, Nicole M. Hooper, and Mark V. White have no disclosures.
Acknowledgments
We would like to acknowledge the contributions of the Medical Research Honors Program team in supervising the project, including Sonia Lobo, PhD, Tracey Pratt, MPH, Laura Mayeski, MT (ASCP) MHA, and Michelle Lemoncelli. We would also like to thank Chelsie Derr for assisting us with submission of the project to and approval by the IRB. The Clinical Research Fund provided funding for me to present this project at
the North Atlantic Primary Care Research Group (NAPCRG) Conference in October 2023.
References
1. Liang CS, Li DJ, Yang FC, Tseng PT, Carvalho AF, Stubbs B, et al. Mortality rates in Alzheimer’s disease and non-Alzheimer’s dementias: a systematic review and meta-analysis. The Lancet Healthy Longevity. 2021 Aug 1;2(8):e479–88.
2. Toot S, Swinson T, Devine M, Challis D, Orrell M. Causes of nursing home placement for older people with dementia: a systematic review and meta-analysis. International Psychogeriatrics. 2016 Nov 3;29(2):195–208.
3. Arthur PB, Gitlin LN, Kairalla JA, Mann WC. Relationship between the number of behavioral symptoms in dementia and caregiver distress: what is the tipping point? International Psychogeriatrics. 2018 Aug;30(8):1099–107.
4. Masopust J, Protopopová D, Vališ M, Pavelek Z, Klímová B. Treatment of behavioral and psychological symptoms of dementias with psychopharmaceuticals: a review. Neuropsychiatric Disease and Treatment. 2018 May;14:1211–20.
5. Lin Y, Chu H, Yang C, Chen C, Chen S, Chang H, et al. Effectiveness of group music intervention against agitated behavior in elderly persons with dementia. International Journal of Geriatric Psychiatry. 2011 Apr 13;26(7):670–8.
6. Svansdottir HB, Snaedal J. Music therapy in moderate and severe dementia of Alzheimer’s type: a case-control study. International Psychogeriatrics. 2006 Dec;18(4):613–21.
7. Ledger AJ, Baker FA. An investigation of long-term effects of group music therapy on agitation levels of people with Alzheimer’s Disease. Aging & Mental Health. 2007 May;11(3):330–8.
8. Ridder HMO, Stige B, Qvale LG, Gold C. Individual music therapy for agitation in dementia: an exploratory randomized controlled trial. Aging & Mental Health. 2013 Apr 27;17(6):667–78.
9. Wall M, Duffy A. The effects of music therapy for older people with dementia. British Journal of Nursing. 2010 Jan 27;19(2):108–13.
10. Cooke M, Moyle W, Shum D, Harrison S, Murfield J. A Randomized Controlled Trial Exploring the Effect of Music on Quality of Life and Depression
in Older People with Dementia. Journal of Health Psychology. 2010 Jul;15(5):765–76.
11. Guetin S, Portet F, Picot MC, Defez C, Pose C, Blayac JP, Touchon J. Impact of music therapy on anxiety and depression for patients with Alzheimer's disease and on the burden felt by the main caregiver (feasibility study). L’Encéphale. 2009 Feb;35(1):57-65.
12. Moreno-Morales C, Calero R, Moreno-Morales P, Pintado C. Music Therapy in the Treatment of Dementia: A Systematic Review and MetaAnalysis. Frontiers in Medicine. 2020 May 19;7:160.
13. Gooding LF, Langston DG. Music Therapy With Military Populations: A Scoping Review. Journal of Music Therapy. 2019;56(4):315–47.
14. Lightstone AJ, Bailey SK, Voros P. Collaborative music therapy via remote video technology to reduce a veteran’s symptoms of severe, chronic PTSD. Arts & Health. 2015 Apr;7(2):123–36.
15. Cole LP, Henechowicz TL, Kang K, Pranjić M, Richard NM, Tian GLJ, et al. Neurologic Music Therapy via Telehealth: A Survey of Clinician Experiences, Trends, and Recommendations During the COVID-19 Pandemic. Frontiers in Neuroscience. 2021 Apr 8;15.
16. Golden TL, Springs S, Kimmel HJ, Gupta S, Tiedemann A, Sandu CC, et al. The Use of Music in the Treatment and Management of Serious Mental Illness: A Global Scoping Review of the Literature. Frontiers in Psychology. 2021 Mar 31;12.
17. Holt-Lunstad J, Smith TB, Layton JB. Social Relationships and Mortality Risk: A Metaanalytic Review. PLoS Medicine. 2010 Jul 27;7(7):e1000316.
18. Umberson D, Karas Montez J. Social Relationships and Health: A Flashpoint for Health Policy. Journal of Health and Social Behavior. 2010 Mar;51(1_ suppl):S54–66.
19. Trimble M, Hesdorffer D. Music and the brain: the neuroscience of music and musical appreciation. BJPsych International. 2017 May;14(2):28–31.
20. Burns A, Brayne C, Folstein M. Key Papers in Geriatric Psychiatry: mini-mental state: a practical method for grading the cognitive state of patients for the clinician. M. Folstein, S. Folstein and P. McHugh, Journal of Psychiatric Research,
1975,12, 189-198. International Journal of Geriatric Psychiatry. 1998 May;13(5):285–94.
21. Newkirk LA, Kim JM, Thompson JM, Tinklenberg JR, Yesavage JA, Taylor JL. Validation of a 26-Point Telephone Version of the Mini-Mental State Examination. Journal of Geriatric Psychiatry and Neurology. 2004 Jun;17(2):81–7.
23. Cummings JL. The Neuropsychiatric Inventory. Neurology. 1997 May;48(5_suppl_6).
24. Thornton M, Travis SS. Analysis of the Reliability of the Modified Caregiver Strain Index. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences. 2003 Mar 1;58(2):S127–32.
25. Kroenke K, Spitzer RL, Williams JBW. The PHQ9. Journal of General Internal Medicine. 2001 Sep;16(9):606–13.
26. Brault A, Vaillancourt G. Group Telehealth Music Therapy With Caregivers: A Qualitative Inquiry. Journal of Patient Experience. 2022 Jun;9.
27. Wang SG, Cevasco-Trotter AM, Silverman MJ, Yuan SH. A narrative review of music therapy for neuropsychiatric symptoms in Alzheimer’s disease and rationale for protocolized music teletherapy. Frontiers in Medicine. 2023 Nov 21;10.
CRISPR-Cas9 Gene Therapy Effects on Inherited Eye Disorders
Chloe E. Pisack1*, Krishna S. Patel1*, and Tyler S. Batista-Nieto1*
¹Geisinger College of Health Sciences, Scranton, PA 18509
*Master of Biomedical Sciences Program
Correspondence: cepisack1@som.geisinger.edu
Abstract
Stargardt’s disease (STGD1), retinitis pigmentosa (RP), and Leber’s congenital amaurosis (LCA) inherited eye conditions are among the main causes of blindness and visual impairment. The primary cause of these diseases, which lead to gradual vision loss, are genetic mutations affecting retinal cells. Gene-editing technologies, especially the CRISPR-Cas9 system, are emerging as promising ways to correct harmful genetic mutations. This review focuses on how effectively CRISPR-Cas9 can be used to treat certain inherited eye conditions. In experiments with patient-derived human induced pluripotent stem cells (hiPSCs), researchers were able to fix STGD1 mutations in the ABCA4 gene using CRISPR-Cas9. The corrected cells retained their ability to differentiate into different cell types and showed minimal off-target effects. This suggests the approach could offer a safe and precise way of treating these conditions. In RP, CRISPR-Cas9 was used to target the RPGR gene in mice models, successfully preserving the shape and function of the photoreceptor cells. The CEP290 mutation in LCA10 was corrected in human patients using the CRISPRCas9 based EDIT-101 in the BRILLIANCE clinical trial, which demonstrated a notable improvement in visual outcomes with minimal adverse effects. Even though the results so far are promising, challenges remain, including long-term safety, the administration of these treatments, and the risk of unintended effects. This review emphasizes the need for further research and clinical trials to improve these therapies. It also highlights the exciting potential of CRISPR-Cas9 as a curative treatment for inherited retinal disorders.
Introduction
The human eye is a very complex organ consisting of 3 major tissues: cornea, lens, and retina (1). Damage to any of these tissues can lead to visual impairments. Retinal disease or damage to the retina is the
primary cause of visual impairments (2). Stargardt’s disease, retinitis pigmentosa, and Leber’s congenital amaurosis are inherited eye disorders caused by damage to the retina or macula (3). Stargardt’s disease is a macular dystrophy since the macula and cones are degenerated. Retinitis pigmentosa and Leber’s congenital amaurosis are retinal dystrophies since the rods are affected more than the cones. Regardless of the dystrophy, there is progressive vision loss in all 3 disorders (3).
Clustered regularly interspaced palindromic repeats (CRISPR)-Cas9 is a gene editing tool that functions to correct errors in the genome (4). CRISPR-Cas9 turns the genes on or off to fix the mutation. CRISPR-Cas9 can correct a disease mutation, thus curing the genetic disorders of the mice. The 2 main components of CRISPR-Cas9 are guide RNA and Cas9. The guide RNA matches the sequence of the target gene, and Cas9 is an endonuclease that makes a double-stranded DNA break at the specific location on the genome (4). After the cut has been made, the cell’s natural DNA repair mechanisms, homology-directed repair (HDR), and non-homologous end-joining (NHEJ) are recruited to repair the double-stranded break, which allows gene editing (5).
Therapeutic gene editing has advanced greatly within the last decade (6). However, gene editing and gene therapy are in their early stages of development. This focused review examines the advancement of gene editing and gene therapy in Stargardt’s disease, retinitis pigmentosa, and Leber’s congenital amaurosis.
Stargardt’s Disease
Stargardt’s disease (STGD1), also known as juvenile macular dystrophy, is an inherited autosomal recessive eye disorder (7). The disease causes vision loss and retinal degeneration in adults and children (7). Early onset of STGD1 is between 1 to 10 years of age (8). The range of late-onset STGD1 is 45 to 72 years of
age (9). The prevalence of the disease is 1 in 8,000 to 10,000 individuals (10). STGD1 causes retinal degeneration and vision loss (7). Such eye impairments are caused by a biallelic mutation on the ABCA4 gene (11). The ABCA4 gene plays a vital role in recycling retinoid byproducts (12). However, a mutation in the ABCA4 gene can prevent the transportation of byproducts, thus leading to toxic buildup in the retina, which causes visual impairments (13).
Human induced pluripotent stem cells (hiPSCs) are stems cells that can differentiate into any cell type, enabling them to model any disease or drug screening (14). The use of hiPSCs is relevant to studying eye-related diseases since access to brain cells is challenging (15). However, with the use of hiPSCs from patients with neurodegenerative disease, conducting studies can be efficient (15).
CRISPR-Cas9 and transcription activator-like effector nucleases (TALENs) are efficient tools for gene editing as both methods involve inducing DNA double-strand breaks and recruit cell repair mechanisms for genetic change (16). A difference between the two methods is that CRISPR-Cas9 relies on single-guide (sgRNA) to target specific DNA sequences, while TALENs use DNA-binding domains (DBDs) to target the specific DNA (17). The ability to have higher cleavage efficiency makes CRISPR-Cas9 preferable over TALENs (7).
The two main ABCA4 gene variants are c.4253+4C>T and c.3211_3212instGT (7). These variants cause a splicing defect and generate a frameshift, respectively (7). Even though there are no current treatments for STGD1, hiPSC studies conducted aim to correct the 2 ABCA4 gene mutations using CRISPR-Cas9. Therefore, using CRISPER-Cas9 holds significant promise in fixing the mutation of the ABCA4 gene in STGD1, and offers hope in treating a wide range of inherited eye disorders (7).
Patient-derived fibroblasts from 2 individuals with STGD1, each harboring heterozygous mutations in ABCA4, were reprogrammed into human induced pluripotent stem cell (hiPSC) lines (18). The pathogenicity of each variant was determined using the ENSEMBL Variant Effect Predictor (VEP) (7,19). VEP is a computational tool used to determine the impact of a genetic mutation with the use of algorithms (7,19). Single-guide RNA (sgRNAs) and TALENS were designed, with sgRNAs being selected based on efficacy with the least off-target effects
(7). CRISPR-Cas9 was introduced into the hiPSCs by transfecting hiPSCs with ribonucleoprotein (RNP) (7). The RNP consisted of Cas9, sgRNA, and single-strand oligodeoxynucleotides (ssODNs), necessary for gene editing (7). The function of ssODNs is to allow precise gene editing by allowing homology-directed repair (HDR) mediated modification to take place (7,20). PCR amplification was completed on the ABCA4 gene target region with the new modification made by CRISPRCas9 (7). Additionally, Sanger sequencing was carried out to determine whether the correction of the ABCA4 gene resulted in off-target alterations (7). Lastly, RTPCR was conducted to determine if the ABCA4 gene had been corrected and if the cells could differentiate into the 3 germ layers (7).
Upon conducting the in silico analysis, the main 2 variants of the ABCA4 mutations were all missense mutations except c.514G>A, which are pathogenic (7). Additionally, c.3211_3212inGT and c.2023G>A are the main gene variant causing STGD1 since these variants appear at a low frequency in the population. The cleavage efficiency for sgRNA/Cas9 was between 15% and 45%, which was higher than that of the DNA cleavage performed by TALENS. Due to the low cutting made by TALENs, CRISPR-Cas9 was used to correct the STGD1 mutations in the ABCA4 gene.
The use of ssODNs plays a critical role in CRISPR-Cas9 gene editing by serving as repair templates during homology-directed repair (HDR) following cleavage by the Cas9 RNP complex (7). ssODNs are precise in carrying out HDR-mediated repairs, and ssODNs repair single-nucleotide substitutions (21). HDRmediated modifications are much more precise than non-homologous end joining (NHEJ) because NHEJ is more error-prone and more likely to generate indels when the two ends of the double-stranded breaks are re-ligated (7).
The gene-edited hiPSC clones were checked to see whether differentiation occurred to form the 3 germ layers. RT-PCR and immunofluorescence indicated that the gene-edited cells can divide into the three germ layers. Thus, indicating that pluripotency is conserved (7).
Two sgRNAs had off-target effects, sgRNA2 had 7, and sgRNA6 had 57 off-targets. After conducting Sanger sequencing, it was confirmed that the editing targeted the correct location of interest and did not cause any other genomic alterations (7).
Even though there have been no therapies to cure visual loss in STGD1 patients as of yet, there have been studies to analyze the effects of CRISPR-Cas9mediated gene editing (22). However, there are issues with an in vivo CRISPR-Cas9 system such as offtarget mutations and double stranded break related oncogenesis (22). The review reported that two sgRNAs had off-target effects, and even though they did not cause any genomic issues, gene editing still caused off-target effects (22).
Additionally, hiPSCs offer a robust in vitro model for investigating neurodegenerative diseases such as STGD1, where primary neural tissues are inaccessible for direct study (15). RNP complexes comprising Cas9, sgRNA, and ssODNs facilitated precise gene editing via homology-directed repair, while the presence of the ssODN also reduced Cas9 re-cutting at successfully edited loci (21,22). In addition, delivering Cas9 via the RNP complex is more efficient for single-base substitution in comparison to plasmid-based Cas9 transfection (20).
The use of ssODN-gene editing results in the ABCA4 gene being edited without causing genomic alterations, thus indicating that CRISPR-Cas9 can be potentially used for the treatment of STGD1 (7). hiPSC-based studies using CRISPR-Cas9 have advanced gene editing research; however, its in vivo application remains nascent, with further investigation needed to assess long-term efficacy and potential adverse effects in treating STGD1 (7).
Retinitis Pigmentosa
Retinitis pigmentosa (RP) is an inherited disorder where damage to the photoreceptors, rods and cones in the eye, result in progressive visual loss. RP is a progressive, long-lasting retinal disorder that typically worsens over time and is classified as either syndromic or non-syndromic, depending on the presence of associated systemic features. Syndromic RP can affect other organs in the body whereas non-syndromic affects the eye only. Most of the RP cases are nonsyndromic, and about 20% to 30% are syndromic (24). In RP, the most common genes mutated are the ones involved in the visual cascade (34). The visual cascade involves the photochemical reaction where light is converted to an electrical signal in the retina (35).
RP has several clinical manifestations, including night blindness, dust-like particles collecting in the
vitreous, posterior cataracts, and fundus damage. Night blindness is one of the symptoms that first occurs because the rods are damaged and cannot properly adjust to dim lighting. Early diagnosis can be detected with dust-like particles in the vitreous. The severity of the posterior cataracts associated with RP corresponds to the age of those who are affected. Lastly, the appearance of the fundus in RP also helps determine the stage of the disease when no other symptoms occur (25). The fundus of the eye is essential to vision because it is composed of the retina, macula, optic, disc, fovea, and blood vessels (36).
The prevalence of RP is estimated at 1 in 3,000 to 1 in 7,0000 individuals (24). Approximately 181 genes have been mapped to chromosomal loci associated with inherited RP, and around 129 genes have been identified at the sequence level, underscoring the disorders extensive genetic heterogeneity (26). A specific mutation that will be discussed in more detail involves the retinitis pigmentosa GTPase regulator (RPGR) which is associated with X-linked retinitis pigmentosa (XLRP) - a form that accounts for 10%20% of all RP cases and is among the most severe (27).
Currently, only 1 approved treatment exists for a subpopulation of patients experiencing RP (31). Other treatments that aim to help manage the condition include gene therapy, vitamin therapy to protect photoreceptors, and retinal transplants for more severe cases (32). Gene editing therapy has been used as a primary way to study the genetic characteristics of many diseases including RP. Gene editing for the retina is ideal because it is easily accessible and is also isolated by the blood retinal barrier (28). CRISPR-CaS9 has shown promising treatments for RP and many other diseases (23).
Recent animal studies have shown how specific gene therapies in rats can lead to promising treatments in humans for RP. The purpose of the investigation was to test the efficacy of short palindromic repeat/ Cas9-mediated gene editing therapy in rodents with RPGR. Mice with Rpgr−/yCas9+/WT genotype received subretinal injections of adenovirus vectors with the sgRNA and donor template, and their therapeutic effects were examined at 1, 6, and 12 months. The Rpgr−/yCas9+/WT genotype consists of the hemizygous knockout mutation in the Rpgr gene with one allele of the Cas9 gene with the other allele being wild type A 6-month-old mouse is equivalent
to a human in their 30s, a 12-month-old mouse is equivalent to a human in their mid-40s and an 18-month-old mouse is equivalent to an old human.
A total of 72 mice were used in the study where mRNA of in vitro-transcribed Cas9 and sgRNA were injected into the zygotes of the mice that were authenticated with PCR and sequencing. After receiving the injection, the mice were given 1% atropine drops and neomycinpolymyxin B-dexamethasone ophthalmic ointments. Atropine drops are a type of medication used to dilate the pupils (28). In order to determine the effects of the injection, histology and immunofluorescence, fundus photography, and DNA analyses were conducted. The immunofluorescence technique allows for the detection of antigens in different tissues (30). In this case the immunofluorescence allows for the detections of RPGR re- expression in the retina.
Close examination of the control and RPGR mice revealed that photoreceptors of the mice with engineered RPGR indeed showed slow retinal degeneration. Retinal fundus imaging revealed that small yellow-white spots appeared within the first 3 months, becoming sparse by 12 months, although pigment deposition persisted. Histological evaluation of retinal sections from WT and RPGR mice showed that in RPGR mice, subtle age-related photoreceptor cell loss was evident by 3 months, progressed by 6 months, and became more pronounced by 12 months. The retinal morphology of mice treated with mutant RPGR with 2 different vectors showed that there was extensive and great retinal photoreceptor preservation. Nine layers of the retina had preserved photoreceptors while the untreated portion of the retina only had 5. The outer nuclear layer of the treated areas was also thicker than those that were not treated (29).
Other gene editing therapies have been tested on RP mutants, however, CRISPR-Cas9 has been shown to be a promising therapy. Some other gene therapies that help treat a certain disease add a functional or partially functional copy of a gene without removing the mutant (or dysfunctional) gene. In contrast, CRISPR-Cas9 can generate specific modifications to a locus and eliminate any defective DNA. The results from the above study show that the CRISPR-Cas9 mediated RPGR gene therapy prevented any photoreceptor degeneration and that the treatment was able to persist for longer periods of time. The study performed
on rats showed promising results in translation to CRISPR-Cas9 in humans.
Leber’s Congenital Amaurosis
Leber’s congenital amaurosis (LCA) is one of the most severe retinal dystrophies associated with childhood blindness (37). LCA is a rare inherited eye disorder prevalent in 1/30,000 people and 1/81,000 newborn babies worldwide (38). Although rare, it is one of the most common inherited retinal diseases (IRDs) with a prevalence of greater than 5% (38). The disease causes severe vision loss and degeneration in the photoreceptor cells, specifically the rods and cones, in the retina (39).
LCA is associated with mutations in approximately 38 different genes (40). One of the most common mutations causing the disease is in the CEP290 gene (40). There are 18 known types of LCA, each caused by a different gene mutation (41). The CEP290 mutation causes LCA10 which is responsible for more than 30% of all cases of LCA globally (42). In the United States, this mutation is responsible for 77% of cases (43). The CEP290 mutation results from a switch in a base sequence from adenine to guanine in intron 26 on chromosome 12 (43). This mutation generates an aberrant splice site, leading to the integration of a cryptic exon into the DNA sequence, resulting in a truncated CEP290 protein which leads to the symptomology associated with LCA10 (43).
Currently, there are no known cures for this mutation, but multiple clinical trials are underway to evaluate possible curative treatments (41). The most impactful clinical trial that has been published is the BRILLIANCE trial (44). This study specifically looked at treating the CEP290 mutation with a CRISPR-Cas9 sequence to regenerate the loss of photoreceptor cells in the retina. This trial was the first human-based CRISPR-Cas9 ocular disease model and used EDIT101, a specific CRISPR-Cas9, that contained an adenoassociated virus 5(AAV5), Staphylococcus aureus Cas9 nuclease, G coupled protein kinase 1, and guide RNAs (44). These factors are required for any CRISPR Cas9 model to help localize the cut in the DNA sequence for the specific mutations (45, 46).
The above study had 14 participants, 2 of which were children age 9 and 14 (44). The participants were either homozygous or heterozygous in their inheritance of LCA10. The study’s main goals were the safety of the treatment and to look at the efficacy
of the gene therapy in treating LCA. The efficacy was analyzed using tests examining four main outcomes: visual acuity, retinal sensitivity, visual-related functional navigation, and vision-related quality of life. All of these factors are impacted by patients with the CEP290 mutation that results in LCA.
The changes in participants’ baseline visual acuity were looked at using the Early Treatment Diabetic Retinopathy Study (ETDRS) and the Berkley Rudimentary Vision Test (BRVT) (44). The ETDRS is a chart that contains multiple rows of equally spaced 5-letter groupings that get increasingly smaller as you move downward (47). The participants who were determined to have reduced visual acuity based on the ETDRS chart used the BRVT to find their visual acuity. The BRVT is a simple and effective test used to look at visual acuity in individuals determined to have significantly reduced visual acuity (48). Retina sensitivity to red, blue, and white light was measured with a full field stimulus test (FST). Visual-related functional navigation was measured using the OraVisual Navigation Challenge (VNC), a mobility test requiring participants to navigate obstacles under different light conditions (49). Finally, the participant’s vision-related quality of life was measured in children using the Children’s Visual Function Questionnaire (CVFQ) and in adults using the National Eye Institute Visual Function Questionnaire–25 (NEI VFQ-25) (44).
EDIT-101 was administered as a single-dose subretinal injection after participants underwent a pars plana vitrectomy and 3 days of oral prednisone. The purpose of the pars plana vitrectomy is to remove some part of the vitreous humor in the eye to allow easier access to the posterior part of the eyeball for the subretinal injection (50). The EDIT-101 was given in either a low dose for cohort 1, an intermediate dose for cohort 2, or a high dose for cohort 3; cohort 4 consisted of the 2 child participants who received the intermediate dose. After the EDIT-101 injection, oral prednisone was continued for 3 weeks in all cohorts.
After 2 years of monitoring the participants for varying amounts of time, the results showed significant improvements due to the insertion of the EDIT-101 CRISPR-Cas9 treatment (44). The treatment was safe, with 22 adverse side effects reported. Most of these side effects were predicted and expected. The treatment was also effective, with 11 out of the 14 participants finding support in at least 1 of the 4 main
efficacy outcomes of visual acuity, retinal sensitivity, visual-related functional navigation, and vision-related quality of life. Six out of the 14 participants found improvement in 2 or more efficacy outcomes.
This study supports future research involving CRISPRCas9 as a potential curative treatment for the genetic mutation of CEP290. Most participants showed efficacy of inserting EDIT-101, a CRISPR-Cas9 gene therapy, with minimal adverse effects. It also shows support for conducting more research on other mutations responsible for LCA, such as CRB1, GUCY2D, and RPE65 (51). The BRILLIANCE clinical trial opens the door for future human studies to be done, as the trial showed gene editing as a possible treatment for mutations that cause other IRDs like LCA.
Overall, the BRILLIANCE clinical trial was a pivotal study in showing genetic therapies' impact on human inherited diseases. Further study in CRISPR-Cas9 models may show the potential to generate treatments for diseases caused by mutations with previously unknown cures.
Conclusion
CRISPR-Cas9, a powerful gene-editing tool, has seen remarkable advancements over the past decade (6). This review focuses on its application in 3 inherited eye disorders- Stargardt's disease (STGD1), retinitis pigmentosa (RP), and Leber’s congenital amaurosis (LCA)- to evaluate the therapeutic potential of CRISPR-Cas9 as a form of gene therapy.
The study with STDG1 showed promising gene editing in hiPSCs and CRISPR-Cas9 (7). However, further studies need to be conducted to do in vivo gene editing with CRISPR-Cas9. The study with RP in mice indicated that CRISPR-Cas9 mediated with RPGR gene therapy prevents photoreceptor degeneration. However, further studies need to be conducted to see the effects of CRISPR-Cas9 on humans (27). The study with EDIT-101 and its efficacy with the CEP290 mutation was effective in the human participants with LCA10 and showed promise for further in vivo studies on humans to correct inherited genetic mutations (44). The findings in these studies demonstrate the effects of CRISPR-Cas9 on STDG1, RP, and LCA and provide strong support for future research into genetic therapy as a possible curative treatment for eye diseases caused by mutations in the human genome.
Disclosures
The authors have no conflicts of interest or financial disclosures relevant to this manuscript.
Acknowledgments
None.
References
1. Graw J. Eye development. Curr Top Dev Biol. 2010; 90:343-86.
2. Zhou C, Li S, Ye L, Chen C, Liu S, Yang H, et al. Visual impairment and blindness caused by retinal diseases: A nationwide register-based study. J Glob Health. 2023; 13:04126.
4. Redman M, King A, Watson C, King D. What is CRISPR/Cas9? Arch Dis Child Educ Pract Ed. 2016; 101(4):213-5.
5. Vasiliou SK, Diamandis EP, Church GM, Greely HT, Baylis F, Thompson C, et al. CRISPR-Cas9 system: Opportunities and concerns. Clin Chem. 2016; 62(10):1304-11.
6. Khalil AM. The genome editing revolution: Review. J Genet Eng Biotechnol. 2020; 18(1):68.
7. Siles L, Martinez-Garcia M, Palfi A, Alifragis P, Irion S, Warre-Cornish K, et al. Efficient correction of ABCA4 variants by CRISPR-Cas9 in hiPSCs derived from Stargardt disease patients. Mol Ther Nucleic Acids. 2023; 32:64-79.
8. Lambertus S, van Huet RA, Bax NM, Hoefsloot LH, Cremers FP, Boon CJ, et al. Early-onset Stargardt disease: Phenotypic and genotypic characteristics. Ophthalmol. 2015; 122(2):335-44.
9. Westeneng-van Haaften SC, Boon CJ, Cremers FP, Hoefsloot LH, den Hollander AI, Hoyng CB. Clinical and genetic characteristics of lateonset Stargardt's disease. Ophthalmol. 2012; 119(6):1199-210.
10. Tsang SH, Sharma T. Stargardt disease. Adv Exp Med Biol. 2018; 1085:139-51.
11. Koenekoop RK. The gene for Stargardt disease, ABCA4, is a major retinal gene: A mini-review. Ophthalmic Genet. 2003; 24(2):75-80.
12. Sun H, Nathans J. Mechanistic studies of ABCR, the ABC transporter in photoreceptor outer segments responsible for autosomal recessive Stargardt disease. J Bioenerg Biomembr. 2001; 33(6):523-30.
14. Poliwoda S, Noor N, Downs E, Schaaf A, Cantwell A, Ganti L, et al. Stem cells: A comprehensive review of origins and emerging clinical roles in medical practice. Orthop Rev (Pavia). 2022; 14(3):37498.
15. Rivetti di Val Cervo P, Besusso D, Conforti P, Cattaneo E. hiPSCs for predictive modelling of neurodegenerative diseases: Dreaming the possible. Nat Rev Neurol. 2021; 17(6):381-92.
16. Gaj T, Gersbach CA, Barbas CF 3rd. ZFN, TALEN, and CRISPR/Cas-based methods for genome engineering. Trends Biotechnol. 2013; 31(7):397405.
17. Boettcher M, McManus MT. Choosing the right tool for the job: RNAi, TALEN, or CRISPR. Mol Cell 2015; 58(4):575-85.
18. Riera M, Patel A, Burés-Jelstrup A, Corcostegui B, Chang S, Pomares E, et al. Generation of two iPS cell lines (FRIMOi003-A and FRIMOi004-A) derived from Stargardt patients carrying ABCA4 compound heterozygous mutations. Stem Cell Res 2019; 36:101389.
19. Fawzy M, Marsh JA. Understanding the heterogeneous performance of variant effect predictors across human protein-coding genes. Sci Rep. 2024; 14(1):26114.
20. Okamoto S, Amaishi Y, Maki I, Enoki T, Mineno J. Highly efficient genome editing for single-base substitutions using optimized ssODNs with Cas9RNPs. Sci Rep. 2019; 9(1):4811.
21. Richardson CD, Ray GJ, DeWitt MA, Curie GL, Corn JE. Enhancing homology-directed genome editing by catalytically active and inactive CRISPR-Cas9 using asymmetric donor DNA. Nat Biotechnol. 2016; 34(3):339-44.
22. Hu X, Zhang B, Li X, Li M, Wang Y, Dan H, et al. The application and progression of CRISPR/Cas9 technology in ophthalmological diseases. Eye (Lond). 2023; 37(4):607-17.
23. Liu W, Liu S, Li P, Yao K. Retinitis pigmentosa: Progress in molecular pathology and biotherapeutical strategies. Int J Mol Sci. 2022; 23 (9):4883
24. Verbakel SK, Van Huet RAC, Boon CJF, Den Hollander AI, Collin RWJ, Klaver CCW, et al. Nonsyndromic retinitis pigmentosa. Prog Retin Eye Res. 2018; 66:157
25. Fahim AT, Daiger SP, Weleber RG, Adam MP, Feldman J, Mirzaa G. Nonsyndromic retinitis pigmentosa overview. GeneReviews[Internet]. 2023
26. Daiger SP, Bowne SJ, Sullivan LS. Perspective on genes and mutations causing retinitis pigmentosa. Arch Opthalmol. 2007; 125(2):151-8
27. Hu S, Du J, Chen N, Jia R, Zhang J, Liu X, et al. In vivo CRISPR/Cas9-mediated genome editing mitigates photoreceptor degeneration in a mouse model of X-linked retinitis pigmentosa. Invest Opthalmol Sci. 2020; 9;61(4):31
28. Cleveland Clinic. Atropine eye solution.[Internet]. Cleveland, OH: Cleveland Clinic; 2024[cited 2024 Nov 17]. Available from https://my.clevelandclinic. org/health/drugs/20729-atropine-eye-solution
29. Mayo Clinic. Neomycin, polymyxin b, and dexamethasone (opthalmic route). [Internet]. Rochester, MN: Mayo Clinic; 2024[cited 2024 Nov 17]. Available from https://www.mayoclinic.org/ drugs-supplements/neomycin-polymyxin-b-anddexamethasone-ophthalmic-route/description/ drg-20406312#
30. Im K, Mareninov S, Diaz MFP, Yong WH. An introduction to performing immunofluorescence staining. Methods Mol Biol. 2019:1897:299-311
31. Cross N, Steen CC, Zegaoui Y, Satherley A, Angelillo L. Current and future treatment of retinitis pigmentosa. Clin Opthalmol. 2020; 16:2909-2921
32. Musarella MA, Macdonald IM. Current concepts in the treatment of retinitis pigmentosa. J Opthalmol 2011:2011:753547
33. Hamel C. Retinitis pigmentosa. Orphanet J Rare Dis. 2006;1:40
34. Phelan JK, Bok D. A brief review of retinitis pigmentosa and the identified retinitis pigmentosa genes. Mol Vis. 2000; 8:6:116-24
36. AAO. American Academy Of Opthalmology [Internet]. Sanfrancisco, CA : AAO; 2020[ cited 2024 Nov 17]. Available from https://www.aao. org/eye-health/anatomy/fundus#:~:text=The%20 fundus%20is%20the%20inside,find%2C%20 watch%20and%20treat%20disease.
37. Russell SR, Drack AV, Cideciyan AV, Jacobson SG, Leroy BP, Van Cauewenbergh C, et al. Intravitreal antisense oligonucleotide sepofarsen in Leber congenital amaurosis type 10: A phase 1b/2 trial. Nat Med. 2022; 28(5):1014–1021.
38. Chiu W, Lin TY, Chang YC, Isahwan-Ahmad Mulyadi Lai H, Lin SC, Ma C, et al. An update on gene therapy for inherited retinal dystrophy: Experience in Leber congenital amaurosis clinical trials. Int J Mol Sci. 2021; 22(9):4534.
39. Coussa RG, Lopez Solache I, Koenekoop RK. Leber congenital amaurosis, from darkness to light: An ode to Irene Maumenee. Ophthalmic Genet. 2017; 38(1):7-15.
40. Huang CH, Yang CM, Yang CH, Hou YC, Chen TC. Leber's congenital amaurosis: Current concepts of genotype-phenotype correlations. Genes (Basel). 2021; 12(8):1261.
41. University of Arizona. Leber Congenital Amaurosis. [Internet]. Arizona: University of Arizona; [cited 2024 Nov 21]. Available from: https://disorders.eyes.arizona.edu/disorders/ Leber-congenital-amaurosis
42. Biasutto P, Dulla K, Adamson P, Schulkens I, Schmidt I, Lane A, et al. QR-110 treatment for Leber’s congenital amaurosis type 10: Restoration of CEP290 mRNA levels and ciliation in LCA10 iPSC-derived optic cups. Invest Ophthalmol Vis Sci. 2017; 58(8):249.
43. Daich Varela M, Cabral de Guimaraes TA, Georgiou M, Michaelides M. Leber congenital amaurosis/early-onset severe retinal dystrophy: Current management and clinical trials. Br J Ophthalmol. 2022;106(4):445-451.
44. Pierce EA, Aleman TS, Jayasundera KT, Ashimatey BS, Kim K, Rashid A, et al. Gene editing for CEP290-associated retinal degeneration. N Engl J Med. 2024; 390(21):1972–1984.
45. Denes CE, Cole AJ, Aksoy YA, Li G, Neely GG, Hesselson D. Approaches to enhance precise CRISPR/Cas9-mediated genome editing. Int J Mol Sci. 2021; 22(16):8571.
46. Ruan GX, Barry E, Yu D, Lukason M, Cheng SH, Scaria A. CRISPR/Cas9-mediated genome editing as a therapeutic approach for Leber congenital amaurosis 10. Mol Ther. 2017; 25(2):331-341.
47. Bailey IL, Lovie-Kitchin JE. Visual acuity testing. From the laboratory to the clinic. Vision Res. 2013; 90: 2–9.
48. Bailey IL, Jackson AJ, Minto H, Greer RB, Chu, MA. The Berkeley rudimentary vision test. Optom Vis Sci. 2012; 89(9):1257–1264.
49. Shapiro A, Corcoran P, Sundstrom C, Angjeli E, Rodriguez JD, Abelson MB, et al. Development and validation of a portable visual navigation challenge for assessment of retinal disease in multi-centered clinical trials. Invest Ophthalmol Vis Sci. 2017; 58(8):3290.
50. Iovino C, Rosolia A, Damiano L, Iodice CM, Di Iorio V, Testa F, et al. Pars plana vitrectomy in inherited retinal diseases: A comprehensive review of the literature. Life (Basel). 2023; 13(6):1241.
Investigating Racial and Ethnic Variation in High Sensitivity C-Reactive Protein Levels Among Individuals with Prediabetes
Jacob George1†‡, Savanna Shaw1†‡, Kevin Sajan1†‡, Alicia Johns2,3, Amanda Young2,3, and Bobbie Johannes2
¹Geisinger College of Health Sciences, Scranton, PA 18509
²Geisinger, Department of Population Health Sciences, Danville, PA 17822
³Geisinger, Biostatistics Core, Danville, PA 17822
†Doctor of Medicine Program
‡Authors contributed equally
Correspondence: jgeorge9@geisinger.edu
Abstract
Background: High sensitivity C-reactive protein (hsCRP) is a marker of inflammation linked to chronic health conditions, including prediabetes. Many factors, including social determinants of health such as socioeconomic status, food insecurity, chronic stress and many others, can affect hs-CRP levels. Given the well-documented health disparities across racial and ethnic minority groups regarding social determinants of health, it’s crucial to investigate how hs-CRP levels vary in people with prediabetes to uncover possible contributing factors that could predispose an individual to develop prediabetes and its sequelae. This study investigated whether hs-CRP levels differ among racial and ethnic groups with prediabetes, potentially reflecting the role of health inequalities.
Methods: Data from the National Health and Nutrition Examination Survey (NHANES) were utilized for adults age 18 and older with prediabetes. Participants were grouped into racial and ethnic minority populations (Mexican American, Other Hispanic, Non-Hispanic Black, Non-Hispanic Asian, and Other Race) and non-minority populations (Non-Hispanic White). A Wilcoxon rank-sum test was performed to compare the distribution of hs-CRP levels after adjusting for sample weights. Further subgroup analyses were conducted comparing CRP levels by minority status for the following subgroups: sex, insurance status, and education status.
Results: In the unweighted sample, racial and ethnic minority populations had significantly higher hsCRP levels than non-minority populations (p=0.03). However, after adjusting for survey weights, this
difference approached but did not reach statistical significance (p=0.09). The median hs-CRP level for minority populations was 2.99 (IQR: 1.22, 5.74), compared to 1.94 (IQR: 1.03, 4.67) in non-minority populations (p = 0.09). Non-Hispanic Black individuals had the highest hs-CRP levels, with a median hs-CRP level of 3.93. CRP levels only seemed to significantly differ in the combined subgroup of high school graduates and individuals with some college or AA degree. In this subgroup, the minority group had higher median CRP levels than the non-minority group (3.10 versus 2.10; p-value = 0.0048).
Conclusion: Our findings present disparities in hs-CRP levels among different racial and ethnic populations with prediabetes, especially in Non-Hispanic Black individuals. Understanding and addressing possible social determinants of health, such as food insecurity or chronic stress, is essential for accurately assessing individual risk and implementing patient-centered preventive strategies.
Introduction
Prediabetes is a significant public health concern in the United States, affecting an estimated 97.6 million adults as of 2021, according to the Centers for Disease Control and Prevention (1). This condition is characterized by blood glucose levels that are elevated above the normal range but have not met the threshold for a diagnosis of Type 2 diabetes (1). The CDC defines prediabetes based on glycated hemoglobin (HbA1C) levels ranging from 5.7% to 6.4%, reflecting average blood glucose concentrations over approximately 3 months (1). Several studies have shown that diagnosis of prediabetes is linked to an increased
risk of developing early forms of nephropathy and chronic kidney disease (2). Additionally, individuals with prediabetes are at markedly increased risk of progression to Type 2 diabetes and its cardiovascular sequelae including coronary heart disease and stroke (3). Large prospective cohorts further demonstrate that prediabetes confers an elevated risk of all-cause mortality relative to normoglycemia, underscoring its clinical significance (3). Thus, early identification of individuals at risk for prediabetes is critical, as timely intervention can delay or prevent the progression to Type 2 diabetes and development of its associated risks (4, 5).
One biomarker that has garnered increased attention for use as a diagnostic test for prediabetes is highsensitivity C-reactive protein (hs-CRP), an acute-phase protein produced by the liver in response to systemic inflammation (6, 7). Elevated hs-CRP levels have been associated with prediabetes, suggesting a potential role for inflammatory processes in its pathogenesis (8). While elevated hs-CRP levels have been linked to the development of prediabetes, limited research has explored how this relationship may vary across different racial and ethnic groups. Existing knowledge indicates that among adults in the United States, there is a disparity of diabetes diagnoses, with minority groups having a higher prevalence in comparison to non-minority groups (9). These disparities have been linked to social determinants of health, for example limited access to care, increased exposure to chronic stress, and residence in poorer neighborhood environments (10–14). Though a social construct, race and ethnicity have notable impacts upon the equity of healthcare outcomes, thus a relevant relationship to explore. This study aims to address this gap in the literature by examining the association between hs-CRP and prediabetes within the context of racial and ethnic variation. There is a focus on examining whether racial and ethnic minority groups (e.g. NonHispanic Black, Hispanic) have higher hs-CRP levels than non-minority populations (Non-Hispanic White), potentially reflecting the role of health inequities in the development of prediabetes.
Methods
Data Collection
Data were obtained from the National Health and Nutrition Examination Survey (NHANES), a crosssectional, multistage probability survey conducted
by the Centers for Disease Control and Prevention to assess the health and nutritional status of the non-institutionalized U.S. population (15). This study used publicly available, de-identified data and was determined not to constitute human subjects research by the Geisinger Institutional Review Board (Study #2024-0576).
Patient Population
Participants included adults age 18 years and older who met the criteria for prediabetes, defined by HbA1c levels of 5.7% to 6.4%. Race and ethnicity were self-reported by participants during standardized NHANES interviews and categorized as Non-Hispanic White, Non-Hispanic Black, Mexican American, Other Hispanic, Non-Hispanic Asian, or Other Race, including multiracial individuals. Categories were selected based on those provided in the NHANES dataset. Race and ethnicity were assessed in this study to evaluate potential disparities in systemic inflammation among individuals with prediabetes, as prior literature suggests associations between race and ethnicity, and chronic disease risk.
Statistical Analysis
To produce nationally representative estimates, we utilized the survey weights developed by NHANES to account for the complex, multistage probability design. Specifically, fasting subsample weights were applied, as hs-CRP is a laboratory variable derived from the fasting blood draw component. For the single NHANES cycle used, weights were adjusted according to the National Center for Health Statistics analytic guidelines. The R “survey” package was used to apply appropriate weights in all statistical components.
Descriptive statistics were performed on the unweighted and weighted samples. Medians, interquartile ranges (IQR), and ranges were reported for continuous variables and frequencies, and percentages for categorical variables. Weighted population characteristics were further examined across race and ethnicity status, and descriptive statistics were provided. A Wilcoxon rank sum test was performed on the unweighted and weighted samples. Further subgroup analyses were conducted comparing hs-CRP levels by minority status for the following subgroups: sex, insurance status, and education status. For education status, the groups were combined to create the following 3 groups (unknown were not considered): less than high school, high school
graduate/some college or associate's degree (AA), and college graduate or above. A p-value of 0.05 was used to determine significance. Analyses were performed using SAS version 9.4 (Cary, North Carolina, USA) and R v4.0.3 (R Foundation for Statistical Computing, Vienna, Austria).
Results
The unweighted study sample consisted of 675 participants, representing a weighted population of 19,280,268 individuals. The median age was 61.0 years (IQR: 50.0 - 70.0) in the unweighted sample and 57.1 years (IQR: 43.7-65.9) in the weighted population, with an age range of 18–80 years in both samples. Median hs-CRP levels were consistent across both samples at 2.30 mg/L, with similar interquartile ranges (unweighted: 1.03-5.45 mg/L; weighted: 1.08-5.31 mg/L). The gender distribution was 41.3% male and 58.7% female in the unweighted sample, compared to 45.7% male and 54.3% female in the weighted population (Table 1).
Educational attainment varied, with the largest proportion of participants being a college graduate or above (unweighted: 37.3%; weighted: 35.3%). The majority of the unweighted and weighted sample had health insurance coverage, with 94.8% of the unweighted sample and 94.0% of the weighted population reporting coverage (Table 1).
Racial and ethnic composition showed that Non-Hispanic Whites comprised the majority (unweighted: 60.0%; weighted: 59.2%), followed by NonHispanic Blacks (unweighted: 11.1%; weighted: 11.0%). Approximately 40% of both the weighted and unweighted samples were identified as belonging to a minority population.
Table 2 stratifies the weighted patient characteristics by racial and ethnic groups. Non-Hispanic Black and Mexican American individuals had the highest median hs-CRP values (3.93 mg/L and 3.64 mg/L, respectively), while education levels and insurance coverage
varied significantly across groups. Insurance coverage was highest among Non-Hispanic Whites (61.0%) and lowest among Mexican Americans (7.0%). The elevated hs-CRP levels observed in the Non-Hispanic Black population contributed substantially to the overall trend of higher systemic inflammation seen within racial and ethnic minority groups.
Hs-CRP levels were compared between racial and ethnic minority and non-minority populations using survey weighted data in Table 3. In the unweighted sample (not represented in Table 3), minority populations had significantly higher hs-CRP levels (p=0.03). In the weighted sample, although median hs-CRP levels were higher in minority groups (2.99 Variables
Table 1. Patient Characteristics
Covered by Insurance
Table 2. Patient Characteristics by Race and Ethnicity, Presented with Row Percentages
vs. 1.94 mg/L in non-minority), the difference was not statistically significant (p=0.09). Within the subgroup analysis, CRP levels only seemed to significantly differ in the combined subgroup of high school graduates and individuals with some college or AA degree. In this subgroup, the minority group had higher median CRP levels than the non-minority group (3.10 versus 2.10; p-value = 0.0048).
Discussion
Overall, our study found that persons identifying as Non-Hispanic Black had higher median hs-CRP levels compared to non-minority population groups. While these differences were statistically significant in the unweighted analysis, they were not significant after applying survey weights (p = 0.09), though a clear trend persisted. These results, when considered with previous studies that linked elevated hs-CRP to impaired glucose regulation, allow us to hypothesize that Non-Hispanic Black populations may have higher impairments of glucose regulation (6–7,16).
Interestingly, in our subgroup analysis our study found significant differences in hs-CRP levels among
people who had completed high school, attended some college, or received an AA degree. Within this group, minority population groups had significantly higher median hs-CRP levels than non-minority population groups. This finding suggests that other social factors may influence and contribute to systemic inflammation even among those with similar education levels, reinforcing the importance of considering how race and other social factors influence health outcomes.
By exploring how hs-CRP levels vary among diverse racial and ethnic groups rather than relying solely on broad demographic categorizations, our findings emphasize the importance of examining prediabetes through the lens of health equity. Social determinants of health such as socioeconomic status (SES), food insecurity, chronic stress, residential segregation, environmental exposure, and unequal healthcare access may contribute to systemic inflammation in these populations, producing biomarker-level disparities, specifically hs-CRP disparities (10–14). For instance, Muscatell et al. demonstrated this in their meta-analysis which found a significant association between lower SES and elevated levels of
Table 3. Median CRP Levels (mg/L) by Race/Ethnicity, Stratified by Sex, Insurance Coverage, and Education Level. Median and interquartile range (IQR) values are reported for minority (N = 270) and non-minority (N = 405) participants. P-values reflect group differences within each stratum.
inflammatory markers (17). These findings underscore the need to further explore upstream structural factors and to implement culturally tailored public health and clinical interventions (18).
Rationale for incorporating social determinants of health into hs-CRP interpretation is further supported by recent work from Modisette et al., who conducted a stratified analysis of hs-CRP predictors across racial and ethnic groups using NHANES data. Their study identified consistent predictors of elevated hs-CRP levels including increased BMI and female sex across most racial and ethnic groups, while also highlighting the influence of socioeconomic factors like poverty, particularly increased among NonHispanic Asian populations. These results reinforce the importance of interpreting hs-CRP levels through a race-conscious and equity-informed lens, as they
suggest that systemic inflammation is not solely biologically driven but also shaped by social determinants of health. Incorporating these insights strengthens the rationale for using hsCRP as a biomarker in prediabetes risk stratification, while emphasizing the need for clinical interpretation that accounts for social and environmental exposures (19).
Considering these disparities, our findings also support the potential utility of hsCRP testing as a widely accessible and affordable biomarker for identifying individuals with prediabetes who may be at greater risk of progression to overt Type 2 diabetes. This is supported by findings from Cheng et al., who demonstrated in a prospective cohort study that elevated hs-CRP levels were significantly associated with increased risk of progression from prediabetes to Type 2 diabetes among middle-aged and older adults (20). However, the clinical interpretation of hs-CRP levels should be approached with caution, integrating frameworks that consider social determinants of health to contextualize differences, rather than solely attributing variations to race (21–22). Such interpretation is essential to avoid misclassification or underestimation of risk, particularly in historically marginalized minority populations (23–24).
Despite the strengths of this study, including use of a nationally representative NHANES dataset and stratification by race/ethnicity, several limitations must be acknowledged. First, the NHANES dataset is cross-sectional in nature, which limits causal inference. Second, we did not control potential confounding variables such as smoking status, physical activity, obesity (BMI), or recent infections, all of which can elevate hs-CRP independently of glycemic status (25). These uncontrolled confounders may have contributed to the attenuation of statistical significance in the weighted analysis, masking underlying relationships.
Future studies should adopt longitudinal designs to track changes in hs-CRP and glycemic control over time, and mechanistic studies should be pursued to better understand the interplay between systemic inflammation and glucose metabolism. Additionally,
incorporating direct measures of socioeconomic status (e.g., income, education, occupation) and environmental exposures would strengthen our ability to delineate the root causes of observed disparities and guide equitable interventions.
Conclusion
This study highlights racial and ethnic differences in hs-CRP levels among individuals with prediabetes, with higher levels observed in minority populations, specifically Non-Hispanic Black populations. These differences persisted even after accounting for demographic factors such as insurance status or education level, underscoring the impact of social determinants of health on systemic inflammation. These findings emphasize the importance of critically interpreting hs-CRP levels with a perspective that acknowledges health disparities; an essential skill to accurately assess patient risk and to develop inclusive, patient-centered intervention strategies.
Disclosures
The authors have no financial disclosures or conflicts of interest to disclose.
Acknowledgments
None.
References
1. Centers for Disease Control and Prevention. Diabetes Data & Research. Atlanta, GA: U.S. Department of Health & Human Services; 2024. Available from: https://www.cdc.gov/diabetes/ php/data-research/index.html
2. Tabák AG, Herder C, Rathmann W, Brunner EJ, Kivimäki M. Prediabetes: a high-risk state for diabetes development. The Lancet. 2012 Jun;379(9833):2279–90.
3. Cai X, Zhang Y, Li M, Wu JHY, Mai L, Li J, et al. Association between prediabetes and risk of all cause mortality and cardiovascular disease: updated meta-analysis. BMJ. 2020;370:m2297.
4. Cefalu WT, Buse JB, Tuomilehto J, Fleming GA, Ferrannini E, Gerstein HC, et al. Update and Next Steps for Real-World Translation of Interventions for Type 2 Diabetes Prevention: Reflections From a Diabetes Care Editors’ Expert Forum. Diabetes Care. 2016 Jun 21;39(7):1186–201.
5. Knowler WC, Barrett-Connor E, Fowler SE, Hamman RF, Lachin JM, Walker EA, et al. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med. 2002;346(6):393–403.
6. Pradhan AD, Manson JE, Rifai N, Buring JE, Ridker PM. C-reactive protein, interleukin 6, and risk of developing type 2 diabetes mellitus. JAMA. 2001 Jul 18;286(3):327–34.
7. Stanimirovic J, Radovanovic J, Banjac K, Obradovic M, Essack M, Zafirovic S, et al. Role of C-reactive protein in diabetic inflammation. Mediators Inflamm. 2022.
8. Kato K, Otsuka T, Saiki Y, Kobayashi N, Nakamura T, Kon Y, et al. Association between elevated C-reactive protein levels and prediabetes in adults, particularly impaired glucose tolerance. Can J Diabetes. 2019;43(1):40–45.e2.
9. Hill-Briggs F, Fitzpatrick SL. (2023). Overview of Social Determinants of Health in the Development of Diabetes. Diabetes care, 46(9), 1590–1598.
10. Ghule A, Kamble TK, Talwar D, Kumar S, Acharya S, Wanjari A, et al. Association of serum high sensitivity C-reactive protein with pre-diabetes in rural population: a two-year cross-sectional study. Cureus. 2021;13(10):e19088.
11. Bailey SR, O'Malley JP, Gold R, Heintzman J, Likumahuwa S, DeVoe JE. (2017). Receipt of diabetes preventive services differs by insurance status at visit. American Journal of Preventive Medicine, 52(6), 735–739.
12. Brook RD, Rajagopalan S, Pope CA, Brook JR, Bhatnagar A, Diez-Roux AV, et al. (2010). Particulate matter air pollution and cardiovascular disease: An update to the scientific statement from the American Heart Association. Circulation, 121(21), 2331–2378.
13. Kramer MR, Hogue CR. (2009). Is segregation bad for your health? Epidemiologic Reviews, 31(1), 178–194.
14. Seligman HK, Laraia BA, Kushel MB. (2010). Food insecurity is associated with chronic disease among low-income NHANES participants. The Journal of Nutrition, 140(2), 304–310.
15. Centers for Disease Control and Prevention (CDC), National Center for Health Statistics (NCHS). National Health and Nutrition
Examination Survey Data. Hyattsville, MD: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention; 2021. Available from: https://www.cdc.gov/nchs/ nhanes/index.htm,
16. Shankar A, Li J. Positive association between highsensitivity C-reactive protein level and diabetes mellitus among US non-Hispanic black adults. Exp Clin Endocrinol Diabetes. 2008;116(8):455–60.
17. Muscatell KA, Brosso SN, Humphreys KL. (2020). Socioeconomic status and inflammation: A metaanalysis. Molecular Psychiatry, 25(9), 2189–2199.
18. Marques-Vidal P, Bochud M, Bastardot F, Luscher T, Ferrero F, Gaspoz JM, et al. Association between inflammatory and obesity markers in a Swiss population-based sample (CoLaus Study). Obes Facts. 2012;5(5):734–44.
19. Modisette R, Yockey A, Fulda K. Do sociodemographic, health behaviors, and clinical factors predict levels of hs-CRP stratified by race and ethnicity? Ann Fam Med. 2024;22(Suppl 1):6727.
20. Cheng ZJ, Wang YF, Jiang XY, Ren WY, Lei SF, Deng FY, et al. High sensitivity C-reactive protein and prediabetes progression and regression in middleaged and older adults: A prospective cohort study. J Diabetes Investig. 2024;15(1):78–86.
21. Nagar SD, Conley AB, Sharma S, Rishishwar L, Jordan IK, Mariño-Ramírez L. Comparing genetic and socioenvironmental contributions to ethnic differences in C-reactive protein. Front Genet. 2021 Oct 18;12:738485.
22. Kaiser Family Foundation (KFF). Use of Race in Clinical Diagnosis and Decision-Making: Overview and Implications. KFF Issue Brief. 2021 Oct 13. Available from: https://www.kff.org/racial-equityand-health-policy/issue-brief/use-of-race-inclinical-diagnosis-and-decision-making-overviewand-implications/
23. National Academies of Sciences, Engineering, and Medicine. Rethinking Race and Ethnicity in Biomedical Research. Washington, DC: The National Academies Press; 2025.
24. Farmer HR, Wray LA, Haas SA. Race, gender, and socioeconomic variations in C-reactive protein using the Health and Retirement Study. J Gerontol B Psychol Sci Soc Sci. 2021 Feb 17;76(3):583–95.
25. Banait T, Wanjari A, Danade V, Banait S, Jain J. (2022). Role of High-Sensitivity C-reactive Protein (Hs-CRP) in Non-communicable Diseases: A Review. Cureus, 14(10), e30225.
Pause and Recharge: Wellness for Medical Students
Cameron Jones1,2,†,‡, Christopher D. Manko1,2,†,‡, Lakshmi Ilango1,2,†, Kevin Xu1,2,†, Dylan Quinn1,2,†, and John Pamula¹
¹Guthrie Robert Packer Hospital, Sayre, PA 18840
²Geisinger College of Health Sciences, Scranton, PA 18509
†Doctor of Medicine Program
‡Authors contributed equally
Correspondence: cmanko@som.geisinger.edu
Abstract
Background: Medical students, especially in their core clerkship/third year of medical school, undergo a significant amount of stress which can impact their studies and ability to care for patients. This study investigates how certain wellness activities can affect student wellness during their core clerkship year.
Methods: The Medical Student Stressor Questionnaire was utilized to assess student wellness. This survey assesses 6 domains of wellness which are academic-related stressors, interpersonal & intrapersonal-related stressors, teaching and learningrelated stressors, social-related stressors, drive & desire-related stressors, and group activities-related stressors. The survey was distributed about halfway into the core clerkship year, followed by numerous wellness activities, and a repeat survey near the end of the year.
Results: Academic stressors saw a significant decrease post-intervention, indicating a notable increase in that domain of wellness. Other domains did not see a significant increase or decrease.
Conclusion: Overall, student wellness events may have an impact on perceived student wellness. Given that wellness is dynamic, changes may have occurred on a smaller day-to-day level as well. Future work should focus on long-term changes as well as other health care student burnout.
Introduction
Medical students often face a number of responsibilities during their studies, especially in their core clerkship/third year. In this transition from student to provider, they are expected to learn a large amount of information quickly and effectively, complete assignments, work with patients and medical teams in numerous new settings and may commit
to extracurricular activities such as research and community service. Medical students in general often experience high levels of stress, burnout, and mental health challenges secondary to demanding academic schedules and clinical responsibilities (1). A systematic review examining numerous multi-institutional studies found that half or more of medical students experience burnout, often characterized by emotional exhaustion, depersonalization, and a reduced sense of personal accomplishment (2). The consequences of stress and burnout in medical students are farreaching, and include mental health disorders such as depression and anxiety, substance abuse, and suicidal ideation (3). One meta-analysis published in JAMA found that the prevalence of depression or depressive symptoms among medical students was 27.2%, and the prevalence of suicidal ideation was 11.1% (4). Additionally, burnout can negatively affect students' academic performance and increase likelihood of dropping out (5). In clinical practice, prior work has shown that physician burnout has been associated with medical errors (6).
These alarming statistics underscore the urgent need for interventions supporting medical student wellness. Institutions such as medical schools and hospitals play an important role here in their development of different curricular events/activities, facilitation of academic culture, and other opportunities to improve wellness. When discussing well-being, it is important to consider the physical, mental, emotional, and social components of wellness and how they affect student performance. This study aims to understand how different wellness activities affect medical student wellness over the course of their core clerkship year.
Methods
This study focused on medical students from Geisinger Commonwealth School of Medicine (Class
of 2026) completing clinical rotations at the Guthrie Robert Packer Hospital which began in February and finished in the clinical space in December. To assess stress levels, we used the Medical Student Stressor Questionnaire (MSSQ), developed at the University of Science Malaysia (7). This tool evaluates stress intensity across 6 domains: academic-related stressors, interpersonal & intrapersonal-related stressors, teaching and learning-related stressors, social-related stressors, drive & desire-related stressors, and group activities-related stressors. The survey was first conducted on Aug. 15, 2024, before implementing 3 wellness interventions, and again on Nov. 20, 2024, after all 3 interventions had taken place. Participation was kept anonymous, and as such preand post-intervention data was not matched.
To help students manage the stress of exams, grading pressure, and heavy coursework, a quiet group study session was organized (event 1). This provided a structured, supportive space where students could study together, share challenges, and stay motivated. The goal was to foster camaraderie and maintain focus leading up to the National Board of Medical Examiners (NBME) clinical knowledge shelf exams. To determine the activities for the remaining 2 wellness interventions, a student poll was conducted. While these events did not directly target exam stress, they aimed to build a sense of community and provide relief from academic pressures. As part of our physical and social wellness initiative, the authors organized a fall hike at Mount Pisgah State Park (event 2). To support social and emotional well-being, the authors planned a trip to the Corning Glass Museum (event 3). In addition to MSSQ survey results, event attendance served as another key metric for evaluating the effectiveness of our interventions. Statistical significance for interventions was tested with an unpaired two-tailed sample t test.
Results
Out of 22 total students at the campus, 15 students completed the pre-intervention MSSQ, and 16 students completed the post-intervention MSSQ. Five students that were a part of conducting the study were not eligible to participate in the survey. One question (question 26) was not included in the surveys distributed.
Initial MSSQ results from pre-intervention data showed that students experienced the most stress
in the following domains, from greatest stress to least stress (with average score out of 4): academicrelated stressors (2.17), teaching and learning related stressors (1.68), group activities related stressors (1.63), social stressors (1.30), drive/desire related stressors (0.58), and interpersonal stressors (0.44). Of the items included in the questionnaire, the most notable stress was attributed to the following items (>= average score of 2.5): tests/examinations (3.07), self-expectation (3.07), quota systems in exams (2.64), receiving poor grades (2.57), uncertainty regarding expectations (2.5), and the large amount of content to be learned (2.5). The lowest pre-intervention questions were verbal or physical abuse by students, teachers, and personnel scoring respectively at 0.21, 0.21, and 0.14.
Participation in events varied, as 12 students attended the first event of quiet study sessions, 6 students attended the second event of outdoor hike, and 10 students attended the final event of the museum trip. However, these numbers represent the total number of participants, which include those students who could not participate in the questionnaire surveys.
Based on post-intervention MSSQ data, changes were seen in each domain of stressors (Table 1). Reductions in stress were observed in the following domains (average score, percent decrease): academic-related stressors (1.76, -18.88%), teaching/related stressors (1.46, -12.64%), social stressors (1.01, -22.28%), and group activity related stressors (1.61, -1.47%).
Increases in stress were observed in 2 domains (average score, percent increase): interpersonal stressors (0.5, +12.50%) and drive/desire-related stressors (0.69, +18.99%). Only the academic stressor domain was found to have a significant change from baseline survey (p = 0.00063). Regarding verbal and physical abuse from students, faculty, and personnel, they continued to be low with post-intervention scores of 0.06, 0.5, and 0.25 respectively. Based on post-intervention data, academic-related stressors remained the highest cause of stress among the domains covered on the MSSQ.
Discussion
This study investigated stress levels among Geisinger Commonwealth School of Medicine medical students at the Guthrie Robert Packer Hospital after intervention with wellness events. Academic stressors found a significant decrease post-intervention.
Compiled pre- vs. post-intervention medical student stressor questionnaire
It is well-known that in general, health care workers including physicians experience burnout (6,8–17). The authors feel it is important to provide understanding on the typical workload and expectations of a third year (or core clerkship) student to best understand where academic stress may stem from. In this formative year, a student rotates across at 7 different core specialties (internal medicine, neurology, psychiatry, surgery, obstetrics/gynecology, pediatrics, and family medicine). This is their first time introduced to most, if not all, of these fields clinically, and as such means navigating a brand-new environment every couple of months. Students are expected to work as providers, and thus take on patients, assessing them, creating differentials and treatment plans, and presenting this work to attendings, nurses, and other involved staff. Additionally, these students must learn new clinical content for each course in preparation of their end of rotation shelf exam. Information is broad with thousands of facts per rotation, including specific treatment guidelines for numerous diseases, screenings, risk factors, and best-next steps in diagnosis and treatment plans. Additionally, students are expected to be up to date on their patients and often must read into additional detail regarding their conditions, as well as ensuring a strong understanding of any procedures the patient may undergo during their hospitalization. On top of that, students are expected to significantly improve their clinical skills throughout the year such as auscultating pathologies, suturing, laparoscopic camera driving, and catheter insertions, to name a few. Furthermore, students are also expected to complete numerous rotation specific assignments to be assessed and graded by school faculty. Grading per rotation involves doing well in provider evaluations, NBME shelf exams, and
assignments. Thus, students must strive to excel in all these components. As one can imagine, this is an extensive workload, pushing students academically, and limiting time for other components of wellness such as social well-being.
Students experienced a significant decrease in stress in the academic stressor domain, with initial preinterventions stressor questions averaging high in tests/grades. Previous literature indicates repeat testing, inadequate preparedness, and time constraints negatively impact student perception of academic stress (1). Decreased reported academic stress in this study may be a factor of increased preparedness resulting from creation of a supportive study environment with the first event. Alternatively, this may have been due to increased clinical knowledge later in the year compared to earlier, as well as greater familiarity with testing processes over time.
Post-intervention results indicated that stress decreased non-significantly in domains of teaching/ learning, social, and group activity stressors, and increased non-significantly in the areas of interpersonal and drive/desire stressors. This indicates that our interventions may not have had a significant impact on wellness domains outside of academics. Non-academic stressors contribute substantially to student well-being or lack thereof and can be difficult to address from an institutional standpoint. Highly individual factors such as familial relationships, socioeconomic status, and physical health can contribute to the mental health and well-being of students, and are often interconnected (18, 19).
From a campus culture standpoint, the lowest preintervention question scored was verbal or physical abuse by other students, faculty, and personnel.
Table 1. Medical Student Stressor Questionnaire Results
While slight increases occurred post-intervention, they were still all below 1. Increases may have been due to poorly perceived interactions during rotations and/or variability amongst participants completing the surveys. Overall, these low scores indicate that the campus emphasized student safety during this year. The Guthrie campus offers student support from campus-specific faculty as well as from overarching school faculty in case of any abusive behavior.
The authors feel the events provided, while not showing statistical significance, still had an individuallevel impact. By organizing the events, students were provided a carved-out time where they could actively focus on wellness, connecting with other students and the local community. Wellness is a dynamic process, that can change both short term and long term. As such, wellness may have increased during the events themselves, and fluctuated afterwards depending on social, emotional, and academic influences. The lack of statistical significance could be attributed to the low sample size. Prior work has shown that other interventions, such as yoga, also had a positive effect on student wellness, further supporting the implementation of wellness activities (20).
This study gives some insight into how our activities may have impacted student wellness. A strength of this study is focusing on multiple domains contributing to wellness as well as different targeted interventions. Limitations to our results include the small sample size, single student type, incomplete survey results (39/40 questions), and single location included. Our study demonstrates some potential effectiveness of interventions aimed at reducing stress levels in students. Future students may benefit from regular surveys at shorter time intervals and closer to events to better evaluate the dynamic process of wellness. Additional work can focus on other targeted interventions to reduce stress in more impacted areas, longer time periods for study, and mixed-methods studies to identify individual themes that could impact wellness. Further studies should explore mitigating stress in medical students at other schools, interns, residents, nursing students, and pharmacy students.
Conclusion
Investigation of medical student wellness revealed that there was a significant increase in wellness from the academic stressors domain for students from beginning to end of the study, possibly due to student
interventions that increased the supportiveness of the environment. Future work should investigate other interventions to decrease stress and investigate wellness for other health care students/providers.
Disclosures
The authors of this study benefited from the wellness activities described above, which was funded by Geisinger Commonwealth School of Medicine. No other financial disclosures are noted.
Acknowledgments
The authors would like to thank Geisinger Commonwealth School of Medicine for their classroom support of this quality improvement project. Their efforts to organize this have enabled the authors to engage in this work during the core clerkship year.
References
1. Weber J, Skodda S, Muth T, Angerer P, Loerbroks A. Stressors and resources related to academic studies and improvements suggested by medical students: a qualitative study. BMC Med Educ [Internet]. 2019 Dec [cited 2025 May 27];19(1):312. Available from: https://bmcmededuc.biomedcentral.com/ articles/10.1186/s12909-019-1747-z
2. IsHak W, Nikravesh R, Lederer S, Perry R, Ogunyemi D, Bernstein C. Burnout in medical students: a systematic review. Clin Teach [Internet]. 2013 Aug [cited 2025 May 27];10(4):242–5. Available from: https:// onlinelibrary.wiley.com/doi/10.1111/tct.12014
3. Glauser W. Medical schools addressing student anxiety, burnout and depression. Can Med Assoc J [Internet]. 2017 Dec 18 [cited 2025 May 27];189(50):E1569–70. Available from: http:// www.cmaj.ca/lookup/doi/10.1503/cmaj.109-5516
4. Rotenstein LS, Ramos MA, Torre M, Segal JB, Peluso MJ, Guille C, et al. Prevalence of Depression, Depressive Symptoms, and Suicidal Ideation Among Medical Students: A Systematic Review and Meta-Analysis. JAMA [Internet]. 2016 Dec 6 [cited 2025 May 27];316(21):2214. Available from: http://jama.jamanetwork.com/ article.aspx?doi=10.1001/jama.2016.17324
5. Prendergast M, Cardoso Pinto AM, Harvey CJ, Muir E. Burnout in early year medical students: experiences, drivers and the perceived value of a reflection-based intervention. BMC Med Educ [Internet]. 2024 Jan 3 [cited 2025 May 28];24(1):7. Available from: https://doi. org/10.1186/s12909-023-04948-0
6. Li CJ, Shah YB, Harness ED, Goldberg ZN, Nash DB. Physician Burnout and Medical Errors: Exploring the Relationship, Cost, and Solutions. Am J Med Qual Off J Am Coll Med Qual. 2023 Aug 1;38(4):196–202.
7. Yusoff MSB, Rahim AFA, Yaacob. MJ. The development and validity of the Medical Student Stressor Questionnaire (MSSQ). ASEAN J Psychiatry. 2010 Jun;11(1).
8. Cohen C, Pignata S, Bezak E, Tie M, Childs J. Workplace interventions to improve well-being and reduce burnout for nurses, physicians and allied healthcare professionals: a systematic review. BMJ Open. 2023 Jun 29;13(6):e071203.
9. Hodkinson A, Zhou A, Johnson J, Geraghty K, Riley R, Zhou A, et al. Associations of physician burnout with career engagement and quality of patient care: systematic review and meta-analysis. BMJ. 2022 Sep 14;378:e070442.
10. Lacy BE, Chan JL. Physician Burnout: The Hidden Health Care Crisis. Clin Gastroenterol Hepatol Off Clin Pract J Am Gastroenterol Assoc. 2018 Mar;16(3):311–7.
11. Macaron MM, Segun-Omosehin OA, Matar RH, Beran A, Nakanishi H, Than CA, et al. A systematic review and meta analysis on burnout in physicians during the COVID-19 pandemic: A hidden healthcare crisis. Front Psychiatry. 2022;13:1071397.
12. Ortega MV, Hidrue MK, Lehrhoff SR, Ellis DB, Sisodia RC, Curry WT, et al. Patterns in Physician Burnout in a Stable-Linked Cohort. JAMA Netw Open. 2023 Oct 2;6(10):e2336745.
13. Patel RS, Sekhri S, Bhimanadham NN, Imran S, Hossain S. A Review on Strategies to Manage Physician Burnout. Cureus. 2019 Jun 3;11(6):e4805.
14. Rothenberger DA. Physician Burnout and WellBeing: A Systematic Review and Framework for Action. Dis Colon Rectum. 2017 Jun;60(6):567–76.
15. Shin P, Desai V, Conte AH, Qiu C. Time Out: The Impact of Physician Burnout on Patient Care Quality and Safety in Perioperative Medicine. Perm J. 2023 Jun 15;27(2):160–8.
16. Taranu SM, Ilie AC, Turcu AM, Stefaniu R, Sandu IA, Pislaru AI, et al. Factors Associated with Burnout in Healthcare Professionals. Int J Environ Res Public Health. 2022 Nov 9;19(22):14701.
17. West CP, Dyrbye LN, Shanafelt TD. Physician burnout: contributors, consequences and solutions. J Intern Med. 2018 Jun;283(6):516–29.
18. Jeyapalan T, Blair E. The Factors Causing Stress in Medical Students and their Impact on Academic Outcomes: A Narrative Qualitative Systematic Review. Int J Med Stud [Internet]. 2024 Jul 9 [cited 2025 May 27];12(2):195–203. Available from: https://ijms.pitt.edu/IJMS/article/view/2218
19. Mirza AA, Baarimah H, Baig M, Mirza AA, Halawani MA, Beyari GM, et al. Academic and non-academic life stressors and their impact on psychological wellbeing of medical students. AIMS Public Health. 2021;8(4):563–80.
20. Thompson C, Meller J, Naqvi M, Adesanya O, Vasylyeva TL. Restorative yoga therapy for thirdyear medical students in pediatrics rotation: Working to improve medical student well-being. J Educ Health Promot. 2023;12:76.
Modification of Dialysis for Management of Elevated Intraocular Pressure: A Case Report
Giovanni
Battistini1† and Daniel Upton2
1Geisinger College of Health Sciences, Scranton, PA 18509
2Geisinger Department of Ophthalmology, Geisinger Health System, Danville, PA 17822
†Doctor of Medicine Program
Correspondence: gbattistini@som.geisinger.edu
Abstract
In this report, we discuss the case of a chronically ill 74-year-old male who developed severe intraocular pressure (IOP) elevations following dialysis treatments after intraocular lens (IOL) exchange. He presented with elevated IOP at 1-week post-op, which was treated topically and orally with success. However, he soon experienced recurrent episodes of severe IOP spikes in the 30–50 mmHg range accompanied by severe pain, nausea, and vomiting. After multiple such episodes, the patient was admitted to the hospital for close observation, where it was found that his IOP severely increased soon after dialysis. After discussion with nephrology, increasing the duration and weekly number of dialysis sessions led to significantly improved IOPs. The patient has done well since that time and was able to return to normal dialysis length and schedule without issue. This case sheds light on possible IOP elevations post dialysis, especially if aqueous outflow is acutely impeded.
Introduction
Ocular dialysis disequilibrium describes a proposed phenomenon whereby intraocular pressure increases during hemodialysis due to disproportionate changes in the osmolality of aqueous humor as opposed to the blood. This imbalance in osmolality is thought to disrupt the equilibrium between production and outflow of aqueous humor in the eye, which determines the intraocular pressure (IOP) (1). Elevations in intraocular pressure can lead to visionthreatening complications such as acute angle closure glaucoma, both in patients with existing glaucoma and those without. Furthermore, other risk factors or acute conditions can contribute to IOP elevations and may also need to be controlled. In the case presented below, a pars plana vitrectomy (PPV) and intraocular lens (IOL) exchange complicated by hyphema and vitreous hemorrhage likely exacerbated the issue,
as both can lead to increased ocular inflammation with blockage of the trabecular meshwork. This case presents an interesting etiology and potential solution for intraocular pressure elevation in an end-stage renal disease (ESRD) patient undergoing hemodialysis.
Case Presentation
A 74-year-old male presented for partially dislocated posterior chamber intraocular lens (PCIOL) in his left eye, which was his only functional eye. He suffered from multiple chronic illnesses, including Merkel cell carcinoma of the right eyelid and right parotid gland (currently in remission), blindness of the right eye due to complications from radiation treatments, chronic obstructive pulmonary disease, atrial fibrillation, gastroesophageal reflux disease, obstructive sleep apnea, coronary artery disease, and ESRD treated with dialysis for several years. Of note, the patient had not had any previous glaucoma or ocular hypertension issues. He subsequently underwent uncomplicated pars plana vitrectomy with IOL exchange with a scleral sutured IOL in the left eye. One day postoperatively the patient had a visual acuity of 20/40, but by day 3 post-operation, his vision had declined. One week post-operation, the patient’s vision was restricted to perception of hand motion only (HMO) with severe pain, and significant vitreous hemorrhage, hyphema, and small nasal choroidal effusion noted on examination. His measured intraocular pressure was 39 mmHg, far exceeding the generally accepted “normal” range of 10 to 21 mmHg. We were able to regain control of the IOP with administration of topical brimonidine, dorzolamide, timolol, Xalatan®, and oral Diamox®. With the patient’s IOP acutely managed, the plan at that time was to observe in the outpatient setting for spontaneous resolution of the previously noted hyphema and vitreous hemorrhage. However, during this observational period, he returned several times for urgent appointments due to severe pain,
in addition to nausea and vomiting. At this time, he was noted to have severe IOP elevations, with maximum recorded IOP reaching 53 mmHg. Given the unclear etiology of the patient’s symptoms, he was admitted to the hospital for pain control and very close observation. Over the course of his inpatient stay, his IOP was checked frequently and noted to only severely spike in the evenings after dialysis. He was evaluated multiple times while on dialysis and never had significant elevation of IOP during treatment itself. At that time, we surmised that fluid shifts were a possible etiology of his symptoms, especially with vitreous humor now absent post PPV. Lengthy discussions with nephrology led to a plan to prolong dialysis treatment times, running the dialysis machine as slowly as possible to hopefully slow electrolyte and fluid shifts into the eye, with the goal of stopping the IOP spikes following dialysis sessions. The patient responded very well to this modified, prolonged dialysis protocol. His IOP never exceeded 24 mmHg and was typically in the middle to upper teens. Furthermore, with the addition of a fourth dialysis day per week, his IOP has been in the middle teens, even with discontinuation of Diamox. We have observed consistent IOP maintenance for months on this regimen. The patient additionally underwent anterior and posterior chamber wash out to treat his persistent vitreous hemorrhage. He is currently doing very well and seeing well from the previously affected eye. He has not had any significant IOP elevations since the alteration of his dialysis protocol.
Discussion
This case provides further evidence for the effect of hemodialysis on IOP, as well as a potential solution for managing IOP increases associated with said treatment. As mentioned above, one of the primary factors thought to contribute to changes in IOP is the changes in fluid osmolarity inherent to dialysis itself. Specifically, as molecules such as urea are removed from the blood, the osmolality of fluids in the body decreases. However, the decrease in aqueous humor osmolality is thought to be disproportional to the decrease in plasma osmolality, thus leading to hypertonicity of the aqueous humor. This hypertonicity leads water to move into the aqueous space (1). Additionally, a population-based study conducted in Malaysia found that the presence of chronic kidney disease may itself be another risk factor for elevated IOP due to impaired aqueous outflow (2).
Regarding management of IOP elevations during dialysis, the available body of literature is somewhat limited. While several possible solutions have been reported in prior studies, there does not appear to be a widely used protocol at this time. Prior suggested solutions have included administration of intravenous glucose, pan-retinal photocoagulation for neovascular glaucoma, and surgical interventions such as Ahmed valve implantation (1, 3–5). A case report by Maja et al. did note an attempt at decreasing dialysis rate as an intervention but ultimately resorted to placement of an Ahmed valve when dialysis modification was ineffective (3).
Of note, current literature related to this topic yields conflicting answers regarding the effects that dialysis may have on IOP. A 2015 study found that IOP decreased significantly after hemodialysis treatment (6). Conversely, literature reviews conducted by Levy et al. and Liakopoulos et al., respectively, were unable to demonstrate a clear correlation between hemodialysis and changes in IOP (7, 8). More recent studies conducted by Jung et al. and Wang et al. suggest that different patient responses in IOP may be due to individual differences in ocular structures such as anterior chamber angle and corneal thickness (9, 10).
While the current literature is divided regarding changes in IOP following dialysis, we believe that this case sheds some light on patient factors that may contribute to elevated IOP. We also hope that the case can provide a framework for effective treatment of elevated IOP in patients who may present similarly. A modification of dialysis protocol as described in this case — increasing treatment frequency and duration with a decreased dialysis rate — could prove invaluable for maintaining eye health in patients with various comorbidities who are also receiving concurrent ophthalmologic care.
Conclusion
In conclusion, we acknowledge that more definitive research is needed regarding the effects of dialysis on IOP, given the lack of consensus in the current literature. Additionally, there is much to explore in terms of the broader utility of dialysis modification in managing possible IOP increases, especially considering the scarcity of recent studies regarding this topic. It is our hope that cases such as this can contribute to further development of a flexible dialysis protocol that can accommodate patients with various
comorbidities, while minimizing risk for exacerbation or development of glaucoma.
Disclosures
Nothing to disclose.
Acknowledgments
We would like to thank Tom Urosevich, OD, for contributing his time, effort, and expertise in reviewing and suggesting edits for this manuscript.
References
1. Lippold CL, Kalarn SP, Swamy RN, Patel AM. Ocular dialysis disequilibrium—Management of intraocular pressure during hemodialysis of open angle glaucoma: A case report and review of the literature. Hemodialysis International. 2019;23(3).
2. Nongpiur ME, Wong TY, Sabanayagam C, Lim SC, Tai ES, Aung T. Chronic Kidney Disease and Intraocular Pressure. Ophthalmology. 2010;117(3):477–83.
3. Maja AK, Lewis CY, Steffen E, Zegans ME, Graber ML. Increased Intraocular Pressure During Hemodialysis: Ocular Dialysis Disequilibrium. Kidney Medicine. 2022;4(9):100526.
4. Frezzotti P, Menicacci C, Bagaglia SA, Mittica P, Toto F, Motolese I. Management of intraocular pressure elevation during hemodialysis of neovascular glaucoma: a case report. BMC Ophthalmol. 2016;16(1):23.
5. Saritas T, Koutsonas A, Walter P, Floege J, Krüger T. Management of Intraocular Hypertension During Hemodialysis by Intravenous Glucose Administration. American Journal of Kidney Diseases. 2014;63(3):500–2.
6. Chelala E, Dirani A, Fadlallah A, Slim E, Abdelmassih Y, Fakhoury H, et al. Effect of hemodialysis on visual acuity, intraocular pressure, and macular thickness in patients with chronic kidney disease. Clin Ophthalmology. 2015;9:109–14.
7. Levy J, Tovbin D, Lifshitz T, Zlotnik M, Tessler Z. Intraocular pressure during haemodialysis: a review. Eye. 2005;19(12):1249–56.
8. Liakopoulos V, Demirtzi P, Mikropoulos DG, Leivaditis K, Dounousi E, Konstas AGP. Intraocular pressure changes during hemodialysis. Int Urol Nephrol. 2015;47(10):1685–90.
9. Jung JW, Yoon MH, Lee SW, Chin HS. Effect of hemodialysis (HD) on intraocular pressure, ocular surface, and macular change in patients with chronic renal failure. Graefes Arch Clin Exp Ophthalmol. 2013;251(1):153–62.
10. Wang F, Wang L, Yu Z, Chen N, Wang D. Effects of Hemodialysis on Intraocular Pressure and Ocular Biological Parameters in Different Angle Structures. Disease Markers. 2022;2022(1):9261653.
Glioblastoma’s Effect on Brain Circuitry
Iletou B. Ehinnou1†‡ and Anthony J. Bragoli2‡
¹Geisinger College of Health Sciences, Scranton, PA 18509
²University of Pittsburgh, Department of Immunology and Microbiology, Pittsburgh, PA 15260
Research over the past few decades has reshaped our understanding of glioblastoma. Rather than existing independently in the brain and physically disrupting brain function, glioblastoma integrates and modifies neuron communication and vasculature to its advantage. Glioblastomas can leverage neuronal signaling, respond to synaptic activity, and remodel their microenvironment to promote their growth. This review consolidates findings from electrophysiology, blood oxygenation level-dependent functional magnetic resonance imaging (BOLD fMRI), magnetoencephalography (MEG), transcriptomic, and metabolic studies to highlight many of the changes that occur in the host brain. For the construction of this review, an electronic search was conducted using studies that fit the following inclusion criteria: (1) The study focused on glioblastomas, (2) the study looked at brain circuitry, cognitive function, or neuronal activity, and (3) the full text of the study was within the two reviewers’ institutional databases. Studies that did not meet these criteria were excluded from this review.
Introduction
It is well established that the brain is one of the most complex, functionally diverse yet precise organs in the human body (2, 22, 27, 39). It is composed of billions of neurons that collectively coordinate sensory perception, emotional responses, higherorder cognition, and motor movements. This system is often referred to as brain circuitry. While neurons are traditionally recognized as the main communicators in the brain’s function, they are far from acting alone. Glial cells, which include astrocytes, oligodendrocytes, and microglia, make up a large portion of the brain’s cellular composition. Neurons and glial cells exist in approximately a 1:1 ratio in the brain (3, 40, 41).
Glial cells are now known to play a central role in regulating and supporting neural activity. They are
involved in maintaining homeostasis, supporting neuron metabolism, defending against immune threats, and remodeling synapses (3, 11, 16, 24).
Recent evidence has revealed that glial cells also play an increasingly important role as modulators of the central nervous system; they are no longer thought of as just passive support cells (16, 41). Because they are crucial to the functions carried out by the brain, when these cells become dysregulated, they can give rise to pathologies like gliomas.
Glioblastoma is the most common and severe type of glioma. Glioblastoma is a poorly understood tumor, and its etiology remains unclear (6, 20, 36). However, several risk factors like exposure to radiation, genetic mutations, age, and certain inherited diseases like LiFraumeni syndrome and neurofibromatosis have been identified as causes (20, 35, 42).
Some early studies, including Feigin and colleagues, introduced the term “gliosarcoma” to describe brain tumors consisting of glial cells and connective tissues, highlighting its unique cellular makeup (5).
A few decades later, Jellinger et al. (1978) classified a broader category of malignant glial tumors as glioblastoma multiforme, based on their diverse morphology and dedifferentiation of astrocyte cells (1). Today, we simply refer to these tumors as glioblastomas, and they remain the most aggressive primary brain tumors in adults (5, 12, 20, 22, 50).
Despite advances in surgeries, radiation, and chemotherapy, glioblastomas remain incurable and are associated with rapid progression, treatment resistance, and poor prognosis (20, 22, 36).
However, more recent studies challenge the traditional view of glioblastomas being simply a mass of proliferating cells. Evidence shows that glioblastoma tumors can make themselves at home, becoming an integral part of the brain as they grow and rewire the brain’s circuitry to their advantage. This review aims to explore and bring together studies on how
glioblastomas interact with and affect the brain circuitry.
Methods
To construct a comprehensive overview of the relationship between glioblastoma and the brain circuitry, we started our search on Google Scholar. We focused on finding foundational studies on normal brain circuitry, glial cells, and how they interact to solidify our understanding. Following this, we conducted a systematic search on key literature addressing the relationship between glioblastomas and brain circuitry using the PubMed database, covering articles published from 1954 to 2024. We intentionally included old studies to understand the historical development of glioblastoma research and how scientific understanding has evolved. Articles were selected based on relevance to our main topic and on citation numbers. We also used a snowballing method by reviewing the references of our primary articles to expand our literature pool. Inclusion criteria were: 1) Studies focusing on glioblastoma; 2) Inclusion of brain circuitry, cognitive function, neuronal activity; and 3) Availability of full text within our institution’s database. Articles without full text availability within the institutional database were excluded from the study. Our exclusion criteria also included studies that were purely molecular without neural context.
Results
Glioblastoma and Brain Circuitry: Integration, Rewiring, and Disruption
Many are beginning to recognize that glioblastomas are not only a mass of rapidly proliferating cells in the brain but also a tumor capable of integrating into neuronal networks and altering their function.
One interesting study by Venkataramani et al. demonstrates that glioma cells can form active synaptic connections with neurons (32). In this study, they used mice with human glioma cells implanted in their brain. Using a calcium imaging technique and a patch-clamp recording, they observed that the glioma cells were expressing AMPA-type glutamate receptors normally found on neurons receiving post-synaptic currents. The calcium imaging showed that those receptors were responding in real time to neuronal firing. Venkataramani and colleagues were able to show that glioma cells exhibited calcium movement
activity in coordination with normal brain neurons. In addition to the expression of AMPA type glutamate receptors, another group, Venkatesh and colleagues, argue that active neurons release neuroligin-3, a synaptic factor that is co-opted by glioma cells to stimulate growth (33, 43).
In parallel to the studies conducted by Venkatesh and colleagues, Tetzlaff et al. used rabies virus-based retrograde tracing, a method that labels cells directly connected via synapses, to find that glioma cells received input from normal neurons across multiple brain regions, and that these connections can vary from patient to patient (29). They claim that inhibiting this synaptic input offers strong therapeutic potential. These findings suggest that glioma tumors do not simply exist passively in the brain, but interact with the neurons to benefit their own survival. Krishna et al. proved this in a study where they showed that high neuronal activity near the tumor promoted tumor growth. They, along with several other groups, observed that tumor cells release proteins like thrombospondin-1 (TSP-1), which stimulate the formation of new excitatory synapses (15, 44, 45, 46). This creates a positive feedback loop: active neurons drive tumor progression, and the tumor modifies the microenvironment to increase neuronal activity.
At first glance, the ability of glioma cells to form connections with neurons may seem unsurprising, as healthy astrocytes do form connections with neurons (16). However, these connections are modulatory and involve astrocytes regulating neurotransmitter concentrations, buffering ions, and releasing gliotransmitters around neuronal synapses to influence neuronal activity, without forming a direct synapse (16). In contrast, glioma cells form true functional excitatory synapses, synapses where activating neurotransmitters are exchanged, with neurons. Neurons transmit glutamatergic signals across those synapses, which activate AMPA receptors on glioma cells, allowing the tumor to detect and respond to neuronal activity (32, 33). These functional synapses allow the tumor to integrate into neuronal circuits and exploit them to promote tumor growth (11, 51).
In addition to forming abnormal connections with neurons, the tumor rewires and damages brain circuitry to its advantage. In a study utilizing functional magnetic resonance imaging (fMRI) on patients performing language tasks, conducted by Krishna et
1. Schematic representation of a zoomed-in view of the glioblastoma tumor microenvironment and glioblastoma cell and neuron interactions. The glioblastoma microenvironment is a complex association of several different glial cells (astrocytes, oligodendrocytes, and microglia), blood vessels, neurons, and glioblastoma cells. Neurons and glioblastoma cells interact in several ways in the tumor microenvironment to alter neuron networks and circuitry.
al., it was observed that in glioma-infiltrated cortical regions, there was still neuronal activity, but this activity was disorganized and spread across the tumorinfiltrated regions. They claimed that this functional rewiring correlates with a decrease in survival rate for patients (15).
Similarly, Bosma et al. used magnetoencephalography (MEG), a neuroimaging technique measuring magnetic fields produced by neuronal activity, and graph theory, a mathematical method modeling pairwise relationships between different groups, to compare cognitive ability between patients with low-grade gliomas and healthy individuals. They found that glioma patients had low clustering on the graph, longer communication paths, and less efficient network structure, especially in theta and beta frequency bands, which are crucial for memory and attention (4). Changes in brain organization caused by these gliomas were linked to worse performance on tasks involving processing speed, attention, and memory.
All these studies show that glioblastomas, even low-grade gliomas, can reorganize the brain network to facilitate their own growth, at the detriment of patients’ cognitive ability and survival.
Neurovascular Decoupling in Glioblastomas: Functional Disruption Beyond the Neuron
In a healthy brain, neuronal activity drives local increases in blood flow, a relationship called
neurovascular coupling (15, 47). However, glioblastomas disrupt this mechanism. The tumor remodels surrounding vasculature and decouples neuronal activity from vascular response, leading to functionally active regions appearing silent on fMRI.
Holodny et al. conducted a functional brain study on patients with glioblastomas near their motor and sensory cortices using blood oxygenation leveldependent (BOLD) fMRI (BOLD detects changes in the ratio of oxygenated to deoxygenated hemoglobin in brain tissue) (17). Surprisingly, these patients showed low or absent activity in those areas, even though later surgical mapping proved that those brain areas were normal. Researchers hypothesize that the tumors might have affected the local vasculature, hence the perceived lack of activity from BOLD fMRI measurements (8).
This raises a concern: functional brain imaging may fail to accurately represent viable tissues in the periphery of the tumor. If normal brain tissues appear inactive due to disrupted vascular responses, that might lead to misclassifying them as non-functional, potentially affecting surgical decisions and treatment plans (7).
A study by Hou et al. reinforced the idea of this decoupling mechanism by examining how gliomas reduced the association between blood vessels and neurons. The study showed that, compared to healthy controls, patients with glioblastomas had reduced activation volume even when the tumor invasion in that specific brain region was minimal. They concluded that this reduction was not due to structural damage but to an insufficient vascular response (10).
This study provides support for the original hypothesis by Holodny et al. that glioblastomas affect blood vessel activity and structure in the periphery of the tumor.
Neuronal Hyperexcitability and Circuit Misfiring in Glioblastoma
Beyond structural invasion and vascular damage, glioblastomas also alter the electrical function of the brain. One of the most consistent findings in glioblastoma studies is that the glioma-infiltrated tissue becomes hyperexcitable (13, 14, 18, 19, 26). These changes can often lead to seizures and impaired cognition. There are several neurological symptoms commonly present in glioma patients, resulting from functional alterations in areas of the cortex that have been infiltrated. The most common neurological
Figure
symptom patients present with is seizures; over 80% of low-grade glioma patients and 50%–60% of highgrade glioma patients suffer from seizures. These can contribute significantly to the deterioration of cognitive function (13, 26).
To investigate how glioblastomas affect neuronal signaling over time, Meyer et al. used 2 glioma mouse models: one slow-growing and the other fast-growing. They monitored neuronal activity using in vivo calcium imaging and glutamate sensors, 2 techniques that allow for real-time data collection. Their results showed that the fast-growing glioblastomas produce spontaneous and unregulated neuronal activity through the accumulation of glutamate. They also found that the irregular activity of the neurons is not just confined to the margin of the tumor. It can extend even into the opposite hemisphere (18). Observations by Köhling et al. observations support Meyer and colleagues’ findings. In their study, they performed a human glioma xenograft into mice and discovered that not only does the tumor generate seizure activities in its proximity, but the seizure activity propagates to healthy tissues as well (14).
Studies like Montgomery et al. used a wide-field calcium imaging system in a mouse model expressing GCaMP (a genetically encoded calcium indicator) to track neural dynamics across the cortex as tumors grew. Initially, activity patterns across the brain were bilaterally synchronized. However, as the tumor infiltrated into deeper regions of the brain, this synchrony became weak. Neural signals near the tumor zone were firing at fast and irregular intervals. Additionally, regions that previously showed coherent fluctuations in calcium signals lost that coordination, particularly in motor and association cortices. One notable finding is that these bursts emerged before the onset of seizures, suggesting that hyperexcitability is not just a late-stage symptom but a fundamental feature of glioma progression (19).
This overactivity of the neurons can place new demands on energy usage in the brain. The next section will better explore how glioblastomas manipulate the neuronal metabolism to enhance their growth.
Metabolic Adaptations That Support Circuit Hijacking
Like most cancer cells, glioblastomas need energy to survive and thrive. To understand the metabolic adaptations necessary for this, Garofano et al. performed an in-depth transcriptomic analysis (a
study of multiple genes at once to observe which ones are actively transcribed) on glioblastoma samples and discovered that these tumors could be divided into metabolic subtypes based on their gene expression profiles. One of the major subtypes showed high expression of genes associated with mitochondrial metabolism, indicating a reliance on oxidative phosphorylation rather than glycolysis for energy production (6). The mitochondrial subtype was also associated with greater tumor aggressiveness and resistance to conventional therapies. While Garofano et al. showed that some glioblastomas rely on mitochondria for energy, Guo et al. (2024) found that under hypoxic conditions, glioma cells rely on glycolysis for energy, which leads to glucose and energy depletion in the brain and causes the release of glutamate in the extracellular environment. The excess glutamate then acts on surrounding neurons, triggering hyperexcitability (7, 48, 49).
Building on metabolic changes induced by glioblastoma, Winkler et al. showed that glioblastoma cells can take advantage of the brain’s activity by altering their own metabolic demands. In areas with high levels of neuronal firing, glioblastoma cells rely less on glucose, using alternative energy sources such as lactate and glutamate, which are typically produced by active neurons. This metabolic flexibility enables the tumor cells to survive and thrive in neural circuits, allowing them to integrate more tightly into the surrounding brain tissue (37). Winkler and colleagues also emphasized that this adaptation is not merely a survival mechanism. It is a way for the tumor to embed itself deeper into the functional circuits of the brain, disrupting normal neural signaling and making the tumor more resistant to treatment. This is supported by Numan et al., where they showed that gliomas preferentially grow in regions with high neuronal activity. Using brain scans from over 400 glioma patients and magnetoencephalography data from healthy people, they found that gliomas most often occurred in areas with high broadband power and elevated spiking activity. This selectivity is related to symptom severity and lower functional performance (21).
Together, these studies suggest that glioblastomas leverage neuronal activities and metabolism under both oxygen-rich and oxygen-poor conditions to take advantage of the high neuronal firing of their environment.
Discussion
This review has examined how glioblastomas invade and reshape components of the brain circuitry. From forming excitatory synapses to mimicking neurons’ behavior, glioblastoma can rearrange the brain network in many dynamic ways. One common theme that emerged across many of the studies in this review is the impact glioblastomas can have on cognition. For example, Krishna et al. and Montgomery et al. found a decrease in high-order function in glioma-infiltrated regions of the brain; however, the exact mechanism remains unclear (15, 19).
One limitation that many of the studies looked at in this review suffer from is that almost all their data comes from rodent models, which might not capture the full complexity of the human brain network. This review also does not cover how different treatments can change the interactions between glioma cells and neuronal networks.
Glioblastomas, being a rapid and aggressive form of cancer, are often caught late in their development; as a result, many of the interactions we have discussed here have already taken root before treatment is administered (15, 19, 36). However, it is worth mentioning that standard therapies like resection, radiation, and chemotherapy can affect these interactions in a variety of ways. Unfortunately, though these treatments remain palliative, with median survival barely exceeding one year (28, 48, 50).
Furthermore, this review does not investigate the role of genetics in glioblastoma invasion; however, without a doubt, this plays a vital role. Studies like Yu et al. demonstrated that PIK3CA variants of glioblastoma, when compared to other variants, can initiate hyperexcitability and remodel synaptic environment during glioma genesis. Also, this review does not explore how different subtypes of glioblastoma differ in their invasion and neuronal remodeling capabilities. The 4 subtypes of glioblastoma, neural progenitorlike (NPC), oligodendrocyte progenitor-like (OPC), astrocyte progenitor-like (AC), and mesenchymallike (MES) are each associated with different genetic alterations to CDK4, EGFR, PDGFRA, and NF1, respectively (1, 9, 20, 25, 30, 31, 34). Glioblastomas can arise from any 4 of those progenitors depending on cues from their environment. A better understanding of the role of genetics in glioblastoma would be beneficial in future treatment development.
Conclusion
This review brings into focus the multifaceted ways in which glioblastomas can interact with and manipulate brain circuitry. They build connections, hijack neuronal signals, and disrupt the communication between blood vessels and neurons. As many studies have revealed, this integration is a strategy to promote the survival of the tumor. Consequently, this integration blurs the line between brain and tumor, complicating both diagnosis and, more importantly, treatment (23). We believe that future research should be dedicated to methods of normalizing neuronal networks and vasculature in glioblastoma-infiltrated brains, to improve patient survival and slow cognitive decline. By understanding the interactions between glioblastomas and the brain, we not only gain insights into how we might better treat this disease but also uncover a deeper knowledge of how brain circuitry is organized and how tumors can adapt to a wide variety of microenvironments.
Disclosures
The authors have no conflicts of interest or financial disclosures relevant to this manuscript.
Acknowledgments
None.
References
1. Aldape K, Zadeh G, Mansouri S, Reifenberger G, von Deimling A. Glioblastoma: pathology, molecular mechanisms and markers. Acta Neuropathol. 2015;129(6):829–848.
2. Bassett DS, Gazzaniga MS. Understanding complexity in the human brain. Trends Cogn Sci. 2011;15(5):200–209.
3. Von Bartheld CS, Bahney J, Herculano-Houzel S. The search for true numbers of neurons and glial cells in the human brain: A review of 150 years of cell counting. J Comp Neurol. 2016;524(18):3865–3895.
4. Bosma I, Reijneveld JC, Klein M, et al. Disturbed functional brain networks and neurocognitive function in low-grade glioma patients: a graph theoretical analysis of resting-state MEG. Nonlinear Biomed Phys. 2009;3(1):9.
5. Feigin IH, Gross SW. Sarcoma arising in glioblastoma of the brain. Am J Pathol. 1954;30(5):1023–1039.
6. Garofano L, Migliozzi S, Oh YT, et al. Pathwaybased classification of glioblastoma uncovers a mitochondrial subtype with therapeutic vulnerabilities. Nat Cancer. 2021;2(2):141–156.
7. Guo X, Qiu W, Li B, Qi Y, Wang S, Zhao R, et al. Hypoxia-induced neuronal activity in glioma patients polarizes microglia by potentiating RNA m6A demethylation. Clin Cancer Res. 2024;30(6):1160–1176.
8. Holodny AJ, Schulder M, Liu WC, Maldjian JA, Kalnin AJ. Decreased BOLD functional MR activation of the motor and sensory cortices adjacent to a glioblastoma multiforme: implications for image-guided neurosurgery. AJNR Am J Neuroradiol. 1999;20(4):609–12.
9. Hoogstrate Y, Draaisma K, Ghisai SA, et al. Transcriptome analysis reveals tumor microenvironment changes in glioblastoma. Cancer Cell. 2023;41(5):678–692.e7.
10. Hou BL, et al. Effect of brain tumor neovasculature on BOLD fMRI activation volume in the primary motor cortex. AJNR Am J Neuroradiol. 2006;15;32(2):489-97.
11. Hua T, Shi H, Zhu M, Chen C, Su Y. Gliomaneuronal interactions in tumor progression: Mechanism, therapeutic strategies and perspectives. Int J Oncol. 2022;61(6):139.
12. Jellinger K. Glioblastoma multiforme: morphology and biology. Acta Neuropathol. 1978;41(2):119–126.
13. Klein M, Engelberts NH, van der Ploeg HM, et al. Epilepsy in low-grade gliomas: the impact on cognitive function and quality of life. Ann Neurol 2003; 54:514–520.
14. Köhling R, Senner V, Paulus W, Speckmann EJ. Epileptiform activity preferentially arises outside tumor invasion zone in glioma xenotransplants. Neurobiol Dis. 2006;22(1):64–75.
15. Krishna S, Keough M, Chaudhury A, et al. Glioblastoma remodeling of human neural circuits decreases survival. Nature. 2023;617(7961):589–97.
16. Linné ML, Aćimović J, Saudargiene A, Manninen T. Neuron–glia interactions and brain circuits. In: Sloot PMA, Hoekstra AG, Panja D, editors. Computational Modeling of Biological Systems . Springer; 2022. p. 77–94.
17. Logothetis NK, Wandell BA. Interpreting the BOLD signal. Annu Rev Physiol. 2004:66:735-69.
18. Meyer J, Yu K, Luna-Figueroa E, et al. Glioblastoma disrupts cortical network activity at multiple spatial and temporal scales. Nat Commun 2024;15:Article 3742.
19. Montgomery MK, Kim SH, Zhao H, et al. Glioma-induced alterations in neuronal activity and neurovascular coupling during disease progression. Cell Rep. 2020;31(1):107500.
20. Neftel C, Laffy J, Filbin MG, et al. An integrative model of cellular states, plasticity, and genetics for glioblastoma. Cell. 2019;178(4):835–849.e21.
21. Numan T, Breed LC, Maciel B, et al. Regional healthy brain activity, glioma occurrence and symptomatology. Brain. 2022;145(11):3654–65.
22. Osoba D, Aaronson NK, Muller M, et al. Effect of neurological dysfunction on health-related quality of life in patients with high-grade glioma. J Neurooncol. 1997;34(3):263–278.
23. Osswald M, Jung E, Sahm F, et al. Brain tumor cells interconnect to a functional and resistant network. Nature. 2015;528(7580):93–98.
24. Park HJ, Friston KJ. Structural and functional brain networks: from connections to cognition. Science. 2013 Nov 1;342(6158):1238411.
25. Patel AP, Tirosh I, Trombetta JJ, et al. Single-cell RNA-seq highlights intratumoral heterogeneity in primary glioblastoma. Science 2014;344(6190):1396–1401.
26. Rudà R, Trevisan E, Soffietti R. Epilepsy and brain tumors. Curr Opin Oncol. 2010;22:611–620.
27. Sultana OF, Bandaru M, Islam MA, et al. Unraveling the complexity of human brain: Structure, function in healthy and disease states. Ageing Res Rev. 2024; 100(102414).
28. Taphoorn MJ, Stupp R, Coens C, Osoba D, Kortmann R, van den Bent MJ, et al. Healthrelated quality of life in patients with glioblastoma: a randomised controlled trial. Lancet Oncol. 2005;6(12):937-44.
29. Tetzlaff CL, et al. Characterizing and targeting glioblastoma neuron-tumor networks with retrograde tracing. Cell. 2024.
30. Tirosh I, Venteicher AS, Hebert C, et al. Singlecell RNA-seq supports a developmental hierarchy in human oligodendroglioma. Nature 2016;539(7628):309–313.
31. Varn FS, Johnson KC, Martinek J, et al. Glioma progression is shaped by genetic evolution and microenvironment interactions. Cell 2022;185(13):2184–2199.e16.
32. Venkataramani V, et al. Glioblastoma hijacks neuronal mechanisms for brain invasion. Cell 2024.
33. Venkatesh HS, Johung T, Caretti V, et al. Neuronal activity promotes glioma growth through neuroligin-3 secretion. Cell. 2015;515(7525):533–537.
34. Verhaak RGW, Hoadley KA, Purdom E, et al. Integrated genomic analysis identifies clinically relevant subtypes of glioblastoma characterized by abnormalities in PDGFRA, IDH1, EGFR, and NF1. Cancer Cell. 2010;17(1):98–110.
35. Weller M, Wick W, Aldape K, et al. Glioma. Nat Rev Dis Primers. 2015;1:15017.
36. Weller M, van den Bent M, Preusser M, et al. EANO guidelines on the diagnosis and treatment of diffuse gliomas of adulthood. Nat Rev Clin Oncol 2021;18:170–186.
37. Winkler F, Venkatesh HS, Amit M, et al. Cancer neuroscience: State of the field, emerging directions. Cell. 2023;186(8):1689–1707.
38. Yu K, Lin CCJ, Hatcher A, et al. PIK3CA variants selectively initiate brain hyperactivity during glioma genesis. Nature. 2020;585(7824):256–260
39. Lou L. Architectures of neuronal circuits. Science. 2021;373(6559).
40. Garcia-Cabezas MA, John YJ, Barbas H, et al. Distinction of Neurons, Glia and Endothelial Cells in the Cerebral Cortex: An Algorithm Based on Cytological Features. Frontiers. 2016;10.
41. Allen NJ and Lyons DA. Glia as architects of central nervous system formation and function. Science. 2018;362(6411):181-185.
42. Yoshikawa MH, Rabelo NN, Telles JPM, et al. Modifiable risk factors for glioblastoma: a systematic review and meta-analysis. Neurosurg Rev. 2023;46(143).
43. Venkatesh HS, Tam LT, Woo PJ, et al. Targeting neuronal activity-regulated neuroligin-3 dependency in high-grade glioma. Nature. 2017; 549:533–537.
44. Daubon T, Léon C, Clarke K, et al. Deciphering the complex role of thrombospondin-1 in glioblastoma development. Nat Commun. 2019;10:1146.
45. Qi C, Lei L, Hu J, Wang G, et al. Thrombospondin-1 is a prognostic biomarker and is correlated with tumor immune microenvironment in glioblastoma. Oncol Lett 2021;21:22.
46. Tanase C, Enciu AM, Codrici E, et al. Fatty Acids, CD36, Thrombospondin-1, and CD47 in Glioblastoma: Together and/or Separately? Int. J. Mol. Sci. 2022;23(2):604
47. Phillips AA, Chan FHN, et al. Neurovascular coupling in humans: Physiology, methodological advances and clinical implications. Journal of Cerebral Blood Flow & Metabolism. 2015;36(4):647664.
48. Lange F, Hörnschemeyer MF, and Kirschstein T. Glutamatergic Mechanisms in Glioblastoma and Tumor-Associated Epilepsy. Cells 2021;10(5),1226.
49. Biegański M and Szeliga M. Disrupted glutamate homeostasis as a target for glioma therapy. Pharmacol. 2024;76:1305–1317.
50. Tykocki T and Eltayeb M. Ten-year survival in glioblastoma. A systematic review. 2018;54:7-13.
Mitral Regurgitation Severity and Echocardiographic Changes at 1 Year Postoperative: A Comparative Study of Surgical Aortic Valve Replacement Versus Transcatheter Aortic Valve Replacement
My N. Nguyen1†, Tariq Ahmad2, and Tyler J. Wallen3
1Geisinger College of Health Sciences, Scranton, PA 18509
2Geisinger Wyoming Valley Medical Center, Cardiology, Wilkes Barre, PA 18711
3Geisinger, Cardiovascular Surgery, Geisinger Center for Aortic Diseases, Wilkes Barre, PA 18711
†Doctor of Medicine Program
Correspondence: mnguyen01@som.geisinger.edu
Abstract
Background: Limited research exists on evolution of coexisting mitral valve regurgitation (MR) severity following transcatheter aortic valve replacement (TAVR) and surgical aortic valve replacement (SAVR) in patients with aortic stenosis (AS). In addition, very few studies have compared changes in echocardiographic measurements from preop to 1- year postop between TAVR and SAVR groups. This study aims to evaluate changes in MR severity and echocardiographic parameters at 1 year postop.
Methods: Patients with AS and MR who underwent TAVR or SAVR from January 2000 to January 2023 were categorized based on preoperative MR severity. A Chi-square test and Fisher’s exact test, where appropriate were conducted to examine the association between demographics or comorbidities and mitral valve regurgitation severity. Survival between two groups was compared using KaplanMeier curves. A Mann-Whitney test was utilized to examine changes in echocardiographic parameters pre- and 1 year postop.
Results: Of 165 patients, 80.0% underwent SAVR. In the SAVR group, 43 (32.6%) patients had coronary artery bypass grafting, 7 (5.3%) underwent mitral valve replacement (MVR), and 3 (2.3%) had endocarditis at surgery. Median age of TAVR patients with mild and moderate/severe MR was 80.0 and 83.5 years, respectively, compared to 71.0 and 76.5 years in SAVR group. Survival curves showed no significant difference between 2 procedures (p=0.0845).
Within patients who underwent TAVR, a significantly higher proportion of patients with moderate/severe MR preop had congestive heart failure compared
to patients with mild MR preop (100% vs 56.0%, p=0.031). For change in MR severity analysis, 25 patients were excluded due to missing follow-up echocardiographic data. Most patients with mild MR in both TAVR (79.2%) and SAVR (66.7%) groups experienced no change in MR severity. A higher percentage of those with moderate/severe MR showed grade improvement (62.5% in TAVR, 91.7% in SAVR without MVR, and 100% in SAVR with MVR). When comparing Echocardiographic parameters, there was no statistically significant difference between SAVR and TAVR groups, regarding Δ change from preop to 1-year postop measurements in median left ventricular ejection fraction (LVEF %) (0 vs 0, p=0.495), left ventricular internal diameter end diastolic (LVIDd cm) (0.01 vs 0.14, p=0.948), left ventricular internal diameter end systolic (LVIDs cm) (-0.34 vs 0.24, p=0.131), left ventricular outflow tract diameter (LVOTd cm) (-0.06 vs -0.10, p=0.838), left atrial dimension (LAd cm) (0.16 vs -0.25, p=0.262), and mitral valve E/A ratio (-0.001 vs 0.10, p=0.469). There were no significant differences in preop comorbidities or operative characteristics examined between improved and not improved MR groups.
Conclusion: Patients with moderate/severe MR who underwent SAVR or TAVR predominantly experienced improved MR grade at one year postoperatively, whereas those with mild MR were likely to remain unchanged. Comparison of echocardiographic changes from preop to 1-year postop shows no significant difference between SAVR and TAVR groups. No predictive preop or operative characteristics were identified. Our findings provide valuable insights for decision-making in performing TAVR and SAVR on patients with AS and MR, particularly those with
moderate to severe MR and multiple comorbidities, who may face higher risks from concomitant valve replacement procedures.
Introduction
Aortic stenosis (AS) is one of the most prevalent clinically significant valvular heart diseases, often coexisting with mitral regurgitation (MR) (1,2). Among patients with valvular heart disease, combined AS and MR is present in approximately 25% of cases. Particularly, coexisting moderate to severe MR is reported in 3% to 74% of patients undergoing transcatheter (TAVR) or surgical aortic valve replacement (SAVR) (3-10). The persistence of MR following aortic valve replacement is a welldocumented contributor to increased morbidity and higher rates of cardiovascular hospitalization (11). While double-valve surgery is an option for severe MR, its associated risks, including increased in-hospital mortality, necessitate careful patient selection (12).
Previous studies indicate that moderate to severe MR is associated with higher two-year mortality among patients undergoing SAVR but not TAVR (13, 14). A large meta-analysis suggests that moderate concomitant MR may increase both early and late mortality after SAVR (13), underscoring the potential need for mitral intervention even in cases of moderate MR. However, evidence also suggests that isolated SAVR can lead to MR improvement due to hemodynamic and left ventricular function alterations (5). Despite these findings, limited data exist on the direct comparison of MR outcomes between SAVR and TAVR in patients with combined aortic and mitral valve disease (14, 15).
Given that persistent MR is associated with increased mortality, identifying preoperative predictors of MR improvement or worsening is critical for optimizing patient management. This study aims to retrospectively analyze long-term MR outcomes following SAVR and TAVR to determine the optimal procedural approach for patients with varying degrees of MR severity. Specifically, we will:
1. Compare the impact of MR on mortality and survival following TAVR and SAVR.
2. Assess changes in MR severity at 1 year post-TAVR and SAVR.
3. Assess changes in echocardiographic parameters at 1 year post-TAVR and SAVR.
4. Identify predictive preop factors for MR improvement.
By addressing these objectives, our findings will provide valuable insights into the management of patients with concomitant AS and MR, facilitating more informed clinical decision-making and potentially improving long-term outcomes.
Methods
Study Population
Data were collected from patients age 18 years and older diagnosed with AS and MR who underwent either TAVR or SAVR. Patients with a history of other valvular diseases or cardiovascular conditions prior to the procedure were excluded from the study.
The initial sample included 185 patients; however, individuals with an "absent" or "unknown" MR status were excluded. Additionally, one patient was removed due to undergoing an aortic valve replacement not associated with aortic stenosis. After these exclusions, the final study cohort consisted of 165 patients. To evaluate MR severity changes from preoperative assessment to one-year postoperative follow-up, 25 additional patients were excluded due to absence of postop echocardiographic values.
Statistical Analysis
Descriptive statistics were utilized to summarize patient characteristics. Categorical variables were presented as frequencies and percentages, while continuous variables were reported as medians with interquartile ranges (IQR). Group comparisons were conducted using the Chi-square test or Fisher’s exact test, as appropriate, to analyze associations between demographic factors, comorbidities, and MR severity within each valve replacement group. The Mann-Whitney test was applied to assess differences in continuous characteristics across preoperative MR severity groups and to evaluate changes in echocardiographic parameters at preoperative and 1-year postoperative intervals. Additionally, a KaplanMeier survival analysis was performed to compare long-term survival outcomes between patients undergoing TAVR and SAVR. This comprehensive approach aims to identify factors predictive of MR improvement post-aortic valve replacement, thereby guiding optimal procedural decisions and enhancing patient outcomes.
Results
Of the 165 patients included in the study, 80% underwent SAVR. The median age of TAVR patients with mild and moderate/severe MR was 80.0 and 83.5 years, respectively, compared to 71.0 and 76.5 years in the SAVR group. The median age in the study population was 74.0 years, with a higher proportion of males (65.5%). Additional patient characteristics are presented in Table 1.
(N=165)
(MV) A point value (113.2 vs 96.7) were higher for patients with mild MR prior to the procedure. Within the SAVR group, the median left ventricular internal diameter end diastolic was slightly higher for patients with moderate/severe MR (5.0 vs 4.7).
Table 4 displays mortality outcomes for each preop MR severity group according to the aortic valve replacement type. Procedural related mortality was defined as death within 30 days of TAVR or SAVR procedure. Within SAVR procedures, a higher proportion of patients with moderate/severe MR had a mortality outcome (92.9% vs 61.5%, p-value=0.002). Survival analysis showed no significant difference between TAVR and SAVR groups (p=0.0845) (Figure 1).
In the analysis of MR severity changes as shown in Table 5, 25 patients were excluded due to missing follow-up echocardiographic data. Within the SAVR group, 43 (32.6%) patients had coronary artery bypass grafting (CABG), 7 (5.3%) underwent mitral valve replacement (MVR), and 3 (2.3%) had endocarditis at surgery. Most patients with mild MR in both the TAVR (79.2%) and SAVR (66.7%) groups experienced no change in MR severity. However, a higher proportion of patients with moderate/severe MR showed improvement (62.5% in TAVR, 91.7% in SAVR without MVR, and 100% in SAVR with MVR).
Table 2 provides a comparison of demographics between preop MR severity levels within TAVR and SAVR procedure groups. Within patients who underwent TAVR, a significantly higher proportion of patients with moderate/severe MR preop had congestive heart failure compared to patients with mild MR preop (100% vs 56.0%, p-value=0.031). Within patients who underwent SAVR, patients who had moderate/severe MR prior to the procedure were significantly older with a median age of 76.5 years (76.5 years vs 71.0 years, p-value=0.005).
Table 3 displays preop echocardiographic results in each preop MR severity group for TAVR and SAVR procedure groups. Within the TAVR group, the ejection fraction (EF) preop value (60.0 vs 42.5) and mitral valve
Table 6 demonstrates changes in preop and 1-year postop echocardiographic results in each preop MR severity group for TAVR and SAVR procedures. Echocardiographic parameter comparisons between SAVR and TAVR groups revealed no statistically significant differences in preoperative to one-year postoperative changes in median left ventricular ejection fraction (LVEF %) (0 vs. 0, p=0.495), left ventricular internal diameter end diastolic (LVIDd cm) (0.01 vs. 0.14, p=0.948), left ventricular internal diameter end systolic (LVIDs cm) (-0.34 vs. 0.24, p=0.131), left ventricular outflow tract diameter (LVOTd cm) (-0.06 vs. -0.10, p=0.838), left atrial dimension (LAd cm) (0.16 vs. -0.25, p=0.262), and mitral valve E/A ratio (-0.001 vs. 0.10, p=0.469).
In order to help identify factors that might predict improvement in MR, patients that underwent aortic valve intervention with mild or greater MR preoperatively were evaluated for improvement in MR. Patients were divided into groups depending on whether there was improvement in MR (n=46) or no improvement/worsening in MR (n=94). Preoperative (comorbidities) and operative characteristics
Table 1. Demographics and Preop MR Severity
Table 2. Comparison of
Ejection fraction (EF) preop
Left ventricular outflow tract diameter
Left atrium dimension
Table 3 continued
Mitral valve E point
Table 3. Preop Echocardiographic Results for Each Preop MR Severity Group by Aortic Valve Replacement Type
(echocardiographic values) were compared. There were no significant differences in pre-operative comorbidities or operative characteristics examined between improved and not improved MR groups.
Discussion
We observed MR improvement in the majority of patients with preop moderate-severe MR, consistent with findings by Kaczorowski et al. and Barbanti et al. (12, 15). Additionally, studies such as those by Waisbren et al. and Wyler et al. (5, 16) highlight variability in MR improvement, emphasizing the influence of functional versus degenerative MR.
The observed MR improvement, regardless of etiology, is particularly relevant for elderly patients with degenerative AS and MR, where double-valve surgery poses higher risks. The absence of significant echocardiographic parameter changes between SAVR and TAVR suggests that MR improvement is not solely dependent on procedural choice. Furthermore, preoperative factors such as atrial fibrillation, pulmonary hypertension, congestive heart failure, diabetes, hypertension, chronic kidney disease, and COPD have been identified as predictors of MR persistence, though these were not statistically significant in our study (15).
Regarding survival analysis between TAVR and SAVR, multiple factors can play a role in mortality rate, including age at procedure, concomitant surgery such as CABG or mitral valve replacement, and presence of endocarditis at the time of procedure.
In practice, changes in MR grade are influenced by various patient risk factors, including age at surgery, presence of comorbidities, and concomitant procedures. Regarding the differences in progression of preoperative MR grade between moderate/severe and mild groups in both procedures, we hypothesized that the types of MR damage (degenerative versus functional) may play at role in these discrepancies. Previous studies showed that etiology of MR may or may not predict reduction in MR. Specifically, there was no statistically significant difference in the proportion of patients that had leaflet abnormalities, mitral annular calcification, or any structural MV disease between groups of patients that experienced improvement in MR compared to those that did not (12). Further investigation and stratification of degenerative versus physiological MR may reveal the underlying pathophysiological cause of these changes, regardless of types of aortic valve replacements involve.
Limitations of this study including lack of echocardiographic measurements postoperatively
Table 5. Changes in MR Severity from Preoperative to 1 Year Postoperative for TAVR and SAVR Groups
and the non-aligned follow up timeline between the two procedures. While there is a specific guideline for TAVR echocardiogram follow-up at 30 days, 6 months, and 1 year, patients with SAVR should follow up at 1 year for biological valve replacement and 5 years for mechanical valve replacement. These discrepancy makes it challenging to obtain echocardiographic data at 1 year postop for every patient. Other limitations include small sample size and the retrospective nature of the analysis. Additionally, information such as etiologies of MR damage and mortality are not always documented on patient chart and can be challenging to identify during chart review.
Our findings highlight the challenges in predicting which patients will experience improvement in MR following aortic valve replacement. These results underscore the need for a prospective evaluation of the clinical benefits of addressing moderate MR at the time of aortic valve intervention. Future studies may focus on prospective analysis that identifies and controls a specific preoperative factor (comorbidities or echocardiographic data) that leads to MR grade improvement, ultimately optimizing patient outcomes. Furthermore, increasing follow-up timeline may give a deeper insight into how MR severity evolves over time.
(months)
Aortic Valve Repl Procedure Type SAVR TAVR Product-Limit Survival Estimates
Conclusion
Patients with moderate/severe MR who underwent SAVR or TAVR predominantly experienced improved MR grade at 1 year postoperatively, whereas those with mild MR were likely to remain unchanged. Comparison of echocardiographic changes from preop to 1-year postop, including LVEF, LVIDd, LVIDs,
Figure 1. Kaplan-Meier curve for SAVR and TAVR groups
Table 4. Mortality Outcomes for Each Preop MR Severity Group by Aortic Valve Replacement Type
LVOTd, and LAd, shows no significant difference between SAVR and TAVR groups. There were no significant differences in pre-operative comorbidities or operative characteristics examined between improved and not improved MR groups. Our findings provide valuable insights for decision-making in performing TAVR and SAVR on patients with AS and MR, particularly those with moderate to severe MR and multiple comorbidities, who may face higher risks from concomitant valve replacement procedures.
Disclosures
All authors declare no conflicts of interest relevant to this study. This research was supported by funding from the Geisinger Commonwealth School of Medicine under CRF number 23-027. The study was approved by the Institutional Review Board (IRB) under protocol number 2023-0995.
Acknowledgments
We sincerely thank Dr. Neil Mehta and Dr. Jeffrey Shuhaiber for their invaluable guidance and expertise in advising on study design and refining data
variables. We would also like to extend our gratitude to Andrea Hattenberger for her assistance in data extraction and Idorenyin Udoeyo for her essential role in data analysis, which greatly contributed to the interpretation of these findings. Additionally, we acknowledge Geisinger Commonwealth School of Medicine for generously providing the research funding necessary to complete this project.
References
1. Nkomo VT, Gardin JM, Skelton TN, Gottdiener JS, Scott CG, Enriquez-Sarano M. Burden of valvular heart diseases: a population-based study. Lancet. 2006 Sep 16;368(9540):1005-11.
2. Supino PG, Borer JS, Preibisz J, Bornstein A. The epidemiology of valvular heart disease: a growing public health problem. Heart Fail Clin. 2006 Oct;2(4):379-93.
3. Barreiro CJ, Patel ND, Fitton TP, Williams JA, Bonde PN, Chan V, Alejo DE, Gott VL, Baumgartner WA. Aortic valve replacement and concomitant mitral valve regurgitation in the elderly: impact on survival and functional outcome.
Table 6. Change in Preop and 1 Year Postop Echocardiographic Results for Each Preop MR Severity Group by Aortic Valve Replacement Type
Circulation. 2005 Aug 30;112(9 Suppl):I443-7.
4. Ruel M, Kapila V, Price J, Kulik A, Burwash IG, Mesana TG. Natural history and predictors of outcome in patients with concomitant functional mitral regurgitation at the time of aortic valve replacement. Circulation. 2006 Jul 4;114(1 Suppl):I541-6.
5. Waisbren EC, Stevens LM, Avery EG, Picard MH, Vlahakes GJ, Agnihotri AK. Changes in mitral regurgitation after replacement of the stenotic aortic valve. Ann Thorac Surg. 2008 Jul;86(1):56-62.
6. Caballero-Borrego J, Gómez-Doblas JJ, CabreraBueno F, García-Pinilla JM, Melero JM, Porras C, Olalla E, De Teresa Galván E. Incidence, associated factors and evolution of non-severe functional mitral regurgitation in patients with severe aortic stenosis undergoing aortic valve replacement. Eur J Cardiothorac Surg. 2008 Jul;34(1):62-6.
7. Harling L, Saso S, Jarral OA, Kourliouros A, Kidher E, Athanasiou T. Aortic valve replacement for aortic stenosis in patients with concomitant mitral regurgitation: should the mitral valve be dealt with? Eur J Cardiothorac Surg. 2011 Nov;40(5):1087-96.
8. Thomas M, Schymik G, Walther T, Himbert D, Lefèvre T, Treede H, Eggebrecht H, Rubino P, Colombo A, Lange R, Schwarz RR, Wendler O. One-year outcomes of cohort 1 in the Edwards SAPIEN Aortic Bioprosthesis European Outcome (SOURCE) registry: the European registry of transcatheter aortic valve implantation using the Edwards SAPIEN valve. Circulation. 2011 Jul 26;124(4):425-33.
9. D'Errigo P, Barbanti M, Ranucci M, Onorati F, Covello RD, Rosato S, Tamburino C, Santini F, Santoro G, Seccareccia F; OBSERVANT Research Group. Transcatheter aortic valve implantation versus surgical aortic valve replacement for severe aortic stenosis: results from an intermediate risk propensity-matched population of the Italian OBSERVANT study. Int J Cardiol. 2013 Sep 1;167(5):1945-52.
10. Gilard M, Eltchaninoff H, Iung B, Donzeau-Gouge P, Chevreul K, Fajadet J, Leprince P, Leguerrier A, Lievre M, Prat A, Teiger E, Lefevre T, Himbert D, Tchetche D, Carrié D, Albat B, Cribier A, Rioufol G, Sudre A, Blanchard D, Collet F, Dos Santos P, Meneveau N, Tirouvanziam A, Caussin C, Guyon P,
Boschat J, Le Breton H, Collart F, Houel R, Delpine S, Souteyrand G, Favereau X, Ohlmann P, Doisy V, Grollier G, Gommeaux A, Claudel JP, Bourlon F, Bertrand B, Van Belle E, Laskar M; FRANCE 2 Investigators. Registry of transcatheter aorticvalve implantation in high-risk patients. N Engl J Med. 2012 May 3;366(18):1705-15.
11. Bonow RO, Carabello BA, Chatterjee K, de Leon AC Jr, Faxon DP, Freed MD, Gaasch WH, Lytle BW, Nishimura RA, O'Gara PT, O'Rourke RA, Otto CM, Shah PM, Shanewise JS; American College of Cardiology/American Heart Association Task Force on Practice Guidelines. 2008 focused update incorporated into the ACC/AHA 2006 guidelines 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 (Writing Committee to revise the 1998 guidelines for the management of patients with valvular heart disease). Endorsed by the Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, and Society of Thoracic Surgeons. J Am Coll Cardiol. 2008 Sep 23;52(13):e1-142.
12. Kaczorowski DJ, Macarthur JW, Howard J, Kobrin D, Fairman A, Woo YJ. Quantitative evaluation of change in coexistent mitral regurgitation after aortic valve replacement. J Thorac Cardiovasc Surg 2013 Feb;145(2):341-7; discussion 347-8.
13. Unger P, Dedobbeleer C, Van Camp G, Plein D, Cosyns B, Lancellotti P. Republished review: Mitral regurgitation in patients with aortic stenosis undergoing valve replacement. Postgrad Med J 2011 Feb;87(1024):150-5.
14. Tzikas A, Piazza N, van Dalen BM, Schultz C, Geleijnse ML, van Geuns RJ, Galema TW, Nuis RJ, Otten A, Gutierrez-Chico JL, Serruys PW, de Jaegere PP. Changes in mitral regurgitation after transcatheter aortic valve implantation. Catheter Cardiovasc Interv. 2010 Jan 1;75(1):43-9.
15. Barbanti M, Webb JG, Hahn RT, Feldman T, Boone RH, Smith CR, Kodali S, Zajarias A, Thompson CR, Green P, Babaliaros V, Makkar RR, Szeto WY, Douglas PS, McAndrew T, Hueter I, Miller DC, Leon MB; Placement of Aortic Transcatheter Valve Trial Investigators. Impact of preoperative moderate/ severe mitral regurgitation on 2-year outcome after transcatheter and surgical aortic valve
replacement: insight from the Placement of Aortic Transcatheter Valve (PARTNER) Trial Cohort A. Circulation. 2013 Dec 24;128(25):2776-84.
16. Wyler S, Emmert MY, Biaggi P, Seifert B, Grünenfelder J, Falk V, Salzberg S. What happens to functional mitral regurgitation after aortic valve replacement for aortic stenosis? Heart Surg Forum. 2013 Oct;16(5):E238-42.
Predictive Scoring Analysis for Identifying Head and Neck Squamous Cell Cancer
Christopher Lloyd1†, My Nguyen1†, and Christian Kauffman2
1Geisinger College of Health Sciences, Scranton, PA 18509
2Geisinger Medical Center, Plastic Surgery Department, Danville, PA 17822
†Doctor of Medicine Program
Correspondence: clloyd@som.geisinger.edu
Abstract
Background: Head and neck squamous cell carcinoma (HNSCC) accounts for an estimated 450,000 deaths annually worldwide, presenting a significant clinical burden (1). Poor survival rates are largely due to the lack of reliable screening tools for early detection, as current methods offer limited insight into tumor invasiveness and metastasis potentials. While previous studies have explored HNSCC risk factors, predicting disease’s likelihood remains challenging due to lack of reliable screening tools. This retrospective study aims to creating a risk stratification scoring system for HNSCC, ultimately improves early detection and management for high-risk patients.
Methods: We developed a risk scoring system using data from the health system Geisinger, incorporating demographic and comorbidity factors such as sex, age, race, BMI, HPV history, smoking, alcohol use, diabetes, and prior diagnoses of leukoplakia or erythroplakia. The study included patients age 18 and older diagnosed with HNSCC between 2000 and 2024, excluding those with HIV/AIDS. Categorical variables were summarized with frequencies and percentages, while medians were reported for continuous variables. We used multiple logistic regression and descriptive analysis to assess the association between each risk factor and HNSCC development.
Results: Of the total 2,435 patients who were diagnosed with HNSCC, 1,607 (66.0%) are patients age 60 years or older at diagnosis. There are 620 females (25.5%) and 1,815 males (74.5%). Mean BMI is 27.5 (SD 7.0) with 707 patients (29.0%) having obesity. One thousand four hundred eighteen patients (58.2%) use tobacco, 949 patients (39.0%) use alcohol, 605 patients (24.8%) have diabetes, 62 patients (2.5%) have history of leukoplakia or erythroplakia, 14 patients (0.6%) have history of HPV, and 59 patients (2.4%) were previously diagnosed
with HNSCC. Among 778 patients with known pack years, 109 (14.0%) smoked less than 10 pack years, 669 (86%) smoked at least 10 pack years, and 536 smoked at least 20 pack years (68.9%). Among 227 patients with known alcohol use frequency, 123 (54.2%) consumed less than 10 drinks per week and 104 (45.8%) consumed at least 10 drinks per week. Type/location of tumor: 621 (25.5%) tongue cancer, 358 (14.7%) tonsillar cancer, 46 (1.9%) nasopharyngeal cancer, 72 (3.0%) oropharyngeal cancer, 54 (2.2%) palatine cancer, 202 (8.3%) lip cancer, 60 (2.5%) gum cancer, 114 (4.7%) floor of mouth cancer, 270 (11.1%) glottic cancer, 259 (10.6%) supraglottic cancer, 54 (2.2%) pyriform sinus cancer, and 325 (13.3%) others. Patients who scored 6 or higher are classified as high risk, 4–6 is medium risk, and 1–3 is low risk of HNSCC.
Conclusion: Tobacco use is identified as the highest risk factor that contributes to developing HNSCC. Patients scoring 6 or more points should undergo more rigorous and proactive screening for early detection. Given the severity of HNSCC, implementing this scoring system is crucial for reducing undiagnosed cases and improving patient outcomes. Future study should focus on which factors are associated with earlier onset of HNSCC and prediction of survival rate based on risk scores.
Introduction
Preventative medicine is becoming the new gold standard of health care, especially in regard to cancer surveillance. Head and neck squamous cell cancer (HNSCC) is a commonly occurring cancer worldwide and can have poor outcomes when not detected and treated early. In order to more efficiently and quickly find at risk patients a predictive scoring system was developed, based on risk factors, social factors, and dietary conditions. This study aims to elucidate those populations at most risk for developing HNSCC.
The yearly incidence rate of head and neck squamous cell carcinoma (HNSCC) stands at an estimated 890,000 new cases with 450,000 deaths occurring annually worldwide, proving to be a burdensome disease (1). 5- and 10-year survival rates range from 50% to 70% for HNSCC depending on location (ex: higher survival rates for p16+ oropharynx 69% vs hypopharynx 51%) and tumor lineage. These troubling survival rates can be attributed to the lack of accurate and precise screening tools available to clinicians to detect earlier stages of tumor growth (2). Furthermore, quality of life deficits present for survivors of HNSCC are considerably more impactful than most other cancers (survivors of other cancers are 23.6 cases per 100,000 individuals) shown to be the second highest rate of suicide (63.4 cases per 100,000 individuals), only second to pancreatic cancer survivors, (86.4 cases per 100,000 individuals) (3). Visibility and physical exam unfortunately only relay limited information on the tumor’s proclivity to invade and metastasize. More information is needed to inform better clinical management for the best prognosis possible.
This retrospective study aims to achieve a more comprehensive pathological and clinical understanding of head and neck squamous cell carcinoma (HNSCC), in order to better serve our population of potentially high-risk patients. By utilizing data from the Geisinger health system, we aimed to develop a scored system based off statewide patient histories (family history, HPV status, diabetes, obesity), demographic (age, sex, ethnicity), social factors (tobacco use, alcohol), and physical exam findings (dysplasia) that will allow a non-biased assessment that can be broadly applied to a population to most quickly and efficiently detect atrisk patients. Furthermore, this study aims to elucidate associations between risk factors such as diabetes and obesity, which have known pathophysiological mechanisms disrupted upregulating oxidative damage, yet little directed research investigating their role in the HNSCC pathogenesis.
Time is critical in oncological cases, particularly when in great proximity to adequate vasculature such as found within the head and neck areas, so it is imperative that an efficient and cost-effective tool be established to ensure that fewer patient cases go unrecognized before their outcome has suffered.
Depending on the stage of cellular degradation and particular hits made to an individual cell line (i.e., activations/inactivation of oncogenes/protooncogenes), particular proteins are expected to
significantly increase in baseline metabolic production. This may even prove to be a useful method in determining significance of oral dysplasia found by providing secondary lab values to support physical exam findings.
This study was crafted to better elucidate the needs of screening for HNSCC, particularly for the patients within Pennsylvania. Many shared social and nutritional aspects overlap for Pennsylvania residents that leave them at high risk of developing HNSCC, particularly rural individuals lacking nearby access to affordable and timely health care screenings. Additionally, even with routine general practitioner exams, thorough inspections of the oral cavity are not routine, thus significantly increasing the risk of development of nascent leukoplakia or metaplastic growths into invasive carcinomas. Furthermore, HNSCC has proven to be a treatable condition if found early enough into disease progression, and with poor 5- and 10-year survival rates, implementation of proper screening tools, practices, and patient education should serve to greatly eliminate the burden of care for late stage HNSCC diagnoses. Below are some common risk factors of HNSCC from previous literature.
HPV
In the United States, where the greatest number of studies have been performed, approximately 70% of all oropharyngeal cancers are attributable to HPV, a much higher proportion than seen worldwide. The overall incidence of HPV-related oropharyngeal cancers was estimated to be 4.8 in 100,000 in 2013–2014, compared with the incidence of HPV-related nonoropharyngeal head and neck cancers of 0.62 per 100,000.
In the United States, the incidence of HPV-related oropharyngeal cancer differs not only by sex but also by age and race. Through 2015 in the United States, the incidence in White men of all ages increased more than any other subgroup, to greater than 18 per 100,000 people. In contrast, the incidence in Hispanic and Black men was 6 and 4 in 100,000, respectively. In White women, the incidence in 2015 was almost 4 per 100,000. In contrast, the incidence in Hispanic and Black women was approximately 2 per 100,000. In 2020, the FDA, using the evidence presented here as a surrogate to predict clinical benefit in preventing HPV associated oropharyngeal cancer, approved the use of the HPV vaccine specifically for the prevention of HPV-related head and neck cancer (4).
Tobacco
Tobacco causes most HNC, and alcohol synergistically increases the risk of HNC conferred by tobacco use. Historically, approximately 90% of patients with HNC have a history of tobacco use, with a 4-fold to 5-fold increase in risk of developing oral cavity, oropharynx, and hypopharynx cancers, and a 10-fold increased risk of laryngeal cancer, among tobacco users. The carcinogenic effects of tobacco are dose-dependent, with the risk of HNC closely related to the frequency, duration, and intensity of cigarette smoking (5).
Alcohol
Alcohol use independently increases the risk of HNC, with an estimated 1% to 4% of cases attributable to alcohol alone and a twofold increase in odds of HNC for drinkers who are never users of tobacco. In particular, alcohol use increases risk of hypopharyngeal cancer when compared with other sites.
However, the larger impact of alcohol is observed in the interaction of alcohol with tobacco use. The combined effects of alcohol and tobacco use have a greater than multiplicative impact in increasing the risk of cancer. Among individuals who smoke 2 or more packs of cigarettes and drink more than 4 alcoholic drinks per day, the risk of HNSCC is increased greater than 35-fold (5, 6).
Male Gender
HNC overall is approximately twofold to threefold more common among men than women in the United States (5).
Methods
Using data across the Geisinger health system, we developed a risk scoring system based on demographic and comorbidity factors, including sex, age, race, BMI, history of HPV, smoking, alcohol use, diabetes, and prior diagnoses of leukoplakia or erythroplakia. Risk factors were selected through existing literature, both individual studies as well as meta-analysis that looked at head and neck squamous cell development. Of note, certain risk factors were excluded, such as the use of betel chew, due to the small sample sizes and lack of documentation likely to be made on patient charts and clinical histories.
Our study included patients age 18 and older diagnosed with HNSCC diagnosis between 2000 and 2024. Those with a history of HIV/AIDS were excluded.
To describe categorical characteristics, the frequency and percentages were reported. Additional limitations on the study include lack of definitive quantities of alcohol and tobacco consumption. Patients were grouped into either low-medium alcohol/tobacco use or high alcohol/tobacco use based on self-reported data during clinical encounters. Furthermore, patients with diabetes mellitus were consolidated into one group, with no differentiation made based on length or severity of diagnosis.
Median and interquartile range (IQR) were reported for continuous characteristics. Multiple logistic regression test and descriptive analysis were conducted to examine the association between each risk factor and likelihood of developing HNSCC. Following analysis of the risk factors and their associations, scores for each risk factor were devised and assigned to correlate with the likelihood of HNSCC development following exposure.
Results
Of the total 2,435 patients who were diagnosed with HNSCC, 1,607 (66.0%) patients age 60 years or older at diagnosis, and 224 (9.2%) patients age < 50 at diagnosis (Table 1). Mean age at diagnosis is 64.8 (SD 12.1). There are 620 females (25.5%) and 1815 males (74.5%). Within the female cohort, 341 (55.0%) patients use tobacco, 201 (32.4%) patients use alcohol, and 177 (28.5%) patients have diabetes (Table 2). Within the male cohort, 1,077 (59.3%) patients use tobacco, 748 (41.2%) patients use alcohol, and 428 (23.6%) patients have diabetes.
Forty-two patients (1.7%) are Black or African American. Within this cohort, 31 (73.8%) patients are male, 12 (28.6%) patients were at least 65 years old at diagnosis, 25 (59.5%) patients were at least 60 years old at diagnosis. Average age at diagnosis is 62.0. Fourteen (33.3%) patients are obese, 25 (59.5%) patients have use tobacco, 16 (38.1%) patients have alcohol use, 11 (26.2%) patients have diabetes, 1 (2.4%) patient had leukoplakia, 1 (2.4%) patient had HPV.
Mean BMI is 27.5 (SD 7.0). 707 patients (29.0%) had obesity with BMI 30 or above. Sixty-two patients (2.5%) have history of leukoplakia or erythroplakia, 14 patients (0.6%) have history of HPV, 1418 patients (58.2%) have tobacco use, 949 patients (39.0%) have alcohol use disorder, 605 patients (24.8%) have diabetes, and 59 patients (2.4%) reported having a history of HNSCC. The highest proportion of patients
Table 1. Demographics of Study Population
(58.2%) who have a history of tobacco use developed HNSCC later in life. Surprisingly, only a small proportion of patients with a history of HPV (0.6%) developed HNSCC.
For patients with previous HNSCC, the odds ratio of recurrent HNSCC with tobacco use is 0.96 (95% CI 0.53 - 1.84). 1017 patients (41.8%) did not smoke, and 1,486 patients (61.0%) did not use alcohol but still developed HNSCC. Among 778 patients with known pack years, 109 (14.0%) smoked less than 10 pack years, 669 (86%) smoked at least 10 pack years, and 536 smoked at least 20 pack years (68.9%).
Among 227 patients with known alcohol use frequency, 123 (54.2%) consumed less than 10 drinks per week and 104 (45.8%) consumed at least 10 drinks per week.
Among 14 patients with HPV, 13 of them are female and only 1 male. Type/location of tumor in our study include 621 (25.5%) tongue cancer, 358 (14.7%) tonsillar cancer, 46 (1.9%) nasopharyngeal cancer, 72 (3.0%) oropharyngeal cancer, 54 (2.2%) palatine cancer, 202 (8.3%) lip cancer, 60 (2.5%) gum cancer, 114 (4.7%) floor of mouth cancer, 270 (11.1%) glottic cancer, 259 (10.6%) supraglottic cancer, 54 (2.2%) pyriform sinus cancer, and 325 (13.3%) others.
Multiple logistic regression and descriptive analysis were used to develop the risk stratification scoring
(24.8%)
(75.2%) Previous leukoplakia or erythroplakia, n (%)
(2.5%)
(97.5%) Previous HNSCC, n (%) Yes
(2.4%)
Table 2. Risk Factors of HNSCC
system, as shown in Table 3. The predictive scoring system for head and neck squamous cell carcinoma (HNSCC) is designed to estimate the probability of an individual developing the condition based on key risk factors. This 9-point scoring system uses a logistic regression model that correlates factors like frequency, odds ratios, and clinical outcomes to predict the likelihood of the disease. Points are awarded based on the following risk factors: male sex (1 point), tobacco use (1 point for less than 10 pack years, 2 points for more than 10 pack years), alcohol use (2 points), diabetes (1 point for a history of diabetes), and previous leukoplakia, erythroplakia, or HNSCC (1 point). For females, a history of HPV
infection also contributes 1 point. Age is another critical factor, with 1 point awarded for being older than 60 years, as 66% of cases occur in this age group. The system divides patients into 3 risk categories: low risk (1–3 points), medium risk (4–6 points), and high risk (7–9 points). Smoking and alcohol consumption are highly correlated with HNSCC development, with 58.2% of diagnosed individuals having a history of smoking less than 10 pack years (earning 1 point), and 86% of diagnosed individuals having a history of smoking more than 10 pack years (earning 2 points). Although previous erythroplakia and HNSCC are highly correlated, only 2.4% of cases represent rediagnosis, with 2.5% of cases being erythroplakia-related. The scoring system aims to help identify high-risk patients, allowing for tailored prevention and treatment strategies.
A logistic regression analysis of age-related risk factors shows that age has a relatively low correlation with the outcome. With a multiple R of 0.0631, the R-squared value is only 0.00399, indicating that age alone explains only a small portion of the variance in the data. The adjusted R-squared is even lower, at 0.00357, further suggesting that age is a minimal predictor in this model. The regression statistics show that the relationship between age and the outcome is statistically significant with an F-value of 9.73 and a p-value of 0.0018.
Discussion
In this study, we identified and analyzed 10 risk factors most likely to be associated with HNSCC development. Occurrences, frequency of associated activities, and comorbidities were used to develop a novel risk stratification scoring system. The use of scoring systems has been widely used to varying degrees of success (7) but is typically field
dependent in terms of its capacity to integrate seamlessly into routine clinical encounters and documentation. Of note, prediction scores have several limitations including inaccurate observations, omitted patient health information, imperfect laboratory testing with the infeasibility of 100% specific and sensitive testing, and varying degrees of patient associated comorbidities. Furthermore, after development of predictive scoring systems, there are potential ethical and economic concerns regarding administering screens or subjecting patients to potentially unnecessary exams and associated fees. Yet it is the understanding of the researchers on this paper that this should not be seen as a limiting factor in the implementation of risk stratification, as preventive medicine has proven to be effective health care. Particularly, preventative medicine and screening tools make ideal domains for integrating risk stratification scoring systems, as typically disease and economic burden is much lower when health concerns are identified earlier in disease progression.
The most significant associated risk factors were tobacco use and alcohol use, 1,418 patients (58.2%) and 949 patients (39.0%), respectively. Our findings reinforce the well-established role of tobacco and alcohol use as primary risk factors for HNSCC, consistent with (8) research, which identified cigarette smoking and alcohol consumption as the strongest predictors of head and neck cancer.
Furthermore, our study results compared appropriately against national averages and suggested likelihood ratios of development seen by other studies (9) Interestingly, when comparing light against heavy tobacco usage, among 778 patients with known pack years, 109 (14.0%) smoked less than 10 pack years and 669 (86%) smoked at least 10 pack years. This suggests potentially a proverbial tipping point in terms of HNSCC development in regard to tobacco usage. This is contrasted against alcohol usage, as a total of 227 patients with known alcohol use frequency, 123 (54.2%) consumed less than 10 drinks per week and 104 (45.8%) consumed at least 10 drinks per week, exhibiting no significant difference in risk regarding total consumption, suggesting exposure and frequency might be more correlated to HNSCC development.
The median age of our cohort was 64.8 years old and the likelihood of male HNSCC development was 3 times as likely. These figures are similarly represented in national meta-analysis data with figures of 66-yearold mean age and a male HNSCC likelihood between
Table 3. Risk Scoring System for HNSCC
2 and 4 times. Within the female cohort, 341 (55.0%) patients use tobacco, 201 (32.4%) patients use alcohol, and 177 (28.5%) patients have diabetes. Within the male cohort, 1,077 (59.3%) patients use tobacco, 748 (41.2%) patients use alcohol, and 428 (23.6%) patients have diabetes. Interestingly, these primary risk factors are comparable between genders, and do not fully explain the significant increase in male driven development. This is contrasted against Park et al. 2022 in 10-year follow-up study of 10 million healthy participants that reported the sex differences in prevalence in HNSCC to be primarily driven by social risk factors such as alcohol and tobacco (10). Contradicting findings between the studies highlight the need for further research to be accomplished on this topic.
Our cohort encompassed 42 patients that were Black or African American, and many statistics regarding diagnosis and risk factors, such as alcohol and tobacco use, compared similarly to the larger cohort. Yet interestingly, Black individuals were noted to have a substantially lower prevalence rate by age 65 (28.6%) when compared against the cohort as a whole (48.7% at 65 or older at time of diagnosis). Additionally, mean age at diagnosis of Black or African American individuals was 62 seen against 66 years old for the entire cohort. The data suggests earlier ages of onset, diagnosis, and progression the Black and African American populations, but due to the small sample size of 42 patients, further research on this topic would be necessary.
While adding to the growing body of literature available for HNSCC, we also highlight important distinctions in risk patterns, particularly regarding several non-traditionally studied risk factors such as obesity, diabetes, and HPV. Both diabetes and obesity had very strong correlations to likelihood of development of HNSCC with 29% cases being obese patients and 24.8% having diabetes.
One of the most striking findings in our study is the relatively low proportion of HNSCC patients with a history of HPV (0.6%), despite the established role of HPV in oropharyngeal cancer. HPV is typically seen with development earlier in life (1), yet the average age of this cohort's HPV positive population was 73 years of age. Of note, the HPV positive sample size was limited with 14 total individuals which may contribute to the significant difference observed. However, our dataset may underrepresent this group due to regional
variations in HPV prevalence, differences in screening practices, underreporting, or data collected prior to expected age of onset for HPV positive patients.
Distinguishing the most common site of invasive growth is of particular importance. It does not only guide clinicians in performing more thorough evaluations, but proper patient education may inform patients that dysplastic and cancerous growth may be commonly found in areas difficult to view on self-assessments areas, leading to encouraging more frequent professional examinations for at-risk populations. Our study found the primary site of HNSCC to be the tongue at 25.5% of all diagnoses’ locations, being a fairly visible site that may be sensitive to disruptive stimulation in addition to lip (8.3%), while still a large percentage of locations were in limited visibility areas such as tonsillar (14.7%), supraglottic (11.1%), and palatine (10.6%). Although higher levels of exposure to social risk factors, such as tobacco and alcohol use, occur within the anterior cavity, a potential explanation could be related to the inability to identify the primary dysplastic growth prior to malignant transformation due to location.
Furthermore, other sites of importance that may not be conventionally thought to be high risk area, including pyriform sinus (2.2%) and nasopharyngeal (1.9%), suggesting that individuals scoring within the high-risk category should encourage communication between dental, ENT, and general practitioners to ensure holistic inspection and screening as occurred. Typically, by the time self-assessment warning signs have been recognized, such as dysphagia (difficulty eating), odynophagia (pain when swallowing) or otalgia (ear pain), tumor progression is quite advanced, highlighting the importance of effective in office examinations and screening methods (1). Of note, 13.3% were not accurately described enough to validate a location for the study.
While our study did not directly examine educational attainment, the significant proportion of patients who developed HNSCC despite not smoking or consuming alcohol suggests that additional social determinants of health, such as occupational exposures, health care access, and nutrition as demonstrated in The INHANCE study (11) may play a role in disease development. These findings highlight the need for broader risk assessment models that incorporate social and environmental factors alongside behavioral risk factors.
Age-related trends in our study also align with previous findings. We observed that the absolute risk of HNSCC increased with age but declined after 60–70 years, potentially due to competing mortality risks from other conditions. Given the high burden of HNSCC in older populations, refining screening strategies to target high-risk individuals while balancing the risks of overdiagnosis remains a crucial challenge.
Prior studies have explored the use of several biomarkers described that were shown to be associated with HNSCC and its development as potential tools in diagnosis. Looking ahead, future research should continue focusing on refining our risk model by integrating molecular and genetic biomarkers to enhance predictive accuracy. Advances in liquid biopsy techniques, including the detection of circulating tumor DNA and protein biomarkers, hold promise for improving early detection in asymptomatic individuals.
For example, IL-8 as a biomarker for oncogenesis has been shown to have a sensitivity of 0.65-0.85 and specificities of 0.79-0.93 in vivo animal studies over the last several years. This is thought to be due to IL-8 having been shown to be elevated in the microenvironment that the cells live in, such as hypoxic or acidic conditions (i.e., smoking, chewing tobacco, poor dental hygiene, alcohol, poor diet). Therefore, IL-8 in turn induces further genetic mutations through over stimulation of pathways controlling cellular growth, division, and maintenance (i.e., NOD1, RIP2 pathways) (12–16). By pairing effective screening methods and tools, such as the novel predictive scoring system with new techniques of salivary or serum screenings, tumor cell development and viability can be properly identified, assessed for risk, and diagnosed efficiently and relatively inexpensively.
Despite the strengths of our study, several limitations should be acknowledged. First, our study is retrospective in nature, which may introduce selection bias and limit the ability to establish causal relationships between risk factors and HNSCC development. Second, our dataset relies on electronic health records from a single health care system, which may not be fully representative of the broader population and could limit the generalizability of our findings. Third, certain risk factors such as genetic predisposition, environmental exposures, and socioeconomic variables were not comprehensively accounted for, potentially underestimating their contributions to HNSCC risk. Lastly, the reliance
on self-reported data for tobacco and alcohol use may introduce recall bias, which could impact the accuracy of our risk stratification model. Future studies incorporating prospective cohort designs and more comprehensive risk factor assessments will be essential to validate and refine our predictive model.
Looking ahead, future research should focus on refining our risk model by integrating molecular and genetic biomarkers to enhance predictive accuracy. Advances in liquid biopsy techniques, including the detection of circulating tumor DNA and protein biomarkers, hold promise for improving early detection in asymptomatic individuals.
Conclusion
Tobacco use is identified as the highest risk factor that contributes to developing HNSCC. Patients scoring 6 or more points should undergo more rigorous and proactive screening for early detection. Given the severity of HNSCC, implementing this scoring system is crucial for reducing undiagnosed cases and improving patient outcomes. Future study should focus on which factors are associated with earlier onset of HNSCC and prediction of survival rate based on risk scores.
Disclosures
All authors declare no conflicts of interest relevant to this study. This research was supported by funding from the Geisinger Commonwealth School of Medicine under CRF number 24-062.
Acknowledgements
We would like to express our gratitude to Dr. Lauren Blaha for her invaluable guidance and expertise in the study design and refinement of data variables. We also acknowledge Jordan Law for his essential support with data analysis. Finally, we are grateful to Geisinger Commonwealth School of Medicine for their generous financial support, which made this project possible.
References
1. Johnson DE, Burtness B, Leemans CR, Lui VWY, Bauman JE, Grandis JR. Head and neck squamous cell carcinoma. Nat Rev Dis Primers. 2020 Nov 26;6(1):92. Erratum in: Nat Rev Dis Primers. 2023 Jan 19;9(1):4.
2. Du E, Mazul AL, Farquhar D, Brennan P, Anantharaman D, Abedi-Ardekani B, Weissler MC, Hayes DN, Olshan AF, Zevallos JP. Long-term Survival in Head and Neck Cancer: Impact of Site, Stage, Smoking, and Human Papillomavirus Status. Laryngoscope. 2019 Nov;129(11):2506-2513.
3. Johnson DE, Burtness B, Leemans CR, Lui VWY, Bauman JE, Grandis JR. Head and neck squamous cell carcinoma. Nat Rev Dis Primers. 2020 Nov 26;6(1):92. doi:
4. Roman BR, Aragones A. Epidemiology and incidence of HPV-related cancers of the head and neck. J Surg Oncol. 2021 Nov;124(6):920-922.
5. Rettig EM, D'Souza G. Epidemiology of head and neck cancer. Surg Oncol Clin N Am. 2015 Jul;24(3):379-96.
6. Di Credico G, Polesel J, Dal Maso L, Pauli F, Torelli N, Luce D, Radoï L, Matsuo K, Serraino D, Brennan P, Holcatova I, Ahrens W, Lagiou P, Canova C, Richiardi L, Healy CM, Kjaerheim K, Conway DI, Macfarlane GJ, Thomson P, Agudo A, Znaor A, Franceschi S, Herrero R, Toporcov TN, Moyses RA, Muscat J, Negri E, Vilensky M, Fernandez L, Curado MP, Menezes A, Daudt AW, Koifman R, Wunsch-Filho V, Olshan AF, Zevallos JP, Sturgis EM, Li G, Levi F, Zhang ZF, Morgenstern H, Smith E, Lazarus P, La Vecchia C, Garavello W, Chen C, Schwartz SM, Zheng T, Vaughan TL, Kelsey K, McClean M, Benhamou S, Hayes RB, Purdue MP, Gillison M, Schantz S, Yu GP, Chuang SC, Boffetta P, Hashibe M, Yuan-Chin AL, Edefonti V. Alcohol drinking and head and neck cancer risk: the joint effect of intensity and duration. Br J Cancer. 2020 Oct;123(9):1456-1463.
7. Chung MS, Choi YJ, Kim SO, Lee YS, Hong JY, Lee JH, Baek JH. A Scoring System for Prediction of Cervical Lymph Node Metastasis in Patients with Head and Neck Squamous Cell Carcinoma. AJNR Am J Neuroradiol. 2019 Jun;40(6):1049-1054.
8. Dhull AK, Atri R, Dhankhar R, Chauhan AK, Kaushal V. Major Risk Factors in Head and Neck Cancer: A Retrospective Analysis of 12-Year Experiences. World J Oncol. 2018 Jun;9(3):80-84.
9. Gormley M, Creaney G, Schache A, Ingarfield K, Conway DI. Reviewing the epidemiology of head and neck cancer: definitions, trends and risk factors. Br Dent J. 2022 Nov;233(9):780-786.
10. Park JO, Nam IC, Kim CS, Park SJ, Lee DH, Kim HB, Han KD, Joo YH. Sex Differences in the Prevalence of Head and Neck Cancers: A 10-Year FollowUp Study of 10 Million Healthy People. Cancers (Basel). 2022 May 20;14(10):2521.
11. Lee YA, Al-Temimi M, Ying J, Muscat J, Olshan AF, Zevallos JP, Winn DM, Li G, Sturgis EM, Morgenstern H, Zhang ZF, Smith E, Kelsey K, McClean M, Vaughan TL, Lazarus P, Chen C, Schwartz SM, Gillison M, Schantz S, Yu GP, D'Souza G, Gross N, Monroe M, Kim J, Boffetta P, Hashibe M. Risk Prediction Models for Head and Neck Cancer in the US Population From the INHANCE Consortium. Am J Epidemiol. 2020 Apr 2;189(4):330-342.
12. Chan LP, Wang LF, Chiang FY, Lee KW, Kuo PL, Liang CH. IL-8 promotes HNSCC progression on CXCR1/2-meidated NOD1/RIP2 signaling pathway. Oncotarget. 2016 Sep 20;7(38):6182061831.
13. Pereira LH, Reis IM, Reategui EP, Gordon C, SaintVictor S, Duncan R, Gomez C, Bayers S, Fisher P, Perez A, Goodwin WJ, Hu JJ, Franzmann EJ. Risk Stratification System for Oral Cancer Screening. Cancer Prev Res (Phila). 2016 Jun;9(6):445-55.
14. Hema Shree K, Ramani P, Sherlin H, Sukumaran G, Jeyaraj G, Don KR, Santhanam A, Ramasubramanian A, Sundar R. Saliva as a Diagnostic Tool in Oral Squamous Cell Carcinoma - a Systematic Review with Meta Analysis. Pathol Oncol Res. 2019 Apr;25(2):447-453.
15. Pall AH, Jakobsen KK, Grønhøj C, von Buchwald C. Circulating tumour DNA alterations as biomarkers for head and neck cancer: a systematic review. Acta Oncol. 2020 Jul;59(7):845-850.
16. Kaskas NM, Moore-Medlin T, McClure GB, Ekshyyan O, Vanchiere JA, Nathan CA. Serum biomarkers in head and neck squamous cell cancer. JAMA Otolaryngol Head Neck Surg. 2014 Jan;140(1):5-11.
Thoracic Kyphosis Correction Using Pre-Bent PatientSpecific Rods in Adolescent Idiopathic Scoliosis: A Retrospective Analysis
Steven J. Grampp1†, Mark Seeley1,2, and Meagan Fernandez2
1Geisinger College of Health Sciences, Scranton, PA 18509
2Department of Orthopaedic Surgery, Geisinger Medical Center, Danville, PA 17822
†Doctor of Medicine Program
Correspondence: stevengrampp58@gmail.com
Abstract
Introduction: Adolescent idiopathic scoliosis (AIS) is a 3-dimensional congenital deformity of the spine. Current surgical treatment involves a multi-segmental posterior pedicle screw instrumentation and spinal fusion. Most reports focus on the surgical results in the coronal plane, but insufficient sagittal correction may lead to malalignment and accelerated distal segment degeneration. Currently, most surgeons contour rods based on experience during surgery without first determining the desired amount of thoracic kyphosis (TK) or measuring the angle created by the rod. In this study, we aim to determine if pre-bent patient-specific rods provide favorable sagittal alignment in AIS patients using normal reference thoracic kyphosis as our primary criteria.
Objectives: The main objective of this study was to assess whether the use of pre-bent patient-specific rods in AIS correction leads to improved sagittal alignment by achieving thoracic kyphosis angles that closely match surgically planned values. We sought to evaluate the degree of alignment across different subgroups classified by initial thoracic kyphosis and compare outcomes with normal reference kyphosis.
Methods: The study included 67 total patients. Normal thoracic kyphosis reference range was established as between 20° and 40°. Patients were then separated into different subgroups based on their initial visit thoracic kyphosis compared to the referenced range, which included hypokyphosis, normal kyphosis, and hyperkyphosis. Thoracic kyphosis was assessed at 3 separate time points via plain radiography: initial visit, pre-operative, and most recent follow-up. Surgically planned target kyphosis was determined prior to surgery and used to correct patients to the reference kyphosis range. Paired T-tests were used to analyze thoracic kyphosis between the different subgroups.
Results: Mean length to last follow-up was found to be 12.8 months. At the last follow-up, the mean increase in thoracic kyphosis was 13° for the group at large. The mean difference between kyphosis at last follow-up and target kyphosis in the subgroups was as follows: hypokyphosis was -4°, normal kyphosis was 0°, and hyperkyphosis group was +6°. Significant changes were found when comparing all subgroups' preoperative thoracic kyphosis angles with their surgical plan, and no significant differences were found when comparing their postoperative thoracic kyphosis angle with their surgical plan.
Conclusion: Prefabricated patient-specific rods achieved thoracic kyphosis angles close to surgically planned values, resulting in favorable sagittal plane alignment. The findings support the use of patientspecific rods to improve sagittal alignment and reduce the risk of malalignment and distal segment degeneration. Next steps for the study include following the cohort of patients over a longer period to assess the degree of degenerative conditions that develop in each subgroup and conducting a comparative analysis with traditional rods in all 3 anatomical planes.
Introduction
Adolescent idiopathic scoliosis (AIS) is a complex 3-dimensional spinal deformity that involves abnormal curvatures in the coronal, sagittal, and axial planes. While the primary focus of scoliosis correction has traditionally been on coronal plane realignment, abnormalities in sagittal alignment are also common and can have significant long-term consequences (1, 2). In the sagittal plane, AIS is frequently associated with thoracic hypokyphosis and lumbar hypolordosis, both of which can alter the natural curvature and biomechanics of the spine (3–6).
Postoperative hypokyphosis has been linked to trunk imbalance, potentially resulting in progressive spinal misalignment and compensatory changes in the pelvis and lower limbs (4, 6). Over time, this imbalance may contribute to early-onset osteoarthritis, disc degeneration, and chronic back pain in adulthood (4, 5). Although advancements in posterior vertebral instrumentation and fusion techniques have significantly improved the ability to achieve coronal plane correction, ensuring optimal sagittal alignment remains a challenge. Studies have reported variability in postoperative sagittal outcomes, with some patients experiencing persistent hypokyphosis or even kyphotic overcorrection, highlighting the need for more refined surgical planning and technique modifications (2–5).
As sagittal balance plays a crucial role in spinal biomechanics and long-term functional outcomes, achieving an appropriate thoracic kyphotic angle is essential for preventing postoperative complications and maintaining proper global spinal alignment. The normal range of thoracic kyphosis (TK) is generally accepted to be between 20° and 40° (8, 9). Geometric relationships have been established between pelvic incidence (PI) and lumbar lordosis (LL), as well as between TK and pelvic parameters (2–5). Achieving optimal sagittal alignment through appropriate restoration of TK is a key goal in AIS corrective surgery (1–5).
Surgeons often rely on personal experience to contour rods intraoperatively, without precisely determining the target kyphosis or measuring the rod’s curvature. However, in pediatric spinal deformity surgery, the use of patient-specific, pre-bent rods has shown promising results in improving sagittal alignment
(9-12). We hypothesize that patient-specific rods will closely achieve the preoperatively planned TK angles, potentially enhancing postoperative sagittal alignment and improving the rods’ mechanical properties.
Methods
Study Design and Patient Selection
This study was approved by the Institutional Review Board (IRB) (20190423) and was in accordance with the ethical requirements set forth in the 1964 Declaration of Helsinki. Patients were retrospectively identified from a single tertiary care center between 2019 and 2022. The study included patients who met the following criteria: instrumentation covering at least the T4 to T12 segments, utilization of machinebent patient-specific rods, availability of standing fullspine anteroposterior and lateral radiographs at the initial visit, preoperative evaluation, and final followup (minimum of 1 year), correction performed with 2 rods with high anchor density, and the use of hybrid constructs comprising pedicle screws and claw hooks.
Surgical Planning
Overall TK was defined as the maximum kyphotic curvature measured between the cranial transition vertebra (junction of TK and cervical lordosis) and the caudal transition vertebra (junction of TK and lumbar lordosis).
Target TK was determined preoperatively using radiographic measurements and PI as a reference parameter. Preoperative TK values were not used in determining the target TK (Figure 1).
Figure 1. Preoperative, surgical plan, and postoperative radiographic images of patient with AIS and thoracic kyphosis
AP and Lateral Images of a Patient with AIS and 12° of Kyphosis in the Thoracic Spine
AP and Lateral Images 12 Months Post Surgery with 36° of Thoracic Kyphosis
Lateral Images of the Surgical Plan to Correct the Patient's AIS, and Contour 38° of Thoracic Kyphosis
Subgroup Classification
To analyze outcomes, patients were categorized based on their preoperative TK values relative to the normal reference range of 20°–40°:
- Hypokyphotic (H Group) – TK < 20°
- Normal kyphosis (N Group) – TK 20°– 40°
- Hyperkyphotic (K Group) – TK > 40°
For each subgroup, the mean TK values were recorded at three time points: preoperative, surgical plan (target TK), and final follow-up (Table 1).
Assessment of Thoracic Kyphosis Changes
To evaluate the effectiveness of patient-specific rods in achieving the planned TK, the following calculations were performed for the entire cohort and each subgroup:
- Expected change in TK = Planned TK –Preoperative TK
- Measured change in TK = Final follow-up TK – Preoperative TK
- Gain difference = Expected change –Measured change. These values were analyzed to determine whether the final TK values matched the preoperative surgical targets (Table 2).
Statistical Analysis
Statistical analysis was performed using IBM SPSS Statistics v25 (IBM Corporation, Armonk, NY).
Descriptive data were reported as means and ranges. Student’s t-tests were used to compare within-group
means (preoperative TK, planned TK, and final followup TK) and to assess between-group differences in TK measurements across the three subgroups. A p-value ≤ 0.05 was considered statistically significant.
Results
A total of 67 patients met the inclusion criteria and were included in the study, with a mean follow-up period of 12.8 months. The cohort was categorized into 3 subgroups based on preoperative TK: hypokyphosis (n = 21), normal kyphosis (n = 37), and hyperkyphosis (n = 9).
Mean TK values were assessed at 3 time points: preoperative, surgical plan (target TK), and postoperative. Comparison of postoperative TK with planned TK demonstrated no statistically significant differences across all subgroups (Table 1).
Significant p-values (≤ 0.05) are in bold
Table 2. Thoracic Kyphosis and Surgical Outcomes Across Subgroups
Table 1. Preoperative vs.
In the overall cohort, the mean difference between planned and achieved postoperative TK was 0° (range: -16.8° to 16.5°), indicating a close match between intended and actual correction. Subgroup analysis revealed:
- Hypokyphotic group: Mean postoperative TK was 4° less than the planned correction.
- Normal kyphosis group: No difference between postoperative and planned TK.
- Hyperkyphotic group: Postoperative TK was 6° less corrected than expected (Table 2)
Discussion
This study evaluated the effectiveness of patientspecific, pre-contoured rods in achieving sagittal alignment in adolescent idiopathic scoliosis (AIS) surgery. Our findings support the hypothesis that these rods facilitate postoperative thoracic kyphosis (TK) values that closely match surgically planned targets, contributing to favorable sagittal plane correction.
Key Findings and Clinical Implications
The results demonstrate that pre-bent patient-specific rods effectively restored TK across all subgroups, with no statistically significant differences between final TK and surgical plan TK in any group. This suggests that using pre-contoured rods based on preoperative spinopelvic parameters enables precise sagittal correction, reducing the variability often seen with manually contoured rods.
Notably, the mean TK increase across the cohort was 13° at final follow-up, indicating that the rods successfully augmented kyphosis where needed. However, subgroup analysis revealed slight discrepancies in correction outcomes.
In the hypokyphosis group, postoperative TK was, on average, 4° less than the planned correction, suggesting that additional overbending or intraoperative modifications may be necessary to fully restore normal kyphosis.
The normal kyphosis group achieved near-perfect alignment, with no significant difference between planned and actual postoperative TK.
In the hyperkyphosis group, postoperative TK remained 6° higher than the planned correction, implying a potential limitation in the ability to fully reduce excessive kyphosis using pre-bent rods alone.
These findings align with prior research indicating that intraoperative rod flattening can lead to a loss of sagittal correction, particularly in hypokyphotic patients, as previously described in studies on spinal deformity correction (1, 2). The results also suggest that preoperative TK alone is not a reliable predictor of final alignment, reinforcing the importance of using PI as a key parameter in surgical planning. The incorporation of PI in preoperative assessments ensures a patient-specific approach to spinal balance, rather than relying solely on preoperative TK, which may be influenced by the scoliotic deformity.
Comparison to Traditional Rod Contouring Techniques
One of the primary challenges in AIS surgery is achieving consistent and reproducible sagittal alignment. Traditional rod contouring techniques rely heavily on surgeon experience and intraoperative estimation, often leading to unpredictable outcomes (13–16). Prior studies have reported high variability in sagittal alignment correction, with some patients experiencing postoperative hypokyphosis, which can predispose them to sagittal imbalance and degenerative changes in adulthood (12, 17–19).
In contrast, employing pre-contoured patient-specific rods provides a consistent approach to attaining the desired thoracic kyphosis angles, reducing the inconsistencies linked to manual rod contouring (11–16). The present study reinforces the notion that industrial pre-bending tailored to individual spinopelvic parameters may lead to more reproducible and predictable outcomes in AIS surgery.
Our results are consistent with prior literature demonstrating that intraoperative rod flattening can contribute to TK loss, particularly in hypokyphotic patients (16, 20). Additionally, studies comparing patient-specific rods to manually contoured rods have shown similar trends, reinforcing the benefit of preoperative planning in optimizing sagittal outcomes (12–14).
Limitations and Future Directions
This study has several limitations. First, the follow-up period (mean 12.8 months) is relatively short, limiting our ability to assess long-term stability of sagittal alignment and the potential for postoperative changes such as rod fatigue, adjacent segment degeneration, or compensatory alignment shifts. Second, the study lacks a control group of patients treated with
manually contoured rods, which would provide a direct comparison to assess whether patient-specific rods truly offer superior sagittal outcomes. Third, rod material and biomechanical properties were not analyzed in this study, though these factors can influence postoperative TK maintenance.
Future research should focus on long-term followup studies to evaluate whether sagittal alignment is maintained over time and whether differences in alignment persist across different patient subgroups. Additionally, a comparative study with traditionally contoured rods would help determine whether patient-specific rods provide a significant advantage in sagittal plane correction. Lastly, investigating the incidence of postoperative complications, such as adjacent segment disease and revision surgery rates, could provide further insight into the long-term benefits of this approach.
Conclusion
This study demonstrates that pre-bent patientspecific rods effectively achieve sagittal alignment in AIS patients, with final TK values closely matching surgically planned targets. While minor discrepancies were observed in the hypokyphosis and hyperkyphosis subgroups, the overall cohort exhibited a mean gain difference of 0°, indicating that the intended correction was largely achieved. These findings support the continued use of preoperative planning based on PI and the application of precontoured rods to enhance sagittal alignment consistency. Further research is needed to explore long-term outcomes and direct comparisons with traditional rod contouring techniques.
Disclosures
Nothing to disclose.
Acknowledgments
We would like to entend our appreciation to the Geisinger Commonwealth School of Medicine's library staff for their assistance in research for this manuscript.
References
1. Pishnamaz M, Migliorini F, Blume C, Kobbe P, Trobisch P, Delbrück H, et al. Long-term outcomes of spinal fusion in adolescent idiopathic scoliosis:
a literature review. Eur J Med Res. 2024 Nov 5;29:534.
2. Addai D, Zarkos J, Bowey AJ. Current concepts in the diagnosis and management of adolescent idiopathic scoliosis. Childs Nerv Syst. 2020 Jun;36(6):1111-9.
3. Cheung JPY. The importance of sagittal balance in adult scoliosis surgery. Ann Transl Med. 2020 Jan;8(2):35.
4. Roussouly P, Labelle H, Rouissi J, Bodin A. Preand post-operative sagittal balance in idiopathic scoliosis: a comparison over the ages of two cohorts of 132 adolescents and 52 adults. Eur Spine J. 2013 Mar;22(Suppl 2):203-15.
5. La Maida GA, Zottarelli L, Mineo GV, Misaggi B. Sagittal balance in adolescent idiopathic scoliosis: radiographic study of spino-pelvic compensation after surgery. Eur Spine J. 2013 Nov;22(Suppl 6):S859-67.
6. Ritzman TF, Floccari LV. The sagittal plane in spinal fusion for adolescent idiopathic scoliosis. J Am Acad Orthop Surg. 2022 Jul 15;30(14):e957-e67.
7. Pediatric Orthopaedic Society of North America. Postural Kyphosis [Internet]. Rosemont (IL): POSNA; [cited 2025 Jul 13]. Available from: https://posna.org/physician-education/studyguide/postural-kyphosis
8. Betz RR. Kyphosis of the thoracic and thoracolumbar spine in the pediatric patient: normal sagittal parameters and scope of the problem. Instr Course Lect. 2004;53:479-84.
9. Ishikawa Y, Kanai S, Ura K, Kokabu T, Yamada K, Abe Y, et al. Development of notch-free, pre-bent rod applicable for posterior corrective surgery of thoracolumbar/lumbar adolescent idiopathic scoliosis. J Clin Med. 2023 Sep 4;12(17):5750.
10. Fukushima A, Ohnishi T, Kokabu T, Abe Y, Tachi H, Endo T, et al. Four-dimensional anatomical spinal reconstruction using pre-bent rods in thoracic adolescent idiopathic scoliosis. Sci Rep. 2025 Jan 2;15(1):378.
11. Solla F, Ilharreborde B, Clément J-L, Rose EO, Monticone M, Bertoncelli CM, et al. Patientspecific surgical correction of adolescent idiopathic scoliosis: a systematic review. Children (Basel). 2024 Jan 15;11(1):106.
12. Jabbouri SS, Joo P, David WB, Jeong S, Moran J, Jonnalagadda A, et al. Pre-contoured patientspecific rods result in superior immediate sagittal plane alignment than surgeon contoured rods in adolescent idiopathic scoliosis. J Spine Surg. 2024 May 29;10(2):177-89.
13. Ohrt-Nissen S, Dahl B, Gehrchen M. Choice of rods in surgical treatment of adolescent idiopathic scoliosis: what are the clinical implications of biomechanical properties? A review of the literature. Neurospine. 2018 Jun 19;15(2):123-30.
14. Lamerain M, Courvoisier A, Glorion C, Vialle R. Sagittal alignment in adolescent idiopathic scoliosis treated by posterior instrumentation with pedicle screws. J Pediatr Orthop. 2016 Mar;36(2):93-100.
15. Bautista AG, Reyes JL, Lee NJ, Fields MW, Sardar ZM, Lenke LG, et al. Patient-specific rods in adolescent and adult spinal deformity surgery: a narrative review. Int J Spine Surg. 2024 Aug 28.
16. Zhang Y, Li H, Wang J, Zhang Y, Zhang X, Zhang Y, et al. A novel patient-specific rod design method for scoliosis surgery based on 3D printing and finite element analysis. Procedia Comput Sci 2022;202:1797-1806.
17. Zhang J, Sun W, Wang J, Qiu Y, Zhu Z. Global sagittal alignment after surgery of right thoracic idiopathic scoliosis: a comparison between normal thoracic kyphosis and hypokyphosis. Sci Rep. 2021 Mar 11;11(1):5660.
18. Scoliosis Research Society. Fixed sagittal imbalance[Internet]. Milwaukee (WI): SRS; [cited 2025 Jul 13]. Available from: https://www.srs.org/ Patients/Conditions/Fixed-Sagittal-Imbalance
19. Park HY, Kim YH, Kim SI, Han SB, Ha KY. Two different types of postoperative sagittal imbalance after long instrumented fusion to the sacrum for degenerative sagittal imbalance. J Neurosurg Spine. 2020 Nov 1;33(5):613-22.
20. Katzouraki G, Vasiliadis ES, Kaspiris A, Evangelopoulos DS, Grivas TB, Pneumatikos SG. Sagittal balance parameters and proximal junctional kyphosis in adolescent idiopathic scoliosis. J Clin Med. 2024 Mar 25;13(7):1895.
An Evaluation of Child Maltreatment Coding within a Large Health Care System in Rural Pennsylvania
Emily
Harasym1†, Michelle Pistner Nixon2, Adam Cook2, H. Lester Kirchner2, Paul Bellino3, and Lisa Bailey-Davis2
1Geisinger College of Health Sciences, Scranton, PA 18509
2Department of Population Health Sciences, Geisinger College of Health Sciences, Danville, PA 17822
3Geisinger Medical Center, Danville, PA 17822
†Doctor of Medicine Program
Correspondence: eharasym@som.geisinger.edu
Abstract
Background: Over 500,000 children were abused or neglected in the United States in 2023, and over 2,000 children died from abuse-related injuries. A decline in child abuse reporting was observed both nationally and in Pennsylvania during the COVID-19 pandemic, likely due to limited interactions between victims and mandated reporters including health care providers. Inconsistencies have been observed when comparing prevalence rates of child maltreatment obtained through electronic health record (EHR) documentation to those obtained from child welfare data. We aimed to determine whether the prevalence of child abuse coded within EHRs mirrored the national underreporting trends observed during the COVID-19 pandemic and to assess the reliability of EHR documentation of child maltreatment for public health surveillance.
Methods: Medical records for pediatric patients were extracted from a large, integrated health care system. Fourteen child maltreatment diagnosis codes were used to identify EHR encounters involving abuse in the emergency department, primary care, well-child, specialist, and hospital settings prior to and during the COVID-19 pandemic. Over 5,300 patients were identified with more than 7,600 encounters since Jan. 1, 2000. Systematic chart reviews were conducted in representative samples to determine the validity of child abuse coding for each encounter. The Pennsylvania Department of Human Services reports child abuse and market share data and estimates of the percent of visits affiliated with this health care system per county were obtained to approximate the number of child abuse reports expected from health care providers within the system. Values were compared to the actual number of cases documented with
maltreatment-associated diagnosis codes in the EHRs.
Results: Preliminary analyses identified trends inconsistent with the anticipated declines in reporting during the pandemic period. Three potential sources of data bias were investigated via targeted review of 83 charts: the use of adult abuse codes for pediatric patients, the overuse of a single sexual abuse code, and the repeating of certain abuse codes over time. This chart review suggested that overlap in the coding criteria for pediatric and adult patients and coding error contributes to the first source of bias. Outside laboratory testing and lab-specific protocol for coding pediatric sexual abuse were responsible for the second source of bias. Of charts reviewed with repeated abuse codes, less than 5% contained evidence of new or continued maltreatment. Just over 50 substantiated reports of child maltreatment were expected from system-affiliated health care providers. Approximately 96 new cases of child maltreatment were documented in EHRs using maltreatment-associated diagnosis codes.
Conclusion: Documentation of child maltreatment within the EHR did not mirror trends observed nationally during the COVID-19 pandemic. While EHR data may improve monitoring of health care utilization and aid in the development of primary prevention strategies, epidemiological investigations are limited by potentially inflationary or inconsistent coding practices. Data biases that decrease reliability include overlap in pediatric and adult coding, laboratoryspecific coding protocol, and inappropriate coding of follow-up visits. As an adjunct to diagnosis codes, EHR-based surveillance may be improved by artificial intelligence-based algorithms programmed to detect or predict pediatric abuse from free text within charts.
Introduction
Child maltreatment remains a pervasive and deadly threat to public health with more than 500,000 reported victims of abuse or neglect and an estimated 2,000 deaths due to child abuse in the United States in 2023 (1). The Centers for Disease Control and Prevention defines child maltreatment as “any act or series of acts of commission or omission by a parent or other caregiver that results in harm, potential for harm, or threat of harm to a child” (2). Child maltreatment is associated with a significant economic burden and the lifetime cost incurred annually for substantiated cases has been estimated to exceed $400 billion in the United States (3). Moreover, childhood trauma contributes to adverse childhood events (ACEs) which can impair pediatric development and may have a profound and enduring impact on adult health outcomes. ACEs have been associated with an increased risk of developing anxiety, depression, substance abuse disorder, respiratory and cardiovascular diseases, and cancer (4, 5).
Early identification of child maltreatment and immediate intervention is paramount to mitigating the long-term effects of abuse. Mandated reporters are trained to identify signs of maltreatment and are legally obligated to report suspected abuse to the appropriate authorities. The majority of reports of alleged abuse arise from mandated reporters, primarily from law enforcement officers, educators, and health care professionals (1). During the COVID-19 pandemic, widespread school closings, stay-at-home orders, and isolation measures instituted to limit disease spread drastically reduced interactions between children and mandated reporters. The pandemic also exacerbated established risk factors for child maltreatment such as social isolation, economic hardship, depression, and stress (6). Despite this, a substantial decrease in reports of suspected child maltreatment was observed internationally (7), in the United States (8), and in Pennsylvania (9) in 2020 (Figure 1). This decrease has been attributed to limited access to mandated reporters in school and extracurricular settings as well as a reduction in routine health care visits (6, 10, 11). Alarmingly, some studies reported an increase in the incidence of substantiated abuse, in the severity of child abuse injuries in the early lockdown period, and in the frequency of maltreatment observed in hospitalized patients (7, 8, 12). In Pennsylvania, reports of
suspected child maltreatment are increasing but still lag behind pre-pandemic totals (9).
Health care providers and emergency medical technicians accounted for more than 8,000 of the nearly 34,000 child abuse reports from mandated reporters in 2023 (9). When appropriately documented, electronic health records (EHRs) may serve as epidemiological data sets for the identification of risk factors for abuse and for surveillance of child abuse diagnosis and reporting. Prevalence rates for child maltreatment obtained from EHRs, however, have been observed to be substantially lower than those obtained from child welfare data (13–16).
Health care providers have identified a number of barriers to both screening for and reporting suspected maltreatment including inadequate training on reporting protocols, distrust of subsequent investigative measures, lack of information about community resources, and office visit time constraints (17–19). One study found that while most primary care physicians felt comfortable examining victims of suspected abuse, they were less comfortable providing definitive opinions on abuse or testifying in court cases regarding abuse (20). These uncertainties are exacerbated by coding practices that largely rely on the International Classification of Diseases, Clinical Modification (ICD-CM). The ICD-CM vocabulary classifies and documents the findings of health care encounters, thus defining and coarsening often unique and sensitive circumstances. The tenth revision (ICD10) expanded child abuse documentation to include codes for both suspected and confirmed child abuse in 2015 (21).
Reliably coding maltreatment within EHRs relies upon the accuracy and completeness of clinician documentation and the competency and scrupulousness of the medical coder. Ambiguity in the diagnostic criteria for child maltreatment, especially when compared to more definitive diagnoses such as fractures or infections, may create documentation and coding challenges. Reported sensitivities, specificities, and predictive values of maltreatment codes have varied among hospital systems, visit settings (inpatient vs. outpatient), and ICD revisions (ICD-9 vs. ICD10). For instance, one study of ICD-9 maltreatment codes among 4 children’s hospitals in the northeast reported sensitivities ranging between 73.5% and 92.4% and specificities between 85.4% and 100% (22). Another reported an ICD-10 maltreatment coding
sensitivity of 55% and specificity of 78% for inpatients and sensitivity and specificity of 22% and 86%, respectively, for outpatients. (23). Documentation errors cited within these studies were due to both provider and coder error.
Traditionally, public health surveillance of child maltreatment has been conducted using Child Protective Services (CPS) data. A number of limitations exist within this dataset, including the overrepresentation of children of low socioeconomic status (SES) and overrepresentation of Black children (24, 25). The use of medical records for surveillance or as an adjunct to CPS data may capture unique populations missed in prior investigations offering valuable insight to enhance public health. Schnitzer et al. report that linking EHRs with child welfare data may identify more than 10% more children than using welfare data alone (26).
Unfortunately, biases in suspecting maltreatment and disproportionalities in health care provider reporting have also been demonstrated. One study reported that while African American children, Hispanic children, and children from high-poverty neighbors were overrepresented in CPS reports from medical providers, Caucasian children and those from lowpoverty neighborhoods were underrepresented (27). Disproportionate use of emergency departments, where acute injuries secondary to child abuse may be identified, by families lacking health insurance may also contribute to trends in over- or underreporting (27). Monitoring ICD-10 code usage for suspected abuse has shown that Black race was an independent risk factor associated with increased suspicion for child abuse and Hispanic race was a protective factor associated with lower suspicion for child abuse when socioeconomic factors and hospital characteristics were controlled (28, 29). Provider bias also impacts the care received by the individual patient. Data has also shown that Black children remain hospitalized longer than other suspected victims of abuse, even when injuries are milder (30). On a larger scale, these biases may influence the allocation of resources for primary and secondary prevention in the community.
The initial aim of the current study was to compare national maltreatment underreporting trends observed during the pandemic to estimates obtained through ICD-10 coding of child maltreatment within EHRs in a large rural Pennsylvania health system. Preliminary analysis revealed inconsistencies in child
maltreatment documentation associated with certain ICD-10 codes which prompted further investigation. The aim of the study was revised to examine the limitations of child maltreatment documentation in the EHR documentation in the EHR using ICD-10 codes.
Methods
Study Population
Institutional Review Board approval was obtained. Medical records were acquired from a pre-existing database within a large, integrated health care system covering 45 counties in rural Pennsylvania. Records for patients less than 18 years old who were receiving care as evidenced by documented height and weight in the EHR from January 2000 onward. Contact dates prior to January 2000 were excluded as fewer encounters existed within this period due to incomplete adoption of electronic recordkeeping. ICD-10 codes of interest pertaining to maltreatment were compiled through a thorough literature search and with the guidance of experts in the field. Fourteen ICD-10 parent codes related to abuse or maltreatment and their relevant subcodes were selected (Table 1). ICD-9 codes documented within patient charts were automatically assigned ICD-10 codes by EHR software. Medical records were filtered to include routine, emergency, and specialist visit encounters containing at least one maltreatment code.
Maltreatment
ICD Codes
T74.0
T74.1
T74.2
T74.4
T76.0
T76.1
T76.2
Y04
Y07
Y09
X95
X93
Z04.72
Y08.89XA
Description
Neglect or abandonment, confirmed
Physical abuse, confirmed
Sexual abuse, confirmed
Shaken infant syndrome
Neglect or abandonment, suspected
Physical abuse, suspected
Sexual abuse, suspected
Assault by bodily force
Perpetrator of assault, maltreatment and neglect
Assault by unspecified means
Assault by other and unspecified firearm and gun discharge
Assault by handgun discharge
Examination following alleged child physical abuse
Assault by other specified means, initial encounter
Table 1. ICD-10 Codes of Interest. Fourteen ICD-10 codes pertaining to maltreatment and their relevant subcodes were compiled after performing a thorough literature search and with the guidance of experts in the field.
Results
Preliminary Findings
Filtering of EHRs by ICD code resulted in more than 5,300 patients with over 7,600 encounters. Initial analysis of extracted data revealed three unexpected findings which prompted further evaluation. First, a drop in maltreatment coding was observed from 2019 to 2021 which preceded the pandemic-related decrease expected in 2020. This drop coincided with a substantial decrease in the use of T74 codes, most notably T74.22XA (child sexual abuse, confirmed, initial encounter), and an increase in T76 codes (Figure 2). Prior to this drop, an increase in coding of T74.22XA was observed from 2013 to 2018 by a specific laboratory group within the health system (Figure 3).
Two additional irregularities were observed which prompted us to perform detailed chart reviews: repetitive use of maltreatment codes over multiple contact dates and use of adult codes for pediatric patients (T74.01, T74.11XA, T74.21XA).
Pre- and Post Pandemic Child Abuse Reports
Reports Substantiated Reports
1. Pre- and Post-Pandemic Child Abuse Reporting. A precipitous decline in child abuse reports was observed in Pennsylvania in 2020 per Pennsylvania Department of Human Services data, likely due to limited contact between children and mandated reporters. Substantiated and total reports of child maltreatment have yet to return to pre-pandemic rates according to the most recent reporting data.
2. Number of Encounters by ICD-10 Parent Code. EHR data was filtered to include pediatric emergency department, primary care, well-child, specialist, and hospital encounters involving select ICD-10 codes beginning Jan. 1, 2000. The most frequently used codes were T74 (abuse, neglect, and other maltreatment, confirmed) and T76 (abuse, neglect, and other maltreatment, suspected). A drastic decrease in the use of T74 codes was observed from 2019 to 2021 with a concomitant increase in the use of T76 codes.
One reviewer systematically reviewed each episode of care to identify the reason for the visit, whether abuse was occurring at that encounter, to classify the type of abuse occurring as physical, sexual, neglect, or other, if applicable, and to describe any injuries. Provider documentation, laboratory testing, and
Representative samples within each of these categories were selected to explore coding rationale. Records containing repeat abuse codes were divided into 2 groups: those containing repetition of the identical abuse code (“true repeats”) and those with multiple encounters with unique abuse codes (“distinct repeats”). These subsets were further divided to elucidate whether time between encounter impacts coding accuracy. Charts with less than 3 months, between 3 and 12 months, and greater than 12 months between abuse code documentation were selected from both cohorts. A representative sample of approximately 5% of pediatric charts with adult abuse codes was selected to rule out miscoding or other documentation anomaly. Targeted chart review subsets are listed in Table 2.
Figure
Figure
all relevant chart notes were considered. Encounters with indeterminate findings were marked as inconclusive. If multiple encounters existed for the same date or abuse code, such as an office visit and lab work drawn on the same date or multiple encounters coded by the laboratory group, all documentation was considered. Pennsylvania Department of Human Services child abuse reporting data and market share data estimating the percent of visits affiliated with this health care system per county were obtained to estimate the number of child abuse reports expected from health care providers within the system. Values were compared to the actual number of cases documented with maltreatment-associated diagnosis codes after filtering out encounters coded for suspected abuse and subsequent visits following an initial disclosure.
Increase in Laboratory Code Usage
T74.4XXA
T74.4XXD T74.4XXS
3. Number of Encounters by T74 Subcode. The drop in T74 coding observed beginning in 2019 was found to be primarily due to decreased usage of T74.22XA (child sexual abuse, confirmed, initial encounter). Prior to this decrease, an increase in T74.22XA was observed from 2013 to 2018.
Twenty-six records were chosen to explore the increase in coding of T74.22XA (Child sexual abuse, confirmed, initial encounter) by a specific laboratory group within the health system observed from 2013 to 2018. Each encounter contained only documentation of laboratory testing conducted. Labs included both nucleic acid and culture-based testing for chlamydia, gonorrhea, syphilis, HIV, hepatitis, and trichomonas as well as screening for pregnancy. Lab tests varied among ordering providers. Providers were determined to be associated with the Children’s Advocacy Center (CAC) through Google search and confirmation by a local expert pediatrician. Labs were likely drawn as part of a CAC investigation of alleged sexual abuse and run by the health system, though this may or may not have been the patient’s primary health care provider. Abuse was described in preceding or subsequent encounters in 11 of 26 records. The remaining 15 records were classified as inconclusive as no documentation of abuse was observed. Seven of 26 records contained multiple encounters with laboratory documentation of T74.22XA. Labs were redrawn
Table 2. Breakdown of Chart Review. Preliminary analysis identified three subsets requiring further evaluation: T74.22XA coding by a specific laboratory group, the use of adult maltreatment codes for pediatric patients, and repeat maltreatment coding across multiple encounters. Targeted chart reviews of each cohort of interest were conducted to evaluate coding rationale or identify potential documentation errors. Representative samples were selected based on prevalence within the data set and cohort complexity.
within 2 months of the initial encounter in 4 records and 3 of these patients had a positive initial test result, likely indicating a test of cure.
Adult Coding of Pediatric Visits
Five records were chosen to evaluate the use of adult codes for pediatric patients. Findings are summarized in Table 3. One encounter was conducted prior to the transition from ICD-9 to ICD-10 and was mapped
Figure
# of Encounters by T74 Subcode
to the updated code (T74.21XA). Two records had multiple encounters on the same day with the same code and all 5 were coded by the laboratory group. Patient ages ranged from 11 to 17. Although overlap exists between appropriate coding for patients aged 15 to 17, the 11-year-old patient was erroneously coded as an adult.
Repeat Abuse Coding
Ten records with repeat coding of unique maltreatment codes were evaluated at time intervals of less than 3 months, between 3 and 12 months, and greater than 1 year between encounters. Six records contained descriptions of abuse documented within visit notes. Abuse was inconclusive in 3 records: 2 encounters with low clinical suspicion for abuse despite parental concern and 1 with laboratory testing only. Of note, 2 records contained documentation of assault by peers (Y09) which was not considered abuse per the scope of this investigation. None of the records with repeat coding of distinct abuse codes contained documentation of repeated abuse. Repeated abuse was inconclusive in 1 record, with ongoing parental concern for abuse despite low clinical suspicion by the pediatrician. No difference in the accuracy of repeat coding was observed among records grouped by time between encounters. Findings are summarized in Table 4.
Forty-two records with repeat coding of identical maltreatment codes were evaluated at time intervals of less than 3 months, between 3 and 12 months, and greater than 1 year between encounters. Thirtyfive records contained documentation of abuse, 5 records were inconclusive for abuse due to insufficient documentation or laboratory testing only, and 2 described physical assault by peers not considered abuse in this investigation. Three records containing true repeat codes with less than 3 months between encounters were labeled inconclusive. All of these records involved laboratory testing only, possibly indicating repeat testing. Inconclusive records with encounters between 3 and 12 months apart included 2 records with insufficient documentation or laboratory testing only and one record with low clinical suspicion of repeat abuse. Two inconclusive records with encounters more than a year apart involved laboratory testing only with insufficient documentation of abuse. Of the 42 records in this cohort, 2 contained documentation of repeat abuse. Thirty-six records within this group contained subsequent encounters
E960.1 Rape 17 Lab Pediatrician
T74.21XS
T74.01XA
T74.01XA
T74.21XA
Adult sexual abuse, confirmed, sequela 15 Lab Gynecologist
Adult neglect or abandonment, confirmed, initial encounter 16 Lab
Adult neglect or abandonment, confirmed, initial encounter
Adult sexual abuse, confirmed, initial encounter
Table 3. Summary of Adult Coding for Pediatric Patients Findings. Five charts were selected to evaluate the use of adult codes for pediatric patients. All encounters were coded by the laboratory group and 2 were coded by additional departments. One chart involved an ICD-9 code which was mapped to the appropriate ICD-10 code. Patient ages ranged from 11 to 17.
< 3
No
< 3 months Yes Physical No
< 3 months Yes
< 3 months Yes
≥ 3 months and ≤ 12 months Yes
≥ 3 months and ≤ 12 months Yes
Sexual No
Sexual No
Physical No
Sexual No
≥ 3 months and ≤ 12 months Inconclusive Sexual Inconclusive
≥ 3 months and ≤ 12 months No N/A No*
> 12 months Inconclusive
> 12 months Yes
*Records indicate physical assault by peers.
Sexual No*
Physical Sexual No
Table 4. Summary of Distinct Repeat Code Findings. Ten charts were selected to evaluate the use of unique maltreatment codes across multiple encounters. Abuse was documented in 7 of these charts. Repeat abuse was not identified in any subsequent encounters although 1 chart was inconclusive as an initial instance of abuse could not be confirmed from EHR documentation. Two charts contained descriptions of assault by peers, which was not considered maltreatment in the context of this study.
coded with the XA suffix indicating an initial encounter for a particular diagnosis code. Evaluation of record documentation indicated that XD (subsequent encounter) or XS (sequelae) suffixes may also have been considered to define encounters in 17 of these records. Findings are summarized in Table 5.
Reasons for repeat abuse coding in cohorts with distinct and true repeat codes are summarized in Table 6. Among records containing distinct repeat
codes with less than 3 months between encounters, 3 of 4 records involved further evaluation of an initial disclosure of abuse: 1 for laboratory testing after alleged sexual abuse, 1 involving an interview by Children & Youth, and 1 after the original documentation of abuse was “misplaced”. The remaining record contained repeat coding for routine follow-up after an initial abuse disclosure without ongoing concern for abuse. In records with distinct repeat codes with 3 to 12 months between encounters, 3 of 4 involved routine or specialist follow-up without concern for ongoing abuse. The remaining record involved parental concern for abuse with low clinical suspicion which was deemed inconclusive. One of the 2 records with repeat coding of distinct maltreatment codes with more than 1 year between encounters involved only laboratory documentation after possible sexual abuse followed by a physical assault which was not considered abuse in this study. The remaining record involved ongoing parental concern for abuse which the patient denied upon interview.
*Two records indicate physical assault by peers.
Table 5. Summary of True Repeat Code Findings. Forty-two charts were selected to evaluate the use of exact repeat maltreatment codes across multiple encounters. Abuse was documented in 35 of these charts. Five records were inconclusive for abuse due to insufficient documentation or laboratory testing only, and 2 described physical assault by peers not considered abuse in this investigation. Three records containing true repeat codes with less than 3 months between encounters were labeled inconclusive as they contained laboratory testing only. Inconclusive records with encounters between 3 and 12 months apart included 2 records with insufficient documentation or laboratory testing only and 1 record with low clinical suspicion of repeat abuse. Two inconclusive records with encounters more than a year apart involved laboratory testing only with insufficient documentation of abuse. Of the 42 records in this cohort, two contained documentation of repeat abuse.
Repeat code type Time between encounters
months
True Repeat 3 months
Of the records with true repeat abuse coding less than 3 months apart, 13 involved routine or specialist follow-ups without documented concern for ongoing abuse. Six involved further evaluation of documented abuse, including Children & Youth interviews or laboratory testing, after the initial disclosure. The remaining 3 records contained multiple encounters for laboratory testing only without sufficient documentation of abuse. Nine records with repeat coding and 3 to 12 months between encounters involved routine or specialist follow-up after the initial disclosure of abuse. Of the remaining records in this cohort, 2 involved repeat
≥ 3 months and ≤ 12 months
> 12 months
Reasons for repeat coding
Second incident (50%)
Ongoing parental concern (50%)
Routine or specialist follow-up (59%)
Evaluation of initial disclosure (27%)
Repeat lab testing (13%)
Routine or specialist follow-up (69%)
Repeat lab testing (15%)
Ongoing parental concern (7%)
Repeat abuse (7%)
Routine or specialist follow-up (57%)
Repeat lab testing (28%)
Repeat abuse (14%)
Table 6. Reasons for Repeat Coding. The majority of encounters with distinct repeats less than 3 months apart involved further evaluation of an initial encounter detailing abuse. The remaining charts involved routine follow-ups not pertaining to abuse. Most distinct repeats with 3 to 12 months between visits documented routine or specialist follow-ups while the remainder involved evaluation for ongoing parental concern for abuse. Distinct repeats with greater than 12 months between visits were equally split between a second instance of abuse and ongoing parental concern. The majority of true repeat coding among all visit intervals involved routine or specialist follow-ups. Other reasons included further evaluation of the initial disclosure, repeat lab testing, ongoing parental concern, or, less frequently, repeat abuse.
lab testing without documentation of abuse, 1 involved ongoing parental concern, and 1 contained documented repeat abuse. Four records with encounters more than a year apart involved routine follow-up without concern for ongoing abuse. Two records in this cohort involved repeat lab testing
without documentation of abuse and the remaining record contained documentation of repeat abuse.
Projected Maltreatment Cases
Market share data approximating the percent of medical visits associated with this health care system across Pennsylvania in 2023 was obtained through Children and Youth. The Pennsylvania Department of Human Services reported that health care workers accounted for approximately 8,100 of the 41,360 child abuse reports in 2023 (9). This rough ratio and market share percentages were applied to the numbers of suspected and substantiated reports supplied by the Department of Human Services in 2023 to estimate the numbers of cases expected from health care workers within this system (Table 7). Approximately 51 cases of substantiated child maltreatment were anticipated to come from health care providers within the system in 2023. An approximate number of child maltreatment cases based upon ICD coding of abuse within EHRs was obtained. Any visit codes associated with the health care system’s labs or with ICD-10 codes containing the words "sequela" or "subsequent" were filtered out to reduce inflation by encounters with insufficient documentation of abuse or follow-up visits without ongoing maltreatment. This resulted in an estimated total of 96 cases of maltreatment from EHR documentation in 2023.
Discussion
Although the initial aim of this study was to compare national maltreatment underreporting trends to estimates of child maltreatment obtained through ICD-10 coding, the focus shifted toward investigating the reliability of EHRs for public health surveillance and the limitations of this application.
Documentation of over 5,300 patients receiving routine, emergency, and specialist evaluations within a large health care system serving 45 counties in Pennsylvania was analyzed. A decrease in ICD coding of child maltreatment was expected to be observed in 2020 in parallel with the precipitous drops reported nationally and statewide (7–9).
Instead, coding for abuse decreased drastically in 2019, with 75% of this decrease due to a reduction in use of T74.22XA (child sexual abuse, confirmed, initial encounter). The majority of T74.22XA codes applied in EHRs were determined to be coded by a specific lab group within the health system. Targeted chart reviews demonstrated that this testing was occurring in conjunction with Children’s Advocacy Center evaluation of alleged sexual abuse. Most of these encounters were standalone screenings for sexually transmitted infections and pregnancy without documentation of confirmed abuse elsewhere in the record. Several encounters appeared to involve repeat lab testing after an initial positive result. The increase and subsequent drop in the use of T74.22XA, however, remains undetermined as no change in laboratory protocol or coding procedure was identified. The accuracy of the use of the “confirmed” code in the workup of possible sexual abuse could not be assessed without requesting outside records. Inclusion of such encounters for public health surveillance data should be carefully considered as these “confirmed” cases may inflate estimated prevalence rates.
Because sexually transmitted infections (STIs) are uncommon in prepubertal children after sexual abuse, the American Academy of Pediatrics (AAP) does not recommend multi-site cultures if a child remains asymptomatic (31). Instead, AAP guidelines recommend evaluating risk on a case by case basis
Table 7. Estimated Suspected and Substantiated Reports of Child Maltreatment from Mandated Reporters within the Health Care System. Market share data estimating the percentage of provider visits associated with the health care system per region were obtained. The number of suspected and substantiated child abuse cases reported by The Pennsylvania Department of Human Services in 2023 and the market share data were used to estimate child abuse cases per region. Approximately 51 cases of substantiated abuse were projected within the health care system footprint.
prior to screening for STIs (31). The variety of laboratory tests observed within this cohort likely represents individual provider judgments after stratifying patient risk for exposure. Infections caused by Neisseria gonorrheae (genital, rectal or pharyngeal), syphilis (genital or rectal), Chlamydia trachomatis, Trichomonas vaginalis, or HIV likely indicate prepubertal abuse if perinatal transmission has been ruled out (32). Adolescents, who are at higher risk, should be screened for all STIs (32). Current guidelines recommend confirmation by culture when cases are being presented in a court of law. This requirement likely contributed to the high number of multi-site cultures observed in this cohort.
All 5 records selected for their use of adult codes for pediatric patients were coded by the laboratory group described above, as well as 2 encounters from additional sources. Only one of these records indicated erroneous coding of an 11-year-old patient. Per the AAP, adult codes may be applied to patients older than 15 (33). These same patients are eligible for pediatric codes until age 17, inclusive, which may complicate EHR surveillance of pediatric abuse. The majority of published studies using ICD codes to monitor maltreatment include only pediatric codes (13, 23, 30, 34). Although children between the ages of 12 and 17 demonstrate the lowest rates of child neglect, this age group is also associated with the highest incidence of sexual abuse (35). Relying on pediatric codes for surveillance of abuse within this cohort may underestimate its prevalence. Future studies should consider inclusion of adult codes filtered by patient age to account for this overlap.
The majority of repeat maltreatment codes documented within EHRs did not show evidence of ongoing or new instances of abuse. Duplicate abuse coding was most often carried over to follow-up visits after an initial disclosure. Repeat codes were also observed when additional investigation, including laboratory testing or Children and Youth evaluation, was required. Almost half of the records containing XA suffixes for initial evaluation of abuse within the identical repeat subset depicted follow-up encounters or evaluations of sequelae of abuse. According to the American Academy of Professional Coders (AAPC), the XA suffix should be applied when the patient is receiving active treatment for a condition (36). This may not always be a patient’s first visit with a provider and this code may be applied to multiple encounters
while receiving treatment. The ICD system includes the XD suffix and the XS suffix to define subsequent encounters and sequelae of diagnoses, respectively. AAPC guidelines recommend use of the XD suffix when a patient is receiving routine care or is in the recovery or healing phase for their condition. The ICD does not establish definitive criteria outlining these phases of treatment and tasks providers with making the distinction based on their assessments and plans. Patients may transition between active and recovery phases repeatedly depending upon their responses to treatments. The XS suffix for sequelae should only be applied when treating a “late effect” directly resulting from the initial diagnosis. For example, feeding difficulties or vision problems may require treatment after an initial shaken baby syndrome diagnosis. Per the AAPC, sequelae can only be coded after resolution of the acute phase of injury or illness.
Identifying actual instances of abuse rather than continued coding of historical abuse remains challenging when utilizing EHR-based datasets. Some studies monitoring ICD code usage such as Negriff et al. include only the first encounter for abuse and exclude all subsequent encounters (29). While this method may more accurately capture the incidence of abuse by removing duplicates, repeat abuse may still occur, as evidenced by the 2 cases observed in this study. In fact, a history of maltreatment has been clearly and consistently identified as a risk factor for future abuse, with recurrence rates of up to 50% (37). Because encounters with XA suffixes do not always capture unique or ongoing instances of abuse, inclusion of these encounters only does not greatly improve prevalence rate estimates. Additionally, pilot studies prior to the transition from ICD-9 to ICD-10 identified suffix coding as a common source of error when coding manually (38). While consideration of only first encounters may more accurately capture incidence rates, the risk of missing cases involving recurrent maltreatment remains.
Market share and Pennsylvania Department of Human Services data roughly estimated 51 cases of substantiated maltreatment in this health care system in 2023 while EHR data resulted in 96. The accuracy of regional utilization estimates of this health system cannot be determined which may impact projected cases. Furthermore, this value was obtained by assuming approximately 20% of all reports of abuse arise from mandated reporters within the
health care system. The 96 cases of maltreatment obtained by tracking ICD usage also had limitations. Attempts to exclude repeat coding of subsequent visits without ongoing abuse may not adequately filter out redundancies and may overlook repeat abuse to the same patient. In addition, this value cannot account for cases of abuse billed by injury or illness and not assigned maltreatment codes.
It is generally accepted that ICD codes and EHR documentation of child maltreatment underestimate true prevalence rates (13–16). Many barriers to accurate documentation and coding of child maltreatment contribute to this underreporting including: difficulty identifying abuse, coding or reporting uncertainties, time constraints, concerns about subsequent investigations, and fear of damaging the patient-provider relationship (16, 17, 19). In recent years, a 4-fold increase in the number of available ICD codes has further complicated rapid and accurate EHR documentation. The ICD coding dictionary has expanded from just over 14,000 codes in the ninth revision to include almost 70,000 codes in the tenth (39). This increase in both complexity and granularity aimed to streamline billing, improve patient safety monitoring, reduce health care fraud, guide public health initiatives, and enhance research endeavors (39). With such vast increases in complexity, providers and coders have expressed a need for computerassisted coding (40).
Despite the vastness of ICD codes, more accurate recording of abuse has been found in free-text sections of EHR rather than documented with diagnosis codes. Studies have shown that emergency department providers more frequently document suspicion for abuse after a fracture within the text of the EHR instead of as the encounter diagnosis (41). This phenomenon has been observed in encounters unrelated to abuse as well, with some studies finding that free text-notes within records served as better indicators of patient health status than ICD diagnoses (42). The chart reviews conducted in this study relied heavily on free text notes to determine whether maltreatment had occurred, even when the newly implemented “suspected” or “confirmed” abuse codes were utilized. In fact, Durand et al. found that providers use these specifiers inconsistently: 6% of cases with a negative forensic evaluation were overcoded as “suspected” while only 63% of verified abuse cases were assigned “confirmed” codes (43)
Accordingly, some researchers are applying artificial intelligence to analyze EHRs to improve identification and prevention of child maltreatment. Natural language processing is one application which combs through chart contents with the ability to synthesize imaging reports, patient-provided information, and documentation from providers, nurses, and social workers to identify language patterns indicative of abuse (44). Negriff et al. found that the use of natural language processing identified a rate of child maltreatment 10 times greater than the rate estimated using ICD codes (45). Lee et al. reported that natural language processing identified almost 300 cases of maltreatment while ICD codes captured only 111 (46). These models may also be trained to more accurately and quickly assign ICD codes to patient encounters (47). Similar methods have been used to predict child physical abuse or recurrence for expedited referral to Child Protective Services (44, 48).
Although these technological advances may improve documentation accuracy and the feasibility of public health surveillance using EHR data, accurate provider documentation remains the cornerstone to improving patient health outcomes.
Limitations
The current study had several limitations. Data was extracted from a single health care system in rural Pennsylvania which limits heterogeneity of the study population, sample size, and generalizability. ICD coding practices and medical evaluations conducted within this system may not be applicable to or representative of other institutions. Targeted chart reviews of each subset were conducted by 1 reviewer and may not adequately represent trends within the entire cohort. While the ICD codes selected to indicate maltreatment aligned with those used in recent published literature, it is well-established that providers frequently document abuse by coding for injuries sustained rather than maltreatment codes (34). It was assumed that a number of cases of abuse would be missed by considering only maltreatment codes as evaluation of sentinel injuries would require additional chart reviews beyond the scope of this study.
Conclusion
In summary, this study illustrated the limitations of ICD codes to reliably capture child maltreatment prevalence rates. Trends observed in EHR-based
datasets may represent systemwide protocols that lack generalizability or are inconsistent with national trends. Additional training for health care providers is needed to better recognize and document child maltreatment and allow for systematic evaluation using EHR data. Promising applications of artificial intelligence may further refine coding utilization and improve the detection or prevent child maltreatment.
Disclosures
All authors declare that they have no conflicts of interest to disclose.
Acknowledgments
This research was supported by the Medical Research Honors Program under the guidance of Ms. Tracey Pratt. The authors would like to extend their gratitude to Iris Johnston and the Health Sciences Library for their expert guidance and support in accessing resources for this manuscript.
References
1. U.S. Department of Health & Human Services AfCaF, Administration on Children, Youth and Families, Children’s Bureau. Child Maltreatment 2023 2023 [Available from: https://www.acf.hhs. gov/cb/data-research/child-maltreatment.
2. Leeb RT PL, Melanson C, Simon T, Arias I. Child Maltreatment Surveillance: Uniform Definitions for Public Health and Recommended Data Elements. Atlanta (GA): Centers for Disease Control and Prevention, National Center for Injury Prevention and Control; 2008.
3. Peterson C, Florence C, Klevens J. The economic burden of child maltreatment in the United States, 2015. Child Abuse Negl. 2018;86:178-83.
4. Goddard A. Adverse Childhood Experiences and Trauma-Informed Care. J Pediatr Health Care. 2021;35(2):145-55.
5. Rokach A, Clayton S. The Consequences of Child Abuse. Healthcare (Basel). 2023;11(11).
6. Rapp A, Fall G, Radomsky AC, Santarossa S. Child Maltreatment During the COVID-19 Pandemic: A Systematic Rapid Review. Pediatr Clin North Am. 2021;68(5):991-1009.
7. Marmor A, Cohen N, Katz C. Child Maltreatment During COVID-19: Key Conclusions and Future
Directions Based on a Systematic Literature Review. Trauma Violence Abuse. 2023;24(2):760-75.
8. Jud A, Orban E, Kaman A, Ravens-Sieberer U, Jarczok M, Li LY, et al. The effects of COVID-19 on the development of reported incidents of child maltreatment over time: A systematic literature review. Child Abuse Negl. 2024;157:107071.
9. Pennsylvania Department of Human Services. Child Protective Services Annual Report 2023 2023 [Available from: https://www.dhs.pa.gov/ docs/Publications/Pages/Child-Abuse-Reports. aspx.
10. Wosik J, Fudim M, Cameron B, Gellad ZF, Cho A, Phinney D, et al. Telehealth transformation: COVID-19 and the rise of virtual care. Journal of the American Medical Informatics Association. 2020;27(6):957-62.
11. COVID-19 Working Group. Successes and Lessons Learned in Responding to the Needs of Pediatricians, Children, and Families During the COVID-19 Pandemic. Pediatrics. 2024;153(6).
12. Huang N, Yang F, Liu X, Bai Y, Guo J, Riem MME. The prevalences, changes, and related factors of child maltreatment during the COVID-19 pandemic: A systematic review. Child Abuse Negl. 2023;135:105992.
13. Karatekin C, Almy B, Mason SM, Borowsky I, Barnes A. Documentation of Child Maltreatment in Electronic Health Records. Clin Pediatr (Phila). 2018;57(9):1041-52.
14. Raghavan R, Brown DS, Allaire BT, Garfield LD, Ross RE, Hedeker D. Challenges in using medicaid claims to ascertain child maltreatment. Child Maltreat. 2015;20(2):83-91.
15. Ravichandiran N, Schuh S, Bejuk M, Al-Harthy N, Shouldice M, Au H, et al. Delayed identification of pediatric abuse-related fractures. Pediatrics. 2010;125(1):60-6.
16. Woodman J, Allister J, Rafi I, de Lusignan S, Belsey J, Petersen I, et al. A simple approach to improve recording of concerns about child maltreatment in primary care records: developing a quality improvement intervention. Br J Gen Pract. 2012;62(600):e478-86.
17. Foster RH, Olson-Dorff D, Reiland HM, BudzakGarza A. Commitment, confidence, and concerns: Assessing health care professionals' child
18. Mandadi AR, Dully K, Brailsford J, Wylie T, Morrissey TK, Hendry P, et al. A national survey of pediatric emergency medicine physicians on improving education in child maltreatment recognition and reporting child abuse & neglect. Child Abuse Negl. 2021;122:105324.
19. Rovi S, Johnson MS. More harm than good? Diagnostic codes for child and adult abuse. Violence Vict. 2003;18(5):491-502.
20. Lane WG, Dubowitz H. Primary care pediatricians' experience, comfort and competence in the evaluation and management of child maltreatment: do we need child abuse experts? Child Abuse Negl. 2009;33(2):76-83.
21. Hunter AA, Livingston N, DiVietro S, Schwab Reese L, Bentivegna K, Bernstein B. Child maltreatment surveillance following the ICD10-CM transition, 2016-2018. Injury Prevention. 2021;27(3):251-6.
22. Hooft AM, Asnes AG, Livingston N, Deutsch S, Cahill L, Wood JN, et al. The Accuracy of ICD Codes: Identifying Physical Abuse in 4 Children's Hospitals. Acad Pediatr. 2015;15(4):444-50.
23. Hughes Garza H, Piper KE, Barczyk AN, Pérez A, Lawson KA. Accuracy of ICD-10-CM coding for physical child abuse in a paediatric level I trauma centre. Inj Prev. 2021;27(S1):i71-i4.
24. Sedlak AJ, Mettenburg, J., Basena, M., Petta, I., McPherson, K., Greene, A., and Li, S. Fourth National Incidence Study of Child Abuse and Neglect (NIS–4): Report to Congress. . 2010.
25. Nadan Y, Spilsbury JC, Korbin JE. Culture and context in understanding child maltreatment: Contributions of intersectionality and neighborhood-based research. Child Abuse Negl. 2015;41:40-8.
26. Schnitzer PG, Slusher P, Van Tuinen M. Child maltreatment in Missouri: combining data for public health surveillance. Am J Prev Med 2004;27(5):379-84.
27. Cort NA, Cerulli C, He H. Investigating health disparities and disproportionality in child maltreatment reporting: 2002-2006. J Public Health Manag Pract. 2010;16(4):329-36.
28. Salimi-Jazi F, Liang NE, Huang Z, Tennakoon L, Rafeeqi T, Trickey A, et al. Racial and Ethnic Disparities in Child Abuse Identification and Inpatient Treatment. JAMA Netw Open. 2024;7(12):e2451588.
29. Negriff S. Incidence of child maltreatment diagnosis in electronic health records of a large integrated healthcare system: 2001-2018. J Child Health Care. 2024;28(2):221-34.
30. Diyaolu M, Ye C, Huang Z, Han R, Wild H, Tennakoon L, et al. Disparities in detection of suspected child abuse. J Pediatr Surg 2023;58(2):337-43.
31. Jenny C, Crawford-Jakubiak JE. The evaluation of children in the primary care setting when sexual abuse is suspected. Pediatrics. 2013;132(2):e55867.
32. Adams JA, Kellogg ND, Farst KJ, Harper NS, Palusci VJ, Frasier LD, et al. Updated Guidelines for the Medical Assessment and Care of Children Who May Have Been Sexually Abused. J Pediatr Adolesc Gynecol. 2016;29(2):81-7.
33. American Academy of Pediatrics. ICD-10-CM Codes and Patient Age. AAP Pediatric Coding Newsletter. 2021;16(9):6-7.
34. Liu Y, Terrell L, Joyce B, Black J, Kaplan SJ, Gifford EJ. Defining suspected child maltreatment using International Classification of Diseases codes: A scoping literature review. Children and Youth Services Review. 2023;149:106907.
35. Guastaferro K, Shipe SL. Child Maltreatment Types by Age: Implications for Prevention. Int J Environ Res Public Health. 2023;21(1).
36. California Medical Association. Coding Corner: “Initial” vs. “subsequent” vs. “sequela” in ICD10-CM coding 2015 [Available from: https:// www.cmadocs.org/newsroom/news/view/ ArticleId/34855/Coding-Corner-Initial-vssubsequent-vs-sequela-in-ICD-10-CM-coding.
37. Hindley N, Ramchandani PG, Jones DP. Risk factors for recurrence of maltreatment: a systematic review. Arch Dis Child. 2006;91(9):744-52.
38. Moczygemba J, Fenton SH. Lessons learned from an ICD-10-CM clinical documentation pilot study. Perspect Health Inf Manag. 2012;9(Winter):1c.
39. Horsky J, Drucker EA, Ramelson HZ. Accuracy and Completeness of Clinical Coding Using ICD10 for Ambulatory Visits. AMIA Annu Symp Proc 2017;2017:912-20.
40. Butz J, Brick D, Rinehart-Thompson LA, Brodnik M, Agnew AM, Patterson ES. Differences in coder and physician perspectives on the transition to ICD-10-CM/PCS: A survey study. Health Policy and Technology. 2016;5(3):251-9.
41. Somji Z, Plint A, McGahern C, Al-Saleh A, Boutis K. Diagnostic coding of abuse related fractures at two children's emergency departments. Child Abuse Negl. 2011;35(11):905-14.
42. Juhra C, Schenkel J, Albert J, Butz N, Born J. In Case of Emergency - Are ICD-10 Codes Enough? Stud Health Technol Inform. 2015;208:195-9.
43. Durand MB, McLaughlin CM, Imagawa KK, Upperman JS, Jensen AR. Identifying Targets to Improve Coding of Child Physical Abuse at a Pediatric Trauma Center. J Trauma Nurs. 2019;26(5):239-42.
44. Annapragada AV, Donaruma-Kwoh MM, Annapragada AV, Starosolski ZA. A natural language processing and deep learning approach to identify child abuse from pediatric electronic medical records. PLoS One. 2021;16(2):e0247404.
45. Negriff S, Lynch FL, Cronkite DJ, Pardee RE, Penfold RB. Using natural language processing to identify child maltreatment in health systems. Child Abuse Negl. 2023;138:106090.
46. Lee RY, Landau AY, Heider PM, Hanson RF, Espeleta HC, Cato KD, et al. Estimating the Prevalence of Child Abuse and Neglect Among Adolescents in Primary Care Through Diagnoses Codes and Free-Text EHR Clinical Notes. J Pediatr Health Care. 2025;39(2):189-97.
47. Paleczny S, Osagie N, Sethi J, Cusimano D. Validity and reliability International Classification of Diseases-10 codes for all forms of injury: A systematic review. PLoS One. 2024;19(2):e0298411.
48. Horikawa H, Suguimoto SP, Musumari PM, Techasrivichien T, Ono-Kihara M, Kihara M. Development of a prediction model for child maltreatment recurrence in Japan: A historical cohort study using data from a Child Guidance Center. Child Abuse Negl. 2016;59:55-65.
A Serial Cross-Sectional Study of Older Adult Caregivers in the United States from 1997 to 2019
SooYoung HS VanDeMark1†
¹Geisinger College of Health Sciences, Scranton, PA 18509
†Doctor of Medicine Program
Correspondence: sooyoung.vdm@gmail.com
Abstract
Background: There is a current and projected discrepancy between the number of older adults requiring care in the United States and the number of health care professionals available to provide that care. This incongruity may shift care responsibilities onto informal, non-professional, and unpaid caregivers — a population that is itself aging. Although international research exists on older adult caregivers, there is a need to explore the U.S.-specific older adult caregiver (OAC, ≥ 64 years old) experience, given this country’s distinct values, culture, and policies surrounding aging and caregiving. This study investigates the demographics and descriptive context of OACs from 1997 to 2019 in the U.S., as well as compares and contrasts OACs with caregivers < 64 years old in 2019.
Methods: Repeated, cross-sectional study using secondary data from national surveys. Participants include 153, 156, 385, 529, and 537 U.S. OACs in 1997, 2004, 2009, 2014, and 2019 respectively, Statistical tests used: Jonckheere-Terpstra (J-T) and Cochran-Armitage tests for trend, chi-square tests of independence, and the Mann-Whitney U-test.
Results: In 1997, the median age of U.S. OACs was 70.0 years old, which rose to 73.0 in 2019. The percentage of OACs grew each year, with OACs being 10–13% of the respondents in 1997 and 2004 to roughly 30% in 2014 and 2019. OACs became more diverse in terms of race and sexual orientation, and more balanced in terms of sex, income, and education. The majority identify their health as good/very good, and most do not think caregiving has impacted their health. The median length of time caregiving decreased from 3 to 2 years from 1997 to 2019. Significant differences were found between OACs versus caregivers < 64 years old in 2019 in the number of instrumental activities of daily life performed; choice to become a caregiver; type of caregiver they
are; if they needed more information managing their own stress; and if they had plans in place for their own future care.
Conclusion: The role of OAC is becoming an increasingly new and shared experience among a more varied group of individuals in the U.S. than previously. Recommendations based on these results include screening patients ≥ 64 years old by a primary care provider (PCP) to determine if a patient identifies as a caregiver or is performing common caregiver duties. Earlier detection of caregiving status could lead to earlier interventions. Additional research should investigate the unique needs and circumstances of OACs in order to provide targeted interventions to improve their lives and the lives of their care recipients.
Introduction
Census data projects that by 2030, 1 in 5 Americans will be over the age of 64, with that estimate increasing to 1 in 4 by 2060 (1, 2). Although COVID-19 mortality rates disproportionately impacted those over the age of 64 (3), it has yet to be determined if or how this will impact mid-century population estimates; thus, it is reasonable to maintain that the U.S. population as a whole will increase in age over the next several decades. However, the U.S. population is not only aging but also undergoing other demographic changes. Additional population level projections include: greater gains in life expectancy for men than women; a longer life expectancy for foreign-born versus nativeborn individuals, regardless of race; and an increase in the number of the very oldest, those over 84 years old — nearly doubling by 2035, and nearly tripling by 2060 to 19 million (1, 2). The U.S. Census projects that by 2060, the average 65-year-old man or woman could expect to have 21 to 24 more years of life (1).
Meanwhile, the number of practicing geriatric health care professionals (e.g., physicians, physician assistants, registered nurses) is not keeping pace
with this demographic change (4–7). This shortage of geriatric-focused professionals is unlikely to be reconciled by 2030 (8). Research has found that older adults are more likely to have chronic health conditions and comorbidities, and thus, utilize the health system more than other age groups (1, 9–11). Importantly, audits of nursing homes, which are often permanent or stopgap residences for older adults, find inadequate numbers of staffing to provide quality care to patients (12). This imbalance in supply and demand creates a void in older adult patient care — a void that is very likely to shift responsibility onto informal, nonprofessional, and unpaid caregivers — a population that itself is aging.
Much of the literature on the current state of caregiving within the United States investigates the definition, identification, and burden of caregiving (13–15). Also common is research on caregivers who are specifically identified through the lens of the care recipient’s (CR) disease or disability (e.g., caregivers for patients with dementia). (16, 17) A 2019 systematic review on the health of caregivers who care for older adults emphasized the necessity for additional research, targeted interventions, and long-term study of caregiver subgroups to potentially counter the idea that caregiving leads to “negative health outcomes” (18). These caregiver subgroups could be defined by intrinsic independent variables (e.g., race or sex) to investigate for similarities and differences, but given the aging U.S. population and likely future shift toward informal, non-paid caregiving as previously mentioned, caregiver age — and older age specifically — seems especially timely and necessary.
Outside of the U.S., researchers have been investigating the older adult caregiver (OAC) subgroup. In Taiwan, a study found OACs used more home services for their care recipients’ (CRs) activities of daily living (ADLs) than younger caregivers. (19) Receiving extra assistance with nursing care — changing catheters, drawing blood — resulted in significant improvement in the OACs’ self-reported health (19). Additionally, older CRs self-reported better health if their caregiver was also older (19).
Another Taiwanese study assessed caregiver burden of OACs using the Neuropsychiatric Inventory Questionnaire (NPI-Q) and shortened Zarit Burden Interview (20). Lower income, more functional impairment of CRs’ ADLs, and higher severity in CRs
irritability and apathy were all independent predictors of increased OAC burden (20). Caregiver age was another independent predictor but with an inverse correlation to burden of caregiving (20). A 2021 Italian investigation used the Caregiver Burden Inventory (CBI) and found that physical burden of caregiving had a significant sex difference with females reporting worse physical health than males (21). OACs also experienced a moderate burden on their time and schedule (21).
A 2020 Swedish study of OACs found an association between caregiving and low mental “health-related quality of life” (22). Factors most associated with low quality of life for caregivers were low financial status, low cognitive ability (based on mini-mental state estimation (MMSE)), and worry about one’s own health (22).
A longitudinal Swedish study compared functional, physical, and mental health of OACs to older non-caregivers at baseline and after 6 years (23). Researchers created 4 tracks based on status (e.g., track 1: caregiver at baseline/caregiver at follow-up), and found high turnover rates in caregiver status after 6 years. Older age correlated with poorer functional, physical, and mental health in all tracks. In contrast to the other studies, no correlation was found between OAC health and low financial status nor gender (23).
These international studies highlight the differences within and between caregiver age groups; however, literature on OACs remains limited (19–21). Few contributions explore the American OACs experience, even though the U.S. has unique cultural norms and public policies regarding caregiving and aging than other countries. This study aims to rectify this gap in knowledge, promote visibility for the OAC subgroup, and provide useful data for designing possible solution to improve OAC health outcomes, which has been shown to positively impact CR health (24).
The hypothesis and specific aims were to evaluate if median OAC age in the U.S. shows a significant increasing trend over time from 1997 to 2019. Additionally, isolate OAC demographics from 1997 to 2019 and highlight any changes. Next, provide descriptive context about OACs and CRs from 1997 to 2019. Lastly, compare a subset of those contextual variables between OACs and younger caregiver (<63 years) in 2019 hypothesizing a significant difference between groups.
Methods
This study employed a repeated cross-sectional design using survey data collected in 1997, 2004, 2009, 2014, and 2019 on behalf of the National Alliance for Caregiving (NAC). Surveys were randomly administered to selected U.S. households via phone and online and conducted in English and Spanish. Deidentified survey data was downloaded from the NAC website, and codebooks to explain the data’s nomenclature were also included (25). Codebooks explaining the data’s nomenclature were included with these data sets.
Criteria for this study for all 5 years included: 1) identifications as a caregiver, 2) fully completed survey, and 3) age ≥ 64. A fourth criterion to include caregivers age 18–63 was only applied to the 2019 survey for age comparison analyses.
Demographic variables include OACs’ age, sex, race, sexual orientation, household income, education, and location. Contextual variables studied were CRs’ age and sex (note: CRs’ race was not collected in any survey); number of CRs per OAC; relationship of OAC to CR; primary illness of CR; current health status; impact of caregiving on one’s health; length of time caregiving; number of tasks completed, categorized as ADLs and instrumental activities of daily living (IADLs); OAC employment status; whether the caregiver felt they had a choice in taking on their responsibilities;
needing more information to manage one’s emotional/ physical stress; type of caregiver (e.g., sole, primary but not sole); if the caregiver had plans in place for their own future care; and weekly hours spent caregiving.
This study used IBM SPSS Statistics 29.0.2.0, Microsoft Excel 16.78.3, GraphPad Prism 10.4.0, and GraphPad QuickCalcs. A non-parametric, independent-sample Jonckheere-Terpstra test evaluated for numeric trends over time. The Cochran-Armitage test looked at categorical trends over time. Chi-square tests of independence analyzed categorical variables and the Mann-Whitney U-test evaluated variable medians. Responses recorded as unknown, refused, and/or do not know, and were <2% of the total responses, they were omitted from analysis. All statistical significance tests used a significance level (alpha) of 0.05 (5%). The Geisinger Institutional Review Board reviewed and determined this study not subject to oversight (#20220708).
Results
Sample sizes of OACs were 153, 156, 385, 529, and 537 for 1997, 2004, 2009, 2014, and 2019, respectively. In 1997, the median age of OACs was 70.0 years old, which rose to 73.0 in 2019, seen in Figure 1. The maximum age for an OAC in these surveys was 99 years old, occurring 6 times in 2004 and once in 2019. The median age of the oldest OACs (≥ 80 years old) was 81.0 in 1997 and 83.0 in 2019.
Figure 1. Median age and age range of older adult caregivers (≥ 64 years old) and oldest adult caregivers (≥ 80 years old). Boxes represent interquartile ranges, horizontal lines within boxes are median ages, and the extending bars display the minimum and maximum ages.
A. Older Adult Caregiver Age by Year
B. Oldest Older Adult (≥80) Caregiver Age by Year
A J-T test of median age did not show a statistically significant trend for either group — neither those ≥ 64 years old (p=.166) nor those ≥ 80 years old (p=.801). However, the percentage of OACs completing surveys grew each year with OACs being 10–13% of the respondents in 1997 and 2004 to roughly 30% in 2014 and 2019, as shown in Table 1. There was also growth among the oldest OACs from 9% in 1997 to 19% in 2019.
Additional OAC demographics from 1997 to 2019 are shown in Table 1. Sex distribution of OACs became more balanced over time. In 1997, OACs were only 24.2% male, but in 2019 were 43.0%. White, non-Hispanic OACs remained the majority across the years. Those identifying as “Asian, Pacific Islander, non-Hispanic OACs” or as “Other” saw increases from 5.9% to 11.5% and 0% to 4.8%, respectively, from 1997 to 2019. In that same time, OACs identifying as “Black, non-Hispanic” and as “Hispanic” decreased from 21.6% to 8.2% and 11.1% to 7.4%, respectively. In 2014 and 2019, 3.8% and 4.3% of OACs identified as lesbian, gay, bisexual, and/or transgender (LGBT). Other demographic changes include a shift from OAC households being majority low-income (<$50k) to more equal distribution across incomes; more OACs reaching a post-high school education; and an increasing majority of OACs living in non-rural areas.
The maximum age of CRs in each of the survey years was 97, 97, 99, 107, and 103 years old respectively, provided in Appendix Table 1, which shows characteristics of the CRs. In 2014, there were two 100-year-olds, five 101-year-olds, and one 107-year-old CRs, and in 2019, there were five 100-yearolds, one 101-year-old, four 102-yearolds, and two 103-year-olds CRs. OACs for these centenarians ranged from 65 to 81 years old in 2014 and 68 to 80 years old in 2019. In 2009, 4.7% of OACs cared for someone under 18 years old, while only 2% did in 2019. The 3 other surveys excluded caregivers of those < 18 years old. A chi-square test of independence found a statistically significant relationship
Sexa No. (%)
Table 1. Characteristics of Older Adult Caregivers (OACs). Percentages may not equal 100% due to rounding. aIn the 1997-2014 surveys, the only choices for caregiver sex were Male or Female. “Refused” was added to the 2019 survey, however no OAC selected that answer. bIn the 1997-2009 surveys, no question was asked regarding a caregiver’s sexual orientation or gender identity. In 2014, participants were asked if they identified as lesbian, gay, bisexual, or transgender (LGBT), and in 2019, the question more broadly asked if a participant identified as LGBT, other sexual orientation, and/or other gender identity. cIn the 1997 survey, no question was asked regarding caregiver’s location.
(
χ2=5.182, df=1, p=0.0228) in the number of CRs under and over 18 years old for 2009 versus 2019. Distribution of CR sex remained roughly 1/3 male to 2/3 female from 2009 to 2019.
The number of individuals cared for by OACs decreased over time. Most OACs had a single CR (1997, 76.5% and 2019, 81.4%) as shown in Appendix Table 2. The 3 most recent surveys found 7.5%, 4.2%, and 5.2% of OACs cared for both an adult and child.
Appendix Table 3 highlights all relationship types between OACs and CRs. The three most common relationships were “friend, neighbor, non-relative,” “spouse,” and “parent,” which changed positions over the years. “Spouse” overtook “parent” for the top spot in 2014 (34.2%) and 2019 (27.7%), shown in Figure 2. Figure 3 highlights a CR’s primary illness, with the 2 most common responses being “old age” and “Alzheimer’s, confusion, dementia, forgetfulness,” followed by “mobility,” “surgery,” “cancer,” and “mental Illness” in varying positions over time.
When OACs rated their current health status, the responses “good” and “very good” were the top choices every year, shown in Figure 4A. “Poor” decreased from 5.77% in 2004 to 1.30% in 2019 and “excellent” also declined nearly 50% from 2004 to 2019. Figure 4B shows that most OACs felt caregiving did not affect their health from 2004 to 2019. The percentage of OACs who felt caregiving made their health better has declined since 2014. The 1997 survey did not ask either question about health status.
The median length of time an OAC spent caregiving was 3 years in 1997, 2004, and 2009, and 2 years in 2014 and 2019. There was no significant trend over time in the length of time spent caregiving (p=0.83). There was no significant trend in the sum of ADLs and IADLs completed by OACs across the 5 years (p=0.480). On average, OACs performed nearly twice as many more IADLs per day than ADLs, as shown in Figure 5.
A significant association was found between OACs who were never employed while caregiving and those who were (χ2=36.39, df=1, p=<0.0001). In 1997, 35% of OACs were employed while caregiving, which decreased to 22% in 2019. No significant trend over time was found for whether OACs felt they had a choice in caregiving (yes/no) with responses roughly 50:50 each year (p=0.4204); and if OACs wanted more help with managing their own physical/emotional stress (yes/no) with the
Table 2. Analysis of Older Adult Caregivers Versus Caregivers < 64 Years Old in 2019. Note: Percentages may not equal 100% due to rounding. *Indicates significant association found with p-values < 0.05.
majority responding “no” each year (p=0.1318).
The 1997 survey did not ask either question.
For 2019, 1,739 responses by caregivers <64 years old were included. A significant difference was found in median number of IADLs performed by OACs versus younger caregivers (5 versus 4 respectively, U=297442, p=0.0082) but found no significant difference in median number of ADLs (p=0.9928) nor sum of ADLs and IADLs
The top 3 CR conditions, in descending order, for both OACs and younger caregivers were “old age,” “Alzheimer’s, confusion, dementia, forgetfulness,” and “mobility” in 2019. For OACs, these were followed by “surgery” and “cancer.” For the younger caregivers, “cancer” then “mental illness” rounded out the top 5.
Discussion
No significant trend found over time was found in the median age of OACs, as hypothesized, nor among the oldest OACs (≥ 80 years old). However, the percentage of OACs and oldest OACs grew each year, which is in concordance with U.S. Census population projections (1, 2). Sex distribution of OACs became increasingly even, possibly due to changing attitudes regarding stereotypical gender roles in the U.S., where women were assumed to take on the role of caregivers, as well as due to the greater projected life expectancy gains for men versus women. As more men live longer, they may find themselves becoming an OAC (1, 2).
(p=0.1367). Shown in Table 2, significant associations were found between caregiver age group and these variables: whether the caregiver felt they had a choice in caregiving (χ2=4.986, df=1, p=0.0256); the type of caregiver they are (χ2 =12.14, df =4, p=0.0164); if they wanted help managing their emotional/physical stress (χ2=16.67, df=1, p=<0.0001); and if plans were in place for their own future care (χ2=90.37, df=1, p=<0.0001).
There was no significant association comparing caregivers by age group and the following variables: current health status (p=0.1536); the impact caregiving had on caregiver’s self-reported health (p=0.1145); and the number of hours spent caregiving (<20 hours versus ≥ 20 hours) (p=0.6428), also shown in Table 2. There was no significant difference in median length of time caregiving for OACs versus younger caregivers (p=0.2495) with median length of time being 2 years for both.
Evolving demographics of OACs — more balanced in terms of sex, income, and education, and more diverse in race and sexual orientation — suggest that the OAC role is becoming a new shared experience among a more varied group of individuals in the U.S. With more people living this experience, it seems an opportune time for further OAC research. In light of these results, a recommended clinical application would be for PCPs would be to screen all patients ≥ 64 years old to determine if the patient identifies as an OAC or is performing common OAC responsibilities. Earlier detection of caregiving status could lead to earlier interventions such as connecting OACs with social workers, offering periods of respite, and emphasizing the need to prioritize self-care.
OACs care for a wide age range of CRs from children to centenarians, providing cross-generational care to parents, children, and sometimes, even grandchildren.
Figure 2. Top 5 relationships of older adult caregivers to their care recipient by year (i.e., “The care recipient is my…”)
Figure 3. Primary illness of an older adult caregiver’s care recipient by year
In 2019, many OACs (45%) identified as being the sole caregiver for their CRs, which is similar to younger caregivers (47%). Intergenerational caregiving may create unexpected and/or complex dynamics between families, e.g., an OAC may face legal barriers when acting as primary caregiver to their grandchild. This type of caregiving may also conflict with an OAC’s sense of self, e.g., identifying as a daughter to then identifying as a caregiver to a parent. Chen et al. found that older CRs (≥ 65 years old) in Taiwan self-reported better health if their caregiver was also older. Research exploring inter- and intragenerational caregiving could elucidate if one has more favorable outcomes for OACs and/or CRs.
The majority of OACs take care of 1 individual, which most recently is identified as the OACs’ spouse. As more people require informal caregiving, OACs may have to choose who to help and opt to care for a spouse over less-intimate relations. Another plausible factor could be the recent cultural and legal shifts surrounding LGBT relationships. Perhaps in later surveys, LGBT caregivers felt more comfortable identifying their CRs as a same-sex spouse instead of referring to them as a non-relative. Further research on the LGBT-identifying caregiver subgroup could help elucidate this point.
A. Caregiver Health Status Over Time
Most CRs of OACs and of the younger caregivers in 2019 need caregiving due to “old age.” More precise questioning regarding old age as a condition may differentiate not only why a CR requires care but also why an OAC does not. This may even help facilitate predictions as to when OACs would need future care themselves. One study contends it may not be the CR’s specific illness/disability that correlates to caregiver burden, but rather the tasks associated with it (15). A possible intervention is for PCPs to inquire about ADLs/IADLs performed, instead of a CR’s specific illness, and provide resources accordingly.
Studies have found contradicting results regarding caregiver burden and its impact on health (13–15, 18–21). In this study, OACs from 2004 to 2019 felt their health was unaffected by their role and rated their health as “good” or “very good” in those years. Individuals who are in good health may be more likely to become caregivers; these results may reflect selection bias (26). Similarly, due to social desirability bias, OACs may not report the negative impacts of caregiving. Without follow-up and independent verification, a self-reported status of “good health” or “no impact on health” will remain in question. Incorporating validated and commonly used tools within the surveys, such as the CBI, NPI-Q, and/or MMSE, would allow for better-quality comparisons and improved understanding regarding caregiver health and burden between studies.
Figure 4. Current Self-reported Health Status and Impact of Caregiving on Health Status by Year. Note: Question was not asked in the 1997 survey.
B. Impact of Caregiving on Health Status Over Time
Figure 5: Average Number of Activities of Daily Life (ADLs) and Instrumental ADLs (IADLs) Performed by an Older Adult Caregiver by Year. Examples.of ADLs include feeding, bathing, dressing, and transferring CRs. IADLs include providing medical care, managing finances, arranging services, shopping and meal preparation, doing housework, and transporting CRs.
OACs spend 2–3 years caregiving, which is comparable to younger caregivers. One study found a high rate of turnover in caregiver status over 6 years and recommended routine and repeat screening of older patients’ caregiver status. (23) That study also found that becoming a caregiver correlated to a positive impact on health (23), which suggests there may be a favorable time to identify, educate, and support OACs to potentially help mitigate any negative outcomes caused by caregiving.
Since 1997, OACs perform a 2:1 ratio of ADL:IADL tasks with an average of 6 tasks per day. ADLs include bathing, dressing, and transferring CRs. IADLs were defined as providing medical care (e.g., giving medications/injections), managing finances (e.g., paying bills, completing insurance claims), arranging services (e.g., home health aides, medical appointments), shopping and meal preparation, doing housework, and transporting CRs (27). This 2:1 ratio may reflect caregiving circumstance — if a CR needed assistance with multiple ADLs, which are often physical in nature, it might necessitate a higher level of care and prompt the CR’s transition from informal to formal care — thus, any caregiver surveyed would be likely perform a low number of ADLs. There was no significant difference in number of ADLs performed by OACs versus younger caregivers, which is perhaps
unexpected as younger caregivers would be presumed to be more able-bodied, and thus, more able to handle the physical demands of ADLs.
A minority (35%) of OACs were employed while caregiving in 1997 and that declined to 22% by 2019. People ≥ 64 years old in the U.S. are more likely to be retired, and older adults who are still employed may be less likely to take on caregiving responsibilities.
A Taiwanese study found OACs self-reported better health if they were unemployed (19). Future research could clarify why fewer OACs are working and caregiving simultaneously, and any related benefits or consequences of such.
The majority of OACs in 2019, unlike their younger counterparts, indicated not needing more help to address their stress. Half of OACs felt they chose to become a caregiver while younger caregivers felt they had less choice. Caregivers also significantly differed by age group in having plans in place for their future health care with OACs more likely to have them. One U.S. study observed >30% of OACs received care while also providing care to others (28). This dual role may prompt OACs to plan for their own future health care needs.
No significant difference was found in self-reported health status and in caregiving’s impact on health status between OACs and younger caregivers negating
Number of Activities
the hypothesis. Interestingly, no difference was found in hours spent caregiving per week, with 34.5% of OACs versus 33.2% of caregivers under 64 years spending more than 20 hours a week caregiving, which suggests that caregivers in their 70s and 80s spend equal time per week caregiving as someone half their age.
Multiple studies found that constraints on an OAC’s time and schedule are a particular stressor contributing to caregiving burden (21, 24). Another study found that it was actually the variety in responsibilities, not solely the hours spent caregiving, that was related to increased burden (15). PCPs could attempt to mitigate these challenges by combining OAC and CR health appointments when appropriate and viewing the OAC and CR as a dyad whose care outcomes are fundamentally linked (8, 24).
Limitations
Surveys are problematic due to: 1) response biases; 2) participant fatigue; and/or 3) inconsistencies in surveys within and across years. An example of irregularity occurred in the 2009 survey, where interviewers were instructed to input caregiver sex based on the sound of the participant’s voice and to not ask directly (25). Moreover, the only answer choices available to interviewers were “male” and “female,” with no option for “other” if the interviewer was unsure.
Conclusion
The primary objective of this research is to promote visibility and advocacy for a potentially vulnerable subgroup of caregivers in the U.S. There is an increasing percentage of caregivers who fall within the older adult age group (≥ 64 years old); and this group has become increasingly diverse in terms of sex, race, income and education. The role of OAC is becoming an increasingly new and shared experience among a more varied group of individuals and may represent an upcoming cultural and social paradigm shift. Some OACs are doing cross-generational caregiving and have multiple CRs which are likely to present with unique challenges. Lastly, significant differences exist in caregiver experience when investigating by age. One clinical recommendation is to screen all primary care patients ≥ 64 years old to determine if a patient identifies as an OAC or performs common OAC responsibilities. Earlier detection of caregiving status could lead to earlier interventions such as connecting OACs with social workers, offering periods of respite, and prioritizing self-care. Continued research
assessing the unique needs of American OACs has the potential to positively impact OACs and their CRs.
Disclosures
None reported.
Acknowledgments
Thank you to Elizabeth Kuchinski, MPH, Mushfiq Tarafder, PhD, MPH, MBBS, and Tina Hockenbury, DO, as well as the Geisinger College of Health Sciences’ Office of Research & Scholarship.
References
1. Medina L, Sabo S, Vespa J. Living Longer: Historical and Projected Life Expectancy in the United States, 1960 to 2060. 2020.
2. Vespa J, Medina L, Armstrong DM. Demographic Turning Points for the United States: Population Projections for 2020 to 2060. 2018, Rev 2020.
3. Goldstein JR, Lee RD. Demographic perspectives on the mortality of COVID-19 and other epidemics. Proceedings of the National Academy of Sciences. 2020;117(36):22035-41.
4. Center GWPS. Projected Future Need for Geriatricians. New York, NY: The American Geriatrics Society; 2017 February 2017.
5. Kozikowski A, Honda T, Segal-Gidan F, Hooker RS. Physician assistants in geriatric medical care. BMC Geriatr. 2020;20(1):449.
6. Smiley RA, Lauer P, Bienemy C, Berg JG, Shireman E, Reneau KA, et al. The 2017 National Nursing Workforce Survey. J Nurs Regul. 2018;9(3):S4-S88.
7. Rowe JW. The US eldercare workforce is falling further behind. Nat Aging. 2021;1(4):327-9.
8. Rose MS, Noelker LS, Kagan J. Improving policies for caregiver respite services. Gerontologist. 2015;55(2):302-8.
9. Boyd CM, Ritchie CS, Tipton EF, Studenski SA, Wieland D. From Bedside to Bench: summary from the American Geriatrics Society/National Institute on Aging Research Conference on Comorbidity and Multiple Morbidity in Older Adults. Aging Clin Exp Res. 2008;20(3):181-8.
10. He W, Larsen L. Older Americans with a disability, 2008-2012. American Community Survey Reports. Washington, DC; 2014.
11. Families Caring for an Aging America. In: Schulz R, Eden J, editors. Families Caring for an Aging America. Washington (DC)2016.
12. Mukamel DB, Saliba D, Ladd H, Konetzka RT. Association of Staffing Instability With Quality of Nursing Home Care. JAMA Network Open 2023;6(1):e2250389-e.
13. Taylor MG, Quesnel-Vallee A. The Structural Burden of Caregiving: Shared Challenges in the United States and Canada. Gerontologist 2017;57(1):19-25.
14. Kuharic M, Mulhern B, Sharp LK, Turpin RS, Pickard AS. Comparison of the EQ-HWB and EQHWB-S With Other Preference-Based Measures Among United States Informal Caregivers. Value Health. 2024;27(7):967-77.
15. Riffin C, Van Ness PH, Wolff JL, Fried T. Multifactorial Examination of Caregiver Burden in a National Sample of Family and Unpaid Caregivers. J Am Geriatr Soc. 2019;67(2):277-83.
16. Sabatini S, Turner SG, Morris RG, Opdebeeck C, Thom JM, Hunt A, et al. Correlates of felt age in caregivers of people with dementia: findings from the IDEAL study. Front Psychol. 2023;14:1287842.
17. Lowers J, Brus K, Smith C, Kavalieratos D, Hepburn K, Perkins MM. "How do You Take that Much Time for One Person's Life?" Experiences of Dementia Caregivers Who are Not Immediate Family. J Appl Gerontol. 2025;44(1):166-75.
18. Bom J, Bakx P, Schut F, van Doorslaer E. The Impact of Informal Caregiving for Older Adults on the Health of Various Types of Caregivers: A Systematic Review. Gerontologist 2019;59(5):e629-e42.
19. Chen MC, Kao CW, Chiu YL, Lin TY, Tsai YT, Jian YZ, et al. Effects of home-based long-term care services on caregiver health according to age. Health Qual Life Outcomes. 2017;15(1):208.
20. Tsai CF, Hwang WS, Lee JJ, Wang WF, Huang LC, Huang LK, et al. Predictors of caregiver burden in aged caregivers of demented older patients. BMC Geriatr. 2021;21(1):59.
21. Spatuzzi R, Giulietti MV, Romito F, Reggiardo G, Genovese C, Passarella M, et al. Becoming an older caregiver: A study of gender differences in family caregiving at the end of life. Palliat Support Care. 2022;20(1):38-44.
22. Fagerstrom C, Elmstahl S, Wranker LS. Analyzing the situation of older family caregivers with a focus on health-related quality of life and pain: a cross-sectional cohort study. Health Qual Life Outcomes. 2020;18(1):79.
23. Wranker LS, Elmstahl S, Cecilia F. The Health of Older Family Caregivers - A 6-Year Follow-up. J Gerontol Soc Work. 2021;64(2):190-207.
24. Sabo K, Chin E. Self-care needs and practices for the older adult caregiver: An integrative review. Geriatr Nurs. 2021;42(2):570-81.
25. Caregiving in the U.S. 2009, 2015, 2020Public Use Data Files. In: (NAC) TNAfC, editor. Caregiving: The National Alliance for Caregiving and the AARP Public Institute.
26. Shaffer KM, Nightingale CL. Comparison of Healthcare Utilization Between Informal Caregivers and Non-Caregivers: An Analysis of the Health Information National Trends Survey. J Aging Health. 2020;32(5-6):453-61.
27. Caregiving in the U.S. 2009, 2015, 2020Public Use Data Files: The National Alliance for Caregiving and the AARP Public Institute; [Available from: https://www.caregiving.org/ research/open-data/].
28. Semere W, Yank V, Lisha NE, Lindquist LA, Huang AJ. Older adults with overlapping caregiving responsibilities and care needs in a U.S. national community-based sample. J Am Geriatr Soc. 2024;72(6):1824-30.
Appendix
Don’t Know/ Refused/ Unknown
Appendix Table 1. Characteristics of Care Recipients (CRs) Whose Caregivers Are ≥ 64 Years Old. Percentages may not equal 100% due to rounding. Race/ethnicity data of care recipient was not collected in any year’s survey. a1997: One response not included because CR’s age was given as “do not know.” Five responses from 2004 and 3 responses from 2009 not included for the same reason. bThe 1997, 2004, 2014 surveys only interviewed caregivers of adults (CR is ≥ 18 years old). cFor CR sex, non-binary was added as a choice in 2019; however, no caregiver selected this answer.
a caregiver to an adult and child?b
Appendix Table 2. Number and Type of Care Recipients of Older Adult Caregivers. Percentages may not equal 100% due to rounding. aRemoved 1 “Do not know” response for 2004. bSurvey question not asked in 1997 and 2004.
Appendix Table 3. Relationship of an Older Adult Caregiver to Their Care Recipient (“The care recipient is my…”). Percentages may not equal 100% due to rounding. aIncludes: nephews, nieces, cousins, grandparent-in-laws and other non-specified relatives.
Afferent Baroreflex Failure Following Neck Radiation: A Case Report
Morgan Glasser1† and Blake Garmon1,2
1Geisinger College of Health Sciences, Scranton, PA 18509
2Geisinger Medical Center, Danville, PA 17822
†Doctor of Medicine Program
Correspondence: mglasser@som.geisinger.edu
Abstract
Afferent baroreflex failure is a rare but serious complication of neck irradiation that causes extreme blood pressure variability. We present the case of a 72-year-old man with oropharyngeal cancer treated with chemotherapy and neck radiation who developed severe blood pressure lability several years after therapy. After a thorough workup ruled out other possible causes, a clinical diagnosis of afferent baroreflex failure was made, despite the absence of formal autonomic testing. Management included long acting central sympatholytics together with as needed medications to treat both hypotensive and hypertensive episodes. This case demonstrates a physiology based treatment strategy as outlined by the Vanderbilt Autonomic Dysfunction Center in the absence of formal guidelines.
Introduction
The baroreflex provides constant and immediate regulation of blood pressure. Baroreceptors are stretch sensitive nerve endings in the carotid sinus and aortic arch. When blood pressure rises, afferent signals increase. This stimulates parasympathetic activity and reduces sympathetic output. Disruption of these pathways by injury to the carotid sinus nerves or autonomic ganglia can cause exaggerated hypertensive responses and heightened sensitivity to vasopressors (1–4).
Neck irradiation is responsible for nearly all cases of afferent baroreflex failure (5). Subclinical impairments may appear soon after radiation therapy (6, 7). Clinically significant dysfunction occurs years later due to progressive fibrosis and damage to the baroreceptor pathways (8). The incidence of this complication remains uncertain.
Afferent baroreflex failure is marked by severe blood pressure instability with alternating hypertensive
crises. Patients often experience hypertensive episodes triggered by mental stress or exertion (9). These episodes are frequently accompanied by facial flushing (10). All patients exhibit hypertension but only some also experience hypotensive episodes. The absence of hypotensive events in some patients may be due to compensatory mechanisms involving aortic baroreceptors and vestibular afferents (11).
Diagnosis of afferent baroreflex failure is primarily clinical and depends heavily on a thorough history, particularly prior neck surgery or radiation, along with the exclusion of alternative diagnoses such as essential hypertension, pheochromocytoma, sympathomimetic drug use, panic disorder, and postural tachycardia syndrome. Ambulatory monitoring of blood pressure and heart rate can reveal the hallmark feature of marked blood pressure variability accompanied by an inappropriate or absent heart rate response. A definitive diagnosis typically requires formal autonomic testing with continuous hemodynamic monitoring during pharmacologic provocation using agents such as phenylephrine and nitroprusside. Patients with afferent baroreflex failure exhibit exaggerated blood pressure responses without the expected compensatory changes in heart rate, reflecting impaired afferent baroreceptor signaling.
Case Presentation
A 72-year-old man presented with significant blood pressure variability. His medical history included severe aortic stenosis, type 2 diabetes mellitus, atrial fibrillation, hypothyroidism, and dyslipidemia. Importantly, the patient received definitive chemotherapy and radiation without surgical resection for stage T2N1M0, p16-positive squamous cell carcinoma of the right tonsil with extension into the soft palate and involvement of right level II lymph nodes. Intensity-modulated radiation therapy was
delivered to a total dose of 6,750 centigray in 30 treatment sessions. The radiation plan included contouring of the primary tumor in the right tonsil with extension into the soft palate, as well as involved right-sided lymph nodes, with dose calculations accounting for adjacent normal tissues. Cisplatin was administered weekly during the course of radiation therapy.
Approximately 4 years after radiation, he began to experience marked blood pressure lability. The patient completed chemoradiation in June 2018. Symptoms of labile blood pressure began in late 2022. He was first evaluated in December 2022. A 24-hour ambulatory blood pressure reading was performed in April 2023. Symptoms have been persistent through the present.
Upon presentation, home readings showed systolic values ranging from 71 to 192 mmHg. Hypotensive episodes were often postural and associated with dizziness, headache, and near syncope. His symptoms were initially attributed to severe aortic stenosis, but they persisted despite surgical valve replacement. At the time of presentation, his antihypertensive regimen included daily oral lisinopril 20 mg and metoprolol succinate ER 200 mg.
Ambulatory blood pressure monitoring revealed a systolic variation of 120 mmHg over a single day, with preservation of the normal nocturnal dip. Figure 1, 24-hour ambulatory blood pressure and heart rate monitoring, displays the 24-hour ambulatory blood pressure and corresponding heart rate data, demonstrating significant blood pressure variability without appropriate compensatory changes in heart rate. Further evaluation included a 24-hour urinary fractionated metanephrine test, which was within normal limits. Laboratory studies showed a creatinine level of 1.2 mg/dL, sodium 133 mmol/L, potassium 5.0 mmol/L, and TSH 2.57 mIU/L. Urine studies revealed an albumin-to-creatinine ratio of 45 mg/g, 24-hour urine creatinine of 1.201 g, and urine sodium of 108 mmol. Urinalysis was unremarkable, and renal artery duplex imaging showed no evidence of renovascular disease.
Discussion
The characteristic blood pressure lability and delayed onset of symptoms following neck irradiation, along with the exclusion of other potential causes, supported a diagnosis of afferent baroreflex failure.
Figure 1. 24-hour ambulatory blood pressure and heart rate monitoring
Although formal autonomic testing was not conducted, the clinical presentation was highly suggestive of afferent baroreflex failure, warranting intervention due to recurrent episodes of both hypertension and hypotension. Nonetheless, the absence of formal autonomic testing remains a significant limitation of this study.
Managing afferent baroreflex failure is challenging due to the unpredictable nature of blood pressure fluctuations. While there are no formal guidelines, the Vanderbilt Autonomic Dysfunction Center outlines a comprehensive, physiology-based treatment strategy in their 2019 JACC review article titled “Blood Pressure Management in Afferent Baroreflex Failure.” This approach focuses on reducing blood pressure variability and improving patient quality of life (11).
The primary treatment involves the use of long-acting central sympatholytic agents, such as guanfacine or methyldopa, to mitigate sympathetic surges. These medications help prevent significant hypertensive spikes. Short-acting sympatholytics, like clonidine, are generally avoided due to the risk of rebound hypertension, though they may be considered for acute hypertensive episodes triggered by stress or exertion. For managing hypotension, the strategy includes increasing salt and fluid intake, the use of midodrine, employing physical counter-maneuvers, and maintaining a supine position with legs elevated. Additionally, for hypertensive crises induced by stress, interventions such as biofeedback, benzodiazepines, or cannabinoids may be beneficial. The overarching goal is not to normalize blood pressure but to reduce the amplitude of hypertensive spikes and minimize episodes of hypotension, thereby enhancing patient well-being.
Our patient is currently being treated with lisinopril 10 mg orally once daily, metoprolol succinate 100 mg orally once daily, amlodipine 5 mg orally twice daily, and guanfacine extended release 1 mg orally at bedtime. As needed, labetalol 100 mg orally is used for hypertensive crises, and midodrine 2.5 mg orally is administered for symptomatic hypotension. Since initiating amlodipine and guanfacine, hypertensive episodes have become less frequent. Mild afternoon hypotension still occurs but is better managed with intermittent midodrine.
Conclusion
Afferent baroreflex failure is a rare but serious late complication of head and neck radiation. Diagnosis requires detailed clinical history, exclusion of mimicking disorders, and recognition of its hallmark feature of marked blood pressure variability years after radiation therapy. Management focuses on attenuating sympathetic surges and preventing hypotensive symptoms. This case highlights the diagnostic and therapeutic complexities of afferent baroreflex failure and the importance of individualized, physiology driven treatment strategies.
Disclosures
The authors have no disclosures.
Acknowledgments
None.
References
1. Kezdi P. Sinoaortic regulatory system; role in pathogenesis of essential and malignant hypertension. AMA Arch Intern Med. 1953;91:26–34.
2. Fagius J, Wallin BG, Sundlöf G, Nerhed C, Englesson S. Sympathetic outflow in man after anaesthesia of the glossopharyngeal and vagus nerves. Brain. 1985;108(Pt 2):423–38.
3. Shannon JR, Jordan J, Black BK, Costa F, Robertson D. Uncoupling of the baroreflex by N(N)-cholinergic blockade in dissecting the components of cardiovascular regulation. Hypertension. 1998;32:101–7.
4. Jordan J, Tank J, Shannon JR, Diedrich A, Robertson D, Biaggioni I. Baroreflex buffering and susceptibility to vasoactive drugs. Circulation. 2002;105:1459–64.
5. Heusser K, Tank J, Luft FC, Jordan J. Baroreflex failure. Hypertension. 2005;45:834–9.
6. Timmers HJ, Wieling W, Soetekouw PM, Bleijenberg G, van der Meer JW, Lenders JW. Hemodynamic and neurohumoral responses to head-up tilt in patients with chronic fatigue syndrome. Clin Auton Res. 2002;12:273–80.
7. Huang CC, Huang TL, Hsu HC, Hsu YH, Chen JC, Lin CC. Long-term effects of neck irradiation on cardiovascular autonomic function: a study in nasopharyngeal carcinoma patients after radiotherapy. Muscle Nerve. 2013;47:344–50.
8. Sharabi Y, Dendi R, Holmes C, Goldstein DS. Baroreflex failure as a late sequela of neck irradiation. Hypertension. 2003;42:110–6.
9. Norcliffe-Kaufmann L, Palma JA, Kaufmann H. Mother-induced hypertension in familial dysautonomia. Clin Auton Res. 2016;26:79–81.
10. Robertson D, Hollister AS, Biaggioni I, Netterville JV, Mosqueda-Garcia R, Robertson RM. The diagnosis and treatment of baroreflex failure. N Engl J Med. 1993;329:1449–55.
11. Biaggioni I, Shibao CA, Diedrich AD, Muldowney JA 3rd, Laffer CL, Jordan J. Blood pressure management in afferent baroreflex failure: JACC review topic of the week. J Am Coll Cardiol. 2019;74(23):2939–47.
Paradoxical Acute Angle Closure Following Topiramate Discontinuation: A Case Report
Carolyn Young1† and Tatiana Franco2
¹Geisinger College of Health Sciences, Scranton, PA 18509
†Doctor of Medicine Program
2Division of Ophthalmology, Geisinger, Pittston, PA 18505
Correspondence: cyoung01@som.geisinger.edu
Abstract
Background: Topiramate is a widely prescribed antiepileptic medication increasingly used for offlabel indications such as migraine prevention, weight loss, and mood stabilization. While topiramate-induced acute angle closure (AAC) is rare, it has become more frequently recognized due to its expanding use. This condition is typically associated with ciliochoroidal effusion and anterior displacement of the lensiris diaphragm. Although most cases occur shortly after therapy initiation, emerging reports suggest angle compromise may also occur after abrupt drug discontinuation.
Methods: We describe 2 cases of bilateral AAC in female patients without prior history of glaucoma. Case 1 presented 2 days after discontinuing topiramate, while Case 2 developed symptoms during active treatment. Both cases were evaluated with slit lamp examination, gonioscopy, tonometry, optical coherence tomography (OCT), and ultra-widefield fundus imaging.
Results: Case 1 showed shallow anterior chambers, optic disc cupping, and normal intraocular pressure (IOP) despite recent drug withdrawal. Case 2 demonstrated acute vision loss, significant myopic shift (>5.00 D), and IOP elevation up to 40 mmHg. Anterior segment OCT revealed quadrant-specific angle narrowing. Both patients improved with discontinuation of topiramate and appropriate medical therapy, including aqueous suppressants and cycloplegics. Anatomical normalization and refractive recovery were documented over follow-up.
Conclusion: These cases broaden the known spectrum of topiramate-induced angle closure by highlighting both paradoxical post-discontinuation and classic acute presentations. Clinicians should maintain a high index of suspicion for AAC in patients with recent or
current topiramate exposure, regardless of IOP. Early recognition and ophthalmology referral are essential to prevent permanent vision loss.
Introduction
Topiramate is a sulfamate-substituted monosaccharide with multiple mechanisms of action, including enhancement of GABA activity, inhibition of voltage-gated sodium channels, and mild carbonic anhydrase inhibition. It is FDA-approved for epilepsy and migraine prophylaxis, but its use has expanded significantly in recent years due to off-label applications in psychiatric disorders, weight loss, and chronic pain management (1–3). As its clinical use broadens, reports of rare but vision-threatening ocular side effects—including acute angle closure and myopic shift—have become more prevalent (4).
Topiramate-induced angle closure is a secondary, nonpupillary block mechanism that differs fundamentally from primary angle closure glaucoma. Rather than iris bombe or anatomical crowding, the underlying pathology involves ciliochoroidal effusion and anterior rotation of the ciliary body, which displaces the lensiris diaphragm forward (4–5). This leads to anterior chamber shallowing, appositional angle closure, and refractive changes such as sudden-onset myopia. While intraocular pressure is often elevated, it may be normal or only mildly increased, contributing to diagnostic uncertainty (6).
Most cases occur within the first two weeks of initiating topiramate therapy and present with acute bilateral visual blurring, eye pain, halos, and headache (7). Prompt recognition and drug cessation typically result in rapid symptom resolution and reversal of anatomical changes. However, emerging reports describe more atypical presentations, including cases occurring after a missed dose or after abrupt discontinuation of the drug (8).
These paradoxical variants can be more subtle, sometimes presenting without pain, without elevated intraocular pressure, or with delayed onset—raising the risk of misdiagnosis or inappropriate treatment.
In this report, we present 2 cases that illustrate the clinical spectrum of topiramate-induced angle closure. The first case involves a paradoxical subacute presentation following abrupt drug discontinuation, in which visual symptoms developed despite normal intraocular pressures and a lack of ocular pain. The second case represents a more classic acute presentation shortly after therapy initiation, with bilateral myopic shift, angle narrowing, and elevated intraocular pressure (IOP). Together, these cases underscore the importance of recognizing variable presentations of topiramate-induced angle closure, and they highlight the role of detailed medication history, multimodal imaging, and timely management in preventing permanent vision loss.
Case Presentation
Case 1
A 37-year-old female with no significant ophthalmic history presented to the Emergency Department (ED) at Geisinger Wyoming Valley Medical Center in October 2023 with bilateral photopsia, visual disturbances described as halos and flare-like patterns around lights, and a dull right-sided headache. The patient also noted a sudden shift in her vision, characterized by increasing difficulty with distance focus. These symptoms developed 2 days after abrupt discontinuation of topiramate, which she had initiated 4 days earlier for migraine prophylaxis and weight management. She denied any history of glaucoma, ocular trauma, autoimmune disease, or systemic inflammatory conditions.
Initial visual acuity was significantly reduced to 20/300 in the right eye (OD) and 20/150 in the left eye (OS), improving to 20/100 OD and 20/70 OS with pinhole correction. Intraocular pressures (IOP) measured by Goldman applanation tonometry were within normal limits: 20 mmHg OD and 18 mmHg OS. Pupils were equal, round, and reactive to light with no afferent pupillary defect. Extraocular motility was full, and visual fields were intact to confrontation. Slit-lamp examination revealed normal lids and lashes, clear corneas with no staining on fluorescein testing, and bilaterally shallow anterior chambers without
cell or flare. The irides were round and reactive, and the lenses were clear. Gonioscopy revealed closed angles in all quadrants bilaterally, with no visible trabecular meshwork or angle structures. Fundoscopic examination revealed large optic nerves with cup-todisc ratios of 0.8 OD and 0.9 OS. The peripheral retina was notable for lattice degeneration in the right eye; maculae and vasculature appeared normal bilaterally.
A broad differential diagnosis was considered. Primary acute angle-closure glaucoma (AACG) was initially suspected due to the shallow anterior chambers and optic nerve cupping; however, this diagnosis was less likely in the absence of elevated IOP, ocular pain, corneal edema, or mid-dilated fixed pupils. The bilateral and symmetric nature of the presentation further favored a secondary angle-closure mechanism. Optic neuritis was also considered given the visual complaints and headache but was ruled out due to the absence of pain with eye movement, preserved color vision, intact visual fields, and normal optic disc appearance aside from cupping. Posterior uveitis was considered but deemed unlikely due to the absence of vitreous inflammation or retinal lesions. Retinal pathology such as central serous chorioretinopathy was also considered but lacked corresponding macular findings on examination.
Ancillary imaging was obtained to further characterize the ocular findings. Optical coherence tomography (OCT) of the retinal nerve fiber layer (RNFL) demonstrated normal thickness across all quadrants in both eyes, with no evidence of glaucomatous damage. Macular OCT revealed preserved foveal contour and intact ganglion cell–inner plexiform layer (GC-IPL) thickness, with no signs of cystoid macular edema, subretinal fluid, or retinal layer disruption (Figure 1). Ultra-widefield fundus imaging demonstrated peripapillary nerve fiber layer striations and suspected vitreoretinal traction: the right eye exhibited localized bundle loss, while the left eye showed retinal striae suggestive of mechanical stress (Figures 2 and 3). Anterior segment OCT revealed significant narrowing and iridocorneal apposition across multiple quadrants bilaterally, confirming anatomical angle closure despite the normal IOP (Figure 4). Standard automated perimetry (Humphrey 24-2) conducted 3 months later showed mild functional asymmetry, with the right eye demonstrating a glaucoma hemifield test result of "Outside Normal Limits" and the left eye "Within Normal Limits." Test reliability was limited by a high rate of false positives, though the findings were
optical coherence tomography
macular
for
1, demonstrating intact foveal contour and preserved retinal architecture in both eyes. No subretinal or intraretinal fluid, macular edema, or outer retinal disruption is observed, supporting structural recovery following the resolution of topiramate-induced acute angle closure.
3. Case 1 widefield fundus imaging of both eyes obtained via Optomap
consistent with mild visual field loss corresponding to the observed optic nerve cupping.
Additional systemic evaluation was conducted to investigate potential non-ophthalmic contributors. Bloodwork revealed elevated inflammatory markers, including a C-reactive protein of 75 mg/L and a white blood cell count of 14.98 ×109/L. Magnetic
resonance imaging (MRI) of the brain demonstrated an enlarged empty sella, but no demyelinating lesions, mass effect, or signs of elevated intracranial pressure. There was no evidence of central nervous system inflammation or compressive pathology. Given the elevated inflammatory markers, the patient was referred to rheumatology and hematology for further evaluation, though no definitive systemic diagnosis was established.
In the context of recent topiramate use followed by abrupt discontinuation, the bilateral angle closure, shallow anterior chambers, myopic shift, and confirmatory imaging findings supported the diagnosis of paradoxical topiramateinduced angle closure. The normal IOP and absence of acute pain or corneal changes suggested a more insidious course. The anatomical findings were consistent with ciliochoroidal effusion leading to anterior displacement of the ciliary body and forward displacement of the lens-iris diaphragm. This mechanism is well described in topiramateassociated ocular syndromes and distinguishes it from primary angleclosure glaucoma. The patient’s symptoms gradually improved with conservative management, including close ophthalmologic monitoring and permanent discontinuation of the medication. No surgical intervention or topical therapy was required.
Case 2
A 43-year-old female presented to the ED in March 2025, with acute bilateral vision loss, excessive tearing, and ocular irritation that began approximately 12 hours prior to evaluation. She denied eye pain, photophobia, trauma, or contact lens use. Her ocular history was notable only for bilateral laser-assisted in situ keratomileusis (LASIK) approximately 20 years prior. One week before presentation, she had initiated topiramate 25 mg daily for weight gain attributed to prior psychotropic medication use. She reported that she had not taken her full dose the previous day.
Figure
ultrawidefield retinal imaging. The left image (OD) demonstrates retinal nerve fiber layer (RNFL) striations with blue arrows indicating the direction of nerve fiber bundle loss. The right image (OS) reveals retinal striae suggestive of vitreoretinal traction or mechanical stress.
Figure 1. Macular optical coherence tomography (OCT) ganglion cell analysis for Case 1, demonstrating preserved ganglion cell-inner plexiform layer (GC-IPL) thickness in both eyes. The values fall within expected ranges, with no significant thinning or asymmetry. These findings support the absence of structural retinal ganglion cell damage despite the acute angle closure episode.
Figure 2. Spectral-domain
(SD-OCT) horizontal
scans
Case
At presentation, uncorrected visual acuity was reduced to 20/100 in both eyes. Pupils were equal and reactive without relative afferent defect. Extraocular motility was intact. Slitlamp examination revealed moderate conjunctival injection and chemosis but no corneal epithelial defect or anterior chamber inflammation. Applanation tonometry revealed markedly elevated intraocular pressures, measured at 39 mmHg OD and 40 mmHg OS. Anterior chambers were shallow in both eyes. Based on the acute, bilateral nature of the symptoms, elevated IOP, and recent topiramate exposure, the leading diagnosis was secondary angle-closure glaucoma, likely medication-induced.
Initial management included cessation of topiramate and initiation of topical therapy consisting of brimonidine tartrate 0.2% and timolol 0.5% twice daily, prednisolone acetate 1% three times daily, and cyclopentolate once daily. On followup the next day, uncorrected visual acuity had decreased to 20/250 OD and 20/300 OS but improved with pinhole to 20/50 OU. Pupils remained reactive and extraocular movements were preserved. Slitlamp examination revealed LASIK flaps bilaterally, clear corneas, and persistent shallow anterior chambers. Chemosis had improved. Intraocular pressures had decreased to 18 mmHg in both eyes. Gonioscopy demonstrated closed angles in all four quadrants bilaterally, with no visible angle structures.
4. Case 1 anterior segment optical coherence tomography (AS-OCT), showing cross-sectional views of the iridocorneal angle in multiple quadrants. Images reveal significant narrowing of the anterior chamber angles with variable degrees of iridocorneal apposition, consistent with topiramate-associated angle compromise. These findings align with the clinical diagnosis of paradoxical angle closure following drug discontinuation and demonstrate residual anatomical crowding despite normalization of intraocular pressure.
5. Macular optical coherence tomography (OCT) ganglion cell analysis for Case 2, revealing preserved ganglion cell-inner plexiform layer (GC-IPL) thickness bilaterally. No focal or diffuse thinning is observed, and measurements remain within expected ranges despite prior angle closure. These findings suggest that, while RNFL thinning was borderline, the central macular ganglion cell layers were structurally intact at the time of imaging.
Multimodal imaging was obtained to further characterize the anatomical changes and evaluate for possible glaucomatous damage. Optical coherence tomography (OCT) of the retinal nerve fiber layer demonstrated borderline thinning, most pronounced in the nasal quadrants of both eyes, which may have been artifact or related to axial elongation. Macular OCT revealed preserved ganglion cell-inner plexiform layer (GC-IPL) thickness bilaterally, with no evidence of
central ganglion cell loss or retinal disruption (Figure 5). Spectral-domain OCT of the macula confirmed intact foveal contour and normal lamination of retinal layers, with no signs of macular edema or subretinal fluid (Figure 6). These findings supported the absence of irreversible retinal or optic nerve damage at the time of imaging.
Ultra-widefield fundus photography captured during early follow-up demonstrated grade 3 peripheral lattice degeneration in both eyes and normalappearing optic nerves without edema, hemorrhage, or
Figure
Figure
Figure 6. Spectral-domain optical coherence tomography (SD-OCT) horizontal macular scans for Case 2, showing preserved foveal contour and normal lamination of retinal layers in both eyes. There is no evidence of macular edema, subretinal fluid, or disruption of the outer retinal layers, consistent with structural integrity following topiramate-associated angle closure.
Figure 7. Case 2 ultra-widefield fundus imaging (Optos) of both eyes obtained during follow-up. The left image (OD) and right image (OS) demonstrate clear visualization of the posterior pole with no optic disc edema or hemorrhage. Peripheral lattice degeneration (grade 3+) is visible in both eyes, most prominently inferotemporally. The nerves appear within normal limits.
pallor (Figure 7). These peripheral retinal findings were considered incidental and unrelated to the acute presentation.
Anterior segment OCT (AS-OCT) provided detailed anatomical confirmation of the angle closure. Imaging showed significant quadrant-to-quadrant variability in angle configuration. In the right eye, iridocorneal touch and anterior segment crowding were evident both temporally and nasally, consistent with ciliochoroidal effusion and forward displacement of the lens-iris diaphragm. In contrast, the left eye demonstrated more open angles in the corresponding quadrants (Figure 8). These findings confirmed an asymmetric but bilateral anatomical predisposition to non-pupillary block angle closure, aligning with known mechanisms of topiramate-induced ocular changes.
Over the next 48 hours, intraocular pressures normalized to 8 mmHg bilaterally. Manifest refraction revealed a pronounced myopic shift: –5.75 D OD and –5.50 D OS, consistent with anterior displacement of the lens-iris diaphragm. Visual acuity
stabilized and slit-lamp examination remained unchanged with shallow anterior chambers but no evidence of inflammation. No corneal edema, uveitis, or optic disc edema was observed throughout the course. Over the following week, the patient’s refractive error showed progressive improvement. One week after presentation, manifest refraction revealed a mild hyperopic shift, and by two weeks, the refraction had returned to near-emmetropic values (–0.25 D OD and plano OS), indicating resolution of the ciliochoroidal effusion. Intraocular pressures remained within normal limits.
Alternative diagnoses were considered, including primary angle-closure glaucoma, LASIK-related ectasia or flap complications, and uveitic glaucoma. However, the acute bilateral presentation, absence of corneal or anterior segment inflammation, recent exposure to topiramate, and dynamic anatomical findings on ASOCT collectively favored a diagnosis of secondary angle closure due to topiramate-induced ciliochoroidal effusion. The rapid normalization of IOP and resolution of refractive error following medication cessation and topical therapy further supported this conclusion.
This case illustrates a classic presentation of topiramate-associated acute angle closure, with bilateral IOP elevation, marked myopic shift, and anterior segment crowding. Multimodal imaging — including RNFL and macular OCT, widefield fundus photography, and anterior segment OCT — was essential for confirming the diagnosis, ruling out glaucomatous progression, and monitoring therapeutic response. The patient recovered fully with prompt medical intervention and discontinuation of the offending agent, avoiding long-term structural or visual sequelae.
Discussion
These 2 cases highlight the diagnostic complexity and clinical variability of topiramate-induced angle closure, a rare but increasingly recognized ocular complication of a widely prescribed systemic
the top
corresponding
Bottom 2 images show AS-OCT for nasal and temporal angles in the left eye similarly with corresponding localization images. The green scan lines indicate the orientation of each cut. Images reveal geographical-dependent variability in angle anatomy: the right eye images (top and second images) show iridocorneal touch and anterior segment crowding, consistent with angle closure; the temporal and nasal quadrants for the left eye (third and fourth images) demonstrate more open angles with visible separation between the iris and cornea. This asymmetry underscores the importance of multi-region imaging when assessing angle status following topiramate discontinuation.
medication. Topiramate is used for epilepsy, migraine prevention, weight loss, and mood stabilization, with expanding off-label use contributing to a rise in associated adverse events (1, 5). While acute angle closure associated with topiramate is typically described as an early complication of treatment, often occurring within 7 to 21 days of initiation, the cases presented here demonstrate that clinically significant anatomical compromise may also occur after abrupt discontinuation or when dosing is inconsistent (7, 8).
Topiramate-induced angle closure is mechanistically distinct from primary angle closure glaucoma. Rather than being caused by pupillary block, the condition arises from ciliochoroidal effusion and anterior displacement of the lens-iris diaphragm, leading to anterior chamber shallowing and secondary angle narrowing (6, 7). This non-pupillary block mechanism is thought to be related in part to topiramate’s weak inhibition of carbonic anhydrase isoenzymes (CA-II and CA-IV), which alters aqueous dynamics and increases uveal permeability (5). These changes are often bilateral and may not consistently produce elevated intraocular
pressure (IOP), making diagnosis challenging when patients present without classic signs of acute glaucoma.
In Case 1, the patient developed shallow anterior chambers, optic nerve cupping, and subjective visual changes 2 days after discontinuing topiramate. Notably, her IOP remained within normal limits throughout the course. These findings underscore that topiramateinduced angle closure may persist beyond active drug use due to the medication’s extended half-life and wide tissue distribution, particularly in individuals with altered renal function (1, 2, 4). The absence of pain or corneal edema further complicated diagnosis, demonstrating the need for high clinical suspicion and detailed anatomical assessment even in atypical presentations.
Case 2 presented more classically, with acute bilateral blurry vision, markedly elevated IOP, and a myopic shift greater than –5.50 diopters. The patient had initiated topiramate one week earlier and had reduced her dose the day before symptom onset. Slit-lamp examination showed chemosis and shallow anterior chambers, and gonioscopy revealed complete angle closure in both eyes. The rapid clinical improvement following discontinuation of the drug and initiation of topical aqueous suppressants and cycloplegics further supported the diagnosis. This case highlights the more well-recognized end of the spectrum, where the diagnosis is clearer and treatment outcomes are typically favorable if prompt therapy is initiated (1, 7).
These 2 cases therefore represent different but complementary patterns of topiramate-induced angle closure: one paradoxical and subacute, occurring after discontinuation with normal IOP, and the other acute and classic, presenting during active therapy with elevated IOP and dramatic refractive shift. Both scenarios support the current understanding that ciliochoroidal effusion, rather than pupillary block, is the underlying mechanism—resulting in anterior segment crowding and angle compromise without necessarily involving intraocular inflammation (5, 6).
Figure 8. Case 2 anterior segment optical coherence tomography (AS-OCT) for temporal and nasal angles in the right eye shown in
2 images above, each with
localization images.
Management of topiramate-induced angle closure differs from primary angle closure. Laser peripheral iridotomy is ineffective, as it does not address the nonpupillary block pathophysiology. Treatment instead focuses on cessation of the offending agent, reduction of aqueous production, and posterior displacement of the lens-iris diaphragm using cycloplegics (7). Topical corticosteroids may be beneficial in cases with inflammatory features, though their role remains less well defined. In both cases presented here, prompt recognition and medical management led to complete resolution of symptoms and anatomical recovery, avoiding the need for surgical intervention.
The existing literature supports a rising incidence of topiramate-related ocular adverse events, particularly as the drug is prescribed more broadly across neurology, psychiatry, and primary care (1, 5, 7). Many prescribers may be unaware of this rare but reversible complication, increasing the risk of delayed diagnosis. According to a systematic review by Abtahi et al., nearly all documented cases of topiramateinduced angle closure occurred within 2–3 weeks of therapy initiation, but late and post-discontinuation presentations may be underreported (7). Additionally, as shown in our first case, angle closure may occur even in the absence of elevated IOP or overt pain —factors that could easily lead to missed or delayed ophthalmology referral if clinicians rely solely on classic glaucoma signs.
Therefore, clinicians in all specialties should maintain a high index of suspicion in patients presenting with acute vision changes, halos, or myopic shift while on or recently off topiramate therapy. Comprehensive ocular evaluation — including slit-lamp examination, gonioscopy, and multimodal imaging when available — is essential to confirm the diagnosis and avoid irreversible visual loss. In both of our cases, early recognition and appropriate management resulted in complete visual and anatomical recovery, underscoring the importance of timely intervention.
Conclusion
Topiramate-induced angle closure represents a distinct and often underrecognized form of secondary angleclosure glaucoma, characterized by ciliochoroidal effusion, anterior displacement of the lens-iris diaphragm, and resultant anterior chamber shallowing without pupillary block (7). The cases presented here underscore the broad clinical spectrum of this
condition — ranging from acute presentations with elevated intraocular pressure (IOP) and pain to subacute or asymptomatic cases occurring even after medication discontinuation.
In Case 1, a patient developed bilateral angle narrowing and visual disturbances several days after abrupt topiramate cessation yet had normal IOP and minimal discomfort. This aligns with emerging literature showing that topiramate-induced ocular changes may persist beyond active use due to prolonged drug half-life and slow redistribution from ocular tissues (8, 13). Notably, Tran et al. described a similar paradoxical presentation of angle closure following drug withdrawal, emphasizing the importance of considering recent topiramate exposure even when therapy has been stopped (8). Moreover, Jalali et al. and Abtahi et al. both highlight that topiramate-induced angle closure is nonpupillary in nature and may not respond to traditional interventions such as laser iridotomy (7, 11).
In contrast, Case 2 exemplifies a more classic presentation: acute bilateral blurry vision, myopic shift, chemosis, and elevated IOP during active topiramate use. This presentation corresponds closely to the majority of cases described in systematic reviews and pharmacovigilance reports, including the original cases reported by Fraunfelder et al (9). However, even within such typical cases, variability exists. As described by Goyal et al., some patients present with rapidly reversible anatomical changes, while others may demonstrate lingering symptoms or asymmetric involvement despite bilateral exposure (12).
Both cases reinforce that the underlying mechanism involves ciliochoroidal effusion — distinct from pupillary block — leading to anterior displacement of the ciliary body and shallowing of the anterior chamber (7, 9). The topiramate package insert also acknowledges angle-closure glaucoma as a rare but serious adverse effect, typically occurring within the first month of use, though exact mechanisms remain incompletely understood (13). In both of our cases, optical coherence tomography (OCT), gonioscopy, and careful refractive assessment were instrumental in confirming the diagnosis and guiding treatment.
Management centers on immediate discontinuation of the drug, use of aqueous suppressants, and cycloplegics to posteriorly rotate the lens-iris diaphragm (7, 9).
Corticosteroids may be added to reduce ciliary body edema, although evidence remains limited to case
reports and small series (11, 12). Importantly, laser peripheral iridotomy is ineffective, and surgical intervention is rarely required if diagnosis is timely (7). In both patients, appropriate medical management resulted in full anatomical and functional recovery. Recent literature emphasizes that demographic risk factors for topiramate-induced angle closure differ from those of primary angle closure. Patients are often younger, female, and without hyperopia or crowded anterior segments (7, 11). Additionally, several reports, including Mansoor et al., describe atypical cases involving prolonged myopic shift and delayed resolution, highlighting the need for extended monitoring even after symptom improvement (10).
Despite its low incidence, the expanding use of topiramate across neurology, psychiatry, and primary care underscores the need for broader awareness of its ocular complications (7, 9). Clinicians should maintain a high index of suspicion when evaluating patients presenting with acute or subacute visual symptoms—including blurred vision, halos, or refractive changes—particularly if they have recently initiated or discontinued topiramate. Early ophthalmologic referral is essential, as prompt intervention can reverse symptoms and prevent permanent vision loss.
2. Shank RP, Gardocki JF, Vaught JL, et al. Topiramate: preclinical evaluation of a structurally novel anticonvulsant. Epilepsia. 1994;35(2):450–460.
3. Maryanoff BE, Nortey SO, Gardocki JF, Shank RP, Dodgson SP. Anticonvulsant O-alkyl sulfamates: 2,3:4,5-bis-O-(1-methylethylidene)-ß-Dfructopyranose sulfamate and related compounds. J Med Chem. 1987;30(5):880–887.
4. Vega D, Maalouf NM, Sakhaee K. Increased propensity for calcium phosphate kidney stones with topiramate use. Expert Opin Drug Saf 2007;6(5):547–557.
5. Choi J, Yoon D, Park M, Joung KI, Shin JY. Topiramate-related adverse events: pattern and signals in the Korea Adverse Event Reporting System, 2010–2017. Medicine (Baltimore). 2020;99(42):e22669.
6. Etminan M, Maberley D, Mikelberg FS. Use of topiramate and risk of glaucoma: a case-control study. Am J Ophthalmol. 2012;153(5):827–830.
7. Abtahi MA, Abtahi SH, Fazel F, et al. Topiramate and the vision: a systematic review. Clin Ophthalmol. 2012;6:117–131. doi:10.2147/OPTH. S27695.
8. Tran HV, Ravinet E, Schnyder C, Reichhart M, Guex-Crosier Y. Blood–brain barrier disruption associated with topiramate-induced angle-closure glaucoma of acute onset. Klin Monbl Augenheilkd. 2006;223(5):425–427.
9. Quist TS, Fortenbach CR, Miller MA, Bettis DI. Topiramate-induced angle closure. EyeRounds.org. Published 2019. Accessed May 21, 2025. https:// webeye.ophth.uiowa.edu/eyeforum/cases/283topiramate-induced-angle-closure.htm
10. Eggebrecht MS, Sriram S, Tamhankar MA. Topiramate-associated bilateral angle-closure glaucoma: a case series and review. Indian J Ophthalmol. 2024;72(4):852–855.
11. Gidal BE, Clark AM, Anders B, Gilliam F. The application of half-life in clinical decision making: comparison of the pharmacokinetics of extendedrelease topiramate (USL255) and immediaterelease topiramate. Epilepsy Res. 2017;129:26–32. doi:10.1016/j.eplepsyres.2016.10.020
13. Lobo AM, Sobrin L, Papaliodis GN, Buxton DF, Kaiser PK. Topiramate-induced acute-onset bilateral myopia and angle-closure glaucoma. Am J Ophthalmol. 2004;137(1):193–195.
2026 Summer Research Immersion Program
Each summer 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:
• GCSOM and Geisinger faculty research seminars
• GCSOM 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:
• Provide students with an immersive research experience under a mentor’s guidance
• Enhance students’ knowledge of the scope and types of research relevant to improving health in the region, nationally, and globally
• Provide research opportunities that span the translational continuum from laboratory based biomedical studies to clinical and public health research conducted with community partners
• Engage students in peer learning and networking
• Enhance students’ skills in oral and written scholarship
Program dates:
SRIP 2026 will be an 8-week program held June 1 – July 24, 2026.
Program deadlines:
Application release date: Dec. 12, 2025
Application submission deadline: Jan. 19, 2026
For more information, contact:
Sonia Lobo, PhD SRIP Program Administrator Associate Dean for Research & Scholarship slobo1@geisinger.edu
Elizabeth Kuchinski, MPH SRIP Director eckuchinski@geisinger.edu
Medical Research Honors Program
Current first-year medical students are eligible to join the Medical Research Honors Program (MRHP). With a mentor’s guidance, you will drive this long-term, thesis focused research experience. By completing the requirements while remaining in good academic standing, you’ll graduate with an honors distinction.
Through the MRHP, you will:
• Advance fundamental scientific knowledge
• Stand apart in competitive residency application fields
• Refine scholarly communication
• Gain a mindset of continual growth and learning
To complete this 4-year program, you must submit a research project proposal, write a thesis, and deliver an oral defense. You will also write abstracts, present posters, and publish findings while building towards your thesis defense. Your research experience is guided by a research mentor, a thesis advisory committee, and the program director. We encourage you to participate in the Summer Research Immersion Program as well.
Application deadline:
Early consideration deadline: May 4, 2026
Final deadline: Sept. 14, 2026
Application packet must include:
• MRHP application form
• Letter of support from research mentor
• CV
• Acknowledgment of mentor’s expectations
• Project proposal: project title, specific aims, hypothesis, background, preliminary data (if available)
Be a mentor
If you would be willing to have a medical student work with you on a long-term, thesis driven research project, email us at mrhp@geisinger.edu or scan the QR code.
Be sure to indicate your willingness to commit time, facilities, and resources to a student as needed throughout the program.
Questions about the MRHP program or mentoring?
Contact: Sonia Lobo, PhD Associate Dean for Research and Scholarship slobo1@geisinger.edu
Tracey Pratt, MPH MRHP Program Manager tpratt@geisinger.edu
Finding your way: Opportunities for student funding
You can find assistance in searching for funding opportunities specifically designed for students at the Office of Research and Scholarship. Funding opportunities may include support for fellowships, internships, research, programming, and collaboration.
The Office of Research and Scholarship can help you locate and qualify funding opportunities as well as assist in application review. Be sure to call or stop by early in the proposal development process so we can work with you to meet your deadline.
Geisinger Clinical Research Fund (CRF)
GSOM is pleased to support student research and professional development through the Clinical Research Fund (CRF).
The Office of Research & Scholarship administers GSOM’s CRF allocation by providing funding for research costs and subscriptions, conference travel reimbursement, and publication fees. As a GSOM student, you may apply for up to $5,000 per calendar year (January –December) from the CRF to support the use of Geisinger research cores (Investigator Initiated Research Operations, Biostatistics, Phenomic Analytics & Clinical Data Core). In addition, you may be able to obtain funds to cover the costs of a research subscription, such as statistical software. You may receive travel support up to $1,000 per calendar year to present your research at a scientific conference. Finally, if your work has been accepted for publication in a scholarly, peer-
reviewed journal, you may be eligible for support with publication fees. Publication support falls under the conference/publication support category, with a cap of $1,000 per student for combined conference/publication support per calendar year.
All student research support must be preapproved by Research & Scholarship staff. Funding is dependent on an annual budget allocation and is not guaranteed. If you're considering applying for research support, contact mrhp@geisinger.edu early in your planning and preparation
Contact information:
Tracey Pratt, MPH Manager of Research Education Resources
Office of Research and Scholarship
Phone: 570-558-3955
Internal extension: 5335
Email: tpratt@geisinger.edu
525 Pine St.
Scranton, PA 18509
570-504-7000
geisinger.edu/gcsom studentresearch@geisinger.edu
From left: Julia C. Kocherzat, Suzanne E. Kozloski, Minhtrinh Cao, Cade V. Neal, Aislin Silver, George G. James, and Damain D. Morris received Excellence in Research Awards for their outstanding abstract submissions at the 2025 Summer Research Symposium.
