Acta Orthopaedica Vol. 90 Issue 3, June 2019

3/19

* Sprowson AP et al. Bone Joint J 2016; 98-B: 1534–1541

www.heraeus-medical.com

Vol. 90, No. 3, 2019 (pp. 191–296)

Bone cement with gentamicin ttamicin i i and clindamycin

ACTA ORTHOPAEDICA

69 %

reduction of deep infections in hip hemiarthroplasty after * fractured neck of femur

Volume 90, Number 3, June 2019

IORT_90_02_COVER - Copy.indd 1

02-05-2019 16:15:56

Acta Orthopaedica is owned by the Nordic Orthopaedic Federation and is the official publication of the Nordic Orthopaedic Federation

E DITORIAL O F FICE

Acta Orthopaedica Department of Orthopedics Lund University Hospital SE–221 85 Lund, Sweden E-mail: acta.ort@med.lu.se Homepage: http://www.actaorthop.org

EDITOR

THE FOUNDATION BOARD OF

Anders Rydholm Lund, Sweden

THE NORDIC O RTHOPAEDIC F EDERATION AND A CTA O RTHOPAEDICA

DEPUTY EDITOR

Peter A Frandsen Odense, Denmark CO-EDITORS

Li Felländer-Tsai Stockholm, Sweden Nils Hailer Uppsala, Sweden Ivan Hvid Oslo, Norway Urban Rydholm Lund, Sweden Bart A Swierstra Wageningen, The Netherlands Eivind Witsø Trondheim, Norway Rolf Önnerfält Lund, Sweden

Peter Frandsen Denmark Ragnar Jonsson Iceland Heikki Kröger Finland Anders Rydholm Sweden Kees Verheyen the Netherlands

WEB EDITOR

Magnus Tägil Lund, Sweden S TATISTICAL EDITOR

Jonas Ranstam Lund, Sweden P RODUCTION MANAGER

Kaj Knutson Lund, Sweden

Vol. 90, No. 3, 2019

SUBSCRIPTION INFORMATION Acta Orthopaedica [print 1745-3674, online 1745-3682] is a peerreviewed journal, published six times a year plus supplements by Taylor & Francis on behalf of Nordic Orthopaedic Federation.

Annual Institutional Subscription, Volume 90, 2019 $1,218

£753

€976

The subscription fee purchases an online subscription. The price includes access to current content and back issues to January 1997 (if available). Printed copies of the journal are provided on request as a free supplementary service accompanying an online subscription. Supplements to the journal are also included in the subscription price.

For more information, visit the journal’s website: http://www.tandfonline.com/IORT Manuscripts should be uploaded at http://www.manuscriptmanager.com/ao/ for further handling at: Acta Orthopaedica Editorial Office, Department of Orthopaedics, Lund University Hospital, SE-221 85 Lund, Sweden Correspondance concerning copyright and permissions should be sent to: Håkan Pårup P.O. Box 3255, SE-103 65 Stockholm, Sweden, Tel: +46 (0)8 440 80 40. Fax: +46 (0)8 440 80 50.

E-mail: hakan.parup@informa.com.

Ordering information: Please contact your local Customer Service Department to take out a subscription to the Journal: USA, Canada: Taylor & Francis, Inc., 530 Walnut Street, Suite 850, Philadelphia, PA 19106, USA. Tel: +1 800 354 1420; Fax: +1 215 207 0050. UK/ Europe/Rest of World: T&F Customer Services, Informa UK Ltd, Sheepen Place, Colchester, Essex, CO3 3LP, United Kingdom. Tel: +44 (0) 20 7017 5544; Fax: +44 (0) 20 7017 5198; Email: subscriptions@tandf.co.uk Dollar rates apply to all subscribers outside of Europe. Euro rates apply to all subscribers in Europe except the UK and Republic of Ireland. If you are unsure which applies, contact Customer Services. All subscriptions are payable in advance and all rates include postage. Journals are sent by air to the USA, Canada, Mexico, India, Japan and Australasia. Subscriptions are entered on an annual basis, i.e., January to December. Payment may be made by sterling check, US dollar check, euro check, international money order, National Giro, or credit card (Amex, Visa and Mastercard).

Back issues: Taylor & Francis retains a two-year back issue stock of journals. Older volumes are held by our official stockists to whom all orders and enquiries should be addressed: Periodicals Service Company, 351 Fairview Ave., Suite 300, Hudson, New York 12534, USA. Tel: +1 518 537 4700; fax: +1 518 537 5899; e-mail: psc@ periodicals.com.

Airfreight and mailing in the USA by agent named Air Business Ltd, c/o Worldnet Shipping Inc., 156-15, 146th Avenue, 2nd Floor, Jamaica, NY 11434, USA. Periodicals postage paid at Jamaica NY 11431.

US Postmaster: Send address changes to Acta Orthopaedica, Air Business Ltd, c/o Worldnet Shipping Inc., 156-15, 146th Avenue, 2nd Floor, Jamaica, NY 11434, USA. Subscription records are maintained at Taylor & Francis Group, 4 Park Square, Milton Park, Abingdon, OX14 4RN, United Kingdom.

Copyright © 2019 The Author(s). Published by Taylor & Francis on behalf of the Nordic Orthopedic Federation. This is an Open Access article distributed under the terms of the Creative Commons Attribution-Non-Commercial License (https://creativecommons.org/licenses/by-nc/3.0 .

Informa UK Limited, trading as Taylor & Francis Group makes every effort to ensure the accuracy of all the information (the “Content”) contained in its publications. However, Informa UK Limited, trading as Taylor & Francis Group, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Informa UK Limited, trading as Taylor & Francis Group. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Informa UK Limited, trading as Taylor & Francis Group shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. Terms & Conditions of access and use can be found at http://www.tandfonline. com/page/terms-and-conditions Indexed/abstracted in: Allied and Complementary Medicine Library (Amed); ASCA (Automatic Subject Citation Alert); Biological Abstracts; Chemical Abstracts; Cumulative Index to Nursing and Allied Health Literature(CINAHL); Current Advances in Ecological and Environmental Sciences; Current Contents/Clinical Medicine; Current Contents/Life Sciences; Developmental Medicine and Child Neurology; Energy Research Abstracts; EMBASE/ Excerpta Medica; Faxon Finder; Focus On: Sports Science & Medicine; Health Planning and Administration; Index Medicus/MEDLINE; Index to Dental Literature; Index Veterinarius; INIS Atomindex; Medical Documentation Service; Nuclear Science Abstracts (Ceased); Periodicals Scanned and Abstracted. Life Sciences Collection; Research Alert; Science Citation Index; SciSearch; SportSearch; Uncover Veterinary Bulletin.

Printed in England by Henry Ling

Acta Orthopaedica

ISSN 1745-3674

Vol. 90, No. 3, June 2019 Shoulder Appropriate care for orthopedic patients: effect of implementation of the Clinical Practice Guideline for Diagnosis and Treatment of Subacromial Pain Syndrome in the Netherlands Low revision rate despite poor functional outcome after stemmed hemiarthroplasty for acute proximal humeral fractures: 2,750 cases reported to the Danish Shoulder Arthroplasty Registry Hip and knee Similar clinical results and early subsidence between the Collum Femoris Preserving and the Corail stem: a randomized radiostereometric study of 77 hips with 2 years’ follow-up Return to work after primary total hip arthroplasty: a nationwide cohort study Not all cemented hips are the same: a register-based (NJR) comparison of taper-slip and composite beam femoral stems Body mass index is associated with risk of reoperation and revision after primary total hip arthroplasty: a study of the Swedish Hip Arthroplasty Register including 83,146 patients Association between patient survival following reoperation after total hip replacement and the reason for reoperation: an analysis of 9,926 patients in the Swedish Hip Arthroplasty Register Resurfacing hip arthroplasty better preserves a normal gait pattern at increasing walking speeds compared to total hip arthroplasty Equivalent hip stem fixation by Hi-Fatigue G and Palacos R+G bone cement: a randomized radiostereometric controlled trial of 52 patients with 2 years’ follow-up Repeated cobalt and chromium ion measurements in patients with large-diameter head metal-on-metal ReCap-M2A-Magnum total hip replacement Implant survival and patient-reported outcome following total hip arthroplasty in patients 30 years or younger: a matched cohort study of 1,008 patients in the Swedish Hip Arthroplasty Register Evaluation of Forgotten Joint Score in total hip arthroplasty with Oxford Hip Score as reference standard Uncemented cups with and without screw holes in primary THA: a Swedish Hip Arthroplasty Register study with 22,725 hips Increased early mortality and morbidity after total hip arthroplasty in patients with socioeconomic disadvantage: a report from the Swedish Hip Arthroplasty Register Cementing does not increase the immediate postoperative risk of death after total hip arthroplasty or hemiarthroplasty: a hospitalbased study of 10,677 patients Total hip arthroplasty, combined with a reinforcement ring and posterior column plating for acetabular fractures in elderly patients: good outcome in 34 patients Complications and readmissions following outpatient total hip and knee arthroplasty: a prospective 2-center study with matched controls Children The development of spasticity with age in 4,162 children with cerebral palsy: a register-based prospective cohort study Arthroscopic versus open, medial approach, surgical reduction for developmental dysplasia of the hip in patients under 18 months of age Information to authors (see http://www.actaorthop.org/)

191

E J D Veen, M Stevens, C T Koorevaar, and R L Diercks

196

A Amundsen, J V Rasmussen, B S Olsen, and S Brorson

202

L J Klein, Gpuretic, M Mohaddes, and J Kärrholm

209

R Laasik, P Lankinen, M Kivimäki, V Aalto, M Saltychev, K Mäkelä, and J Vahtera H A Kazi, S L Whitehouse, J R Howell, and A J Timperley

214 220

A S Sayed-Noor, S Mukka, M Mohaddes, J Kärrholm, and O Rolfson

226

P Cnudde, E Bülow, S Nemes, Y Tyson, M Mohaddes, and O Rolfson

231

D M J M Gerhardt, T G Ter Mors, G Hannink, and J L C Van Susante P B Jørgensen, M Lamm, K Søballe, and M Stilling

237 243

H Mäntymäki, P Lankinen, T Vahlberg, A Reito, A Eskelinen, and K Mäkelä

249

M Mohaddes, E Nauclér, J Kärrholm, H Malchau, D Odin, and O Rolfson

253

A Larsson, O Rolfson, and J Kärrholm

258

V Otten, S Mukka, K Nilsson, S Crnalic, and J Kärrholm

264

R J Weiss, J Kärrholm, O Rolfson, and N P Hailer

270

E Ekman, I Laaksonen, K Isotalo, A Liukas, T Vahlberg, and K Mäkelä

275

T Lont, J Nieminen, A Reito, T-K Pakarinen, I Pajamäki, A Eskelinen, and M K Laitinen

281

K Gromov, C C Jørgensen, P B Petersen, P KjærsgaardAndersen, P Revald, A Troelsen, H Kehlet, and H Husted

286

O Lindén, G Hägglund, E Rodby-Bousquet, and P Wagner

292

S Duman, Y Camurcu, H Sofu, H Ucpunar, D Akbulut, and T Yildirim

Acta Orthopaedica 2019; 90 (3): 191–195

191

Appropriate care for orthopedic patients: effect of implementation of the Clinical Practice Guideline for Diagnosis and Treatment of Subacromial Pain Syndrome in the Netherlands Egbert J D VEEN 1, Martin STEVENS 1, Cornelis T KOOREVAAR 2, and Ron L DIERCKS 1 1 Department

of Orthopedic Surgery, University of Groningen, University Medical Center Groningen; 2 Department of Orthopedic Surgery and Traumatology, Deventer Hospital, Deventer, the Netherlands Correspondence: ejdveen@gmail.com Submitted 2018-08-23. Accepted 2019-01-31.

Background and purpose — The multidisciplinary Clinical Practice Guideline for diagnosis and treatment of subacromial pain syndrome (SAPS) was created in 2012 by the Dutch Orthopedic Association. In brief, it stated that SAPS should preferably be treated nonoperatively. We evaluated the effect of the implementation of the guideline on the number of shoulder surgeries for SAPS in the Netherlands (17 million inhabitants).

Patients and methods — An observational study was conducted with the use of aggregated data from the national database of the Dutch Health Authority from 2012 to 2016.

Information was collected on patients referred to and seen at orthopedic departments. Data from the following Diagnoses Related Groupings were analyzed: 1450 (tendinitis supraspinatus) and 1460 (rotator cuff tear).

Results — In 2016 fewer patients were diagnosed with tendinitis supraspinatus than in 2012—a decrease from 49,491 to 44,662 (10%). Of the patients diagnosed with tendinitis, 14% were treated surgically in 2012; this number dropped to 9% by 2016. More patients with a rotator cuff tear were diagnosed in 2016 than in 2012, an increase from 17,793 to 23,389 (32%), fewer were treated surgically: 30% in 2012, compared with 25% in 2016.

Interpretation — After introducing the multidisciplinary Clinical Practice Guideline “Diagnosis and treatment of subacromial pain syndrome,” a decrease in shoulder surgeries for related diagnoses was observed in the Netherlands. The introduction and dissemination of this guideline seems to have contributed to the implementation of more appropriate health care and prevention of unnecessary surgeries.

Shoulder pain is a frequent complaint in the general population, with an incidence of 0.8–2.3% and a lifetime prevalence of up to 67% (Urwin et al. 1998, Luime et al. 2004). It is mainly seen in women over age 45 (Greving et al. 2012). The most frequent complaint is pain at the shoulder with overhead activities, and pain at night.

Neer (1983)developed the concept of “impingement syndrome,” also called rotator cuff disease, bursitis, and supraspinatus tendinitis. None of these names cover the complex origin of subacromial pain with a painful arc, which nowadays is called “subacromial pain syndrome” (SAPS) (Papadonikolakis et al. 2011, Diercks et al. 2014b). Most of the symptoms usually resolve within a few months. Some patients show persistent symptoms despite physiotherapy and are referred to orthopedic surgeons to discuss open or arthroscopic bursectomy, acromioplasty, and/or rotator cuff repair.

In recent years increasing scientific evidence shows that patients’ results from surgical interventions are not better than treatment with physiotherapy and/or steroid injections (Dorrestijn et al. 2009, Björnsson Hallgren et al. 2017, Ketola et al.

2017). A randomized controlled trial (RCT) showed no benefit of acromioplasty compared with sham surgery or nonoperative treatment (Beard et al. 2018).

A clinical practice guideline for diagnosis and treatment of subacromial pain syndrome based on the available scientific evidence was created by the Dutch Orthopedic Society in 2012. The major recommendations were: SAPS should preferably be treated nonoperatively; patients who do not respond to exhaustive nonoperative treatment can be offered surgery; asymptomatic rotator cuff tears should not be treated surgically; when surgical repair of symptomatic rotator cuff tears is considered, the size of the tear, the condition of the muscles, and age and activity level of the

© 2019 The Author(s). Published by Taylor & Francis on behalf of the Nordic Orthopedic Federation. This is an Open Access article distributed under the terms of the Creative Commons Attribution-Non-Commercial License (https://creativecommons.org/licenses/by/4.0) DOI 10.1080/17453674.2019.1593641

192

patient are important factors to consider in the context of patient expectations; surgical treatment of tendinosis calcarea is not recommended.

To disseminate the guideline, presentations were given to the Dutch Orthopedic Society, the Rehabilitation Society and the Dutch Shoulder and Elbow Society. The guidelines were published in the Dutch Orthopedic Journal, the Dutch General Medical Journal (Diercks et al. 2014a), and Acta Orthopedica 2014 (Diercks et al. 2014b). Multiple presentations were held for national and regional symposia, physical therapists, and GPs. We have now examined whether the referral and treatment patterns have changed following the presentation of the clinical practice guideline for SAPS.

Patients and methods This observational study was conducted with use of aggregated data from the national database of the Dutch Health Authority from 2012 until 2016. All patients seen by a medical specialist in the Netherlands have specific codes registered for every diagnosis and treatment. The following diagnosisrelated groupings (DRG) (in Dutch DBC, Diagnose Behandel Combinatie) are applicable to the SAPS guideline: • 1450: tendinitis supraspinatus/ biceps, i.e., impingement; • 1460: rotator cuff/ biceps tendon tear.

We excluded the code 1480 “AC/SC disorders” and 1487 “other enthesiopathy of shoulder/elbow.”

The following surgical codes are related: • 38100: acromioplasty; • 38177 surgery on shoulder bursa.

To examine whether the treatment regimens for SAPS have changed since presentation of this new guideline we extracted data from the Dutch Health Authority (NZA) (Zorgautoriteit 2016). After choosing a DRG all declared subsequent surgical procedures can be found and calculated. The NZA is an autonomous administrative authority under the Dutch Ministry of Health, Welfare and Sports. The NZA has a database with nationwide data of all patients diagnosed at any Dutch hospital, and all interventions such patients underwent within the chosen diagnostic criteria (Zorgautoriteit 2018). As stated before, registry in this database is mandatory using a fixed list with diagnoses to choose from. Every Dutch orthopedic surgeon is obliged to use these codes for billing of the costs at the insurance companies.

This list has not been changed during the study period. Only 1 DRG can be chosen for the shoulder complaint at the first visit. This database starts from 2012 and contains only anonymous and aggregated data. We looked within the groups of the above-mentioned DRGs from January 1, 2012 to December 31, 2016 and registered the number of patients who had subsequent surgery. The numbers in the database were complete for 2012, 2013, and 2014. As a result of the ongoing billing process at the time of this study the numbers for the year 2015

Acta Orthopaedica 2019; 90 (3): 191–195

were 90% completed and the numbers for 2016 were 75% completed. The numbers of these years are extrapolated to 100% in order to make a valid comparison.

The Dutch healthcare insurance system requires referral by a GP before a patient can visit a medical specialist such as an orthopedic surgeon. To calculate the incidence of DRGs, and thus the trend in referrals by mostly GPs, information was gathered from the Dutch Central Bureau of Statistics (Statline 2018).

Additionally, an online survey (see Supplementary data) was performed with a small cohort of GPs and orthopedic surgeons. GPs were randomly selected from a database of the university and the orthopedic surgeons were selected from the Dutch orthopedic association database. All of the invited GPs (n = 33) and orthopedic surgeons (n = 23) filled in the form. They were asked about their experiences with shoulder complaints, the guideline, and if they changed their treatment strategies as a result of the guideline.

Statistics Descriptive statistics were used for the annual incidence rates of referred patients for a specific DRG per 100,000 inhabitants. To get an impression of the effect of the dissemination of the guideline, total numbers of DRGs in the Netherlands were calculated and compared with the baseline year (2012) and with each subsequent year. 95% confidence intervals (CI) of the difference between these 2 proportions were calculated (Fleiss et al. 2013). SAS version 9.4 (SAS Institute, Cary, NC, USA) was used.

Ethics, funding, and potential conflicts of interest This study was reviewed and approved by the medical ethical committee of University Medical Center Groningen (register: 201501203-2018/259). There was no special funding for this study and there is no potential conflict of interest to be declared by any of the authors.

Results Incidence From 2012 to 2016 a total of 237,960 patients were diagnosed by orthopedic surgeons with DRG 1450 “tendinitis supraspinatus/biceps, i.e., impingement” and 97,900 patients with DRG 1460 “rotator cuff/biceps tendon tear.” In 2016, fewer patients were diagnosed with a tendinitis supraspinatus/ biceps, i.e., impingement (DRG 1450) compared with 2012, a decrease of 10%. More patients with rotator cuff or biceps tendon tear (DRG 1460) were diagnosed in 2016 than in 2012, an increase of 32% (Table 1). The referral pattern to orthopedic departments changed between 2012 and 2016. For DRG 1450 the incidence decreased from 2.96 to 2.63 per 100,000 inhabitants; for DRG 1460 the incidence increased from 1.06 to 1.38 per 100,000 inhabitants.

Acta Orthopaedica 2019; 90 (3): 191–195

193

Table 1. Numbers and incidences (per 100,000 inhabitants in the Netherlands) of DRG divided by nonoperative and operative treatment

Surgery (%) 40 35

Tendinitis supraspinatus/biceps, Rotator cuff or biceps tendon tear

i.e., impingement (DRG 1450) (DRG 1460) Non- Non Referred Incidence operative Operative Referred Incidence operative Operative

Tendinitis/impingement Rotator cuff tear

30 25 20

2012 49,591 2.96 42,443 7,148 2013 48,945 2.92 42,076 6,869 2014 47,401 2.82 41,375 6,026 2015 a 47,361 2.80 42,551 4,810 2016 a 44,662 2.63 40,585 4,077 a

17,793 1.06 12,597 5,196 20,130 1.20 14,191 5,939 21,156 1.26 15,015 6,141 23,643 1.40 17,337 6,306 23,389 1.38 17,593 5,796

Numbers extrapolated to 100%.

Table 3. Differences in patients who had surgery for DRG 1460 (rotator cuff or biceps tendon tear) for each year compared with 2012 and subsequent years

Baseline versus Difference (Confidence interval)

Baseline versus Difference (Confidence interval)

2012 2012 2012 2012 2013 2014 2015

2012 2012 2012 2012 2013 2014 2015

a

0.00 (–0.01 to 0.00) –0.02 (–0.02 to –0.01) a –0.04 (–0.05 to –0.04) a –0.05 (–0.06 to –0.05) a –0.01 (–0.02 to –0.01) a –0.03 (–0.03 to –0.02) a –0.01 (–0.01 to –0.01) a

Significant difference.

10 5 0

Table 2. Differences in patients who had surgery for DRG 1450 (tendinitis supraspinatus/biceps, i.e., impingement) for each year compared with 2012 and subsequent years

2013 2014 2015 2016 2014 2015 2016

15

a

2013 2014 2015 2016 2014 2015 2016

0.00 (–0.01 to 0.01) 0.00 (–0.11 to 0.01) –0.03 (–0.03 to –0.02) a –0.04 (–0.05 to –0.04) a 0.00 (–0.01 to 0.00) –0.02 (–0.03 to –0.02) a –0.02 (–0.03 to –0.01) a

2012

2013

2014

2015

2016

Figure 1. Percentages of surgeries of total referred patients for each DRG.

Table 4. Surgical procedures

Shoulder Year Acromioplasty bursectomy 2012 9,025 2013 8,625 2014 8,263 2015 6,803 2016 5,310

1,974 2,377 2,545 2,488 2,335

Significant difference.

Surgery for DRG 1450 tendinitis supraspinatus/ biceps, i.e., impingement Of the patients diagnosed with DRG, 1,450 14% underwent surgery in 2012. This decreased to 9% in 2016 (Table 1). This is a statistically significant drop when comparing 2012 with 2014, 2015, and 2016, but also when comparing the subsequent years with each other (Table 2).

Surgery for DRG 1460 rotator cuff or biceps tendon tear Of the 1,460 patients diagnosed with a DRG, 30% underwent surgery in 2012. The percentage of referred patients who had surgery decreased to 25% in 2016 (Table 1). This is a statistically significant drop when comparing 2012 with 2015 and 2016, but also when comparing 2014, 2015, and 2016 with each other (Table 3).

Surgical codes When looking at surgical codes a statistically significant decrease in acromioplasties (41%) and an increase in bursectomies (18%) is seen over the years (Table 4).

An overview of the percentages of the total referred patients for each DRG treated surgically is depicted in Figure 1.

To gain an impression of the experiences with the guideline an online survey (see Supplementary data) was performed on

shoulder surgeons and GPs. 23 shoulder surgeons and 33 GPs were reached. All but 2 shoulder surgeons were familiar with the guideline and 19 considered it helpful with treating their patients with SAPS; 19 surgeons stated that less than 10% of the SAPS patients were treated surgically after the guideline was published.

2 out of 33 GPs were familiar with the SAPS guideline but 11 of the GPs stated that they had changed their treatment the past years; more patients are treated nonoperatively and not referred to an orthopedic specialist.

Discussion This study was conducted to investigate the effect of the introduction of the Clinical Practice Guideline for Subacromial Pain Syndrome. The results show that after publication of the guideline, the number of patients referred with the diagnosis of “impingement” or “tendinitis” decreased by 10%. Surgery decreased, by 37% and 17% respectively, in the SAPS and rotator cuff tear-related group. The proportion of surgical treatment for “tendinitis” and “rotator cuff tear” decreased from 18% in 2012 to 15% in 2016. Also, a decrease in acromioplasties was observed. Despite most GPs not being familiar with the SAPS guideline they had changed their practice with fewer referrals.

194

Our results are in line with international shifts in the treatment of SAPS and rotator cuff tears, most likely as a result of emerging evidence. In Finland a drop in acromioplasties is seen from 1998 to 2011 (Paloneva et al. 2015). In Australia the number of patients having a rotator cuff repair increased over the years 2001–2013 (Thorpe et al. 2016). The same trend is seen in the United States, with a decrease in acromioplasties and a rise in rotator cuff surgeries (Mauro et al. 2012). The international rise in patients undergoing surgical repair for a rotator cuff tear may be the result of improved surgical options during the latest 15 years.

The Clinical Practice Guideline for Diagnosis and Treatment of Subacromial Pain Syndrome was completed in 2012.

Since 2012 more scientific evidence supports the recommendations of the Clinical Practice Guideline. The most recent RCT with surgical, sham surgical, and nonoperative treatments for SAPS showed no benefit of surgery (Beard et al.

2018). The RCT of Farfaras et al. (2016) showed no difference between open acromioplasty, arthroscopic acromioplasty, and physiotherapy in the treatment of SAPS after 2–3 years, but somewhat better results for the surgical groups compared with the physiotherapy group after long-term follow-up (Farfaras et al. 2018). Ketola et al. (2017) saw no benefit of surgical treatment for SAPS after 10 years follow-up of an RCT. A review of 2014 saw a benefit of physiotherapy compared with controls (Gebremariam et al. 2014).

Several recent RCTs showed no benefit of surgery for asymptomatic degenerative rotator cuff tears compared with nonoperative treatment (Moosmayer et al. 2014, Lambers Heerspink et al. 2015). No difference was seen either when these treatments were combined with an acromioplasty (Kukkonen et al. 2015).

Several institutes have recognized elements that increase the impact of clinical guidelines, like the standards of trustworthiness developed by the IOM (Institute of Medicine) of the American National Academies, and derivative products like physician–patient guides that help provide more practical information. The Dutch guideline fulfills these conditions.

Nevertheless, several clinical guidelines like those produced

by NICE (National Institute for Health and Care Excellence) in the UK and other national bodies appear to play a limited part in orthopedic decision-making (Grove et al. 2016).

Although formal codified knowledge in the form of clinical guidelines still appears to play a modest part in orthopedic surgery clinical practice decision-making, the coincidence of new high-level scientific evidence provided by well-designed and performed RCTs and the development of clinical guidelines will have an impact on orthopedic clinical decision-making (Khan et al. 2013). We observed this effect in our study period after implementation of our clinical guideline and the publication of several RCTs that confirmed the conclusions of our guideline.

The results of the survey show that the treatment strategies of the orthopedic surgeons are roughly in line with the guide-

Acta Orthopaedica 2019; 90 (3): 191–195

lines; fewer patients were treated surgically. However, only 2 out of 33 GPs were familiar with the SAPS guideline. All GPs used the National General Practitioner Guideline (NHG) “shoulder complaints” from 2008. We found a decline in referrals from GPs, but still it is unclear whether this can be attributed to the guideline.

Clinical orthopedic practice is difficult to change, as shown in our study: SAPS is still treated surgically in 10% of cases.

This is also seen in the treatment of degenerative meniscal tears. Several studies and clinical guidelines indicate that arthroscopic debridement is of no benefit (Sihvonen et al.

2013, Thorlund et al. 2015) but arthroscopies on patients with degenerative meniscal tears are still performed in the Netherlands (Rongen et al. 2018).

One of the flaws is that the data of this study were derived from the database of the NZA, which started in 2012 with no information preceding that. The effects we found may be the result of a trend based on earlier reports. Another limitation is the extrapolation of the database numbers for 2015 and 2016 to compare them with the preceding years. Although this may influence the total number of patients with that diagnosis, the relative number of surgical vs. nonoperative treatments is not influenced because this is only recorded within the diagnostic group. The surgical codes are used for a sole procedure or as part of other surgery, such as arthroscopic lateral clavicular resection, therefore the numbers are not always limited to DRG 1450 and 1460 but may also be registered from other DRGs. On the basis of registration inaccuracies, a distal clavicle resection could have been performed additional to an acromioplasty, or vice versa. As the aim of the study was to identify the number of procedures before and after the publication of the guideline, and the registration system did not change, this will not have had an effect on the study results.

In summary, the introduction and dissemination of this guideline seem to have contributed to implementation of more appropriate healthcare and prevention of unnecessary surgeries. Although GPs refer fewer patients for SAPS, their education can still be improved.

Supplementary data The online survey is available as supplementary data in the online version of this article, http://dx.doi.org/10.1080/ 17453674.2019.1593641 EV, MS, CK, and RD all participated in the conception and design of the study. EV was responsible for acquisition of the data. EV and MS did the statistical analysis. All authors critically revised the manuscript for important intellectual content and approved the final version of the manuscript.

The authors would like to thank thank Roy Stewart and Ruth Rose for their contribution to the study and the analysis.

Acta thanks Lars Evert Adolfsson and Jeppe Vejlgaard Rasmussen for help with peer review of this study.

Acta Orthopaedica 2019; 90 (3): 191–195

Beard D J, Rees J L, Cook J A, Rombach I, Cooper C, Merritt N, Shirkey B A, Donovan J L, Gwilym S, Savulescu J. Arthroscopic subacromial decompression for subacromial shoulder pain(CSAW): a multicentre, pragmatic, parallel group, placebo-controlled, three-group, randomised surgical trial.

Lancet 2018; 391(10118): 329-38.

Björnsson Hallgren, H C, Adolfsson, L E, Johansson K, Öberg B, Peterson A, Holmgren T M. Specific exercises for subacromial pain: good results maintained for 5 years. Acta Orthop 2017, 88(6), 600-5.

Diercks R L. Practice guideline “diagnosis and treatment of the subacromial pain syndrome.” Ned Tijdschr Geneeskd 2014a; 158: A6985.

Diercks R, Bron C, Dorrestijn O, Meskers C, Naber R, de Ruiter T, Willems J, Winters J, van der Woude, Henk Jan. Guideline for diagnosis and treatment of subacromial pain syndrome: a multidisciplinary review by the Dutch Orthopaedic Association. Acta Orthop 2014b; 85(3): 314-22.

Dorrestijn O, Stevens M, Winters J C, van der Meer K, Diercks R L. Conservative or surgical treatment for subacromial impingement syndrome? A systematic review. J Shoulder Elbow Surg 2009; 18(4): 652-60.

Farfaras S, Sernert N, Hallström E, Kartus J. Comparison of open acromioplasty, arthroscopic acromioplasty and physiotherapy in patients with subacromial impingement syndrome: a prospective randomised study. KSSTA 2016; 24(7): 2181-91.

Farfaras, S, Sernert N, Rostgard-Christensen L, Hallstrom E, Kartus J T. Subacromial decompression yields a better clinical outcome than therapy alone: a prospective randomized study of patients with a minimum 10-year follow-up. J Sports Med 2018; 46(6): 1397-407.

Fleiss J L, Levin B, Paik M C. Statistical methods for rates and proportions.

Chichester: Wiley; 2013.

Gebremariam L, Hay E M, van der Sande R, Rinkel W D, Koes B W, Huisstede B M. Subacromial impingement syndrome: effectiveness of physiotherapy and manual therapy. Br J Sports Med 2014; 48(16): 1202-8.

Greving K, Dorrestijn O, Winters J C, Groenhof F, Van der Meer K, Stevens M, Diercks R L. Incidence, prevalence, and consultation rates of shoulder complaints in general practice. Scand J Rheumatol 2012; 41(2): 150-5.

Grove A, Johnson R, Clarke A, Currie G. Evidence and the drivers of variation in orthopaedic surgical work: a mixed method systematic review. Health Syst Policy Res 2016; 3: 1.

Ketola S, Lehtinen J T, Arnala I. Arthroscopic decompression not recommended in the treatment of rotator cuff tendinopathy: a final review of a randomised controlled trial at a minimum follow-up of ten years. Bone Joint J 2017; 99(6): 799-805.

Khan H, Hussain N, Bhandari M. The influence of large clinical trials in orthopedic trauma: do they change practice? J Orthop Trauma 2013; 27(12): e274.

Kukkonen J, Joukainen A, Lehtinen J, Mattila K T, Tuominen E K, Kauko T, Aarimaa V. Treatment of nontraumatic rotator cuff tears: a randomized controlled trial with two years of clinical and imaging follow-up. J Bone Joint Surg Am 2015; 97(21): 1729-37.

195

Lambers Heerspink F O, van Raay J J, Koorevaar R C, van Eerden P J, Westerbeek R E, van ’t Riet E, van den Akker-Scheek I, Diercks R L. Comparing surgical repair with conservative treatment for degenerative rotator cuff tears: a randomized controlled trial. J Shoulder Elbow Surg 2015; 24(8): 1274-81.

Luime J J, Koes B W, Hendriksen I, Burdorf A, Verhagen A P, Miedema H S, Verhaar J. Prevalence and incidence of shoulder pain in the general population: a systematic review. Scand J Rheumatol 2004; 33(2): 73-81.

Mauro C S, Jordan S S, Irrgang J J, Harner C D. Practice patterns for subacromial decompression and rotator cuff repair: an analysis of the American Board of Orthopaedic Surgery database. J Bone Joint Surg Am 2012; 94(16): 1492-9.

Moosmayer S, Lund G, Seljom U S, Haldorsen B, Svege I C, Hennig T, Pripp A H, Smith H. Tendon repair compared with physiotherapy in the treatment of rotator cuff tears: a randomized controlled study in 103 cases with a five-year follow-up. J Bone Joint Surg Am 2014; 96(18): 1504-14.

Neer C S. Impingement lesions. Clin Orthop Rel Res 1983; (173): 70-7.

Paloneva J, Lepola V, Karppinen J, Ylinen J, Aarimaa V, Mattila V M. Declining incidence of acromioplasty in Finland. Acta Orthop 2015; 86(2): 220-4.

Papadonikolakis A, McKenna M, Warme W, Martin B I, Matsen III F A. Published evidence relevant to the diagnosis of impingement syndrome of the shoulder. J Bone Joint Surg Am 2011; 93(19): 1827-32.

Rongen J J, van Tienen T G, Buma P, Hannink G. Meniscus surgery is still widely performed in the treatment of degenerative meniscus tears in the Netherlands. Knee Surg Sports Traumatol Arthrosc 2018; 26(4): 1123-9.

Sihvonen R, Paavola M, Malmivaara A, Itälä A, Joukainen A, Nurmi H, Kalske J, Järvinen T L. Arthroscopic partial meniscectomy versus sham surgery for a degenerative meniscal tear. N Engl J Med 2013; 369(26): 2515-24.

Statline C. Centraal bureau voor de statistiek. Bevolking kerncijfers Identifier: 37296ned, http://opendata.cbs.nl/statline Thorlund J B, Juhl C B, Roos E M, Lohmander L S. Arthroscopic surgery for degenerative knee: systematic review and meta-analysis of benefits and harms. BMJ 2015; 350: h2747.

Thorpe A, Hurworth M, O’Sullivan P, Mitchell T, Smith A. Rising trends in surgery for rotator cuff disease in Western Australia. Aust NZ J Surg 2016; 86(10): 801-4.

Urwin M, Symmons D, Allison T, Brammah T, Busby H, Roxby M, Simmons A, Williams G. Estimating the burden of musculoskeletal disorders in the community: the comparative prevalence of symptoms at different anatomical sites, and the relation to social deprivation. Ann Rheum Dis 1998; 57(11): 649-55.

Zorgautoriteit. Open Data Van De Nederlandse Zorgautoriteit 2016; http:// www.opendisdata.nl Zorgautoriteit. 2018; https://www.nza.nl/english

196

Acta Orthopaedica 2019; 90 (3): 196–201

Low revision rate despite poor functional outcome after stemmed hemiarthroplasty for acute proximal humeral fractures: 2,750 cases reported to the Danish Shoulder Arthroplasty Registry Alexander AMUNDSEN 1, Jeppe V RASMUSSEN 1, Bo S OLSEN 1, and Stig BRORSON 2 1 Department

of Orthopaedic Surgery, Herlev-Gentofte University Hospital, Herlev; 2 Department of Orthopaedic Surgery, Zealand University Hospital, Køge, Denmark Correspondence: alexander.amundsen.03@regionh.dk Submitted 2018-09-26. Accepted 2019-01-11.

Background and purpose — The revision rate of stemmed hemiarthroplasty (SHA) for acute proximal humeral fractures is low, but does not necessarily reflect the functional outcome. We report the revision rate of SHA for acute proximal humeral fractures and the proportion of arthroplasties that are not revised despite low functional outcome scores.

Patients and methods — The Danish Shoulder Arthroplasty Registry was used to identify all patients with a proximal humeral fracture that was treated with a SHA between January 1, 2006 and December 31, 2015. Information on demographics, surgical procedures, and revisions was collected by the registry. The Western Ontario Osteoarthritis of the Shoulder (WOOS) index at 1 year was used as functional outcome score. We converted the score to a percentage of a maximum score with 100 being the best.

Results — 2,750 SHAs in 2,719 patients were included.

Mean age was 72 years (SD 11); 79% were women. Mean WOOS at 1 year was 55 (SD 26). A total of 101 (4%) arthroplasties were revised, and the 10-year cumulative implant survival rate was 95%. The Cox regression model showed a statistically significant impact on implant survival of age, but not of sex or arthroplasty brand. A WOOS score below 30 and 50 was reported in 11% and 25% of patients, respectively.

Interpretation — We found a high implant survival rate, but also a high proportion of patients with a low functional outcome score 1 year after surgery.

Proximal humeral fractures account for approximately 10% of all fall-related fractures in adults (Court-Brown et al. 2017) and the incidence increases with age, with females over 80 years having an incidence rate of 379 per 100,000 person years (Launonen et al. 2015a). A rise in the incidence of proximal humeral fractures is expected with an increasing elderly population.

Most proximal humeral fractures can be managed non-surgically (Handoll and Brorson 2015, Launonen et al. 2015b).

However, in the case of head split fractures or fracture dislocations, an arthroplasty may be considered because of a high risk of avascular necrosis and post-traumatic osteoarthritis. The stemmed hemiarthroplasty (SHA) has traditionally been preferred. It may result in satisfactory long-term pain relief, but results for postoperative shoulder movement have been less predictable (Antuna et al. 2008). The reason for this may be related to impaired rotator cuff function and non-union of the tuberosities (Kralinger et al. 2004, Greiner et al. 2009, Boileau et al. 2013, Giovale et al. 2014, Hashiguchi et al. 2015).

Revision rates after SHA are low, ranging from 1% to 9% (Fevang et al. 2009, Namdari et al. 2013, Brorson et al. 2017).

However, revision rates do not necessarily reflect the functional outcome as some arthroplasties are never revised due to patient- or surgery-related factors.

We report the revision rate of SHA for acute proximal humeral fractures and the proportion of arthroplasties that are not revised despite low functional outcome scores.

Patients and methods The Danish Shoulder Arthroplasty Registry (DSR) was used to identify all patients with proximal humeral fractures treated with SHA between January 1, 2006 and December 31, 2015 © 2019 The Author(s). Published by Taylor & Francis on behalf of the Nordic Orthopedic Federation. This is an Open Access article distributed under the terms of the Creative Commons Attribution-Non-Commercial License (https://creativecommons.org/licenses/by/4.0) DOI 10.1080/17453674.2019.1597491

Acta Orthopaedica 2019; 90 (3): 196–201

Number of SHAs 400

197

Table 1. Hierarchy of reasons for revision where more than one reason was reported Hierarchy of reasons for revision

300

I Infection – an infection that requires revision of the arthroplasty II Periprostethic fracture – fracture that requires revision of the arthroplasty III Dislocation and instability IV Loosening – loosening of any arthroplasty component V Rotator cuff problem VI Others – glenoid wear, biomechanical problems including overstuffing, and pain with no other complication

200

100

0

2006 2007 2008 2009 2010 2011 2012 2013 2014 2015

Year Figure 1. Annual number of SHAs for acute proximal humeral fractures in Denmark from 2006 to 2015.

(Figure 1). Only patients who were surgically managed within 2 weeks from the date of injury were included.

The DSR is a national registry that was established in 2004.

All surgeons performing shoulder arthroplasty, at hospitals or private clinics in Denmark, have been obliged to report to the DSR since 2006. Information concerning demographics and the surgical procedure is reported to the registry by the surgeon at the time of the operation (Rasmussen et al. 2012).

The 1-year functional outcome is assessed with the Western Ontario Osteoarthritis of the Shoulder (WOOS) index. The WOOS is managed by the DSR, which sends patients a questionnaire 10–14 months after surgery. A single reminder was sent to non-responders. DSR uses the WOOS questionnaire for all patients with shoulder arthroplasties regardless of indication for surgery. Currently, the DSR does not include information on fracture classification or other functional outcome scores in the database.

WOOS is a patient-administrated quality of life questionnaire consisting of 19 questions divided into 4 categories (physical symptoms, sports and work, lifestyle, and emotions). Each question is designed as a visual analogue scale, ranging from 0 to 100 points, with 100 being the worst. In this study the total WOOS scores are converted to percentages, with 100 being the best. A questionnaire was marked as incomplete if 1 or more questions were unanswered. The Danish translated version of WOOS has previously been validated for osteoarthritis patients (Rasmussen et al. 2013), but is yet to be validated for fracture patients.

Patients who died or whose surgery was revised within 1 year of surgery were not sent a WOOS questionnaire. If revision occurred later than 1 year after surgery the WOOS score of the primary arthroplasty procedure was included in the analysis.

A revision was defined as removal or exchange of the hemiarthroplasty component or the addition of a glenoid compo-

nent. The revision procedure is linked to the primary procedure using a unique civil registration number given at birth.

Information regarding the revision procedure including the indication for revision is reported to the registry by the surgeon at the time of the revision procedure. It is possible for the surgeon to report more than 1 indication for revision. In these cases, we used a hierarchy to classify the indication for revisions (Table 1). This was based on a previously reported hierarchy of reasons for revisions made by the Nordic Arthroplasty Register Association (NARA), which has been adapted by the DSR (Rasmussen et al. 2016).

The vital status of patients was obtained through the Danish National Register of Persons and linked to the data from the DSR using the civil registration number.

Statistics Hazard ratios were calculated using Cox’s proportional hazards regression model with 95% confidence intervals (CI).

Age, sex, year of surgery, and arthroplasty brand were included in the model. Log–log plots and Schoenfeld residuals were used to check that the proportional hazards assumption was fulfilled. The 10-year cumulative implant survival rate was illustrated using the Kaplan–Meier method with revision as the endpoint. A log-rank test was used to compare the implant survival rates of different age groups. The 1-way Anova test was used to compare the mean WOOS score of different age groups. A chi-square test was used to compare response rates between men and women.

The presumption of independence is violated when we include bilateral procedures in the survival analyses and, even though this may theoretically have consequences, no practical problems have been shown when analyzing arthroplasty register data (Ranstam et al. 2011). Additionally, the underlying assumption of no competing risks, which forms the basis of the Cox regression model and Kaplan–Meier method, is violated in survival analyses of arthroplasty data as patients are censored if they die.

SPSS version 22.0 (IBM Corp, Armonk, NY, USA) was used to perform the analyses. P-values were 2-tailed and a p-value < 0.05 was set as the level of statistical significance.

198

Acta Orthopaedica 2019; 90 (3): 196–201

Table 2. Demographics, proportion of revisions, and mean WOOS score overall and for each age group Total Sex, n (%) Male 570 (21) Female 2,180 (79) Prosthesis brand, n (%) Depuy Global FX 952 (34) Zimmer Bigliani-Flatow 840 (31) Tornier Aequalis 268 (10) Biomet Comprehensive 245 (9) Others 445 (16) Revision, n (%) 101 (4) WOOS Complete, n (%) 1,525 (60) Score, mean (SD) 55 (26)

< 55 years 55–74 years ≥ 75 years 104 (59) 73 (41)

310 (24) 1,002 (76)

156 (12) 1,105 (88)

48 (27) 67 (38) 22 (12) 15 (9) 25 (14) 9 (5)

495 (38) 372 (28) 141 (11) 98 (7) 206 (16) 70 (5)

409 (32) 401 (32) 105 (8) 132 (11) 214 (17) 22 (2)

87 (52) 53 (24)

780 (62) 54 (26)

658 (58) 57 (25)

Table 3. Risk of revision. Cox regression model a

Age

Univariate Adjusted relative relative risk (95% CI) risk (95% CI)

≥ 75 55–75 < 55

1.0 (reference) 2.9 (1.8–4.7) 2.7 (1.3–5.9)

1.0 (reference) 2.9 (1.8–4.7) 3.3 (1.5–7.5)

a

Sex, prosthesis brand (Depuy Global FX, Zimmer Bigliani-Flatow, Tornier Aequalis Fx, Biomet Comprehensive Fx) and year of surgery (2006–2007, 2008–2009, 2010–2011, 2012–2013, 2014–2015) were included in the multiple model and showed no statistical significance.

Frequency 70

Table 4. Reasons for revision. Values are frequency (percent) Total Dislocation Loosening Infection Fracture Rotator cuff failure Others a Missing Total a Including

< 55 years 55–75 years ≥ 75 years

60

50

21 (0.8) 2 11 8 2 (0.1) 0 2 0 15 (0.5) 1 11 3 1 (0.0) 0 1 0 30 (1.1) 4 19 7 21 (0.8) 2 17 2 11 (0.4) 0 9 2

40

101 (3.7) 9 70 22

10

pain of unknown cause.

30

20

0 0

0

20 30 40 50 60 70 80 90 100

WOOS score

Ethics, funding, and potential conflicts of interest Permission to handle and store data was obtained from the Danish Data Protection Agency (date 03.07.2014, j.nr. 200758-0015). According to the regulations in Denmark, this study did not need permission from the National Committee on Health Research Ethics. There was no conflicts of interest to be declared related to this study.

Results 2,750 SHAs in 2,719 patients were included; 79% were women and the mean age was 72 years (SD 11) (Table 2). The use of SHA was stable in the period from 2007 to 2012, with peaks in 2008, 2010, and 2013. A decrease in SHA was found from 2013 to 2015 (Figure 1).

WOOS 157 (6%) patients died and 37 (1%) patients underwent revision within 1 year of surgery, leaving 2,556 (93%) arthroplasties available for follow-up. WOOS was completed in 1,525 SHAs (60%) with a mean WOOS of 55 (SD 26). Incomplete and missing WOOS questionnaires accounted for 197 (8%)

Figure 2. Distribution of WOOS scores. Red lines signify a WOOS score of 30 and 50.

and 834 (33%) respectively. WOOS was completed by 54% male and 61% female patients (p < 0.01). A WOOS score below 30 and 50 was reported in 303 (11%) and 676 (25%) patients, respectively (Figure 2). There were no stastically significant differences in WOOS score between age groups (p = 0.08).

Revision and prosthesis survival 101 (4%) SHAs were revised. Patients who were younger than 55 and patients between 55 and 74 years had a 2.7 (CI 1.3–5.9) and 2.9 (CI 1.8–4.7) times higher risk of revision compared with patients who were older than 75 years (Table 3). The most common indications for revision were rotator cuff problem (1%) and dislocation (0.8%) (Table 4).

The 10-year cumulative implant survival rate was 95% (CI 94–96). For patients who were younger than 55 years, patients between 55 and 75 years, and patients who were older than 75 years the survival rates were 94% (CI 89–97), 93% (CI 91–95), and 98% (CI 96–99), respectively (Figure 3).

Acta Orthopaedica 2019; 90 (3): 196â&#x20AC;&#x201C;201

199

Cumulative implant survival rate (%) 100

97.5

95.0

92.5

90.0

< 55 years

87.5 0

1

2

3

55â&#x20AC;&#x201C;75 years 4

5

6

> 75 years 7

8

9

10

Years after index operation Figure 3. The 10-year cumulative implant survival rate and CI for patients younger than 55 years (blue), patients between 55 and 74 years (green), and patients older than 75 years (red) (p < 0.01).

Discussion We found a high implant survival rate, but also a high proportion of patients with a low functional outcome score 1 year after surgery. This adds to the debate concerning the use of SHA and other prostheses in patients with displaced proximal humeral fractures.

Olerud et al. (2011) conducted a randomized controlled trial (RCT), investigating quality of life, function, and pain in 55 patients with displaced 4-part fractures treated non-surgically or surgically with SHA. The patients, with a mean age of 77 years, were followed up 2 years postoperatively. The study found that the health-related quality of life score (EQ-5D 0.81 vs. 0.65), the Disabilities of the Arm, Shoulder and Hand (DASH) score (30 vs. 37), and pain assessment (VAS 15 vs. 25) were in favor of SHA compared with non-surgically treated patients. However, the only statistically significant difference was found in EQ-5D and the score differences in DASH and VAS are too small to be considered clinically significant. Range of motion and Constant score were similar between the 2 groups.

In a RCT comparing non-surgical and SHA-treated Neer 4-part fractures in 50 patients older than 65 years, Boons et al.

(2012) found that there was no clear benefit for SHA compared with non-surgical treatment. Constant score and Simple Shoulder Test were similar 3 or 12 months after surgery. Both groups had improved strength 12 months after surgery and the nonsurgically treated group had better abduction strength at the 3and 12-month follow-up compared with SHA-treated patients.

However, the non-surgically treated patients experienced more pain than SHA patients at the 3-month follow-up, but this difference was not detectable at 12 months postoperatively. Furthermore a Cochrane review found that patients surgically

managed for displaced proximal humeral fractures involving the humeral neck did not have a better outcome compared with non-surgically managed patients 1 and 2 years after surgery (Handoll and Brorson 2015). The evidence of this review did, however, not cover fracture dislocations or head split fractures, as there are no RCTs concerning these indications. In addition, the review found that surgically managed patients were likely to have greater need of subsequent surgery.

The proportion of revisions in our study corresponds with the 3% reported by both the NARA group (Brorson et al.

2017) and in a study of 422 Norwegian patients undergoing SHA for acute proximal humeral fractures (Fevang et al.

2009). In contrast, 9% revisions were found in a systematic review including 7 studies comprising 263 SHAs for proximal humeral fractures. However only 1 study was prospective and only 2 studies had a final follow-up of more than 30 patients (Namdari et al 2013). In our study the most common reason for revision was failure of the rotator cuff. Degeneration or rupture of the rotator cuff or nonunion of the tuberosities are associated with proximal migration, which can be painful and can restrict movement of the shoulder.

The implant survival rate in our study reflects previously reported rates. The NARA group found 1-, 5-, and 10-year implant survival rates of 0.99, 0.96, and 0.95 respectively for SHA used for proximal humeral fractures (Brorson et al. 2017). Another study from the United States (Farng et al.

2011) reported a 10-year cumulative survival rate of 94% for 5,044 patients with proximal humeral fractures. However, 6% of patients were treated with total shoulder arthroplasty. An improvement of implant survival to a final 5-year cumulative survival rate of 95% was found in a study of 751 acute fractures, of which 86% were treated with SHA (Fevang et al.

2015).

The low revision rate is in contrast to the high number of patients reporting a low WOOS score 1 year after surgery.

The reason for this is unknown, but surgeons might hesitate to revise because of age, comorbidity, or low functional demands or because the revision procedure can be challenging. Thus, for patients with proximal humeral fractures the revision rate alone does not necessarily reflect the effect of the shoulder arthroplasty. The reporting of satisfaction after SHA for proximal humeral fractures varies in the literature and many studies do not account for the assessment used. The patient satisfaction self-assessment scale was used in a study of 51 patients, with a mean age of 71 years and of whom 39 were female, who were treated with SHA for 3- or 4-part proximal humeral fractures (Valenti et al. 2017). This study found that 13 of 51 patients reported poor satisfaction after 18 months. The age and distribution of female patients in this study is similar to our study. In a study by Boileau et al. (2002), 66 patients were asked if they were very satisfied, satisfied, disappointed, or unhappy with the functional outcome 27 months after SHA for 3- and 4-part proximal humeral fractures. 29 patients were either disappointed or unhappy.

200

We found differences in WOOS score between age groups, but none were statistically significant. A study on the social implications of SHA for proximal humeral fractures in patients older than 70 years found that 85% of patients lived in their own environment and managed daily life despite poor shoulder function (Dietrich et al. 2007). Furthermore, the difference in functional outcome between younger and elderly patients might be due to differences in general health condition and in the mechanisms of trauma. Most proximal humeral fractures occur in elderly patients with osteoporosis suffering low-energy trauma, while the force causing the fractures among younger patients is higher. Younger patients might still be working or performing activities that require good shoulder function. Thus, the outcome may not meet their expectations.

The strength of this study is the high number of patients and the WOOS questionnaire, providing valuable information on the 1-year functional outcome. The use of data on a national level is associated with high external validity. This study has limitations. The completeness of the WOOS questionnaire was 60%, which leaves a large group of non-responders. A study on the reliability of patient-reported outcome in DSR showed that the non-responders do not appear to bias the overall result (Polk et al. 2013). However, that study was based on all indications for shoulder arthroplasty and the study did not manage to contact 18% of the patients. The DSR does not have information on general health condition, radiographs, or classification of the fractures in the database. We have information only on the patient-reported outcome at 1 year, but the functional outcome might improve after 1 year. The study does not provide information on functional outcome after revision arthroplasties. This is important, as some revisions lead to a good functional outcome and cannot be considered as persisting failures. Lastly, WOOS was invented for patients with osteoarthritis and has not been validated for patients treated with shoulder arthroplasty for proximal humeral fractures.

In summary, we found a high implant survival rate, but also a high proportion of patients with a low functional outcome score 1 year after surgery. The reason for this is unknown, but surgeons might hesitate to revise because of age, comorbidity, or low functional demands or because the revision procedure can be challenging. Young age was associated with an increased risk of revision compared with older patients, but the functional outcome at 1 year was similar.

AA obtained permissions, analyzed data, and prepared the manuscript of this study. AA and JVR did the statistical analyses. Reading and proofing of the manuscript was done by JVR, BSO, and SB. The authors would like to thank all surgeons in Denmark for reporting to DSR and thank the DSR for providing us with data.

Acta thanks Antti Loosi and Sari Ponzer for help with peer review of this study.

Acta Orthopaedica 2019; 90 (3): 196â&#x20AC;&#x201C;201

Antuna S A, Sperling J W, Cofield R H. Shoulder hemiarthroplasty for acute fractures of the proximal humerus: a minimum five-year followup. J Shoulder Elb Surg 2008; 17(2): 202-9. doi: 10.1016/j.jse.2007.

06.025.

Boileau P, Krishnan S G, Tinsi L, Walch G, Coste J S, Mole D. Tuberosity malposition and migration: reasons for poor outcomes after hemiarthroplasty for displaced fractures of the proximal humerus. J Shoulder Elb Surg 2002; 11(5): 401-12.

Boileau P, Winter M, Cikes A, Han Y, Carles M, Walch G, Schwartz D G. Can surgeons predict what makes a good hemiarthroplasty for fracture?

J Shoulder Elb Surg 2013; 22(11): 1495-506. doi: 10.1016/j.jse.2013.

04.018.

Boons H W, Goosen J H, van Grinsven S, van Susante J L, van Loon C J. Hemiarthroplasty for humeral four-part fractures for patients 65 years and older: a randomized controlled trial. Clin Orthop Relat R 2012; 470(12): 3483-91. doi: 10.1007/s11999-012-2531-0.

Brorson S, Salomonsson B, Jensen S L, Fenstad A M, Demir Y, Rasmussen J V. Revision after shoulder replacement for acute fracture of the proximal humerus. Acta Orthop 2017; 88(4): 446-50. doi: 10.1080/17453674.2017.1307032.

Court-Brown C M, Clement N D, Duckworth A D, Biant L C, McQueen M M. The changing epidemiology of fall-related fractures in adults. Injury 2017; 48(4): 819-24. doi: 10.1016/j.injury.2017.02.021.

Dietrich M, Meier C, Zeller D, Grueninger P, Berbig R, Platz A. Primary hemiarthroplasty for proximal humeral fractures in the elderly: long-term functional outcome and social implications. Eur J Trauma Emerg Surg 2007; 33(5): 512-19. doi: 10.1007/s00068-007-6134-5.

Farng E, Zingmond D, Krenek L, Soohoo N F. Factors predicting complication rates after primary shoulder arthroplasty. J Shoulder Elb Surg 2011; 20(4): 557-63. doi: 10.1016/j.jse.2010.11.005.

Fevang B T, Lie S A, Havelin L I, Skredderstuen A, Furnes O. Risk factors for revision after shoulder arthroplasty: 1,825 shoulder arthroplasties from the Norwegian Arthroplasty Register. Acta Orthop 2009; 80(1): 83-91.

Fevang B T, Nystad T W, Skredderstuen A, Furnes O N, Havelin L I. Improved survival for anatomic total shoulder prostheses. Acta Orthop 2015; 86(1): 63-70. doi: 10.3109/17453674.2014.984113.

Giovale M, Mangano T, Roda E, Repetto I, Cerruti P, Kuqi E, Franchin F. Shoulder hemiarthroplasty for complex humeral fractures: a 5 to 10-year follow-up retrospective study. Musculoskelet Surg 2014; 98(Suppl 1): 27-33. doi: 10.1007/s12306-014-0319-y.

Greiner S H, Diederichs G, Kroning I, Scheibel M, Perka C. Tuberosity position correlates with fatty infiltration of the rotator cuff after hemiarthroplasty for proximal humeral fractures. J Shoulder Elb Surg 2009; 18(3): 431-6. doi: 10.1016/j.jse.2008.10.007.

Handoll H H, Brorson S. Interventions for treating proximal humeral fractures in adults. Cochrane DB Syst Rev 2015(11): Cd000434. doi: 10.1002/14651858.CD000434.pub4.

Hashiguchi H, Iwashita S, Ohkubo A, Takai S. The outcome of hemiarthroplasty for proximal humeral fractures is dependent on the status of the rotator cuff. Int Orthop 2015; 39(6): 1115-19. doi: 10.1007/s00264-0152758-y.

Kralinger F, Schwaiger R, Wambacher M, Farrell E, Menth-Chiari W, Lajtai G, Hubner C, Resch H. Outcome after primary hemiarthroplasty for fracture of the head of the humerus: a retrospective multicentre study of 167 patients. J Bone Joint Surg Br 2004; 86(2): 217-9.

Launonen A P, Lepola V, Saranko A, Flinkkila T, Laitinen M, Mattila V M. Epidemiology of proximal humerus fractures. Arch Osteoporos 2015a; 10: 209. doi: 10.1007/s11657-015-0209-4.

Launonen A P, Lepola V, Flinkkila T, Laitinen M, Paavola M, Malmivaara A. Treatment of proximal humerus fractures in the elderly: a systemic review of 409 patients. Acta Orthop 2015b; 86(3): 280-5. doi: 10.3109/17453674.2014.999299.

Namdari S, Horneff J G, Baldwin K. Comparison of hemiarthroplasty and reverse arthroplasty for treatment of proximal humeral fractures: a systematic review. J Bone Joint Surg Am 2013; 95(18): 1701-8. doi: 10.2106/ jbjs.l.01115.

Acta Orthopaedica 2019; 90 (3): 196â&#x20AC;&#x201C;201

Olerud P, Ahrengart L, Ponzer S, Saving J, Tidermark J. Hemiarthroplasty versus nonoperative treatment of displaced 4-part proximal humeral fractures in elderly patients: a randomized controlled trial. J Shoulder Elb Surg 2011; 20(7): 1025-33. doi: 10.1016/j.jse.2011.04.016.

Polk A, Rasmussen J V, Brorson S, Olsen B S. Reliability of patient-reported functional outcome in a joint replacement registry: a comparison of primary responders and non-responders in the Danish Shoulder Arthroplasty Registry. Acta Orthop 2013; 84(1): 12-17. doi: 10.3109/17453674.2013.

765622.

Ranstam J, Karrholm J, Pulkkinen P, Makela K, Espehaug B, Pedersen A B, Mehnert F, Furnes O. Statistical analysis of arthroplasty data, I: Introduction and background. Acta Orthop 2011; 82(3): 253-7. doi: 10.3109/17453674.2011.588862.

Rasmussen J V, Jakobsen J, Brorson S, Olsen B S. The Danish Shoulder Arthroplasty Registry: clinical outcome and short-term survival of 2,137

201

primary shoulder replacements. Acta Orthop 2012; 83(2): 171-3. doi: 10.3109/17453674.2012.665327.

Rasmussen J V, Jakobsen J, Olsen B S, Brorson S. Translation and validation of the Western Ontario Osteoarthritis of the Shoulder (WOOS) index: the Danish version. Patien Relat Outcome Meas 2013; 4: 49-54. doi: 10.2147/ prom.s50976.

Rasmussen J V, Brorson S, Hallan G, Dale H, Aarimaa V, Mokka J, Jensen S L, Fenstad A M, Salomonsson B. Is it feasible to merge data from national shoulder registries? A new collaboration within the Nordic Arthroplasty Register Association. J Shoulder Elb Surg 2016; 25(12): e369-e77. doi: 10.1016/j.jse.2016.02.034.

Valenti P, Aliani D, Maroun C, Werthel J D, Elkolti K. Shoulder hemiarthroplasty for proximal humeral fractures: analysis of clinical and radiographic outcomes at midterm follow-up: a series of 51 patients. Eur J Orthop Surg Tr 2017; 27(3): 309-15. doi: 10.1007/s00590-017-1927-7.

202

Acta Orthopaedica 2019; 90 (3): 202–208

Similar clinical results and early subsidence between the Collum Femoris Preserving and the Corail stem: a randomized radiostereometric study of 77 hips with 2 years’ follow-up Liesbeth J KLEIN, Goran PURETIC, Maziar MOHADDES, and Johan KÄRRHOLM

Department of Orthopaedics, Institute of Clinical Sciences, Sahlgrenska Academy, University of Gothenburg, Sweden Correspondence: Liesbeth_klein@hotmail.com Submitted 2018-06-04. Accepted 2018-12-17.

Background and purpose — Femoral neck preserving hip replacement has been suggested to improve clinical results and facilitate late revision. We compared the 2-year outcome and radiostereometric pattern of femoral head migration between the Collum Femoris Preserving (CFP) stem and the Corail stem.

Patients and methods — 83 patients were randomized to either a CFP stem or a Corail stem. All patients received the same cup. At 2 years clinical outcomes were assessed using validated scoring systems and plain radiographs.

2-year migration was determined using radiostereometric analysis.

Results — At 2 years the clinical outcomes (Oxford Hip Score, Harris Hip Score, SF-36, EQ5D-VAS, satisfaction VAS, and pain VAS) were similar between the 2 groups. The radiographic measurements showed that the femoral neck was resected around 1 cm more proximally with use of CFP stems (p < 0.001). The proximal–distal and medial–lateral migration of the femoral head center was similar. The Corail stem showed increased posterior displacement after 1 year, but no difference was found between the absolute translations in the anterior–posterior direction (p = 0.2). 2 CFP stems were revised due to loosening within the first 2 years.

None of the Corail stems was revised.

Interpretation — In the 2-year perspective clinical outcomes suggested no obvious advantages with use of the CFP stem. The magnitude of the early stem migration was similar, but the pattern of migration differed. The early revisions in the CFP are a cause of concern.

The stems used most frequently in primary hip replacement surgery have a length of about 13–15 cm. Removal of such a stem, if uncemented and ingrown, might become difficult should any late infection or instability problems occur.

The concept of femoral neck preserving hip replacement with the use of a short stem was introduced for young and active patients, who, partly due to their longer life expectancy, may require multiple revisions, which would be facilitated due to the higher femoral neck osteotomy. The more proximal physiological load distribution can be expected to decrease proximal bone resorption, which could facilitate fixation of a revision stem. Preservation of the femoral neck will also imply that the native anteversion of the femur is easier to preserve during the operation, which could result in improved function and overall clinical results.

The Collum Femoris Preserving (CFP) stem was introduced

by Pipino and Calderale in the 1980s and has been evaluated in multiple studies. So far the clinical documentation of the CFP stem indicates a stable fixation and good 2–9-year results (Röhrl et al. 2006, Briem et al. 2011, Nowak et al. 2011, Kress et al. 2012, Lazarinis et al. 2013, Hutt et al. 2014, Li et al.

2014, You et al. 2015). To our knowledge there are no published randomized studies comparing the CFP with a welldocumented standard stem.

We speculated that a more conservative resection of the femoral neck could lead to better clinical outcomes. Therefore, we compared the neck-preserving CFP stem with the conventional Corail stem in a randomized controlled trial. Our primary aim was to compare the clinical outcomes at 2 years, using Oxford Hip Score between the 2 groups. Second, we used radiostereometric analysis (RSA) to compare fixation between the 2 stem designs.

© 2019 The Author(s). Published by Taylor & Francis on behalf of the Nordic Orthopedic Federation. This is an Open Access article distributed under the terms of the Creative Commons Attribution-Non-Commercial License (https://creativecommons.org/licenses/by/4.0) DOI 10.1080/17453674.2019.1577344

Acta Orthopaedica 2019; 90 (3): 202–208

ENROLLMENT

203

Assessed for eligibility n = 458 Excluded Declined to participate n = 375 Randomized n = 83

ALLOCATION

Allocated to CFP (n = 41) Received allocated intervention (n = 41)

Allocated to Corail (n = 42) Received allocated intervention (n = 42)

FOLLOW-UP

Lost to follow-up (n = 2): – due to revision surgery, 2

Lost to follow-up (n = 3): – unknown reason, 2 – dead, 1

performed or supervised by a surgeon with long experience of uncemented THR. All patients were operated using a direct lateral approach with the patient in the lateral decubitus position.

Full weight-bearing was encouraged directly postoperatively.

2 patients were lost to follow-up. 1 patient died before the 2nd year follow-up due to brain metastases. 2 patients in the CFP group underwent stem revision due to loosening between the 1 and 2 years’ follow up. The cup shell was left in place in both cases. All other patients were followed for at least 2 years.

Clinical outcome measures The Oxford Hip Score and Harris Hip Score were conducted preoperatively and after 12 and ANALYSIS 24 months of follow-up. The University of Los Analyzed with RSA (n = 38) Analyzed with RSA (n = 39) Angeles California activity scale (UCLA) was Excluded from analysis (n = 1): Excluded from analysis (n = 0) assessed preoperatively, after 3 months, and after – technical RSA problems, 1 12 and 24 months. Quality of life was determined Figure 1. Consort flow diagram by the SF-36 (containing a physical health component summery score [SF-36 p] and a mental health component summery score [SF-36 m]), Patients and methods EQ-5D-VAS, a visual analogue scale (VAS) for Study design and population pain, and postoperatively satisfaction with the outcome of surWe conducted a randomized controlled trial at the Sahlgren- gery. These scores were determined preoperatively, and after ska University Hospital, Mölndal. We included patients with 3, 12, and 24 months. The EQ-5D was scored according to a painful hip and radiological evidence of osteoarthritis who the UK tariffs. The UCLA questionnaire was scored using the were eligible for hip arthroplasty. Other inclusion criteria were English scoring tool. hip anatomy suitable for both designs according to preoperative planning and age between 35 and 75 years. Exclusion criteria Radiography were previous treatment with cortisone and low expected activ- Postoperatively and after 12 and 24 months, standard pelvic, ity rate due to other diseases such as generalized joint disease. anteroposterior (AP), and lateral radiographs were obtained.

458 patients visiting our outpatient clinic between May 2012 We determined the length of the remaining femoral neck, the and May 2014 fulfilled the primary inclusion criteria. Of these, neck resorption ratio, and the position of the tip of the stem 83 patients (83 hips) with radiographic appearance of the prox- in the femoral canal. The remaining neck was measured from imal femur judged suitable for uncemented fixation accepted to the middle of the lesser trochanter to the proximal calcar. participate. Patients were randomly assigned to either a CFP or Neck resorption and the position of the tip were expressed Corail stem with the use of envelopes. 41 hips received a CFP in ratios (Figure 2). Radiolucent lines around the stem were and 42 a Corail stem (Figure 1). All hips were operated with a determined according to Gruen. Radiographs were examined Delta-TT cup (Lima, Italy). The CFP stem (LINK, Germany) by 2 of the authors (LK and GP), with an agreement of 0.74 is available in 6 sizes with 2 different curvatures, 2 CCD angles (Cohen’s kappa).

(117 or 126 degrees) and with or without a calcium phosphate coating. In our study only coated stems were used. The Corail Radiostereometric analysis (RSA) stem (DePuy Synthes, USA) is a conventional, uncemented, During surgery, 7–9 0.8 mm tantalum markers were placed in hydroxyapatite-coated collarless straight stem. It is available the proximal femoral bone. Translations of the femoral head in 11 sizes. Since 2009 it has been the most frequently used represented migration of the stem. Uniplanar radiographs were exposed at a median of 2 days (0–5) after surgery, using uncemented stem in Sweden for primary THR. The mean age of the 83 patients (53 men) at operation was 2 detectors with an angle of about 40° between the X-ray tubes 58 years (35–73). 76 patients had primary osteoarthritis, 5 and with use of cage 77 (RSA Biomedical, Umeå, Sweden). secondary osteoarthritis due to dysplasia, 1 idiopathic femo- Follow-up investigations were performed 3, 6, 12, and 24 ral head necrosis, and 1 femoral head necrosis after trauma. months after surgery. To determine the precision of the RSA 14 different surgeons performed the operations. They were all measurements we conducted double examinations postopera-

204

Acta Orthopaedica 2019; 90 (3): 202–208

Table 1. Mean error of rigid body fitting and condition number

Mean error of rigid body fitting (mm) Condition number

Figure 2. Method of measuring the remaining neck (a) and the position of tip of the stem. We measured the distance between the tip of the stem and the inner cortex and calculated the ratio between these distances. Ratio between lateral and medial distance is b/c.The neck resorption ratio (NRR) is calculated by dividing the distance between the medial tip of the collar and the medial apex of the remaining neck (d) by the length of a straight line traced from the medial tip of the collar to the apex of the lesser trochanter (e).

tively of 76 hips and calculated the 99% prediction interval of the precision based on the presumption of zero motion between repeated exposures. The medial–lateral, proximal– distal, and anterior–posterior translation of the femoral head center could be measured with a precision of 0.18, 0.18, and 0.45 mm respectively. This tolerance interval corresponded to the 99% confidence interval of the error around a supposed mean 0 value of the error (no systematic bias between double examinations). The analysis of movement of the stem was performed using the UMRSA analysis software 6.0 (RSA Biomedical, Umeå, Sweden). The median values and ranges of the mean error of rigid body fitting and condition number are presented in Table 1. The center of the femoral head was used to measure translations of the stem; hence stem rotation could not be analyzed. RSA analysis of stem migration up to 2 years was performed on 39 CFP stems and 38 Corail stems due to insufficient number of stable bone markers in 1 patient.

Statistics Our primary outcome was the Oxford Hip Score. The secondary outcome was distal stem migration measured with RSA. A power analysis, assuming normally distributed data, performed before the study started, indicated that 30 patients in each group would give us the possibility to detect a difference of 4 points on the OHS between the groups with a power of 80%. A corresponding analysis of distal migration of the femoral head center indicated that we could detect a group difference of 0.4 mm based on a presumed standard deviation of 0.5 mm in each group. All outcomes were analyzed using IBM

CFP Corail median range median range 0.2 30

0.04–0.4 17–151

0.2 33

0.04–0.3 18–93

SPSS Statistics 23 (IBM Corp, Armonk, NY, USA). Most of the clinical parameters recorded did not follow a normal distribution. Therefore we used the Mann–Whitney test to compare the clinical outcomes between the Corail and CFP group.

Comparisons were done on preoperative data and results at 3 months and 2 years. Comparison of RSA data was performed at 2 years with use of a Mann–Whitney test. In addition, the results of repeated ANOVA test on 37 CFP and 35 Corail stems with complete RSA data on all 4 follow-up occasions are presented. P-values < 0.05 were regarded to represent a statistically significant difference.

Ethics, registration, funding, and potential conflicts of interests The study followed the Helsinki declaration (Ethical approval 243-12, Regional ethical committee Gothenburg, Sweden).

Informed consent was obtained from all patients. The study was registered in ClinicalTrials.gov (NCT02983526) and followed the CONSORT Statement. Institutional support was received from LINK, Germany, LIMA, Italy, Ingabritt and Arne Lundbergs Research Foundation, and LUA/ALF, Sweden. No conflicts of interest were declared.

Results Clinical outcomes The characteristics of the groups were nearly similar at baseline (Table 2). No statistically or clinically important differences were found between the 2 groups after 3 months, except from estimation of general health, where patients were asked to value their general health at the moment of questioning in comparison with the last 12 months of their life. At 3 months, 35 of the patients with a Corail stem valued their health to be better compared with 26 in the CFP group (p = 0.04). This result in favor of the Corail stem had disappeared at the 2-year follow-up. At this time the clinical outcomes had improved compared with the preoperative measurements, being similar between the 2 groups (Table 3).

Radiographic outcomes The postoperative radiographs showed a mean preservation of the proximal femur of 37 mm (SD 5.4) in patients with a CFP stem, compared with 28 mm (SD 5.4) in the Corail group (p < 0.001). At the 2-year follow-up 39 CFP and 39 Corail stems

Acta Orthopaedica 2019; 90 (3): 202–208

205

Table 2. Patient characteristics and preoperative clinical variables

CFP (n = 41) n median (range)

Age Harris Hip Score a Oxford Hip Score SF-36 p SF-36 m EQ-5D EQ-VAS Pain VAS UCLA score General health b better the same worse missing answers

41 61 (35–73) 23 53 (22–74) 39 21 (8–45) 37 26 (14–49) 37 57 (24–71) 38 0.5 (0–0.8) 36 60 (10–95) 38 70 (20–93) 38 4 (2–10) 38 2 14 22 3

Table 3. Clinical outcomes after 2 years

Corail (n = 42) n median (range)

42 58 (43–73) 22 52 (32–83) 41 20 (2–35) 38 26 (13–52) 38 50 (2–71) 40 0.2 (–0.6 to 0.8) 40 60 (5–95) 40 64 (5–85) 41 4 (2–10) 40 0 12 28 2

Harris Hip score 35 100 (44–100) Oxford Hip score 38 44 (19–48) SF-36 p 38 48 (13–57) SF-36 m 38 55 (17–66) EQ-5D 38 0.8 (–0.2 to 1) EQ-VAS 38 85 (20–100) Pain VAS 38 7 (0–73) Satisfaction VAS 38 95 (9–100) UCLA-activity score 38 6 (2–10 General health a 38 better 18 the same 15 worse 5 missing answers 1

a High

number of missing Harris Hip Score was caused by logistic problems (poor communication to study secretaries and failures to scan these forms). b Number of patients that valued their general health as better / the same / worse than the last 12 months and missing answers.

a General

CFP (n = 39) n median (range)

Corail (n = 39) pn median (range) value 38 100 (48–100) 38 45 (15–48) 38 43 (18–63) 38 54 (18–63) 38 0.8 (–0.4 to 1) 38 85 (20–100) 39 2 (0–81) 39 97 (0–100) 39 6 (2–10) 39 15 14 10 0

0.7 0.9 0.9 0.7 0.2 0.8 0.4 0.7 0.6 0.2

health, see Table 2

Table 4. Median and mean translation (mm) of the center of the femoral head at 2 years

Translations

median

CFP (n = 39) range mean (95%CI)

Medial (+) / lateral (–) 0.1 –0.6 to 2 0.2 (0.1 to 0.3) Proximal (+) / distal (–) –0.2 –1 to 0.2 –0.3 (–0.4 to –0.2) Anterior (+) / posterior (–) –0.1 –2 to 1 –0.1 (–0.2 to 0.2)

were radiographically analyzed. The mean neck resorption ratio at 2 years was 0.00 in the Corail group and 0.04 (SD 0.1) in the CFP group with 7 patients who showed neck resorption (p = 0.003) compared with none in the Corail group. The median lateral–medial ratio of the position of the tip of the stem after 2 years was 1 (0.4–2.0) in the CFP group and 0.7 (0.4–1.5) in the Corail group (p < 0.001), meaning that the tip of the Corail stems was placed more laterally. The median anterior–-posterior ratio was 1.1 (0.6–3.0) after 2 years in the CFP group and 1.3 (0.5–3.3) in the Corail group, meaning both of the stems were placed more posteriorly, without any statistically significant difference between them (p = 0.4). 9 Corail stems showed radiolucent lines on plain radiographs at 2-year follow-up, most of them less than 15% of the total stem circumference facing bone in Gruen regions 1, 7, 8, 9, 10, and 14. In 2 stems the radiolucent lines had an extension between 15% and 30%. 3 CFP stems showed radiolucency around the stem. In 1 stem it occupied 5% of the stem–bone interface, localized to Gruen region 1. In 2 hips the lines had an extension of 20% and 25% corresponding to regions 8, 9, 10, and 12. Median radiolucency in the CFP and Corail groups was 0% for both stem designs and on both the AP (p = 0.05) and lateral (p = 0.5) views.

median

Corail (n = 38) range mean (95%CI)

0.1 –0.2 to 4 0.3 (0.1 to 0.5) –0.1 –6 to 0.3 –0.5 (–0.9 to –0.1) –0.2 –12 to 0.9 –0.7 (–1 to 0)

p-value 0.7 0.7 0.02

RSA results After 2 years the median proximal–distal translation of the center of the femoral head was similar in both groups (Table 4). The femoral head center showed a mean medial translation in both groups during the first 2 years (Figure 3). The Corail stem showed an increased mean posterior displacement compared with the CFP stem (p = 0.02). However, if the direction of the movement was disregarded, the median absolute movement in the anterior–posterior direction was similar between groups (p = 0.2). This indicates that the center of the femoral head of the CFP stem moved both posteriorly and anteriorly, whereas there was mainly a tendency to posterior displacement in the Corail group. Further studies of all follow-up occasions (repeated measure ANOVA, 37 CFP and 35 Corail stems) showed no statistically significant differences between the groups, either when signed (p ≥ 0.08) or absolute (p ≥ 0.2) migration data were studied.

We also examined the number of stems that showed RSA migration above the individual 99% detection limit and in any of the 3 directions (medial–lateral, proximal–distal and/ or anterior–posterior). Between the postoperative radiostereometric examination and the examination at 2 years movements above the detection limit were observed in 31 CFP stems and

206

Acta Orthopaedica 2019; 90 (3): 202–208

Proximal (+)/distal (–) translation, mm 1

Anterior (+)/posterior (–) translation, mm

CFP

0

5

CFP

2

–2

–2

–4

–3

–6

–4

–8

3

2

1

–10

–5

–12

–6

CFP

4

0

–1

Medial (+)/lateral (–) translation, mm

0

–14 –7

0

6

12

18

0

24

6

18

24

Anterior (+)/posterior (–) translation, mm

Proximal (+)/distal (–) translation, mm Corail

0

Corail

2

0

6

12

18

24

Months after surgery Medial (+)/lateral (–) translation, mm 5

Corail

4

0

–1

–1

Months after surgery

Months after surgery 1

12

–2

–2

–4

–3

–6

3

2

–8

–4

1

–10

–5

–12

–6

0

–14 –7

0

6

12

18

24

0

6

Months after surgery

12

18

24

Months after surgery

–1

0

6

12

18

24

Months after surgery

Figure 3. Migration of all individual prosthesis along the 3 different axes (mm).

in 30 Corail stems. 9 CFP stems and 9 Corail stems showed detectable movement between the first and second year. Thus, 30 CFP and 29 Corail stems had stabilized within 1 year.

Revisions and complications 1 patient who received a Corail stem had an intraoperative fissure, which was treated with cerclage wires. 2 CFP stems were revised due to loosening at 15 and 21 months. All cultures sampled during the 2 revisions were negative and repeated tests of C-reactive protein were normal. 1 of these 2 patients died 23 months after the operation due to cancer. There were no dislocations or infections.

Discussion Previous reports of the CFP stem showed good short- and mid-term results (Röhrl et al. 2006, Briem et al. 2011, Nowak et al. 2011, Kress et al. 2012, Lazarinis et al. 2013, Hutt et al.

2014, Li et al. 2014, You et al. 2015). Whether the CFP stem improves the outcome in terms of hip function and patient satisfaction compared with a conventional stem has not been

studied previously. Despite a rather extensive clinical evaluation we were not able to demonstrate any benefits for the CFP stem in the 2-year perspective. Overall there are very few randomized studies of short stems. Tomaszewski et al. (2013) compared the clinical outcomes of patients operated with an ultra-short stem (Proxima) with a control group who received a classic design. He concluded that patients in the Proxima group had a better clinical status and a greater quality of life.

In another study of the same short stem design Salemyr et al.

(2015) did not find any difference in the clinical results after 2 years. Hube et al. (2004) conducted a randomized clinical trial in which they compared the Mayo short stem with the ABG stem in 93 patients. The follow-up was short, only 3 months, when they observed higher HHS scores with use of the Mayo stem.

Our study has several limitations. The follow-up is short, which means that late-occurring problems usually related to wear, loosening, and periprosthetic fractures cannot be accounted for. The 2 patients with early loosening are a cause of concern but still too few for any certain conclusions. In 1 of these cases a generalized malignant disease was subsequently diagnosed, which could have had influenced the heal-

Acta Orthopaedica 2019; 90 (3): 202–208

ing potential of the bone immediately after the index operation when the malignancy was unknown. Another limitation is the participation of 11 surgeons, some of whom operated on only a few patients, even if this was done with an experienced colleague as assistant. Corail stems supplemented with tantalum markers or model-based RSA were not available to us, which restricted the radiostereometric evaluation to measurement of femoral head translations. This was the original way to measure stem translations (Mjöberg 1986). Even if lack of stem markers means that rotations not could be studied we think that the information obtained is sufficiently good to measure the most important parameter, namely subsidence (Kärrholm et al. 1994). Femoral head translations can only serve to assume the direction of any stem rotations. Distal and medial translation of the head center can be interpreted as varus tilt, and posterior translation as retroversion or posterior tilt of the stem. Because of different stem shape and neck length, the axis of rotation might vary depending on stem design, which will make it still more difficult to speculate in what way head translations mirror the direction and magnitude of rotation.

Finally, some patients did not complete their clinical questionnaires resulting in 5% missing answers, with 1 patient having 10% missing answers. Strengths of our study are the randomized design, inclusion of comparatively many patients, a wide spectrum of clinical outcome parameters used, and measurements of stem migration with high resolution.

Previous studies of the CFP stem showed an improvement of the HHS to at least 82 points (You et al. 2015) but mostly to 90 or more (Röhrl et al. 2006, Briem et al. 2011, Nowak et al. 2011, Kress et al. 2012, Lazarinis et al. 2013, Hutt et al.

2014, Li et al. 2014). In our study the HHS in the CFP patients improved to a median of 93 at the 2-year follow-up. We do not think that the minor differences observed between the 2 groups studied at the 3 months’ follow-up had any clinical relevance.

Previous studies evaluating the CFP stem indicate a stable fixation and good short and intermediate term results on durability (Röhrl et al. 2006, Briem et al. 2011, Nowak et al. 2011, Kress et al. 2012, Lazarinis et al. 2013, Li et al. 2014, You et al. 2015). Hutt et al. (2014) reported a survivorship of 100% after a mean follow-up of 9 years. Survivorship of the CFP stem in our study was 95% after 2 years. As mentioned above, comorbidity might have contributed to 1 of our 2 revisions, perhaps causing inferior bone quality and/or compromised healing potential. Concerning the 2nd revision we think that this stem was slightly undersized. In the annual report of the Swedish Hip Arthroplasty Register 2015, the 5-year survival rate of the CFP stem was 97%, which was significantly lower than observed in the control group (99%; contemporary uncemented stems of standerd length) (SHAR 2015).

We found radiolucent lines in the proximal Gruen zones in several patients in both groups, which corresponds to previous observations (Röhrl et al. 2006, Nowak et al. 2011, You et al.

2015). The relative amount of neck resorption we observed

207

after 2 years was higher than reported by Pipino and Molfetta (1993). In their study 1% of the 200 observed stems showed neck resorption after a follow-up of 1 to 6 years.

Using the RSA techniques, we found similar absolute translations along any of the axes between the groups. However, we found a statistically significant difference between the mean anterior–posterior motion. We noticed that the center of the femoral head in the Corail group moved posteriorly in most cases, while the center of the femoral head in the CFP group moved both posteriorly and anteriorly. We assume that this translation is a result of the rotation of the stem into retroor anteversion. The 2 other RSA studies performed on the CFP stem both showed retroversion of the stem using the mean translation and rotation (Röhrl et al. 2006, Lazarinis et al. 2013). The range of the data published by Lazarinis et al.

(2013) (–0.26 to 0.55 mm) suggest that the CFP stem moved both into retro- and in anteversion. The increased retroversion observed by us in the Corail group could be an effect of a shorter remaining femoral neck and the shape differences of the 2 stems studied. So far we do not know whether these different patterns of anterior–posterior femoral head translations have any clinical relevance.

2 years after surgery, the mean proximal–distal translation of the femoral head was –0.3 mm in the CFP group. The previous RSA studies of the CFP stem measured smaller mean subsidence (–0.05 and –0.13 mm) (Röhrl et al. 2006, Lazarinis et al. 2013). In these studies subsidence was measured at the center of the stem, which means that the measured values will be less influenced by any varus angulation of the stem. In the study by Röhrl et al. (2006) patients were advised to only partially weight bear during the first 6 weeks, which also could have had an influence.

The migration of the Corail stem along the 3 different axes reported in our study is in line with an RSA study regarding the Corail stem (Campbell et al. 2011). Further follow-up and probably studies of more cases are necessary to determine the limit of acceptable subsidence of the CFP stem not to be associated with an increased risk of clinical loosening. It might be that the presence of continuous migration and especially subsidence past 1 year is a more negative prognostic sign than the magnitude of migration up to 6 months or 1 year, provided that the stem stabilizes within this time period. Von Schewelov et al. (2012) observed distal migration in the majority of cases with maximum values up to 21 mm of the Corail stem inserted in patients with femoral neck fracture. All stems were stabilized during the 2nd year of observation, which could support this theory. Similar observations and use of continuous subsidence as a prognostic bad sign are also supported by previous observations of various uncemented stems followed for short periods with RSA (Luites et al. 2006, Ström et al. 2007, BaadHansen et al. 2011). We also observed a continuous subsidence in our 2 revised cases, but still the knowledge in this field is too limited to allow for any certain conclusions. Long-term studies of larger patient populations will be necessary to map

208

out in more detail the prognostic value of early RSA recordings of uncemented stems.

The CFP stem has been on the market for about 20 years and has been used all over the world. Extensive evaluation shows no difference in clinical results, but we found 2 fixation failures in the CFP group within the first 2 years. Therefore we think that the CFP stem should be used with caution. Careful preoperative planning might be still more important for these types of stems to select the correct size and shape and avoid under-sizing. Whether the use of CFP stem at index surgery gives an advantage during revision surgery is yet to be investigated.

LJK: Collection of clinical data, RSA measurements, statistical analysis, writing manuscript. GP: Data collection, follow-up, study design, review of manuscript. MM: Follow up, review of manuscript. JK: Study design, review of manuscript, statistical and methodological aspects.

The authors would like to thank Bita Shareghi for her help on the RSA measurements.

Acta thanks Lene Bergendal Solberg and Marc J Nieuwenhuijse for help with peer review of this study.

Baad-Hansen T, Kold S, Olsen N, Christensen F, Soballe K. Excessive distal migration of fiber-mesh coated femoral stems. Acta Orthop 2011; 82(3): 308-14. doi: 10.3109/17453674.2011.574562.

Briem D, Schneider M, Bogner N, Botha N, Gebauer M, Gehrke T, Schwantes B. Mid-term results of 155 patients treated with a Collum Femoris Preserving (CFP) short stem prosthesis. Int Orthop 2011; 35(5): 655-60. doi: 10.1007/s00264-010-1020-x.

Campbell D, Mercer G, Nilsson K G, Wells V, Field J R, Callary S A. Early migration characteristics of a hydroxyapatite-coated femoral stem: an RSA study. Int Orthop 2011; 35(4): 483-8. doi: 10.1007/s00264-009-0913-z.

Hube R, Zaage M, Hein W, Reichel H. [Early functional results with the Mayo hip, a short stem system with metaphyseal–intertrochanteric fixation]. Der Orthopade 2004; 33(11): 1249-58. doi: 10.1007/s00132-004-0711-7.

Hutt J, Harb Z, Gill I, Kashif F, Miller J, Dodd M. Ten year results of the Collum Femoris Preserving total hip replacement: a prospective cohort study of seventy five patients. Int Orthop 2014; 38(5): 917-22. doi: 10.1007/s00264-013-2212-y.

Kärrholm J, Borssen B, Lowenhielm G, Snorrason F. Does early micromotion of femoral stem prostheses matter? 4-7-year stereoradiographic follow-up of 84 cemented prostheses. J Bone Joint Surg Br 1994; 76(6): 912-7.

Acta Orthopaedica 2019; 90 (3): 202–208

Kress A M, Schmidt R, Nowak T E, Nowak M, Haeberle L, Forst R, Mueller L A. Stress-related femoral cortical and cancellous bone density loss after Collum Femoris Preserving uncemented total hip arthroplasty: a prospective 7-year follow-up with quantitative computed tomography. Arch Orthop Trauma Surg 2012; 132(8): 1111-19. doi: 10.1007/s00402-012-1537-0.

Lazarinis S, Mattsson P, Milbrink J, Mallmin H, Hailer N P. A prospective cohort study on the short Collum Femoris-Preserving (CFP) stem using RSA and DXA: primary stability but no prevention of proximal bone loss in 27 patients followed for 2 years. Acta Orthop 2013; 84(1): 32-9. doi: 10.3109/17453674.2013.765623.

Li M, Hu Y, Xie J. Analysis of the complications of the Collum Femoris Preserving (CFP) prostheses. Acta Orthop Traumatol Turc 2014; 48(6): 623-7. doi: 10.3944/AOTT.2014.13.0060.

Luites J W, Spruit M, Hellemondt G G, Horstmann W G, Valstar E R. Failure of the uncoated titanium ProxiLock femoral hip prosthesis. Clin Orthop Relat Res 2006; 448: 79-86. doi: 10.1097/01.blo.0000224011.12175.83.

Mjöberg B. Loosening of the cemented hip prosthesis” the importance of heat injury. Acta Orthop 1986; (Suppl 221): 1-40.

Nowak M, Nowak T E, Schmidt R, Forst R, Kress A M, Mueller L A. Prospective study of a cementless total hip arthroplasty with a Collum Femoris Preserving stem and a trabeculae oriented pressfit cup: minimun 6-year follow-up. Arch Orthop Trauma Surg 2011; 131(4): 549-55. doi: 10.1007/ s00402-010-1189-x.

Pipino F, Molfetta L. Femoral neck preservation in total hip replacement. Ital J Orthop Traumatol 1993; 19(1): 5-12.

Röhrl S M, Li M G, Pedersen E, Ullmark G, Nivbrant B. Migration pattern of a short femoral neck preserving stem. Clin Orthop Relat Res 2006; 448:738. doi: 10.1097/01.blo.0000224000.87517.4c.

Salemyr M, Muren O, Ahl T, Boden H, Eisler T, Stark A, Skoldenberg O. Lower periprosthetic bone loss and good fixation of an ultra-short stem compared to a conventional stem in uncemented total hip arthroplasty. Acta Orthop 2015; 86(6): 659-66. doi: 10.3109/17453674.2015.1067087.

SHAR. Swedish Hip Artroplasty Register annual report 2015, page 56.

ISBN 978-91-980507-9-0. https://shpr.registercentrum.se/shar-in-english/ annual-reports-from-the-swedish-hip-arthroplasty-register/p/rkeyyeElz Ström H, Nilsson O, Milbrink J, Mallmin H, Larsson S. The effect of early weight bearing on migration pattern of the uncemented CLS stem in total hip arthroplasty. J Arthroplasty 2007; 22(8): 1122-9. doi: 10.1016/j. arth.2006.11.015.

Tomaszewski W, Kotela I, Kawik L, Bednarenko M, Lorkowski J, Kotela A. Quality of life of patients in the evaluation of outcomes of short stem hip arthroplasty for hip osteoarthritis. Orthop Traumatol Rehabil 2013; 15(5): 439-57. doi: 10.5604/15093492.1084359.

Von Schewelov T, Ahlborg H, Sanzen L, Besjakov J, Carlsson A. Fixation of the fully hydroxyapatite-coated Corail stem implanted due to femoral neck fracture: 38 patients followed for 2 years with RSA and DEXA. Acta Orthop 2012; 83(2): 153-8. doi: 10.3109/17453674.2011.641107.

You R J, Zheng W Z, Chen K, Lv H S, Huang D F, Xiao Y Z, Yang D Y, Su Z Q. Long-term effectiveness of total hip replacement with the Collum Femoris Preserving prosthesis. Cell Biochem Biophys 2015; 72(1): 43-7. doi: 10.1007/s12013-014-0401-y.

Acta Orthopaedica 2019; 90 (3): 209–213

209

Return to work after primary total hip arthroplasty: a nationwide cohort study Raul LAASIK 1, Petteri LANKINEN 2, Mika KIVIMÄKI 3–5, Ville AALTO 3, Mikhail SALTYCHEV 6, Keijo MÄKELÄ 2, and Jussi VAHTERA 7 1 Department

of Orthopaedics and Trauma, Tampere University Hospital, Tampere, Finland; 2 Department of Orthopedics and Traumatology, Turku University Hospital and University of Turku, Turku, Finland; 3 Finnish Institute of Occupational Health, Helsinki, Finland; 4 Clinicum, Faculty of Medicine, University of Helsinki, Helsinki, Finland; 5 Department of Epidemiology and Public Health, University College London, London, UK; 6 Department of Physical and Rehabilitation Medicine, Turku University Hospital and University of Turku, Turku, Finland; 7 Department of Public Health, University of Turku, Turku, Finland Correspondence: Laasik.raul@gmail.com Submitted 2018-02-07. Accepted 2018-12-29.

Background and purpose — While the number of working-age patients undergoing total hip arthroplasty (THA) is increasing, the effect of the surgery on patients’ return to work (RTW) is not thoroughly studied. We aimed to identify risk factors of RTW after THA among factors related to demographic variables, general health, health risk behaviors, and socioeconomic status.

Patients and methods — We studied 408 employees from the Finnish Public Sector (FPS) cohort (mean age 54 years, 73% women) who underwent THA. Information on demographic and socioeconomic variables, preceding health, and health-risk behaviors was derived from linkage to national health registers and FPS surveys before the operation. The likelihood of return to work was examined using Cox proportional hazard modeling.

Results — 94% of the patients returned to work after THA on average after 3 months (10 days to 1 year) of sickness absence. The observed risk factors of successful return to work were: having < 30 sick leave days during the last year (HR 1.8; 95% CI 1.4–2.3); higher occupational position (HR 2.2; CI 1.6–2.9); and BMI < 30 (HR 1.4; CI 1.1–1.7).

Age, sex, preceding health status, and health-risk behaviors were not correlated with RTW after the surgery.

Interpretation — Most employees return to work after total hip arthroplasty. Obese manual workers with prolonged sick leave before the total hip replacement were at increased risk of not returning to work after the surgery.

In 2015, a third of primary THAs in Finland were performed on patients under the age of 65 years (Finnish Arthroplasty Register).

Given the need for a longer working career and patients’ expectations of improved mobility after the surgery, the return to work (RTW) has been suggested to be an important marker of surgery success (Kuijer et al. 2009, Malviya et al.

2014, Tilbury et al. 2014). The RTW has positive effects on patients’ physical and mental health and social and economic benefits if employed (Liang et al. 1986, Koenig et al. 2016).

Of the patients who undergo THA, 68% to 95% return to work (Mobasheri et al. 2006, Bohm 2010, Nunley et al. 2011, Tilbury et al. 2014, Hoorntje et al. 2018). Due to a growing number of annual procedures, however, a substantial number of patients are not able to continue working after THA. We aimed to identify risk factors of unsuccessful RTW after THA among factors related to demographic variables, general health, health-risk behaviors, and socioeconomic status in a large nationwide cohort of public sector employees.

Patients and methods The patients were identified from the Finnish Public Sector (FPS) study, a nationwide register- and survey-based cohort amongst employees of 10 municipalities and 6 hospital districts covering a wide range of occupations—from city mayors and doctors to semiskilled cleaners, nurses, and teachers (Airaksinen et al. 2017). The cohort members had been employed for a minimum of 6 months in the participating organizations between 1991 and 2005 (n = 151,901) (Figure 1). Since 1997, repeated questionnaire data with approximately 4-year intervals have been collected from all employees at work at the time of the survey. The information on base-

© 2019 The Author(s). Published by Taylor & Francis on behalf of the Nordic Orthopedic Federation. This is an Open Access article distributed under the terms of the Creative Commons Attribution-Non-Commercial License (https://creativecommons.org/licenses/by/4.0) DOI 10.1080/17453674.2019.1591081

210

Acta Orthopaedica 2019; 90 (3): 209–213

Finnish Public Sector cohort n = 151,901 THA surgery 1996–2011 n = 1,692 (1%) Patients who responded to survey before surgery n = 408 Returned to work within 1 year after THA n = 383

Failed to returned to work within 1 year after THA (n = 25): – died, 0 – retired to pension, 3 – sick leave continued, 22

Figure 1. Patient selection.

line characteristics before the surgery was derived from the survey responses, the employers’ records, and national health registers. All participants have been linked to surgical data on THA from the National Care Register for Health Care, maintained by the National Institute for Health and Welfare, as well as the National Sickness Absence Register maintained by the Social Insurance Institution of Finland. The linkage data were available until December 31, 2011.

Type of surgery and patient characteristics Of the FPS cohort members, 1,690 underwent a unilateral THA between 1999 and 2011 (Figure 1). Of these, 408 had responded to a survey before the surgery and were included in the study. THA was performed in patients mainly because of primary hip osteoarthritis (n = 369, 90%). Risk factors were measured on average 3.6 years (SD 1.9, range 1.0–9.9) before the operation. The type of surgery was defined as NFB30, NFB40, NFB50, NFB60, NFB62, or NFB99 according to the NOMESCO Classification of Surgical Procedures Version 1.14 by the Nordic Medico-Statistical Committee.

Return to work (RTW) RTW was determined as the number of days between the date of discharge and the date of the end of the sick leave. Return to work was defined here both as full and part time. All Finnish residents aged 16 to 67 years are legitimized to receive daily allowances due to medically certified sickness absence.

All sickness absence periods are medically certified, and they are encoded in the register with start and end dates. The RTW was also dichotomized as yes vs. no.

Risk factors of RTW The participants’ demographic variables (sex, age) and occupational grade at the time of the surgery were obtained from the employers’ registers. The spectrum of occupational grades was coded according to the International Standard Classification of Occupation (ISCO) and then downsized to 3 groups: higher-grade non-manual workers (e.g., teachers, physicians), lower-grade non-manual workers (e.g., registered nurses, technicians), and manual workers (e.g., cleaners, maintenance

workers). Marital status (married or cohabiting vs. single, divorced, or widowed) was obtained from a baseline questionnaire. Age was defined in full years at the time of THA. Information on health and health behaviors was obtained from the baseline questionnaire and national health registers. Physical activity was defined as average weekly hours of leisure-time physical activity (walking, brisk walking, jogging, and running, or similar), including commuting, during the previous year (Kujala et al. 1998). The hours per week spent on activity at each intensity level were multiplied by the average energy expenditure of each activity expressed in metabolic equivalent of task (MET). Physical activity was categorized into 2 groups, “low” (≤ 14 MET-hours/week) and “high” activity (> 14 MET-hours/week). Alcohol consumption was categorized according to the habitual frequencies of drinking beer, wine, and spirits as “none”, “moderate,” and “heavy” consumption. The cut-off for heavy alcohol consumption was set as 210 g/week (Rimm et al. 1999). Smoking status was dichotomized as “currently smoking” vs. “quit or never smoked.” Self-reported body weight and height were used to calculate BMI, which was used to identify obese (BMI ≥ 30) and non-obese (BMI < 30) participants. Psychological distress was measured with the 12-item version of the General Health Questionnaire (GHQ) (Goldberg et al. 1997) with 3 or more positive response set as a cut-off point of psychological distress (“yes” vs. “no”). Participant rated their general health on a 5-point scale (from 1 = “good health” to 5 = “worst health”), and the self-rated health was then dichotomized by categorizing response scores 1 and 2 as good and scores 3 to 5 as poor general health. The data on the presence of diabetes, coronary heart disease, asthma, chronic obstructive pulmonary disease, or rheumatoid arthritis were obtained from the Drug Reimbursement Register, which contains information on persons entitled to special reimbursement for treatment of chronic health conditions. The presence of comorbidity was then dichotomized as “yes” vs. “no.”

Statistics The participants were followed from the date of discharge to the date when an employee returned to work, was granted a disability pension, an old-age pension, died, or end of study (December 31, 2011), whichever came first.

Cox proportional hazard models were used to study the associations between baseline characteristics and return to work. We examined the associations separately for each risk factor adjusted for age and sex. The results are presented as hazard ratios (HR) and their 95% confidence intervals (CI).

Proportional hazards assumption of the Cox model was fulfilled concerning all other parameters studied, except socioeconomic status, alcohol consumption, and perceived general health. For these 3 parameters, Schemper’s weighted Cox regression was used (Schemper et al. 2009).

All analyses were performed using the SAS statistical software, version 9.1.3 (SAS Institute, Cary, NC, USA).

Acta Orthopaedica 2019; 90 (3): 209–213

211

Baseline characteristics of the patients and their associations with the rate of return to work after THA. Hazard ratios (HR) and their 95% confidence intervals (CI) are derived from Cox proportional hazard analyses. Schemper’s weighted Cox regression was used concerning occupational status, alcohol consumption and self-related health because proportional hazards assumption was not fulfilled Separately

THA analyzed a

n (%) HR (CI) Age b (0 missing) 54.3 (SD 6.5) Sex (0 missing) Men 110 (27) Women 298 (73) (ref) Married or cohabiting (6 missing) Yes 315 (78) No 87 (22) Obese (BMI >30) (11 missing) No 297 (75) Yes 100 (25) Current smoking (11 missing) No 324 (82) Yes 73 (18) High alcohol consumption (4 missing) No 360 (89) Yes 44 (11) Comorbidities (0 missing) No 349 (86) Yes 59 (14) Psychological distress (3 missing) No 289 (71) Yes 116 (29) Self-rated health (5 missing) Good 214 (53) Poor 189 (47) Physically active (3 missing) (MET-hours/week >14) Yes 267 (66) No 138 (34) Occupational status (3 missing) Higher level non-manual 115 (28) Lower level non-manual 137 (34) Manual 153 (38) Preoperative sickness absence c (0 missing) No 277 (68) Yes 131 (32) Year of surgery (0 missing) 2007–2011 243 (60) 1998–2006 165 (40) a b c

Probability of return to work

Probability of return to work

1

1

0.8

0.8

0.6

0.6

0.4

0.4

0.2

0.2

Higher level non-manual Lower level non-manual Manual

1.0 (1.0–1.0) 1.2 (0.9–1.5) 1.0 (ref) 1.1 (0.9–1.5) 1.0 (ref)

0

A

0 0

100

200

300

400

Days after index operation B

Probability of return to work

1

1.4 (1.1–1.7) 1.0 (ref)

0.8

0.8

0.9 (0.7–1.2) 1.0 (ref)

0.6

0.6

0.9 (0.6–1.5) 1.0 (ref)

0.4

0.4 Preoperative sickness abscence > 30 days: No Yes

0.2

1.2 (0.9–1.5) 1.0 (ref) 1.0 (0.8–1.3) 1.0 (ref)

200

300

400

Days after index operation

0.2

0

1.0 (0.8–1.2) 1.0 (ref)

100

Probability of return to work

1

1.3 (1.0–1.8) 1.0 (ref)

0

BMI > 30 No Yes

0 0

C

100

200

300

400

Days after index operation

0

D

100

200

300

400

Days after index operation

Figure 2. Probability of return to work overall (A), or according to occupational status (B), preoperative sickness absence (C), and obesity (D). The expected dispersion of survival curves between categories of these patient characteristics was observed across the entire follow-up period (log-rank test: p < 0.001 for occupational status and sickness absences, p = 0.01 for obesity).

2.2 (1.6–2.9) 1.3 (1.0–1.6) 1.0 (ref)

Results 1.8 (1.4–2.3) 1.0 (ref) 1.2 (1.0–1.4) 1.0 (ref)

Age and sex adjusted if appropriate Increase in age >30 days of sickness absence within the 1-year period preceding the operation

Ethics, funding, and potential conflicts of interest The ethics committee of the Hospital District of Helsinki and Uusimaa approved the study (ethical approval: 22.3.2011, Dnor 60/13/03/00/2011). All investigations were conducted in conformity with ethical principles of research. MK received a grant from the Academy of Finland (311492) for the conduct of the study. No competing interests were declared.

The majority, 73% of patients, were women (Table). The occupational status was evenly distributed among the 3 groups.

The average age at the time of surgery was 54 years (32–67).

15% of patients had some chronic medical comorbidity and 53% reported good general health (Table).

Risk factors of RTW After the surgery, 94% (n = 383) of the patients returned to work after a mean of 103 days (10–354) of sickness absence (Figure 2). Of the prognostic variables examined, occupational status, sickness absences before the operation, and obesity were statistically significantly associated with RTW. Patients with higher non-manual occupational status had a 2.2 (CI 1.6–2.9) times higher probability of RTW than manual workers adjusted for age and sex (Table). Low compared with high number of sick leaves (< 30 days) before the surgery was associated with a 1.8 (CI 1.4–2.3) times higher probability of

212

RTW. Also, non-obese patients were more likely to return to work than obese patients (HR 1.4 (CI 1.1–1.7).

Figure 2 illustrates the RTW curves by occupational status, sickness absences before the operation, and obesity. The expected dispersion of curves between categories of these patient characteristics was observed across the entire followup period (log-rank test: p < 0.001 for occupational status and sickness absences, p = 0.01 for obesity).

Demographic variables (age and sex), from socioeconomic factors marital status, health behaviors (smoking, physical activity, and high alcohol consumption), chronic medical comorbidities (asthma, diabetes mellitus, rheumatoid arthritis, and coronary artery disease), or psychological distress were not statistically significantly associated with RTW (Table).

Discussion Successful return to work has been identified as a crucial outcome marker for patients after THA (Kuijer et al. 2009, Malviya et al. 2014, Tilbury et al. 2014). Larger patient series are needed to assess the principal factors for successful return to work and to evaluate the interaction of these factors using multivariate analysis.

Our findings are in line with previous studies. Stigmar et al.

(2017) reported that, for Swedish patients who started a sick leave period from day 7 before surgery and who had had no ongoing sick leave from day 30 to day 8 prior to THA surgery, the median postoperative sick leave period for women and men was similar, 3 months. In our study, for patients who did not have > 30 days’ sickness absence within the 1-year period preceding the operation the median postoperative sick leave was similar.

The employment status and sick leaves before the surgery have been suggested to be important determinants of returning to work after hip replacement (Jensen et al. 1985, Johnsson and Persson 1986, Mobasheri et al. 2006, Bohm 2010, Cowie et al. 2013). Bohm (2010) did not observe a correlation between RTW and sex, smoking status, or high alcohol consumption.

Lower socioeconomic statuses have been reported to be associated with a higher incidence of osteoarthritis and a higher disability level (Thumboo et al. 2002, Wetterholm et al. 2016).

In a Swedish cohort study, osteoarthritis was diagnosed earlier and the incidence rate of hip replacement surgery was higher among patients with higher income and higher educational level (Wetterholm et al. 2016). While some previous studies have produced conflicting evidence on the influence of obesity on THA outcome (Moran et al. 2005, Namba et al. 2005, Davis et al. 2011, Alvi et al. 2015), others have reported that obesity might also affect RTW (Cowie et al. 2013).

Identifying factors predicting RTW after THA may help in patient selection and setting adequate goals for medical and occupational rehabilitation after the surgery. Numerous factors have been suggested as potential risk factors of unsuccessful RTW after THA: health behaviors, age, educational

Acta Orthopaedica 2019; 90 (3): 209–213

level, comorbidities, surgical complications, and type of work, as well as the length of sick leaves before the surgery (Nevitt et al. 1984, Jensen et al. 1985, Johnsson and Persson 1986, Suarez et al. 1996, Mobasheri et al. 2006, Kuijer et al. 2009, Nunley et al. 2011, Cowie et al. 2013, Malviya et al. 2014, Tilbury et al. 2014, Leichtenberg et al. 2016, Hoorntje et al.

2018). Previous studies in this field of research have mostly been conducted on relatively small samples exploiting separately either register-based data or data obtained from surveys and, in most cases, risk factors have been measured close to the time of surgery, blocking out the long-term effects of risks (Kuijer et al. 2009, Nunley et al. 2011, Malviya et al. 2014, Tilbury et al. 2014, Kleim et al. 2015, Bardgett et al. 2016a).

The relative importance of the multiple risk factors of not returning to work has therefore remained unclear.

In relation to socioeconomic status and preoperative sick leave, our findings were in line with the conclusions of 3 previous systematic reviews and 1 meta-analysis on the subject (Kuijer et al. 2009, Malviya et al. 2014, Tilbury et al. 2014, Hoorntje et al. 2018). Contrary to findings of these reviews, we did not find any effect of age on RTW. In relatively young patients (mean age 50 years) Nunley et al. (2011) reported that younger age, physical activity, and employment at the time of surgery were associated with RTW. Cowie et al. (2013), Jensen et al. (1985), and Bohm (2010) have also reported on a correlation between RTW and age, physical condition, and comorbidities. Workplace conditions have also been found to be associated with work performance after the THA (Kuijer et al. 2009, Malviya et al. 2014, Tilbury et al. 2014).

The reasons for differences between our results and earlier reports may lie in the diversities in studied samples that are, especially when insurance issues are involved, bounded to national health care and insurance systems.

Strengths and weaknesses Compared with previous literature, our study has notable strengths. The cohort size was large enough to employ sophisticated statistical methods. The studied sample came from a well-characterized occupational cohort and represented a wide range of occupations. The comprehensive data have been gathered on health and health-risk behaviors before the surgery. All the data were linked to reliable national health registers including detailed information on objective data on the date of the operation and the beginning and end dates of all periods of sickness absence and comorbidities, enabling accurate estimation of the timing of return to work. Many risk factors of unsuccessful RTW, such as occupational status, sickness absences before the operation, and comorbid medical conditions, were measured objectively from the registers.

The generalizability of these findings may be affected by the differences in national welfare, pension, and workers’ compensation schemes (Scott et al. 2017). The studied cohort was limited to employees in the public sector and dominated by women. No data on workplace adjustments before or after

Acta Orthopaedica 2019; 90 (3): 209–213

the surgery were available. The motivation for RTW, which is always a potential source of uncertainty in studies on such a subject, remains unknown. No data on possible complications and revision surgery were available. Prosthetic joint infections, periprosthetic fractures, or dislocations may remarkably influence the RTW. Another limitation is that we were not able to assess patients’ activity levels, patient-reported THA satisfaction, or functional outcome scores. The RTW may also be influenced by patients’ interactions with healthcare professionals as well as patients’ pre-surgery expectations (Bardgett et al. 2016b), which were not known.

In summary, more than 9 out of 10 working-aged patients returned to work after THA. Obese manual workers with prolonged sick leave before the total hip replacement were at increased risk of not returning to work after the surgery. These risk factors should be accounted for when planning THA and post-surgery rehabilitation. All the other studied socioeconomic factors were not associated with the return to work rate.

Orthopedic surgeons should consider referring patients at risk for no RTW for additional work-directed care.

RL, PL, KM, and JV designed the study. RL, PL, VA, MK, KM, and JV carried out the analytical aspects of the study, including statistical analysis and modeling. RL, PL, MS, KM, and JV drafted the manuscript. KM and JV contributed equally to the study, and are joint senior authors.

Acta thanks Martin Englund and Paul Kuijer for help with peer review of this study.

Airaksinen J, Jokela M, Virtanen M, Oksanen T, Pentti J, Vahtera J, Koskenvuo M, Kawachi I, Batty G D, Kivimaki M. Development and validation of a risk prediction model for work disability: multicohort study. Sci Rep 2017; 7(1): 13578.

Alvi H M, Mednick R E, Krishnan V, Kwasny M J, Beal M D, Manning D W. The effect of BMI on 30 day outcomes following total joint arthroplasty. J Arthroplasty 2015; 30(7): 1113-17.

Bardgett M, Lally J, Malviya A, Kleim B, Deehan D. Patient-reported factors influencing return to work after joint replacement. Occup Med (Lond) 2016a; 66(3): 215-21.

Bardgett M, Lally J, Malviya A, Deehan D. Return to work after knee replacement: a qualitative study of patient experiences. BMJ Open 2016b; 6(2): e007912,2015-007912.

Bohm E R. The effect of total hip arthroplasty on employment. J Arthroplasty 2010; 25(1): 15-18.

Cowie J G, Turnbull G S, Ker A M, Breusch S J. Return to work and sports after total hip replacement. Arch Orthop Trauma Surg 2013; 133(5): 695700.

Davis A M, Wood A M, Keenan A C, Brenkel I J, Ballantyne J A. Does body mass index affect clinical outcome post-operatively and at five years after primary unilateral total hip replacement performed for osteoarthritis? A multivariate analysis of prospective data. J Bone Joint Surg Br 2011; 93(9): 1178-82.

Finnish Arthroplasty Register (FAR). Http://www.thl.fi/far Goldberg D P, Gater R, Sartorius N, Ustun T B, Piccinelli M, Gureje O, Rutter C. The validity of two versions of the GHQ in the WHO study of mental illness in general health care. Psychol Med 1997; 27(1): 191-7.

Hoorntje A, Janssen K Y, Bolder S B T, Koenraadt K L M, Daams J G, Blankevoort L, Kerkhoffs G M M J, Kuijer P P F M. The effect of total hip arthroplasty on sports and work participation: a systematic review and meta-analysis. Sports Med 2018; 48(7): 1695-726.

213

Jensen J S, Mathiesen B, Tvede N. Occupational capacity after hip replacement. Acta Orthop Scand 1985; 56(2): 135-7.

Johnsson R, Persson B M. Occupation after hip replacement for arthrosis.

Acta Orthop Scand 1986; 57(3): 197-200.

Kleim B D, Malviya A, Rushton S, Bardgett M, Deehan D J. Understanding the patient-reported factors determining time taken to return to work after hip and knee arthroplasty. Knee Surg Sports Traumatol Arthrosc 2015; 23(12): 3646-52.

Koenig L, Zhang Q, Austin M S, Demiralp B, Fehring T K, Feng C, Mather R C, 3rd, Nguyen J T, Saavoss A, Springer B D, Yates A J, Jr. Estimating the societal benefits of THA after accounting for work status and productivity: a Markov model approach. Clin Orthop Relat Res 2016; 474(12): 2645-54.

Kuijer P P, de Beer M J, Houdijk J H, Frings-Dresen M H. Beneficial and limiting factors affecting return to work after total knee and hip arthroplasty: a systematic review. J Occup Rehabil 2009; 19(4): 375-81.

Kujala U M, Kaprio J, Sarna S, Koskenvuo M. Relationship of leisure-time physical activity and mortality: the Finnish twin cohort. JAMA 1998; 279(6): 440-4.

Leichtenberg C S, Tilbury C, Kuijer P, Verdegaal S, Wolterbeek R, Nelissen R, Frings-Dresen M, Vliet Vlieland T. Determinants of return to work 12 months after total hip and knee arthroplasty. Ann R Coll Surg Engl 2016; 98(6): 387-95.

Liang M H, Cullen K E, Larson M G, Thompson M S, Schwartz J A, Fossel A H, Roberts W N, Sledge C B. Cost-effectiveness of total joint arthroplasty in osteoarthritis. Arthritis Rheum 1986; 29(8): 937-43.

Malviya A, Wilson G, Kleim B, Kurtz S M, Deehan D. Factors influencing return to work after hip and knee replacement. Occup Med (Lond) 2014; 64(6): 402-9.

Mobasheri R, Gidwani S, Rosson J W. The effect of total hip replacement on the employment status of patients under the age of 60 years. Ann R Coll Surg Engl 2006; 88(2): 131-3.

Moran M, Walmsley P, Gray A, Brenkel I J. Does body mass index affect the early outcome of primary total hip arthroplasty? J Arthroplasty 2005; 20(7): 866-9.

Namba R S, Paxton L, Fithian D C, Stone M L. Obesity and perioperative morbidity in total hip and total knee arthroplasty patients. J Arthroplasty 2005; 20(7 Suppl 3): 46-50.

Nevitt M C, Epstein W V, Masem M, Murray W R. Work disability before and after total hip arthroplasty. assessment of effectiveness in reducing disability. Arthritis Rheum 1984; 27(4): 410-21.

Nunley R M, Ruh E L, Zhang Q, Della Valle C J, Engh C A Jr, Berend M E, Parvizi J, Clohisy J C, Barrack R L. Do patients return to work after hip arthroplasty surgery. J Arthroplasty 2011; 26(6 Suppl): 92,98.e1-3.

Rimm E B, Williams P, Fosher K, Criqui M, Stampfer M J. Moderate alcohol intake and lower risk of coronary heart disease: Meta-analysis of effects on lipids and haemostatic factors. BMJ 1999; 319(7224): 1523-8.

Schemper M, Wakounig S, Heinze G. The estimation of average hazard ratios by weighted Cox regression. Stat Med 2009; 28: 2473-89.

Scott C E H, Turnbull G S, MacDonald D, Breusch S J. Activity levels and return to work following total knee arthroplasty in patients under 65 years of age. Bone Joint J 2017; 99-B(8): 1037-46.

Stigmar K, Dahlberg L E, Zhou C, Jacobson Lidgren H, Petersson I F, Englund M. Sick leave in Sweden before and after total joint replacement in hip and knee osteoarthritis patients. Acta Orthop 2017; 88(2): 152-7.

Suarez J, Arguelles J, Costales M, Arechaga C, Cabeza F, Vijande M. Factors influencing the return to work of patients after hip replacement and rehabilitation. Arch Phys Med Rehabil 1996; 77(3): 269-72.

Thumboo J, Chew L H, Lewin-Koh S C. Socioeconomic and psychosocial factors influence pain or physical function in Asian patients with knee or hip osteoarthritis. Ann Rheum Dis 2002; 61(11): 1017-20.

Tilbury C, Schaasberg W, Plevier J W, Fiocco M, Nelissen R G, Vliet Vlieland T P. Return to work after total hip and knee arthroplasty: a systematic review. Rheumatology (Oxford) 2014; 53(3): 512-25.

Wetterholm M, Turkiewicz A, Stigmar K, Hubertsson J, Englund M. The rate of joint replacement in osteoarthritis depends on the patient’s socioeconomic status. Acta Orthop 2016; 87(3): 245-51.

214

Acta Orthopaedica 2019; 90 (3): 214–219

Not all cemented hips are the same: a register-based (NJR) comparison of taper-slip and composite beam femoral stems Hussain A KAZI 1, Sarah L WHITEHOUSE 2, Jonathan R HOWELL 1, and A John TIMPERLEY 1,3 1 Princess

Elizabeth Orthopaedic Centre, Royal Devon and Exeter NHS Foundation Trust, Exeter, UK 2 Queensland University of Technology (QUT), Brisbane, Queensland, Australia; 3 University of Exeter, Exeter, UK Correspondence: jtimperley@mac.com Submitted 2017-10-24. Accepted 2019-01-14.

Background and purpose — No difference in outcome has been demonstrated comparing cemented taper-slip and composite beam designs in short-term randomised trials; we assessed outcome differences using a registry analysis.

Patients and methods — All cemented stems with > 100 implantations were identified in the National Joint Registry of England and Wales from April 1, 2003 to September 31, 2013 and categorised as taper-slip or composite beam.

Survival analyses using Kaplan–Meier and Cox regression were performed.

Results — We identified 292,987 cemented arthroplasties, of which 16% (47,586) were composite beam stems, with taper-slip stems making up the remainder (n = 245,401).

There was a statistically significant increased chance of revision in the composite beam group compared with the taper-slip group (1.7% vs 1.3%, p < 0.001) but statistically no significant differences of survival estimates (p = 0.06).

When the 2 groups were segregated to delineate the most implanted model in each category, the differences became more profound with the most implanted taper-slip stem (Exeter V40) showing statistically and clinically significant superior 8-year survival: 97.9% compared with 97.6% for all other taper-slip; 97.5% for the most implanted composite beam (Charnley cemented stem); and 97.7% for all other composite beam.

Interpretation — There was an increased incidence of revision for composite beam stems. The most implanted taper-slip stem demonstrated significant survival advantage vs. all other stems.

Cemented femoral stems can be divided into designs that achieve fixation as a composite beam and those that function as a taper-slip device (Shah and Porter 2005). Taper-slip stem designs function by controlled stem subsidence within the cement mantle whereas composite beam stems seek mechanical interlock at all interfaces including fixation between the stem and cement.

Radiostereometry studies (Alfaro-Adrian et al. 2001) have shown differences between taper-slip and composite beam stems with respect to their migration and micromotion. Polished tapered stems subside within cement, with no movement occurring at the cement–bone interface. In contrast composite beam stems subside over smaller distances but crucially this occurs at both the stem–cement and cement–bone interfaces.

Movement of the cement in relation to bone indicates that fixation at the cement–bone interface is compromised and the cement cannot be osseointegrated (Schmalzried et al. 1992).

Despite the findings in in vitro and implant retrieval studies (Huiskes et al. 1998, Verdonschot and Huiskes 1998, Howell et al. 2004), most in vivo reports have failed to determine a difference in outcome between composite beam and taper-slip designs (Lachiewicz et al. 2008, Jayasuriya et al. 2013), most likely due to small numbers.

We investigated revision rates in the UK for primary cemented hips by prosthesis subgroup of taper-slip and composite beam stems.

Patients and methods The National Joint Registry of England & Wales (and latterly Northern Ireland and Isle of Man) (NJR) was established in 2002. Patient demographics and surgical details are recorded, with mortality information being updated biannually and sub-

© 2019 The Author(s). Published by Taylor & Francis on behalf of the Nordic Orthopedic Federation. This is an Open Access article distributed under the terms of the Creative Commons Attribution-Non-Commercial License (https://creativecommons.org/licenses/by/4.0) DOI 10.1080/17453674.2019.1582680

Acta Orthopaedica 2019; 90 (3): 214–219

215

Table 1. Design features of different cemented stems (after Huiskes et al. 1998) Design

Force closed (taper-slip)

Shape closed (composite beam)

Surface Finish Polished Roughened/matt Taper + +/– Collar – + Ridges/flanges/profiles – +

sequent revisions linked to the primary operation, with more than 94% completeness reported (Porter 2017).

We performed an approved retrospective cohort study of the NJR dataset. Data were requested to provide information regarding potential confounding factors. The study population included all validated cemented primary total hip operations performed in England and Wales from April 1, 2003 to September 30, 2013, as this request preceded Northern Ireland and the Isle of Man joining the NJR (2013 and 2015 respectively). The mean length of follow-up for this cohort was 4.2 years (0–12).

Using the criteria by Huiskes (1998) (Table 1), stem designs were subdivided in terms of whether they were taper-slip or composite beam using published data (predominantly surface finish). Only stems with >100 implantations were included.

In order to remove bias of metal-on-metal hips in analysing the effect of stem design on outcome, the definitive analysis was performed excluding metal-on-metal and ‘unknown’ bearing couples.

The most commonly implanted stems of both designs were then separated in order to examine whether stems with the same design philosophy function in an identical fashion giving equivocal results. The final analysis therefore comprised 4 cohorts: most implanted taper-slip (Exeter, Stryker Orthopaedics, Mahwah, NJ); all other taper-slip; most implanted composite beam (Charnley, DePuy Orthopaedics, Warsaw, IN); and all other composite beam.

Statistics Statistical analysis was performed using IBM SPSS (Version 22, IBM Corp, Armonk, NY, USA) and NCSS (NCSS 10 Statistical Software (2015). NCSS, LLC. Kaysville, UT, USA, ncss.com/software/ncss). Cox regression analysis (using the Enter method where all variables are added as a single block) was used to identify revision rates within subgroups and factors influencing these rates. Hazard ratios (HR) and 95% confidence intervals (CI) are presented. Frequencies were compared using the chi-squared (χ2) test and continuous variables compared using analysis of variance (ANOVA). Confounding factors were investigated: age, sex, ASA grade, procedure type (routine/complex), diagnosis, approach, and bearing couple.

Surgeon grade and provider type (public or private) were not provided by the NJR. Data validation was performed prior to

analysis by scrutiny of the data, including categorisation of stem types, use of cement, examination of missing and invalid responses according to surgical details, and coding and validation of diagnosis and reasons for revision. Following validation, there were minimal missing values (5 for sex) other than for approach, where these cases were treated as a separate group in order to determine whether any bias existed. The 5 cases with missing sex were excluded from the Cox regression model. Kaplan–Meier survival curves were constructed with cut-off at 8 years where the appropriate effective number of cases at risk remained, utilising the guidance stipulated by Pocock et al. (2002) and Lettin et al. (1991) and cumulative survival compared using the log-rank test. Competing risk analysis was not adopted as it is more appropriate when the risk of death is high (Gillam et al. 2010) and may not be the most appropriate for estimating implant failure (Sayers et al.

2018).

Ethics, funding, and potential conflicts of interests This work was approved by the National Joint Registry Research Sub-Committee. The work involves de-identified data so is exempt from IRB approval.

JRH and AJT have received or will receive benefits for personal or professional use from Stryker Corporation. In addition, benefits have been directed to a research fund and educational institution with which SLW, JRH, and AJT are associated. No funding was received specifically for this project, and there was no input from any commercial interest for any aspect of this study.

Results 292,987 primary cemented hip replacements were included.

Composite beam stems accounted for 16% (47,586 hips), with the remainder being taper-slip stems. Exeter V40 was the commonest taper-slip design and Charnley cemented stem the commonest composite beam design (Table 2, see Supplementary data). There was a tendency for composite beam stems to be used in slightly older patients (mean 73.6 years) than taper-slip (mean 71.9 years) (Table 3, see Supplementary data) although this is unlikely to be clinically relevant. There was a higher proportion of deaths (17.2%) in the composite beam group compared with 10.5% in the taper-slip group (Tables 3 and 4, see Supplementary data), but more detailed exploration is beyond the scope of this project. Ignoring the deaths in both groups, there was a statistically significant increased chance of revision in the composite beam group compared with the taper-slip group (1.7% vs. 1.3%, p < 0.001) (Table 3, see Supplementary data).

Kaplan–Meier survival curves were constructed comparing the 2 groups (Figure 1). Both design philosophies had similar curves; log-rank test, p = 0.06: taper-slip stem 97.9% (CI 97.8–98.0); composite beam 97.6% (CI 97.4–97.8) at 8 years.

216

Acta Orthopaedica 2019; 90 (3): 214–219

Figure 1. Kaplan–Meier survival curve for taper-slip and composite beam stems. Taperslip 97.9% (CI 97.8–98.0) and composite beam 97.6% (97.4–97.8) 8-year survival.

Figure 2. Kaplan–Meier survival curve for most implanted taper-slip, all other taper-slip, most implanted composite beam and all other composite stems. Exeter 97.9% (CI 97.8–98.0), all other taper-slip 97.6% (97.4–97.8), Charnley 97.5% (CI 97.2–97.8), and all other composite beam 97.7% (CI 97.4–98.0) 8-year survival.

Figure 3. Plot of survival functions for each group when adjusted for confounders using Cox regression.

Table 5. Reasons for revision by stem type. Values are frequency (%) (multiple reasons allowable)

Total Reason for revision n = 292,987 Aseptic loosening stem Aseptic loosening socket Dislocation Stem fracture Infection Stem lysis Pain Peri-prosthetic fracture stem Other Total

Exeter All other Charnley All other (taper-slip) taper-slip (composite beam) composite beam n = 176,189 n = 69,043 n = 23,141 n = 24,445 p-value

490 169 (0.1) 97 (0.1) 722 395 (0.2) 147 (0.2) 1140 638 (0.4) 296 (0.4) 50 30 (0.02) 16 (0.02) 921 516 (0.3) 181 (0.3) 127 48 (0.03) 28 (0.04) 570 285 (0.2) 157 (0.2) 437 215 (0.1) 192 (0.3) 1,035 559 (0.3) 270 (0.4) 5,492 (1.9) 2,855 (1.6) 1,384 (2.0)

The dataset was further analysed comparing the most implanted taper-slip stem (Exeter), all other taper-slip stems, most implanted composite beam stem (Charnley), and all other composite beam stems in 4 separate cohorts (Table 4, see Supplementary data). Reasons for revision are shown (Table 5).

The risk of aseptic loosening and stem lysis was higher for composite beam stems than taper-slip varieties (Table 5), as were the rates of revision for infection. There was a difference in the risk of peri-prosthetic fracture between the most implanted taper-slip stem design (0.1%) and all other taperslip stems (0.3%), both higher than the composite beam groups, which was statistically significant (p < 0.001, chisquared test). All other reasons for revision were of similar incidence between the 2 stem designs.

148 (0.6) 83 (0.4) 108 (0.5) 2 (0.01) 125 (0.5) 32 (0.14) 74 (0.3) 18 (0.1) 104 (0.4) 694 (3.0)

76 (0.3) 97 (0.4) 98 (0.4) 2 (0.01) 99 (0.4) 19 (0.08) 54 (0.2) 12 (0.05) 102 (0.4) 559 (2.3)

< 0.001 < 0.001 0.02 0.3 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001

When the dataset was further subdivided to assess all 4 groups, however, the survival curves changed, with a superior survival for the most commonly implanted taper-slip stem compared with all other taper-slip (p < 0.001) and most commonly implanted composite beam (p = 0.01), (Figure 2, Table 6).

Finally, in order to adjust for known confounders (age, sex, diagnosis, ASA grade, procedure type, approach, and bearing couple), Cox regression analysis was performed (Table 7, see Supplementary data) and adjusted survival curves plotted (Figure 3), indicating the superior results of the most implanted taper-slip (Exeter) group (all other taper-slip HR 1.2 [CI 1.1– 1.3]; Charnley HR 1.2 [CI 1.0–1.3]; other composite beam HR 1.2 [CI 1.1–1.3]). These results remained consistent when taperslip and composite beam were compared; HR 1.1 (CI 1.0–1.2).

Acta Orthopaedica 2019; 90 (3): 214–219

217

Table 6. Survival rates between groups Stem type

Kaplan–Meier 8-year survival (95% CI)

Most implanted taper-slip (Exeter) All other taper-slip Most implanted composite beam (Charnley) All other composite beam

97.9% (CI 97.8–98.0) 97.6% (CI 97.4–97.8) 97.5% (CI 97.2–97.8) 97.7% (CI 97.4–98.0)

Discussion This is the first study to assess the performance of cemented femoral components over the first 10 years of NJR data.

Our analysis on almost 300,000 THAs in the NJR initially showed similar results between taper-slip and composite beam cemented stems, as predicted from theoretical studies (Schmalzried et al. 1992, Alfaro-Adrian et al. 2001). However, closer examination identified clear differences within those groups when the most implanted of each group was separated out, so clearly the performance of an individual stem design cannot be predicted by a simple categorisation between taperslip and composite beam.

A study using the Finnish Arthroplasty Register compared the outcomes of the 12 most popular cemented stem designs over a 25-year period. Both the Exeter and Muller straight stem achieved greater than 90% survivorship at 15 years with aseptic loosening as an endpoint (Makela et al. 2008). This again suggests that good results, in terms of survivorship, are possible when composite beam and taper-slip stems are used.

2 randomised trials have been performed comparing stems with different design philosophies. Lachiewicz et al. (2008) enrolled 201 patients (219 hips) and found no differences at 5 years comparing taper-slip and a roughened pre-coat stem in terms of revision for loosening or failure. Jayasuriya et al.

(2013) compared a composite beam design (Charnley) with a double-tapered (Exeter) and triple-tapered (C-stem) design.

At the 2-year review of 120 patients, no difference in bone remodelling or outcomes between the 3 groups was found.

Numerous cohort (Van Eynde et al. 2010, Broden et al.

2015), case control (Sarvilinna et al. 2005), randomised trials (Lachiewicz et al. 2008) and registry studies (Hailer et al.

2010, Thien et al. 2014) have compared revision rates and peri-prosthetic fracture rates for cemented and uncemented components and have compared peri-prosthetic fracture rates by cemented fixation type. Overall the risk of peri-prosthetic fracture is higher with uncemented stems. In a study of 437,629 patients in the Nordic Arthroplasty Register Association the relative risk for peri-prosthetic fracture in the uncemented group was 8.7 with the risk increasing with increasing age (Thien et al. 2014). Amongst cemented stem designs there is evidence that peri-prosthetic fracture rates are higher in those with a polished tapered stem after hip fracture (Sarvilinna et al. 2005). Thien et al. (2014) in a registry analysis revealed a higher peri-prosthetic fracture risk for a polished

tapered stem when compared with a composite beam counterpart. We confirmed that there is a statistically significant difference in peri-prosthetic fracture risk between taper-slip and composite beam stems but this risk is offset by the decreased risk for revision for other indications.

Harris (1998), an advocate of roughened pre-coated stems, reviewed the results of various stem designs and postulated that roughening per se was not deleterious due to the multiple series and designs demonstrating good outcomes. He made the point that specific stem geometry issues may lead to poorer results with some designs more than others. We did not separate the results of different brands of stems that function as composite beam devices but it is worth noting that, even within a single brand, differences in results have been described that have their origins in modifications to the shape and surface finish of the implant (Dall et al. 1993).

Polished tapered stems, be they double- or triple-tapered, have demonstrated excellent long-term results due to their taper-slip geometry and mode of action. The Exeter stem, the most implanted stem identified in the series described, is a polished double-tapered design, earlier iterations of which have shown excellent results at up to 17 years follow-up and beyond in both the design centre (Carrington et al. 2009, Petheram et al. 2016) and independent units (Hook et al. 2006, Young et al. 2009). These results have held true in both the general population and those under 50 years old at the time of surgery. Other designs of collarless, polished, tapered stems also have good published results in the literature (Purbach et al. 2013, Junnila et al. 2016) but we have identified in this registry analysis that, overall, the results of the Exeter stem were statistically significantly better than those of other stems combined. This may be due to some poorly performing stems included in this group, but individual brand comparisons were beyond the scope of this study. The results for almost all indications for revision were improved when the market-leading stem was implanted.

Whilst individual studies are useful, the use of registry data has been suggested as a more powerful tool in measuring outcomes for the generalist/non-specialist (Palan et al. 2016).

Our study highlights the difference between brands of implant that are assumed to function with the same design philosophy, although is limited by the fact that more detailed, individual brand-specific analysis was beyond the scope of this study.

Large registry studies are able to detect small differences in outcome, although the difference between statistical and clinical significance should be considered, as well as the potential effect of bias (Whitehouse et al. 2014). Although the differences between the groups were small in our study, a difference of 0.5% at 8 years may be clinically significant when attempting to maximise the effectiveness of this highly successful procedure, and highlights that not all stems of a similar philosophy behave in exactly the same manner.

Limitations of our study include the small number of data entries submitted to the NJR in the early years of its existence

218

and the fact that some centres had poor submission compliance data submission, potentially skewing results. Similarly, the revision rates may be higher than reported due to unreliable NJR compliance with data submission at revision surgery (Porter 2017). This is unlikely to skew the findings if the failure to report was equivalent across all stem designs. Residual confounding may also remain due to the limitations of data capture within the NJR (e.g., the use of the Charlson index for comorbidities would be preferential to ASA grade but is not part of the minimum dataset) or inclusion in the analysis (e.g., surgeon experience or Trust preference may dictate which implant is used).

In summary, this large registry review study showed a significant survival advantage of the most popular taper-slip design over all other groups of patients. Future research efforts should focus on brand/design comparison rather than comparing outcomes in different fixation philosophies as this provides more accurate data and results as demonstrated in this paper.

Even these comparisons may be skewed by other confounders relevant only at brand level (Junnila et al. 2016).

Supplementary data Tables 2–4 and 7 with descriptive statistics and results from the Cox regression analysis, and NJR disclaimer are available as supplementary data in the online version of this article, http://dx.doi.org/10.1080/17453674.2019.1582680

HAK: Drafting of the manuscript; data interpretation. SLW: Study conception, data analysis, data interpretation; methodology development; editing of the manuscript. JRH: Concept development, editing of the manuscript; data interpretation. AJT: Study conception, concept development; senior manuscript editor; data interpretation.

The authors would like to thank the patients and staff of all the hospitals in England, Wales and Northern Ireland who have contributed data to the National Joint Registry. They are grateful to the Healthcare Quality Improvement Partnership (HQIP), the NJR Research Sub-Committee, and staff at the NJR Centre for facilitating this work.

Acta thanks Rüdiger Weiss and Wierd P Zijlstra for help with peer review of this study.s

Alfaro-Adrian J, Gill H S, Murray D W. Should total hip arthroplasty femoral components be designed to subside? A radiostereometric analysis study of the Charnley Elite and Exeter stems. J Arthroplasty 2001; 16(5): 598-606.

Broden C, Mukka S, Muren O, Eisler T, Boden H, Stark A, Skoldenberg O. High risk of early periprosthetic fractures after primary hip arthroplasty in elderly patients using a cemented, tapered, polished stem. Acta Orthop 2015; 86(2): 169-74.

Carrington N C, Sierra R J, Gie G A, Hubble M J, Timperley A J, Howell J R. The Exeter Universal cemented femoral component at 15 to 17 years: an update on the first 325 hips. J Bone Joint Surg Br 2009; 91(6): 730-7.

Dall D M, Learmonth I D, Solomon M I, Miles A W, Davenport J M. Fracture and loosening of Charnley femoral stems: comparison between first-generation and subsequent designs. J Bone Joint Surg Br 1993; 75(2): 259-65.

Acta Orthopaedica 2019; 90 (3): 214–219

Gillam M H, Ryan P, Graves S E, Miller L N, de Steiger R N, Salter A. Competing risks survival analysis applied to data from the Australian Orthopaedic Association National Joint Replacement Registry. Acta Orthop 2010; 81(5): 548-55.

Hailer N P, Garellick G, Karrholm J. Uncemented and cemented primary total hip arthroplasty in the Swedish Hip Arthroplasty Register. Acta Orthop 2010; 81(1): 34-41.

Harris W H. Long-term results of cemented femoral stems with roughened precoated surfaces. Clin Orthop Relat Res 1998; (355): 137-43.

Hook S, Moulder E, Yates P J, Burston B J, Whitley E, Bannister G C. The Exeter Universal stem: a minimum ten-year review from an independent centre. J Bone Joint Surg Br 2006; 88(12): 1584-90.

Howell J R, Jr., Blunt L A, Doyle C, Hooper R M, Lee A J, Ling R S. In vivo surface wear mechanisms of femoral components of cemented total hip arthroplasties: the influence of wear mechanism on clinical outcome. J Arthroplasty 2004; 19(1): 88-101.

Huiskes R, Verdonschot N, Nivbrant B. Migration, stem shape, and surface finish in cemented total hip arthroplasty. Clin Orthop Relat Res 1998; (355): 103-12.

Jayasuriya R L, Buckley S C, Hamer A J, Kerry R M, Stockley I, Tomouk M W, Wilkinson J M. Effect of sliding-taper compared with composite-beam cemented femoral prosthesis loading regime on proximal femoral bone remodeling: a randomized clinical trial. J Bone Joint Surg Am 2013; 95(1): 19-27.

Junnila M, Laaksonen I, Eskelinen A, Pulkkinen P, Ivar Havelin L, Furnes O, Marie Fenstad A, Pedersen A B, Overgaard S, Karrholm J, Garellick G, Malchau H, Makela K T. Implant survival of the most common cemented total hip devices from the Nordic Arthroplasty Register Association database. Acta Orthop 2016; 87(6): 546-53.

Lachiewicz P F, Kelley S S, Soileau E S. Survival of polished compared with precoated roughened cemented femoral components: a prospective, randomized study. J Bone Joint Surg Am 2008; 90(7): 1457-63.

Lettin A W, Ware H S, Morris R W. Survivorship analysis and confidence intervals: an assessment with reference to the Stanmore total knee replacement. J Bone Joint Surg Br 1991; 73(5): 729-31.

Makela K, Eskelinen A, Pulkkinen P, Paavolainen P, Remes V. Cemented total hip replacement for primary osteoarthritis in patients aged 55 years or older: results of the 12 most common cemented implants followed for 25 years in the Finnish Arthroplasty Register. J Bone Joint Surg Br 2008; 90(12): 1562-9.

Palan J, Smith M C, Gregg P, Mellon S, Kulkarni A, Tucker K, Blom A W, Murray D W, Pandit H. The influence of cemented femoral stem choice on the incidence of revision for periprosthetic fracture after primary total hip arthroplasty: an analysis of National Joint Registry data. Bone Joint J 2016; 98-B(10): 1347-54.

Petheram T G, Whitehouse S L, Kazi H A, Hubble M J, Timperley A J, Wilson M J, Howell J R. The Exeter Universal cemented femoral stem at 20 to 25 years: a report of 382 hips. Bone Joint J 2016; 98-B(11): 1441-9.

Pocock S J, Clayton T C, Altman D G. Survival plots of time-to-event outcomes in clinical trials: good practice and pitfalls. Lancet 2002; 359(9318): 1686-9.

Porter M. National Joint Registry data quality audit. J Trauma Orthopaedics 2017; 5(3): 42.

Purbach B, Kay P R, Siney P D, Fleming P A, Wroblewski B M. The C-stem in clinical practice: fifteen-year follow-up of a triple tapered polished cemented stem. J Arthroplasty 2013; 28(8): 1367-71.

Sarvilinna R, Huhtala H, Pajamaki J. Young age and wedge stem design are risk factors for periprosthetic fracture after arthroplasty due to hip fracture: a case-control study. Acta Orthop 2005; 76(1): 56-60.

Sayers A, Evans J T, Whitehouse M R, Blom A W. Are competing risks models appropriate to describe implant failure? Acta Orthop 2018; 89(3): 256-8.

Schmalzried T P, Kwong L M, Jasty M, Sedlacek R C, Haire T C, O’Connor D O, Bragdon C R, Kabo J M, Malcolm A J, Harris W H. The mechanism of loosening of cemented acetabular components in total hip arthroplasty: analysis of specimens retrieved at autopsy. Clin Orthop Relat Res 1992; 274: 60-78.

Shah N, Porter M. Evolution of cemented stems. Orthopedics 2005; 28(8 Suppl.): s819-25.

Acta Orthopaedica 2019; 90 (3): 214â&#x20AC;&#x201C;219

Thien T M, Chatziagorou G, Garellick G, Furnes O, Havelin L I, Makela K, Overgaard S, Pedersen A, Eskelinen A, Pulkkinen P, Karrholm J. Periprosthetic femoral fracture within two years after total hip replacement: analysis of 437,629 operations in The Nordic Arthroplasty Register Association database. J Bone Joint Surg Am 2014; 96(19): e167.

Van Eynde E, Hendrickx M, Scheerlinck T. Uncemented femoral stem design influences the occurrence rate of postoperative fractures after primary hip arthroplasty: a comparison of the Image and Profile stems. Acta Orthop Belg 2010; 76(2): 189-98.

219

Verdonschot N, Huiskes R. Surface roughness of debonded straight-tapered stems in cemented THA reduces subsidence but not cement damage. Biomaterials 1998; 19(19): 1773-9.

Whitehouse S L, Bolland B J, Howell J R, Crawford R W, Timperley A J. Mortality following hip arthroplasty: inappropriate use of National Joint Registry (NJR) Data. J Arthroplasty 2014; 29(9): 1827-34.

Young L, Duckett S, Dunn A. The use of the cemented Exeter Universal femoral stem in a district general hospital: a minimum ten-year follow-up. J Bone Joint Surg Br 2009; 91(2): 170-5.

220

Acta Orthopaedica 2019; 90 (3): 220–225

Body mass index is associated with risk of reoperation and revision after primary total hip arthroplasty: a study of the Swedish Hip Arthroplasty Register including 83,146 patients Arkan S SAYED-NOOR 1, Sebastian MUKKA 1, Maziar MOHADDES 2,3, Johan KÄRRHOLM 2,3, and Ola ROLFSON 2,3 1 Department 3 Department

of Surgical and Perioperative Sciences, Umeå University, Umeå; 2 Swedish Hip Arthroplasty Register, Centre of Registers, Gothenburg; of Orthopaedics, Institute of Clinical Sciences, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden Correspondence: arkansam@yahoo.com Submitted 2018-11-05. Accepted 2019-02-01.

Background and purpose — The prevalence of obesity is on the rise, becoming a worldwide epidemic. The main purpose of this register-based observational study was to investigate whether different BMI classes are associated with increased risk of reoperation within 2 years, risk of revision within 5 years, and the risk of dying within 90 days after primary total hip arthroplasty (THA). We hypothesized that increasing BMI would increase these risks.

Patients and methods — We analyzed a cohort of 83,146 patients who had undergone an elective THA for primary osteoarthritis between 2008 and 2015 from the Swedish Hip Arthroplasty Register (SHAR). BMI was classified according to the World Health Organization (WHO) into 6 classes: < 18.5 as underweight, 18.5–24.9 as normal weight, 25–29.9 as overweight, 30–34.9 as class I obesity, 35–39.9 as class II obesity, and ≥ 40 as class III obesity.

Results — Both unadjusted and adjusted parameter estimates showed increasing risk of reoperation at 2 years and revision at 5 years with each overweight and obesity class, mainly due to increased risk of infection. Uncemented and reversed hybrid fixations and surgical approaches other than the posterior were all associated with increased risk. Obesity class III (≥ 40), male sex, and increasing ASA class were associated with increased 90-day mortality.

Interpretation — Increasing BMI was associated with 2-year reoperation and 5-year revision risks after primary THA where obese patients have a higher risk than overweight or normal weight patients. As infection seems to be the main cause, customizing preoperative optimization and prophylactic measures for obese patients may help reduce risk.

The prevalence of obesity is on the rise, becoming a worldwide epidemic. Currently, more than two-thirds of Americans are classified as obese (Yang and Colditz 2015). In obese patients, total hip arthroplasty (THA) can be challenging because the extensive adipose tissue can compromise optimal surgical technique, prolong operative time, and increase intraoperative bleeding and risk for postoperative complications (Bowditch and Villar 1999, Liu et al. 2015, Wooten and Curtin 2016, Krauss et al. 2018). The effect of BMI on functional outcome, quality of life, and complication rate following THA has been investigated in a number of studies (Vincent et al. 2012, Liu et al. 2015, Haynes et al. 2017, Barrett et al. 2018). As BMI increases, the functional improvement and quality of life after THA may deteriorate. Based on this presumed increased risk, the American Association of Hip and Knee Surgeons Workgroup released a statement recommending that arthroplasty operations in patients with a BMI > 40 be delayed, especially in the setting of other comorbid conditions (Workgroup of the American Association of Hip and Knee Surgeons Evidence Based Committee 2013). However, only a limited number of studies reporting perioperative complications have been based on population-based cohorts (Murgatroyd et al. 2014, Ward et al. 2015, Husted et al. 2016, Wagner et al. 2016, Jung et al.

2017, Werner et al. 2017, Zusmanovich et al. 2018, DeMik et al. 2018, Jeschke et al. 2018).

The main purpose of this register-based cohort study was to investigate whether under- or overweight, separated into BMI classes, is associated with increased risk of reoperation within 2 years, risk of revision within 5 years, and the risk of dying within 90 days after primary total hip arthroplasty. We hypothesized that increasing BMI would negatively influence the reoperation, revision, and mortality risks.

© 2019 The Author(s). Published by Taylor & Francis on behalf of the Nordic Orthopedic Federation. This is an Open Access article distributed under the terms of the Creative Commons Attribution-Non-Commercial License (https://creativecommons.org/licenses/by/4.0) DOI 10.1080/17453674.2019.1594015

Acta Orthopaedica 2019; 90 (3): 220–225

221

All THR 2008–2015 n = 127,663

Patients and methods Study design and setting The Swedish Hip Arthroplasty Register (SHAR) was launched in 1979 to prospectively monitor THAs performed in Sweden and to evaluate the performance of implants, fixation methods, and surgical techniques. The register covers all publicly and privately funded hospitals performing THA. The completeness of registration for primary THAs is between 97% and 99%. A unique patient identifier, the personal identity number, provides information on date of birth and sex. For each operation, participating hospitals record variables such as implant article number, type of fixation, and surgical approach. In 2008, information on American Association of Anesthesiologists’ physical status classification (ASA), weight, and height were added to the routine data collection.

We followed the STROBE guidelines (von Elm et al. 2014).

Patient selection Patients included in this observational study met the following criteria: primary osteoarthritis (International Classification of Diseases [ICD] M16.0 and M16.1) operated with THA between January 1, 2008 and December 31, 2015 using traditional (not resurfacing) implants with uncemented, cemented, hybrid, or reversed hybrid fixation. In patients with bilateral THA during the study period, we included only records concerning the first THA. Patients with missing documentation regarding BMI or ASA class were excluded. BMI was classified according to the World Health Organization (WHO) into 6 classes: < 18.5 as underweight, 18.5–24.9 as normal weight, 25–29.9 as overweight, 30–34.9 as class I obesity, 35.0–39.9 as class II obesity, and ≥ 40 as class III obesity.

Outcome measures Reoperation is defined as any kind of subsequent open surgical procedure related to the inserted arthroplasty, no matter whether the arthroplasty, or any of its parts, is replaced, extracted, or left untouched. Revision is defined as a subsequent procedure where at least 1 part of the prosthesis is exchanged, added to, or extracted. All revisions are also classified as reoperations, but not all reoperations are revisions.

The outcome measures of this study include: 1. Reoperations within the first 2 years from the index THA procedure, including all types of open surgical procedures to the hip and for any reason; 2. Revisions within the first 5 years from the index THA procedure. For first-time procedures, a revision in the SHAR is defined as exchange or removal of one or more implant component(s); 3. 90-day mortality. The mortality data are obtained by crossmatching data from SHAR with the Swedish Population Register, governed by the Swedish Tax Office.

Excluded (n = 44,517): – second hip, 14,853 – resurfacing, 1,010 – not OA, 23,140 – ASA and/or BMI missing, 5,514 Study group n = 83,146

Figure 1. Flowchart of patients through the study.

Causes of reoperation and revision were categorized into loosening/osteolysis, dislocation, infection, and other.

Confounders A priori, we decided to include the age, sex, ASA class, fixation method, and surgical approach as confounders. These variables have previously demonstrated association with both exposure and outcome and are not considered to be in the causal pathway between potential risk factors and outcome.

Statistics Survival estimates (with 95% confidence intervals [CI]) for not being reoperated within 2 years, not revised within 5 years, and being alive within 90 days were calculated using Kaplan–Meier survival analysis. The assumption of proportionality was checked graphically. Simple and multiple Cox regression analyses were applied to calculate unadjusted and adjusted hazard ratios (HR). We adjusted for age, sex, ASA class, fixation method, and surgical approach at primary surgery. R version 3.4.4 (https://www.r-project.org) was used to perform all analyses.

Ethics, funding, and potential conflicts of interests The study was conducted in accordance with the ethical principles of the Helsinki Declaration and was approved by the Regional Ethical Review Board in Gothenburg, Sweden (decision 271-14). There was no external funding for the project and no competing interest to declare.

Results 127,663 primary THAs, registered in SHAR between January 1, 2008 and December 31, 2015 were primarily included.

After exclusion of resurfacing arthroplasties, second hip THA, patients with secondary OA, and those with missing data, 83,146 patients (mean age 69 years, 57% females) remained for analysis (Figure 1, Table 1). The majority of patients were normal weight or overweight. Age at operation decreased and the ASA class increased with increasing

Acta Orthopaedica Orthopaedica 2019; 2019; 90 90 (3): (3): 220–225 Acta 220–225

222 222

Table 1. Demography per BMI class

Factor Number of patients Age, mean (SD) Female sex, n (%) ASA, n (%) I II III IV/V Fixation, n (%) All cemented All uncemented Hybrid Reversed hybrids Surgical approach, n (%) Posterior Direct lateral Other

Underweight

Normal weight

Overweight

Class I obesity

Class II obesity

Class III obesity

All patients

579 73 (11) 516 (89)

25,718 70 (10) 16,721 (65)

36,301 69 (10) 18,360 (51)

15,751 67 (9) 8,568 (54)

3,939 ( 65 (9) 2,455 (62)

858 64 (9) 590 (69)

83,146 69 (10) 47,210 (57)

125 (22) 330 (57) 116 (20) 8 (1)

7,998 (31) 14,519 (56) 3,120 (12) 81 (0.3)

9,147 (25) 22,356 (62) 4,675 (13) 123 (0.3)

2,499 (16) 10,120 (64) 3,067 (19) 65 (0.4)

272 (6.9) 2,224 (57) 1,416 (36) 27 (1)

54 (6) 391 (46) 398 (46) 15 (2)

20,095 (24) 49,940 (60) 12,792 (15) 319 (0.4)

464 (80) 46 (8) 16 (3) 53 (9)

18,146 (71) 3,839 (15) 562 (2) 3,171 (12)

24,342 (67) 6,342 (17) 676 (2) 4,941 (14)

10,359 (66) 2,890 (18) 281 (2) 2,221 (14)

2,532 (64) 775 (20) 65 (2) 567 (14)

539 (63) 186 (22) 20 (2) 113 (13)

56,382 (68) 14,078 (17) 1,620 (2) 11,066 (13)

273 (47) 253 (44) 53 (9)

13,044 (51) 10,859 (42) 1,813 (7)

19,049 (53) 15,046 (41) 2,205 (6)

8,496 (54) 6,435 (41) 818 (5)

2,119 (54) 1,643 (42) 177 (5)

477 (56) 353 (41) 28 (3)

43,458 (52) 34,589 (42) 5,094 (6)

Figure 2. Kaplan–Meier 2-year reoperation estimates by BMI class (including CIs).

Figure 3. Kaplan–Meier 5-year implant survival estimates by BMI class (including CIs). For color codes, see Figure 2.

weight and obesity class. The dominating fixation technique was cemented and a posterior approach was used in nearly half of the operations.

Table 2 (see Supplementary data) presents survival estimates at 2 years for reoperation, 5 years for revision, and 90 days for mortality among the 6 BMI classes.

Risk of reoperation within 2 years The probability of reoperation increased in overweight and obesity classes I–III (Figure 2). Both unadjusted and adjusted parameter estimates showed increasing risk of reoperation at 2 years with each overweight and obesity class, mainly due to increased risk of infection, whereas the HR for underweight was similar to the reference category normal weight (Tables 2 and 3, see Supplementary data). The 2-year risk of reopera-

Figure 4. Kaplan–Meier 90-day mortality estimates by BMI class. For color codes, see Figure 2.

tion was higher in men and increased with higher ASA class.

Uncemented fixation and reversed hybrid fixations, and other surgical approaches than the posterior were all associated with increased risk.

Risk of revision within 5 years The probability of not being revised was lower in BMI overweight and obesity classes I–III (Figure 3). Both unadjusted and adjusted parameter estimates showed increasing risk of revision at 5 years with each overweight and obesity class, mainly due to increased risk of infection, while the HR for underweight was similar to the reference category normal weight (Tables 2 and 3, see Supplementary data). The 5-year risk of revision was higher in men and increased with higher ASA class. Uncemented fixation and reversed hybrid fixa-

Acta Orthopaedica 2019; 90 (3): 220–225

tions, and other surgical approaches than the posterior were all associated with increased risk.

90-day mortality Underweight and obesity class III were associated with higher mortality compared with the other BMI classes (Figure 4).

However, HRs for BMI classes were not statistically significantly higher compared with normal weight (Table 3, see Supplementary data). In the multiple regression model, only obesity class III (≥ 40), male sex, and increasing ASA class were associated with increased risk.

Discussion The impact of bodyweight on the occurrence and progression of hip OA as well as on the early and late results of THA has been in focus during the last 2 decades. Several studies have shown that overweight may be associated with hip OA symptoms, motivating THA at early ages (Harms et al. 2007, Changulani et al. 2008, Gandhi et al. 2010). Furthermore, increased BMI has been linked to higher risks for perioperative complications such as bleeding, infection, and dislocation, even though there is still a debate over the validity of some of these results (Vincent et al. 2012, Liu et al. 2015, Haynes et al. 2017, Barrett et al. 2018). The concept of obesity paradox, i.e., the favourable and protective effect of obesity on some aspects of the outcome of THA, has been raised (Shaparin et al. 2016, Zhang et al. 2018).

By definition, obesity class I and II would yield an ASA class of II or above and obesity class III would yield ASA class III or above. Although there was a clear pattern with higher ASA class for the obesity levels, not all patients were classified as per the definition. This highlights the interrater variation in assessment of ASA class and possible local traditions among anesthesiologists in how the ASA classification system is applied (Sankar et al. 2014). Nevertheless, we believe this also reflects the result of an overall assessment of perioperative risk factors where some otherwise healthy obese patients are classified lower than as defined by the classification system.

In our study, patients with class III obesity were younger.

This concurs with other studies (Harms et al. 2007, Changulani et al. 2008, Gandhi et al. 2010). Changulani et al. (2008) studied the relationship between obesity and age among 1,025 THA patients and found that the morbidly obese were 10 years younger on average than those with a normal BMI. In their systematic review, Haynes et al. (2017) also found that obesity was associated with younger age at time of primary THA. These findings agree with a review of registry data from the Mayo Clinic (Singh and Lewallen 2014), which showed a decrease in the mean age of patients undergoing primary THA by 0.7 years. This was inversely associated with an increase of 1.6 in the BMI of primary THA patients over the same study

223

period (1993–2005). A possible explanation for this association might be the increased pain sensitivity, high-level forces/ wear on the joint surface, and the lower physical activity in morbidly obese patients.

We found an association of BMI class with increasing risk of reoperation within 2 years and revision within 5 years, mainly due to increased risk of infection (Table 2). This concurs with a recent report using German nationwide billing data for inpatient hospital treatment covering more than 130,000 THAs. In this report, Jeschke et al. (2018) found increased overall postoperative complication and 1-year revision rates with higher BMI class. Similar to our results, they found that 90-day mortality increased only in class III obesity patients.

Other studies have demonstrated increased postoperative infection and dislocation rates both after primary and revision THA with increasing BMI (Vincent et al. 2012, Pulos et al.

2014, Houdek et al. 2015, Liu et al. 2015, Haynes et al. 2017, Barrett et al. 2018, Kennedy et al. 2018). Contrary to the above-mentioned results, some reports showed comparable postoperative complication rates across BMI classes (Davis et al. 2011, McCalden et al. 2011, Watts et al. 2016). This divergence requires further attention and analysis. Increased BMI is associated or has a causal relationship with medical comorbidities such as diabetes mellitus, and cardiovascular disorders, as well as antibiotic resistance. Moreover, THA in obese patients may be more surgically demanding as the voluminous deep adipose tissue, weak fatty-infiltrated periarticular soft-tissue envelope, and obscured anatomical landmarks may result in suboptimal positioning of THA components, prolonged operative time, and wound problems (Elson et al. 2013, Hanly et al. 2016). Higher weight increases load and forces on THA components, which potentially increases the risk of wear and implant loosening. However, a sedentary lifestyle might counteract this risk for wear and aseptic loosening. The above-mentioned parameters may, at least partly, explain the negative impact of increased BMI on 2-year reoperation and 5-year revision outcomes. Interestingly, we found comparable risks for reoperation within 2 years and revision within 5 years due to mechanical complications (loosening and dislocation) among the BMI classes (Table 2, see Supplementary data). Patients with increased BMI may also have some positive aspects such as adequate nutrition, careful preoperative preparation and surgical technique usually performed by more experienced surgeons, more active postoperative medical care, and physical rehabilitation and follow-up. These aspects can be protective to some extent, but apparently not for patients with a very high BMI such as class III obesity.

This study has some limitations. SHAR does not capture all reoperations. The completeness of registrations of revisions is higher than that of other reoperations where components are not removed, exchanged, or added. Also, there are missing data in the reporting of weight and height. There is no reason to suspect a systematic underreporting of BMI or reop-

224

erations based on BMI. However, the most common cause for a reoperation without revising implants is periprosthetic infection. Given that high BMI is a risk factor for infection, the underreporting may distribute differently between BMI classes. Hence, the higher risk of reoperation associated with increasing BMI class may be underestimated. Despite the comprehensive set of variables included in SHAR, parameters such as smoking, type of comorbidities, nutritional status, OA severity, surgical complexity, and surgeon experience were not available. Therefore, as with most register-based studies, residual confounding likely exists. Also, we did not correct for multiple testing; however, confounders were selected a priori and based on previous established relationships. The methods for measuring weight and height are heterogeneous and include estimates by health care professionals, patientreported values, and actual measurements at the preoperative assessment. Another limitation pertains to the use of BMI as a surrogate measure for excess fat although it does not distinguish between the distributions of fat, muscles, and bone mass. On average, women have greater amounts of total body fat than men with an equivalent BMI, while muscular highly trained athletes may have a high BMI because of increased muscle mass. Furthermore, there are numerous factors related to genetics, and the physical and social environment including comorbidities that will influence the body mass index. Thus, BMI could be viewed as a proxy for known and unknown factors related to the health status of the patient, but is as such attractive to use because it can be easily measured. Multiple comparisons among the BMI classes is another limitation. These limitations are counterbalanced by the strength of study: a nationwide large study group using a register with high completeness and validity. The inclusion of not only revisions as an endpoint but also any reoperation adds to the strength of the study.

In summary, BMI classes were associated with reoperation and revision risks after primary THA, where morbidly obese patients have more than a doubled risk than obese or normal weight patients. As infection seems to be the main cause, customizing preoperative optimization and prophylactic measures for obese patients may help reduce risk. Furthermore, many clinical aspects could be addressed such as adequate antibiotic prophylaxis, e.g., weight-adjusted as well as the use of incisional vacuum-assisted closure in overweight patients. Although BMI is a well-established risk factor for complications following THA, this is the first study focusing on BMI and outcomes in a Swedish context. This will help inform surgeons and their patients on risks related to BMI classes.

Supplementary data Tables 2–3 are available as supplementary data in the online version of this article, http://dx.doi.org/10.1080/17453674.

2019.1594015

Acta Orthopaedica 2019; 90 (3): 220–225

ASN and OR conceived the study and defined the analysis plan with input from all other co-authors. ASN, OR, and SM drafted the manuscript. All authors interpreted results and reviewed the manuscript.

The authors would like to thank all orthopedic surgeons and administrative personnel at Swedish hospitals for their contribution of data and engagement in the register. They also thank registry coordinator Pär Werner who performed statistical analyses.

Acta thanks Jeppe Lange and Mogens Laursen for help with peer review of this study.

Barrett M, Prasad A, Boyce L, Dawson-Bowling S, Achan P, Millington S, Hanna S A. Total hip arthroplasty outcomes in morbidly obese patients: a systematic review. EFORT Open Rev 2018; 3(9): 507-12.

Bowditch M G, Villar R N. Do obese patients bleed more? A prospective study of blood loss at total hip replacement. Ann R Coll Surg Engl 1999; 81(3): 198-200.

Changulani M, Kalairajah Y, Peel T, Field R E. The relationship between obesity and the age at which hip and knee replacement is undertaken. J Bone Joint Surg Br 2008; 90-B: 360–3.

Davis A M, Wood A M, Keenan A C, Brenkel I J, Ballantyne J A. Does body mass index affect clinical outcome post-operatively and at five years after primary unilateral total hip replacement performed for osteoarthritis? A multivariate analysis of prospective data. J Bone Joint Surg Br 2011; 93(9): 1178-82.

DeMik D E, Bedard N A, Dowdle S B, Elkins J M, Brown T S, Gao Y, Callaghan J J. Complications and obesity in arthroplasty: a hip is not a knee. J Arthroplasty 2018; 33(10): 3281-7.

Elson L C, Barr C J, Chandran S E, Hansen V J, Malchau H, Kwon Y M. Are morbidly obese patients undergoing total hip arthroplasty at an increased risk for component malpositioning? J Arthroplasty 2013; 28(8 Suppl.): 41-4.

Gandhi R, Wasserstein D, Razak F, Davey J R , Mahomed N N. BMI independently predicts younger age at hip and knee replacement. Obesity (Silver Springs) 2010; 18(12): 2362-6.

Hanly R J, Marvi S K, Whitehouse S L, Crawford R W. Morbid obesity in total hip arthroplasty: redefining outcomes for operative time, length of stay, and readmission. J Arthroplasty 2016; 31(9): 1949-53.

Harms S, Larson R, Sahmoun A E, Beal J R. Obesity increases the likelihood of total joint replacement surgery among younger adults. Int Orthop 2007; 31(1): 23-6.

Haynes J, Nam D, Barrack R L. Review: Obesity in total hip arthroplasty: does it make a difference? Bone Joint J 2017; 99-B (1 Suppl. A): 31-6.

Houdek M T, Wagner E R, Watts C D, Osmon D R, Hanssen A D, Lewallen D G, Mabry T M. Morbid obesity: a significant risk factor for failure of two-stage revision total hip arthroplasty for infection. J Bone Joint Surg Am 2015; 97(4): 326-32.

Husted H, Jørgensen C C, Gromov K, Kehlet H, Lundbeck Foundation Center for Fast-track Hip and Knee Replacement Collaborative Group. Does BMI influence hospital stay and morbidity after fast-track hip and knee arthroplasty? Acta Orthop 2016; 87(5): 466-72.

Jeschke E, Citak M, Günster C, Halder A M, Heller K D, Malzahn J, Niethard F U, Schräder P, Zacher J, Gehrke T. Obesity increases the risk of postoperative complications and revision rates following primary total hip arthroplasty: an analysis of 131,576 total hip arthroplasty cases. J Arthroplasty 2018; 33(7): 2287-92.

Jung P, Morris A J, Zhu M, Roberts S A, Frampton C, Young S W. BMI is a key risk factor for early periprosthetic joint infection following total hip and knee arthroplasty. N Z Med J 2017; 130(1461): 24-34.

Kennedy J W, Young D, Meek D R M, Patil S R. Obesity is associated with higher complication rates in revision total hip arthroplasty. J Orthop 2018; 30: 15(1): 70-2.

Acta Orthopaedica 2019; 90 (3): 220–225

Krauss E S, Cronin M, Dengler N, Simonson B G, Altner K, Daly M, Segal A. Semin Thromb Hemost 2018 Dec 19. [Epub ahead of print].

Liu W, Wahafu T, Cheng M, Cheng T, Zhang Y, Zhang X. Review: The influence of obesity on primary total hip arthroplasty outcomes: a meta-analysis of prospective cohort studies. Orthop Traumatol Surg Res 2015; 101(3): 289-96.

McCalden R W, Charron K D, MacDonald S J, Bourne R B, Naudie D D. Does morbid obesity affect the outcome of total hip replacement? An analysis of 3290 THRs. J Bone Joint Surg Br 2011; 93(3): 321-5.

Murgatroyd S E, Frampton C M, Wright M S. The effect of body mass index on outcome in total hip arthroplasty: early analysis from the New Zealand Joint Registry. J Arthroplasty 2014; 29(10): 1884-8.

Pulos N, McGraw M H, Courtney P M, Lee G C. Revision THA in obese patients is associated with high re-operation rates at short-term follow-up.

J Arthroplasty 2014; 29(9 Suppl.): 209-13.

Sankar A, Johnson SR, Beattie W S, Tait G, Wijeysundera D N. Reliability of the American Society of Anesthesiologists physical status scale in clinical practice. Br J Anaesth 2014; 113(3): 424-32.

Shaparin N, Widyn J, Nair S, Kho I, Geller D, Delphin E. Does the obesity paradox apply to early postoperative complications after hip surgery? A retrospective chart review. J Clin Anesth 2016; 32: 84-91.

Singh J A, Lewallen D G. Increasing obesity and comorbidity in patients undergoing primary total hip arthroplasty in the US: a 13-year study of time trends. BMC Musculoskelet Disord 2014; 15: 441.

Vincent H K, Horodyski M, Gearen P, Vlasak R, Seay A N, Conrad B P, Vincent K R. Review: Obesity and long term functional outcomes following elective total hip replacement. J Orthop Surg Res 2012; 7: 16. von Elm E, Altman D G, Egger M, Pocock S J, Gøtzsche P C, Vandenbroucke J P; STROBE Initiative. The Strengthening the Reporting of Observational

225

Studies in Epidemiology (STROBE) Statement: guidelines for reporting observational studies. Int J Surg 2014; 12(12): 1495-9.

Wagner E R, Kamath A F, Fruth K M, Harmsen W S, Berry D J. Effect of body mass index on complications and reoperations after total hip arthroplasty. J Bone Joint Surg Am 2016; 98(3): 169-79.

Ward D T, Metz L N, Horst P K, Kim H T, Kuo A C. Complications of morbid obesity in total joint arthroplasty: risk stratification based on BMI. J Arthroplasty 2015; 30(9 Suppl.): 42-6.

Watts C D, Houdek M T, Wagner E R, Lewallen D G, Mabry T M. Morbidly obese vs nonobese aseptic revision total hip arthroplasty: surprisingly similar outcomes. J Arthroplasty 2016; 31(4): 842-5.

Werner B C, Higgins M D, Pehlivan H C, Carothers J T, Browne J A. Super obesity is an independent risk factor for complications after primary total hip arthroplasty. J Arthroplasty 2017; 32(2): 402-6.

Wooten C, Curtin B. Morbid obesity and total joint replacement: is it okay to say no? Orthopedics 2016; 39(4): 207-9.

Work Group of the American Association of Hip and Knee Surgeons Evidence Based Committee. Review: Obesity and total joint arthroplasty: a literature based review. J Arthroplasty 2013; 28(5): 714-21.

Yang L, Colditz G A. Prevalence of overweight and obesity in the United States, 2007–2012. JAMA Intern Med 2015; 175(8): 1412-3.

Zhang J C, Matelski J, Gandhi R, Jackson T, Urbach D, Cram P. Can patient selection explain the obesity paradox in orthopaedic hip surgery?

An analysis of the ACS-NSQIP Registry. Clin Orthop Relat Res 2018; 476(5): 964-73.

Zusmanovich M, Kester B S, Schwarzkopf R. Postoperative complications of total joint arthroplasty in obese patients stratified by BMI. J Arthroplasty 2018; 33(3): 856-64.

226

Acta Orthopaedica 2019; 90 (3): 226–230

Association between patient survival following reoperation after total hip replacement and the reason for reoperation: an analysis of 9,926 patients in the Swedish Hip Arthroplasty Register Peter CNUDDE 1,2,3, Erik BÜLOW 1,2, Szilard NEMES 2, Yosef TYSON 1,4, Maziar MOHADDES 1,2, and Ola ROLFSON 1,2 1 Swedish Hip Arthroplasty Register, Gothenburg, Sweden; 2 Department of Orthopaedics, Institute of Clinical Sciences, Sahlgrenska Academy, University of Gothenburg, Sweden; 3 Department of Orthopaedics, Hywel Dda University Healthboard, Prince Philip Hospital, Llanelli, UK; 4 Section of Orthopaedic Surgery, Department of Surgical Sciences, Uppsala University Hospital, Uppsala, Sweden Correspondence: peter.cnudde@icloud.com Submitted 2018-11-19. Accepted 2019-02-12.

Background and purpose — The association between long-term patient survival and elective primary total hip replacement (THR) has been described extensively. The long-term survival following reoperation of THR is less well understood. We investigated the relative survival of patients undergoing reoperation following elective THR and explored an association between the indication for the reoperation and relative survival.

Patients and methods — In this observational cohort study we selected the patients who received an elective primary THR and subsequent reoperations during 1999–2017 as recorded in the Swedish Hip Arthroplasty Register. The selected cohort was followed until the end of the study period, censoring or death. The indications for 1st- and eventual 2nd-time reoperations were analyzed and the relative survival ratio of the observed survival and the expected survival was determined.

Results — There were 9,926 1st-time reoperations and of these 2,558 underwent further reoperations. At 5 years after the latest reoperation, relative survival following 1sttime reoperations was 0.94% (95% CI 0.93–0.96) and 0.90% (CI 0.87–0.92) following 2nd-time reoperations. At 5 years patients with a 1st-time reoperation for aseptic loosening had higher survival than expected; however, reoperations performed for periprosthetic fracture, dislocation, and infection had lower survival.

Interpretation — The relative survival following 1stand 2nd-time reoperations in elective THR patients differs by reason for reoperation. The impact of reoperation on life expectancy is more obvious for infection/dislocation and periprosthetic fracture.

While the risk of dying and life expectancy following a primary THR has been studied extensively, patient survival after further surgical interventions is virtually unexplored (Jones et al. 2018, Yao et al. 2018). What happens to life expectancy if patients undergo reoperation and does the clinical indication for the reoperation influence life expectancy? So far, little is known about death following reoperation or revision after THR. The increasing age at the time of reoperation, the increased complexity of the surgery, and the timing of surgery might influence life expectancy. The relative survival method has been developed to provide better insights into the relation between a study population and a general population (Stare et al. 2005).

The Swedish Hip Arthroplasty Register (SHAR), a reliable source of information on longitudinal outcome (Kärrholm 2010, Cnudde et al. 2016) combined with national aggregated data from Statistics Sweden provide a platform for this study, which investigates the relative survival of patients undergoing reoperation following elective THR and the influence of the indication for the reoperation on the relative survival. Additionally, it investigates time- and indication-dependent patterns for 1st- and 2nd-time reoperations.

Patients and methods Data sources For this study, we used prospectively collected data on all patients who underwent a 1st-time reoperation following elective THR in 1999–2017 as recorded in SHAR. Patients who had their primary THR before 1999 were excluded. All surgical- and patient-related variables could be accessed and

© 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group, on behalf of the Nordic Orthopedic Federation. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits ­unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

DOI 10.1080/17453674.2019.1597062

Acta Orthopaedica 2019; 90 (3): 226–230

227

Table 1. Demographics of the study population. Values are mean years (SD) unless otherwise stated

n Age at primary THR Time to 1st reoperation Time from 1st to 2nd reoperation Age at 1st reoperation Age at 2nd reoperation Women, n (%) Dead, n (%)

Aseptic Periprosthetic loosening Dislocation fracture

Infection

Other

Unknown

3,558 1,782 1,574 2,065 877 60 61 (11) 69 (11) 70 (11) 68 (12) 60 (13) 60 (16) 8.0 (4.4) 3.2 (4.0) 5.2 (4.4) 1.6 (3.1) 4.1 (4.0) 6.4 (5.3) 2.0 (2.6) 1.7 (2.9) 1.6 (2.4) 0.6 (1.5) 1.5 (2.2) 1.4 (3.1) 70 (11) 73 (11) 76 (12) 70 (11) 65 (12) 67 (16) 69 (11) 73 (11) 73 (13) 69 (11) 67 (13) 69 (14) 1,790 (50) 1,045 (59) 884 (56) 869 (42) 478 (55) 40 (57) 566 (16) 818 (46) 711 (45) 645 (31) 136 (16) 10 (14)

p-value < 0.001 for all comparisons.

analysed from the database. However, for this analysis we concentrated solely on age, sex, and indication for surgery at the time of primary THR and age at the time of and indications for further operations. A reoperation is defined as any further surgery to the hip regardless of whether implant components are exchanged, removed, added or not, whereas a revision is defined as a reoperation where implant components are exchanged, removed, and/or added. Cause for reoperation was categorized into aseptic loosening, dislocation, periprosthetic fracture, infection, and other causes. Closed reductions, aspirations and isolated tissue biopsies are not included in this definition.

The selected cohort was followed until the end of the study period (December 31, 2017), censoring, or death.

Statistics Continuous variables were summarized as means (SD), categorical variables as percentages. Subsequently, we summarized and illustrated survival with the help of relative survival curves with 95% confidence intervals (CI) (Pohar Perme et al.

2012). We used R version 3.5 for statistical analyses (R Core Team (2018), R: A language and environment for statistical computing, Vienna, Austria, https://www.R-project.org) with the “relsurv” package for statistical analysis and applied the Pohar Perme method for calculating the relative survival.

Relative survival The relative survival was based on comparison of patients who underwent reoperation with aggregated data at national level from the general population (http://www.mortality.org).

For any given time point onward, we estimated the relative survival ratio based on the observed survival in the patient group divided by the observed survival in general population matched on age, sex, and year of birth. The observed survival of the general population was extracted from publicly available mortality tables tabulated for birth year and sex. The formula has previously been published (Cnudde et al. 2018a). A relative survival of 1 indicates that the exposure of interest, here the reoperation or the condition causing it, does not affect the survival in any measurable way. It does not mean that all

patients survive. A relative survival of less than 1 indicates excess hazard for the patients, while values above 1 indicate better survival than expected.

Ethics, funding, and potential conflicts of interests This study is part of a research project with the overall aim to perform a multidimensional longitudinal outcomes assessment following total hip replacement. Ethical review approval was obtained on April 7, 2014 from the Regional Ethical Review Board in Gothenburg, Sweden (entry number 271-14). The study was in part financed by grants from the Swedish state under the agreement between the Swedish government and the county councils, the ALF agreement (ALFGBG-522591). No competing interests declared.

Results Using the SHAR databases, 278,309 primary THRs were identified in the study period from January 1, 1999 to December 31, 2017. There were 9,926 1st-time reoperations, of which 7,581 were 1st-time revision procedures. Of these 2,558 underwent further reoperations, of which 1,541 were subsequent revisions. There were patients undergoing subsequent procedures (up to 19 recorded reoperations and 8 revisions).

Patients’ demographics and indications for reoperations are presented in Table 1. Patients who underwent reoperation for infection and aseptic loosening were generally younger at the time of their surgery than patients undergoing the procedure for dislocation and periprosthetic fracture. The indications for reoperations varied according to sex, with more females undergoing reoperations for periprosthetic fractures and dislocations whereas more males had reoperations for infection.

The median follow-up time was 8.4 years (0–18) from primary surgery.

Patients undergoing 1st-time reoperation had a lower survival rate compared with the general population for the whole study period. The relative survival was 98% (CI 98–99) at 1 year, 94% (CI 93–96) at 5 years, 80% (CI 75–86) at 10 years, and 61% (CI 50–74) at 15 years. Relative survival was worse

228

Acta Orthopaedica 2019; 90 (3): 226–230

Indication for reoperation Aseptic loosening

Periprosthestic fracture

Other

Dislocation

primary to first reoperation first to second reoperation

Infection 0

1

2

3

4

5

6

7

8

Median years to reoperation

Figure 1. Relative survival after 1st-time reoperation per indication at the time of the reoperation (truncated at 10 years).

Figure 2. Relative survival (and confidence intervals) after the 2nd-time reoperation per indication at the time of the reoperation (truncated at 10 years).

for 2nd-time reoperations compared with 1st-time reoperations with 97% (CI 96–98) at 1 year, 90% (CI 87–92) at 5 years, 69% (CI 56–85) at 10 years, and 49% (CI 34–71) at 15 years.

We stratified the relative survival per indication for reoperation (Figure 1). 1st-time reoperations for aseptic loosening had similar survival to the general population. In fact, the relative survival ratio implied a 4% increased survival compared with the general population at 5 years. Relative survival following reoperations performed for periprosthetic fracture was worse compared with aseptic loosening or other causes up to 15 years. Up to 5 years, relative survival following 1st-time reoperations for dislocation and infection was worse than for aseptic loosening and other causes (Table 2, see Supplementary data).

The relative survival following a 2nd-time reoperation (Figure 2) was only marginally better if the reoperation was performed for aseptic loosening within the 1st year (Table 3, see Supplementary data). Further reoperations for infection, periprosthetic fracture, and dislocation had lower survival with the lowest survival in cases of re-reoperation for dislocation and periprosthetic fracture. The number of patients at risk is below 100 at 10 years in the group of the re-reoperations with the exception of infections and other (non-further specified) indications.

There was a difference in time between primary THR and reoperation depending on the indication for the surgery (at the time of the reoperation), with a shorter time interval in the case of infection, dislocation, and periprosthetic fracture in the 1st postoperative year. A later and a more gradual increase in cases of loosening could be seen at later stages postoperatively. Further reoperations peaked very early after the 1st reintervention with the shortest interval when the 1st reoperation was performed for infection (Figure 3).

We also studied the influence of the clinical diagnosis at the time of primary THR on further reoperations (Table 4, see

Figure 3. Median time between the primary THR and the 1st-time reoperation and between 1st- and 2ndtime reoperation for the different indications.

Supplementary data). Aseptic loosening was the most frequent indication at the time of reoperation when the primary THR was performed for almost all clinical indications except when a THR was performed as a result of trauma complications.

Infections and dislocations frequently led to further infections for the same reasons (Table 5, see Supplementary data). Many treatment-resistant infections end up leading to an excision arthroplasty of the THR (classified as other procedure within the Tables and Figures).

Discussion We found that the survival following 1st- and 2nd-time repeat surgery in elective THR is influenced by the reason for the reoperation. Patients undergoing reoperations for aseptic loosening had a survival that is better compared with the survival of the general population up to 5 years following the reoperation. Reoperations for dislocation and periprosthetic fractures were associated with a worse survival compared with the survival of the general population, also visible in the case of a further reoperation.

THR can fail for a variety of reasons in isolation or through a combination of factors. It is well known that the patientreported outcomes after revision are worse compared with those of a primary procedure (Lubbeke et al. 2007, Postler et al. 2017). It is also likely that activity levels, as well as a potential to return to work, is affected by the revision THR (Scott et al. 2018). Nevertheless, the crude indicators of success have been re-revision/reoperation and/or mortality following the surgical procedure. The risk of further reoperations after a reoperation has been alluded to in the annual reports of the SHAR (https://shpr.registercentrum.se) and a 20% risk of a subsequent reoperation has been described. In the majority of cases a 1st reoperation occurs early within the 1st couple of

Acta Orthopaedica 2019; 90 (3): 226â&#x20AC;&#x201C;230

years with a shortening of time interval when further reoperations are needed.

Mortality can be calculated in different ways and the effect of the procedure on life expectancy will be influenced by a multitude of factors. Short-term mortality has been described

by Jones et al. (2018) based on a comparison of patients who underwent revision surgery and were compared with patients awaiting revision surgery. They found a higher mortality rate in patients who underwent revision surgery. This is likely attributed to the effect of the surgery. Yao et al. (2018) described long-term mortality following revision THR using the Mayo database and found a slightly higher mortality than in the general population. Our findings differ from the Mayo findings as we have an improved relative survival until 5 years postoperatively and from 5 years onward a survival that does not differ from the general population. The survival in patients reoperated for infections, periprosthetic fractures, and dislocations are, however, all worse than in the general population.

The Mayo group combined aseptic loosening, bearing wear, and dislocation in a single group. Whilst we were not able to look at the influence of comorbidity, we did study the effect of further reoperations.

Relative survival methods have been used to describe the cancer prognosis at population level. The overall relative survival 5 years after a 1st-time reoperation is comparable with the 5-year relative survival of melanomas (94% and 95%) in females in Australia and Sweden. Our 5-year relative survival after 2nd-time reoperation is slightly worse compared with the Australian and Swedish 5-year relative survival after diagnosis of breast cancer in women (91% and 92%). The overall 10-year relative survival for all cancers in Australia and Sweden is 61% and 69%. Our 10-year relative survival after reoperation is worse than the 10-year relative survival for breast cancer in Australia and Sweden (85% and 86%) and comparable to the relative survival after cervical cancer (70% and 76%) (https://ncci.canceraustralia.gov.au/outcomes/relative-survival-rate/10-year-relative-survival and https://www. cancerfonden.se/cancer-i-siffror). These comparisons illustrate the severity of suffering THR complications leading to reoperation.

Despite the ongoing improvement in outcomes of THR, some issues remain. There is a gradual increase in patients undergoing reoperations for infection and dislocation (Cnudde et al. 2018b). These predominantly occur in the early postoperative stages and are associated with an increased mortality.

There is also an accepted knowledge that revision THRs do worse than primary THRs (Espehaug et al. 1998, Lie et al.

2004). Therefore the right implant choice at the time of the primary procedure (Thien et al. 2014), management strategies at the time of the primary surgery, and subsequent surgeries will have to be developed and followed in an attempt to get the best outcomes and to avoid complications such as dislocation, infection, and periprosthetic fractures. Such complications not only put a burden on the patient but also have an impact on the

229

scarce health care resources, and the question will have to be asked which departments and surgeons will be best placed to organise and provide the treatment and ensure the best possible outcomes.

The risk of subsequent re-reoperations is quite prominent in the case of infection and dislocation, with further surgery for the same reasons being the most common indication if surgery takes place. This reiterates the complexity of this problem and the need to develop management pathways for the prevention and adequate treatment of infections and dislocations. Bigger head size and dual mobility cups have been advocated in the case of dislocation (Mohaddes et al.

2017). New pathways advocating a timely diagnosis, rigorous debridement, local delivery of antibiotics, and prolonged use of systemic antibiotics are now considered the mainstay of treatment protocols in the case of periprosthetic joint infections.

Strengths and limitations This study is based on data from a validated national quality register with linkage to the administrative databases.We deliberately selected the period 1999â&#x20AC;&#x201C;2017 as 1999 coincides with the year when more details of the implants used became available within the SHAR, and represents the time when increased use of more contemporary implants and head sizes was noticed, reflecting current practice. We acknowledge that selecting only reoperations following primary surgeries performed during this period does not reflect the actual proportion of different reasons for reoperations undertaken in Sweden; the distribution is skewed towards early and midterm complications.

In our analysis there was no matching for comorbidity and socioeconomic status. A previous study of our research group has described the effect of both comorbidity and socioeconomic status on both mortality and revision after THR (Cnudde et al. 2018c). We are aware that the indication at the time of primary surgery probably influences the reason for subsequent revisions with those who have a THR for complications after trauma or acute fracture having a higher risk of dislocation, periprosthetic fracture, and infection (Cnudde et al. 2018c).

Whether the risk of earlier death is linked to the reoperation per se or to the increased risk of reoperation in frail patients is something that will need further study.

In summary the impact of reoperation on life expectancy is obvious for infection/dislocation and periprosthetic fracture.

Treatment strategies should be developed to prevent these complications occurring and surgical pathways for the reoperations should be developed in order to improve outcomes for patients.

Supplementary data Tables 2â&#x20AC;&#x201C;5 are available as supplementary data in the online version of this article, http://dx.doi.org/10.1080/17453674.

2019.1597062

230

PC, SN, and OR conceived and planned the study. EB and SN processed the data and performed the statistical analysis. PC prepared the 1st draft of the manuscript. All authors cooperated in the analysis and interpretation of the data and the writing of the manuscript.

The authors would like to thank the orthopedic teams at the different centres in Sweden as well as the register coordinators at the Registercentrum Västra Götaland for their ongoing support with the data provision and collection.

Acta thanks Ove Furnes and Per Kjærsgaard-Andersenfor help with peer review of this study.

Cnudde P, Rolfson O, Nemes S, Karrholm J, Rehnberg C, Rogmark C, Timperley J, Garellick G. Linking Swedish health data registers to establish a research database and a shared decision-making tool in hip replacement.

BMC Musculoskelet Disord 2016; 17(1): 414. doi: 10.1186/s12891-0161262-x.

Cnudde P, Rolfson O, Timperley A J, Garland A, Karrholm J, Garellick G, Nemes S. Do patients live longer after THA and is the relative survival diagnosis-specific? Clin Orthop Relat Res 2018a; 476(6): 1166-75. doi: 10.1007/s11999.0000000000000097.

Cnudde P, Nemes S, Bulow E, Timperley J, Malchau H, Karrholm J, Garellick G, Rolfson O. Trends in hip replacements between 1999 and 2012 in Sweden. J Orthop Res 2018b; 36(1): 432-42. doi: 10.1002/ jor.23711.

Cnudde P H J, Nemes S, Bulow E, Timperley A J, Whitehouse S L, Karrholm J, Rolfson O. Risk of further surgery on the same or opposite side and mortality after primary total hip arthroplasty: a multi-state analysis of 133,654 patients from the Swedish Hip Arthroplasty Register. Acta Orthop 2018c; 89(4): 386-93. doi: 10.1080/17453674.2018.1475179.

Espehaug B, Havelin L I, Engesaeter L B, Langeland N, Vollset S E. Patient satisfaction and function after primary and revision total hip replacement.

Clin Orthop Relat Res 1998; (351): 135-48.

Acta Orthopaedica 2019; 90 (3): 226–230

Jones M D, Parry M, Whitehouse M R, Blom A W. Early death following revision total hip arthroplasty. Hip Int 2018; 28(4): 400-6. doi: 10.5301/ hipint.5000593.

Kärrholm J. The Swedish Hip Arthroplasty Register (http://www.shpr.se).

Acta Orthop 2010; 81(1): 3-4. doi: 10.3109/17453671003635918.

Lie S A, Havelin L I, Furnes O N, Engesaeter L B, Vollset S E. Failure rates for 4762 revision total hip arthroplasties in the Norwegian Arthroplasty Register. J Bone Joint Surg Br 2004; 86(4): 504-9.

Lubbeke A, Katz J N, Perneger T V, Hoffmeyer P. Primary and revision hip arthroplasty: 5-year outcomes and influence of age and comorbidity. J Rheumatol 2007; 34(2): 394-400.

Mohaddes M, Cnudde P, Rolfson O, Wall A, Karrholm J. Use of dual-mobility cup in revision hip arthroplasty reduces the risk for further dislocation: analysis of seven hundred and ninety one first-time revisions performed due to dislocation, reported to the Swedish Hip Arthroplasty Register. Int Orthop 2017; 41(3): 583-8. doi: 10.1007/s00264-016-3381-2.

Pohar Perme M, Stare J, Esteve J. On estimation in relative survival. Biometrics 2012; 68(1): 113-20. doi: 10.1111/j.1541-0420.2011.01640.x.

Postler A E, Beyer F, Wegner T, Lutzner J, Hartmann A, Ojodu I, Gunther K P. Patient-reported outcomes after revision surgery compared to primary total hip arthroplasty. Hip Int 2017; 27(2): 180-6. doi: 10.5301/hipint.5000436.

Scott C E H, Turnbull G S, Powell-Bowns M F R, MacDonald D J, Breusch S J. Activity levels and return to work after revision total hip and knee arthroplasty in patients under 65 years of age. Bone Joint J 2018; 100-B(8): 1043-53. doi: 10.1302/0301-620X.100B8.BJJ-2017-1557.R2.

Stare J, Henderson R, Pohar M. An individual measure of relative survival. J Roy Stat Soc Series C (Applied Statistics) 2005; 54(1): 115-26.

Thien T M, Chatziagorou G, Garellick G, Furnes O, Havelin L I, Makela K, Overgaard S, Pedersen A, Eskelinen A, Pulkkinen P, Karrholm J. Periprosthetic femoral fracture within two years after total hip replacement: analysis of 437,629 operations in the Nordic Arthroplasty Register Association database. J Bone Joint Surg Am 2014; 96(19): e167. doi: 10.2106/jbjs.m.00643.

Yao J J, Maradit Kremers H, Abdel M P, Larson D R, Ransom J E, Berry D J, Lewallen D G. Long-term mortality after revision THA. Clin Orthop Relat Res 2018; 476(2): 420-6. doi: 10.1007/s11999.0000000000000030.

Acta Orthopaedica 2019; 90 (3): 231–236

231

Resurfacing hip arthroplasty better preserves a normal gait pattern at increasing walking speeds compared to total hip arthroplasty Davey M J M GERHARDT 1, Thijs G TER MORS 1, Gerjon HANNINK 2, and Job L C VAN SUSANTE 1 1 Department of Orthopedics, Rijnstate Hospital, Arnhem; 2 Department of Operating Rooms, Radboud University Medical Center, Nijmegen, The Netherlands Correspondence: jvansusante@rijnstate.nl Submitted 2018-09-10. Accepted 2019-02-02.

Background and purpose — Gait analysis performed under increased physical demand may detect differences in gait between total (THA) versus resurfacing hip arthroplasty (RHA), which are not measured at normal walking speed.

We hypothesized that patients after RHA would reach higher walking speeds and inclines compared with THA. Additionally, an RHA would enable a more natural gait when comparing the operated with the healthy contralateral hip.

Patients and methods — From a randomized controlled trial comparing THA with RHA with at least 5 years’ followup patients with a UCLA score of more than 3 points (n = 34) were included for an instrumented treadmill gait analysis.

25 patients with a unilateral implant (primary analysis—16 THA versus 9 RHA) and 9 patients with a bilateral implant (sub-analysis—n = 5 RHA + THA; n = 4 THA + THA). Spatiotemporal parameters, ground reaction forces, and range of motion were recorded at increasing walking speeds and inclines. Functional outcome scores were obtained.

Results — At a normal walking speed of 1.1 m/s and at increasing inclines no differences were recorded in gait between the 2 groups with a unilateral hip implant. With increasing walking speed the RHA group reached a higher top walking speed (TWS) (adjusted difference 0.07 m/s, 95% CI –0.11 to 0.25) compared with THA. Additionally, RHA patients tolerated more weight on the operated side at TWS (155 N, CI 49–261) and as such weight-bearing approached the unaffected contralateral side. For the RHA group a “between leg difference” of 8 N (CI 3–245) was measured versus –129 N (CI –138 to –29) for THA (adjusted difference 144 N, CI 20–261). Hip flexion of the operated side at TWS was higher after RHA compared with THA (adjusted difference 8°, CI 1.7–14).

Interpretation — In this study RHA patients reached a higher walking speed, and preserved a more normal weight acceptance and a greater range of hip flexion against their contralateral healthy leg as compared with patients with a THA.

Increasing numbers of young patients choose hip arthroplasty instead of accepting hip impairment. In an attempt to increase implant durability and future revision options the metal on metal (MoM) resurfacing hip arthroplasty (RHA) was introduced, improving implant stability with the use of larger femoral head diameters and preservation of femoral bone stock (Amstutz et al. 2004, Grigoris et al. 2006, Gerhardt et al. 2015). Patients benefit from regaining hip function near to normal as gait analysis studies and questionnaires have shown (Mont et al. 2007, Bisseling et al. 2015). However, the use of RHA has decreased over the past decade due to concerns about adverse reactions to metal debris (Langton et al. 2010).

Still, the hip resurfacing concept, restoring patients’ mobility particularly in young active patients, remains relevant since previous studies have reported somewhat better functional outcome after RHA versus THA (Pollard et al. 2006, Heilpern et al. 2008, Haddad et al. 2015). So far, only 2 randomized controlled trials have been performed comparing postoperative gait between RHA and THA (Lavigne et al. 2010, Petersen et al. 2011). In these studies, the clinically perceived benefit of RHA compared with conventional THA on patient mobility and gait could not be confirmed. However, these studies may not be entirely conclusive since a limited number of patients were enrolled and measurements were done at normal walking speed. More modern gait analysis does allow assessment of patients’ gait pattern at increasing walking speeds and inclines. The advantage of using an instrumented treadmill is the ability to continuously increase speed and walking incline to detect gait differences that may not be detected at a normal or slow walking speed.

In this study a modern instrumented treadmill assisted gait analysis after RHA versus THA was performed where spatiotemporal, kinematic, and kinetic data could be continuously monitored under increasing walking speed and incline. We hypothesized that in this way RHA patients would still prove to preserve a more normal gait pattern of the operated leg similar to the gait pattern of the healthy contralateral leg.

© 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group, on behalf of the Nordic Orthopedic Federation. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits ­unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

DOI 10.1080/17453674.2019.1594096

Acta Orthopaedica Orthopaedica2019; 2019; 90 90 (3): (3): 231–236 231–236 Acta

232 232

Assessed for eligibility at 5 years postoperative RHA (n = 34)

THA (n = 26)

RHA excluded from gait analysis (n = 13): – not willing to participate (well-functioning implant), 2 – lost to follow-up, 1 – metastatic cancer disease, 1 – Parkinson´s disease, 1 – symptomatic contralateral hip osteoarthritis, 2 – symptomatic knee osteoarthritis, 2 – chronic obstructive pulmonary disease, 1 – cup revision, 2 – radicular back pain, 1

THA excluded from gait analysis (n = 13): – not willing to participate (well-functioning implant), 4 – lost to follow-up, 2 – claudicatio intermittens, 1 – polymyalgia rheumatic, 1 – symptomatic contralateral hip osteoarthritis, 2 – chronic obstructive pulmonary disease, 1 – bursitis trochanterica, 2

Included in gait analysis RHA (n = 21)

Figure 1. Flowchart of study.

Primary analysis: Unilateral implants n = 25 (RHA, 16 and THA, 9)

Patient and methods The study group included RHA and THA patients from a larger randomized controlled trial to compare RHA against a conventional small-diameter MoM THA with at least a complete 5-year follow up (Smolders et al. 2010). 34 RHA and 26 THA patients were available to participate in this instrumented gait analysis follow-up study. Only relatively active patients, who did not use walking aids during daily living, with more than 3 points according to the university of California at Los Angeles (UCLA) activity score at 5 years’ follow-up were approached. Exclusion criteria were contralateral hip osteoarthritis, presence of a total knee arthroplasty, or any musculoskeletal disorder affecting patients’ gait other than the hip implant. 34 patients could be included for gait analysis.

Patients were categorized into 2 groups: (1) 25 patients with a unilateral hip implant (16 RHA and 9 THA), and (2) 9 patients with a bilateral hip implant (4 THA+THA and 5 RHA+THA).

The primary analysis concerns patients with a unilateral hip implant, comparing THA with RHA; the secondary analysis concerns patients with bilateral implants. (Figure 1, Table 1).

Functional questionnaires 2 weeks prior to the gait analysis all patients completed the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC), EuroQol 5D (EQ-5D), UCLA activity score, and the Oxford Hip Score (OHS) to establish patientreported outcome after surgery (Table 2).

Gait analysis Patients were assessed by a Gait Real-time Analysis Interactive Lab (GRAIL; Motek Medical, Amsterdam, the Netherlands).

THA (n = 13)

Secondary analysis: Bilateral implants n=9 (RHA+THA, 5 and THA+THA, 4)

Table 1. Clinical details of the 2 groups of patients with a unilateral hip implant

Factor

RHA (n = 16)

THA (n = 9)

p-value

Sex (women/men) Mean body mass index (SD) Length, cm (SD) Weight, kg (SD) Mean age at surgery (SD) Mean follow-up in years (SD)

5/11 26 (3) 177 (13) 82 (19) 52 (10) 6.3 (1)

1/8 28 (5) 180 (9) 91 (21) 57 (8) 6.2 (0)

0.4 a 0.2 b 0.5 b 0.3 b 0.2 b 0.9 b

a Fisher’s exact probability b Student’s t-test.

test

Table 2. Clinical scores according to the UCLA activity score, Oxford Hip Score (OHS: best–worst 12–60 points scoring), EQ-5D visual analogue scale, and WOMAC hip score (best to worst 0–94 points scoring). Values are mean (SD) Unilateral OHS UCLA EQ-5D VAS WOMAC

RHA (n = 16) 14 (3) 6.9 (2.4) 80 (7) 4 (5)

THA (n = 9) 14 (2) 7.3 (2.4) 78 (10) 4 (5)

p-value 0.9 0.7 0.6 0.8

Student’s t-test was performed.

A 3D motion capture system with an instrumented dual-belt treadmill was employed, with force plates underneath both belts to record the kinetics of each step, left and right, independently at increasing speed and inclination (maximum 10 degrees). A motion-capture system with 24 anatomic placed body markers on the lower extremities, pelvis, and spine is

Acta Acta Orthopaedica Orthopaedica2019; 2019; 90 90 (3): (3): 231–236 231–236

Force Y – Ground reaction force (N)

Maximum weight acceptance Maximum push off Mid support

Impulse

Time during gait cycle (s)

Figure 2. Illustration of the ground reaction force (N) plotted against time (s) during a gait cycle from heel-strike to toe-off, with the maximum weight acceptance (first peak), mid support, maximum push (second peak) off, and the impulse (area under the curve).

integrated in the GRAIL to record changes in body position and range of motion of the hip joint during each gait cycle.

Since body markers were placed on anatomical landmarks including the lateral femoral epicondyle, greater trochanter of the femur, anterior and posterior superior iliac spine, sacral bone, and Th10, hip flexion and extension could be monitored after correction for concomitant spine motion. The 3D marker trajectories were collected (100 Hz) with a 10-camera 3D motion capture system (Vicon Nexus, Oxford Metrics Ltd, Oxford, UK) and processed in D-flow (Motekforce Link, Amsterdam, the Netherlands). Kinetic data were collected from the force plates (Forcelink, 12 channels, sample frequency 1000 Hz) during the stance phase according to the gait analysis protocol published by Aqil et al. (2013) resulting in 4 variables for analysis: maximum weight acceptance, mid support, maximum push off, and impulse. Maximum weight acceptance and maximum push off are the first and second force peaks in the stance phase with the mid support force being the lowest point between both peaks. The impulse is defined as the total force throughout the stance phase or the area under the curve (Figure 2). Kinetic data obtained from the force plates was normalized for bodyweight towards a standard 80 kg.

With the motion capture system the spatiotemporal parameters speed, step length, stride time, and cadence, as well as the kinematic data of hip range of motion, were continuously recorded. All patients were tested by 2 independent physical therapists, blinded for the implant type and side. Testing followed a 6-minute acclimatization period at a fixed 1.11 m/s (4 km/h) speed to eliminate inconsistencies in the patients’ gait due to lack of warming up (Matsas et al. 2000, Wiik et al.

2012). Patients walked wearing shoes, without any support or walking aids on the treadmill.

Flat ground walking started at 1.11 m/s and was increased by 0.28 m/s (1 km/h) every 20 s up to the patient’s top walking

233 233

speed (TWS). After a 5-minute break, patients were asked to walk uphill at a fixed speed of 1.11 m/s, with increments of 1° every 20 s. The treadmill was inclined up to the patient’s maximum walking incline (TWI), with a maximum treadmill incline of 10°, which corresponds to intensive hiking.

Statistics All data extracted from the GRAIL system were analyzed using MATLAB (R2014b, The MathWorks Inc, Natick, MA, USA). Force calculations were performed only on correctly measured steps. A correctly measured step was defined as a deviation from the mean of all steps of less than 2 times the standard deviation.

Descriptive statistics were used to summarize the data.

Independent samples t-tests and non-parametric independent t-tests were used to assess differences between baseline scores. As the participants were selected from a previously performed randomized controlled trial (Smolders et al. 2010), no prior power calculation was performed. Linear regression was used to test between-group differences (RHA vs. THA) in spatiotemporal, kinetic, and kinematic parameters while adjusting for sex, age at surgery, BMI, and UCLA score.

Results are presented as means with 95% confidence intervals (CI). Additional analyses were performed on the 9 patients with bilateral hip implants; an RHA combined with a THA (n = 5), or a bilateral THA (n = 4). In these small groups, paired t-tests were used to assess the between-leg differences regarding the ground reaction forces in both groups, with kinematic and spatiotemporal parameters. Differences were considered statistically significant with a p-value < 0.05. All statistical analyses were performed using R version 3.5.1 (R Foundation for Statistical Computing, Vienna, Austria).

Ethics, registration, funding, and potential conflicts of interest Approval from the regional ethics committee was obtained for the gait analysis (LTC 2015-0576/METC NL50830.091.14).

Written informed consent was obtained from all patients. The study was registered at ClinTrials.gov (NCT02484781). An independent institutional research grant was obtained (Rijnstate Vriendenfonds). There are no conflicts of interest to be reported by any of the authors.

Results Primary analysis Clinical outcome Patient characteristics and surgical data for the primary analysis are summarized in Table 1. 16 unilateral RHA and 9 unilateral THA were included. Clinical outcome scores are presented in Table 2. Overall both groups presented with a good OHS, UCLA score, EQ-5D VAS, and WOMAC without clinically relevant differences.

234

Primary gait analysis (patients with a unilateral hip implant) 1. Fixed speed (1.11 m/s), no incline (0°). No statistically significant differences were measured between the RHA and the THA regarding the spatiotemporal, kinetic, and kinematic parameters recorded when walking at a normal speed of 1.11 m/s (4 km/h) (Tables 3–5, see Supplementary data).

In addition, no statistically significant discrepancy between the implanted hip and the contralateral hip was observed irrespective of the type of implant.

2. Increasing walking speed towards TWS with no incline (0°). The RHA group reached a higher TWS than the THA group, 2.03 m/s (SD 0.2) versus 1.92 m/s (0.2), respectively (adjusted difference 0.07 m/s, CI –0.11 to 0.25). No differences were seen regarding the stride time, stride length, or cadence, both parameters showed similar adaptation towards TWS (Table 3, see Supplementary data). For the THA group an increasing walking speed coincided with an increasing discrepancy in weight acceptance between the implanted (977 N (103)) and the contralateral healthy hip (1,106 N (85)) (∆ = –129 N at TWS (CI –138 to –29). As for the RHA group a minor between-legs difference was measured of 1,132 N (153) versus 1,124 N (114) for the contralateral side (∆ = 8 N at TWS (CI 3–245). Thus the adjusted difference in weight acceptance between legs at TWS for the RHA group and the THA group was 144 N (CI 20–261). For the other ground reaction forces recorded, no statistically significant differences were seen at TWS (Table 4a, see Supplementary data). Regarding hip range of motion at an increasing walking speed up to patients’ TWS, hip flexion increased from 35° (6) towards 47° (5) after RHA versus 33° (8) towards 39° (6) after THA. Hip flexion of the operated leg at TWS was higher after RHA (adjusted difference 8° (CI 1.7–14). Additionally, a minor between leg difference was measured (adjusted difference 2.8, CI –7.4 to 1.8) (Table 5, see supplementary data).

3. Fixed walking speed and an increasing incline towards TWI. At an increasing incline with a fixed normal walking speed (1.11 m/s) no statistically significant differences were seen between RHA and THA at patients’ TWI (Table 3, see Supplementary data). Regarding the ground reaction forces measured, no differences between groups or between the operated and the contralateral hip were measured at increasing inclines (Table 4a, see Supplementary data). No difference was measured in patients’ hip range of motion between legs or between groups for hip flexion and extension (Table 5, see Supplementary data).

Secondary gait analysis: patients with bilateral hip implants The group of 5 bilateral patients (4 women) with an RHA on one side and a THA on the other side had a mean BMI of 28 (4), mean UCLA of 7.8 (0.8) and a mean age at surgery of 60 (4) years. From the small number of patients, obviously no statistically significant differences could be detected in

Acta Orthopaedica 2019; 90 (3): 231–236

between-leg differences regarding the ground reaction forces, stride length, stride time, and hip range of motion. Overall, at normal walking speed and TWI all mean values were comparable between the two legs (Table 4b and 6, see Supplementary data). This also accounted for TWS, except that the difference in maximum weight acceptance between the two legs increased in favor of the side with an RHA (adjusted difference 45 N (CI –63 to 153).

The group of 4 patients (3 women) with a bilateral THA had a mean BMI of 24 (2), mean UCLA of 7.4 (1), and mean age at surgery of 54 years (12). Again, overall no between-leg differences were measured at normal walking speed, TWS, or TWI regarding the ground reaction forces, stride length, stride time, and hip range of motion (Table 4b and 6, see Supplementary data). This time the difference in weight acceptance between the two legs (THA and THA) was less (2 N, CI –82 to 87).

Discussion Postoperative gait differences between RHA and THA remain controversial regarding presumed benefits for RHA. In this treadmill-assisted gait analysis indeed no differences in gait pattern were measured at a fixed flat walking speed of 1.11 m/s, nor at increasing inclines with a fixed flat walking speed.

However, with increasing speeds towards patients’ TWS, patients with an RHA had a weight acceptance on the operated hip similar to the weight acceptance of the healthy contralateral hip. In the group of patients with a THA this weight acceptance on the operated leg was relatively lower, resulting in a higher between-leg difference at top walking speed. In addition, a greater range of motion in the hip joint was measured and a trend towards a higher top walking speed and a greater stride length was observed after RHA. The primary gait analyses focused on patients with a unilateral hip implant.

In an attempt to maximize the potential inclusion of patients available for gait analysis a secondary gait analysis was also performed on patients with bilateral hip implants. The 2 secondary gait analyses of patients with a bilateral hip implants (RHA+THA and bilateral THA) confirmed overall the outcome of the primary analyses without clear between-leg differences in the evaluated outcome parameters, in particular for the bilateral THA group. Interestingly, patients with an RHA and a THA revealed a similar weight acceptance between legs at normal walking speed whereas at TWS the difference in weight acceptance increased in favor of the RHA side. With the small sample size this difference was statistically not significant.

2 randomized controlled trials on gait analysis have already reported on similar postoperative walking speed and gait restoration after RHA versus THA without a statistically significant difference between groups (Lavigne et al. 2010, Petersen et al. 2011). However, in contrast to our study Petersen et al.

(2011) assessed gait adaptation at patients’ comfortable walk-

Acta Orthopaedica 2019; 90 (3): 231–236

235

Force Y (kN) – RHA at 4 km/h walking speed

Force Y (kN) – RHA at top walking speed

Force Y (kN) – THA at 4 km/h walking speed

Force Y (kN) – THA at top walking speed

1.6

1.6

1.6

1.6

1.4

1.4

1.4

1.2

1.2

1.2

1.2

1.0

1.0

1.0

1.0

0.8

0.8

0.8

0.8

0.6

0.6

0.6

0.6

0.4

0.4

0.4

0.4

0.2

0.2

0.2

0.2

0

0

0

Healthy leg Operated leg

1.4

a

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

Time during gait cycle (sec)

0

0.1

0.2

0.3

0.4

0.5

0.6

Time during gait cycle (sec)

0

b

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

Time during gait cycle (sec)

0

0

0.1

0.2

0.3

0.4

0.5

0.6

Time during gait cycle (sec)

Figure 3. (a) Typical example of gait adaptation towards top walking speed (TWS) after RHA. Similar leg differences are seen at 4 km/h and at TWS (in this case at 8 km/h) between the healthy and operated leg.

(b) Typical example of gait adaptation towards top walking speed (TWS) after THA. Minor between leg differences are measured at 4 km/h between the healthy and operated leg; however, at TWS this between-leg difference increases, in this case at 7 km/h.

ing speed at 6 and 12 weeks after surgery in 30 patients (15 RHA versus 15 THA). A respective non-significant difference in speed increase from 6 to 12 weeks after surgery was observed of 1.19 (0.3) to 1.32 (0.2) m/s (RHA) versus 1.10 m/s (0.3) to 1.25 m/s (0.2) (THA). No differences in kinematic and kinetic parameters were measured between groups at these comfortable walking speeds. Lavigne et al. (2010) reported on a non-significant difference in comfortable walking speeds favoring RHA (n = 24) at 12 months postoperatively. In that study, however, RHA was compared with a large diameter head THA (n = 15) instead of a more conventional small diameter head THA. Both these earlier studies differ importantly from our study as most parameters were evaluated only at normal, comfortable walking speeds and weight acceptance of the operated leg was not assessed. For this reason we feel that these earlier studies missed the encountered differences in our study. A more recent non-randomized study by Wiik et al. (2012) also used gait analysis on an instrumented treadmill to compare 22 RHA, 22 THA, and a control group (n = 23). They reported on a top walking speed (TWS) of 2.06 m/s (0.22) after RHA versus 1.90 m/s (0.19) after THA (p < 0.05), whereas the control group without a hip arthroplasty reached 2.08 m/s (SD 0.17). As such that study also confirmed a better performance for RHA with increasing walking speeds; however, it may have been biased since cohorts were selected. The strength of our study is that these findings were also confirmed in patients from a randomized study.

The use of a large femoral head diameter in RHA and the absence of an intramedullary stem in the femur may explain the perceived more natural gait frequently claimed by patients after RHA. Our findings objectively support this assumption as indicated by a more natural postoperative gait restoration seen after RHA compared with conventional THA (Figure 3 shows typical examples of gait adaptation after RHA and THA). Besides the larger range of motion that can be obtained with a larger femoral head diameter (Burroughs et al. 2005), in particular the absence of an intramedullary stem did allow

for weight acceptance on the operated leg comparable to the non-operated contralateral side. Earlier studies already recognized this tendency towards a more physiological gait after RHA (Daniel et al. 2004, Mont et al. 2007, Aqil et al. 2013).

For example, Aqil et al. (2013) performed an instrumented treadmill-assisted gait analysis in 9 patients with bilateral hip arthroplasties, THA (head diameters range from 28 to 38mm) on one side versus RHA contralaterally. A strong correlation between increasing speeds and increasing between-leg differences in ground reaction force was also described.

The presence of an intramedullary stem obviously stiffens the femur, which in turn decreases weight tolerance of that femur. The fact that with RHA such a stemmed device can be avoided appears to be a benefit. It should be noted, however, that only with gait analysis using increasing walking speeds could this benefit for RHA be quantified and as such it may not be as clinically significant. In addition, this benefit for RHA has to be balanced against the increasing concern around metal-onmetal articulation (Smith et al. 2012, Bolognesi and Ledford 2015). For the future it may remain interesting to focus on research allowing for the resurfacing concept whilst avoiding metal-on-metal articulations (Van Susante et al. 2018).

An important strength of this study is that patients were recruited from a randomized trial and that selection bias could be avoided, which may be a risk when cohort series are compared. In addition the gait analysis was performed on an advanced instrumented treadmill with the use of motion capture, which ensured computerized measurements and provided a simultaneous registration of spatiotemporal, kinetic, and kinematic measurements.

Some potential limitations also have to be discussed. We decided to perform an additional gait analysis on patients already included in an RCT with 5 years’ follow-up completed. Thus the available number of patients was predetermined and could not be increased to improve the power of the study. In addition, patient inclusion was based on the UCLA score and the patient’s comorbidities recorded in their

236

medical file that could have influenced the gait analysis. This might have induced selection bias and introduced imbalances in covariates, although patient demographics and functional scores were similar. However, in our analyses we adjusted for these imbalances. Moreover, in a secondary analysis, bilateral cases were analyzed and confirmed our findings. Besides, due to the selection criteria a rather homogeneous study group of relatively active individuals was established, which may also have strengthened the study potential to identify an implantrelated difference in this rather small number of patients.

For the statistical analysis of the ground reaction forces patient bodyweight was normalized to 80 kg. Since the main focus of this study was detecting potential inter-patient (between leg) and not intra-patient differences this correction for bodyweight did not bias the results. Finally, since the number of available patients was predetermined no prior power calculation for this study was performed.

In summary this study confirms that at a normal walking speed (1.11 m/s) no major differences in patient postoperative gait pattern can be expected comparing RHA with conventional THA. However, with increasing walking speeds RHA patients preserved a more normal weight acceptance and a greater range of hip flexion against their contralateral healthy leg as compared with patients with a THA. We believe that maintenance of a large femoral head diameter and avoidance of stiffening the femur with an intramedullary stem are the main contributors to this benefit for RHA. Obviously, the concerns around adverse reaction to metal debris from a metalon-metal articulation so far remain an important disadvantage for RHA and should be balanced against this benefit in gait; future innovations avoiding metal-on-metal articulation in resurfacing remain interesting.

Supplementary data Tables 3–6 are available as supplementary data in the online version of this article, http://dx.doi.org/10.1080/17453674.

2019.1594096

The study was designed by DG and JLCvS. DG and TM carried out the data processing. The quality control of data was performed by DG, TM, and GH. DG and GH performed the statistical analysis. All authors interpreted the results. The manuscript preparation and editing were done by DG. All the authors reviewed the final manuscript.

Acta thanks Håkan Hedlund and Marketta Henriksson for help with peer review of this study.

Amstutz H C, Ebramzadeh E, Sarkany A, Le Duff M, Rude R. Preservation of bone mineral density of the proximal femur following hemisurface arthroplasty. Orthopedics 2004; 27(12): 1266-71.

Aqil A, Drabu R, Bergmann J H, Masjedi M, Manning V, Andrews B, Muirhead-Allwood S K, Cobb J P. The gait of patients with one resurfacing and one replacement hip: a single blinded controlled study. Int Orthop 2013; 37(5): 795-801.

Acta Orthopaedica 2019; 90 (3): 231–236

Bisseling P, Smolders J M, Hol A, van Susante J L. Metal ion levels and functional results following resurfacing hip arthroplasty versus conventional small-diameter metal-on-metal total hip arthroplasty: a 3 to 5 year follow-up of a randomized controlled trial. J Arthroplasty 2015; 30(1): 61-7.

Bolognesi M P, Ledford C K. Metal-on-metal Total Hip Arthroplasty: patient evaluation and treatment. J Bone Joint Surg Am 2015; 23(12): 724-31.

Burroughs B R, Hallstrom B, Golladay G J, Hoeffel D, Harris W H. Range of motion and stability in total hip arthroplasty with 28-, 32-, 38-, and 44-mm femoral head sizes. J Arthroplasty 2005; 20(1): 11-19.

Daniel J, Pynsent P B, McMinn D J. Metal-on-metal resurfacing of the hip in patients under the age of 55 years with osteoarthritis. J Bone Joint Surg Br 2004; 86(2): 177-84.

Gerhardt D M, Smolders J M, Rijnders T A, Hol A, van Susante J L. Changes in bone mineral density and femoral neck narrowing in the proximal femur three to five years after hip resurfacing versus conventional total hip arthroplasty. J Arthroplasty 2015; 30(2): 308-14.

Grigoris P, Roberts P, Panousis K, Jin Z. Hip resurfacing arthroplasty: the evolution of contemporary designs. Proc Inst Mech Eng H 2006; 220(2): 95-105.

Haddad F S, Konan S, Tahmassebi J. A prospective comparative study of cementless total hip arthroplasty and hip resurfacing in patients under the age of 55 years: a ten-year follow-up. Bone Joint J 2015; 97-B(5): 617-22.

Heilpern G N, Shah N N, Fordyce M J. Birmingham hip resurfacing arthroplasty: a series of 110 consecutive hips with a minimum five-year clinical and radiological follow-up. J Bone Joint Surg Br 2008; 90(9): 1137-42.

Langton D J, Jameson S S, Joyce T J, Hallab N J, Natu S, Nargol A V. Early failure of metal-on-metal bearings in hip resurfacing and large-diameter total hip replacement: a consequence of excess wear. J Bone Joint Surg Br 2010; 92(1): 38-46.

Lavigne M, Therrien M, Nantel J, Roy A, Prince F, Vendittoli P A. The John Charnley Award: The functional outcome of hip resurfacing and large-head THA is the same: a randomized, double-blind study. Clin Orthop Relat Res 2010; 468(2): 326-36.

Matsas A, Taylor N, McBurney H. Knee joint kinematics from familiarised treadmill walking can be generalised to overground walking in young unimpaired subjects. Gait Posture 2000; 11(1): 46-53.

Mont M A, Seyler T M, Ragland P S, Starr R, Erhart J, Bhave A. Gait analysis of patients with resurfacing hip arthroplasty compared with hip osteoarthritis and standard total hip arthroplasty. J Arthroplasty 2007; 22(1): 100-8.

Petersen M K, Andersen N T, Mogensen P, Voight M, Soballe K. Gait analysis after total hip replacement with hip resurfacing implant or Mallory-head Exeter prosthesis: a randomised controlled trial. Int Orthop 2011; 35(5): 667-74.

Pollard T C, Baker R P, Eastaugh-Waring S J, Bannister G C. Treatment of the young active patient with osteoarthritis of the hip: a five- to seven-year comparison of hybrid total hip arthroplasty and metal-on-metal resurfacing. J Bone Joint Surg Br 2006; 88(5): 592-600.

Smith A J, Dieppe P, Howard P W, Blom A W, National Joint Registry for England and Wales. Failure rates of metal-on-metal hip resurfacings: analysis of data from the National Joint Registry for England and Wales. Lancet 2012; 380(9855): 1759-66.

Smolders J M, Hol A, Rijnders T, van Susante J L. Changes in bone mineral density in the proximal femur after hip resurfacing and uncemented total hip replacement: a prospective randomised controlled study. J Bone Joint Surg Br [Randomized Controlled Trial] 2010; 92(11): 1509-14.

Van Susante J L C, Verdonschot N, Bom L P A, Tomaszewski P, Campbell P, Ebramzadeh E, Schreurs B W. Lessons learnt from early failure of a patient trial with a polymer-on-polymer resurfacing hip arthroplasty. Acta Orthop 2018; 89(1): 59-65.

Wiik A, Tankard S, Carten R, Lewis A, Amis A, Cobb J. An instrumented treadmill enables measurement of gait at high speeds, suggesting hip resurfacing almost matches normal controls, while hip replacement does not. In: ORS Annu Meet 2012. (Unpublished data)

Acta Orthopaedica 2019; 90 (3): 237–242

237

Equivalent hip stem fixation by Hi-Fatigue G and Palacos R+G bone cement: a randomized radiostereometric controlled trial of 52 patients with 2 years’ follow-up Peter B JØRGENSEN 1,2, Martin LAMM 1, Kjeld SØBALLE 1,2, and Maiken STILLING 1,2 1 Department of Orthopedic Surgery, Aarhus University Hospital; 2 Department of Clinical Medicine, Aarhus University, Denmark Correspondence: pbjr@clin.au.dk Submitted 2018-03-12. Accepted 2019-02-08.

Background and purpose — Long-term fixation of cemented femoral stems relies on several factors including cement adhesion and fatigue. Hi-Fatigue is a newer thirdgeneration bone cement with low-viscosity properties at room temperature, good mechanical strength, and stable bone–cement interface in a laboratory testing environment.

Palacos bone cement has excellent 10-year survival and is considered gold standard. We compared stem subsidence after fixation with Hi-Fatigue and Palacos bone cements using radiostereometry.

Patients and methods — In a patient-blinded randomized controlled trial, 52 patients (30 women) at mean age 76 years (71–87) with osteoarthrosis and no osteoporosis received Hi-Fatigue G or Palacos R+G cement fixation of collarless, polished, double-tapered stems (CPT). Tantalum beads were inserted in the periprosthetic bone. Supine stereoradiographs were obtained postoperatively, 3 months, 6 months, 1 year, and 2 years after surgery. Oxford Hip Score (OHS) and VAS pain were recorded preoperatively and 1 and 2 years after surgery. Cement working times and properties were registered.

Results — At 2 years, mean stem subsidence of 1.12 mm (95% CI 0.96–1.29) for Hi-Fatigue and 1.19 mm (CI 1.03– 1.34) for Palacos was similar. Likewise, stem version was comparable between cement groups. Mean OHS and VAS pain were similar between cement groups.

Cement working times were similar between cement groups, but the mean curing time was longer for Hi-Fatigue (13.7 min) than for Palacos (11.6 min).

Interpretation — We found similar and generally low migration of CPT femoral stems inserted with Hi-Fatigue and Palacos bone cement until 2 years’ follow-up, which indicates a good long-term survival of polished taper femoral stems inserted with both cement types.

Palacos bone cement was introduced in the early 1970s, is widely used, and has a reported 10-year implant survival rate between 92.6% and 98.8%, when used in hip arthroplasties (Junnila et al. 2016), and may be considered a gold standard bone cement (van der Voort et al. 2015, Junnila et al. 2016).

Hi-Fatigue G bone cement is a newer third-generation bone cement, with a low initial viscosity but without long-term follow-up data.

Both Hi-Fatigue G and Palacos R+G bone cements surpass international standards for stability tests according to ISO 5833:2002 (ISO 2002). They both contain gentamicin sulphate (Hi-Fatigue G 0.55/40 g, Palacos 0.8/40 g) and are sterilized using ethylene oxide. The radiopaque medium for both cements is zirconium dioxide, but the concentration is lower in Hi-Fatigue G (12 %) compared with Palacos (17 %), which acts positively for the mechanical stability of HiFatigue G bone cement (Arora et al. 2013). Compared with Palacos R+G, Hi-Fatigue G bone cement has a lower initial viscosity that allow for good cement-to-bone penetration, interface strength, and a better fatigue life, minimizing the risk of cement failure (Rey et al. 1987, Stone et al. 1996, Race et al. 2006, Tanner 2008). These features may lead to better implant fixation with Hi-Fatigue G bone cement compared with Palacos R+G. RSA can be used to measure implant migration with respect to tantalum markers in the surrounding bone as part of a phased introduction (Nelissen et al. 2011). Stem subsidence and retroversion have been shown to be good predictors of implant survival and this has further been used and suggested as a standard in the evaluation of new bone cements (Karrholm et al. 1994, Hauptfleisch et al. 2006). We hypothesized less subsidence of polished femoral stems fixed with Hi-Fatigue G bone cement compared with polished femoral stems fixed with Palacos bone cement.

© 2019 The Author(s). Published by Taylor & Francis on behalf of the Nordic Orthopedic Federation. This is an Open Access article distributed under the terms of the Creative Commons Attribution-Non-Commercial License (https://creativecommons.org/licenses/by/4.0) DOI 10.1080/17453674.2019.1595390

238

Patients and methods The study design was a patient-blinded, randomized, controlled study.

Between November 2010 and March 2014, 52 patients (30 women) with a mean age of 76 years (71–87) were included in the study (Figure 1). The criteria for inclusion were primary osteoarthritis, age 71 years and above, preoperative T-score above –2.5 (meaning no osteoporosis), and informed consent. The exclusion criteria were neuromuscular and vascular disease, former proximal femoral fracture, osteonecrosis of the femoral head, pharmacological (NSAID, estrogens, cortisone), metabolic bone disease, senile dementia, alcohol or drug abuse, major psychiatric disease, metastatic cancer/radiation or chemotherapy, poor dental status (infection risk), spine disease, or severe systemic disease (i.e. hemiparesis or Parkinson’s disease). Sample size calculation indicated 24 patients per group based on a clinically relevant difference in subsidence of 0.33 mm (SD 0.39) with a power of 90% and alpha set to 0.05 (Karrholm et al. 1994, Glyn-Jones et al. 2006). To balance postoperative dropout, we aimed for 25 patients per group. To compensate for dropouts during the inclusion period of the study we included an additional 2 patients.

Randomization was done within 6 blocks of 10 patients (5 CPT stems fixed with Hi-Fatigue G bone cement, and 5 CPT stems fixed with Palacos R+G bone cement) by drawing concealed labels from sequentially numbered closed envelopes. Randomization was done in theater. The surgeon was not blinded to the type of cement used. 27 patients received stem fixation by Hi-Fatigue G bone cement (Zimmer Biomet) and 25 patients received stem fixation by Palacos R+G bone cement (Heraeus).

The components used were the collarless, polished, doubletapered CPT (12–14 conus Cr-Co) femoral stem (Zimmer Biomet), which has shown excellent long-term results (DHR 2016), the cementless Trilogy Fiber-Mesh Cup (Zimmer Biomet) with optional screw fixation and a highly crosslinked Longevity polyethylene liner (Zimmer Biomet). Femoral heads were CoCr size 36 mm.

The cement was stored in the theater at 20° C (18.3–21.5) and at 44% humidity (17–78) for at least 24 hours before surgeries, and in similar general storage conditions with temperature 20° C (17.5– 21.4) and at 48% humidity (24–78) with some seasonal variation. Both types of bone cement were vacuum mixed with the closed MixiGun system (Zimmer Biomet). The cement curing time was monitored using a digital timer with 1-second resolution. Nurses and surgeons also evaluated the consistency and user friendliness of each bone cement type on a numeric scale from 1 to 9.

All procedures were performed by 6 experienced hip surgeons. A preoperative plan was made using standardized digital radiographs with a 30 mm metal ball marker, at the level of the greater trochanter, and the AGFA OT3000 digital tem-

Acta Orthopaedica 2019; 90 (3): 237–242

Table 1. Baseline demographics. Values are mean (95% CI) or frequency Baseline demographics Hi-Fatigue

Palacos

n Age Sex (male/female), n T-score BMI Oxford Hip Score Pain rest Pain activity

27 76 (75–78) 9/18 –1.1 (–1.5 to –0.8) 29 (27–31) 22 (19–25) 3.4 (2.4–4.4) 6.3 (5.4–7.1)

25 76 (74–78) 13/12 –0.9 (–1.3 to –0.5) 29 (27–31) 25 (20–28) 3.5 (2.2–4.8) 5.7 (4.5–6.8)

Oxford Hip Score: 0–48, 48 being best

plating software [AGFA, Vancouver], for optimal correction of lateral and vertical offsets, and optimal postoperative leg length. A posterolateral approach was used. During surgery 1-mm tantalum beads were inserted into the peri-prosthetic femoral bone (lesser and greater trochanter). Peroperative leg length and stability was assessed before and after implantation using trial components (the femoral rasp, and trial heads). All patients received prophylactic antibiotics: intravenous cefuroxime 1.5 g preoperatively and 1.5 g 3 times in the postoperative 24 hours, and thrombo-prophylactic treatment: subcutaneous Arixtra (fondaparinux) 2.5 mg/day or Innohep (tinzaparin) 4,500 anti-Xa IE/day postoperatively until discharge.

Postoperatively the patient was mobilized with full weightbearing and walking aids as needed, using a “fast track” protocol.

Pre- and postoperative characteristics of the study population (Table 1) Patients assessed for study participation and follow-up of randomized participants are shown in the CONSORT flowchart (Figure 1). 1 patient was excluded during surgery because the MixiGun jammed twice during application of cement, due to a human error. At 2 years’ follow-up, there have been no revisions due to aseptic implant loosening. 1 patient suffered a traumatic periprosthetic fracture 18 months after operation and received revision of the stem and osteosynthesis of the fracture. 2 patients suffered hip dislocation within the first 3 months, 1 of these combined with avulsion of the greater trochanter. Another patient had avulsion of the greater trochanter post-surgery without known trauma. Both avulsions were treated nonoperatively. There was 1 periprosthetic infection 1 month postoperatively treated with soft tissue debridement and change of acetabular liner and metal head. This patient had a full recovery but died of causes unrelated to the periprosthetic infection 2 months before 2-year follow-up. The clinical scores (OHS and VAS pain) were similar at 1-year follow-up and 2-year follow-up between groups (Figure 2).

At 2-year follow-up, VAS pain and OHS correlated neither with subsidence (rho < 0.2) nor with retroversion (rho < 0.01) (Table 2, see Supplementary data).

Acta Orthopaedica 2019; 90 (3): 237–242

ENROLLMENT

Assessed for eligibility n = 287 Excluded (n = 235): – met exclusion criteria, 95 a – declined, 71 – not hip arthroplasty, 37 – other, 25 Randomized n = 52

ALLOCATION

Allocated to control (n = 27) Received allocated interventiion (n = 27)

Allocated to intervention (n = 25) Received allocated interventiion (n = 25)

FOLLOW-UP

Lost to follow-up at 5 years (n = 2): – died between 1 and 2 years, 1 – declined to attend 2-year follow-up, 2

Lost to follow-up at 5 years (n = 1): – revision due to fracture between 1 and 2 years, 1

ANALYSIS

Analyzed at endpoint (n = 24)

Analyzed at endpoint (n = 24)

Excluded from analysis (n = 1): – inadequate RSA-marker configuration

Excluded from analysis (n = 0)

Figure 1. Consort flow chart. a In 2012 patients taking medication with vitamin K antagonists were removed from the list of contraindications, because its relevance disappeared with increased use of NOAK. Until then 16 patients were excluded due to use of vitamin K antagonists.

Primary outcome measure: RSA A standard RSA setup of 2 synchronized ceiling-fixed roentgen tubes (Arco-Ceil/Medira; Santax Medico, Aarhus, Denmark) angled toward each other at 40° were used. All radiographs were fully digital (Fuji CR, image size 35 x 43) and were stored in DICOM file format without compression.

During this study the RSA equipment was replaced with a newer direct digital dedicated stereo X-ray system, AdoraRSA suite (NRT, Aarhus, Denmark) with CXDI-70C detectors (Canon, Tokyo, Japan). The configuration remained: 2 ceiling-mounted X-ray tubes at 40 degrees angle. We used a uniplanar carbon calibration box (Box 24, Medis Specials, Leiden, Netherlands). Patients had loaded full weight on the hip prosthesis before radiostereometric examinations, which were recorded with the patient supine.

Model-Based RSA 3.34 (RSAcore, Leiden, the Netherlands) was used for RSA analysis using EGS-models (Kaptein et al. 2006). The upper limit for mean rigid body fitting error (RBE), which is the stability limit for markers used in the analysis, was per default 0.5 mm in the software. The mean rigid body error of the bone markers was 0.19 (CI 0.16–0.22).

The mean condition number (dispersion of the bone markers in the femur) was 39 (CI 31–46). The difference in matching of the EGS hip-stem model to the CPT stem (model pose estimation) in the stereoradiographs was mean 0.10 mm (CI 0.01–0.11).

Double examination stereoradiographs were obtained for all patients to document the clinical precision (Valstar et al.

2005). The double examination stereoradiographs were performed with complete repositioning of the patient and the

239

radiographic equipment between examinations. The postoperative stereoradiograph was used as the reference in the migration analysis of the double examinations, and the expected difference in displacement between the 2 calculations represents the systematic error of the RSA system (bias) and should be zero.

Based on clinical double examination stereoradiographs the precision of RSA expressed as the coefficient of repeatability (CR) was approximately 0.2 mm for translations including TT and 0.3 mm for translation on the z-axis, less than 1° for rotation about the X and Z axes, and—as expected—less precise for rotation about the Y-axis 2.1° (Table 3, see Supplementary data). 1 patient was excluded based on inadequate marker configuration with a high condition number. Another 3 patients had condition numbers between 150 and 250, which was above the suggested 150 upper limit, but the configuration of the marker models was visually good (not linear) and acceptable and RSA data were included in the study (Valstar et al. 2005, ISO 2013).

Evaluation of femoral stem migration by radiostereometric analysis was performed at 3 months, 6 months, 1 year, and 2 years with postoperative stereoradiographs as baseline.

Secondary outcome measures Radiographic evaluation of the cementing was performed by one blinded assessor (PBJ). Cement distribution was evaluated on the postoperative AP and Axial radiographs as either “excellent” (A), “Slight radiolucency at the cement–bone interface” (B), “Radiolucency involving 50–99% of the cement–bone interface” (C) and “Radiolucency in 100% of the cement–bone interface in any projection or failure to fil the canal such that the tip was not covered” (D), as proposed by Barrack et al. (1992).

Patient-reported outcome measures consisted of pain measured during rest and activity on a Visual Analogue Scale (VAS) from 0 to 10 (10 being the worst), and Oxford Hip Score (OHS) was between 0 and 48 (0 being worst) (Paulsen et al. 2012). VAS pain and OHS score were evaluated before operation and at 1 and 2 years after operation.

Statistics All continuous variables were evaluated for normality using qqplots. The groups were then compared using Student’s t-test or non-parametric tests (Mann–Whitney U-test and Spearman’s rank correlation coefficient), as appropriate. The primary RSA endpoint was y-translation (subsidence). The secondary RSA endpoints were the remaining individual migrations along and rotations about the single axes, the summed migration in terms of total translation (TT = sqrt(Tx2+Ty2+ Tz2)), total rotation (TR = sqrt(Rx2+ Ry2+ Rz2)) and Maximum Total Point Motion (MTPM) (Selvik 1989, Valstar et al. 2005), and the clinical data. For OHS and pain scores, we present mean values and 95% confidence intervals (CI) for comparison with the literature. Statistical significance was assumed at p < 0.05.

240

Acta Orthopaedica 2019; 90 (3): 237–242

Oxford Hip Score

Y-axis translation (mm)

Retroversion (°)

0

50

2.5 Hi–Fatigue Palacos

Hi–Fatigue Palacos 2.0

40 –0.5

1.5

30

1.0 20

–1.0 0.5

10 Hi–Fatigue Palacos 0

0

0 –1.5 12

24

Months after index operation

Figure 2. Oxford Hip Scores: preoperative, 12, and 24 months postoperatively. Both HiFatigue G (p < 0.001) and Palacos R+G (p < 0.001) achieved significant increase from 0 to 12 months with no statistically significant difference between groups (p = 0.3).

0

3

6

12

24

Months after index operation

Figure 3. Mean subsidence (negative y-translation) of CPT stems inserted with Hi-Fatigue G and Palacos bone cements. Confidence intervals are presented in error bars, for graphical use only.

Stata/SE version 13.1 (StataCorp, College Station, TX, USA) was used for statistical computations.

Ethics, registration, funding, and potential conflicts of interest The study was approved by the Central Denmark Region Committee on Biomedical Research Ethics (Journal no.

M-20100112; issue date: May 27, 2010) and it was registered with ClinicalTrials.gov (NCT01289834) and the Data Protection Agency (2007-58-0010, issue 19 Oct 2010). The trial was performed in compliance with the Helsinki II Declaration. The RSA analyses of this study were funded by Zimmer Biomet.

All authors report no conflicts of interest.

Results Primary outcome Subsidence was greater than the detection limit (0.16 mm) for all patients but not statistically significantly different for CTP stems fixed with Hi-Fatigue G bone cement and Palacos R+G bone cement respectively after both 1 year (p = 0.2) and 2 years (p = 0.7) (Figure 3, Tables 3 and 4, see Supplementary data). The estimated 2-year difference of 0.06 mm (CI –0.17 to 0.3) in subsidence between the groups excludes the clinically relevant difference of 0.33 mm. Therefore, we find the subsidence in both groups to be similar.

Likewise, all other signed translations and rotations and summed migration measures (TT, TR, MTPM) were similar between cement groups (Table 4, see Supplementary data, Figure 4).

0

3

6

12

24

Months after index operation

Figure 4. Mean y-rotation (retroversion) of CPT stems inserted with Hi-Fatigue G and Palacos bone cements. Confidence intervals are presented in error bars, for graphical use only.

At 1-year follow-up 7 patients (all Palacos R+G) surpassed the subsidence threshold of 1.2 mm, which Karrholm et al.

(1994) showed to give a 50% risk of later revision in anatomic stems. At 2-year follow-up, a total of 14 patients (7 Palacos R+G) exceeded the subsidence threshold of 1.2 mm, but all had low pain scores (mean 0.8, CI 0.1–1.5) and good Oxford Hip Score (mean 45, CI 42–47).

Secondary outcomes Both Hi-Fatigue G and Palacos R+G showed good cement distribution (whiteout), but Palacos R+G was more often classified with slight radiolucency (n = 10) than Hi-Fatigue G (n = 1). There were no differences in stem position in the 2 groups (Table 5, see Supplementary data).

The surgical time (n = 51) was on average 85 minutes (50– 150). 1 operation took a particularly long time because some equipment was contaminated and needed sterilization during the surgery, but this was not related to use of the bone cement.

There were no statistically significant differences in working times for mixing and waiting for readiness to use of the bone cements (p > 0.4). Mean time of stem insertion was 16 seconds shorter for Palacos R+G (3:49) than for Hi-Fatigue G (4:05) and the mean time until total curing was 2:08 minutes longer for Hi-Fatigue G (13:46) than for Palacos R+G bone cement (11:35) (Table 6, see Supplementary data). The final curing time of Hi-Fatigue G correlated with the theater temperature and storing temperature (Table 7, see Supplementary data). No other correlations were found between mixing time, time to apply cement and stem, or total curing time and storage temperature or temperature in the theater in either of the 2 bone cements (rho < 0.24, p > 0.3).

Acta Orthopaedica 2019; 90 (3): 237–242

Discussion The key finding of this study was similar 2-year stem subsidence and clinical results in patients operated with a collarless polished double-tapered femoral stem that was fixed with either the newer Hi-Fatigue G bone cement or Palacos R+G bone cement.

Migration The primary outcome of this study registered with ClinicalTrials.gov was femoral component migration in broader terms.

Until now, no migration direction or pattern has been identified to clearly predict later loosening of CPT stems. However, subsidence of CPT stems has been associated with inferior cementing quality (Yates et al. 2008) and was therefore chosen in our study as the primary effect parameter. Using a single migration measure as the primary outcome also limited the risk of multiplicity testing issues.

Some subsidence is expected with the collarless polished double-tapered design, unlike the shape-closed stem designs. However, progressive and excessive migration is considered to be a proxy measure for later aseptic loosening (Nieuwenhuijse et al. 2012). In our study both bone cements provided acceptable stem fixation and the migration measures were comparable to similar RSA studies of cemented, collarless, polished, tapered femoral stems. In both cement groups the mean subsidence was more than the reported threshold of 0.33 mm (mean 0.82, CI 0.72–0.92) at 6 months, but at 2-year follow-up subsidence was less than reported threshold values that are predictive of later revision of cemented shape-closed stems (Lubinius SP1) (Karrholm et al. 1994).

The subsidence in both groups was less than 1.03 mm at 1-year follow-up, which is in accordance with other findings of 0.8 mm subsidence with the CPT stem in the first year (Yates et al. 2008). At 2-year follow-up both cement groups had subsided less than 1.2 mm. This is similar to 2-year findings of 1.36 mm of the C-stem (DePuy) (von Schewelov et al.

2014) and 1.42 mm and 0.92 mm of the Exeter stem (Stryker) (Nieuwenhuijse et al. 2012, Murray et al. 2013). These studies found no late revisions after 10 years. In Nordic countries the CPT stem was used in 6,222 THAs (5,630 with Palacos) from 1995 to 2013. These patients (mean age 73) had a 98.7% 10-year implant survivorship with aseptic loosening as endpoint (Junnila et al. 2016).

For both cement groups in our study, stem retroversion was below 1.8 degrees at 2-year follow-up, which is in accordance with reported retroversion of other similar polished stem types, such as findings of 1.42 degrees retroversion for the Exeter stem (Stryker) (Nieuwenhuijse et al. 2012), 1.6 degrees retroversion of the C-stem (DePuy) (von Schewelov et al. 2014) and 1.58 degrees for C-stem and 1.43 degrees for Exeter stem (Flatoy et al. 2015) at 2 years’ follow-up.

241

Clinical evaluation The clinical evaluation of the slow-curing Hi-Fatigue G bone cement revealed that curing time for Hi-Fatigue G was more than 2 minutes longer as compared with Palacos R+G bone cement. This may seem like a short time, but it is in fact a longer waiting time, where the surgeon needs to apply manual pressure support on the femoral stem in the bone cement during cement curing.

Our patients reached a mean OHS score slightly better than the Danish background population-based OHS score of 40 and better than the threshold (OHS = 40) correspondent to acceptable symptoms after THR (Paulsen et al. 2012, Keurentjes et al. 2014).

Cement curing The mixing time and waiting phase in our Palacos R+G group correspond to the in-vitro findings of Dall et al. (2007) who presented mixing time of 50 seconds and waiting phase of 55 seconds. The working phase of 383 seconds and setting time of 76 seconds in the Dall study is, however, a bit lower compared with our in-vivo findings, and could be explained by a higher storage temperature in the Dall study resulting in shorter working time (Kuehn et al. 2005). In our study, the relatively low theater temperature combined with good correlation between storing/theater temperature curing time for HiFatigue G can also explain why Hi-Fatigue G exhibits longer curing time than Palacos R+G. Slow-curing Hi-Fatigue G bone cement shows better fatigue test in comparison with Palacos R+G bone cement (Tanner 2008). This factor may be important for long-term survival of cemented implants, and short-term (2 years) follow-up RSA in this study revealed only small translations and rotation of CPT stems inserted with Hi-Fatigue G bone cement, which gives positive expectations for long-term survival (Olerud et al. 2014, Meinardi et al. 2016).

Strengths and weaknesses The generalizability of the study results translates into elderly (> 70 years) non-osteoporotic hip osteoarthritis patients treated with cemented CPT femoral stem. Yet cemented femoral stems are typically used in older more fragile and often osteoporotic patients, i.e. in the treatment of displaced intracapsular femoral neck fracture by hemiarthroplasty or total hip arthroplasty. Another limitation in this study is a high number of exclusions and non-consenters.

The typical reason for declining to participate in the study was poor health and inability to show for several follow-ups, and potentially there is a selection bias towards more fit and mobile elderly patients in the study compared with the typical elderly THA patient.

We expected less stem subsidence with slow-curing HiFatigue G cement because of better bone penetration. However, we found no difference in stem subsidence between cement groups in this study. We did not put markers in the bone cement, and thus we do not know if the measured subsid-

242

ence took place in the stem-mantle zone or the bone cement junction (Stefansdottir et al. 2004).

In conclusion, this study showed equivalent femoral stem subsidence with Hi-Fatigue G as compared with Palacos R+G bone cement until 2 years’ follow-up. We found 2 minutes longer curing time for Hi-Fatigue G compared with Palacos R+G bone cement. Migrations were similar to other studies of cemented polished stems, and clinical outcomes were above the threshold for acceptable symptoms after THR. Based on this study we expect similar long-term results for fixation of the CPT stem with Hi-Fatigue G and Palacos R+G bone cements.

Supplementary data Tables 2–7 are available as supplementary data in the online version of this article, http://dx.doi.org/10.1080/17453674.

2019.1595390 PBJ and MS wrote the manuscript and performed the statistical analyses.

ML, KS and MS contributed to planning, interpretation of data, and critical review of the manuscript.

Note: All product names are for identification purposes only, and may be trademarks of their respective owners.

Acta thanks Stephan Maximilian Röhrl and Olof Sköldenberg for help with peer review of this study.

Arora M, Chan E K S, Gupta S, Diwan A D. Polymethylmethacrylate bone cements and additives: a review of the literature. World J Orthop 2013; 4(2): 67-74.

Barrack R L, Mulroy R D Jr, Harris W H. Improved cementing techniques and femoral component loosening in young patients with hip arthroplasty: a 12-year radiographic review. J Bone Joint Surg Br 1992; 74(3): 385-9.

Dall G F, Simpson P M, Breusch S J. In vitro comparison of Refobacin-Palacos R with Refobacin Bone Cement and Palacos R + G. Acta Orthop 2007; 78(3): 404-11.

DHR. National Report 2016. The Danish Hip Arthroplasty Registry; 2016.

Flatoy B, Rohrl S M, Rydinge J, Dahl J, Diep L M, Nordsletten L. Triple taper stem design shows promising fixation and bone remodelling characteristics: radiostereometric analysis in a randomised controlled trial. Bone Joint J 2015; 97-B(6): 755-61.

Glyn-Jones S, Alfaro-Adrian J, Murray D W, Gill H S. The influence of surgical approach on cemented stem stability: an RSA study. Clin Orthop Relat Res 2006; 448: 87-91.

Hauptfleisch J, Glyn-Jones S, Beard D J, Gill H S, Murray D W. The premature failure of the Charnley Elite-Plus stem: a confirmation of RSA predictions. J Bone Joint Surg Br 2006; 88(2): 179-83.

ISO. International Standard ISO 5833:2002. Implant for surgery: acrylic resin cements. Geneva: ISO; 2002.

ISO. International standard ISO 16087:2013. Implants for surgery: roentgen stereophotogrammetric analysis for the assessment of migration of orthopaedic implants. Geneva: ISO; 2013.

Junnila M, Laaksonen I, Eskelinen A, Pulkkinen P, Ivar Havelin L, Furnes O, Marie Fenstad A, Pedersen A B, Overgaard S, Karrholm J, Garellick G, Malchau H, Makela K T. Implant survival of the most common cemented total hip devices from the Nordic Arthroplasty Register Association database. Acta Orthop 2016; 87(6): 546-53.

Acta Orthopaedica 2019; 90 (3): 237–242

Kaptein B L, Valstar E R, Spoor C W, Stoel B C, Rozing P M. Model-based RSA of a femoral hip stem using surface and geometrical shape models.

Clin Orthop Relat Res 2006; 448: 92-7.

Karrholm J, Borssen B, Lowenhielm G, Snorrason F. Does early micromotion of femoral stem prostheses matter? 4–7-year stereoradiographic follow-up of 84 cemented prostheses. J Bone Joint Surg Br 1994; 76(6): 912-17.

Keurentjes J C, Van Tol F R, Fiocco M, So-Osman C, Onstenk R, KoopmanVan Gemert A W, Poll R G, Nelissen R G. Patient acceptable symptom states after total hip or knee replacement at mid-term follow-up: thresholds of the Oxford hip and knee scores. Bone Joint Res 2014; 3(1): 7-13.

Kuehn K D, Ege W, Gopp U. Acrylic bone cements: composition and properties. Orthop Clin North Am 2005; 36(1): 17-28, v.

Meinardi J E, Valstar E R, Van Der Voort P, Kaptein B L, Fiocco M, Nelissen R G. Palacos compared to Palamed bone cement in total hip replacement: a randomized controlled trial. Acta Orthop 2016; 87(5): 473-8.

Murray D W, Gulati A, Gill H S. Ten-year RSA-measured migration of the Exeter femoral stem. Bone Joint J 2013; 95-B(5): 605-8.

Nelissen R G, Pijls B G, Karrholm J, Malchau H, Nieuwenhuijse M J, Valstar E R. RSA and registries: the quest for phased introduction of new implants.

J Bone Joint Surg Am 2011; 93(Suppl 3): 62-5.

Nieuwenhuijse M J, Valstar E R, Kaptein B L, Nelissen R G. The Exeter femoral stem continues to migrate during its first decade after implantation: 10-12 years of follow-up with radiostereometric analysis (RSA). Acta Orthop 2012; 83(2): 129-34.

Olerud F, Olsson C, Flivik G. Comparison of Refobacin bone cement and palacos with gentamicin in total hip arthroplasty: an RSA study with two years follow-up. Hip Int 2014; 24(1): 56-62.

Paulsen A, Odgaard A, Overgaard S. Translation, cross-cultural adaptation and validation of the Danish version of the Oxford hip score: assessed against generic and disease-specific questionnaires. Bone Joint Res 2012; 1(9): 225-33.

Race A, Miller M A, Clarke M T, Mann K A, Higham P A. The effect of lowviscosity cement on mantle morphology and femoral stem micromotion: a cadaver model with simulated blood flow. Acta Orthop 2006; 77(4): 607-16.

Rey R M Jr, Paiement G D, McGann W M, Jasty M, Harrigan T P, Burke D W, Harris W H. A study of intrusion characteristics of low viscosity cement Simplex-P and Palacos cements in a bovine cancellous bone model. Clin Orthop Relat Res 1987; (215): 272-8.

Selvik G. Roentgen stereophotogrammetry. Acta Orthop Scand 1989; 60(Suppl. 232): 1-51.

Stefansdottir A, Franzen H, Johnsson R, Ornstein E, Sundberg M. Movement pattern of the Exeter femoral stem: a radiostereometric analysis of 22 primary hip arthroplasties followed for 5 years. Acta Orthop Scand 2004; 75(4): 408-14.

Stone J J, Rand J A, Chiu E K, Grabowski J J, An K N. Cement viscosity affects the bone cement interface in total hip arthroplasty. J Orthop Res 1996; 14(5): 834-7.

Tanner K. Fatigue testing of four bone cements for AAP Biomaterials GmbH Co. Test Report. London: Department of Materials, Queen Mary University of London; 2008.

Valstar E R, Gill R, Ryd L, Flivik G, Borlin N, Karrholm J. Guidelines for standardization of radiostereometry (RSA) of implants. Acta Orthop 2005; 76(4): 563-72. van der Voort P, Pijls B G, Nieuwenhuijse M J, Jasper J, Fiocco M, Plevier J W, Middeldorp S, Valstar E R, Nelissen R G. Early subsidence of shapeclosed hip arthroplasty stems is associated with late revision: a systematic review and meta-analysis of 24 RSA studies and 56 survival studies. Acta Orthop 2015; 86(5): 575-85. von Schewelov T, Carlsson A, Sanzen L, Besjakov J. Continuous distal migration and internal rotation of the C-stem prosthesis without any adverse clinical effects: an RSA study of 33 primary total hip arthroplasties followed for up to ten years. Bone Joint J 2014; 96-b(5): 604-8.

Yates P J, Burston B J, Whitley E, Bannister G C. Collarless polished tapered stem: clinical and radiological results at a minimum of ten years’ followup. J Bone Joint Surg Br 2008; 90(1): 16-22.

Acta Orthopaedica 2019; 90 (3): 243–248

243

Repeated cobalt and chromium ion measurements in patients with large-diameter head metal-on-metal ReCap-M2A-Magnum total hip replacement Heikki MÄNTYMÄKI 1, Petteri LANKINEN 2, Tero VAHLBERG 3, Aleksi REITO 4, Antti ESKELINEN 4, and Keijo MÄKELÄ 2 1 Department

of Orthopaedics, Tampere University Hospital and University of Turku; 2 Department of Orthopaedics and Traumatology, Turku University Hospital and University of Turku; 3 Department of Biostatistics, University of Turku; 4 Coxa Hospital for Joint Replacement and University of Tampere, Finland Correspondence: heikki.mantymaki@pshp.fi Submitted 2018-10-24. Accepted 2019-02-08

Background and purpose — Whole blood (WB) cobalt (Co) and chromium (Cr) ion levels have a major role in the follow-up of metal-on-metal total hip replacement (MoM THR). We investigated, first, if there was a change in WB Co or Cr levels over repeated measurements in patients with ReCap-M2A-Magnum THR, and, second, determined how many patients had WB Co or Cr levels that exceeded the safe upper limits (SUL) in the repeated whole blood metal ion assessment.

Patients and methods — A Recap-M2A-Magnum THR was used in 1,329 operations (1,188 patients) at our institution between 2005 and 2012. We identified all patients (n = 319) with unilateral ReCap-M2A-Magnum implants who had undergone at least 2 repeated metal ion measurements with the first blood sample taken mean 5.5 years (1.8–9.3) after surgery and the second taken mean 2 years (0.5–3) after the first.

Results — The median WB Co and Cr ion levels decreased in repeated measurements from 1.40 (0.40–63) ppb to 1.10 (0.20–68) ppb and from 1.60 (0.60–13.0) ppb to 1.10 (0.30–19.0) ppb, respectively. 7% of the Co ion values exceeded SUL at the initial measurement, and 7% at the control measurement. The proportion of Cr ion values exceeding the safe upper limit (SUL) decreased during the measurement interval from 5% to 4%.

Interpretation — Repeated metal ion measurements in unilateral ReCap-M2A-Magnum patients in a mean 2-year time interval did not show any increase. Long-term ion levels are, however, not yet known.

Metal debris-related local reaction (adverse reaction to metal debris, ARMD) is a well-known complication of largediameter head metal-on-metal total hip replacement (MoM THR). More than 1 million MoM THRs were performed before widespread concerns about ARMD were raised (Lombardi et al. 2012). The Australian Orthopaedic Association’s National Joint Replacement Registry (AOANJRR) was the first to report early failures of MoM THR (AOANJRR 2008).

Despite the AOANJRR’s report, it took approximately 4 years for the European orthopedic community to adequately react to the issue (MHRA 2012). In Finland, the Finnish Arthroplasty Society recommended, in 2012, not to continue implantation of MoM THRs (Finnish Arthroplasty Society 2015).

ARMD is the most frequent cause of revision surgery among patients with large-diameter head MoM THR (Finnish Arthroplasty Register n.d.). Despite the implantations of MoM THRs having ceased, there are a large number of patients with MoM THR still in situ and these patients require regular followup. Earlier studies have shown the relationship between high whole blood (WB) metal ion levels and failure of the MoM THR (Hart et al. 2014). Therefore, WB metal ion assessments have been used to detect ARMD, first to predict the failure of the implant and second to evaluate patients’ metal ion burden (Hannemann et al. 2013, Finnish Arthroplasty Society 2014).

It is recommended that patients with large-diameter head MoM THR should be regularly monitored with clinical examination, and when necessary with metal ion measurements and MARS-MRI (magnetic artifact reduction sequence-MRI) (Medicines and Healthcare products Regulatory Agency 2012,

© 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group, on behalf of the Nordic Orthopedic Federation. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits ­unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

DOI 10.1080/17453674.2019.1595469

244

Health Canada 2012, European Federation of National Associations of Orthopaedics and Traumatology [Effort] 2012).

However, it is unclear for how long and how often (interval) patients with a large-diameter head MoM THR should be screened. There are reports of repeated metal ion measurements in a few MoM implants (van Der Straeten et al. 2013b, Reito et al. 2014), which have shown that it is useful to perform regular metal ion measurements. The tribology and failure rate of different designs of MoM THR prosthesis varies (AOANJRR 2016). Therefore, studies with different implant designs are needed. As far as we know, there are no study reports on repeated whole blood metal ion measurements after implantation of the large-diameter head ReCap-M2A-Magnum THR. The primary aims of our study were therefore to investigate: 1) if there is a substantial change in whole blood Co or Cr levels in repeated measurements performed a mean 24 months (7 to 36) after the initial measurement in patients operated on with ReCap-M2A-Magnum THR; and 2) what proportion of patients with unilateral ReCap-M2AMagnum THR have whole blood Co or Cr levels exceeding the safe upper limits in a mean 2-year time interval in the repeated measurements (chromium (Cr) 4.6 ppb, cobalt (Co) 4.0 ppb) (van der Straeten et al. 2013a).

Patients and methods We established a screening program at our institution for MoM THR to identify patients with ARMD. The screening was done according to the follow-up protocol recommended by the Finnish Arthroplasty Society (2014). The screening included an Oxford Hip Score (OHS) questionnaire, anteroposterior and lateral radiographs of the hip, and whole blood (WB) Cr and Co ion concentration measurements. Patients with moderate or poor OHS score, and/or patients with WB Cr or Co concentration > 5 ppb were referred for MRI using magnetic artifact reduction sequence (MARS). These patients were also clinically examined by a senior orthopedic surgeon at our outpatient clinic. Revision surgery for ARMD was considered if the patient had severe hip symptoms, such as pain, clicking, and swelling, and there was a clear pseudotumor on MRI. Revision surgery was also considered if an asymptomatic patient had very high WB metal ion levels (> 10 ppb) to avoid symptoms of Co poisoning (Rizzetti et al. 2009). All patients who were not revised were scheduled for annual or biennial repeat visits. Borderline cases were evaluated more frequently.

For this study, we identified all patients with unilateral ReCap-M2A-Magnum implants (1,047 patients). The stem used was the Biomet Bi-Metric (Biomet Orthopedics Inc, Warsaw, IN, USA). From the 264 patients without any ion measurements, 70 had been revised and 85 had died. Of these, 7 patients had been revised first and had died afterwards (Figure 1). The rest of these patients with a unilateral

Acta Orthopaedica 2019; 90 (3): 243–248

Patients operated on with M2A-ReCap-Magnum THR n = 1,188 patients with 1,329 hips Excluded Bilateral M2A-ReCap-Magnum THR n = 141 patients with 282 hips Patients with unilateral M2A-ReCap-Magnum THR n = 1,047 Excluded (n = 728): – revised, dead or lost to prior screening, 264 – attended only 1 follow-up visit, 447 – met exclusion criteria, 8 – other MoM implant in contralateral hip, 9 Patients included in the study n = 319

Figure 1. Flow chart of the study.

ReCap-M2A-Magnum implant without any ion measurements were lost to follow-up. 783 patients made the first follow-up visit including WB metal ion measurements. From the 111 patients without the second follow-up visit 12 patients had been revised, 3 had died, and 96 were lost to follow-up (Figure 1). 336 patients (336 hips) had undergone 2 follow-up visits.

9 patients had bilateral MoM hip devices and were excluded.

All participating patients had their blood samples taken from the antecubital vein using a 21-gauge BD Vacutainer Eclipse blood collection needle (Becton, Dickinson and Co, Franklin Lakes, NJ, USA). The first 10 mL tube of blood was used for analysis of standard laboratory tests such as C-reactive protein and erythrocyte sedimentation rate measurement. The second blood sample was taken in Vacuette NH trace elements tube (Greiner Bio-One GmbH, Kremsmünster, Austria) containing sodium heparin. Cobalt and chromium analyses from whole blood were performed using an accredited method with Inductively Coupled Plasma Mass Spectrometry (ICP-MS, VITA Laboratory, Helsinki, Finland in collaboration with Medical Laboratory of Bremen, Germany). The detection limit for Cr was 0.2 ppb and for Co 0.2 ppb. The intra-assay variation for WB Cr and Co was 2.2% and 2.7% and inter-assay variation was 6.7% and 7.9%, respectively.

Statistics 319 patients met the criteria for this study with at least 2 repeated metal ion measurements. The mean time elapsing from the first metal ion assessment (initial measurement) to the second (control measurement) was 2.0 years (SD 0.5, range 0.6–3.0). All unilateral ReCap-M2A-Magnum patients operated on at our institution are considered here as the control group, whereas those patients with 2 WB ion measurements are referred to as the study group. A 2-sample t-test was used to test the difference in age, inclination angle, and femoral head diameter between the control and study groups.

Sex distribution was compared using a chi-square test. Demographics were similar between the groups (Table 1).

Acta 2019; 90 Acta Orthopaedica Orthopaedica 2019; 90 (3): (3): 243–248 243–248

245 245

Table 1. Comparison of demographic variables between the control group (= overall unilateral ReCap Magnum THR group, n = 1,047) and the study group (n = 319) Study group Female patients (%) Age (SD) Median femoral head diameter (SD), mm Mean acetabular inclination (SD)

Whole blood Co concentration (ppb) p = 0.6

p = 0.03

p < 0.0001 p < 0.0001

59 64 (9)

55 65 (10)

0.2 0.2

49 (4) 43 (7)

49 (4) 43 (8)

0.4 1

Whole blood Cr concentration (ppb) p = 0.5

p = 0.004

2.0

2.0

1.6

1.6

1.2

1.2

0.8

0.8

p = 0.0001

p < 0.0001

p < 0.0001

p = 0.007

0.4

0.4

First measurement Second measurement

First measurement Second measurement

0

Control group p-value

2 n=6

3 n = 29

4 n = 55

5 n = 138

6 n = 91

Years to first measurement

0

2 n=6

3 n = 28

4 n = 55

Table 2. Differences in WB CO and CR levels (ppb)

WB Co, n = 319 median geometric mean WB Cr, n = 317 median geometric mean

6 n = 90

Years to first measurement

Control

p-value

1.4 (0.4–63) 1.5

1.1 (0.2–68) 1.2

< 0.001

1.6 (0.6–13) 1.7

1.1 (0.3–19) 1.2

< 0.001

Frequency

Frequency 60

60

50

50

40

40

30

30

20

20

10

10

0

5 n = 138

Initial

< –5.0 –5.0 to –3.1

–3.0 to –1.1

–1.0 to –0.1

0.0 to 1.0

1.1 to 3.0

3.1 to 5.0

> 5.0

Change (ppb) between Co measurements

0

< –5.0 –5.0 to –3.1

–3.0 to –1.1

–1.0 to –0.1

0.0 to 1.0

1.1 to 3.0

3.1 to 5.0

> 5.0

Change (ppb) between Cr measurements

Figure 2. Geometric mean whole blood Co values (left) and Cr levels (right) divided across the follow-up time before initial measurement.

Figure 3. Frequency distribution of change in whole blood Co levels (left) and Cr levels (right) compared with the initial measurement.

The time elapsing from the index operation to the first metal ion measurement (initial) is referred to as follow-up time. Mean follow-up time between the index operation and the first metal ion measurement was 5.5 years (range 1.8 to 9.3 years). Patients were divided into follow-up time interval groups according to the time elapsing from the index operation to the first metal ion assessment. The time elapsing from the first metal ion measurement (initial) to the second measurement (control) in the same patient is referred to as the measurement interval. Thus total follow-up is defined as follow-up time plus measurement interval. The individual change in 2 consecutive metal ion measurements from the same patients was modelled using a random coefficient model. Log-transformed ion values were used in conditional models due to positively skewed distribution of ion levels.

Results are expressed as geometric means for better interpretability. SUL values for WB Co were 4.0 ppb and for WB Cr 4.6 ppb as reported earlier (van der Straeten et al. 2013a).

P-values lower than 0.05 in a 2-tailed test were considered statistically significant.

The change over a 2-year measurement interval was calculated and plotted as frequency distributions for both metal ions separately.

Arthroplasty Association (Finnish Arthroplasty Society 2014).

It was a register study, and the patients were not directly contacted. Therefore, approval by the local ethical committee was not needed. This study was financially supported by the Competitive State Research Financing of the Expert Responsibility area of Tampere University Hospital. Outside this study, HM has received travel/accommodation expenses from DePuy Synthes. AE received research funding from Zimmer Biomet and DePuy Synthes and consultancy fees from Zimmer Biomet. AR reports personal fees from a paid lecture. PL, KTM, and TV have nothing to disclose. No benefits in any form have been received related directly or indirectly to this article.

Ethics, funding, and potential conflicts of interest The study was based on the national recommendation for systematic screening of MoM THR patients given by the Finnish

Results There was a statistically significant decrease in repeated WB Co and Cr values (Table 2).

Geometric mean of WB Co and Cr levels did not change in the 2-year follow-up group. However, there were only 6 measurements in the 2-year follow-up group. Geometric mean of WB Co and Cr values showed statistically significant decrease in the 3- to 6-year follow-up groups (Figure 2).

Both WB Co and Cr concentrations remained within ±1 ppb of their initial value in most patients (86% for Co, 81% for Cr), with no trends towards increasing values (Figure 3).

246

Acta Orthopaedica 2019; 90 (3): 243–248

Change in Co concentration (ppb)

Change in Cr concentration (ppb)

Log Co concentration (ppb)

Log Cr concentration (ppb)

20

20

5

3.0

15

15

10

10

5

5

0

2.5

4

2.0 3

1.5

2

1.0

1

0.5

0

–5

–5

–10

–10

–15

–15

–1

–20

–20

–2

0

10

20

30

40

50

60

70

Initial Co concentration (ppb)

0.0

0

–0.5

0

2

4

6

8

10

12

14

Initial Cr concentration (ppb)

–1.0 –1.5

First

Measurement

Second

First

Measurement

Second

Figure 4. Changes in Co ion levels (left) and Cr ion levels (right) compared with the initial measurement.

Figure 5. Spaghetti plots for individual Co and Cr values at initial and control measurements. Values are naturally log-transformed.

6.6% of the Co ion measurements exceeded SUL at the initial measurement. The proportion increased slightly, being 7% at the control measurement. The proportion of SUL exceeding Cr ion levels decreased during the measurement interval from 5% to 4%.

The Co and Cr levels decreased over time and stayed mostly below the SUL if the initial value was low. The exceptions were those with high values already at the start (Figure 4).

The cobalt value increased from safe value to value above the safe limit in 8 patients, whereas the chromium value increased from safe value to value above the safe limit in 6 patients.

Spaghetti plots for individual Co and Cr values at initial and control measurements are presented in Figure 5. Values are naturally log-transformed.

had been revised and because of that did not undergo a second metal ion measurement. Most patients with very high WB ion levels initially had also been re-operated to avoid cobalt poisoning, and were not included. Therefore our study patients included mostly those with relatively low initial WB ion values. The progress of WB metal ion concentrations in patients with very high initial values is not known. Our findings are not generalizable to other MoM devices.

From the 111 patients with only 1 follow-up visit, 12 had been revised and 3 had died. We acknowledge that there were many patients with only 1 metal ion measurement. However, this is about the same proportion as in the work of Reito et al.

(2016b). Some of the patients were elderly with remarkable comorbidity and they may have chosen not to come for the repeated measurements even though the possibility was provided. In Finland the healthcare system refers patients mostly to the university hospital in their own district.

The group-level results may not be relevant from a single patient perspective. For the patient, it is more relevant to know if the metal level in his/her blood is high or not, what the expected change is in a repeated measurement, and which levels will raise concern. Even if only 5% of all patients have dangerously high blood levels it may be worth measuring the blood of all patients a second time or more. Therefore, we assessed our data additionally by modelling the individual change. However, also on individual level, the increase in ion levels on repeated measurements was rare.

The Medicines and Healthcare products Regulatory Agency (MHRA) and Health Canada have recommended a cutoff level of serum cobalt and chromium of 7 ppb (MHRA 2012, Health Canada 2012). MHRA even recommends repeated testing within 3 months of abnormal results. European guidelines suggest that ion concentrations between 2 ppb and 7 ppb are of concern (EFORT 2012). The US Food and Drug Administration in the USA and Therapeutic Goods Administration in Australia do not state any cut-off ion concentration thresholds (US FDA 2013, TGA 2012). According to Hart et al. (2011) a cut-off level of 7 ppb shows good specificity, but relatively low sensitivity. Lardanchet et al. (2012) suggested a cobalt cut-off

Discussion We found that median or geometric mean WB Co and Cr levels in repeated metal ion measurements in unilateral ReCap-M2A-Magnum patients in a mean 2-year time interval did not show notable increase. Long-term ion levels are, however, not yet known.

A limitation of our study was that the inclusion criterion used was arbitrary. We aimed to study changes in WB metal ion levels by repeated measurements, and the practical measurement interval was 2 years. The time frame from the first measurement to the second was not constant, however, in our patients. Therefore, we were compelled to select a time range, and 7 to 36 months (mean 2 years) was deemed most suitable.

It is possible that a longer time range between the measurements such as 5 or 10 years might give different results. The long-term implant survivorship of the M2A-ReCap-Magnum THR, or long-term metal ion values of these patients, are not yet known. Further research is needed prior to determining whether a metal ion screening program of patients with this device should be ceased.

Another limitation of our study was also that most patients with severe hip symptoms and a clear pseudotumor in MRI

Acta Orthopaedica 2019; 90 (3): 243–248

level of 8 ppb. Van der Straeten et al. (2013a) defined SUL for unilateral hip resurfacing (HR) patients at Cr 4.6 ppb and Co 4.0 ppb, and for bilateral HR patients at Cr 7.4 ppb and Co 5.0 ppb. For our study purposes we decided to use the cut-off levels suggested by van der Straeten et al. (2013a). We do not think using other cutoff levels as SUL would change our message.

Of all MoM devices, metal ion levels of ASR (DePuy, Warsaw, IN, USA) THR and HR have been scrutinized most thoroughly. Reito and co-workers (2014) assessed 254 unilateral patients, of whom 156 had received an ASR XL THR and 98 patients an ASR HR (n = 254). The second blood sample was taken 8 to 16 months after the first. In the majority of HR patients both WB Cr and Co concentrations remained within ±1 ppb in the second measurement and the majority of the values also remained below the SUL. However, in the THR group there was a significant increase in WB Co levels over the measurement interval and 32% of the patients exceeded the SUL during the measurement interval. They concluded that it is useful to perform regular WB metal ion measurements in ASR XL THR patients, although not in ASR HR patients (Reito et al. 2014). In the current study we were not able to reproduce this finding in unilateral ReCap Magnum THR patients. The measurement interval of our study was even longer (mean 2 years) than in the study of Reito et al.

(2014), and the decreasing tendency of WB ion levels was clear. The poorer performance of the ASR device may explain the difference in WB ion level development compared with the ReCap Magnum THR (Seppänen et al. 2018). Our data did not include any ReCap HR devices.

In another study Reito et al. (2016b) showed that there is also a substantial increase in repeated Co and Cr level measurements in patients with bilateral ASR THR, but not with HR patients. Additionally, 21% of THR patients had WB Co ion levels already exceeding the SUL in the first measurement (Reito et al. 2016b). Our current study did not include bilateral procedures, so these previous findings could not be verified using ReCap Magnum THRs. Further research is needed concerning bilateral ReCap Magnum devices to assess WB ion level development tendency.

Matharu et al. (2015) have previously raised concern regarding variable protocols worldwide in MoM THR screening. They have suggested further research to clarify blood metal ion thresholds, and whether thresholds differ between implants (Matharu et al. 2015). Our current results strengthen this impression. Implant-specific thresholds seem to be more effective to detect ARMD (Matharu et al. 2016a, 2016b, 2017).

We might need specific thresholds for different implants to have better sensitivity and specificity in ARMD screening.

Our findings imply that patients with unilateral M2AReCap-Magnum THR with WB metal ion levels below the SUL do not benefit from routine metal ion level screening, at least in a mean 2-year interval. We are aware that these findings cannot be generalized to other LDH MoM THR brands.

Due to the previous studies of repeated metal ion measure-

247

ments in ASR, the universal protocol may not be sufficient for all THR designs (Reito et al. 2016a). Implant-specific thresholds might be needed in the future to detect ARMD in different THR designs. Further, we do not know how the WB metal ion levels develop in the long term in unilateral ReCapMagnum patients. Wear and corrosion of the bearing surface and the trunnion may well increase in long-term follow-up.

Further research is needed to assess the long-term benefits of WB ion measurements and to determine the specific thresholds to detect ARMD in M2A-ReCap-Magnum THR patients.

Further research is also needed to determine how frequently WB metal ion concentrations needs to be evaluated in these patients.

AR and AE designed the protocol and methods. KTM performed the surgery and recorded the intraoperative data. TV analyzed the data and did the statistics. PL, TV, and HM collected the data. PL, KTM, TV, and HM wrote the manuscript. All authors contributed to the revision of the manuscript.

Acta thanks Harald Brismar and Jan A N Verhaar for help with peer review of this study.

Australian Orthopaedic Association National Joint Replacement Registry (AOANJRR): Annual Report 2008. Adelaide, S. Australia: Australian Orthopaedic Association; 2008. https://aoanjrr.sahmri.com/documents/10180/42662/Annual%20Report%202008?version%20 = 1.1&t = 1349406277970 Australian Orthopaedic Association National Joint Replacement Registry (AOANJRR): Annual Report 2016. Adelaide, S. Australia: Australian Orthopaedic Association; 2016. https://aoanjrr.sahmri.com/annualreports-2016 European Federation of National Associations of Orthopaedics and Traumatology (EFORT), European Hip Society (EHS), Arbeitsgemeinschaft Endoprothetik (AE) and Deutsche Arthrosehilfe (DAH). Consensus statement.

2012. “Current evidence on the management of metal-on-metal bearings.” https://www.efort.org/wp-content/uploads/2013/10/2012_05_10_MoM_ Consensus_statement1.pdf Finnish Arthroplasty Register (FAR). Available at: www.thl.fi/far Finnish Arthroplasty Society. MoM hip arthroplasty follow up and interpretation of follow up results November 17, 2014. http://www.suomenartroplastiayhdistys.fi/files/faa_mom_fu_recommendations.pdf Finnish Arthroplasty Society. Recommendation for the use and follow-up of patients with metal-on-metal (MoM) hip arthroplasty 2015. http://www. suomenartroplastiayhdistys.fi/index.php?page = 1074&lang = 1.

Hannemann F, Hartmann A, Schmitt J, Lützner J, Seidler A, Campbell P, Delaunay C P, Drexler H, Ettema H B, García-Cimbrelo E, Huberti H, Knahr K, Kunze J, Langton D J, Lauer W, Learmonth I, Lohmann C H, Morlock M, Wimmer M A, Zagra L, Günther KP. European multidisciplinary consensus statement on the use and monitoring of metal-on-metal bearings for total hip replacement and hip resurfacing. Orthop Traumatol Surg Res 2013; 99: 263–71.

Hart A J, Sabah S A, Bandi A S, Maggiore P, Tarassoli P, Sampson B, Skinner J A. Sensitivity and specificity of blood cobalt and chromium metal ions for predicting failure of metal-on-metal hip replacement. J Bone Joint Surg Br 2011; 93(10): 1308–13.

Hart A J, Sabah S A, Sampson B, Skinner J A, Powell J J, Palla L, Pajamäki K J, Puolakka T, Reito A, Eskelinen A. Surveillance of patients with metalon-metal hip resurfacing and total hip prostheses: a prospective cohort study to investigate the relationship between blood metal ion levels and implant failure. J Bone Joint Surg Am 2014; 96:1091–9.

248

Health Canada, Government of Canada Metal-on-metal hip implants—information for orthopaedic surgeons regarding patient management following surgery—for health professionals; 2012. http://healthycanadians.gc.ca/ recall-alert-rappel-avis/hc-sc/2012/15835a-eng.php Lardanchet J F, Taviaux J, Arnalsteen D, Gabrion A, Mertl P. One-year prospective comparative study of three large-diameter metal-on-metal total hip prostheses: serum metal ion levels and clinical outcomes. Orthop Traumatol Surg Res 2012; 98(3): 265-74.

Lombardi A V Jr, Barrack R L, Berend K R, Cuckler J M, Jacobs J J, Mont M A, Schmalzried T P. The hip society: algorithmic approach to diagnosis and management of metal-on-metal arthroplasty. J Bone Joint Surg Br 2012; 94(11 Suppl. A): 14–18.

Matharu G S, Mellon S J, Murray D W, Pandit H G. Follow-up of metalon-metal hip arthroplasty patients is currently not evidence based or cost effective. J Arthroplasty 2015; 30: 1317–23.

Matharu G S, Berryman F, Brash L, Pynsent P B, Dunlop D J, Treacy R B. Can blood metal ion levels be used to identify patients with bilateral Birmingham Hip Resurfacings who are at risk of adverse reactions to metal debris? Bone Joint J 2016a; 98-B: 1455-62.

Matharu G S, Berryman F, Brash L, Pynsent P B, Treacy R B, Dunlop D J. The effectiveness of blood metal ions in identifying patients with unilateral Birmingham Hip Resurfacing and Corail-Pinnacle metal-on-metal hip implants at risk of adverse reactions to metal debris. J Bone Joint Surg Am 2016b; 98: 617–26.

Matharu G S, Berryman F, Judge A. Blood metal ion thresholds to identify patients with metal-on-metal hip implants at risk of adverse reactions to metal debris: an external multicenter validation study of Birmingham Hip Resurfacing and Corail-Pinnacle implants. J Bone Joint Surg Am 2017; 99: 1532-9.

Medicines and Healthcare products Regulatory Agency (MHRA).

Medical device alert: All metal-on-metal (MoM) hip replacements (MDA/2012/036), 2012. https://assets.publishing.service.gov.uk/media/ 5485abf6ed915d4c10000273/con155767.pdf Reito A, Moilanen T, Puolakka T, Pajamäki J, Eskelinen A. Repeated metal ion measurements in patients with high risk metal-on-metal hip replacement. Int Orthop 2014; 38(7): 1353-61.

Acta Orthopaedica 2019; 90 (3): 243–248

Reito A, Lainiala O, Elo P, Eskelinen A. Prevalence of failure due to adverse reaction to metal debris in modern, medium and large diameter metal-onmetal hip replacements—the effect of novel screening methods: systematic review and metaregression analysis. PLoS One 2016a Mar 1; 11(3): e0147872. Review.

Reito A, Lainiala O, Nieminen J, Eskelinen A. Repeated metal ion measurement in patients with bilateral metal on metal (ASR™) hip replacements.

Orthop Traumatol Surg Res 2016b; 102(2): 167-73.

Rizzetti M C, Liberini P, Zarattini G, Catalani S, Pazzaglia U, Apostoli P, Padovani A. Loss of sight and sound. Could it be the hip? Lancet 2009; 373(9668): 1052.

Seppänen M, Laaksonen I, Pulkkinen P, Eskelinen A, Puhto A P, Kettunen J, Leskinen J, Manninen M, Mäkelä K. High revision rate for large-head metal-on-metal THA at a mean of 7.1 years: a registry study. Clin Orthop Relat Res 2018; 476(6): 1223-30.

NJR Editorial Board. National Joint Registry for England, Wales and Northern Ireland 13th Annual Report 2014. http://www.njrcentre.org.uk/ njrcentre/Portals/0/Documents/England/Reports/13th%20Annual%20 Report/07950%20NJR%20Annual%20Report%202016%20ONLINE%20 REPORT.pdf Therapeutic Goods Administration, Department of Health, Australian Government Metal-on-metal hip replacement implants — information for general practitioners, orthopaedic surgeons and other health professionals, 2012. http://www.tga.gov.au/hp/information-devices-mom-hip-implants.htm US Food and Drug Administration. Medical devices: metal-on-metal hip implants, Information for Orthopaedic Surgeons, 2013. http://www.fda. gov/MedicalDevices/ProductsandMedicalProcedures/ImplantsandProsthetics/MetalonMetalHipImplants/ucm241667.htm Van Der Straeten C, Grammatopoulos G, Gill H S, Calistri A, Campbell P, De Smet K A. The 2012 Otto Aufranc Award: The interpretation of metal ion levels in unilateral and bilateral hip resurfacing. Clin Orthop Relat Res 2013a; 471(2): 377-85.

Van Der Straeten C, Van Quickenborne D, De Roest B, Calistri A, Victor J, De Smet K. Metal ion levels from well-functioning Birmingham Hip Resurfacings decline significantly at ten years. Bone Joint J. 2013b; 95-B(10): 1332-8.

Acta Orthopaedica 2019; 90 (3): 249–252

249

Implant survival and patient-reported outcome following total hip arthroplasty in patients 30 years or younger: a matched cohort study of 1,008 patients in the Swedish Hip Arthroplasty Register Maziar MOHADDES 1,2, Emma NAUCLÉR 1, Johan KÄRRHOLM 1,2, Henrik MALCHAU 1,2, Daniel ODIN 1, and Ola ROLFSON 1,2 1 Swedish Hip Arthroplasty Register, Gothenburg, Sweden; 2 Department of Orthopaedics, Institute of Clinical Sciences, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden Correspondence: maziar.mohaddes@gmail.com Submitted 2018-11-30. Accepted 2019-02-14.

Background and purpose — The outcome of total hip arthroplasty (THA) in younger patients is suggested to be inferior compared with the general THA population. There is a lack of studies with long-term follow up for very young patients. We report on implant survival and patient-reported outcome in patients aged 30 years or younger.

Patients and methods — Data on THAs performed in Sweden between the years 2000 and 2016 were included.

There were 504 patients 30 years or younger with complete demographic and surgical data (study group). A matched comparison group older than 30 years was identified. Implant survival was analyzed using the Kaplan–Meier method.

Patient-reported outcome was analyzed in a subgroup of patients.

Results — The 10-year and 15-year implant survivorship for the study group was 90% and 78%, respectively. The corresponding figures for the patients older than 30 years were 94% and 89%. The median preoperative EQ-5D index was lower in the study group; the improvement in EQ-5D index was similar between the study and the comparison groups.

The preoperative EQ-VAS was lower and the improvement in EQ-VAS at 1 year was larger in the study group.

Interpretation — The promising 10-year implant survival and 1-year improvement in patient-reported outcome suggests that THA is a feasible option in the patients 30 years or younger.

The outcome of total hip arthroplasty (THA) in the younger population is suggested to be inferior compared to the THA general population (AOANJRR 2018, Bayliss et al. 2017, Kärrholm et al. 2018, NJR 2018). For example, the lifetime risk of having a revision following THA for patients aged 50 to 54 years at primary surgery is reported to be 17% and 30% for females and males respectively (Bayliss et al. 2017). In a systematic review, Walker et al. (2016) in a meta-analysis on patients aged 30 years or younger found a revision rate of 5% with a mean follow-up of 8 years. They highlighted lack of register studies with long-term follow-up for the very young patients. The sparse long-term reports and a belief in inferior outcomes in younger patients might create difficulties for surgeons and patients when deciding whether THA is a feasible option.

We analyzed long-term implant survival and patientreported outcomes at 1 year in patients aged 30 years or younger, registered in the Swedish Hip Arthroplasty Register.

A propensity-score-matched group of patients older than 30 years were included for comparison.

Patients and methods The Swedish Hip Arthroplasty Register (SHAR) is a national register with full coverage. In the annual report for 2017, the register reported a completeness of 98% for primary THAs. (Kärrholm et al. 2018).

We identified all primary THAs in individual patients aged 30 years or younger, operated on between January 2000 and

© 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group, on behalf of the Nordic Orthopedic Federation. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits ­unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

DOI 10.1080/17453674.2019.1599776

250

Acta Orthopaedica 2019; 90 (3): 249–252

Primary THA in SHAR 2000–2016 n = 208,316

≤ 30 years old n = 687 Excluded diagnoses (n = 75): – tumor, 14 – fracture/trauma, 56 – other, 5 Excluded due to other reasons a n = 108 ≤ 30 years old included in the study n = 504

> 30 years old n = 207,629 Excluded diagnoses (n = 28,496): – tumor 1,351 – fracture/trauma, 26,959 – other, 184

Ethics, funding, and potential conflicts of interest This study is a part of a larger research project which has been reviewed and approved by the Regional Ethical Review Board in Gothenburg (2014-04-09, 271-14). The study was not financed by any external funding.

The authors declare no conflicts of interest.

Excluded due to other reasons a n = 9,272 > 30 years old included in the study n = 169,861

Results

The average age of the patients in the study and the comparison group at the time of priFigure 1. Patient selection. Numbers given in this figure are patients having a THA. In mary surgery was 25 (SD 4) years and 54 a bilaterally operated cases only the first hip was included. Excluded: metal-on-metal (SD 13) years respectively. articulation, femoral head size > 36 mm or data missing on fixation, femoral head size or articulation.

60% were women in both groups (Table 1). Inflammatory joint disease and osteoDecember 2016 as reported to the SHAR. In patients operated arthritis (OA) following childhood disease were the most bilaterally only the first hip operated was included. Surger- common indications for surgery. Uncemented fixation was ies performed due to a fracture (n = 56), tumor (n = 14), or most common, followed by reversed hybrids in both study and with unspecific diagnosis (n = 5) were excluded. Further, all comparison group (Table 1). The average follow-up was 8 (SD surgeries performed with a metal-on-metal hip replacement 5) years in both groups. The average time from primary THA and those with missing data on fixation, femoral head size, and articulation (n = 108) were excluded. The study group Table 1. Demographic and surgical data. Numbers are given as n consisted of 504 patients 30 years or younger. During the (%) unless otherwise stated

same period, 207,629 surgeries performed in patients older than 30 years had been reported to the register (Figure 1).

Demographics ≤ 30 years > 30 years For comparison a propensity-score-matched group older than and surgical data (n = 504) (n = 504) 30 years (n = 504) was included. Patient-reported outcomes Women 300 (60) 300 (60) were reported to SHAR on a national basis from 2008. The Age, mean (SD) 25 (4) 54 (13) end of the study was defined as March 23, 2018 (the date Diagnosis Primary OA 50 (10) 49 (10) when the data repository was created), revision or death, Inflammatory arthritis 119 (24) 123 (24) whichever occurred first. Primary outcome was implant surOA following childhood disease 120 (24) 119 (24) vival at 15 years. Secondary outcomes were implant survival Avascular necrosis 73 (15) 73 (15) Other 142 (28) 140 (28) at 10 years and patient-reported outcomes pre- and at 1 year Year of operation postoperatively.

Statistics Normally distributed data are reported as mean (SD). Nonnormally distributed data are presented with median (IQR).

The propensity score matching was done using sex, diagnosis, implant fixation, articulation, year of operation, and femoral head size. A 1:1 nearest-neighbor matching using logistic regression was applied to estimate the propensity scores.

Non-parametric testing was performed to compare the patientreported outcomes. Statistical significance was set at p < 0.05.

Unadjusted Kaplan–Meier survival analysis was performed to analyze implant survival with revision, defined as exchange or removal of parts or all of the implant. Survival data are presented as percentage and 95% confidence interval (CI). In the study group (30 years or younger) a univariable Cox regression analysis was performed to study the influence of fixation, articulation, and femoral head size on risk of revision.

2000–2004 133 (26) 144 (29) 2005–2009 115 (23) 109 (22) 2010–2014 170 (34) 167 (33) 2015–2016 86 (17) 84 (17) Fixation Cemented 75 (15) 76 (15) Hybrid 35 (7) 36 (7) Reversed hybrid 80 (16) 73 (15) Uncemented 314 (62) 319 (63) Articulation Ceramic on ceramic 131 (26) 141 (28) Ceramic on polyethylene 22 (4) 22 (4) Ceramic on polyethylene (x-linked) 6 (1) 5 (1) Metal on polyethylene 291 (58) 285 (57) Metal on polyethylene (x-linked) 54 (11) 51 (10) Femoral head size < 28 mm 62 (12) 54 (11) 28 mm 233 (46) 239 (47) 32 mm 161 (32) 168 (33) 36 mm 48 (10) 43 (9) Follow-up, mean (SD) 8 (5) 8 (5) OA = osteoarthritis, x-linked = highly cross-linked.

Acta Orthopaedica 2019; 90 (3): 249–252

251

Table 3. Patient-reported outcomes. Values are median (IQR) unless otherwise stated

EQ-5D-3L-index, n preoperative 1 year postoperative delta EQ-VAS, n preoperative 1 year postoperative delta

≤ 30 years 139 0.09 (–0.02 to 0.59) 0.70 (0.60 to 1.00) 0.51 (0.18 to 0.76) 153 44 (30 to 61] 80 (60 to 90) 29 (10 to 45)

> 30 years 198 0.16 (0.06 to 0.69) 0.80 (0.70 to 1.00) 0.36 (0.17 to 0.71) 219 50 (35 to 70) 80 (65 to 90) 20 (3 to 37)

p-value 0.003 0.04 0.1 0.006 0.3 0.02

Due to a migration from EQ-5D 3 level to EQ-5D 5 level, there was a discrepancy in the number of patients with valid EQ-5D-3L index and EQ-VAS.

Figure 2. Survival probability with implant revision as end-point.

Age

0

Number at risk after index operation (years) 3 6 9 12

s15

≤ 30 504 421 325 212 136 66 > 30 504 416 307 195 128 70

Table 2. Reason for revision and type of revision performed

Reason/type

≤ 30 years (n = 53)

Reason for revision, n Aseptic loosening 34 Dislocation 1 Fracture 0 Infection 6 Others 9 Type of revision, n Cup + stem exchange 9 Cup exchange 22 Extraction 6 Femoral head exchange 1 Liner ± head exchange 12 Stem exchange 3

> 30 years (n = 29) 14 3 1 7 3 4 13 3 1 5 3

to revision was 8 (SD 5) years in the study group and 6 (SD 5) years in the comparison group.

The 15- and 10-year implant survivorship for patients aged 30 years or younger was 78% (CI 71–84) and 90% (CI 87–94) respectively. The corresponding figures for the comparison group were 89% (CI 84–93) and 94% (CI 91–97) (Figure 2).

53 out of 504 patients in the study group had been revised, and the most common reason for revision was aseptic loosening (n = 34). Corresponding figures were 29 and 14 in the comparison group (Table 2). Applying the Cox regression, the risk of revision in the study group was lower if uncemented fixation had been used (HR: 0.3, CI 0.2–0.7) when compared with cemented fixation. The limited number of events in different

strata with regards to articulation and head size did not allow for meaningful statistical analysis.

Median (IQR) preoperative EQ-5D index was 0.09 (–0.02– 0.59) and 0.16 (0.06–0.69) in the study and the comparison group respectively (p = 0.003). The corresponding figures 1 year postoperatively were 0.7 (0.6–1.0) and 0.8 (0.7–1.0) (p = 0.04). In the study group the median preoperative EQ-VAS (44 [30–61]) was lower (p = 0.006). At 1 year, EQ-VAS was similar between the 2 groups (p = 0.3); the improvement in EQ-VAS (29 [10–45]) was larger (p = 0.02) in the study group (Table 3).

Discussion Our study represents a nationwide analysis of young patients operated with a total hip arthroplasty after the turn of the 21st century. We found acceptable long-term results with regards to implant survival. Patient-reported outcomes in the younger patients at 1 year following THA were satisfactory.

Early reports on young patients operated with THA revealed a substantially higher risk of revision (Chandler et al. 1981, Dorr et al. 1983, Sarmiento et al. 1990) than comparable publications studying older patients. In 1990 Dorr et al. reported on 81 patients between 15 and 45 years of age operated with a THA. At an average follow up of 9 years 29 of the patients had been revised. Young age has since been considered a relative contraindication. Also, according to later data from national joint registries, the risk of revision is much higher in younger patients (Bayliss et al. 2017). In our study the implant survival was markedly higher than in the aforementioned studies. This could partly reflect the improvement in surgical techniques, tribology of the implants, and fixation technique (Barrack et al. 1992, Walker et al. 2016). These improvements could partly account for the higher implant survival being reported in recent single-center series (Makarewich et al. 2018, S ­ chreurs et al. 2018). Schreurs et al. (2018) analyzing 180 hips operated with a cemented hip replacement reported a 10-year implant survival of 87% (Schreurs et al. 2018). Makarewich et al.

(2018) reported a 10-year implant survival of 89% in patients

252

younger than 30 years. Due to concerns regarding generalizability when single-center studies are performed we decided to report on data from a national registry with high completeness. We found implant survival at 10 years consistent with recent reports. We are not certain whether differences in comorbidity, indications for surgery and number of previous surgeries are similar between our and previous studies. In our analysis the 15-year implant survival was 78%. Including metal-on-metal cases the 15- and 10-year implant survival were 76% and 89%. Thus there is still room for improvement for outcomes in the younger patients.

To our knowledge there are no previous reports on patientreported outcomes in younger patients. Before operation both the EQ-5D index and EQ-VAS were lower in the younger group. This finding might reflect a reluctance by surgeons and perhaps also by patients to accept operation with a THA due to an expected high risk of future revisions. The corresponding values at 1-year post-surgery did not differ between the groups, hence the improvement was larger in the younger group. It could be argued that the large improvement and the comparable patient-reported outcomes at 1 year is a warrant for the success of THA with regards to risk of late re-operations (Eneqvist et al. 2018).

There are several limitations to our study. First it could be argued that only the most skilled surgeons are operating on younger patients, making a comparison with a larger cohort of patients operated by all surgeons difficult. We do not have access to surgeon-specific data in the register, thus not allowing us to adjust for this supposed discrepancy. Although this limitation might benefit the implant survival in the younger group it should not have influenced the patient-reported outcomes (Jolbäck et al. 2018). Further, including data from national registries with high coverage and completeness will contribute to increased generalizability of our conclusions when compared with single-center series. Second, in our analysis more than half of the patients in the study group were operated with nonhighly cross-linked polyethylene (PE), mainly during the start of the study period. This could partly explain the less favorable implant survival at 15 years. It could be expected that the supposed long-term benefits of the highly cross-linked PE are more pronounced in the younger study group. If so, the results might be even better for these patients in the long term, but this hypothesis needs further investigation. Third, in performing an observational study there may be residual confounders such as case complexity and presence of previous surgical interventions that we could not account for. To what extent such factors might have influenced our results remains unknown.

In summary we found acceptable implant survival in the long term in a cohort of 504 patients reported to the Swedish Hip Arthroplasty Register and promising patient-reported outcomes at 1 year. Our findings support the use of THA as a feasible option in young patients with severe symptoms at least in cases where no other joint-preserving surgery is expected to be successful.

Acta Orthopaedica 2019; 90 (3): 249–252

MM, HM, JK, and OR conceived and planned the study. DO and EN performed statistical analyses. MM drafted the manuscript with subsequent substantial inputs from all co-authors. All authors interpreted the findings.

The authors would like to thank the orthopedic teams at the different centers in Sweden as well as the register coordinators at the Center of Registries, Västra Götaland for the high quality and integrity of the data being collected in the Swedish Hip Arthroplasty register.

Acta thanks Willem Schreurs and Claus Varnum for help with peer review of this study.

AOANJRR. Australian Orthopaedic Association National Joint Replacement Registry (AOANJRR). Hip, Knee & Shoulder Arthroplasty: 2018 Annual Report. Australian Orthopaedic Association National Joint Replacement Registry (AOANJRR). Hip, Knee & Shoulder Arthroplasty: 2018 Annual Report. https://aoanjrr.sahmri.com/documents/10180/576950/ Hip%2C%20Knee%20%26%20Shoulder%20Arthroplasty; 2018; 2018.

Barrack R L, Mulroy R, Harris W H. Improved cementing techniques and femoral component loosening in young patients with hip arthroplasty: a 12-year radiographic review. Bone Joint J 1992; 74(3): 385-9.

Bayliss L E, Culliford D, Monk A P, Glyn-Jones S, Prieto-Alhambra D, Judge A, et al. The effect of patient age at intervention on risk of implant revision after total replacement of the hip or knee: a population-based cohort study.

Lancet 2017; 389(10077): 1424-30.

Chandler H P, Reineck F, Wixson R, McCarthy J. Total hip replacement in patients younger than thirty years old: a five-year follow-up study. J Bone Joint Surg Am 1981; 63(9): 1426-34.

Dorr L, Takei G, Conaty J. Total hip arthroplasties in patients less than fortyfive years old. J Bone Joint Surg Am 1983; 65(4): 474-9.

Dorr L D, Luckett M, Conaty J P. Total hip arthroplasties in patients younger than 45 years. A nine- to ten-year follow-up study. Clin Orthop Relat Res 1990; (260): 215-19.

Eneqvist T, Nemes S, Bülow E, Mohaddes M, Rolfson O. Can patientreported outcomes predict re-operations after total hip replacement? Int Orthop 2018; 42(2): 273-9.

Jolbäck P, Rolfson O, Mohaddes M, Nemes S, Kärrholm J, Garellick G, et al. Does surgeon experience affect patient-reported outcomes 1 year after primary total hip arthroplasty? A register-based study of 6,713 cases in western Sweden. Acta Orthop 2018; 89(3): 265-71.

Kärrholm J, Mohaddes M, Odin D, Vinblad J R, C, Rolfson O. The Swedish Hip Arthroplasty Register, Annual Report 2017. The Swedish Hip Arthroplasty Register, Annual Report 2017. https://registercentrum.blob.core. windows.net/shpr/r/-rsrapport-2017-S1xKMzsAwX.pdf; 2018; 2018.

Makarewich C A, Christensen M, Anderson M B, Gililland J M, Pelt C E, Peters C L. Ten-year survivorship of primary total hip arthroplasty in patients 30 years of age or under: successful, but higher risk of revision.

Bone Joint J 2018; 100(Suppl. 1): 50–.

NJR. National Joint Registry for England, Wales, Northern Ireland and Isle of Man 15th Annual Report 2018. National Joint Registry for England, Wales, Northern Ireland and Isle of Man 15th Annual Report 2018. http://www. njrreports.org.uk/Portals/0/PDFdownloads/NJR%2015th%20Annual%20 Report%202018.pdf; 2018; 2018.

Sarmiento A, Ebramzadeh E, Gogan W, McKellop H. Total hip arthroplasty with cement: a long-term radiographic analysis in patients who are older than fifty and younger than fifty years. J Bone Joint Surg Am 1990; 72(10): 1470-6.

Schreurs B, Colo E, Schmitz M, Rijnen W, Gardeniers J. Total hip arthroplasty in patients under 30 years: a long-term report of 180 hips. Bone Joint J 2018; 100(Suppl. 1): 48–.

Walker R P, Gee M, Wong F, Shah Z, George M, Bankes M J, et al. Functional outcomes of total hip arthroplasty in patients aged 30 years or less: a systematic review and meta-analysis. Hip Int 2016; 26(5): 424-31.

Acta Orthopaedica 2019; 90 (3): 253–257

253

Evaluation of Forgotten Joint Score in total hip arthroplasty with Oxford Hip Score as reference standard Amanda LARSSON 1, Ola ROLFSON 1,2, and Johan KÄRRHOLM 1,2 1 Department

of Orthopaedics, Institute of Clinical Sciences, Sahlgrenska Academy, University of Gothenburg, Gothenburg; 2 The Swedish Hip Arthroplasty Register, Registercentrum, Västra Götaland, Gothenburg, Sweden Correspondence: amanda.em.larsson@gmail.com Submitted 2018-11-29. Accepted 2019-02-14.

Background and purpose — Total hip arthroplasty (THA) is performed mainly because of pain. To evaluate the result after surgery, different questionnaires measuring the patient-reported outcome regarding quality of life are used.

Forgotten Joint Score (FJS), designed to chart postoperative symptoms, was developed to find subtle differences between patients who report that their operated hip is “very good” or “excellent.” We evaluated whether FJS provides additional information compared with the Oxford Hip Score (OHS) and ceiling and floor effects with use of these instruments. We also studied level of internal consistency for OHS and FJS, and the reproducibility of the FJS. Patients and methods — 111 patients who underwent unilateral primary THA in 2015 were included. The participants answered 2 questionnaires: Forgotten Joint Score and Oxford Hip Score. Floor and ceiling effects were recorded for each of the instruments and agreement between them.

The FJS was studied with respect to reproducibility and level of internal consistency.

Results — OHS ceiling effect (31%) was higher compared with FJS (21%), whereas the OHS seemed to provide a more nuanced picture of patients with an inferior clinical result. Floor effect for FJS was 3% and 0% for OHS. The degree of explanation was 68% between the 2 questionnaires (linear regression, r2 = 0.68). FJS items had a high internal consistency (Cronbach’s a = 0.93) and reproducibility (Pearson correlation = 0.87, ICC = 0.93); 92 patients answered on 2 distributions of the FJS questionnaires, 19 patients had identical answers.

Interpretation — OHS had a larger ceiling effect than FJS, which could indicate that FJS is a more fine-tuned instrument to separate patients with good to excellent outcome after THA. The high internal consistency of FJS indicates that the items of the instrument consistently cover the construct of joint awareness.

The aim with total hip arthroplasty (THA) is to relieve pain, and improve joint mobility, physical ability, and quality of life. Since patients’ expectations on the postoperative pain and functional outcomes have changed over the past 20 years (Hamilton et al. 2017), it is of importance to apply validated methods to measure patient-reported outcomes (PROs) after surgery (Behrend et al. 2012). Questionnaires measuring PROs should preferably exhibit low ceiling and floor effects.

Critical ceiling and floor effects are regarded to be present if 15% or more of the population reach the maximum or minimum score of a scale, respectively (Terwee et al. 2007). Ceiling and floor effects could make it difficult to study the development over time, since the true results and changes at follow up are concealed.

The commonly used Oxford Hip Score (OHS) and the more recently developed Forgotten Joint Score (FJS) are 2 validated metrics for the evaluation of THA. OHS focuses on the preoperative status, while FJS primarily was designed to chart the symptoms postoperatively (Hamilton et al.

2016). Wylde et al. (2005) explored different weaknesses of the OHS questionnaire, e.g., that patients experienced that some questions did not have a clear meaning. The patients also commented on the difficulty of answering according to their “average pain” during the past 4 weeks, since their pain sometimes fluctuated based on current medication and level of physical activity. Some of the questions in OHS are so called “double-barreled questions,” meaning there is more than 1 claim in each question. This could result in difficulty in interpreting the answers, since some patients marked more than 1 of the possible answers for each question. In addition, the OHS has been criticized for exhibiting ceiling effects at postoperative follow-ups (Hamilton et al. 2016, 2017). The FJS was developed in 2012 as a reaction to the shortcomings of established measures following joint replacement and its use is growing.

© 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group, on behalf of the Nordic Orthopedic Federation. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits ­unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

DOI 10.1080/17453674.2019.1599252

254

Acta Orthopaedica 2019; 90 (3): 253–257

Selected patients Unilateral THA performed 2015 n = 200 11 patients died and 10 were replaced Patients invited n = 199 Excluded (n = 88): – did not respond, 76 – incomplete answers, 12 Patients with complete answers n = 111 (56%) Did not respond to retest n = 14 Included in FJS test–retest n = 97

Figure 1. Participation.

The Swedish Hip Arthroplasty Register established a program collecting PROs in 2002 (Rolfson et al. 2011). The Registry direction is currently investigating new or alternative measures to include in the program. As both the newly developed FJS and the well-established OHS are tentative candidates, we sought to investigate these measures in a typical THA population. Hence, the objective of this study was to evaluate whether Forgotten Joint Score (FJS) provides more information compared with, or in addition to, Oxford Hip Score (OHS) regarding the clinical outcome after THA. Specifically, we aimed to compare ceiling and floor effects between OHS and FJS, to investigate the level of internal consistency for OHS and FJS, and to test the reproducibility of the FJS.

Table 1. Patient demographics: ASA classification and Charnley category in 200 patients (200 hips) primarily selected to be invited

All n = 200

Age < 65 Age ≥ 65 n = 99 n = 101

Male/female 100/100 50/49 50/51 Age, median 66 57 75 range 17–97 17–64 66–97 Diagnosis Primary osteoarthritis 135 66 69 Inflammatory joint disease 1 1 0 Fracture 36 10 26 Sequelae childhood hip disease 10 10 0 Femoral head necrosis 18 12 6 Charnley category a A 84 44 40 B 79 32 8 C 18 10 47 missing 19 13 6 ASA 1 49 39 10 2 118 49 69 3 30 9 21 4–5 0 0 0 missing 3 2 1 a Filled

in by patient 1 year after the index operation.

1 year after the primary index operation. 10 to 14 days after return of the questionnaire, the FJS was sent out once again to evaluate its reproducibility. Approximately 1 month after the first invitation to participate was sent out, patients who had not responded received a phone call reminder and an offer to receive a new set of questionnaires. If the patient declined participation or did not answer after 2 phone calls no further attempts to reach the patients were made. Numbers included were calculated based on an estimated response rate of 75%, i.e., 150 patients.

Study population (Figure 1) 200 patients who had undergone unilateral THA at the Department of Orthopaedics, Sahlgrenska University Hospital in Mölndal during 2015 were invited to participate. The population was selected from a consecutive series of THA with stratification for age and sex: half of the invited patients were over 65 years old and the other half were 65 years old or younger; half of the invited patients in each age group were females. All types of diagnoses (Table 1) and patients who previously had been operated on their opposite hip were included in the study.

1 of the 200 patients had been revised at the time when the questionnaires were sent out. This patient was also included.

Patient-reported outcome measures Oxford Hip Score The OHS, developed in 1996 (Dawson et al. 1996), is a patient-centered, 12 item-questionnaire with questions concerning pain and physical ability that the patient experienced during the past 4 weeks. The OHS originally used a scoring system ranging between 1 and 5 (worst–best). Since 2007, OHS items range from 0 to 4 where 4 is the best, which leads to a total score ranging from 0 to 48, where 48 equals the best outcome (Murray et al. 2007). When interpreting the answers and calculating the overall score of OHS, a maximum of 2 missing values are accepted. If the patient fills in more than 1 answer per question, the worst response should be used when calculating the total score (Nilsdotter and Bremander 2011).

Sample and logistics Invited patients were asked to complete 2 questionnaires, FJS and OHS, which were sent out by ordinary mail at the beginning of September 2017. Questionnaires were filled in at least

Forgotten Joint Score The Forgotten Joint Score (FJS) is a joint-specific questionnaire developed in 2012 (Behrend et al. 2012) with the aim to measure PRO of joint disorders (Hamilton et al. 2017). FJS is

Patients and methods

Acta Orthopaedica Orthopaedica 2019; 90 (3): (3): 253–257 253–257 Acta 2019; 90

255 255

designed to measure the ability of the patient Forgotten Joint Score Difference between FJS and OHS to “forget” about their problematic joint after 100 20 treatment. FJS is available in 3 versions: hip, knee, and shoulder. Studies imply that older 80 0 questionnaires do not provide quite as variegated a picture of the results, as they mostly 60 –20 differ between “good” and “bad.” Behrend y = –59.59 + 2.99x et al. (2012), however, states that since FJS –40 40 differ between “good,” “very good,” and “excellent” on a 5-grade Likert scale rang–60 20 ing from “never” to “mostly,” it could reduce –80 the risk of ceiling effects. As opposed to, for 0 example OHS, FJS is a questionnaire that 0 20 40 60 80 100 0 8 16 24 32 40 48 Mean of FJS and OHS Oxford Hip Score focus on the awareness, instead of the pain, of the affected joint (Hamilton et al. 2017). Figure 2. Linear regression analysis. The Figure 3. Bland–Altman limits of agreement.

4 missing values are regarded as acceptable degree of explanation between the 2 instruwhen the scores are summarized and trans- ments reached 0.68. formed to a scale ranging from 0 to 100, where a high value indicate that the patient tends to be less aware of the affected joint when performing by grants from the Swedish state under the agreement between daily activities (Behrend et al. 2012). According to the official the Swedish government and the county councils, the ALF website of FJS (http://www.forgotten-joint-score.info/), the agreement (ALFGBG- 522591 and 721791, respectively). The developers have undertaken translation into several languages authors have no conflict of interest. (including Swedish) and subsequently linguistically validated the translated forms based on the Principles of Good Practice for the Translation and Cultural Adaption Process for PatientResults Reported Outcomes (PRO) Measures (Wild et al. 2005).

Ceiling and floor effects of FJS and OHS Reasons for non-participation Of the 199 patients primarily invited, 111 (56%) answered Of those patients receiving a reminder phone call, 30 did not the questionnaires. Of these, 52% were female. The median answer. A further 16 patients accepted to participate in the age was 69 (19–100). Mean OHS was 42 (SD 9, median study, but did not, despite their positive answer, return any 45, IQR 40–48) and mean FJS was 64 (SD 32, median 71, questionnaires (Table 2, see Supplementary data).

IQR 42–93). 68% to 94% chose option 1 (best) for each of the questions in the OHS questionnaire. The corresponding Ceiling and floor effect proportion for FJS ranged from 39% to 77%. The responses A critical ceiling or floor effect was regarded to be present on the FJS forms were more scattered among the different if 15% or more of the patients chose the best or worst pos- options, and all response options were chosen for all quessible answer when answering a specific question (Terwee et tions, which differs from OHS, where no patients chose al. 2007). Ceiling or floor effects at or above this level indicate option 5 (worst) on 3 questions (questions 2, 3, and 4. Tables limited content validity and reduced reliability.

3 and 4, see Supplementary data). The total ceiling effect for OHS was 31% and FJS 21%. The floor effect did not reach Statistics 15% for either questionnaire, but was slightly higher for FJS Cronbach’s a was used to measure the internal consistency (FJS 3%, OHS 0%). among the 7 questions that were assessed to measure the same health dimension in the 2 questionnaires. The relationship Level of internal consistency (Cronbach’s a) and between total OHS and FJS was assessed using simple linear agreement between questionnaires regression. The intraclass correlation coefficient (ICC) and Pear- 113 patients responded to the first distribution of FJS (Cronson’s correlation were calculated to evaluate the reproducibility bach’s a = 0.97) and 111 patients answered the OHS (Cronof the questionnaires. Statistical analyses were performed using bach’s a = 0.93). In the linear regression analysis, the degree IBM SPSS Statistics version 24 (IBM Corp, Armonk, NY, USA). of explanation between the 2 instruments reached 68% (r2 = 0.68, Figure 2). The mean difference between FJS and OHS Ethics, funding, and potential conflicts of interest was 22, i.e., FJS was mean 22 lower than OHS when conThe study was approved by the Regional Ethical Review verted to a 0–100 scale (Bland–Altman limits of agreement, Board in Gothenburg (ref 607-17). OR and JK were financed Figure 3).

256

Forgotten Joint Score reproducibility The Pearson correlation coefficient and the intraclass correlation (ICC) for the reproducibility of the FJS were 0.87 (95% CI 0.75–0.96) and 0.93 (95% CI 0.89–0.95).

Discussion This pilot study emanated from the Swedish Hip Arthroplasty Register to explore and evaluate new instruments with potential use for specific or general purposes in the follow-up after THA. We found that the FJS had an acceptable and good reproducibility. It also had less of a ceiling effect when compared with the OHS. Because of these properties, the FJS could be considered for many purposes and especially in the evaluation of new implants claimed to provide improved patient satisfaction and/or clinical performance.

Ceiling and floor effects of FJS and OHS Studies have shown that OHS has a risk for ceiling effects postoperatively (Hamilton et al. 2016, 2017), which also was found by us. These ceiling effects were greater for OHS for each of the 7 correlating questions as well as the overall ceiling effect, which further support our hypothesis that FJS is a more nuanced instrument when measuring the results after THA. According to Terwee et al. (2007), a ceiling effect is present if more than 15% of the participants score the “best” result, corresponding to the lowest possible score for OHS and FJS. In a study of patients operated with either total knee or total hip arthroplasty, Hamilton et al. (2016) found a ceiling effect in the latter group of 8% at 6 months, which increased to 10% at 1 year, i.e., well below this limit (Hamilton et al. 2016).

In our study, the ceiling effect for the OHS and FJS were higher at 31% and 21% respectively. The reason for this is not known, but longer follow-up could have contributed. Neither of the 2 questionnaires reached 15% for floor effect, though 3% achieved the maximum score on FJS. Several patients achieved high total FJS values, which indicates high awareness of their hip joint in daily life. Thus, the FJS provides a better differentiation of hip-related symptoms in patients who are generally satisfied with their THA operation. So far, the majority of studies which have compared the FJS with other scoring systems, have found that the ceiling effect of FJS is smaller or about the same as the reference used. Matsumoto et al. (2015) compared FJS with the WOMAC and the Japanese Orthopaedic Association Hip Disease Evaluation Questionnaire (JHEQ) in 108 patients operated with THA about 2.5 years after the operation. A low ceiling effect was reported for the FJS (4%) and for the JHEQ (3%), (Matsumoto et al. 2015).

Hamilton et al. (2016) suggested that FJS is more responsive to change than OHS based on findings of a more pronounced change in the FJS than for the OHS between the 6and 12-month follow-up. They also noticed that the measured

Acta Orthopaedica 2019; 90 (3): 253–257

ceiling effect was nearly doubled for OHS compared with FJS (21% and 10% respectively). In another study by Hamilton et al. (2017), they found a pronounced floor effect for FJS when used preoperatively. 22% of the THA patients achieved the lowest score. These number differs from OHS, where no floor effects were shown preoperatively. The ceiling effect, however, was approximately half for FJS 1 year postoperatively compared with OHS. Reproducibility and internal consistency To examine the reproducibility of FJS, a comparison was made between the answers of the 1st and 2nd distribution of FJS. Overall, patients chose the same response option for both questionnaires in the test–retest analysis. Fewer patients, however, answered on the second distribution of FJS (n = 92). 19 patients had identical answers on both FJS 1 and FJS 2 and the internal consistency was high (ICC = 0.93).

Thomsen et al. (2016) found good reliability in test–retest of FJS total score (ICC = 0.91). When FJS was compared with the Oxford Knee Score (OKS), there was a high level of internal consistency (Cronbach’s a = 0.96). Behrend et al.

(2012) also observed that FJS had high internal consistency (Cronbach’s a = 0.95). Both these studies and ours indicate that the items of FJS consistently measure the construct of joint awareness.

Weaknesses of this study Our study was performed in a Swedish population. To what extent this circumstance has influenced our result is not known, but we think that the influence of differences in ethnicity between nations or groups is weak, at least in Europe.

Since this was a cross-sectional study, change over time was not analyzed, which is a limitation of this study, as well as the fact that only primary THA patients were included. The response frequency was lower compared with the Register’s ordinary PROM routine (Rolfson 2016), which could be due to a number of extra questionnaires with stated research purpose outside normal routines.

Another possible limitation of this study was that a moderately sized population was included. The calculated participation level was 75%, i.e., 150 respondents, and the actual participation level was 56%. 123 of the patients answered the FJS completely and 111 patients answered the OHS completely. The response frequency probably would have been higher if the study took place over a longer period and more attempts could have been made to reach the patients who did not answer.

Despite the fact that the FJS had a lower ceiling effect than the OHS, 1 of 5 nevertheless reached the maximum score for all items included. This finding raises the question whether still more sensitive instruments should be used, especially in situations when implants used in surgical procedures supposed to address high-demand patients are operated. It could be that new instruments used to evaluate patients with femo-

Acta Orthopaedica 2019; 90 (3): 253–257

roacetabular impingement have still less ceiling effect in an arthroplasty population than the FJS (Thorborg et al. 2011, Griffin et al. 2012, Mohtadi et al. 2012). These questionnaires do, however, include items related to high-demand sport and work-related activities, these not being relevant for a majority of the population operated with total hip arthroplasty. Thus, these instruments might be less suitable to use in arthroplasty registers.

Conclusion We found that FJS could be used as an alternative to OHS in primary THA. The answers on FJS are more scattered than on OHS, which could indicate that FJS provides a more variegated picture of the clinical results in this population. The ceiling effect of FJS was lower compared with that of OHS, which provides valuable information, not least in a field where new implants with proposed superior performance are continuously introduced and patient expectations in terms of the results tend to increase. On the other hand, FJS might, due to its floor effect, be less suitable to predict the need for future revision surgery. Further studies in the Swedish population of patients with THA are needed, preferably with a larger study population, to establish whether the Forgotten Joint Score is sensitive to change and if it could be included in the Swedish Hip Arthroplasty Register PROMs routine with maintained completeness.

Supplementary data Tables 2–4 are available as supplementary data in the online version of this article, http://dx.doi.org/10.1080/17453674.

2019.1599252 JK conceived and planned the study. AL managed questionnaire logistics.

AL and OR performed the statistical analyses. AL drafted the manuscript.

All authors interpreted the results.

Acta thanks Ove Furnes for help with peer review of this study.

Behrend H, Giesinger K, Giesinger J M, Kuster M S. The “Forgotten Joint” as the ultimate goal in joint arthroplasty: validation of a new patient-reported outcome measure. J Arthroplasty 2012; 27(3): 430-6.e1.

257

Dawson J, Fitzpatrick R, Carr A, Murray D. Questionnaire on the perceptions of patients about total hip replacement. J Bone Joint Surg Br 1996; 78(2): 185-90.

Griffin D R, Parsons N, Mohtadi N G, Safran M R. A short version of the International Hip Outcome Tool (iHOT-12) for use in routine clinical practice. Arthroscopy 2012; 28(5): 611-16; quiz 6-8.

Hamilton D F, Giesinger J M, MacDonald D J, Simpson A H, Howie C R, Giesinger K. Responsiveness and ceiling effects of the Forgotten Joint Score-12 following total hip arthroplasty. Bone Joint Res 2016; 5(3): 87-91.

Hamilton D F, Loth F L, Giesinger J M, Giesinger K, MacDonald D J, Patton J T, et al. Validation of the English language Forgotten Joint Score-12 as an outcome measure for total hip and knee arthroplasty in a British population. Bone Joint J 2017; 99-b(2): 218-24.

Matsumoto M, Baba T, Homma Y, Kobayashi H, Ochi H, Yuasa T, et al. Validation study of the Forgotten Joint Score-12 as a universal patient-reported outcome measure. Eur J Orthop Surg Traumatol 2015; 25(7): 1141-5.

Mohtadi N G, Griffin D R, Pedersen M E, Chan D, Safran M R, Parsons N, et al. The development and validation of a self-administered quality-of-life outcome measure for young, active patients with symptomatic hip disease: the International Hip Outcome Tool (iHOT-33). Arthroscopy 2012; 28(5): 595-605; quiz 6-10.e1.

Murray D W, Fitzpatrick R, Rogers K, Pandit H, Beard D J, Carr AJ, et al.

The use of the Oxford hip and knee scores. J Bone Joint Surg Br 2007; 89(8): 1010-4.

Nilsdotter A, Bremander A. Measures of hip function and symptoms: Harris Hip Score (HHS), Hip Disability and Osteoarthritis Outcome Score (HOOS), Oxford Hip Score (OHS), Lequesne Index of Severity for Osteoarthritis of the Hip (LISOH), and American Academy of Orthopedic Surgeons (AAOS) Hip and Knee Questionnaire. Arthritis Care Res (Hoboken) 2011; 63(Suppl 11): S200-7.

Rolfson O. Höftprotesregistret Årsrapport; 2016.

Rolfson O, Kärrholm J, Dahlberg L, Garellick G. Patient-reported outcomes in the Swedish Hip Arthroplasty Register: results of a nationwide prospective observational study. J Bone Joint Surg Br 2011; 93(7): 867-75.

Terwee C B, Bot S D M, de Boer M R, van der Windt D A W M, Knol D L, Dekker J, et al. Quality criteria were proposed for measurement properties of health status questionnaires. J Clin Epidemiol 2007; 60(1): 34-42.

Thomsen M G, Latifi R, Kallemose T, Barfod K W, Husted H, Troelsen A. Good validity and reliability of the forgotten joint score in evaluating the outcome of total knee arthroplasty. Acta Orthop 2016; 87(3): 280-5.

Thorborg K, Holmich P, Christensen R, Petersen J, Roos E M. The Copenhagen Hip and Groin Outcome Score (HAGOS): development and validation according to the COSMIN checklist. Br J Sports Med 2011; 45(6): 478-91.

Wild D, Grove A, Martin M, Eremenco S, McElroy S, Verjee–Lorenz A, et al.

Principles of good practice for the translation and cultural adaptation process for patient–reported outcomes (PRO) measures: report of the ISPOR Task Force for Translation and Cultural Adaptation. Value Health 2005; 8(2): 94-104.

Wylde V, Learmonth I D, Cavendish V J. The Oxford hip score: the patient’s perspective. Health Qual Life Outcomes 2005; 3: 66.

258

Acta Orthopaedica 2019; 90 (3): 258–263

Uncemented cups with and without screw holes in primary THA: a Swedish Hip Arthroplasty Register study with 22,725 hips Volker OTTEN 1, Sebastian MUKKA 1, Kjell NILSSON 1, Sead CRNALIC 1, and Johan KÄRRHOLM 2 1 Department

of Surgical and Perioperative Sciences (Orthopaedics), Umeå University, Umeå; 2 Department of Orthopaedics, Institute of Surgical Science, Sahlgrenska University Hospital, Gothenburg University, Mölndal, Sweden Correspondence: volker.otten@umu.se Submitted 2018-11-21. Accepted 2019-02-15.

Background and purpose — Uncemented cups in total hip arthroplasty (THA) are often augmented with additional screws to enhance their primary stability. We investigated whether there is a difference in the risk for revision between cups with screw holes and cups without screw holes.

Patients and methods — We analyzed the risk for cup revision of uncemented cups registered in the Swedish Hip Arthroplasty Register (SHAR) between 2000 and 2017 with respect to the presence of screw holes. Only patients with primary osteoarthritis (OA) were included. 22,725 cups, including 12,354 without screw holes and 10,371 with screw holes, were evaluated. Revision rates at 2 and 10 years after the primary operation were analyzed.

Results — At a median follow-up time of 3.4 years (0–18), 459 cup revisions were reported. The main reasons for cup revision during the whole observation time were infection, 52% of all cup revisions, and dislocation, 26% of all cup revisions. The survival rate with cup revision due to aseptic loosening as endpoint was 99.9% (95% CI 99.8–99.9) at 2 years for both cups with and cups without screw holes, and the survival rates at 10 years were 99.5% (CI 99.3–99.7) and 99.1% (CI 98.6–99.5), respectively. Cups without screw holes showed a decreased risk of revision due to any reason at both 2 years (adjusted hazard ratio [HR] 0.6, CI 0.5–0.8) and 10 years (HR 0.7, CI 0.5–0.9).

Interpretation — We found a very low revision rate for aseptic loosening with modern, uncemented cup designs.

Cups with screw holes had an increased risk of revision due to any reason in patients with primary OA

Uncemented cups have gained popularity in recent years, even in countries traditionally using cemented fixation (Kärrholm et al. 2017). Initial stability is essential for good long-term results. Therefore, uncemented cups are often augmented with screws or pegs. However, screw holes in acetabular cups have also been discussed as potential routes for synovial fluid, which might lead to osteolysis (Aspenberg and van der Vis 1998, Iorio et al. 2010). Additionally, using screws for cup fixation increases the cost and operation time. Hence, the surface of uncemented cups has been further developed towards a rougher finish to improve primary stability and long-term bone ingrowth even without any extra augmentation.

In a recently published randomized controlled study with a minimum follow-up of 14 years comparing cups with and without screw holes, we found no difference in implant migration (Otten et al. 2016). To our knowledge, there has not been any published study investigating the risk for cup revision depending on the use of cups with or without screws.

We analyzed the survival of uncemented cups with or without screw holes in the Swedish Hip Arthroplasty Register (SHAR). Our hypothesis was that screw fixation reduces the risk for early aseptic loosening but increases the risk for late failure because of potentially increased risk for osteolysis due to screw holes (Iorio et al. 2010). Revision rates due to aseptic loosening and for any reason at 2 and 10 years after primary operation were analyzed.

Patients and methods Study design, source of data, and terminology This study is based on data obtained from the SHAR. Primary and revision hip arthroplasties performed in Sweden have been registered since 1979. During the last decade, the completeness of primary surgeries has been approximately 98–99% and 94% for revisions. Since 1999, detailed informa-

© 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group, on behalf of the Nordic Orthopedic Federation. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits ­unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

DOI 10.1080/17453674.2019.1599777

Acta Orthopaedica 2019; 90 (3): 258–263

259

Number or operations

Primary THA procedures 2000–2017 using uncemented cups (SHPR database) n = 46,047 Excluded (n = 23,322): – resurfacing arthroplasty, 2,806 – dual mobility cups, 217 – cups designs used less than 500 times, 3,199 – registration about screw holes missing, 178 – cup designs with screw holes used < 100 times or without screw holes used < 100 times, 8,500 – liner material or design unknown, 76 – ceramic liner, 444 – metal liner, 164 – standard PE liner, 2,295 – head size unknown, 42 – head size < 28 mm, 160 – head size > 36 mm, 181 – other diagnosis than primary OA or unknown, 5,060 Analyzed n = 22,725 Without screw holes, 12,354 With screw holes, 10,371

3,000 CUPS without screw holes with screw holes

2,000

1,000

0 2000 2002 2004 2006 2008 2010 2012 2014 2016

Year of operation

Figure 2. Number of operations per year.

Table 1. Baseline characteristics of the study population

Figure 1. Procedures not included in the final analysis.

tion regarding the prosthesis design, such as the presence of screw holes, has been registered in the SHAR. The guidelines of the STROBE statement were followed.

In this study, we classified cups without any holes and cups with 1 central hole, used for the cup impactor, into the category referred to as “cups without screw holes.” Cups with holes placed in a sector of the shell or over the whole cup area (multi-hole cups) were classified into “cups with screw holes” regardless of whether the holes were intended for screws or pegs, as long as the holes penetrated the entire thickness of the shell. Cup revision is defined as any reoperation of the hip where 1 or more components of the cup (shell, liner, or both) were removed or exchanged. “Aseptic loosening” as the reason for revision includes aseptic loosening, osteolysis, liner instability, technical reasons (explain), and wear.

Failure due to insufficient primary stability most often occurs within the first 2 years. Hence, survival of the cups up to a 2-year follow-up was analyzed. Aseptic loosening due to osteolysis is a late complication that might be influenced by the presence of screw holes. Therefore, we also evaluated the 10-year survival.

Characteristics of the study population Between January 1, 2000, and December 31, 2017, 46,047 uncemented cups used in primary total hip arthroplasty (THA) were reported to the SHAR, including 63 different cup designs.

During the 18-year study period, 30 of these cup designs were reported in less than 100 cases and for 17 cup designs in even less than 10 cases. Only cases using cups that have been frequently used both with and without screw holes and that are still available today were included in this study. After applying the exclusion criteria described in Figure 1, 22,725 hips

Screw holes No Yes n (%) n (%)

Totals n (%) p-value a

Sex

Male 6,916 (56) 5,405 (52) 12,321 (54) < 0.001 Female 5,438 (44) 4,966 (48) 10,404 (46) Age < 45 403 (3.3) 406 (3.9) 809 (3.6) < 0.001 45–64 7,428 (60) 6,395 (62) 13,823 (61) 65–74 3,662 (30) 2,607 (25) 6,269 (28) ≥ 75 861 (7.0) 963 (9.3) 1,824 (8.0) Side Right 6,550 (53) 5,572 (54) 12,122 (53) 0.3 Left 5,804 (47) 4,799 (46) 10,603 (47) Approach b

Posterolateral 7,769 (63) 3,252 (31) 11,021 (49) < 0.001 Direct lateral 4,016 (33) 7,008 (68) 11,024 (49) Other 464 (3.8) 97 (0.9) 561 (2.5) Type of stem:

Uncemented 10,580 (86) 8,715 (84) 19,295 (85) 0.001 Cemented 1,774 (14) 1,656 (16) 3,430 (15) Cup coating No HA 6,662 (54) 3,483 (34) 10,145 (45) < 0.001 HA 5,692 (46) 6,888 (66) 12,580 (55) Head size 28 mm 1,385 (11) 2,434 (24) 3,819 (17) < 0.001 32 mm 7,585 (61) 7,128 (69) 14,713 (65) 36 mm 3,384 (27) 809 (7.8) 4,193 (19) Head material

Metal 9,260 (75) 9,542 (92) 18,802 (83) < 0.001 Ceramic 3,094 (25) 828 (8.0) 3,922 (17) Sum

12,354 (100) 10,371 (100) 22,725 (100)

Distribution of possible confounding factors between cups with and without screw holes a Pearson’s chi-square. b Missing data in 119 cases.

in 19,840 patients were used for analysis (Table 1), including 8 different cup designs, 12,354 cups without screw holes and 10,371 cups with screw holes (Table 2, see Supplementary data). Both hips were included in 2,885 patients. Cups without screw holes were used in 1,305 of these bilateral THA patients,

260

and cups with screw holes were used in 1,217 bilateral THA patients. The remaining 363 patients had 1 of each type of cup used on each on the 2 sides. The median follow-up time for cups without screw holes was 2.6 years (0–16) and for cups with screw holes was 5.2 years (0–18) (Table 2, see Supplementary data). The number of procedures and the proportion of cups without screw holes increased over time (Figure 2).

Operations from 81 different hospitals are included, of which 30 hospitals used cups with screw holes in more than 90% of cases, and 30 hospitals used cups with screw holes in less than 10% of cases.

Adjusting for confounders Poor bone quality or abnormal anatomy could lead to the use of cups with screws in primary THA. Neither of these factors is registered in the SHAR. We assumed that older patients and women, in general, have poorer bone quality and patients who had dysplasia or sequelae after hip diseases, as well as patients with other secondary arthrosis, are more likely to have abnormal anatomy. Therefore, we used the patient’s age and sex to adjust the dataset and excluded all diagnoses other than primary osteoarthritis (OA). To various extents, the presence of a hydroxyapatite (HA) coating, the head size, and the head material will contribute to the revision risk (Lazarinis et al.

2010, Cross et al. 2012, Dahl et al. 2012), and these factors were used as covariates in a Cox regression.

Liner material influences the amount of wear. During the last few years, almost exclusively PE liners with x-linked polyethylene (PE) have been used in uncemented cups in Sweden.

Hence, only cups with the x-linked PE liner were included in the analysis.

Statistics Continuous variables are described as the means, medians, and ranges. Comparisons between groups were performed using independent Student’s t-tests, Welch’s t-tests, or 1-way ANOVA. Categorical data were analyzed with chi-square tests.

Survival of the cup was calculated using life tables with 95% confidence intervals (CIs). For comparison of the survival of cups with and without screw holes, the log-rank test was used.

Adjusted hazard ratios (HRs) with CIs were calculated using multivariable Cox regression models.

The total observation time comprised 18 years (2000–2017).

However, the number of cups at risk for revision after 10 years was only 1,846, and the distribution of cups with and cups without screw holes was unequal between cup design (Table 4, see Supplementary data); therefore, the risk for revision was not calculated beyond 10 years.

The E-value was calculated to define the minimum strength of association of the HR that an unmeasured confounder would need to have with both the treatment and the outcome to fully explain away the association between screw holes and cup revision on the measured covariates (VanderWeele and Ding 2017).

Acta Orthopaedica 2019; 90 (3): 258–263

A p-value < 0.05 was considered statistically significant.

Statistical analysis was performed using SPSS Statistics software, version 24.0 (IBM Corp, Armonk, NY, USA).

Ethics, funding, and potential conflicts of interest The study was approved by the Regional Ethics Committee in Gothenburg (dnr 348-17). No competing interests are declared. The research was funded by SHAR and grants from the regional agreement on medical training and clinical research (ALF) between Västerbotten County Council and Umeå University.

Results Early revisions within 2 years after primary operation The survival rate with cup revision for aseptic loosening within 2 years was 99.9% (CI 99.8–99.9) for both groups (Table 3).

When including all reasons for cup revision, the survival rate for cups without screw holes was 98.6% (CI 98.4–98.8) and for cups with screw holes was 98.4% (CI 98.2–98.7).

The crude HR for the risk of cup revision due to aseptic loosening of cups without screw holes compared with cups with screw holes was 0.8 (CI 0.4–1.7). After adjusting for sex, age, surgical approach, type of stem fixation, presence of HA coating, head size, head material, and cup design, we found that screw holes had no influence on cup revision due to aseptic loosening (HR = 0.6, CI 0.2–1.8). The crude HR for cup revision for any reason was 0.8 (CI 0.7–1.0) for cups without screw holes. After adjustment for the covariates, cups without screw holes showed a lower risk for cup revision for any reason with an HR of 0.6 (CI 0.5–0.8) (Table 3) and an E-value of 2.7 (CI 1.8–3.4). The influence of other patient- or prothesis-related factors on the revision rate is presented in Table 5 (see Supplementary data).

In 293 out of 22,725 hips, cup revision, including exchange (or extraction) of the cup, the liner, or both, was performed within the first 2 years after the primary operation. These early revisions were caused by infection in 58% of cases, dislocation in 27%, and aseptic loosening in 9%. Other complications, such as fracture, implant failure, or pain, caused less than 6% of early cup revisions.

The proportion of various reasons for early revision did not differ significantly between cups with and without screw holes (p = 0.2) but differed significantly between cup designs (p = 0.03). For example, all revisions of the Tritanium cup were performed due to infection, while the main reason for revision of the Continuum cup was dislocation (47%).

Revisions within 10 years after primary operation The overall 10-year survival rate for aseptic loosening was 99.1% (CI 98.6–99.5) for cups without screw holes and 99.5% (CI 99.3–99.7) for cups with screw holes (Table 3). However, the risk for cup revision due to any reason was still lower for

Acta Orthopaedica 2019; 90 (3): 258–263

261

Table 3. Survival after 2 and 10 years and hazard ratio for revision

Endpoint

2-year survival (95% CI)

Cup revision for aseptic loosening Without screw holes 99.9 (99.8–99.9) With screw holes 99.9 (99.8–99.9) Cup revision for any reason Without screw holes 98.6 (98.4–98.8) With screw holes 98.4 (98.2–98.7) Revision (cup or stem) for any reason Without screw holes 98.0 (97.7–98.2) With screw holes 97.7 (97.4–98.0)

10-year survival (95% CI)

Crude HR Adjusted HR Crude HR Adjusted HR (0–2 years) (0–2 years) (0–10 years) (0–10 years) (95% CI) p-value (95% CI) p-value (95% CI) p-value (95% CI) p-value

99.1 (98.6–99.5) 99.5 (99.3–99.7)

0.8 (0.4–1.7) 0.6 1.0 (ref)

0.6 (0.2–1.8) 0.4 1.0 (ref)

1.2 (0.7–2.1) 0.5 1.0 (ref)

0.9 (0.5–1.8) 0.8 1.0 (ref)

96.5 (95.8–97.2) 96.8 (96.3–97.2)

0.8 (0.7–1.0) 0.09 0.6 (0.5–0.8) 0.002 1.0 (0.8–1.2) 0.7 1.0 (ref) 1.0 (ref) 1.0 (ref)

0.7 (0.5–0.9) 0.004 1.0 (ref)

95.0 (94.1–95.9) 95.4 (94.9–95.9)

0.9 (0.7–1.0) 0.09 0.6 (0.5–0.8) 0.001 1.0 (0.8–1.1) 0.6 1.0 (ref) 1.0 (ref) 1.0 (ref)

0.7 (0.6–0.8) < 0.001 1.0 (ref)

The adjusted hazard ratio was calculated based on a Cox regression model with gender, age, surgical approach, type of stem fixation, cup coating, head size, head material, and cup design as covariates.

Number of cups at risk for revision in cups without/with screw holes was 12,354/10,371 at 0 years, 7,228/7,984 at 2 years and 3,47/1,499 at 10 years after primary operation.

cups without screw holes, with an adjusted HR of 0.7 (CI 0.5– 0.9) of E-value of 2.2 (CI 1.5–3.4) (Table 3).

Between the 2- and 10-year follow-up, 152 cup revisions were registered. 41% of these revisions were caused by infection, 26% by dislocation, and 20% because of aseptic loosening. The distribution of these revisions did not differ statistically significantly between patients who underwent unilateral or bilateral operations.

Discussion We found similar risk both at 2 years and 10 years for revision because of aseptic loosening between cups with and without screw holes in patients with primary OA. However, the risk for cup revision for any reason at both 2 and 10 years was higher when a cup with screw holes was used.

Screw fixation of uncemented cups increases stability in simulated models (Hsu et al. 2007) and cadaver studies (Won et al. 1995) and is therefore used with the intention to reduce the risk for loosening. This theory might still hold true for patients with abnormal anatomy, fractures, revision settings, and cups without porous coating or trabecular surfaces.

Additional screw fixation of cups with a porous coating or trabecular surfaces has previously been investigated in small populations and did not reduce migration in radiostereometric analysis (RSA) studies (Minten et al. 2016, Otten et al. 2016).

A review of 5 articles with a total of more than 1,000 patients and a follow-up time of up to 5 years also did not show any difference in revision rate or osteolysis between cups with and without screw fixation (Ni et al. 2013). Even in the present study with a substantially larger population and operations performed in more than 80 different hospitals, the cups without screw hole did not show a higher risk for early revision due to aseptic loosening. In contrast, we found a higher risk for revision due to any reason for cups with screw holes.

Using screws has some potential risks. Both prolonged operation time (Pepe et al. 2017) and increased likelihood of receiving a blood transfusion (Colacchio et al. 2017) have been reported in recent studies. Inserting screws in the acetabulum might even be a risk for damaging intrapelvic vessels (Ohashi et al. 2017). A report from a smaller group of patients describe a higher risk for osteolysis around cups with screw holes (Iorio et al. 2010).

During the last 2 decades, the number of uncemented cups used per year in Sweden has substantially increased. In the last decade, cups designed without screw holes have been used more often in operations for primary OA than have cups with holes. Data from the current study show that this development did not increase the risk of aseptic loosening. In contrast, we found a higher HR for cup revision for any reason if the cup had screw holes, at both 2 and 10 years. The main reason for both early and late revision was infection. A longer operation time when using screws might have caused a higher risk for infection, and there might also have been differences related to patient selection, which are not possible to adjust for in a register study. The E-value shows that unmeasured confounders need to increase the risk for revision by 2.7 times and must be 2.7 times more common in the group of patients who received a cup with screw hole to fully explain away the differences at 2 years. Of the possible confounding factors available, female sex, the use of cemented stems, and HA coating of the cups reduced the risk for cup revision for any reason. However, these 3 risk-reducing factors were overrepresented in the group of cups with screw holes. Nevertheless, cups with screw holes showed a higher risk for revision due to any reason both at 2 years and 10 years after operation.

The second most common reason for revision in this study was dislocation. The larger the head size, the lower the risk for revision due to dislocation (Hailer et al. 2012). In particular, 22 mm heads, which were used as standard size in the early days of modern hip arthroplasty to reduce wear (Charn-

262

ley et al. 1969), had a higher risk for dislocation. In cups with cross-linked PE liners, head sizes up to 36 mm do not seem to increase wear (Howie et al. 2016). Therefore, a head size of 28–36 mm, depending on the cup size, seems to be the optimal size when using metal or ceramic heads and a cross-linked PE liner. In this study, only head sizes 28, 32, and 36 mm were included. We did not find any difference in the risk for cup revision for any reason between these 3 head sizes. However, a larger proportion of 28 mm heads was used in the group of cups with screw holes, and this might still have influenced the risk for revision. TMT cups have been reported to have a higher revision rate due to dislocation (Hailer 2018, Laaksonen et al.

2018). Our results are concurrent with these reports.

HA coating has been discussed in several other papers. There is still no consensus on how HA coating influences the risk for cup revision. Several studies have shown an increased revision rate for HA-coated cups (Stilling et al. 2009, Lazarinis et al.

2010). The main reason for the higher revision rate of cups with HA coating seems to be failure of the liner (Lazarinis et al. 2012). Older cup designs have more often been used with HA coating and, at the same time, with standard but not crosslinked PE liners. When adjusting for potential confounders, including the type of liner, in a larger register-based study, a similar risk of aseptic loosening was found for cups with or without HA coating (Lazarinis et al. 2017). Unadjusted data from our study show a slight advantage for cups with HA coating, but when adjusting for several potential confounders, including cup design, there was no statistically significant difference in the risk for revision.

The main limitation of this study is the lack of detailed patient- or surgeon-related information that might have influenced the decision to use screws. Register data do not provide any information regarding the reason why a specific implant was chosen in the individual case. Perhaps surgeons chose cups with the possibility of augmentation with screws in more difficult cases. Excluding all diagnoses other than primary OA reduces the variety of some of these factors. A comparison of the included hospitals showed significant differences in the use of cups with screw holes. It is unlikely that these differences between hospitals can be explained fully by case mix.

Another limitation of our study is that only 2 of the cup designs, with unequal distribution between the groups, had a follow-up time of more than 10 years. Therefore, no reliable analysis was possible beyond 10 years of follow-up, and general conclusions about the long-term consequences of using cups with screw holes should be made with caution. A new analysis of the registry data will be necessary when a sufficient number of cases with several different cup designs have been followed for more than 10–15 years to obtain robust long-term data.

17% of patients were registered with bilateral cups. This subgroup decreases the variance within the groups and can increase the risk for type 1 errors. However, we did not find a statistically significant difference in HR for revision between

Acta Orthopaedica 2019; 90 (3): 258–263

cups in patients with unilateral THA and cups in patients with bilateral THA. The incidence of aseptic loosening was very low, making it statistically uncertain to adjust for a large number of potential confounders. The larger number of revisions for any reason gives better statistical strength.

In conclusion, we found that the revision rate for modern and frequently used uncemented cups was very low, and cup revision due to aseptic loosening within 2 years was extremely rare. We could not show that the use of cups designed for additional fixation with screws had any advantages in standard patients. In contrast, cups with screw holes increased the risk of cup revision for any reason. Notably, our study mainly embraces the first decade after the operation. Longer followup is needed to evaluate whether this conclusion remains valid during the second decade.

Supplementary data Tables 2, 4–5 are available as supplementary data in the online version of this article, http://dx.doi.org/10.1080/17453674.

2019.1599777

VO: study design, data collection, statistical analysis, data analysis, manuscript writing. SM: statistical analysis, data analysis, manuscript writing. KN and SC: data analysis, manuscript writing. JK: study design, data collection, statistical analysis, data analysis, manuscript writing, study supervision.

Acta thanks Ross W Crawford for help with peer review of this study.

Aspenberg P, van der Vis H. Fluid pressure may cause periprosthetic osteolysis: particles are not the only thing. Acta Orthop 1998; 69(1): 1-4.

Charnley J, Kamangar A, Longfield M. The optimum size of prosthetic heads in relation to the wear of plastic sockets in total replacement of the hip.

Medical Biol Eng 1969; 7(1): 31-9.

Colacchio N D, Cleary M X, Reid D, Trofa D, Pevear M E, Smith E L J C. Direct anterior approach total hip arthroplasty requires less supplemental acetabular screw fixation and fewer blood transfusions than posterior approach. Curr Orthop Pract 2017; 28(4): 404-8.

Cross M B, Nam D, Mayman D J. Ideal femoral head size in total hip arthroplasty balances stability and volumetric wear. HSS J 2012; 8(3): 270-4.

Dahl J, Soderlund P, Nivbrant B, Nordsletten L, Rohrl S M. Less wear with aluminium-oxide heads than cobalt-chrome heads with ultra high molecular weight cemented polyethylene cups: a ten-year follow-up with radiostereometry. Int Orthop 2012; 36(3): 485-90. doi: 10.1007/s00264-011-1334-3.

Hailer N. 20 years of porous tantalum in primary and revision hip arthroplasty: time for a critical appraisal. Acta Orthop 2018; 89(3): 254-5.

Hailer N P, Weiss R J, Stark A, Kärrholm J. The risk of revision due to dislocation after total hip arthroplasty depends on surgical approach, femoral head size, sex, and primary diagnosis: an analysis of 78,098 operations in the Swedish Hip Arthroplasty Register. Acta Orthop 2012; 83(5): 442-8.

Howie D W, Holubowycz O T, Callary S A. The wear rate of highly crosslinked polyethylene in total hip replacement is not increased by large articulations: a randomized controlled trial. J Bone Joint Surg Am 2016; 98(21): 1786-93. doi: 10.2106/jbjs.15.01248.

Hsu J-T, Chang C-H, Huang H-L, Zobitz M E, Chen W-P, Lai K-A, An K-N. The number of screws, bone quality, and friction coefficient affect acetabular cup stability. Med Eng Phys 2007; 29(10): 1089-95.

Acta Orthopaedica 2019; 90 (3): 258–263

Iorio R, Puskas B, Healy W L, Tilzey J F, Specht L M, Thompson M S. Cementless acetabular fixation with and without screws: analysis of stability and migration. J Arthroplasty 2010; 25(2): 309-13. doi: http://dx.doi. org/10.1016/j.arth.2009.01.023.

Kärrholm J, Lindahl H, Malchau H, Mohaddes M, Nemes S, Rogmark C, Rolfson O. The Swedish Hip Arthroplasty Register Annual Report 2016; 2017.

Laaksonen I, Lorimer M, Gromov K, Eskelinen A, Rolfson O, Graves S E, Malchau H, Mohaddes M. Trabecular metal acetabular components in primary total hip arthroplasty: higher risk for revision compared with other uncemented cup designs in a collaborative register study including 93,709 hips. Acta Orthop 2018; 89(3): 259-64.

Lazarinis S, Karrholm J, Hailer N P. Increased risk of revision of acetabular cups coated with hydroxyapatite. Acta Orthop 2010; 81(1): 53-9. doi: 10.3109/17453670903413178.

Lazarinis S, Karrholm J, Hailer N P. Effects of hydroxyapatite coating of cups used in hip revision arthroplasty. Acta Orthop 2012; 83(5): 427-35. doi: 10.3109/17453674.2012.720117.

Lazarinis S, Mäkelä K T, Eskelinen A, Havelin L, Hallan G, Overgaard S, Pedersen A B, Kärrholm J, Hailer N P. Does hydroxyapatite coating of uncemented cups improve long-term survival? An analysis of 28,605 primary total hip arthroplasty procedures from the Nordic Arthroplasty Register Association (NARA). Osteoarthritis Cartilage 2017; 25: 1980-7.

Minten M J, Heesterbeek P J, Spruit M. No effect of additional screw fixation of a cementless, all-polyethylene press-fit socket on migration, wear, and

263

clinical outcome: a 6.5-year randomized radiostereometric analysis followup report. Acta Orthop 2016; 87(4): 363-7.

Ni S H, Guo L, Jiang T L, Zhao J, Zhao Y G. Press-fit cementless acetabular fixation with and without screws. Int Orthop 2013. doi: 10.1007/s00264013-2075-2. Epub 2013 Aug 28.

Ohashi H, Kikuchi S, Aota S, Hakozaki M, Konno S. Surgical anatomy of the pelvic vasculature, with particular reference to acetabular screw fixation in cementless total hip arthroplasty in Asian population: a cadaveric study. J Orthop Surg (Hong Kong) 2017; 25(1): 2309499016685520. doi: 10.1177/2309499016685520.

Otten V T, Crnalic S, Röhrl S M, Nivbrant B, Nilsson K G. Stability of uncemented cups—long-term effect of screws, pegs and HA coating: a 14-Year RSA follow-up of total hip arthroplasty. J Arthroplasty 2016; 31(1): 156-61.

Pepe M, Kocadal O, Erener T, Ceritoglu K, Aksahin E, Aktekin C N. Acetabular components with or without screws in total hip arthroplasty. World J Orthop 2017; 8(9): 705.

Stilling M, Rahbek O, Søballe K. Inferior survival of hydroxyapatite versus titanium-coated cups at 15 years. Clin Orthop Relat Res 2009; 467(11): 2872-9.

VanderWeele T J, Ding P. Sensitivity analysis in observational research: introducing the e-value. Ann Intern Med 2017; 167(4): 268-74. doi: 10.7326/ M16-2607.

Won C H, Hearn T, Tile M. Micromotion of cementless hemispherical acetabular components. Does press-fit need adjunctive screw fixation? Bone Joint J 1995; 77(3): 484-9.

264

Acta Orthopaedica 2019; 90 (3): 264–269

Increased early mortality and morbidity after total hip arthroplasty in patients with socioeconomic disadvantage: a report from the Swedish Hip Arthroplasty Register Rüdiger J WEISS 1,2, Johan KÄRRHOLM 2,3, Ola ROLFSON 2,3, and Nils P HAILER 2,4 1 Department of Molecular Medicine and Surgery, Section of Orthopaedics and Sports Medicine, Karolinska Institutet, Karolinska University Hospital, Stockholm; 2 Swedish Hip Arthroplasty Register, Gothenburg; 3 Department of Orthopaedics, Institute of Clinical Sciences, Sahlgrenska Academy, University of Gothenburg, Gothenburg; 4 Department of Surgical Sciences, Section of Orthopaedics, Uppsala University Hospital, Uppsala, Sweden Correspondence: rudiger.weiss@sll.se Submitted 2018-11-29. Accepted 2019-02-16.

Background and purpose — Socioeconomic status is associated with the outcome of major surgery. We investigated the association of socioeconomic status with the risk of early mortality and readmissions after primary total hip arthroplasty (THA).

Patients and methods — We obtained information on income, education, immigration, and cohabiting status as well as comorbidities of 166,076 patients who underwent primary THA due to primary osteoarthritis (OA) from the Swedish Hip Arthroplasty Register, the Swedish National Inpatient Register and Statistics Sweden. Multivariable Cox regression models were fitted to estimate the adjusted risk of mortality or readmissions within 90 days after index surgery.

Results — Compared with patients on a low income, the adjusted risk of 30-day mortality was considerably lower in patients on a high income (hazard ratio [HR] 0.5, 95% confidence interval [CI] 0.3–0.7) and in those on a medium income (HR 0.7, CI 0.6–0.9). Similar risk reductions were found for the endpoint 90-day mortality. Patients with a high income had a lower adjusted risk of readmission for cardiovascular reasons than those with a low income (HR 0.7, CI 0.6–0.9), as had those with a higher level of education (adjusted HR 0.7, CI 0.6–0.9). Patients with higher socioeconomic status had a lower degree of comorbidities than socioeconomically disadvantaged patients. However, adjusting for socioeconomic confounders in multivariable models only marginally influenced the predictive ability of the models, as expressed by their area under the curve.

Interpretation — Income and level of education are strongly associated with early mortality and readmissions after primary THA, and both parameters are closely connected to health status. Since adjustment for socioeconomic confounders only marginally improved the predictive ability of multivariable regression models our findings indicate that comorbidities may under certain circumstances serve as an acceptable proxy measure of socioeconomic background.

Socioeconomic status is strongly associated with both access to and the outcome after major surgical interventions, including total joint replacements that represent one of the most common surgical interventions in developed countries (Kurtz et al. 2007, Hunt et al. 2013, Berstock et al. 2014, Katz et al.

2017).

The relationship between health and socioeconomic status is complex since socially deprived total hip arthroplasty (THA) patients have greater comorbidity and a higher prevalence of smoking compared with more affluent patients (Jenkins et al.

2009, Clement et al. 2011). Even in a European healthcare system with less pronounced social inequality, socioeconomically disadvantaged individuals are less healthy (Padyab et al.

2013). It was therefore postulated almost 2 decades ago that “performance measures should be stratified by socioeconomic position…” (Fiscella et al. 2000), but such stratification is rarely presented in studies on short-term complications after THA surgery.

Therefore we examined (1) how personal income, level of education, cohabiting, and immigration status affect early mortality and readmission rates after primary THA surgery due to OA, and (2) whether adjustment for such socioeconomic confounders substantially changes the attained risk estimates and the predictive ability of regression models.

Early mortality was defined as 30- and 90-day mortality, and hospital readmissions were divided into those that occurred for cardiovascular disease, or for any reason.

Patients and methods Data sources We extracted all patients operated with a primary THA from the Swedish Hip Arthroplasty Register (SHAR) 1992–2012 but excluded patients operated for reasons other than primary OA of the hip (Figure 1, see Supplementary data). To avoid

© 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group, on behalf of the Nordic Orthopedic Federation. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits ­unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

DOI 10.1080/17453674.2019.1598710

Acta Orthopaedica 2019; 90 (3): 264–269

dependency issues only the first operation was accounted for in patients with bilateral THA. National coverage of the SHAR is complete and completeness of registration has been found to be stable around 96–98% (SHAR 2015, Soderman et al. 2000).

Age at index surgery was divided into 3 age groups (< 60, 60–75, and > 75 years). Socioeconomic factors including personal income and level of education were obtained from Statistics Sweden. Income was divided into low, middle, and high along tertiles, and level of education was divided into the categories low (9 years or less of school education), middle (high school level), and higher education (university level). Hospital type at the index operation was divided into university, county, and private hospitals. Immigration status was categorized as non-immigrant or immigrant (defined as a patient born outside of Sweden). Cohabiting status was categorized as living with a partner, irrespective of marital status, or living alone.

Data on medical comorbidities, dates of death, and readmissions (for cardiovascular reasons or for any reason) were obtained from the Swedish National Inpatient Register (SNPR) where data on contacts with healthcare providers are recorded together with International Classification of Diseases (ICD)and procedural codes (Ludvigsson et al. 2011). Categories of comorbidity were computed using the Charlson Comorbidity Index (CCI) (Charlson et al. 1987, de Groot et al. 2003) by analysing registered ICD codes from the year preceding the index procedure but excluding the inpatient period during which the index procedure was performed. Comorbidity was categorized into 3 levels: low (CCI 0), moderate (CCI 1 or 2), and high (CCI > 2) (Deyo et al. 1992, Quan et al. 2011).

Patients who died or emigrated during follow-up were identified through the Total Population Register (Statistics Sweden).

Outcomes Primary outcome was early mortality (30- and 90-day mortality), which was defined as death occurring within 30 or 90 days after the index procedure. Secondary endpoints were readmission due to cardiovascular reasons or for any reason within the first 90 days after the index procedure. Cardiovascular disease was defined as myocardial infarction, chronic heart failure, peripheral vascular disease, and/or cerebrovascular disease (Quan et al. 2011).

Statistics Follow-up started on the day of surgery and ended on the day of death, emigration, or censorship at December 31, 2012, whichever came first. Continuous data were described using medians and ranges; observed and expected frequencies were analysed using the chi-square test. 95% confidence intervals (CI) were used to describe estimation uncertainty.

Kaplan–Meier survival analysis was performed to estimate unadjusted survival. Cox proportional hazard models were fitted for each primary or secondary outcome at a time to calculate hazard ratios (HR) with CI, and confounders were

265

Table 1. Description of the study population Factor

Number %

Total 166,076 Women 93,855 57 Men 72,221 43 Median age at index surgery (range) 70 (16–100) Median follow-up, years (range) 7 (0–22) Year of surgery 1992–1998 44,746 27 1999–2005 54,222 33 2006–2012 67,108 40 Income Low 57,397 35 Middle 55,151 33 High 53,528 32 Education Low 78,422 47 Middle 57,594 35 Higher 30,060 18 Cohabiting status Cohabiting 97,257 59 Non-cohabiting 68,819 41 Immigration status Non-immigrant 153,403 92 Immigrant 12,673 8 Charlson Comorbidity Index Low 139,197 84 Moderate 23,925 14 High 2,954 2 Hospital type County 125,801 76 Private 21,342 13 University 18,933 11 Implant fixation Cemented 136,554 82 Uncemented 13,233 8 a Other 16,289 10 Revision after index surgery No 157,698 95 Yes 8,378 5 a Hybrid,

reversed hybrid, and resurfacing.

subsequently included in multivariable regression models to calculate adjusted HR. The choice of covariates included in multivariable regression models was based on assessment of relevance and non-interference using directed acyclic graphs.

The assumption of proportionality of hazards was investigated graphically and by calculating scaled Schoenfeld residuals.

We used C-statistics to assess the area under the curve (AUC) and thus the ability of a given model to predict outcomes.

The observation period was divided into 3 different time periods (1992–1998, 1999–2005 and 2006–2012) to investigate temporal trends in the demography, comorbidity, surgical technique, and mortality in the investigated cohort.

The level of statistical significance was set at p < 0.05, but CIs were consistently used to assess estimation uncertainty.

Characteristics of the study population We identified 166,076 patients (57% females) with primary OA operated with a THA at a median age at surgery of 70

266

Acta Orthopaedica 2019; 90 (3): 264–269

Table 2. Hazard ratio (HR) for mortality up to 30 days after THA

Crude Adjusted Factor % n HR (95% CI) p-value HR (95% CI) p-value Age < 60 years 0.1 16 Ref. Ref. 60–75 years 0.1 115 2.2 (1.3–3.7) 0.003 1.7 (1.0–2.8) > 75 years 0.5 264 9.3 (5.6–15.4) < 0.001 5.1 (3.1–8.6) Sex Female 0.2 176 Ref. Ref. Male 0.3 219 1.6 (1.3–2.0)) < 0.001 1.9 (1.5–2.4) Year of surgery 1992–1998 0.4 174 Ref. Ref. 1999–2005 0.2 123 0.6 (0.5–0.7) < 0.001 0.6 (0.5–0.8) 2006–2012 0.1 98 0.4 (0.3–0.5) < 0.001 0.3 (0.3–0.4) Income Low 0.4 213 Ref. Ref. Middle 0.2 127 0.6 (0.5–0.8) < 0.001 0.7 (0.6–0.9) High 0.1 55 0.3 (0.2–0.4) < 0.001 0.5 (0.3–0.7) Education Low 0.3 261 Ref. Ref. Middle 0.2 100 0.5 (0.4–0.7) < 0.001 0.9 (0.7–1.1) Higher 0.1 34 0.3 (0.2–0.5) < 0.001 0.8 (0.6–1.2) Cohabiting status Cohabiting 0.2 191 Ref Ref. Non-cohabiting 0.3 204 1.5 (1.2–1.8) < 0.001 1.4 (1.1–1.8) Immigration status Non-immigrant 0.2 374 Ref. Ref. Immigrant 0.2 21 0.7 (0.4–1.1) 0.1 0.8 (0.5–1.2) Charlson Comorbidity Index Low 0.1 193 Ref. Ref. Moderate 0.7 159 4.8 (3.9–5.9) < 0.001 4.4 (3.6–5.4) High 1.5 43 10.6 (7.6–14.7) < 0.001 9.9 (7.0–13.9) Hospital type County 0.3 314 Ref. Ref. Private 0.1 25 0.5 (0.3–0.7) < 0.001 0.9 (0.6–1.3 University 0.3 56 1.2 (0.9–1.6) 0.2 1.0 (0.8–1.4

0.05 < 0.001 < 0.001 0.001 < 0.001 0.006 < 0.001 0.4 0.4 0.001

Within the first 30 days after surgery 395 (0.2%) patients died, giving an unadjusted 30-day survival of 99.8% (CI 99.7–99.8).

709 (0.4%) patients died within 90 days, resulting in an unadjusted 90-day survival of 99.6% (CI 99.5–99.6). We identified 1,208 readmissions for cardiovascular reasons within 90 days (0.7% of the study population), and 27,197 (16% of the study population) were readmitted to hospital for any reason within this time frame.

Ethics, funding, and potential conflicts of interests Ethical approval was granted by the Regional Ethical Review Board in Gothenburg (approval number: 2013/360–13). No external funding was received. No competing interests were declared.

Results

Socioeconomic status and early postoperative mortality When compared with patients in the low< 0.001 income category, patients in the high-income < 0.001 category had about half the risk of death within 30 days after THA surgery (adjusted 0.5 HR 0.5, CI 0.3–0.7), and patients in the 0.9 middle-income category also had a lower Number of events = 395; adjusted for age, sex, year of surgery, income, education, adjusted risk with an HR of 0.7 (CI 0.6–0.9; cohabiting status, immigration status, Charlson Comorbidity Index and hospital type.

Table 2). 90-day mortality showed similar risk reductions for patients in the high- and middle-income categories when compared with those in the lowest income category Table 3. Level of income and education divided by Charlson Comor(Table 4, see Supplementary data). In contrast, the level of edubidity Index. Values are frequency (%) cation was not associated with statistically significant variation in the adjusted risk of 30- or 90-day mortality after surgery

Charlson Comorbidity Index (Tables 2 and 4, see Supplementary data). Patients who were Factor Low Moderate High non-cohabiting had an increased risk of both 30- (adjusted HR Income 1.4, CI 1.1–1.8) and 90-day (adjusted HR 1.6, CI 1.4–1.9) morLow 47,581 (83) 8,822 (15) 994 (2) tality when compared with patients living with a partner.

Middle 45,440 (82) 8,559 (16) 1,152 (2) High Level of education Low Middle Higher

46,176 (86)

6,544 (12)

808 (2)

64,849 (83) 48,442 (84) 25,906 (86)

12,098 (15) 8,119 (14) 3,708 (12)

1,475 (2) 1,033 (2) 446 (2)

years (16–100). About half of the patients had a low level of education (47%; Table 1). Patients with a higher level of education or high income had a slightly lower level of comorbidities (Table 3).

0.3

Readmissions within 90 days Patients in the high-income category had a lower risk of readmission for cardiovascular reasons than those in the lowincome category (adjusted HR 0.7, CI 0.6–0.9), and a similar risk reduction was found for patients with a higher level of education (adjusted HR 0.7, CI 0.6–0.9). Patients living without a partner had an increased adjusted risk of readmission (HR 1.2, CI 1.1–1.4) compared with those living with a partner. Immigrants had a slightly increased risk of readmission for cardiovascular reasons compared with non-immigrants, but this find-

Acta Orthopaedica 2019; 90 (3): 264–269

ing failed to reach the level of statistical significance (HR 1.2, CI 1.0–1.5, p = 0.09; Table 5, see Supplementary data).

Patients with a high income (adjusted HR 0.9, CI 0.9–0.9) had a reduced risk for readmission for any reason, whereas patients living alone (adjusted HR 1.4, CI 1.3–1.4) and immigrants had an increased risk for readmission for any reason (adjusted HR 1.2, CI 1.1–1.2) (Table 6 and Figure 2, see Supplementary data).

Access to healthcare and temporal changes in demography, comorbidity, and mortality The majority of the patients were operated in county hospitals (76%), followed by private hospitals (13%) and university hospitals (11%). A higher than expected proportion of patients with a high income (observed count n = 11,035, expected count n = 6,879; p < 0.001 derived from chi-square test) and of those with a higher level of education (observed count n = 6,692, expected count n = 3,863; p < 0.001) were operated upon in private hospitals. Patients operated in private hospitals had a lower adjusted risk of 90-day mortality (adjusted HR 0.7, CI 0.5–1.0; Table 4, see Supplementary data) when compared with patients in county hospitals.

Patients operated during 1999–2005 had a lower risk of early death compared with those operated 1992–1998, and those who had THA surgery during the latest period of 2006–2012 had the lowest risk of all sub-cohorts (Tables 2 and 4, see Supplementary data).

Impact of adjustment for socioeconomic factors on the predictive ability of multivariable models The regression model fitted for the endpoint 90-day mortality that included the covariates age, sex, comorbidity, income, and education had an AUC of 0.77 (CI 0.76–0.79), whereas the model without the covariates income and education had an AUC of 0.76 (CI 0.74–0.78). For the regression model fitted for the endpoint readmission for cardiovascular reasons within 90 days that included the covariates age, sex, comorbidity, income, and education we estimated an AUC of 0.74 (CI 0.73–0.76), and we attained a similar estimate of 0.73 (CI 0.72–0.75) for the model without the covariates income and education.

Discussion Main findings This study indicates that socioeconomic disadvantage is associated with an increased risk of both early postoperative mortality and early readmission after THA surgery. However, we also found that adjustment for socioeconomic status results in small changes in the attained risk estimates and only slightly improved C-statistics when modelling early mortality and readmissions after THA.

267

Limitations and strengths Our study is subject to incompleteness of data registration and registration errors. Misclassification bias is another limitation since our estimates of comorbidity and causes for readmissions are entirely based upon ICD codes registered by healthcare providers. If there are changes or errors in the coding of diagnoses, these will translate into erroneous measures of comorbidity.

For instance, the comorbidity of patients analyzed in this study seems to have progressed over time (Table 7 and Supplementary data), a finding that could be related to an actual increase in the proportion of patients suffering from a higher burden of comorbidities during the last decade. On the other hand, caregiver reimbursement based on the DRG system may also have triggered a more zealous way of coding comorbidities, a phenomenon described as “physician code creep” (Seiber 2007).

We lack data on lifestyle-related confounders of our main outcome measures, such as body mass index, smoking, alcohol habits, and physical activity. Since these confounders may differ between socioeconomic strata they can profoundly influence the observed associations between socioeconomic variables and our outcome measures.

Mutually adjusted effect estimates derived from multivariable regression models are presented in our analysis. It has to be noted that the presented risk estimates for specific confounders can either represent total-effect or direct-effect estimates, and effect estimates related to the primary exposure cannot be interpreted in the same way as effect estimates of secondary risk factors (Westreich and Greenland 2013). We attempted to limit the problems associated with the presence of effect mediators and colliders by devising directed acyclic graphs prior to regression analyses, but the interdependency of socioeconomic status and comorbidity is strong and our reported effect estimates of secondary risk factors must therefore be interpreted with caution.

Strengths of our study includes the prospective data collection within 3 well-validated, population-based registers, creating high external validity. We chose to include only patients with primary OA of the hip as the reason for THA insertion, an import selection since some patients receiving THA for other reasons, such as those with a fracture of the femoral neck, are older and have a high burden of comorbidity, whereas those receiving a THA due to sequelae of childhood hip disorders are often younger and, possibly, fitter than those with OA. Taken together, by restricting our study population to patients with primary OA of the hip we avoided a large amount of residual confounding that would have been difficult to adjust for. The combined in-depth information on income, achieved level of education, and ethnicity is an additional valuable facet of our study, since this enables investigation of socioeconomic factors on a more profound level.

Main findings related to previous literature The overall 30- and 90-day mortality was low in our study, which is in accordance with other studies analysing early

268

mortality after primary THA surgery for OA (Aynardi et al.

2009, Berstock et al. 2014). In our study, lower socioeconomic status was associated with increased 90-day mortality, which is also in agreement with US studies on this topic, although the odds ratio in 1 of these studies was slightly higher than in ours (Mahomed et al. 2003, Clement et al.

2011), an effect that is possibly attributable to larger disparity in income and education in the United States compared with Sweden.

It is described that socioeconomically deprived patients operated with a THA have greater comorbidity than more affluent patients, and deprived patients also have a higher prevalence of smoking, which has a profound impact on cardiovascular and respiratory comorbidity (Jenkins et al.

2009). Other studies indicate that patients with low incomes have a higher risk of acute adverse medical events after THA surgery (Agabiti et al. 2007). The association of lower socioeconomic status with higher mortality among hip fracture patients is known (Barone et al. 2009, Kristensen et al.

2017). Our finding of an association between early mortality after THA and non-cohabiting status is an expected finding when considering the fact that widowers have an increased mortality compared with married individuals, and that married persons have a longer life expectancy compared with unmarried persons (Hu and Goldman 1990, Shor et al. 2012).

To have information on this parameter is a strength of our analysis; this facet of socioeconomic status is rarely reported in other studies.

We found a readmission rate for any reason of 16% within 90 days after surgery, which can be compared with a prevalence of 30-day readmissions of 7% after hip replacements in a US study (Brooke et al. 2015). A study on the risk of death, readmissions to hospital, and wound infections after elective primary THA in United States Medicare patients supports our findings (Mahomed et al. 2003).

THA patients in the higher socioeconomic strata were overrepresented in private hospitals. Some authors claim that wealthier patients have easier access to private sector care, which is associated with shorter waiting times (Agabiti et al. 2007), and this seems also to be true within the Swedish healthcare system.

The ability of the adjusted regression models to predict their respective outcomes was investigated by assessing model C-statistics, a term used to describe the AUC. We found AUC that varied between 0.7 and 0.8, which indicates a relatively good predictive capacity when compared with an investigation of 3 different comorbidity measures to predict the incidence of reoperations within 2 years after primary THA surgery, where the maximal AUC was 0.5 (Gordon et al. 2013). However, when models without adjustment for socioeconomic background variables were re-fitted we attained similar AUC to the previously fitted models with adjustment for socioeconomic background.

Acta Orthopaedica 2019; 90 (3): 264–269

Conclusion Our findings support the notion that patients with socioeconomic disadvantage run a substantially higher risk of both early mortality and readmissions after THA surgery, even within the setting of a relatively egalitarian healthcare system.

However, perhaps partly due to the fact that poor socioeconomic status is associated with a higher degree of comorbidities, adjusting for comorbidities may suffice in analyses where detailed information on socioeconomic background is lacking.

Supplementary data Tables 4–7 and Figures 1–2 are available as supplementary data in the online version of this article, http://dx.doi.org/ 10.1080/17453674.2019.1598710 The authors would like to thank all Swedish orthopaedic surgeons and secretaries who contributed data.

RJW and NPH: study design, database preparation, data analysis, manuscript drafting and editing. JK and OR: study design and final manuscript editing.

Acta thanks Per Kjærsgaard-Andersen and Stephan Maximilian Röhrl for help with peer review of this study.

Agabiti N, Picciotto S, Cesaroni G, Bisanti L, Forastiere F, Onorati R, Pacelli B, Pandolfi P, Russo A, Spadea T, Perucci C A, Italian Study Group on Inequalities in Health C. The influence of socioeconomic status on utilization and outcomes of elective total hip replacement: a multicity population-based longitudinal study. Int J Qual Health Care 2007; 19 (1): 37-44.

Aynardi M, Pulido L, Parvizi J, Sharkey P F, Rothman R H. Early mortality after modern total hip arthroplasty. Clin Orthop Relat Res 2009; 467 (1): 213-18.

Barone A P, Fusco D, Colais P, D’Ovidio M, Belleudi V, Agabiti N, Sorge C, Davoli M, Perucci C A. Effects of socioeconomic position on 30-day mortality and wait for surgery after hip fracture. Int J Qual Health Care 2009; 21 (6): 379-86.

Berstock J R, Beswick A D, Lenguerrand E, Whitehouse M R, Blom A W. Mortality after total hip replacement surgery: a systematic review. Bone Joint Res 2014; 3 (6): 175-82.

Brooke B S, Goodney P P, Kraiss L W, Gottlieb D J, Samore M H, Finlayson S R. Readmission destination and risk of mortality after major surgery: an observational cohort study. Lancet 2015; 386 (9996): 884-95.

Charlson M E, Pompei P, Ales K L, MacKenzie C R. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chron Dis 1987; 40 (5): 373-83.

Clement N D, Muzammil A, Macdonald D, Howie C R, Biant L C. Socioeconomic status affects the early outcome of total hip replacement. J Bone Joint Surg Br 2011; 93 (4): 464-9. de Groot V, Beckerman H, Lankhorst G J, Bouter L M. How to measure comorbidity. a critical review of available methods. J Clin Epidemiol 2003; 56 (3): 221-9.

Deyo R A, Cherkin D C, Ciol M A. Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Epidemiol 1992; 45 (6): 613-9.

Fiscella K, Franks P, Gold M R, Clancy C M. Inequality in quality: addressing socioeconomic, racial, and ethnic disparities in health care. JAMA 2000; 283 (19): 2579-84.

Acta Orthopaedica 2019; 90 (3): 264â&#x20AC;&#x201C;269

Gordon M, Stark A, Skoldenberg O G, Karrholm J, Garellick G. The influence of comorbidity scores on re-operations following primary total hip replacement: comparison and validation of three comorbidity measures.

Bone Joint J 2013; 95-B (9): 1184-91.

Hu Y R, Goldman N. Mortality differentials by marital status: an international comparison. Demography 1990; 27 (2): 233-50.

Hunt L P, Ben-Shlomo Y, Clark E M, Dieppe P, Judge A, MacGregor A J, Tobias J H, Vernon K, Blom A W, National Joint Registry for England W, Northern I. 90-day mortality after 409,096 total hip replacements for osteoarthritis, from the National Joint Registry for England and Wales: a retrospective analysis. Lancet 2013; 382 (9898): 1097-104.

Jenkins P J, Perry P R, Yew Ng C, Ballantyne J A. Deprivation influences the functional outcome from total hip arthroplasty. The Surgeon: J Roy Coll Surg Edinburgh and Ireland 2009; 7 (6): 351-6.

Katz J N, Winter A R, Hawker G. Measures of the appropriateness of elective orthopaedic joint and spine procedures. J Bone Joint Surg Am 2017; 99 (4): e15.

Kristensen P K, Thillemann T M, Pedersen A B, Soballe K, Johnsen S P. Socioeconomic inequality in clinical outcome among hip fracture patients: a nationwide cohort study. Osteoporos Int 2017; 28 (4): 1233-43.

Kurtz S, Ong K, Lau E, Mowat F, Halpern M. Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030.

J Bone Joint Surg Am 2007; 89 (4): 780-5.

Ludvigsson J F, Andersson E, Ekbom A, Feychting M, Kim J L, Reuterwall C, Heurgren M, Olausson P O. External review and validation of the Swedish national inpatient register. BMC Public Health 2011; 11: 450.

269

Mahomed N N, Barrett J A, Katz J N, Phillips C B, Losina E, Lew R A, Guadagnoli E, Harris W H, Poss R, Baron J A. Rates and outcomes of primary and revision total hip replacement in the United States medicare population. J Bone Joint Surg Am 2003; 85-A (1): 27-32.

Padyab M, Malmberg G, Norberg M, Blomstedt Y. Life course socioeconomic position and mortality: a population register-based study from Sweden.

Scand J Public Health 2013; 41 (8): 785-91.

Quan H, Li B, Couris C M, Fushimi K, Graham P, Hider P, Januel J M, Sundararajan V. Updating and validating the Charlson comorbidity index and score for risk adjustment in hospital discharge abstracts using data from 6 countries. Am J Epidemiol 2011; 173 (6): 676-82.

SHAR. Swedish Hip Arthroplasty Register, Annual Report 2015, http://www. shpr.se/en/.

Seiber E E. Physician code creep: evidence in Medicaid and State Employee Health Insurance billing. Health Care Financing Rev 2007; 28 (4): 83-93.

Shor E, Roelfs D J, Curreli M, Clemow L, Burg M M, Schwartz J E. Widowhood and mortality: a meta-analysis and meta-regression. Demography 2012; 49 (2): 575-606.

Soderman P, Malchau H, Herberts P, Johnell O. Are the findings in the Swedish National Total Hip Arthroplasty Register valid? A comparison between the Swedish National Total Hip Arthroplasty Register, the National Discharge Register, and the National Death Register. J Arthroplasty 2000; 15 (7): 884-9.

Westreich D, Greenland S. The table 2 fallacy: presenting and interpreting confounder and modifier coefficients. Am J Epidemiol 2013; 177 (4): 292-8.

270

Acta Orthopaedica 2019; 90 (3): 270–274

Cementing does not increase the immediate postoperative risk of death after total hip arthroplasty or hemiarthroplasty: a hospitalbased study of 10,677 patients Elina EKMAN 1, Inari LAAKSONEN 1, Kari ISOTALO 1, Antti LIUKAS 2, Tero VAHLBERG 3, and Keijo MÄKELÄ 1 1 Department of Orthopaedics and Traumatology, Turku University Hospital; 2 Department of Anaesthesiology and Intensive Care, University of Turku, Pohjola Hospital; 3 Department of Clinical Medicine, Biostatistics, University of Turku, Turku, Finland Correspondence: elina.ekman@tyks.fi Submitted 2018-11-26. Accepted 2019-02-21.

Background and purpose — It has been suggested that cemented arthroplasty is associated with increased periand postoperative mortality due to bone cement implanting syndrome, especially in fracture surgery. We investigated such an association in elective total hip arthroplasty (THA) patients and hemiarthroplasty (HA) patients treated for femoral neck fracture.

Patients and methods — All 10,677 patients receiving elective THA or HA for fracture in our hospital between 2004 and 2015 were identified. Mortality rates for cemented and uncemented THA and HA were compared at different times postoperatively using logistic regression analysis. Analysis was adjusted for age, sex, ASA class, and year of surgery.

Results — Adjusted 10- and 30-day mortality after cemented THA was comparable to that of the uncemented THA (OR 1.7; 95% CI 0.3–8.7 and OR 1.6; CI 0.7–3.6, respectively). There was no statistically significant difference in the adjusted 2-day mortality in the cemented HA group when compared with the uncemented group. However, in a subgroup analyses of ASA-class IV HA patients there was a difference, statistically not significant, during the first 2 days postoperatively in the cemented HA group compared with the uncemented HA group (OR 2.1; CI 0.9–4.7).

Interpretation — Cementing may still be a safe option in both elective and hip fracture arthroplasty. Excess mortality of cemented THA and HA in the longer term is comorbidity related, not due to bone cement implantation syndrome.

However, in the most fragile HA patient group caution is needed at the moment of cementing.

Total hip arthroplasty (THA) is considered a safe procedure and over recent years early postoperative mortality has decreased (Hunt et al. 2013). In a recent systematic review, mortality during the first 30 postoperative days after THA was 0.3% (Berstock et al. 2014).

Conversely, femoral neck fracture is associated with high peri- and postoperative mortality. 30-day mortality has been reported to be as high as 5–10% (Keating et al. 2006, Moja et al. 2012, Smith et al. 2014). This is largely explained by most of these patients being fragile with several comorbidities, such as cardiovascular diseases, cognitive impairment, and poor pre-fracture mobility.

THA can be performed as cemented, uncemented, hybrid (uncemented cup and cemented stem) or reverse hybrid (cemented cup and uncemented stem) whereas hemiarthroplasty can be performed with or without bone cement.

Cemented THA has superior implant survival rates compared with uncemented THA in long term follow-up in elderly patients (Issack et al. 2003, Buckwalter et al. 2006, Morshed et al. 2007, Mäkelä et al. 2014). In addition, national guidelines in Finland, Sweden, and the UK recommend cemented HA in femoral neck fracture (National Clinical Guideline 2011, Rogmark et al. 2014, Current Care Guidelines 2017). However, some surgeons hesitate to use bone cement due to the possibility of bone cement implantation syndrome (BCIS), which may cause cardiovascular disturbances, pulmonary embolism, and at worst death of the patient (Donaldson et al. 2009).

We investigated whether the use of bone cement is associated with higher immediate mortality in patients treated with

© 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group, on behalf of the Nordic Orthopedic Federation. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits ­unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

DOI 10.1080/17453674.2019.1596576

Acta Orthopaedica 2019; 90 (3): 270–274

THA or HA in Turku University Hospital.

We also assessed separately whether there is a difference in the early postoperative mortality in the most fragile (ASA-class IV–V) patients treated with cemented or uncemented HA.

Patients and methods

271

Table 1. Baseline characteristics for the patient cohorts

Number of patients Number of operations Percentage of females Mean age (SD) range ASA class 1–2, n (%) ASA class 3, n (%) ASA class 4, n (%)

Uncemented Cemented Uncemented Hybrid Cemented HA HA THA THA THA 1,142 1,173 67 81 (10) 37–107 115 (10) 707 (60) 349 (30)

1,868 1,935 74 83 (8) 40–103 194 (10) 1,234 (64) 501 (26)

4,855 5,563 56 65 (10) 13–96 3,303 (59) 2,163 (39) 86 (2)

612 640 68 75 (8) 38–91 258 (40) 363 (57) 18 (3)

1,274 1,366 73 77 (6) 42–97 494 (36) 819 (60) 52 (4)

Source of data This study was performed at Turku Univer- THA = total hip arthroplasty. HA = hemiarthroplasty. sity Hospital, Turku, Finland. Patients operated for OA, rheumatoid arthritis, psoriatic arthritis, juvenile arthritis, unspecified arthritis, or femoral to compare the mortality in the cemented HA group comneck fracture (ICD-10 codes M16.0 - M16.9, M05.8, M05.9, pared with the uncemented HA group and the mortality in the M06.0, M07.3, M08.0, M08.3, M13.9, S72.0) with unce- hybrid THA and cemented THA groups compared with the mented, cemented, or hybrid THA (ICD-10 codes NFB30, uncemented THA group. A random intercept logistic model NFB40, and NFB50) or with cemented or uncemented HA was used to account for the dependency between operations (ICD-10 code NFB10, NFB20) were included in the study. performed for the same patient. Analyses were adjusted for During the study period from January 1, 2004 to May 8, 2015, potential confounding factors age, gender, ASA class, and 7,569 primary THAs and 3,108 HAs were performed in Turku year of surgery. In addition, subgroup analysis for patients of University Hospital. For every patient the preoperative diag- ASA class IV was applied to compare the mortality between nosis and sex, age, and ASA class at the time of the operation the cemented and the uncemented HA groups. This analysis were recorded. Time of death was obtained from the National was also adjusted for age, sex, and year of surgery. P-values Causes of Death Statistics maintained by Statistics Finland. less than 0.05 were considered statistically significant. Statistical analyses were done with SAS System for Windows, verStudy population sion 9.4 (SAS Institute Inc., Cary, NC, USA).

Of the 7,569 primary THAs 74% were uncemented,18% cemented, and 8.5% hybrid. 60% of the THA operations Ethics, funding, and potential conflicts of interest were performed on women and the most common preopera- Ethical approval was granted by the Regional Ethical Review tive diagnosis was OA (75%). Of the HAs 38% were unce- Board in Turku (approval number THL/926/5.05.00/2017). mented and 62% were cemented. During the study period the This research received no specific grant from any funding use of cemented THA decreased in Finland whereas the use agency in the public, commercial, or not-for-profit sectors. of uncemented THA increased (Finnish Arthroplasty Register The authors declare no conflicts of interest.

[FAR] 2017). The use of uncemented HA increased a little while the use of cemented HA remained constant (Yli-Kyyny et al. 2014). In the HA group 71% of the operations were performed on women and all these operations were performed Results due to femoral neck fracture. In all study groups, ASA classes The number of deaths for each group is presented in Table 2.

I, IV, and V were highly uncommon and we therefore grouped There were no statistically significant differences in mortalASA classes I–II and IV–V together. The baseline characteris- ity at any time point when comparing hybrid THA with uncetics for the study groups are given in Table 1. mented THA (Table 3). There were no deaths during the 1st 2 A third-generation cementing technique (washing the bone days postoperatively in the uncemented THA group, 1 (0.2%) with pulsed lavage, use of an intramedullary plug and ret- in the hybrid group, and 3 (0.2%) in the cemented group (Table rograde insertion of the cement) was used in all operations. 2). There were more deaths in the cemented THA group when Simultaneous bilateral THAs were not included. There were comparing with the uncemented THA group after adjusting 1,123 patients who had both hips operated on; all operations the groups for age, sex, ASA class, and year of surgery at 180 were performed on a different day and they were included in days (OR 2.0; CI 1.0–3.7) postoperatively (Table 3). No stathe study as separate procedures. tistically significant difference was found at other time points.

With unadjusted data there were more deaths in the Statistics cemented HA group (50 deaths, 2.6%) when compared with Mortality at 2 days, 10 days, 30 days, 90 days, 180 days, and the uncemented HA group (22 deaths, 1.9%) during the 1st one year was assessed. Binary logistic regression was used 2 days postoperatively (Table 2). Of these patients 30 in the

272

Acta Orthopaedica 2019; 90 (3): 270–274

Table 2. Number of deaths

Uncemented Cemented Uncemented Hybrid Cemented HA HA THA THA THA

Discussion

We found no statistically significant difference in the adjusted early postoperative Number of operations 1,173 1,935 5,563 640 1,366 mortality after cemented THA compared Number of deaths, n (%) 0–2 days 22 (1.9) 50 (2.6) 0 1 (0.2) 3 (0.2) with uncemented or hybrid THA. Further, 0–10 days 61 (5.2) 99 (5.1) 4 (0.1) 2 (0.3) 5 (0.4) we found no statistically significant dif0–30 days 105 (9.0) 171 (8.9) 6 (0.1) 2 (0.3) 9 (0.7) ferences in the adjusted mortality between 0–90 days 173 (14.8) 303 (15.7) 17 (0.3) 5 (0.8) 19 (1.4) 0–180 days 227 (19.4) 388 (20.1) 29 (0.5) 8 (0.3) 29 (2.1) cemented and uncemented HA at any time 0–365 days 279 (23.8) 503 (26.0) 50 (0.9) 14 (2.2) 41 (3.0) point. However, in the subgroup analyses of ASA class IV HA patients there was a THA = total hip arthroplasty. HA = hemiarthroplasty. difference that did not quite reach our criteria for statistical significance during the 1st Table 3. Postoperative mortality risk for uncemented THA com2 days postoperatively in the cemented HA group compared pared with hybrid and cemented THA, OR , 95% CI, and p-value (adjusted for age, sex, ASA class, and year of surgery) with the uncemented HA group. Based on our results, cementing may be a safe option in both elective and fracture hip surgery. However, in the most fragile HA patient group caution is

Uncemented Hybrid THA Cemented THA

THA (ref.) OR (95% CI) p-value OR (95% CI) p-value needed at the moment of cementing.

Cementing is the gold standard for implant fixation, espe0–2 days 1 NA NA cially in elderly patients treated for femoral neck fractures.

0–10 days 1 3.2 (0.5–20.2) 0.2 1.7 (0.3–8.7) 0.5 0–30 days 1 1.7 (0.6–4.9) 0.4 1.6 (0.7–3.6) 0.3 Bone cement has been thought to strengthen bone from inside 0–90 days 1 1.7 (0.6–4.9) 0.4 1.6 (0.7–3.6) 0.3 and, therefore, to decrease the risk for periprosthetic fracture, 0–180 days 1 1.7 (0.7–3.9) 0.2 2.0 (1.0–3.7) 0.04 osteolysis, and loosening. All major registries show lower 0–365 days 1 1.5 (0.8–2.9) 0.2 1.6 (0.9–2.7) 0.08 revision rates for cemented implants in elderly patients with NA = Not available due to zero deaths during the first two days postOA (Swedish Hip Arthroplasty Register 2013, AOANJRR operatively in the uncemented THA group.

2016, NJR 2016, FAR 2017). Additionally, there is evidence that cementing the stem reduces postoperative pain and leads to better mobility (Parker et al. 2010). Cementing may also Table 4. Postoperative mortality risk for cemented HA compared decrease the risk of reoperation when compared with uncewith uncemented HA, OR, 95% CI, and p-value (adjusted for age, sex, ASA class, and year of surgery) mented hemiarthroplasty in hip fracture patients (Gjertsen et al. 2012, Yli-Kyyny et al. 2014). Due to these data, the proportion of cemented stems has been increasing recently and 62%

Uncemented Cemented THA

THA (ref.)

OR (95% CI) p-value of the HA patients in this study were cemented. Earlier studies reported that cementing of the hip device was associated 0–2 days 1 1.4 (0.8–2.3) 0.3 with a risk of BCIS increasing perioperative morbidity and 0–10 days 1 1.0 (0.7–1.4) 1.0 0–30 days 1 1.0 (0.8–1.3) 1.0 mortality (Coventry et al. 1974, Ereth et al. 1992, Parvizi et 0–90 days 1 1.1 (0.9–1.3) 0.5 al. 1999). It has been suggested that the risk of BCIS might be 0–180 days 1 1.0 (0.8–1.2) 0.9 increased in hip fracture patients who are, in general, old and 0–365 days 1 1.1 (0.9–1.3) 0.4 fragile and have several comorbidities (Keating et al. 2006, Moja et al. 2012). Improvements in surgical and anesthesiolcemented HA group and 10 in the uncemented group were ogy techniques and implants, the use of low molecular weight classified as ASA class IV. Age, sex, ASA class, and year of heparins (LMWHs) in the 1980s, and operating room sterility surgery adjusted mortality did not differ statistically signifi- have significantly reduced overall mortality risks associated cantly between the groups during the 1st 2 postoperative days with hip arthroplasty.

(OR 1.4; CI 0.8–2.3) (Table 4). In addition, there was no staThere are earlier studies reporting increased early postopertistically significant difference in the mortality rate between ative mortality in patients treated with cemented HA (Parvizi cemented and uncemented HA at any other time point either. et al. 1999, Yli-Kyyny et al. 2014). We found a higher proporIn the subgroup analyses of patients of ASA class IV there tion of perioperative deaths (0–2 days postoperatively) in the was a difference in mortality that did not reach statistical sig- cemented HA group than in the uncemented HA group. It is nificance during the 1st 2 postoperative days in the cemented possible that these numbers include deaths due to BCIS; noneHA group when compared with the uncemented HA group theless, this could not be confirmed as we did not have access (OR 2.1; CI 0.9–4.7). No statistically significant difference in to the cause of death. However, this difference vanished after adjusting data for age, sex, and ASA class, suggesting that mortality was found thereafter either.

Acta Orthopaedica 2019; 90 (3): 270–274

the difference was not due to cementing. This is in line with registry studies from Australia and UK where there has not been an increase in early postoperative mortality when comparing cemented and uncemented implants (Costa et al. 2011, Costain et al. 2011). Also, in studies reporting increased early postoperative mortality when using bone cement, the risk disappeared after the first postoperative week or even reversed to a lower mortality for those treated with a cemented prosthesis (Costain et al. 2011, Yli-Kyyny et al. 2014). In our study, in the most fragile patient group (ASA class IV) we found a tendency toward higher mortality during the 1st 2 days when bone cement was used. However, most of these patients survive the immediate postoperative period and long-term outcome may motivate cementation.

We found in the adjusted data an increased risk of death in patients treated with cemented THA when compared with patients treated with uncemented THA at 180 days postoperatively. This late mortality, however, is not explained by BCIS. It is probably due to the baseline differences in the treatment groups: patients treated with cemented THA were older than patients treated with uncemented THA. Our finding is in line with an earlier study that found no increase in mortality with cemented THA compared with uncemented THA during the first 30 postoperative days (Parvizi et al. 2001).

Our study has several limitations. First, we do not have information concerning perioperative resuscitations due to BCIS that did not lead to the patient’s death. It is possible that there is more morbidity due to cementing, which might affect to patient’s quality of life. Second, we did not have the causes of death. Therefore, we do not know the absolute number of deaths due to BCIS. However, we focused on overall mortality. Also, besides ASA class, we did not have information on patients’ comorbidities known to affect the risk of death (such as dementia or congestive heart failure) and therefore study groups could not be adjusted for these. Third, the early mortality rate after THA is low and it is possible that in a larger population some smaller differences in the mortality could be detected. Further, some surgeons may have hesitated to use bone cement due to the possibility of BCIS. This may cause some selection bias to our results, although we think of minor importance.

Lastly, that differences in the mortality rates did not reach statistical significance does not exclude excess mortality in a group, especially when confidence intervals are large. However, our data imply that it is unlikely bone cementing would increase mortality rates in THA patients, although we cannot exclude some excess mortality in HA patients.

In summary, there was no statistically significant difference in adjusted perioperative and short-term postoperative mortality between patients treated with cemented HA or THA and patients treated with uncemented HA or THA or hybrid THA in our material. Cementing may still be a safe option in both elective and fracture hip surgery. Excess mortality of cemented THA and HA in the longer term is comorbidity related, not due to BCIS.

273

KM designed and coordinated the study. EE collected the data and drafted the manuscript. IL helped to draft the manuscript. TV calculated the statistics. KM, IL, KI, AL, TV, and EE contributed to the interpretation of the data and results and to the preparation of the manuscript. All authors read the final manuscript.

Acta thanks Anne Garland and Sven-Erik Ricksten for help with peer review of this study.

AOANJRR. Annual Report 2016. AOANJRR; 2016.

Berstock J R, Beswick A D, Lenguerrand E, Whitehouse M R, Blom A W. Mortality after total hip replacement surgery: a systematic review. Bone Joint Res 2014; 3(6): 175-82.

Buckwalter A E, Callaghan J J, Liu S S, Pedersen D R, Leinen J A, Johnston R C, Goetz D D, Sullivan P M. Results of Charnley total hip arthroplasty with use of improved femoral cementing techniques. J Bone Joint Surg Am 2006; 88(7): 1481-5.

Costa M L, Griffin X L, Pendleton N, Pearson M, Parsons N. Does cementing the femoral component increase the risk of peri-operative mortality for patients having replacement surgery for a fracture of the neck of femur? J Bone Joint Surg Br 2011; 93-B(10): 1405-10.

Costain D J, Whitehouse S L, Pratt N L, Graves S E, Ryan P, Crawford R W. Perioperative mortality after hemiarthroplasty related to fixation method.

Acta Orthop 2011; 82(3): 275-81.

Coventry M B, Beckenbaugh R D, Nolan D R, Ilstrup D M. 2,012 total hip arthroplasties: a study of postoperative course and early complications. J Bone Joint Surg Am 1974; 56(2): 273-84.

Current Care Guidelines. Hip fracture. http://www.kaypahoito.fi/web/kh/ suositukset/suositus?id=hoi50040 Donaldson A J, Thomson H E, Harper N J, Kenny N W. Bone cement implantation syndrome. Br J Anaesth 2009; 102(1): 12-22.

Ereth M H, Weber J G, Abel M D, Lennon R L, Lewallen D G, Ilstrup D M, Rehder K. Cemented versus noncemented total hip arthroplasty: embolism, hemodynamics, and intrapulmonary shunting. Mayo Clin Proc 1992; 67(11): 1066-74.

Finnish Arthroplasty Register (FAR). http://www.thl.fi/far; 2017.

Gjertsen J-E, Lie S A, Vinje T, Engesæter L B, Hallan G, Matre K, Furnes O. More re-operations after uncemented than cemented hemiarthroplasty used in the treatment of displaced fractures of the femoral neck: an observational study of 11,116 hemiarthroplasties from a national register. J Bone Joint Surg Br 2012; 94(8): 1113-19.

Hunt L P, Ben-Shlomo Y, Clark E M, Dieppe P, Judge A, MacGregor A J, Tobias J H, Vernon K, Blom A W. 90-day mortality after 409 096 total hip replacements for osteoarthritis, from the National Joint Registry for England and Wales: a retrospective analysis. Lancet 2013; 382(9898): 1097-104.

Issack P S, Botero H G, Hiebert R N, Bong M R, Stuchin S A, Zuckerman J D, Di Cesare P E. Sixteen-year follow-up of the cemented spectron femoral stem for hip arthroplasty. J Arthroplasty 2003; 18(7): 925-30.

Keating J F, Grant A, Masson M, Scott N W, Forbes J F F, Forbes J F F. Randomized comparison of reduction and fixation, bipolar hemiarthroplasty, and total hip arthroplasty: treatment of displaced intracapsular hip fractures in healthy older patients. J Bone Joint Surg Am 2006; 88(2): 249-60.

Moja L, Piatti A, Pecoraro V, Ricci C, Virgili G, Salanti G, Germagnoli L, Liberati A, Banfi G. Timing matters in hip fracture surgery: patients operated within 48 hours have better outcomes. A meta-analysis and metaregression of over 190,000 patients. PLoS One 2012; 7(10): e46175.

Morshed S, Bozic K J, Ries M D, Malchau H, Colford J M. Comparison of cemented and uncemented fixation in total hip replacement: a meta-analysis. Acta Orthop 2007; 78(3): 315-26.

Mäkelä K T, Matilainen M, Pulkkinen P, Fenstad A M, Havelin L, Engesaeter L, Furnes O, Pedersen A B, Overgaard S, Kärrholm J, Malchau H, Garellick G, Ranstam J, Eskelinen A. Failure rate of cemented and uncemented total hip replacements: register study of combined Nordic database of four nations. BMJ 2014; 348(January): f7592.

274

National Clinical Guideline C. National Institute for Health and Clinical Excellence: Guidance. Manag Hip Fract Adults 2011. London: NIHCE; 2011.

NJR. NJR 13th Annual Report. Hemel Hempstead: National Joint Registry; 2016.

Parker M J, Gurusamy K S, Azegami S. Arthroplasties (with and without bone cement) for proximal femoral fractures in adults. Cochrane Database Syst Rev 2010; (6): CD001706.

Parvizi J, Holiday A D, Ereth M H, Lewallen D G. The Frank Stinchfield Award. Sudden death during primary hip arthroplasty. Clin Orthop Relat Res 1999; (369): 39-48.

Parvizi J, Johnson B G, Rowland C, Ereth M H, Lewallen DG. Thirty-day mortality after elective total hip arthroplasty. J Bone Joint Surg Am 2001; 83-A(10): 1524-8.

Acta Orthopaedica 2019; 90 (3): 270–274

Rogmark C, Fenstad A M, Leonardsson O, Engesæter L B, Kärrholm J, Furnes O, Garellick G, Gjertsen J-E. Posterior approach and uncemented stems increases the risk of reoperation after hemiarthroplasties in elderly hip fracture patients. Acta Orthop 2014; 85(1): 18-25.

Smith T, Pelpola K, Ball M, Ong A, Myint P K. Pre-operative indicators for mortality following hip fracture surgery: a systematic review and metaanalysis. Age Ageing 2014; 43(4): 464-71.

Swedish Hip Arthroplasty Register. Annual Report; 2013.

Yli-Kyyny T, Sund R, Heinänen M, Venesmaa P, Kröger H. Cemented or uncemented hemiarthroplasty for the treatment of femoral neck fractures?

Acta Orthop 2014; 85(1): 49-53.

Acta Orthopaedica 2019; 90 (3): 275–280

275

Total hip arthroplasty, combined with a reinforcement ring and posterior column plating for acetabular fractures in elderly patients: good outcome in 34 patients Tõnis LONT 1, Jyrki NIEMINEN 1, Aleksi REITO 2, Toni-Karri PAKARINEN 1,2, Ilari PAJAMÄKI 2, Antti ESKELINEN 1, and Minna K LAITINEN 3 1 Coxa Hospital for Joint Replacement, Tampere, Finland; 2 Department of Orthopaedics and Traumatology, Unit of Musculoskeletal Surgery, Tampere University Hospital, Tampere, Finland; 3 Department of Orthopaedics and Traumatology, Helsinki University Hospital, Helsinki, Finland Correspondence: tonis.lont@coxa.fi Submitted 2018-09-23. Accepted 2019-02-22.

Background and purpose — Low-energy acetabulum fractures are uncommon, and mostly occur in elderly patients. Determining the optimal operative treatment for such fractures is challenging. Here we investigated whether acutely performed total hip arthroplasty plus posterior column plating (THA) reduced complications and reoperations compared with open reduction and internal fixation (ORIF) in elderly patients with acetabular fractures.

Patients and methods — We retrospectively reviewed the records of 59 patients, > 55 years of age, with complex acetabular fractures, caused by low-energy trauma, treated between January 2008 and September 2017. Of these patients, 34 underwent acute THA, and 25 ORIF alone.

Patient and implant survival were compared between groups using Kaplan–Meier survival analysis and Cox multiple regression. Functional outcomes assessed by Oxford Hip Score (OHS) were compared between the THA patients and those 9 ORIF patients who underwent secondary THA due to posttraumatic hip osteoarthritis (OA) during follow-up.

Results — Overall patient survival was 90% (95% CI 82–98) at 12 months, and 64% (CI 47–81) at 5 years. Of 25 ORIF patients, 9 required secondary THA due to posttraumatic OA. Large fragments on the weight-bearing acetabular dome upon imaging predicted ORIF failure and secondary THA. The acute THA group and secondary THA group had similar 12-month OHS. Interpretation — Acute THA including a reinforcement ring resulted in fewer reoperations than ORIF alone in elderly patients with acetabular fractures. These findings support acute THA as first-line treatment for complex acetabular fractures in elderly patients.

Low-energy acetabulum fractures are rare, and mostly occur in elderly patients with comminuted and complex fracture patterns. They are associated with unfavorable prognostic signs, such as articular impaction and fragmentation, pre-existing osteoarthritis (OA) of the hip, and osteopenia (Anglen et al.

2003, Laflamme et al. 2011). Acetabular fracture treatment depends on the fracture type as well as on patient-related factors, such as comorbidities. Treatment options in elderly patients include a nonoperative approach, open reduction and internal fixation (ORIF), and acute total hip arthroplasty (THA) with or without simultaneous ORIF (Daurka et al. 2014).

Nonoperative treatment is optimal for non-displaced fractures and in patients with severe comorbidities (Guerado et al. 2012). Surgery for low-energy displaced acetabular fractures is challenging. Poor bone quality may impede stable osteosynthesis. Poor prognostic indicators for ORIF include anteromedial dome impaction, poor reduction and fixation of the weight-bearing dome, and associated pelvic fractures (Anglen et al. 2003, Laflamme et al. 2011). Recent reports describe promising results with the combination of THA and ORIF (Boelch et al. 2017, Ortega-Briones et al. 2017, Salama et al. 2017). However, no study has compared the outcomes of these different treatment options; the optimal treatment for low-energy comminuted acetabular fractures in the elderly population remains unclear.

In this retrospective study we examined whether acutely performed THA including posterior column plating would result in fewer complications and reoperations than ORIF alone in elderly patients with comminuted acetabular fractures. In addition we compared functional outcome Oxford Hip Score between patients healed by THA and patients healed with secondary THA after a failed ORIF.

© 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group, on behalf of the Nordic Orthopedic Federation. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits ­unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

DOI 10.1080/17453674.2019.1597325

276

Acta Orthopaedica 2019; 90 (3): 275–280

Fracture patterns according to Letournel Anterior wall Anterior column Anterior column + femoral neck Posterior wall Posterior wall + femoral neck Posterior wall + dislocation Posterior column Transverse Posterior wall + posterior column Posterior wall + posterior column + femoral neck Transverse + posterior wall Transverse + posterior wall + femoral neck T-type T-type + femoral neck Anterior column + hemitransverse Both columns

Acute THA + ORIF ORIF

0

2

4

6

8

10 12 14 16 18 20

Frequency

Figure 1. Fracture patterns according to Letournel classification.

Patients and methods We retrospectively reviewed the records of all patients over 55 years of age who were diagnosed with and treated for a low-energy comminuted acetabular fracture at our hospitals between January 1, 2008, and September 1, 2017. Patients were identified from a prospectively maintained database that identifies and records all patients referred to and managed in the unit. The study population comprised 59 patients, of whom 25 were treated with ORIF alone and 34 with acute THA including posterior column plating. There were no definitive radiographic criteria for the treatment of patients with either ORIF or THA. Patients who underwent ORIF were treated at the Tampere University Hospital, Tampere, Finland. All patients who underwent THA were operated on at the Coxa Hospital for Joint Replacement, Tampere, Finland, which performs all arthroplasties in the same region. Both hospitals are located in the same building and share the same emergency room. Acute trauma patients are discussed and treatment decisions are made in a collective meeting pragmatically by treating physicians.

Fracture patterns according to Letournel classification were determined based on preoperative radiographs (Figure 1). The majority of patients exhibited a complex T-type fracture with central femoral head protrusion (Figures 2). Patient demographics were similar between the treatment groups, except that follow-up was longer in the ORIF group than in the THA group. Based on the Charlson Comorbidity Index (CCI), patients in the THA group tended to have more severe comorbidities than patients in the ORIF group, but this difference was not statistically significant (median CCI of 5 vs. 4; p = 0.1). A causal directed acyclic graph was used to investigate confounding factors (Figure 3, see Supplementary data). Table 1 summarizes the patient characteristics.

ORIF procedures were performed by senior pelvic trauma surgeons. Patients were administered preoperative antibiotic prophylaxis, and placed under general anesthesia. Surgery was initiated with the patient in a supine position. An anterior

A

B Figure 2. A. 75-year-old patient with T-type fracture and central protrusion of the femoral head.

Red line indicates the Gull sign.

B. Computed tomography showing quadrilateral surface comminution and central protrusion of femoral head.

C. The same patient at last follow up. The femoral head had been morsellized and used as filling

C

intrapelvic (AIP) approach was used through a low midline incision. When necessary, a lateral window/first window of an ilioinguinal approach was used to fix the anterior and lateral parts of the pelvis. Thereafter, the patient was repositioned into the lateral decubitus position, and a Kocher–Langenbeck approach was used to access the posterior acetabular fracture components, when necessary.

Acute THA was performed by experienced revision arthroplasty surgeons together with pelvic trauma surgeons. Patients were placed under spinal anesthesia, and a dose of preoperative antibiotic prophylaxis was infused 30 min prior to surgery. A Kocher–Langenbeck approach was used. The posterior column was supported by adding posterior column plating and a GAP II reinforcement ring (Stryker, Mahwah, NJ, USA).

Acta Orthopaedica 2019; 90 (3): 275–280

277

Table 1. Patient demographics. Values are mean (range) unless otherwise specified

Characteristic

Total

Primary operative strategy THA + ORIF ORIF only

p-value

Number 59 34 25 Age 70 (56–92) 71 (56–92) 69 (58–83) Follow-up, years 2.6 (0–9) 1.4 (0–6) 4.2 (0–9) BMI 26 (15–42) 25 (15–42) 26 (20–36) Women, n 17 10 7 Estimated blood loss, L 1.2 (0.4–2.7) 1.1 (0.4–2.7) 1.4 (0.3–3.7) Operating time, min 190 (97–321) 169 (97–310) 218 (120–321) Stem, n Cementless 22 13 9 Cemented 21 13 Acetabular component, n Cemented 3 3 0 Cemented constrained 34 31 3 Cementless 6 0 6 Heart disease, n 14 8 6 Neurologic disease, n 10 7 3 Alcohol abuse, n 6 5 1 Diabetes, n 13 7 6 Trauma, n Falling from the same level 45 25 20 Fall from high 3 2 1 Motor vehicle accident 6 5 1 Biking 5 2 3 Complications, n Secondary OA leading to THA – 9 Dislocation 1 – Periprosthetic fracture 1 – Infection 0 1 CCI, median 5 5 4 THA: total hip arthroplasty; ORIF: open reduction and internal fixation; BMI: body mass index; OA: osteoarthritis; CCI: Charlson Comorbidity Index

Various components were used in both the femur and acetabulum throughout the study period, depending on the implant selected by the hospital (Table 1). In all acute THA cases, morselized autograft bone transplantation from the resected femoral head was performed using an impaction grafting technique (Hosny et al. 2017). In the first acute THA case, anterior column reduction and fixation was performed using an AIP approach. Additional anterior fixation was not applied in any subsequent patients.

Thromboprophylaxis started at 6 hours postoperatively and continued until 4 weeks postoperatively. Mobilization was started on the first postoperative day. ORIF-treated patients were mobilized with partial-weight-bearing walking aids for 6 weeks, after which full weight-bearing was allowed. For patients who underwent acute THA, the goal was partial weight-bearing for 6 weeks, but if patients were unable to follow restrictions, weight-bearing was allowed as tolerated.

During follow-up visits, patients were clinically evaluated and subjected to pelvic radiographs. The OHS was administered at outpatient clinic visits, or via a routine letter request sent to patients with THA at 12 months. All complications encountered during follow-up were recorded.

0.2 0.01 0.2 0.4 0.4 0.01 0.01 0.01

0.6 0.3 0.2 0.5 0.5

0.01

0.1

Statistics Patient and implant survival rates were assessed using the Kaplan–Meier method. Between-group comparisons were performed using the log-rank test.

The Cox regression model was used to identify independent factors affecting patient survival. Continuous variables were reported as mean and 95% confidence interval (CI), and compared between groups by t-test. Differences in proportions were assessed using Fisher’s exact test. Follow-up time was calculated from the date of surgery to the date of the most recent revision, follow-up, or death.

We calculated the CI for relative risks.

All analyses were performed using SPSS Statistics 24.0 (IBM Corp, Armonk, NY, USA), and a p-value of < 0.05 was considered significant.

Ethics, funding, and potential conflicts of interest This retrospective study was approved by the local chair of the audit department.

The study was funded by the Competitive Research Funding of Tampere University Hospital. No competing interests are declared.

Results Overall patient survival after fracture was 90% (CI 82–98) at 12 months, and 58% (CI 38–77) at 5 years. In the ORIF group, 12-month survival was 91% (CI 73–100) and 5-year survival was 86% (CI 63–87). In the acute THA group, the respective survival rates were 89% (CI 77–100) and 30% (CI 2–62) (Figure 3). Univariable analysis revealed that mortality was statistically significantly associated to some extent with low CCI, neurological disease, and alcoholism. In multiple regression analysis, all predictors showed plausible effects on mortality, but the null hypothesis could not be rejected based on the HRs and their corresponding CIs. Of the 25 ORIF patients, 9 developed posttraumatic OA necessitating secondary THA at a median of 12 months (7–37) after fracture surgery. Kaplan–Meier analysis revealed that implant survival was 74% (CI 54–94) at 12 months and 52% (CI 30–75) at 2 years in the ORIF group, and was 100% at 12 months and 91% (CI 74–108) at 2 years in the acute THA group (Figure 4). After adjustment for other variables, ORIF was associated with a hazard ratio (HR) of 12. However, the 95% confidence interval indicated that the data were consistent with a wide range of plausible HRs, from 1.4 to 91.

278

Acta Orthopaedica 2019; 90 (3): 275–280

Cumulative patient survival

Cumulative implant survival

1.0

1.0

0.8

0.8

0.6

0.6

0.4

0.4

0.2

0.2

0.0 0

20

40

60

80

100

120

Between groups, OHD differed by a mean of 2.1 points (CI −2.4 to 6.6).

The OHS exceeded 37 points in 75% of patients. In the acute THA group, 2 patients developed complications: 1 patient had a periprosthetic fracture leading to femoral component revision, and 1 patient had recurrent dislocations that were treated with closed reductions. In the secondary THA group, 2 patients developed deep infections leading to two-stage revision arthroplasty. These groups are compared in Table 2.

0.0 0

20

40

60

80

100

120

Months after index operation

Months after index operation

Figure 3. Postoperative patient survival stratified by surgical group.

Figure 4. Implant survival stratified by surgical group.

Discussion

Osteoporotic fragility fractures, especially in the pelvic area, are increasingly common as the elderly population grows (Kim et al. 2015, Rinne et al. 2017). Fragility fractures are more frequently com ORIF with Mean minuted, and these fractures and low

Acute THA + ORIF secondary THA difference p-value bone quality are associated with poor outcomes with ORIF; thus, the operative Number 34 9 Age 70 (56–87) 65 (58–74) 3 0.1 strategy differs from that used in younger Follow-up, months 15 (1–82) 72 (15–113) 3 0.0 patients with good bone quality (Anglen BMI 23 (15–42) 28 (21–33) 1 0.4 et al. 2003, Ferguson et al. 2010, Salama Women, n 10 3 1 Estimated blood loss, L 1.1 (0.4–2.7) 1.1 (0.700–2.0) 60 0.8 et al. 2017). Recently, the more straightOperating time, min 169 (97–310) 143 (100–269) 26 0.2 forward approach of acute THA combined Oxford Hip Score at 12 months 41 (33–46) 42 (6–48) 2 0.3 with ORIF has gained popularity as an Trauma, n 0.4 Falling from the same level 25 6 alternative to tedious and time-consuming Fall from high 2 – fracture reduction and internal fixation Motor vehicle accident 5 1 with osteosynthesis (Boelch et al. 2017, Biking, n 2 2 Complications, n 0.2 Ortega-Briones et al. 2017, Salama et al.

Infection – 2 2017). In our retrospective study, we comDislocation 1 – pared ORIF and acute THA for the treatPeriprosthetic fracture 1 –

ment of osteoporotic acetabular fractures For abbreviations, see Table 1. in elderly patients. We found that acute THA including posterior column plating with a reinforcement ring resulted in a All ORIF-treated patients showed good reduction quality, low complication rate and carried a low risk for revision. This with a gap of less than 2–3 mm. Patients with unsuccessful appears to be a safe procedure in elderly patients with comORIF tended to show a gull/gull-wing radiological sign and minuted low-energy fractures of the acetabulum. ORIF alone a considerable fragment in the acetabular weight-bearing area carried a relatively high risk of posttraumatic OA necessitatdome, which was observed in 6 of the 9 patients requiring ing a secondary THA. secondary THA. Among the 16 patients successfully treated To our knowledge, this is the first study to directly comwith ORIF alone, 2 also exhibited a fragment in the weight- pare the two most common treatment options for these fracbearing dome. The follow-up duration was shorter in the acute tures. The reasons to choose acute THA over more convenTHA group than in the ORIF group, with a mean difference in tional osteosynthesis may not be obvious. Risk factors for follow-up time of 3 months (CI 2–4) (Table 2). Radiographic osteosynthesis failure are poorly defined in the literature, but follow-up revealed that all fractures healed, and the acetabular generally include marginal impaction, femoral head damage, autologous bone grafts were well incorporated in patients in severe fracture comminution, and the presence of a gull sign the acute THA group (Figure 2C).

(Anglen et al. 2003, Kreder et al. 2006, Gary et al. 2011,

Table 2. Comparison of patients treated with acute THA+ORIF and those treated by ORIF alone followed by secondary THA due to posttraumatic OA. Values are median (range) unless otherwise specified

Acta Orthopaedica 2019; 90 (3): 275–280

Daurka 2014, Li and Tang 2014). In our study, fragment impaction in the acetabular weight-bearing-area dome was a sign of an unfavorable prognosis in ORIF-treated patients, which is in accordance with current literature (Anglen et al.

2003, Laflamme et al. 2011). A large fragment in the anteromedial dome was detected upon imaging in 6 of the 9 patients who were initially treated with ORIF and subsequently required secondary THA. The overall rate of secondary THA in the ORIF group was one-third in our study, which is much higher than the 10% rate reported in the literature. However, prior reports have usually included young patients or all fracture types (Tannast et al. 2012). Among elderly patients, the rate of secondary THA after ORIF widely varies from 19% up to as high as 100% in elderly patients with fractures exhibiting femoral head or acetabular impaction. Elderly patients infrequently have both femoral head and acetabular impaction (Kreder et al. 2006, Archdeacon et al. 2013, Clarke-Jenssen et al. 2017). The high rate of secondary THA in our study was likely due to the comminuted fracture patterns seen in our elderly patients. One major issue with ORIF is the prolonged restriction on weight-bearing. Some elderly patients do not comply with the weight-bearing restriction, resulting in either failures or permanent immobilization (Archdeacon et al. 2013).

Patients had a median age of 70, but there were relatively few complications and none leading to perioperative or immediate postoperative death. The rate of infection was generally low, but was higher in the secondary THA group compared with the acute THA group. The rate of dislocation was also surprisingly low in all patients. Only 1 patient in the acute THA group had a recurrent dislocation, which was not revised because the patient evidently abused alcohol.

Only the first patient in the acute THA group underwent plate fixation of both the anterior and posterior columns. Increasing evidence supports the practice of using only posterior plating (Anglen et al. 2003, Carroll et al. 2010, Herscovici et al. 2010, Salama et al. 2017). After the first patient, the remaining acute THA cases received plate fixation of only the posterior column, resulting in reduced operation times, an easier approach, and simplified postoperative rehabilitation that diminished postoperative complications. There were no revisions or complications due to instability, supporting the hypothesis that posterior column plating was sufficient to stabilize the pelvis in these patients. Our findings and the current literature emphasize the importance of posterior column stabilization and prevention of central migration during treatment using a posterior column plate and an anti-protrusion cage. It remains unclear why the anterior column does not require fixation (Herscovici et al. 2010, Guerado et al. 2012, Rickman et al. 2012, Boelch et al. 2017, Ortega-Briones et al.

2017, Salama et al. 2017).

Our study has several limitations, including those inherent to its retrospective non-randomized design. Additionally, the number of patients is small, limiting the statistical power of

279

our analyses. Mobility and functional scores were not preoperatively assessed. The results may have also been influenced by the short-term follow-up for the acute THA group. However, a properly conducted survivor analysis that accounted for follow-up time revealed a decreased failure rate in the acute THA group.

The majority of injuries in the present study were caused by low-energy falls from the same level. There were also several injuries involving low-speed bicycle and motor vehicle accidents, but these injuries were mostly caused by patient confusion. The population was most similar to the hip fracture population (Panula et al. 2011), but with an improved overall survival of 90% at 12 months, decreasing to 64% at 5 years.

This improved survival might be biased because most fragile patients with acetabular fractures are not considered for operative treatment, and were thus excluded from our study. During the first postoperative year, survival was equal between the 2 treatment groups. Subsequently, the survival rate exhibited a more rapid decrease in the acute THA group compared with the osteosynthesis group, although this difference was not statistically significant. The retrospective design of this study reveals differences in patients between the treatment groups, but also demonstrates the pragmatic decision-making indicated by the CCI. Patients in the acute THA group had more comorbidities, and thus required a long-lasting surgical procedure that avoided the high revision rate in the ORIF group.

The OHS is a validated, reliable, and well-established assessment tool for evaluating the outcome of THA (Dawson et al. 1996). In our study, OHS differed between treatment groups by a mean of 2 points, and did not differ between the acute THA group and the secondary THA group. In a recent study, Hamilton et al. (2018) defined “treatment success” following THA based on an OHS threshold value of 37.5 points, since over 90% of THA patients with an OHS value of over 37.5 points expressed satisfaction with the surgical outcome (Hamilton et al. 2018). In our study, 75% of patients had an OHS exceeding this “treatment success” level, which can be regarded as a good outcome in this older and fragile patient group.

In conclusion, our present results demonstrated that acute THA, performed simultaneously with a reinforcement ring and ORIF, resulted in fewer reoperations, improved implant survival, and yielded a good functional outcome when compared with ORIF alone in elderly patients with complex osteoporotic acetabular fractures. We prefer this acute THA procedure as first-line treatment in this patient population, especially when preoperative radiographs reveal a large dome fragment. However, this surgery is complex and requires a multidisciplinary team with a mixed skill set.

Supplementary data Figure 3 is available as supplementary data in the online version of this article, http://dx.doi.org/10.1080/17453674.2 019.1597325

280

The present study was planned and designed by TL, JN, and MKL. Statistical analyses were performed by AR, TL, and MKL. The manuscript was written by TL and MKL. All authors participated in the data interpretation, and the critical revision of the manuscript.

Acta thanks Olav Røise and Morten Schultz Larsen for help with peer review of this study.

Anglen J O, Burd T A, Hendricks K J, Harrison P. The “Gull Sign”: a harbinger of failure for internal fixation of geriatric acetabular fractures. J Orthop Trauma 2003; 17(9): 625-34.

Archdeacon M T, Kazemi N, Collinge C, Budde B, Schnell S. Treatment of protrusio fractures of the acetabulum in patients 70 years and older. J Orthop Trauma 2013; 27(5): 256-61. doi: 10.1097/BOT.0b013e 318269126f.

Boelch S P, Jordan M C, Meffert R H, Jansen H. Comparison of open reduction and internal fixation and primary total hip replacement for osteoporotic acetabular fractures: a retrospective clinical study. Int Orthop 2017; 41(9): 1831-7. doi: 10.1007/s00264-016-3260-x.

Carroll E A, Huber F G, Goldman A T, Virkus W W, Pagenkopf E, Lorich D G, Helfet D L. Treatment of acetabular fractures in an older population. J Orthop Trauma 2010; 24(10): 637-44. doi: 10.1097/BOT.0b013e3181ceb685.

Clarke-Jenssen J, Roise O, Storeggen S A O, Madsen J E. Long-term survival and risk factors for failure of the native hip joint after operatively treated displaced acetabular fractures. Bone Joint J 2017; 99-b(6): 834-40. doi: 10.1302/0301-620x.99b6.bjj-2016-1013.r1.

Daurka J S, Pastides P S, Lewis A, Rickman M, Bircher M D. Acetabular fractures in patients aged > 55 years: a systematic review of the literature.

Bone Joint J 2014; 96-B(2): 157-63. doi: 10.1302/0301-620X.96B2.32979.

Dawson J, Fitzpatrick R, Carr A, Murray D. Questionnaire on the perceptions of patients about total hip replacement. J Bone Joint Surg Br 1996; 78(2): 185-90.

Ferguson T A, Patel R, Bhandari M, Matta J M. Fractures of the acetabulum in patients aged 60 years and older: an epidemiological and radiological study. J Bone Joint Surg Br 2010; 92(2): 250-7. doi: 10.1302/0301620X.92B2.22488.

Gary J L, Lefaivre K A, Gerold F, Hay M T, Reinert C M, Starr A J. Survivorship of the native hip joint after percutaneous repair of acetabular fractures in the elderly. Injury 2011; 42(10): 1144-51. doi: 10.1016/j. injury.2010.08.035.

Guerado E, Cano J R, Cruz E. Fractures of the acetabulum in elderly patients: an update. Injury 2012; 43(Suppl 2): S33-41. doi: 10.1016/S00201383(13)70177-3.

Acta Orthopaedica 2019; 90 (3): 275–280

Hamilton D F, Loth F L, MacDonald D J, Giesinger K, Patton J T, Simpson A H, Howie C R, Giesinger J M. Treatment success following joint arthroplasty: defining thresholds for the Oxford Hip and Knee Scores. J Arthroplasty 2018; 33(8): 2392-7. doi: 10.1016/j.arth.2018.03.062.

Herscovici D Jr, Lindvall E, Bolhofner B, Scaduto J M. The combined hip procedure: open reduction internal fixation combined with total hip arthroplasty for the management of acetabular fractures in the elderly. J Orthop Trauma 2010; 24(5): 291-6. doi: 10.1097/BOT.0b013e3181b1d22a.

Hosny H A H, El-Bakoury A, Fekry H, Keenan J. Mid-term results of Graft Augmentation Prosthesis II cage and impacted allograft bone in revision hip arthroplasty. J Arthroplasty 2017. doi: 10.1016/j.arth.2017.11.060.

Kim J W, Herbert B, Hao J, Min W, Ziran B H, Mauffrey C. Acetabular fractures in elderly patients: a comparative study of low-energy versus high-energy injuries. Int Orthop 2015; 39(6): 1175-9. doi: 10.1007/s00264-015-2711-0.

Kreder H J, Rozen N, Borkhoff C M, Laflamme Y G, McKee M D, Schemitsch E H, Stephen D J. Determinants of functional outcome after simple and complex acetabular fractures involving the posterior wall. J Bone Joint Surg Br 2006; 88(6): 776-82. doi: 10.1302/0301-620X.88B6.17342.

Laflamme G Y, Hebert-Davies J, Rouleau D, Benoit B, Leduc S. Internal fixation of osteopenic acetabular fractures involving the quadrilateral plate.

Injury 2011; 42(10): 1130-4. doi: 10.1016/j.injury.2010.11.060.

Li Y L, Tang Y Y. Displaced acetabular fractures in the elderly: results after open reduction and internal fixation. Injury 2014; 45(12): 1908-13. doi: 10.1016/j.injury.2014.09.004.

Ortega-Briones A, Smith S, Rickman M. Acetabular fractures in the elderly: midterm outcomes of column stabilisation and primary arthroplasty.

BioMed Res Int 2017; 2017:4651518. doi: 10.1155/2017/4651518.

Panula J, Pihlajamaki H, Mattila VM, Jaatinen P, Vahlberg T, Aarnio P, Kivela S L. Mortality and cause of death in hip fracture patients aged 65 or older: a population-based study. BMC Musculoskeletal Disorders 2011; 12: 105. doi: 10.1186/1471-2474-12-105.

Rickman M, Young J, Bircher M, Pearce R, Hamilton M. The management of complex acetabular fractures in the elderly with fracture fixation and primary total hip replacement. Eur J Trauma Emerg Surg 2012; 38(5): 511-6. doi: 10.1007/s00068-012-0231-9.

Rinne P P, Laitinen M K, Huttunen T, Kannus P, Mattila V M. The incidence and trauma mechanisms of acetabular fractures: a nationwide study in Finland between 1997 and 2014. Injury 2017; 48(10): 2157-61. doi: 10.1016/j. injury.2017.08.003.

Salama W, Mousa S, Khalefa A, Sleem A, Kenawey M, Ravera L, Masse A. Simultaneous open reduction and internal fixation and total hip arthroplasty for the treatment of osteoporotic acetabular fractures. Int Orthop 2017; 41(1): 181-9. doi: 10.1007/s00264-016-3175-6.

Tannast M, Najibi S, Matta J M. Two to twenty-year survivorship of the hip in 810 patients with operatively treated acetabular fractures. J Bone Joint Surg Am 2012; 94(17): 1559-67. doi: 10.2106/jbjs.k.00444.

Acta Orthopaedica 2019; 90 (3): 281–285

281

Complications and readmissions following outpatient total hip and knee arthroplasty: a prospective 2-center study with matched controls Kirill GROMOV 1,3, Christoffer Calov JØRGENSEN 3, Pelle Baggesgaard PETERSEN 2,3, Per KJÆRSGAARDANDERSEN 4,3, Peter REVALD 3, Anders TROELSEN 1, Henrik KEHLET 2,3, and Henrik HUSTED 1,3 1 Department

of Orthopedic Surgery, Copenhagen University Hospital, Hvidovre; 2 Section of Surgical Pathophysiology, Rigshospitalet; 3 Lundbeck Foundation Centre for Fast-track Hip and Knee Arthroplasty; 4 Departement of Orthopedic Surgery, Vejle Hospital, Vejle, Denmark Correspondence: kirgromov@gmail.com Submitted 2018-11-25. Accepted 2019-01-10.

Background and purpose — Outpatient arthroplasty has gained popularity in recent years; however, safety concerns still remain regarding complications and readmissions. In a prospective 2-center study we investigated early readmissions with overnight stay and complications following outpatient total hip (THA) and total knee arthroplasty (TKA) compared with a matched patient cohort with at least 1 postoperative night in hospital.

Patients and methods — All consecutive and unselected patients scheduled for THA or TKA at 2 participating hospitals were screened for potential day of surgery (DOS) discharge. Patients who fulfilled the DOS discharge criteria were discharged home. Patients discharged on DOS were matched on preoperative characteristics using propensity scores to patients operated at the same 2 departments prior to the beginning of this study with at least 1 overnight stay. All readmissions within 90 days were identified.

Results — It was possible to match 116 of 138 outpatients with 339 inpatient controls. Median LOS in the control cohort was 2 days (1–9). 7 (6%) outpatients and 13 (4%) inpatient controls were readmitted within 90 days. Readmissions occurred between postoperative day 2–48 and day 4–58 in the outpatient and control cohorts, respectively. Importantly, we found no readmissions within the first 48 hours and no readmissions were related to the DOS discharge.

Interpretation — Readmission rates in patients discharged on DOS may be similar to matched patients with at least 1 overnight stay. With the selection criteria used, there may be no safety signal associated with same-day discharge.

Outpatient arthroplasty has gained popularity in recent years as the fast-track concept has evolved with optimized logistics and perioperative treatment leading to reduced length of stay in hospital (LOS) worldwide (Kehlet 2013, Berend et al. 2018a).

The popularity of outpatient arthroplasty is further fueled by increased focus on value-based treatments and cost-effectiveness, as outpatient procedures have been shown to have favorable financial benefits compared with in-hospital procedures (Lovald et al. 2014, Husted et al. 2018). Also, studies have shown that outpatient arthroplasty is feasible both for total hip arthroplasty (THA) and total knee arthroplasty (TKA), even in an unselected patient population (Goyal et al. 2016, Gromov et al. 2017). However, same-day discharge is reserved for a few selected patients and less then 1% of hip and knee arthroplasties in the United States are being performed as outpatient procedures (Basques et al. 2017, Courtney et al. 2017). While the majority of the studies have shown outpatient arthroplasty to be safe in a carefully selected patient population (Goyal et al. 2016, Courtney et al. 2017, Nelson et al. 2017), safety concerns still remain regarding complications and readmissions (Lovecchio et al. 2016, Courtney et al. 2018). Therefore, more studies are needed to investigate early complications following outpatient THA and TKA in a well-defined patient population with complete follow-up.

This prospective 2-center study investigated early readmissions requiring at least one night in hospital and types of complications following outpatient THA and TKA compared with a matched patient cohort treated within a standard fast-track setup requiring at least 1 night in hospital.

© 2019 The Author(s). Published by Taylor & Francis on behalf of the Nordic Orthopedic Federation. This is an Open Access article distributed under the terms of the Creative Commons Attribution-Non-Commercial License (https://creativecommons.org/licenses/by/4.0) DOI 10.1080/17453674.2019.1577049

282

Patients and methods All consecutive and unselected patients referred to the 2 participating departments and scheduled for unilateral THA or TKA were screened for participation in the study between December 2015 and September 2017. Outpatient surgery was defined as discharge to own home on day of surgery (DOS). Excluded were patients with ASA score > 2, patients with sleep apnea requiring treatment due to safety concerns if those patients were to be sent home with opioids, and patients operated as #3 in the operating room, as they were unlikely to be back on the ward in time to fulfill the functional discharge criteria on DOS. Finally, an adult had to be present at home for at least 24 hours following discharge in order for the patients to participate in outpatient surgery. Patients who fulfilled the DOS discharge criteria (Table 1, see Supplementary data) were discharged home. Detailed patient eligibility and fulfillment of discharge criteria for a part of this cohort has previously been published (Gromov et al. 2017). The prospective outpatient cohort thus consisted of patients discharged on DOS from the 2 departments during the study period.

The in-patient control cohort consisted of propensity score matched TKA (n = 134) and THA (n = 205 patients operated at the same 2 departments from January 2013 to November 2015 with at least 1 overnight stay—thus prior to introduction of outpatient THA and TKA surgery). All surgeries were performed in a standardized fast-track setup (Husted 2012) by surgeons specialized in THA and TKA surgery. The standard surgical protocol for both THA and TKA included spinal anesthesia, standardized fluid management, use of preoperative intravenous tranexamic acid (TXA), preoperative singleshot high-dose methylprednisolone, and absence of drains.

Mechanical thromboprophylaxis and extended oral thromboprophylaxis were not used. All THAs were performed using a standard posterolateral approach. All TKAs were performed with a standard medial parapatellar approach without the use of tourniquet with application of local infiltration analgesia (LIA). Rivaroxaban was used as oral thromboprophylaxis starting 6 to 8 hours postoperatively and continuing daily until discharge if LOS < 5 days. Patients discharged on DOS received oral thromboprophylaxis for 2 days.

The patients were transferred from the postoperative recovery unit to the ward after a few hours, where immediate mobilization was attempted allowing full weight-bearing. Physiotherapy was started on the day of surgery and continued until discharge. After the patients were back in the ward, the nurse and the physiotherapist screened them for fulfillment of discharge criteria (discharge criteria for DOS discharge consisted of functional criteria that applied to all patients during the entire period as well as additional criteria for DOS discharge (Table 1, See supplementary data). All patients were discharged to their own homes. The treatment of all patients from 2013 and throughout the study was standardized at both

Acta Orthopaedica 2019; 90 (3): 281–285

departments and was the same for all patients—both patients participating in outpatient surgery and those not participating.

Both hospitals are reporting to the Lundbeck Foundation Centre for Fasttrack Hip and Knee Replacement Database (LCDB) (Jørgensen and Kehlet 2013b). All data on patient characteristics and comorbidities were thus prospectively recorded in the LCDB. Using the patients’ unique social security number (Central Office of Civil Registration), we obtained information on postoperative LOS and 90-day readmissions and mortality from the Danish National Patient Registry (Andersen et al. 1999).

As reporting to the Danish National Patient Registry is mandatory for hospitals to receive reimbursement, complete followup is assured. All patients with LOS > 4 days had their medical records examined to determine the reason for prolonged LOS. All unplanned admissions with overnight stay following discharge within 90 days postoperatively were evaluated using discharge or patient records and included as readmissions if related to index surgery. Reasons and timing for readmission were recorded.

Statistics The propensity score (PS) was calculated using a logistic regression model. Covariates entered in the model were: age, sex, BMI, hospital, procedure, living situation, use of walking aids, smoking, alcohol use > 2 units/day (1 unit = 12 g of alcohol), pharmacologically treated cardiovascular disease, pharmacologically treated pulmonary disease (PD), Type II diabetes (DM), use of anticoagulants, pharmacologically treated psychiatric disease, previous venous thromboembolism (VTE), previous stroke, family history with VTE, and hypertension. As no patients in the outpatient cohort had Type 1 diabetes or preoperative anemia, patients with these 2 comorbidities were excluded from the control cohort. A 3:1 greedy nearest neighbor matching algorithm was used, excluding both cases and controls outside the area of common support. In 3 cases (3%), only 1 acceptable match was found and in 3 cases (3%) only 2 acceptable matches were found.

For evaluation of successful matching we calculated the standardized differences.

Comparison of the outpatient and historical cohorts was done using the Mantel–Haenszel common odds ratio for weighted related data with a p-value of < 0.05 considered significant.

Confidence intervals (CI) were all calculated as 95% CI. Ethics, registration, funding, and potential conflict of interest Permission to store and review patient data without prior informed consent was obtained from the Danish Data protection Agency (RH-2017-132) and the Danish National Board of Health (3-3013-56/2/EMJO). LCBD is registered on ClinicalTrials.gov (NCT01515670) as an ongoing study registry on preoperative patient characteristics and postoperative morbidity. This study was supported by a research grant from Zimmer

Acta Orthopaedica 2019; 90 (3): 281–285

Patients scheduled for unilateral THA or TKA screened December 2015 – September 2017 n = 1,065 (THA/TKA = 575/490) Eligibility criteria for outpatient surgery not fulfilled n = 655 (THA/TKA = 355/300) Eligible for outpatient surgery n = 410 (THA/TKA = 220/190) Discharged on day of surgery n = 138 (THA/TKA = 80/58) Excluded from analysis (n = 22): – not in the LCBD database, 13 – missing 1 or more variables, 6 – could not be matched, 3 Included in the matching analysis n = 116 (THA/TKA = 70/46)

Patient inclusions and exclusions

Biomet and the Lundbeck Foundation, which did not participate in the investigation, data analysis or data interpretation.

Authors declare no conflict of interest related to this study.

Results 1,065 patients were screened for eligibility during the study period, 410 patients were eligible for outpatient surgery, and 138 patients were discharged on DOS. 13 patients were not registered in the LCBD, 6 patients were missing 1 or several variables for matching, and 3 patients could not be matched, leaving 116 patients in the outpatient cohort for matching analysis (Figure). The control cohort consisted of 339 patients matched on the available comorbidities. Median LOS in the control cohort was 2 days (1–9). 4 patients in the control cohort had a LOS > 4 days (Table 3, see Supplementary data).

7 (6%, CI 3–12) outpatients and 13 (4%, CI 2–6) inpatient controls were readmitted within 90 days (OR 1.6, CI 0.7–4), respectively (Tables 4 and 5, see Supplementary data). Suspicion of deep venous thrombosis was the most common reason for readmission in both groups: 29% and 38% in the outpatient and control cohort respectively. Readmissions occurred between postoperative day 2 and 48 and day 4 and 58 in the outpatient and control group respectively (Tables 4 and 5, see Supplementary data). Of the 22 patients discharged on DOS who were excluded from analysis, 1 was readmitted due to dislocation of the operated hip on postoperative day 72.

283

Table 2. Preoperative characteristics of the total cohort of outpatients and the matched cohorts of outpatients and non-outpatients.

Frequency (%) unless otherwise specified

Characteristic

PS-matched PS-matched outpatients non-outpatients n = 116 n = 339 STD a

Age, mean (SD) 61 (11) 62 (10.4) BMI, mean (SD) 28 (5) 28 (5) Hospital A 76 (66) 219 (65) B 40 (35) 120 (35) Procedure THA 70 (60) 205 (61) TKA 46 (40) 134 (40) Sex Female 47 (41) 147 (43) Male 69 (60) 192 (57) Social situation Living with others 102 (88) 300 (89) Living alone 14 (12) 39 (12) Use of walking aid Yes 7 (6) 20 (94) No 109 (94) 319 (6) Smoking Yes 23 (20) 78 (23) No 93 (80) 261 (77) Alcohol > 2 units daily Yes 16 (14) 44 (13) No 100 (87) 295 (87) Preoperative anemia Yes 0 0 No 116 (100) 339 (100) Pharmacologically treated psychiatric disorder Yes 8 (7) 22 (94) No 108 (93) 317 (7) Pharmacologically treated pulmonary disease Yes 12 (10) 29 (9) No 104 (90) 310 (91) Pharmacologically treated cardiac disease Yes 8 (7) 22 (7) No 108 (93) 317 (93) Type-2 diabetes Yes 1 (1) 1 (0) No 115 (99) 338 (100) Any anticoagulants Yes 6 (5) 21 (6) No 110 (95) 318 (94) Previous venous thromboembolic event Yes 5 (4) 12 (4) No 111 (96) 327 (97) Previous stroke Yes 9 (8) 30 (9) No 107 (92) 309 (91) Family with venous thromboembolic event Yes 8 (7) 20 (6) No 108 (93) 319 (94) Hypertension Yes 36 (31) 111 (33) No 80 (69) 228 (67)

0.09 0.01 0.2 0.0 0.05 0.02 0.0 0.06 0.02

0.0 0.01 0.05 0.01 0.06 0.03 0.03 0.03 0.03 0.03

a STD:

standardized difference; a STD of > 0.1 was chosen as indicative of imbalance.

PS: propensity score. SD: standard deviation. BMI: body mass index.

THA: total hip arthroplasty TKA: total knee arthroplasty.

Discussion In this prospective 2-center cohort study with a matched control cohort, we found that 6% outpatients and 4% inpatient

controls were readmitted within 90 days following THA and TKA. No patient was readmitted within 48 hours after surgery.

284

Previous studies with complete follow-up in a similar setup have reported 90 days readmission rates of between 9% and 15% (Husted et al. 2010, Jørgensen and Kehlet 2013b). As patients in our study were selected for outpatient surgery and thus healthier compared with the average patient population, lower readmission rates are expected. Readmission rates found in our study are higher compared with some readmission rates previously published following outpatient THA and TKA from the United States, as a recent review by Hoffmann et al. (2018) reported only 2% 90-day readmission rates with similar low readmission rates reported by several other studies (Goyal et al. 2016, Basques et al. 2017, Berend et al. 2018b).

Contrary to this, other studies from the United States have reported similar or even higher readmission rates following outpatient THA and TKA compared with our study: Springer et al. (2017) and Richards et al. (2018) reported 30-day readmission rates of 8%, while Berger et al. (2005) reported a 6% readmission rate within 90 days. The discrepancy in readmission rates following outpatient arthroplasty highlights variation in patient cohorts, definitions of readmission, follow-up method, and definitions of outpatient surgery (Saleh et al.

2019). When comparing outcomes and safety aspects between studies describing outpatient surgery, it is important to be aware of the definition of outpatient surgery itself as some studies use < 24 hour stay as the definition, while others —like this study, define outpatient surgery as DOS discharge (Vehmeijer et al. 2018). It could be argued that we should have used all hospital contacts instead of only readmissions with overnight stay in hospital as a study outcome. However, while this is of economic and logistical interest when evaluating the potential benefits of DOS discharge, in our opinion increases in planned hospital visits or readmissions not requiring overnight stay in hospital are of less relevance when discussing potential safety issues. Our finding that suspicion of deep venous thrombosis, ruled out by ultrasound, was the main reason for readmission is in accordance with previous studies (Husted et al. 2010), and may explain the higher readmission rates in the current study compared with some other studies, as visits to the emergency department with an overnight stay for diagnostics may not always be considered a readmission (Saleh et al. 2019), and may be diagnosed in the primary sector depending on the local logistical setup. Further on, as per department policy, we encourage all patients to contact the department in case of problems, which leads to potential readmissions that also would have been treated in the primary sector in a different setup. Finally, discharge location plays a role when comparing readmission. All patients in our study were discharged to their own home without any additional care, which might explain increased readmission rates compared with studies where patients are discharged to nursing care facilities or similar.

Timing of readmission is of specific interest for outpatient arthroplasty. A main safety consideration following DOS discharge is potential serious complications that are better treated during hospital admission, and potentially fatal if occurring at

Acta Orthopaedica 2019; 90 (3): 281–285

the patient’s own home such as thromboembolic events and pulmonary embolism, myocardial infection, and stroke (Jørgensen and Kehlet 2016, Petersen et al. 2018). This was highlighted by Parvizi et al. (2001), who warned against early discharge from the hospital while pointing out that the majority of serious complications occurred in the early postoperative period within 4 days, while the patients are still in hospital.

This has since been disputed, as studies have shown fast-track THA and TKA with a mean LOS of 1–2 days to be safe and with very low rates of thromboembolic events (Jørgensen and Kehlet 2013b, Petersen et al. 2018). A possible explanation for reduced complication rates in a fast-track setup is early mobilization, which reduces the risk for TE events even with short TE prophylaxis during hospital stay only (Jørgensen et al.

2013a). No complications in our study occurred within the first postoperative 48 hours, and the complications that did occur would most likely have resulted in readmissions even if the patients had a minimum of 1 overnight stay. We thus believe that the risk of serious complications in the early postoperative period is very low for patients with no or few systemic comorbidities that are deemed eligible for outpatient surgery. This is supported by Jørgensen and Kehlet. (2016), who showed that only 11% of all early (< 1 week) TE embolic events occurred in patients without pre-, peri-, or postoperative disposition, meaning that 89% of early TE events occurred in patients who would not be eligible for outpatient surgery.

2 readmissions in the outpatient cohort were due to periprosthetic fracture after a fall. It is possible to speculate that such a complication could potentially be avoided if the patients were kept in hospital longer for better mobilization and achievement of steady gait. However, fractures due to a fall occurred both during admission and after discharge in the control cohort, thus we do not believe that such complications can be prevented by a longer readmission, as they are related to patient characteristics rather that short LOS (Jørgensen and Kehlet 2013a).

2 readmissions in the outpatient cohort in our study were due to urinary retention on PO day 2 and 7. However, both patients had spontaneous urination prior to discharge (as required by discharge criteria); therefore those readmissions would most likely not be avoided even if the patients had a minimum of 1 overnight stay. Consequently, we believe that readmissions in the outpatient cohort were not related to the early DOS discharge.

The low number of patients discharged on DOS is the main limitation of this study, thus making it difficult to draw definite conclusions on safety aspects for outpatient surgery. Our eligibility criteria were very broad, thus eligible patients who were not discharged on the DOS most likely differed from the patients who were discharged on DOS in a variety of parameters. To account for those potential differences we used the propensity score matching to match outpatients to a similar patient group (Table 2) from a retrospective cohort. As mentioned above, serious complications that are better handled

Acta Orthopaedica 2019; 90 (3): 281–285

if occurring in hospital compared with patients’ own homes are the main safety concerns of outpatient surgery. However, such complications (pulmonary embolism, myocardial infection, and stroke) are extremely rare (< 0.1%) even in an unselected THA and TKA population (Jørgensen and Kehlet 2016, Petersen et al. 2018), and presumably even more rare in a healthier selected population of patients deemed eligible for outpatient surgery (Jørgensen and Kehlet 2016). Thus, very large cohorts of patients discharged on DOS are required to make definite statements on safety aspects of outpatient surgery and an RCT on safety aspects of outpatient surgery would not be possible. Therefore we believe that continuous monitoring of complications following outpatient surgery in prospective cohort studies may be the optimal approach.

In summary, we found comparable readmission rates in patients discharged on DOS and matched patients with at least 1 overnight stay within a fast-track setup. Most importantly, we did not find any readmissions within the first 48 hours and found no readmissions to be related to the DOS discharge.

Supplementary data Tables 1, 3, 4, and 5 are available as supplementary data in the online version of this article, http://dx.doi.org/10.1080/ 17453674.2019.1577049

KG, HH, and HK planned the study. KG, PKA, PR, AT, and HH were responsible for the logistical setup and collected the data. CCJ, PBP, and HK analyzed the data. KG wrote the first draft of the paper. All authors revised the paper.

Acta thanks Maziar Mohaddes and Stephan Vehmeijer for help with peer review of this study.

Andersen T F, Madsen M, Jørgensen J, Mellemkjoer L, Olsen J H. The Danish National Hospital Register: a valuable source of data for modern health sciences. Dan Med Bull 1999; 46(3): 263-8.

Basques B A, Tetreault M W, Della Valle C J. Same-day discharge compared with inpatient hospitalization following hip and knee arthroplasty. J Bone Joint Surg 2017; 99(23): 1969-77.

Berend K R, Lombardi A V, Berend M E, Adams J B, Morris M J. The outpatient total hip arthroplasty. Bone Joint J 2018a; 100-B(1 Suppl. A): 31-5.

Berend M E, Lackey W G, Carter J L. Outpatient-focused joint arthroplasty is the future: The Midwest Center for Joint Replacement experience. J Arthroplasty 2018b; 33(6): 1647-8.

Berger R A, Sanders S, Gerlinger T, Della Valle C, Jacobs J J, Rosenberg A G. Outpatient total knee arthroplasty with a minimally invasive technique.

J Arthroplasty 2005; 20(7, Suppl. 3): 33-8.

Courtney P M, Boniello A J, Berger R A. Complications following outpatient total joint arthroplasty: an analysis of a national database. J Arthroplasty 2017; 32(5): 1426-30.

Courtney P M, Froimson M I, Meneghini R M, Lee G-C, Della Valle C J. Can total knee arthroplasty be performed safely as an outpatient in the Medicare population? J Arthroplasty 2018; 33(7): S28-31.

Goyal N, Chen A F, Padgett S E, Tan T L, Kheir M M, Hopper R H, Hamilton W G, Hozack W J. Otto Aufranc Award: A multicenter, randomized study

285

of outpatient versus inpatient total hip arthroplasty. Clin Orthop Relat Res 2016; 475(2): 364-72.

Gromov K, Kjærsgaard-Andersen P, Revald P, Kehlet H, Husted H. Feasibility of outpatient total hip and knee arthroplasty in unselected patients: a prospective 2-center study. Acta Orthop 2017; 88(5): 516-21.

Hoffmann J D, Kusnezov N A, Dunn J C, Zarkadis N J, Goodman G P, Berger R A. The shift to same-day outpatient joint arthroplasty: a systematic review. J Arthroplasty 2018; 33(4): 1265-74.

Husted H. Fast-track hip and knee arthroplasty: clinical and organizational aspects. Acta Orthop 2012; 83(Suppl. 316): 1-39.

Husted H, Otte K S, Kristensen B B, Orsnes T, Kehlet H. Readmissions after fast-track hip and knee arthroplasty. Arch Orthop Trauma Surg 2010; 130(9): 1185-91.

Husted H, Kristensen B B, Andreasen S E, Skovgaard Nielsen C, Troelsen A, Gromov K. Time-driven activity-based cost of outpatient total hip and knee arthroplasty in different set-ups. Acta Orthop 2018; 89(5) 515-21.

Jørgensen C C, Kehlet H. Fall-related admissions after fast-track total hip and knee arthroplasty: cause of concern or consequence of success? Clin Interv Aging 2013a; 8: 1569-77.

Jørgensen C C, Kehlet H, Lundbeck Foundation Centre for Fast-track Hip and Knee Replacement Collaborative Group. Role of patient characteristics for fast-track hip and knee arthroplasty. Br J Anaesth 2013b; 110(6): 972-80.

Jørgensen C C, Kehlet H, Lundbeck Foundation Centre for Fast-track Hip and Knee replacement collaborative group. Early thromboembolic events ≤ 1week after fast-track total hip and knee arthroplasty. Thromb Res 2016; 138: 37-42.

Jørgensen C C, Jacobsen M K, Soeballe K, Hansen T B, Husted H, Kjærsgaard-Andersen P, Hansen L T, Laursen MB, Kehlet H. Thromboprophylaxis only during hospitalisation in fast-track hip and knee arthroplasty, a prospective cohort study. BMJ Open 2013; 3(12): e003965.

Kehlet H. Fast-track hip and knee arthroplasty. Lancet 2013; 381(9878): 1600-2.

Lovald S T, Ong K L, Malkani A L, Lau E C, Schmier J K, Kurtz S M, Manley M T. Complications, mortality, and costs for outpatient and short-stay total knee arthroplasty patients in comparison to standard-stay patients. J Arthroplasty 2014; 29(3): 510-15.

Lovecchio F, Alvi H, Sahota S, Beal M, Manning D. Is outpatient arthroplasty as safe as fast-track inpatient arthroplasty? A propensity score matched analysis. J Arthroplasty 2016; 31(9): 197-201.

Nelson S J, Webb M L, Lukasiewicz A M, Varthi A G, Samuel A M, Grauer J N. Is outpatient total hip arthroplasty safe? J Arthroplasty 2017; 32(5): 1439-42.

Parvizi J, Sullivan T A, Trousdale R T, Lewallen D G. Thirty-day mortality after total knee arthroplasty. J Bone Joint Surg Am 2001; 83-A(8): 115761.

Petersen P B, Kehlet H, Jørgensen C C, Lundbeck Foundation Centre for Fasttrack Hip and Knee Replacement Collaborative Group. Myocardial infarction following fast-track total hip and knee arthroplasty; incidence, time course and risk factors: a prospective cohort study of 24,862 procedures.

Acta Orthop 2018; 89(6): 603-9.

Richards M, Alyousif H, Kim J-K, Poitras S, Penning J, Beaulé P E. An evaluation of the safety and effectiveness of total hip arthroplasty as an outpatient procedure: a matched-cohort analysis. J Arthroplasty. 2018; 33(10): 3206-10.

Saleh A, Faour M, Sultan A A, Brigati D P, Molloy R M, Mont M A. Emergency department visits within thirty days of discharge after primary total hip arthroplasty: a hidden quality measure. J Arthroplasty 2019; 34(1): 20-6.

Springer B D, Odum S M, Vegari D N, Mokris J G, Beaver W B. Impact of inpatient versus outpatient total joint arthroplasty on 30-day hospital readmission rates and unplanned episodes of care. Orthop Clin North Am 2017; 48(1): 15-23.

Vehmeijer S B W, Husted H, Kehlet H. Outpatient total hip and knee arthroplasty. Acta Orthop 2018; 89(2): 141-4.

286

Acta Orthopaedica 2019; 90 (3): 286–291

The development of spasticity with age in 4,162 children with cerebral palsy: a register-based prospective cohort study Olof LINDÉN 1, Gunnar HÄGGLUND 1, Elisabet RODBY-BOUSQUET 1,2, and Philippe WAGNER 2 1 Department of Clinical Sciences, Lund, Orthopedics, Lund University, Sweden; 2 Centre for Clinical Research, Uppsala University, Region Västmanland, Västerås, Sweden Correspondence: olof.linden@med.lu.se Submitted 2018-06-15. Accepted 2019-01-18.

Background and purpose — Spasticity is often regarded as a major cause of functional limitation in children with cerebral palsy (CP). We analyzed the spasticity development with age in the gastrosoleus muscle in children with CP. Children and methods — This is a longitudinal cohort study of 4,162 children (57% boys) with CP born in 1990– 2015, monitored using standardized follow-up examinations in the Swedish surveillance program for CP. The study is based on 57,953 measurements of spasticity of the gastrosoleus muscle assessed using the Ashworth scale (AS) in participants between 0 and 15 years of age. The spasticity was analyzed in relation to age, sex, and Gross Motor Function Classification System (GMFCS) levels using a linear mixed model. Development of spasticity with age was modeled as a linear spline.

Results — The degree of spasticity increased in most children over the first 5 years of life. At 5 years of age, 38% had an AS level of ≥ 2. The spasticity then decreased for 65% of the children during the remaining study period. At 15 years of age only 22% had AS ≥ 2. The level of spasticity and the rate of increase and decrease before and after 5.5 years of age were higher in children at GMFCS IV–V. Abbreviations ARC: Annual rate of change AS: Ashworth scale CI: 95% confidence interval CP: Cerebral palsy CPUP: Swedish surveillance program for cerebral palsy GMFCS: Gross Motor Function Classification System ITB: Intrathecal baclofen pump MAS: Modified Ashworth scale SDR Selective dorsal rhizotomy TAL: Achilles tendon lengthening

Interpretation — The degree of spasticity of the gastrosoleus muscle often decreases after 5 years of age, which is important for long-term treatment planning and should be considered in spasticity management.

Spasticity and muscle weakness are common manifestations of cerebral palsy (CP; Dev Med Child Neurol 2002; 44(9): 63340.). The spastic CP subtypes represent 75–80% of the population with CP (Westbom et al. 2007). Himmelmann et al. (2007) reported spasticity in 69% of children with dyskinetic CP. There are several definitions of spasticity. The most commonly used is probably the definition by Lance et al. (1980): “a velocity-dependent increase in tonic stretch reflexes (muscle tone) with exaggerated tendon jerks, resulting from hyperexcitability of the stretch reflex, as one component of the upper motor neuron syndrome.” Muscle weakness, present in all individuals with CP, has several etiologies: the CP-specific background, through inactivity, and as a consequence of interventions, i.e., muscle or tendon lengthening (Segal et al. 1989).

Spasticity may result in a limited range of active joint motion, with reduced gross and fine motor function and pain.

However, spasticity may sometimes improve function by compensating for muscle weakness (Hicks et al. 2008).

Several studies have demonstrated that in children with CP the spasticity increases during the first years of life (Mutch et al. 1992, Kuban and Leviton 1994, Hägglund and Wagner 2008). A cross-sectional study using data from 547 children in southern Sweden showed that the degree of spasticity in the gastrosoleus muscle reached a peak at 4–5 years of age then decreased each year up to 12 years of age (Hägglund and Wagner 2008).

The Swedish surveillance program for individuals with CP (www.cpup.se) was initiated in 1994 in the south of Sweden.

© 2019 The Author(s). Published by Taylor & Francis on behalf of the Nordic Orthopedic Federation. This is an Open Access article distributed under the terms of the Creative Commons Attribution-Non-Commercial License (https://creativecommons.org/licenses/by/4.0) DOI 10.1080/17453674.2019.1590769

Acta Orthopaedica 2019; 90 (3): 286–291

Table 1. The modified Ashworth scale (MAS) Score

Description of muscle tone

0 Normal tone, no increase in tone 1 Slight increase in tone—a catch and release at the end of the range of motion 1+ Slight increase in tone—catch, followed by minimal resistance in remainder of range 2 More marked increase in tone through most of range 3 Considerable increase in tone, passive movement difficult 4 Affected parts rigid in flexion or extension

Since 2005, CPUP has been a national health care quality register approved by the National Board of Health and Welfare in Sweden. CPUP covers more than 95% of all individuals with CP in Sweden born in or after 2000 (Hägglund et al. 2005, Alriksson-Schmidt et al. 2017). CPUP comprises standardized repeated follow-up examinations, including measurement of spasticity according to the modified Ashworth scale (MAS) (Table 1) (Bohannon and Smith 1987).

The previous study from Hägglund and Wagner (2008) included a cross-sectional analysis on a cohort from southern Sweden. Therefore we now analyzed the development of spasticity with age, sex, and gross motor function in the total population of children with CP in Sweden, including a longitudinal analysis.

Children and methods Children with suspected CP are included in the CPUP program as early as possible, usually before 2 years of age. The diagnosis of CP is confirmed at 4 years of age by a neuropediatrician. Children without CP are then removed from the register. The proportion of children in CPUP without CP is less than 2% (Westbom et al. 2007). The definition of CP by Mutch et al. (1992) was used in the early years and was replaced in 2007 with the definition proposed by Rosenbaum et al. (2007).

The child’s local physiotherapist classifies his/her gross motor function according to the Gross Motor Function Classification System (GMFCS) (Palisano et al. 2008). Children at GMFCS level I are examined once a year up to 6 years of age, then every second year. Those at GMFCS levels II–V are examined twice a year up to 6 years of age, then once a year. The examination includes measurement of muscle tone according to the MAS (Bohannon and Smith 1987). Measurement of tone in the gastrosoleus muscle is standardized and performed supine lying with hips and knees extended, described in a manual and recorded on a standard form (www.cpup.se).

In this study, we used all measurements of spasticity of the gastrosoleus muscle at 0–15 years of age in all children in CPUP born between January 1990 and March 2014 that were reported up to September 2016. For the analysis, MAS levels 1 and 1+ were combined into one category corresponding to

287

the original Ashworth scale (AS) as there is a lack of evidence supporting an ordinal relationship between level 1 and level 1+ (Johnson 2001). Measurements from both legs of each child were included in the study, except for those children (21%) with unilateral spastic CP. For these, AS level was summed over the follow-up period and any side with a non-zero sum was subsequently chosen for inclusion. The analysis was performed using the total cohort including children treated with botulinum toxin injections and oral baclofen. Analyses were performed with and without the children treated with selective dorsal rhizotomy (SDR), intrathecal baclofen pump (ITB), and Achilles tendon (gastrosoleus) lengthening (TAL).

Statistics First, the data were grouped in 1-year intervals and plotted in bar charts according to the last known GMFCS level and sex.

The last GMFCS assignment was used for all previous evaluations. Second, to produce a parsimonious model for the somewhat complicated 3-level hierarchical structure of the study data, measurements from 1 leg in each child were chosen for the analysis. The selection was done randomly in those for whom both legs were included. The selected leg was used for all evaluations. For the data selected, a 2-level mixed effects model (Fitzmaurice et al. 2004) was fitted with levels for leg and measurement. Development of spasticity with age was modelled using a linear spline with a knot at the age of 5.5.

The knot was chosen by means of searching a grid with half year-spacing and subsequent inspection of the likelihood of the resulting model. The knot that yielded the greatest likelihood was chosen for the analysis. The model also included a random intercept (i.e., the AS at the starting point of the follow-up), a random slope for the annual rate of change (ARC) before the age of 5 (ARC < 5.5) and a random slope for the annual rate of change from the age of 5.5 and beyond (ARC ≥ 5.5). The random effects covariance matrix was unstructured, allowing for estimation of all random effect and covariances.

The development of spasticity with age was estimated from the mixed model for each individual child. As a final step, the model was adjusted for differences between sexes and birth years. It also included interaction terms between GMFCS level, ARC < 5.5, and ARC ≥ 5.5 to evaluate the differences in AS development with age for different levels of gross motor function. The differences in ARC between GMFCS levels was tested by fitting 1 model including the GMFCS interaction terms, and 1 without, subsequently testing which model best fit the data, by use of a likelihood ratio test. Furthermore, in order to justify pooling of the data from unilateral cases with those of remaining CP types, an indicator variable was created to identify the unilateral cases. The indicator variable was then included in the final model both as an intercept and as an interaction term together with ARC variables to evaluate possible differences in development between unilateral cases and remaining study participants. As with GMFCS, the difference was evaluated using a likelihood ratio test. A p-value > 0.05

288

Acta Orthopaedica 2019; 90 (3): 286–291

Table 2. Number of children according to sex and GMFCS level

Distribution – Ashworth scale (%) 100

Sex Unclassified I Boys Girls Total

GMFCS level II III IV

0 1 2 3 4

90

V

Total

170 963 365 205 308 370 2,381 157 731 251 130 234 278 1,781 327 1,694 616 335 542 648 4,162

80 70 60 50 40

was interpreted as a lack of evidence of a difference in development between unilateral and remaining cases.

Our choice to treat the outcome scale as continuous is based on the general rule of thumb which states that an ordinal scale may approximate a continuous scale when it has 5 levels or more (Norman 2010). An important issue using this approach with the current data is that the distributional assumptions of the statistical model, which was developed for continuous outcome data, are fulfilled. To evaluate possible deviations from the model assumptions in terms of non-normal residual distribution, residuals on both levels were examined graphically. In order to evaluate the impact of possible deviations on study conclusions all models were compared with and without the use of Huber–White robust standard errors. To further evaluate the impact of the choice of scale on the study conclusions, a sensitivity analysis was performed comparing the results of our original model with those using a mixed effects logistic regression model (Fitzmaurice et al. 2004). Results were additionally compared with and without the use of weights to account for the random selection of one leg in most study participants. All models were fitted using the maximum likelihood approach (Fitzmaurice et al. 2004) in Stata v. 14 (StataCorp, College Station, TX, USA).

Ethics, funding, and potential conflicts of interest The study was approved by the Medical Research Ethics Committee at Lund University (LU-433-99). Verbal consent to use the data for research was provided by all families participating in CPUP. The study was funded by Stiftelsen för bistånd åt rörelsehindrade i Skåne. The authors declare no conflict of interest.

Results The study was based on 57,953 measurements in 4,162 children (2,381 boys, 1,781 girls). The distribution according to sex and the GMFCS is described in Table 2 and the distribution of measurements related to age is described in Table 3 (see Supplementary data). 65 children (1,584 measurements) had been treated with SDR, 67 (1,496 measurements) with ITB, and 430 (8,106 measurements) with TAL. There were 327 children (2,429 measurements) without reported GMFCS level. Most of these children were registered before the introduction of the GMFCS in 1999.

30 20 10 0

1

2

3

4

5

6

7

8

9 10 11 12 13 14 15

Age

Figure 1. Degree of spasticity of the gastrosoleus muscle according to the Ashworth scale relative to age in the total sample of 57,953 measurements in 4,162 children.

MAS 2.0 GMFCS V IV III II I

1.8 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0

0.5

5.5

10

16

Age

Figure 2. Prediction of the development of spasticity with age in relation to GMFCS level, using a mixed-model analysis.

The cross-sectional data showed an increasing proportion of children with AS ≥ 2 up to 5 years of age, followed by a reduced proportion up to 15 years of age (Figure 1). When those children treated with SDR, ITB, or TAL were excluded, the same pattern was seen, but the proportion of children with AS ≥ 2 was reduced from 38% to 35% at 5 years of age and was 22% and 23%, respectively, at 15 years of age. Boys and girls showed similar development.

In a mixed model, excluding adjustment variables for sex and birth year, the intercept, i.e., the muscle tone at the starting point of the model, was higher for children at GMFCS levels III–V (1.36–1.37) than those at levels II (1.08) and I (0.85) (Figure 2). The ARC < 5 was 0.04, (CI 0.03–0.05), which means that the muscle tone increased on average 0.20 AS levels during the period up to 5 years of age. This increase varied from 0.00 in children with GMFCS level III to 0.38 for those at GMFCS V. After 5 years of age the mean ARC ≥ 5

Acta Orthopaedica 2019; 90 (3): 286–291

was –0.04 (CI –0.05 to –0.04), describing an average reduction in muscle tone of 0.4 AS levels up to 15 years of age (see Figure 2). The standard deviation of ARC < 5 was 0.20 and ARC ≥ 5 0.11 in the study population. The typical development of increasing muscle tone up to 5.5 years of age was estimated from the model to be seen in 57% of the population and decreasing muscle tone after 5 years of age in 68%.

The results from the likelihood ratio test showed the variation in ARC with GMFCS level to be significant (p < 0.001).

There was no significant difference in development between unilateral and remaining cases (p = 0.7).

The mean muscle tone was lower in girls and in children born later in the study period (Table 4, see Supplementary data). The differences were small but statistically significant (adjusted for age and GMFCS level).

Discussion Our main finding was that the degree of spasticity, as measured with the Ashworth scale, increased in most children during their first 5 years of life, followed by a decrease during the remaining follow-up period to 15 years of age. The intercept was higher, and the rate of increase and decrease were more pronounced, in children at higher GMFCS levels.

In the statistical analysis, all sensitivity analyses indicated that our conclusions were robust for the choice of analysis method (data not shown). We consequently concluded that our simple approach, without sampling weights or robust standard errors, was appropriate. However, as we chose to model the development with age assuming a constant ARC before and after 5.5 years of age, primarily for reasons of facilitating the statistical analysis, potential variation in the ARC with age may have influenced our results. For example, if the ARC peaked in the first few years of life, children included late in the study period may have been observed to have little or no increase in spasticity with age, making us underestimate the proportion of children with a positive ARC < 5.5. Indeed, the estimated proportion of children increasing before the age of 5.5 was less than those decreasing beyond the age of 5.5, which may appear to be contradictory. However, the question may be clarified by Figure 1, showing the proportion of measurements in children at 5.5 years with any spasticity to be approximately 80%.

It is well known that children with CP often have low or normal muscle tone in the newborn period and then develop increasing spasticity in the first years of life (Mutch et al. 1992, Kuban and Leviton 1994). The decrease in spasticity found in this study after a peak at 5 years of age supports our previous findings from southern Sweden (Hägglund and Wagner 2008).

As a parallel to this development several studies have shown an increase in spasticity over time after stroke or traumatic brain injury in adults (Wissel et al. 2013). Not uncommonly the spasticity progresses rapidly in these situations but then

289

declines during the recovery stages (Naghdi et al. 2010). The initial development of spasticity after a stroke is hypothesized to be a positive development, suggesting that the nervous system is beginning to initiate repair mechanisms to restore muscle tone and movement (Li and Francisco 2015). The exact mechanisms of this process are unclear but are suggested to involve rearrangement and plasticity at both the spinal and cerebral levels. As the exact mechanism behind spasticity in CP are not fully known, it is possible also to assume a similar development of the spasticity in children with CP. A previous study from our group, based on examinations of children in southern Sweden with CP during 1995–2006 (Hägglund and Wagner 2008), showed that 47% had AS ≥ 2 with a peak level at 4 years of age, compared with 38% with a peak level at 5 years of age in the present study. This is explained by the lower levels of spasticity found in the later birth cohorts (Table 4, see Supplementary data). The proportion of children in GMFCS IV and V was also lower in the later birth cohorts. This is consistent with findings from epidemiological studies of the CP panorama in western Sweden, which show a decreased rate of bilateral spastic CP and an increased frequency of unilateral spastic CP in recent years (Himmelmann and Uvebrant 2014). The mixed model we used compensates for the changes in AS and GMFCS distribution in the different birth cohorts.

When children treated with SDR, ITB, or TAL were excluded, the same trend of spasticity development was observed. The children treated with SDR or ITB represent only 3% of the population and the reduced spasticity after these treatments did not explain the overall decline in spasticity over time. Children treated with botulinum toxin A were included in the analyses. Many of these were treated more than 3 months before their examination. It cannot be excluded that some children treated with botulinum toxin have been reported with a lower muscle tone than they would have had without treatment. A study based on all children in CPUP in Sweden showed that about 20% had been treated with botulinum toxin A injected in the gastrosoleus muscle during the study period 2014–2015 (Franzen et al. 2017). The number of children treated with botulinum toxin A in the gastrosoleus muscle according to age at treatment showed the same pattern as the development of spasticity, with a peak at 4–6 years of age. Hence, because most treatments were done during the age period that registered the highest degree of spasticity, compensation for a possible treatment effect would more likely strengthen the pattern of increasing spasticity followed by decreasing spasticity with increasing age.

A spastic muscle will not stretch to the same degree as a muscle with normal tone. Therefore, spasticity may inhibit growth in the length of the muscle, resulting in the development of muscle contracture, with a decreasing range of joint motion (Rang et al. 1986). Both spasticity and contracture of the gastrosoleus muscle may result in toe walking. However, spasticity and contracture may also compensate for weak-

290

ness of the gastrosoleus muscle. The combination of reduced muscle tone and increased body weight with age might result in a change from toe walking to a calcaneal or crouch gait with increased dorsiflexion of the ankle joint during mid-stance. As an effect functional gait can be compromised and a continuous deterioration in walking function in the adult with CP has been reported (Opheim et al. 2009). The pros and cons of spasticity and the need for spasticity-reducing treatment may thus vary with age.

The AS is the most widely used instrument for assessing spasticity. Some studies have shown problems with interrater reliability with the scale (Mutlu et al. 2008). However, a recent systematic review and meta-analysis concluded that inter- and intra-rater agreement was satisfactory (MeseguerHenarejos et al. 2018). The reliability is slightly higher for the original scale, which was used in the present study (Mutlu et al. 2008). Low reliability would, in the absence of systematic errors, only increase the observed variation of the spasticity measurements in the population with increased confidence interval widths.

The validity of the AS has been analyzed in several studies (Scholtes et al. 2006). With AS it could sometimes be difficult to distinguish spasticity from contracture, especially at AS levels 3 and 4 (Alhusaini et al. 2010). Contractures of the gastrosoleus muscle, measured as a reduced range of foot dorsiflexion, usually increase with age (Nordmark et al. 2009).

This increasing contracture could underestimate the decline in spasticity as measured with AS after 5 years of age. Contracture could, however, also to some extent contribute to the increased spasticity registered in children up to 5 years of age.

The normally increasing degree of spasticity during the first years of life in children with CP is well known and has been confirmed in earlier studies (Mutch et al. 1992, Kuban and Leviton 1994, Hägglund and Wagner 2008).

A limitation in this study was that GMFCS levels were not reported for 327 children (8%), most of whom were followed before the publication of the GMFCS. The development of spasticity was not analyzed in relation to CP subtypes. In CPUP, classification of the CP subtype by a neuropediatrician is missing for about 20% of registrants. In a previous study from southern Sweden, where CP subtype was reported in all cases, children with the spastic and dyskinetic subtypes showed similar development of spasticity (Hägglund and Wagner 2008). The cohort of children included in the present study had been followed in CPUP for different lengths of time. This is compensated for in the model where each child’s follow-up time is treated separately.

In summary, spasticity, as measured using the Ashworth scale, increases in most children with CP up to 5 years of age followed by a decrease up to the age of 15 years. This information is important for long-term treatment planning. It also demonstrates the importance of control groups when analyzing spasticity-reducing treatments.

Acta Orthopaedica 2019; 90 (3): 286–291

Supplementary data Tables 3 and 4 are available as supplementary data in the online version of this article, http://dx.doi.org/10.1080/17453674.2 019.1590769 Study design: OL, GH, ERB, PW. Data collection: OL, GH, ERB. Data analysis: GH, PW. Manuscript preparation: OL, GH, ERB, PW. Acta thanks Thomas Dreher and Freeman Miller for help with peer review of this study.

Prevalence and characteristics of children with cerebral palsy in Europe. Dev Med Child Neurol 2002; 44(9): 633-40.

Alhusaini A A, Dean C M, Crosbie J, Shepherd R B, Lewis J. Evaluation of spasticity in children with cerebral palsy using Ashworth and Tardieu scales compared with laboratory measures. J Child Neurol 2010; 25(10): 1242-7.

Alriksson-Schmidt A I, Arner M, Westbom L, Krumlinde-Sundholm L, Nordmark E, Rodby-Bousquet E, Hägglund G. A combined surveillance program and quality register improves management of childhood disability.

Disabil Rehabil 2017; 39(8): 830-6.

Bohannon R W, Smith M B. Interrater reliability of a modified Ashworth scale of muscle spasticity. Phys Ther 1987; 67(2): 206-7.

Fitzmaurice G M, Laird N M, Ware J H. Applied longitudinal analysis. Hoboken, NJ: Wiley; 2004.

Franzen M, Hägglund G, Alriksson-Schmidt A. Treatment with Botulinum toxin A in a total population of children with cerebral palsy: a retrospective cohort registry study. BMC Musculoskelet Disord 2017; 18(1): 520.

Hägglund G, Wagner P. Development of spasticity with age in a total population of children with cerebral palsy. BMC Musculoskelet Disord 2008; 9: 150.

Hägglund G, Andersson S, Duppe H, Lauge-Pedersen H, Nordmark E, Westbom L. Prevention of dislocation of the hip in children with cerebral palsy: the first ten years of a population-based prevention programme. J Bone Joint Surg Br 2005; 87(1): 95-101.

Hicks J L, Schwartz M H, Arnold A S, Delp S L. Crouched postures reduce the capacity of muscles to extend the hip and knee during the single-limb stance phase of gait. J Biomech 2008; 41(5): 960-7.

Himmelmann K, Uvebrant P. The panorama of cerebral palsy in Sweden. XI. Changing patterns in the birth-year period 2003-2006. Acta Paediatr 2014; 103(6): 618-24.

Himmelmann K, Hagberg G, Wiklund L M, Eek M N, Uvebrant P. Dyskinetic cerebral palsy: a population-based study of children born between 1991 and 1998. Dev Med Child Neurol 2007; 49(4): 246-51.

Johnson G R. Measurement of spasticity. In: Johnson G R, Barnes M P, editors. Upper motor neurone syndrome and spasticity. Cambridge: Cambridge University Press; 2001. p. 79–95.

Kuban K C, Leviton A. Cerebral palsy. N Engl J Med 1994; 330(3): 188-95.

Lance J W. Spasticity: disorder of motor control Miami, FL: Symposia Specialists; Chicago: Year Book Medical Publishers; 1980.

Li S, Francisco G E. New insights into the pathophysiology of post-stroke spasticity. Front Hum Neurosci 2015; 9: 192.

Meseguer-Henarejos A B, Sanchez-Meca J, Lopez-Pina J A, Carles-Hernandez R. Inter- and intra-rater reliability of the Modified Ashworth Scale: a systematic review and meta-analysis. Eur J Phys Rehabil Med 2018; 54(4): 576-90.

Mutch L, Alberman E, Hagberg B, Kodama K, Perat M V. Cerebral palsy epidemiology: where are we now and where are we going? Dev Med Child Neurol 1992; 34(6): 547-51.

Acta Orthopaedica 2019; 90 (3): 286–291

Mutlu A, Livanelioglu A, Gunel M K. Reliability of Ashworth and Modified Ashworth scales in children with spastic cerebral palsy. BMC Musculoskelet Disord 2008; 9: 44.

Naghdi S, Ansari N N, Mansouri K, Hasson S. A neurophysiological and clinical study of Brunnstrom recovery stages in the upper limb following stroke. Brain Inj 2010; 24(11): 1372-8.

Nordmark E, Hägglund G, Lauge-Pedersen H, Wagner P, Westbom L. Development of lower limb range of motion from early childhood to adolescence in cerebral palsy: a population-based study. BMC Med 2009; 7: 65.

Norman G. Likert scales, levels of measurement and the “laws” of statistics.

Adv Health Sci Educ Theory Pract 2010; 15(5): 625-32.

Opheim A, Jahnsen R, Olsson E, Stanghelle J K. Walking function, pain, and fatigue in adults with cerebral palsy: a 7-year follow-up study. Dev Med Child Neurol 2009; 51(5): 381-8.

Palisano R J, Rosenbaum P, Bartlett D, Livingston M H. Content validity of the expanded and revised Gross Motor Function Classification System.

Dev Med Child Neurol 2008; 50(10): 744-50.

291

Rang M, Silver R, de la Garza J. Cerebral palsy. In: Lovell W W, Wood R, editors. Pediatric Orthopaedics, 2nd ed. Philadelphia: JB Lippincott; 1986. p. 345–96.

Rosenbaum P, Paneth N, Leviton A, Goldstein M, Bax M, Damiano D, Dan B, Jacobsson B. A report: the definition and classification of cerebral palsy April 2006. Dev Med Child Neurol Suppl 2007; 109: 8-14.

Scholtes V A, Becher J G, Beelen A, Lankhorst G J. Clinical assessment of spasticity in children with cerebral palsy: a critical review of available instruments. Dev Med Child Neurol 2006; 48(1): 64-73.

Segal L S, Thomas S E, Mazur J M, Mauterer M. Calcaneal gait in spastic diplegia after heel cord lengthening: a study with gait analysis. J Pediatr Orthop 1989; 9(6): 697-701.

Westbom L, Hägglund G, Nordmark E. Cerebral palsy in a total population of 4–11-year-olds in southern Sweden: prevalence and distribution according to different CP classification systems. BMC Pediatr 2007; 7: 41.

Wissel J, Manack A, Brainin M. Toward an epidemiology of poststroke spasticity. Neurology 2013; 80(3 Suppl 2): S13-19.

292

Acta Orthopaedica 2019; 90 (3): 292–296

Arthroscopic versus open, medial approach, surgical reduction for developmental dysplasia of the hip in patients under 18 months of age Serda DUMAN 1, Yalkin CAMURCU 2, Hakan SOFU 3, Hanifi UCPUNAR 2, Deniz AKBULUT 4, Timur YILDIRIM 4 1 Diyarbakir

Selahaddin Eyyubi State Hospital, Department of Orthopaedics and Traumatology, Diyarbakir; 2 Erzincan University Faculty of Medicine, Department of Orthopaedics and Traumatology, Erzincan; 3 Medical Park Bahçelievler Hospital, Department of Orthopaedics and Traumatology, Istanbul; 4 Baltalimani Bone and Joint Diseases Education and Research Hospital, Department of Pediatric Orthopaedics, Istanbul Correspondence: serdaduman@yahoo.com Submitted 2018-06-06. Accepted 2019-02-15.

Background and purpose — The value of arthroscopic surgical reduction in developmental hip dysplasia is poorly known. We compared the clinical and radiographic efficacy of arthroscopic and medial open surgical reduction in patients less than 18 months of age with developmental hip dysplasia.

Patients and methods — 54 patients with a mean age of 11 months who were treated by Ludloff’s medial open reduction technique (28 hips, Group L) or arthroscopic surgical reduction technique (26 hips, Group A) were evaluated in this case series. Data on age, sex, preoperative Tönnis grade, operative time, estimated blood loss, residual leg length discrepancy, range of motion (ROM), acetabular index (AI) angle, coverage ratio of the femoral head, continuity of Menard–Shenton line, re-dislocation rate, McKay classification, and Kalamchi–MacEwen avascular necrosis (AVN) classification were collected.

Results — Preoperatively, the mean AI angle was 39° in Group L and 37° in Group A. At the latest follow-up, the mean AI was 26° in both groups. The mean femoral head coverage ratio was 79% in Group L and 80% in Group A. The Menard–Shenton line was intact in all patients. Residual leg length discrepancy or limited ROM was not detected in any patients. 4 patients in Group L and 2 in Group A were diagnosed with type 2 AVN. Interpretation — Arthroscopic surgical reduction in patients aged 6–18 months revealed promising clinical and radiographic outcomes similar to medial open reduction using Ludloff’s technique.

Arthroscopic surgical reduction in developmental dysplasia of the hip (DDH) is an innovative technique (Bulut et al. 2005, Öztürk et al. 2013, Eberhardt et al. 2014). Iliopsoas tenotomy, capsular release, and excision of the ligamentum teres, as well as pulvinar tissue, can easily be performed arthroscopically (Eberhardt et al. 2015), preferably in children aged younger than 18 months. Previous reports in the literature have focused on the detailed surgical technique and outcomes of this approach (Eberhardt et al. 2012, 2015, Öztürk et al. 2013).

We compared the clinical and radiographic efficacy of arthroscopic and open surgical reduction in children aged younger than 18 months with DDH.

Patients and methods This case series evaluated the clinical and radiographic data of patients who underwent surgical treatment for DDH. The inclusion criteria were primary DDH treated surgically using a medial open or an arthroscopic approach in patients aged 6–18 months with a minimum postoperative follow-up of 24 months. Of 68 patients we excluded 14 for different reasons (Figure 1).

Arthrography was performed in all patients under 1 year of age, and we decided to perform surgery when concentric reduction could not be achieved. The accepted criterion for adequate reduction of the hip was a concentrically reduced cartilaginous femoral head with < 4 mm of lateralization (L distance). Lateralization was measured on arthrograms as the distance between the iliac bone and the surface of the cartilaginous femoral head on a line drawn from the inferior end of the ilium to the center of the cartilaginous femoral head, designated as distance L (Figure 2). More than 4 mm of medial dye

© 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group, on behalf of the Nordic Orthopedic Federation. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits ­unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

DOI 10.1080/17453674.2019.1599775

Acta Orthopaedica 2019; 90 (3): 292–296

Patients who underwent surgical treatment for DDH before 18 months of age n = 68 Excluded (n = 14): – pelvic osteotomy during surgical reduction, 2 – past medical history of hip surgery, 1 – teratologic or neuromuscular hip dislocation, 3 – hip dysplasia as part of a syndrome, 1 – lost to follow-up, 4 – incomplete clinical follow-up records, 3 Analyzed n = 54 (medial open reduction, 28 arthroscopic reduction, 26)

Figure 1. Flow chart demonstrating the excluded patients.

pooling, a thickened limbus that in some way interferes with reduction, and constricted capsule indicated lack of concentric reduction (Leveuf 1947). Patients older than 1 year underwent surgery (open or arthroscopically) without closed reduction.

Additionally, we used the clinical criterion of Ramsey’s safe zone, described as the arc of motion through which the hip remains reduced without forced abduction. To determine the safe zone of Ramsey, the hip was adducted to the point of re-dislocation, and that position was noted. Then, the hip was again reduced and extended until it dislocated, and again point of dislocation was noted as well as the required internal rotation to maintain reduction. The minimum range for the safe zone should be measured as 25°. In the case where the stable zone exceeded the safe zone, excessive forced abduction was needed to obtain concentric reduction, or the safe zone was measured as less than 25°, surgery was indicated.

These explicit criteria were consistent between 2 senior pediatric orthopedic surgeons who received training in the same department.

During the study period (2014–2016), 214 patients aged 6–18 months who were diagnosed with DDH underwent treat-

293

ment at our hospital. Among these, 146 (68%) underwent closed reduction and 68 children (32%) underwent surgery.

After exclusions this study included 47 female and 7 male patients. The mean age of the patients was 11 months (6–17).

All patients were diagnosed with unilateral hip joint involvement. For this study the patients were categorized into 2 groups based on the surgical technique used by the 2 authors (1 author routinely performed medial approach open reduction and the other routinely performed arthroscopic reduction). Medial approach open reduction based on the Ludloff technique (Ludloff 1908) was performed in 28 hips (Group L), and arthroscopic reduction was performed in 26 hips (Group A). 14 patients in Group L and 7 patients in Group A received neonatal splinting prior to study enrollment. 2 patients in Group A had failed closed reduction at another center. Age at the time of surgery, sex, the Tönnis grade (Tönnis 1976), and the AI angle measured on anteroposterior pelvic radiographs were recorded in each patient preoperatively.

Surgical technique Medial approach open reduction was performed through a transverse incision measuring 5 cm in length. After dissecting the adductor longus muscle, broad exposure of the surgical field was achieved with dissection of the pectineus muscle.

Tenotomy of the iliopsoas was performed, and the joint capsule was incised. Subsequently, we excised the ligamentum teres, as well as the pulvinar tissue, followed by incision of the transverse acetabular ligament. The femoral head was then reduced into the acetabular cavity. Post-reduction stability was confirmed, and the surgical incision was sutured. A pelvipedal cast was applied in the human position of 100° flexion, < 50° of abduction, and < 10° of internal rotation.

Arthroscopic reduction was performed under general anesthesia with the patient placed in the supine position (Figure 3).

An assistant ensured that the hip was maintained in a position of 90° flexion and 40–60° of abduction, and no traction

Figure 2. A. L distance (red arrow). The distance between the iliac bone and the surface of the cartilaginous femoral head (b) on a line drawn from the inferior end of the ilium (c) to the center of the cartilaginous femoral head (a). B. Arthrogram of a dislocated hip with elongated L distance. C. Arthrogram of a hip with an acceptable reduction.

294

Acta Orthopaedica 2019; 90 (3): 292–296

Figure 3. Arthroscopic technique demonstrating the portals (A), intra-articular pulvinar (B), hypertrophic ligamentum teres (C), intraoperative fluoroscopic image during incision of the transverse acetabular ligament (D), and intraoperative radiographic control for concentric reduction (E).

was needed. A medial sub-adductor portal located 1 cm lateral and 1 cm ventral to the ischial tuberosity in the palpable gap between the adductors and the ischiocrural muscles was used to place the 2.7 mm 70° arthroscope intra-articularly under fluoroscopic control (Eberhardt et al. 2015). Arthroscopic evaluation of the hip joint was performed. An anterolateral portal was placed 2 cm distal to the superior iliac spine and 1 cm lateral from a line drawn through the superior iliac spine and the middle of the patella. Radiofrequency ablation and a shaver were used to remove the ligamentum teres and the pulvinar tissue. Arthroscopic scissors were used to incise the transverse acetabular ligament. After intraoperative examination, capsular release with radiofrequency ablation was performed in 5 patients in whom the safe zone was inadequate.

Iliopsoas tenotomy was not performed owing to the risks reported by a previous study (Eberhardt et al. 2015). Reduction of the hip joint was then performed, and post-reduction stability was confirmed. Portal incisions were sutured, and a pelvipedal cast was applied in the human position of 100° flexion, < 50° of abduction, and < 10° of internal rotation.

Operative time (min) and the estimated blood loss (mL) were recorded in each patient. Postoperative radiography was performed in all patients to confirm concentric reduction of the femoral head into the acetabular cavity. The pelvipedal cast was removed at the end of the 3rd postoperative month, and a Denis Browne orthosis was applied for an additional 8 weeks with the hip maintained in 30–45° of abduction.

All children were followed-up clinically for at least 24 months (24–30) postoperatively, Residual leg length discrepancy, range of motion (ROM) of the operated hip joint, and changes in the AI (between pre- and postoperative examinations) were assessed in all children. The continuity of the

Menard–Shenton line was assessed postoperatively. Coverage ratio of the femoral head inside the acetabulum was also measured (Heyman et al. 1950). The McKay classification (Berkeley et al. 1984) was used to categorize the clinical outcomes in each child. The Kalamchi–MacEwen classification (Kalamichi and MacEwen 1980) was used to evaluate any postoperative avascular necrosis (AVN). All radiographic and clinical measurements or classifications were evaluated and recorded at least twice by 2 authors (YC, DA) who were blinded to the surgical technique.

Statistics Owing to the descriptive nature of the analysis used in this case series, descriptive statistics were presented using medians with ranges (minimum–maximum). Frequencies and percentages were calculated to express categorical variables.

Ethics, funding, and potential conflicts of interest This study was performed following approval from the institutional ethical review board (33216249-604.01.02-E.37840).

Informed consent was obtained from the parents of all children included in this study. No funding was received for this study.

All authors declare that they have no conflicts of interest.

Results (Tables 1 and 2) The mean operative time was similar in the two groups. However, the mean estimated blood loss in Group A was lower than that in Group L. (No vascular or nerve injury occurred in any patient.) The Menard–Shenton line was intact in all patients postoperatively. On the basis of the McKay classification for

Acta Orthopaedica 2019; 90 (3): 292–296

295

Table 1. Demographic data of patients

Demographic data

Table 2. Perioperative data of patients. Values are median (range) unless otherwise specified

Medial open Arthroscopic reduction reduction All patients (n = 28) (n = 26) (n = 54)

Sex (female/male) 25/3 Age (months), median (range) 11 (6–17) Side (right/left) 10/18

22/4 12 (7–17) 10/16

47/7 11 (6–17) 20/34

functional evaluation of hips, the children were categorized as follows: 18 patients as Grade I, 9 as Grade II, and 1 patient as Grade III in Group L and 16 patients as Grade I and 6 as Grade II in Group A. Residual leg length discrepancy or limited ROM of the operated hip joint was not detected in any patient during follow-up.

Type 2 AVN (Kalamchi–MacEwen AVN classification) was diagnosed in 4 patients in Group L and 2 patients in Group A. Re-dislocation occurred in 1 patient in Group 1 at the end of 1 month postoperatively.

Discussion Simple closed reduction is the first-line treatment for DDH in patients aged 6 months to 1 year (Race and Herring 1983, Terjesen and Halvorsen 2007). Open reduction should be considered only if closed reduction cannot be performed.

Medial open surgical reduction is a choice for the management of patients younger than 18 months with DDH. The minimal incision and minimal blood loss are advantages of this approach. Limited exposure of the hip joint is a disadvantage.

Several studies have described successful outcomes with this technique (Morcuende et al. 1997, Okano et al. 2009). However, injury to the medial circumflex femoral artery with consequent avascular necrosis (AVN) of the femoral head is the most important risk associated with the medial approach open reduction (Pospischill et al. 2012). Reportedly, re-dislocation, possible need for secondary pelvic osteotomy (secondary to limited acetabular re-modeling), limitations of ROM, and subsequent lateralization of the femoral head are other possible complications.

Arthroscopic reduction of DDH is not a common treatment because it requires clinical experience in pediatric orthopedics and arthroscopic surgery. Previous reports have described the details of the surgical technique and the outcomes of arthroscopic treatment of DDH (Eberhardt et al. 20015, Kitano et al. 2010, Öztürk et al. 2013). Arthroscopic reduction was introduced by Gross (1977). McCarthy and MacEwen (2007) reported the outcomes of arthroscopic surgery in 3 children with DDH and reported that 1 child required secondary surgery owing to failure of arthroscopic reduction. Bulut et al. (2005) reported favorable clinical and radiographic outcomes of arthroscopically assisted surgical reduction for DDH

Medial open Arthroscopic reduction reduction All patients (n = 28) (n = 26) (n = 54)

Preoperative data Tönnis classification, n Grade I 0 0 Grade II 10 10 Grade III 14 13 Grade IV 4 3 Acetabular index angle (°) 40 (24–49) 38 (28–44) Operative data Operative time (minutes) 34 (30–40) 32 (30–40) Estimated blood loss (mL) 35 (15–55) 9 (5–15) Postoperative data Acetabular index angle (°) 26 (11–39) 27 (19–36) Coverage ratio of femoral head (%) 80 (0–100) 80 (50–100)

0 20 27 7 39 (24–49) 32 (30–40) 18 (5–55) 26 (11–39) 80 (0–100)

performed in 4 patients younger than 18 months in 2005 and Öztürk et al. (2013) reported similar outcomes in 9 patients in 2013. Their technique involved an anterolateral skin incision and iliopsoas tenotomy followed by exposure of the hip joint capsule to directly create anteromedial and anterolateral arthroscopic portals. We performed arthroscopic reduction without dissecting the joint capsule. Additionally, the aforementioned study used a 4.0 mm 30° arthroscope, whereas we preferred the 2.7 mm 70° arthroscope, which offers better intra-articular visualization.

Eberhardt et al. (2012) reported outcomes of arthroscopic hip reduction using sub-adductor and anterolateral portals in 5 infants. Eberhardt et al. (2014) also published another study that evaluated 9 patients (mean age 21 months) in whom surgical intervention included arthroscopic reduction combined with open peri-acetabular osteotomy. They reported promising outcomes with a mean follow-up of 15 months, particularly in patients with type 2 or 3 hips based on the Tönnis classification. Eberhardt et al. (2015) published a detailed analysis of the obstacles preventing arthroscopic reduction along with their previous experience. However, these studies included only 15 children aged under 18 months and also included 15 children aged older than 18 months. Furthermore, pelvic osteotomy was performed concomitant with arthroscopic intervention in a few children in their case series. Therefore, the effects of pelvic osteotomy on the outcomes were excluded.

Iliopsoas tenotomy is routinely performed by many surgeons to reduce the risk of AVN; however, a limited number of studies have reported the effects of iliopsoas tenotomy. Yüksel et al. (2009) showed that only one-third of the tenotomized iliopsoas were reattached and that the remaining underwent atrophy. Eberhardt et al. (2014) also reported successful results in children who did not undergo tenotomy. Notably, in our study, tenotomy was not performed in Group A. This, however, did not affect the results observed in the groups.

296

In light of our results as well as outcomes of previous studies, arthroscopic surgical reduction of DDH seems a promising and effective alternative to medial approach open reduction in pre-walking children. The major advantages of this technique are lesser blood loss and that it does not require wide surgical dissection of the joint capsule.

The major limitation of our study is the retrospective evaluation of prospectively followed patient groups. However, the retrospective study design and comparison of the different techniques performed by 2 different surgeons exclude any patient selection bias. The strength of this study is that it is the first to compare an arthroscopic technique with the commonly used Ludloff technique.

In conclusion, arthroscopic surgical reduction of DDH showed promising clinical and radiographic outcomes that were similar to those observed with a medial approach open reduction in patients aged 6–18 months diagnosed with DDH. SD, undertook study design, manuscript preparation, manuscript writing, and is the corresponding author. YC performed measurements, and analysis of the literature. HS performed statistical analysis and manuscript editing.

HU did data collection. DA performed the measurements. TY, the senior author, did the last revision of the manuscript.

Acta thanks Colin F Moseley and Ralph Sakkers for help with peer review of this study.

Berkeley M E, Dickson J H, Cain T E, Donovan M M. Surgical therapy for congenital dislocation of the hip in patients who are twelve to thirty-six months old. J Bone Joint Surg Am 1984; 66: 412-20.

Bulut O, Oztürk H, Tezeren G, Bulut S. Arthroscopic-assisted surgical treatment for developmental dislocation of the hip. Arthroscopy 2005; 21: 574-9.

Eberhardt O, Fernandez F F, Wirth T. Arthroscopic reduction of the dislocated hip in infants. J Bone Joint Surg Br 2012; 94-B: 842-7.

Eberhardt O, Wirth T, Fernandez F F. Arthroscopic reduction and acetabuloplasty for the treatment of dislocated hips in children of walking age: a preliminary report. Arch Orthop Trauma Surg 2014; 134: 1587-94.

Acta Orthopaedica 2019; 90 (3): 292–296

Eberhardt O, Wirth T, Fernandez F F. Arthroscopic anatomy of the dislocated hip in infants and obstacles preventing reduction. Arthroscopy 2015; 31(6): 1052-9.

Gross R. Arthroscopy in hip disorders in children. Orthop Rev 1977; 6: 43-9.

Heyman C H, Herndorn C H. Legg–Perthes disease: a method for the measurement of the roentgenographic result. J Bone Joint Surg Am 1950; 32 A(4): 767-78.

Kalamchi A, MacEwen G D. Avascular necrosis following treatment of congenital dislocation of the hip. J Bone Joint Surg Am 1980; 62: 876-88.

Kitano T, Imai Y, Morita M, Nakagawa K, Wada M, Sakai T, Eguchi T, Kurada T. New treatment method for developmental dysplasia of the hips after walking age: arthroscopic reduction with limboplasty based on the findings of preoperative imaging. J Orthop Sci 2010; 15: 443-51.

Leveuf J. Primary congenital subluxation of the hip. J Bone Joint Surg Am 1947; 29A: 149-62.

Ludloff K. Zur blutigen Einrenkung der angeborenen Huftluxation. Zeitschr Orthop Chir 1908; 22: 272-6.

McCarthy J J, MacEwen G D. Hip arthroscopy for the treatment of children with hip dysplasia: a preliminary report. Orthopedics 2007; 30: 262-4.

Morcuende J A, Meyer M D, Dolan L A, Weinstein S L. Long-term outcome after open reduction through an anteromedial approach for congenital dislocation of the hip. J Bone Joint Surg Am 1997; 79: 810-17.

Okano K, Yamada K, Takahashi K, Enomoto H, Osaki M, Shindo H, et al.

Long-term outcome of Ludloff’s medial approach for open reduction of developmental dislocation of the hip in relation to the age at operation. Int Orthop 2009; 33: 1391-6. Öztürk H, Oztemur Z, Bulut O, Tezeren G, Bulut S. Arthroscopic-assisted surgical treatment for developmental dislocation of the hip before the age of 18 months. Arch Orthop Trauma Surg 2013; 133: 1289-94.

Pospischill R, Weninger J, Ganger R, Altenhuber J, Grill F. Does open reduction of the developmental dislocated hip increase the risk of osteonecrosis?

Clin Orthop Relat Res 2012; 470: 250-60.

Race C, Herring J A. Congenital dislocation of the hip: an evaluation of closed reduction. J Pediatr Orthop 1983; 3: 166-72.

Terjesen T, Halvorsen V. Long-term results after closed reduction of late detected hip dislocation: 60 patients followed up to skeletal maturity. Acta Orthop 2007; 78: 236-46.

Tönnis D. Normal values of the hip joint for the evaluation of X-rays in children and adults. Clin Orthop Relat Res 1976; (119): 9-47.

Yüksel H Y, Yilmaz S, Aksahin E, et al. The evaluation of hip muscles in patients treated with one-stage combined procedure for unilateral developmental dysplasia of the hip: Part I: MRI evaluation. J Pediatr Orthop 2009; 29: 872-8.

3/19

* Sprowson AP et al. Bone Joint J 2016; 98-B: 1534–1541

www.heraeus-medical.com

Vol. 90, No. 3, 2019 (pp. 191–296)

Bone cement with gentamicin ttamicin i i and clindamycin

ACTA ORTHOPAEDICA

69 %

reduction of deep infections in hip hemiarthroplasty after * fractured neck of femur

Volume 90, Number 3, June 2019

IORT_90_02_COVER - Copy.indd 1

02-05-2019 16:15:56