Acta Orthopaedica, Vol. 91, issue 5, 2020

5/20 ACTA ORTHOPAEDICA

LONGER IMPLANT SURVIVAL. WITH THE RIGHT BONE CEMENT.

23%

lower revision risk* with PALACOS® R+G compared to other bone cements

* Calculated difference of cumulated revision rates in hip arthroplasty at 13 years of implantation

www.heraeus-medical.com

10034

NJR Data Supplier Feedback (summary reports); Cumulative revision rates (2007–2020) status February 2020. Current report accessible at http://herae.Us/njr-data We thank the patients and staff of all the hospitals in England, Wales, Northern Ireland and the Isle of Man who have contributed data to the National Joint Registry. We are grateful to the Healthcare Quality Improvement 3DUWQHUVKLS +4,3 WKH 1-5 6WHHULQJ &RPPLWWHH DQG VWDII DW WKH 1-5 &HQWUH IRU IDFLOLWDWLQJ WKLV ZRUN 7KH YLHZV H[SUHVVHG UHSUHVHQW WKRVH RI +HUDHXV 0HGLFDO *PE+ DQG GR QRW QHFHVVDULO\ UHƃHFW WKRVH RI WKH 1DWLRQDO -RLQW Registry Steering Committee or the Health Quality Improvement Partnership (HQIP) who do not vouch for how the information is presented.

Vol. 91, No. 5, 2020 (pp. 501–619)

The element of success in joint replacement

Volume 91, Number 5, October 2020

2006_19476_AZ_PALACOS_EOS_2.0_Acta_Orthop_Europe_215x280_EN.indd 1 TF-IORT200146.indd 1

14.07.20 13:39

24-10-2020 13:23:59

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. 91, No. 5, 2020

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.

Airfreight and mailing in the USA by agent named WN Shipping USA, 156-15, 146th Avenue, 2nd Floor, Jamaica, NY 11434, USA. Periodicals postage paid at Jamaica NY 11431.

Annual Institutional Subscription, Volume 91, 2020

US Postmaster: Send address changes to Acta Orthopaedica, WN Shipping USA, 156-15, 146th Avenue, 2nd Floor, Jamaica, NY 11434, USA.

$1,291

£798

€1,035

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: Maria Montzka, Portfolio Manager – Medicine P.O. Box 3255, SE-103 65 Stockholm, Sweden, Tel: +46 (0)760 14 24 68. Fax: +46 (0)8 440 80 50.

E-mail: maria.montzka@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.

Subscription records are maintained at Taylor & Francis Group, 4 Park Square, Milton Park, Abingdon, OX14 4RN, United Kingdom.

Copyright © 2020 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. 91, No. 5, October 2020 Guest editorial Hope for better days: 2 studies aiming to answer the remaining questions around dual mobility cups

501

J Verhaar

Perspective Pulmonary function in patients with spinal deformity: have we been ignorant?

503

M L van Hooff, N te Hennepe, and M de Kleuver

506 514

O Wolf, S Mukka, M Notini, M Möller, N P Hailer, and the DUALITY GROUP L W A H van Beers, B C H van der Wal, T G van Loon, D J F Moojen, M F van Wier, A D Klaassen, N W Willigenburg, and R W Poolman

520

L R Hedman and L Felländer-Tsai

523

R Kelc, M Vogrin, and J Kelc

527

P Martinkevich, L L Larsen, T Græsholt-Knudsen, G Hesthaven, M B Hellfritzsch, K K Petersen, B Møller-Madsen, and J D Rölfing T Basso, H Dale, H Langvatn, G Lønne, I Skråmm, M Westberg, T S Wik, and E Witsø M J Temmesfeld, R B Jakobsen, and P Grant

Study protocols, dual mobility cup The DUALITY trial—a register-based, randomized controlled trial to investigate dual mobility cups in hip fracture patients Study protocol: Effectiveness of dual-mobility cups compared with uni-polar cups for preventing dislocation after primary total hip arthroplasty in elderly patients — design of a randomized controlled trial nested in the Dutch Arthroplasty Registry COVID-19 Perspective: Simulation-based skills training in non-performing orthopedic surgeons: skills acquisition, motivation, and flow during the COVID-19 pandemic Perspective: Cognitive training for the prevention of skill decay in temporarily non-performing orthopedic surgeons Physical child abuse demands increased awareness during health and socioeconomic crises like COVID-19: A review and education material Virus transmission during orthopedic surgery on patients with COVID19 – a brief narrative review Does a surgical helmet provide protection against aerosol transmitted disease?

Orthopedic surgery residents’ perception of online education in their programs during the COVID-19 pandemic: should it be maintained after the crisis?

Reflections: The personal and professional impact of COVID-19 on orthopedic surgery trainees: reflections from an incoming intern, current intern, and chief resident The COVID-19 pandemic in Singapore: what does it mean for arthroplasty?

Impact of the COVID-19 pandemic on orthopedic trauma workload in a London level 1 trauma center: the “golden month”: The COVid Emergency Related Trauma and orthopaedics (COVERT) Collaborative Perspective: Challenges and adaptations in training during pandemic COVID-19: observations by an orthopedic resident in Singapore Surgical intervention in patients with proximal femoral fractures confirmed positive for COVID-19—a report of 2 cases Hip and pelvis Implant migration and bone mineral density measured simultaneously by low-dose CT scans: a 2-year study on 17 acetabular revisions with impaction bone grafting Patients with hip resurfacing arthroplasty are not physically more active than those with a stemmed total hip A small number of surgeons outside the control-limit: an observational study based on 9,482 cases and 208 surgeons performing primary total hip arthroplasties in western Sweden The incidence of pelvic fractures and related surgery in the Finnish adult population: a nationwide study of 33,469 patients between 1997 and 2014 Children Development of the annual incidence rate of fracture in children 1980–2018: a population-based study of 32,375 fractures Time trends in pediatric fractures in a Swedish city from 1950 to 2016

534 538 543

F Figueroa, D Figueroa, R Calvo-Mena, F Narvaez, N Medina, and J Prieto

547

D N Bernstein, N Greene, and I O Ibrahim

551

J Decruz, S Prabhakar, B T K Ding, and R Kunnasegaran

556

C Park, K Sugand, D Nathwani, R Bhattacharya, and K M Sarraf

562

W-S Foong, H L T Teo, D H B Wang, and S Y J Loh

567

S K Song, W K Choi, and M R Cho

571

H Stigbrand, K Brown, H Olivecrona, and G Ullmark

576 581

J Jelsma, M G M Schotanus, I T A F Buil, S M J Van Kuijk, I C Heyligers, and B Grimm P Jolbäck, E Nauclér, E Bülow, H Lindahl, and M Mohaddes

587

P P Rinne, M K Laitinen, P Kannus, and V M Mattila

593

A V Larsen, E Mundbjerg, J M Lauritsen, and C Faergemann

598

E Bergman, V Lempesis, J-Å Nilsson, L Jephsson, B E Rosengren, and M K Karlsson

Combined massive allograft and intramedullary vascularized fibula transfer: the Capanna technique for treatment of congenital pseudarthrosis of the tibia A new standard radiographic reference for proximal fibular height in children Correspondence New 3-dimensional implant application as an alternative to allograft in limb salvage surgery: a technical note on 10 cases Information to authors (see http://www.actaorthop.org/)

605

S C M van den Heuvel, H A H Winters, K H Ultee, N Zijlstra-Koenrades, and R J B Sakkers

611

A Frommer, M Niemann, G Gosheger, G Toporowski, A Laufer, M Eveslage, J N Brรถking, R Rรถdl, and B Vogt

617

P Nayak versus J W Park and H G Kang

Acta Orthopaedica 2020; 91 (5): 501–502

501

Guest editorial

Hope for better days: 2 studies aiming to answer the remaining questions around dual mobility cups

Readers of scientific papers are mostly interested in new findings. By scanning the paper’s title, reading the summary and, when attracted by the conclusions, reading the full article, we hope to find new information that will help us to treat our patients better. In this issue of Acta Orthopaedica there are 2 excellent papers, which I would recommend reading, but I need to warn readers beforehand. They do not contain new data yet, but are still worth the time investment required to read them (Van Beers et al. 2020, Wolf et al. 2020).

Both articles concern the value of dual mobility cups (DMC). DMCs came onto the market to reduce the risk of dislocation after total hip arthroplasty. A spherical liner in a DMC encloses the metal femoral head and articulates with a thin metal shell, which is fixed to the acetabular bone. The first DMCs were developed in France, and publications from French colleagues suggested better range of motion and reduced dislocation rates. In many countries the enticing concept of DMC was adopted, especially in patients with high risk of dislocation, and in many patients after revision hip surgery. Based on observational studies, the concept seems to work by reducing dislocation risks. But there are no well-powered randomized trials showing this clearly and there are still safety concerns because of the large polyethylene liner in a metal shell. Wear, aseptic loosening, periprosthetic infection, and intraprosthetic dislocation are all potential complications not yet studied in great detail.

Due to the pioneering work of Scandinavian orthopedics, there are many well-run national implant registers worldwide, but unfortunately they cannot provide all the answers to the above questions. Implant registers report only dislocations that lead to revisions, but without surgery there is no information on the actual dislocation rate. The only solution for the problem is a proper randomized controlled trial (RCT) and both papers give detailed descriptions of such study designs.

Wolf et al. (2020) aim to perform a national (Swedish), multicenter, register-based RCT comparing a DMC with a standard cup in patients > 65 years with a non-pathological, displaced femoral neck fracture. Van Beers et al. (2020) plan to perform a similar national (Dutch) RCT but they compare a DMC with normal cup in all patients ≥ 70 years undergoing elective primary hip arthroplasty. The power calculation of both studies leads to quite a high number of patients for each

study—1,100 in the Dutch study and 1,600 in the Swedish study—and this explains why multiple centers are needed, to include sufficient patients.

The papers illustrate excellently how much energy the preparation of such studies requires from the researchers. However, not only the research plan but also the weakness of the studies are already openly and thoroughly considered in the Discussion section of the respective articles. This shows that even the best research plan is still a compromise and several good studies are necessary to reach the right conclusions.

A real challenge for all studies of this size is the follow-up.

How to keep track of all the patients? For this reason, both studies are nested in the national arthroplasty registries (the Swedish Hip Arthroplasty Register and the Dutch Arthroplasty Register). After final study follow-up, all participants remain traceable in the arthroplasty registers for evaluation of long-term survival and mortality. Both studies will trace complications leading to further surgery. The Swedish study will use the Swedish National Patient Register to detect all dislocations, not only those leading to surgery. The researchers of the Swedish study have a clear advantage over the Dutch study, which intends to detect these dislocations with a questionnaire sent at 3-month, 1-year, and 2-year follow-up. Importantly, both studies also include health-economic evaluation of the use of dual mobility cups, which is relevant for society. Will the increased cost associated with DMC reduce total costs, including those caused by dislocations?

The weakness of many orthopedic interventions is that their scientific foundation is weak and the complications and healthcare economic consequences are insufficiently studied. dual mobility cups are a clear example. They are promising but the proof is doubtful. Based on ambition and using the best scientific tools as well as the excellent options from the national arthroplasty registers, the planned studies from Sweden and the Netherlands give us hope of a more knowledgeable future.

Jan Verhaar Department of Orthopaedics and Sports Medicine Erasmus University Medical Center Rotterdam The Netherlands email: j.verhaar@erasmusmc.nl

© 2020 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 10.1080/17453674.2020.1817663

502

Van Beers L W A H, Van der wal B C H, Van Loon T G, Moojen D J F, Van Wier M F, Klaassen A D, Willigenburg N W, Poolman R W. Study protocol: effectiveness of dual-mobility cups compared with unipolar cups for preventing dislocation after primary total hip arthroplasty in elderly patients: design of a randomized controlled trial nested in the Dutch Arthroplasty Registry. Acta Orthopaedica 2020; 91 (5): 514-9.

Acta Orthopaedica 2020; 91 (5): 501–502

Wolf O, Mukka S, Notini M, Moller M, Hailer N P and Duality Group. Study protocol: The DUALITY trial—a register-based, randomized controlled trial to investigate dual mobility cups in hip fracture patients. Acta Orthopaedica 2020; 91 (5): 506-13.

Acta Orthopaedica 2020; 91 (5): 503–505

503

Perspective

Pulmonary function in patients with spinal deformity: have we been ignorant?

Miranda L VAN HOOFF 1,2, Niek TE HENNEPE 1, and Marinus DE KLEUVER 1 1 Department

of Orthopedics, Radboud University Medical Center, Nijmegen, The Netherlands, 2 Department Research, Sint Maartenskliniek, Nijmegen, The Netherlands Correspondence: Miranda.vanHooff@radboudumc.nl Submitted 2020-04-30. Accepted 2020-06-18.

Spine deformity refers to a broad spectrum of abnormal spinal curvatures, which are prevalent in all ages, and are seen by family physicians, orthopedists, and spine specialists. Pulmonary symptoms such as shortness of breath with exertion and reduced exercise tolerance are commonly experienced in both adolescents and adults with a spinal deformity. As yet, these clinically relevant pulmonary symptoms are not routinely monitored and may have health implications later in the patient’s life as pulmonary function gradually deteriorates with age. This Perspective aims to create awareness among care providers and researchers: attention should be paid to this underexposed domain when consulting with patients who have a spinal deformity, and adequate measurement instruments need to be developed to ultimately enhance the quality of care delivered to these patients.

Adolescent idiopathic scoliosis (AIS) In adolescence, idiopathic scoliosis is the most common type of scoliosis, occurring in about 2–3% of adolescents aged 16 years or younger (Weinstein et al. 2008). Various symptoms, such as back pain, reduced self-image, physical disability, and cardiopulmonary compromise, are well reported in AIS (Koumbourlis 2006, Weinstein et al. 2008). The effect of AIS on pulmonary function has been recognized and could lead, when untreated, to disability secondary to pulmonary symptoms, such as shortness of breath during daily functioning or exercise intolerance (Weinstein et al. 1981, Tsiligiannis and Grivas 2012, Weinstein 2019). Furthermore, it seems an important contributing factor to avoidance of activities and exercise training in patients with AIS (Lenke et al. 2002). Surprisingly, however, it is not routinely quantified/measured, and in all scientific publications this domain is rarely reported and extremely underexposed. Have we been ignorant?

In research, attempts have been made to objectively quantify pulmonary function in patients with AIS using clinical pulmo-

nary function tests (PFTs). Even though a decrease in values of total lung capacity may be seen, dissociation exists between the measured pulmonary deficits and symptoms experienced by the patients (Tsiligiannis and Grivas 2012). Although PFTs are valuable to investigate and monitor patients with suspected or known respiratory pathology and to evaluate patients prior to major surgery (Ranu et al. 2011), conflicting evidence exists regarding their clinical value for both clinicians and patients in routine care for AIS patients. As such, obtaining routine PFTs for long-term patient surveillance and/or quantifying treatment effects is not standard practice as they lack clinical relevance, and are time consuming and costly to obtain.

Adult spinal deformity (ASD) Symptomatic ASD refers to various degenerative, progressive conditions and affects the thoracic or thoracolumbar spine throughout the aging process (Diebo et al. 2019). In younger adults the most common spinal deformity is persistent idiopathic scoliosis, whereas in middle-aged and older adults de novo degenerative lumbar scoliosis or adult degenerative scoliosis are more common (Silva and Lenke 2010, Diebo et al. 2019). Given its prevalence, with rapid increases expected over the coming decades, the disorder is of growing interest in health care. Global disparities in both assessment and treatment of ASD exist, resulting in high costs for society (Diebo et al. 2019).

Despite the (limited) knowledge regarding pulmonary function in adolescents, even less evidence is available regarding the effects of ASD on pulmonary function and the impact of (surgical) interventions for ASD on pulmonary function (Lehmann et al. 2015). A natural decline in pulmonary function is seen with aging but seems more pronounced in patients with untreated spinal deformity (Weinstein et al. 1981). Surgery in ASD patients has been reported to result in a significant deterioration in (clinical) PFTs two years following surgical

© 2020 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 10.1080/17453674.2020.1786267

504

correction (Lehmann et al. 2015). However, here too it is not routinely quantified/measured, and in scientific publications this domain is rarely reported and extremely underexposed.

Pulmonary function: clinicians’ and patients’ perspective Clinicians’ perspective Recently, for both AIS (De Kleuver et al. 2017) and ASD (Faraj et al. 2019) a standard outcome set was developed. The relevance of routinely assessing pulmonary function in both AIS and ASD was recognized by a worldwide group of expert clinicians, but such a patient-relevant measure is not currently available.

Patients’ perspective To obtain a first impression of pulmonary problems as experienced by patients with AIS and ASD, an anonymous exploratory survey was performed during an information day for adolescents and adults with scoliosis in the Netherlands. Questions were related to pulmonary symptoms (including description in own words); limitations in daily functioning due to pulmonary symptoms; worsening of symptoms with increased fatigue; differences during the course of the day (answer options yes/no). Patient characteristics included age, concomitant respiratory disease, and previous surgical treatment for AIS or ASD. When previous spine surgery had been performed it was asked whether symptoms were different after surgery (relevant difference in terms of improvement and worsening, or no difference).

After brief instruction, in total 58 patients completed the survey, aged 43 years (SD 12; categorized 10–17 years [n = 9]; 18–24 years [n = 10]; 25–40 years [n = 6], and ≥ 40 years [n = 33]). 3 patients reported having a respiratory disease (COPD [n = 1]; asthma [n = 2]). Among the 58 patients, 26 experienced pulmonary symptoms (age < 40 years [8/25]; ≥ 40 years [15/33]). Most patients described their symptoms as “breathlessness” (10/26) or “fatigue”/“fatigue due to limited endurance” (5/26). Daily functioning of 19/26 patients was limited due to the pulmonary problems and 18/26 patients reported worsening of symptoms with increased fatigue. Fourteen patients underwent scoliosis surgery and four of them experienced a relevant difference before and after the surgery. Although this survey undoubtedly has selection bias, these preliminary results indicate that pulmonary symptoms are prevalent and were commonly experienced in about half of both adolescent and adult patients.

Continuous outcome monitoring: pulmonary function In the era of value-based healthcare, to monitor the quality of the full cycle of care for patients with spinal disorders, routine outcome monitoring through outcome registries is valuable (Van Hooff et al. 2015). These outcome registries are based on a standard set of patient-relevant outcomes, i.e. patientreported and clinician-based outcomes that matter to patients (Van Hooff et al. 2015). Based on the above, an outcome measure that includes patients’ experience regarding pulmonary function is clearly needed.

Acta Orthopaedica 2020; 91 (5): 503–505

Relevance of patient-reported outcome measure (PROMs): the way forward?

As yet, the theoretical construct of pulmonary function in both AIS and ASD, in terms of for example shortness of breath, and/or reduced exercise tolerance and/or respiratory fatigue, is not clearly understood. As such, no adequate methods are available that take patients’ perspective into account to quantify this in routine clinical practice. Clinical PFTs lack clinical relevance, as they do not cover the patients’ perspective, are time consuming, and are expensive to obtain in routine clinical daily practice. An adequate PROM might be a good alternative to assess pulmonary function in patients with spinal deformity.

However, when following guidelines for the development of PROMs it takes several years to develop an adequate PROM in different languages and in terms of validity, reliability, and responsiveness (measurement properties). A general PROM development process, in which both clinicians and patients are involved, consists of several iterative steps that require mixed qualitative and quantitative (longitudinal) study designs (De Vet et al. 2014, Mokkink et al. 2019): definition of the theoretical construct to be measured, item generation, generating and selecting items, development of scales and scoring methods, initial pilot testing (feasibility and usability), and clinical field testing to evaluate its validity, reliability, and responsiveness (measurement properties). Our patient survey has demonstrated that we may indeed have been ignorant of an important aspect of adolescent and adult spinal deformity patients’ lives, and that we need to explore this further. Meanwhile, whilst work is performed to develop an adequate PROM, we recommend that care providers who see adolescents and adults with a spinal deformity should be aware of, pay attention to, and address pulmonary symptoms experienced, such as shortness of breath or reduced exercise tolerance.

De Kleuver M, Faraj S S A, Holewijn R M, Germscheid N M, Adobor R D, Andersen M, Tropp H, Dahl B, Keskinen H, Olai A, Polly D W, Van Hooff M L, Haanstra T M. Defining a core outcome set for adolescent and young adult patients with a spinal deformity. Acta Orthop 2017; 88(6): 612-18.

De Vet H C W, Terwee C B, Mokkink L Bv, Knol D L. Measurement in medicine. 3rd ed. Cambridge: Cambridge University Press; 2014.

Diebo B G, Shah N V, Boachie-Adjei O, Zhu F, Rothenfluh D A, Paulino C B, Schwab F J, Lafage V. Adult spinal deformity. Lancet 2019; 394(10193): 160-72. doi: 10.1016/S0140-6736(19)31125-0. Epub 2019 Jul 11.

Faraj S S A, Van Hooff M L, Haanstra T M, Wright A, Plly D, Glassman S D, De Kleuver M. A153: Core set of outcome measures for Adult Spinal Deformity surgery: A global consensus-based approach. Global Spine J 2019; 9(2S): 83S. doi: 1177/2192568219839730 Koumbourlis A C. Scoliosis and the respiratory system. Paediatr Respir Rev 2006; 7: 152-60.

Lehman R A, Kang D G, Lenke L G, Stallbaumer J J, Sides B A. Pulmonary function following adult spinal deformity surgery: minimum two-year follow-up. Bone Joint Surg Am 2015; 97: 32-9.

Lenke L G, White D K, Kemp J S, Bridwell K H, Blanke K M, Engsberg J R. Evaluation of ventilatory efficiency during exercise in patients with idiopathic scoliosis undergoing spinal fusion. Spine 2002; 27: 2041-5.

Acta Orthopaedica 2020; 91 (5): 503â&#x20AC;&#x201C;505

Mokkink L B, Prinsen C A C, Patrick D L, Alonso J, Bouter L M, De Vet H C W, Terwee C B. COSMIN study design checklist for patient-reported outcome measurement instruments. Version July 2019. Available from: https://www.cosmin.nl/wp-content/uploads/COSMIN-study-designingchecklist_final.pdf (accessed May 31, 2020).

Ranu H, Wilde M, Madden B. Pulmonary function tests. Ulster Med J 2011; 80(2): 84-90.

Silva F E, Lenke L G. Adult degenerative scoliosis: evaluation and management. Neurosurg Focus 2010; 28: E1 . doi: 10.3171/2010.1.FOCUS09271 Tsiligiannis T, Grivas T. Pulmonary function in children with idiopathic scoliosis. Scoliosis 2012; 7: 7.

505

Van Hooff M L, Jacobs W C, Willems P C, Wouters M W, De Kleuver M, Peul, W C, Ostelo R W, Fritzell P. Evidence and practice in spine registries.

Acta Orthop 2015; 86 (5): 534-44.

Weinstein S L, Zavala D C, Ponseti I V. Idiopathic scoliosis: long term follow up and prognosis in untreated patients. J Bone Joint Surg Am 1981; 63: 702-12.

Weinstein S L, Dolan L A, Cheng J C, Danielsson A, Morcuende J A. Adolescent idiopathic scoliosis. Lancet 2008; 371(9623): 1527-37. doi: 10.1016/ S0140-6736(08)60658-3.

Weinstein S L. The natural history of adolescent idiopathic scoliosis. J Pediatr Orthop 2019; 39(6, Suppl. 1): S44-S46. doi: 10.1097/BPO. 0000000000001350.

506

Acta Orthopaedica 2020; 91 (5): 506–513

Study protocol: The DUALITY trial—a register-based, randomized controlled trial to investigate dual mobility cups in hip fracture patients Olof WOLF 1, Sebastian MUKKA 2, Maja NOTINI 1, Michael MÖLLER 3, Nils P HAILER 1, and the DUALITY GROUP a 1 Department of Surgical Sciences, Orthopaedics, Uppsala University; 2 Department of Surgical and Perioperative Science (Orthopaedics), Umeå University; 3 Institute of Clinical Science, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden Correspondence: olof.wolf@surgsci.uu.se Submitted 2020-05-08. Accepted 2020-05-18. a Collaborators: Maria MANNBERG, Department of Surgical Sciences, Orthopaedics, Uppsala University; Olof SKÖLDENBERG, Karolinska Institutet, Department of Clinical sciences at Danderyd Hospital, Stockholm; Kamal KADUM, Maziar MOHADDES, and Ola ROLFSON, Institute of Clinical Sciences, Sahlgrenska Academy, University of Gothenburg, Gothenburg; Nicole JESSEN, Department of Orthopaedics, Sunderby Hospital, Luleå; Daniel STAM, Department of Orthopaedics, Hallands Hospital, Halmstad; Mohammadamin AGHAEE, Department of Orthopaedics, Lidköping Hospital, Lidköping; Jörg SCHILCHER, Department of Orthopaedics and Department of Biomedical and Clinical Sciences, Linköping University, Linköping; Elin NEMLANDER, Department of Orthopaedics, Ljungby Hospital, Ljungby, Sweden.

Background and purpose — Physically and mentally fit patients with a displaced femoral neck fracture (FNF) are mostly treated with total hip arthroplasty (THA). Dislocation is a severe and frequent complication in this group, and dual mobility cups (DMC) were developed to reduce the risk of dislocation after THA. The DUALITY trial investigates whether the use of DMC in FNF patients treated with a THA reduces the risk of dislocation.

Patients and methods — The trial is a national, multicenter, register-based, randomized controlled trial (rRCT).

Patients ≥ 65 years with a non-pathological, displaced FNF (Type Garden 3–4/AO 31-B2 or B3) who are suitable for a THA according to local guidelines are assessed for eligibility using the web-based registration platform of the Swedish Fracture Register (SFR). 1,600 patients will be randomized 1:1 to either insertion of a DMC (intervention group) or a standard cup (control group). The study is pragmatic in that the choice of implant brands, surgical approach, and periand postoperative protocols follow the local routines of each participating unit. All outcome variables will be retrieved after linkage of the study cohort obtained from the SFR with the Swedish Hip Arthroplasty Register and the National Patient Register.

Outcomes — The primary outcome is the occurrence of any dislocation of the index joint treated with closed or open reduction within 1 year after surgery, expressed as a relative risk when comparing groups, and a risk reduction of at least 45% is considered clinically relevant. Secondary outcomes are the relative risk of any reoperation of the index joint, periprosthetic joint infection, and mortality within 90 days and 1 year. Patient-reported outcomes and health economics are evaluated.

Start of trial and estimated duration — The DUALITY trial started recruiting patients in January 2020 and will continue for approximately 5 years.

Trial registration — The trial is registered at clinicaltrials.gov (NCT03909815; December 12, 2019).

Most patients with a displaced femoral neck fracture (FNF) are treated with an arthroplasty, and those who are independently mobile, have few comorbidities, and are cognitively intact commonly receive a total hip arthroplasty (THA) rather than a hemiarthroplasty (Bhandari and Swiontkowski 2017).

However, FNF patients treated with a THA often suffer from dislocation, which results in long-lasting impairment of quality of life (Enocson et al. 2009a). The incidence of dislocations after THA performed due to FNF is up to 13% (Jobory 2020), thus being much higher than after THA performed due to osteoarthritis (Johansson et al. 2000, Phillips et al. 2003, Meek et al. 2006, Skoldenberg et al. 2010). Most dislocations occur during the first postoperative year (Phillips et al. 2003, Meek et al. 2006, Hailer et al. 2012), and small femoral head sizes, the posterior surgical approach, comorbidity burden, and male sex are all associated with an increased risk of dislocation (Jolles et al. 2002, Phillips et al. 2003, Meek et al. 2006, Conroy et al. 2008, Enocson et al. 2009b, Kim et al. 2009, Hailer et al. 2012, Ko and Hozack 2016).

Dual mobility cups (DMC) were introduced in order to improve joint stability. In the DMC, a spherical polyethylene liner encloses the metal femoral prosthesis head of standard diameter (mostly 22 or 28 mm), and this liner is mobile within an external metal shell in order to increase range of motion

© 2020 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 10.1080/17453674.2020.1780059

Acta Orthopaedica 2020; 91 (5): 506–513

and jumping distance (Caton and Ferreira 2017, Cuthbert et al. 2019). According to a systematic review of observational studies on primary and revision THA performed on a multitude of indications the use of DMC is associated with comparatively low dislocation rates (Darrith et al. 2018). A prospective study on FNF patients operated on with DMC reports a relatively low dislocation rate of 1.4% within 9 months (Adam et al. 2012), but without a comparison group. A largescale observational study on 9,040 FNF patients treated with THA estimates a considerably reduced risk of revision due to dislocation in patients operated on with DMC compared with those receiving standard cups (Jobory et al. 2019), but the low rate of revisions due to dislocation of 0.8% in that study is contradicted by a considerably higher dislocation rate of 4.7% in a smaller observational study (Tabori-Jensen et al. 2019).

It must be noted that arthroplasty register-based studies only report revisions due to dislocation, but not the incidence of dislocations per se, such that the true dislocation rate is consistently underestimated in arthroplasty register studies (Devane et al. 2012).

Further to doubts related to the efficacy of the DMC concept there are concerns regarding its safety. The concept of a large polyethylene liner ensheathed between two metal surfaces might increase polyethylene wear (Tabori-Jensen et al.

2018) and the subsequent risk of aseptic loosening (Caton et al. 2014), and an increased risk of periprosthetic joint infections (PJI) is also reported after the use of DMC (Kreipke et al. 2019), although this notion has been contested (Jobory et al. 2019). A design-specific complication of DMC is intraprosthetic dislocation (Philippot et al. 2013, Darrith et al.

2018). Thus, although observational evidence indicates that DMC confers a reduced risk of dislocation, both the efficacy and safety of DMC in patients with FNF are uncertain, and no high-level evidence study has yet been conducted (Griffin et al. 2016).

The primary aim of this trial is to investigate whether the use of DMC reduces the risk of dislocation after THA surgery performed due to FNF when compared with standard cups.

As secondary endpoints we shall analyze whether the risk of the adverse events any reoperation, periprosthetic joint infection, and mortality is increased after use of the DMC, patientreported outcomes measured are compared, and the question as to whether use of the costlier DMC is cost-efficient will be addressed.

Patients and methods Study design The DUALITY trial is a multicenter, register-nested, randomized controlled trial (rRCT; James et al. 2015). Patients with a displaced FNF who are eligible for a THA according to local guidelines are randomized 1:1 to intervention (DMC) or control treatment (standard cup).

507

Table 1. Screening questions within the SFR platform This patient is eligible for inclusion in the Duality trial for randomization to receive a standard cup or a dual mobility cup for a Garden 3–4 fracture. Answer the following questions for screening.

• Is the patient already treated for the fracture?

• Is the patient suitable for a total hip arthroplasty?

• Can both treatments (standard and dual mobility cup) be performed for this patient?

• Has the patient given informed consent?

Study subjects and eligibility criteria The SFR study platform detects eligible patients based on age (≥ 65 years) and type of fracture (type 3 or 4 according to Garden [Kazley et al. 2018], AO types 31-B2 or B3) during registration of the injured patient and automatically alerts the admitting physician of the possibility to screen the patient for eligibility, a method of register-based screening and inclusion that is also used for the first orthopedic rRCT in Sweden, the HipSTHer trial (Wolf et al. 2020). Inclusion criteria are eligibility for a THA according to local guidelines and routines, availability of both treatment options, and signed, informed patient consent (Table 1). Unavailability of both treatment options can be due to implants being out of stock or the lack of the individual surgeon’s competence to use either implant type. Exclusion criteria are cognitive impairment, previous inclusion of a contralateral THA in the ongoing trial, delayed fracture surgery (date of injury more than seven days prior to date of screening), pathological or stress fracture of the femoral neck, and fracture adjacent to a previous ipsilateral hip implant, such as a previously inserted screw or plate.

Randomization and blinding Randomization is also performed by use of the study platform incorporated in the SFR. Subjects will be randomized to receive either a DMC (intervention group) or a standard cup (control group), using an allocation sequence hidden from all involved healthcare providers and provided by a trial-independent statistician. There will be no patient or physician blinding.

Surgical intervention The trial design is pragmatic, which implies that the choice of implant brands, fixation methods, surgical approach, pre-, peri-, and postoperative routines are based on the participating hospitals’ preferences. Nonetheless, the study protocol requires all participating units to maintain their chosen regime across both intervention and control groups, thus ensuring that only the type of intervention varies per unit.

In Sweden, two-thirds of FNF patients who receive a THA are operated on via a direct lateral approach (Swedish Hip Arthroplasty Register 2018) according to Hardinge (1982) or Gammer (1985), and the remaining third via a posterior approach. Surgical approach can vary by surgeon, but indi-

508

vidual surgeons must maintain the same approach for both study groups. If the posterior approach is used, the posterior capsule and short external rotators should be repaired, but if individual surgeons choose to abstain from this recommendation, they are free to do so, provided they maintain this regime across treatment groups.

Implants DMC The 3 cup brands Avantage (Zimmer Biomet, Warsaw, IN, USA), Polar (Smith & Nephew, London, UK), and Ades (Zimmer Biomet) account for 97% of the DMC used in Swedish FNF patients, and none other than these are currently used at the participating units (Swedish Hip Arthroplasty Register 2018). Should novel DMCs be introduced during the trial period they may be used in study participants, provided that an adequate introduction to the specifics of each implant has been given by the manufacturer. For smaller cup sizes, below 50 mm for the most common DMC, only liners with an inner diameter of 22 mm are available, necessitating the use of 22 mm femoral heads in patients operated on with small cup diameters, whereas 28 mm heads are used in combination with all medium- to large-sized cups.

Standard cups The variation in the use of standard cups in Swedish FNF patients is slightly larger, with the Lubinus (Waldemar Link, Hamburg, Germany), Marathon (DePuy Synthes, Warsaw, IN, USA), Exeter RimFit (Stryker, Kalamazoom MI, USA), and Lubinus IP (Waldemar Link) cups being the most common.

As for DMC, the smallest cup sizes require the use of femoral head sizes of 22 mm.

Stem components The Lubinus SP2 (Waldemar Link), Exeter (Stryker), and MS-30 (Zimmer) stems are used in more than 90% of Swedish FNF patients, and the stem type that represents the local standard for FNF patients who receive a THA will be used at each participating unit.

Postoperative treatment Weight bearing will be allowed according to local routines at participating units in both study groups. Postoperative mobilization will start day 0 or 1, which today is the standard of care.

Hip precautions can vary between units but must be consistent across groups, ensured by instructions in the study protocol stating that the same educational material, oral information, and rehabilitation regime are presented to all study participants within a given unit, regardless of whether DMC or standard cups were inserted.

Withdrawal of patients from the trial Participants are free to withdraw from the trial at any time without any adverse consequences to further treatment.

Acta Orthopaedica 2020; 91 (5): 506–513

Table 2. ICD-10 and NOMESCO codes defining primary and secondary endpoints Endpoint Codes ICD Dislocation Periprosthetic

joint infection

NOMESCO Dislocation Periprosthetic

joint infection Any reoperation

M24.3, M24.4, M24.4F, S73.0, T93.3 M00.0, M00.0F, M00.1, M00.2, M00.2F, M00.8, M00.8F, M00.9, M00.9F, M86.0F, M86.1F, M86.6, M86.6F, T81.4, T84.5, T84.5F, T84.5X, T84.7, T84.7F NFH00, NFH02, NFH20, NFH21, NFH22 NFSx, NFA12, TNF05, TNF10 Any of the codes above, and: NFA00-22, NFA31-32, NFCx, NFF01–12, NFL09–19, NFL39–49, NFL69–99, NFM09–29, NFM49, NFM79–99, NFTx, NFWx

Already collected data on patients who choose to withdraw their consent to participate in the trial will be retained in the study database, but no additional data including data derived from cross-matching of the SFR database with other registries will be added. Patients who withdraw from the study will not be replaced.

Endpoints Primary endpoint The primary endpoint is the occurrence of any dislocation treated with closed or open reduction of the index joint within one year. Dislocation is treated as a binary categoric variable that is registered together with an underlying time-to-event variable. The occurrence of dislocations is determined by linking the study cohort derived from the SFR with the Swedish Hip Arthroplasty Register (SHAR) and the Swedish National Patient Register (NPR). In the SHAR, reoperations and revisions of the index joint are registered, including all open reductions, but excluding closed reductions that are not reported to this register. In the NPR, both closed and open reductions including laterality are registered using International Classification of Diseases (ICD)-10 and NOMESCO codes (Table 2), thus indicating a diagnosis of dislocation and/or its treatment.

The presence of a contralateral THA is expected in about 20% of the study participants (Swedish Hip Arthroplasty Register 2018), and to avoid false-positive events due to errors in laterality coding medical charts of all study participants who have been identified as having experienced dislocations will be assessed, and it will thus be ascertained on which joint the reduction or revision procedure was performed.

Secondary endpoints Secondary endpoints are the relative risk of any reoperation of the index joint, PJI of the index joint, and mortality within 90 days and 1 year in the intervention compared with the control group. Any reoperation of the index THA

Acta Orthopaedica 2020; 91 (5): 506â&#x20AC;&#x201C;513

is defined as the occurrence of any surgical procedure performed on the previously treated hip within 1 year after surgery. Reoperations are registered in the SHAR, but the occurrence of reoperations will be additionally verified by cross-matching study participants with the NPR and searching for ICD-10- and NOMESCO-codes (Table 2) indicative of reoperations. A PJI is defined based on the registration of ICD-10 and NOMESCO codes obtained from the NPR (Table 2). Deaths and dates of death are registered in the NPR, allowing for the calculation of 90-day and 1-year mortality. Patient-reported outcomes will be assessed by use of EQ-5D domain score (5 levels) and by the EQ5D-visual analogue scale (VAS) on a 0â&#x20AC;&#x201C;100 numeric scale.

Both parameters are routinely collected in the SHAR and will be assessed 1 year after index surgery. Procedural costs for both intervention and control treatment will be recorded at representative sites, as will procedural costs for closed reductions and reoperations.

Data collection Baseline data on age, sex, injury mechanism, fracture classification, time of diagnosis obtained by radiography, and time and type of surgical treatment are transcribed from the SFR to the study database. Answers given by the admitting physician in response to the screening questions will be saved to the study database in order to enable an analysis of reasons underlying the failure to include eligible patients in the trial.

Postoperatively, procedural details on the type of surgical approach, type of cup and stem fixation, cement brands, cup and stem brands, cup and femoral head diameter, femoral neck length, and stem size are registered in the SHAR according to national routines. In addition to these procedural details body mass index (BMI) and American Society of Anesthesiologists (ASA) class are recorded in the SHAR. After trial completion, the study cohort obtained from the SFR will be linked to information on the study participants registered in the SHAR and the NPR, and these data will be entered into a common research database.

Data quality assurance A study monitor will have regular contacts with all participating units in order to (1) verify the presence of informed consent forms signed by participating subjects, (2) confirm that the team at each participating unit adheres to the study protocol, (3) specifically verify that inclusion and exclusion criteria are consistent, and (4) assist locally responsible investigators regarding technical issues with the study platform. A locally responsible study coordinator ensures that all personnel involved in the treatment of trial subjects at each participating unit are adequately informed and trained regarding protocol requirements, and that the standardization of surgery and postoperative treatment across treatment groups is adhered to.

The steering committee will have no access to outcomes until the database is locked.

509

Estimated sample size and power Scenario 1 For our power calculation, we assume that the 1-year incidence of dislocation after insertion of a standard THA after FNF is 7%, thus slightly lower than the 8% dislocation rate described in Swedish FNF patients treated with a THA (Jobory 2020). For the intervention group operated on with a DMC we assume a relative risk of 0.5, giving an incidence of dislocation of 3.6%. This risk reduction is based on the relative risk of dislocations after the use of a DMC estimated in previous observational studies, ranging from 0.3 to 0.5 (Hailer et al. 2012, Tarasevicius et al. 2013, Bensen et al. 2014, Jobory et al. 2019).

Scenario 2 A recent study from Denmark that investigates the use of DMC in FNF patients reports a dislocation rate of 4.7% after a mean follow up of 5.4 years (Tabori-Jensen et al. 2019). To account for this alternative, more pessimistic scenario, we calculate power based on the assumption of a 1-year dislocation rate of 8% in the control group and 4.5% in the intervention group, giving a relative risk of 0.55.

Sample size was determined by simulations under a simplified assumption of a constant risk during the 1-year follow-up, with 25% of the control-arm patients having an event risk of 6.4% (no risk factors), 50% of patients having a 7.4% event risk (one risk factor), and 25% of patients having a 8.5% event risk (two risk factors), corresponding to sex and surgical approach as independent risk factors associated with an increased risk of dislocation (Hailer et al. 2012). Random censoring due to death was assumed to occur exponentially at 10%/year. This assumption is based on a Swedish study on hip fracture patients treated with a THA (Hailer et al. 2016) and is also in line with mortality data in patients treated with THA due to FNF that is reported by the SHAR (Swedish Hip Arthroplasty Register 2018).

With a sample size of n = 1,600 patients, the trial has 88% power to detect a reduction in 1-year dislocation rates from 7% to 3.6%, equaling a hazard ratio of 0.5 (scenario 1), and 83% power to detect a reduction from 8% to 4.5%, equaling a hazard ratio of 0.55 (scenario 2).

Statistics Analysis will be performed using the intention-to-treat principle including all randomized patients according to randomized treatment. The primary outcome is the adjusted risk of dislocation treated by open or closed reduction within 1 year.

The cumulative unadjusted incidence of dislocations will be estimated using the Kaplanâ&#x20AC;&#x201C;Meier method per randomized treatment group. The relative hazard of dislocation in the intervention compared with the control group will be estimated by Cox regression models adjusted for sex and surgical approach and will be presented as a hazard ratio with 95% profile like-

510

lihood confidence interval and a two-sided likelihood-ratio p-value. With the registry-nested follow-up, we assume that follow-up will be complete, but in the rare case that a patient has incomplete follow-up he or she will be considered censored at last known follow-up. Death before dislocation will be handled as censoring at day of death.

The secondary endpoints any reoperation, PJI, and mortality will be analyzed and described in the same way as the primary endpoint. Supplementary sensitivity analyses will be performed for all event endpoints. These analyses will primarily use logistic regression with the same covariates as the primary analysis, and as a supplement risk differences with Wald confidence intervals will be computed. To investigate sensitivity to baseline covariates, unadjusted Cox regression models will be fitted. Sensitivity analyses to investigate the impact of censoring by death, in addition to analyzing death as an outcome, will include analyses of the composite of dislocation and death performed similarly to the primary endpoint analysis. Estimation of the risk of dislocation after 1 year will be investigated in an additional sensitivity analysis including patients with follow-up exceeding 1 year. Randomized and actual treatments will be described in a CONSORT diagram, and additional per-protocol analyses will be undertaken as sensitivity analyses. The threshold of statistical significance will be set at a two-sided p-value of 0.05. Secondary endpoints will be presented without formal multiplicity adjustment.

EQ-5D domain scores (5 levels) at 1 year after index surgery will be summarized using descriptive frequency tables by randomized treatment. They will be analyzed by using proportional odds logistic regression adjusted for the baseline domain score as a categorical variable and presented as the common odds ratio for all cut-points. For the primary presentation and analysis, missing domain scores due to death will be considered a separate category. For the adjusted analysis, missing baseline scores will be imputed using multiple imputation. Sensitivity analyses using observed cases only will also be provided.

EQ-5D VAS score at 1 year after index surgery will be presented using tables of medians and quartiles as well as empirical cumulative distribution plots of VAS score and linear change in VAS from baseline. The VAS score will be analyzed using proportional odds logistic regression adjusted baseline score as a numerical variable modelled as a restricted cubic spline. Missing baseline scores will be imputed using multiple imputation. Outcome scores that are missing due to death will primarily be imputed as 0, with no imputation of other missing scores.

For all event outcome variables, pre-defined subgroup/ interaction analyses to assess the homogeneity of the treatment contrast will be performed for sex, age, ASA class, and BMI, and for the procedural characteristics femoral neck length, cup diameter, femoral head diameter, type of cup, type of stem, type of cement, and surgical approach. For categori-

Acta Orthopaedica 2020; 91 (5): 506–513

cal subgroup indicators, events will be described in each subgroup as for the entire population, and the treatment contrast in each subgroup will be estimated using a Cox proportional hazard model with treatment, subgroup, indicator, and interaction, and presented with nominal 95% confidence intervals for each subgroup and the interaction p-value. For age, sex, and BMI, the interaction model will use restricted cubic spline modelling, and present the results as a curve of treatment contrast by covariate with 95% pointwise confidence bands and the interaction p-value. Treatment comparison is not relevant for subgroups that are specific to a single treatment arm. For such subgroups descriptive statistics including Kaplan–Meier plots will be presented for each subgroup.

For health economic studies, Markov modelling based on the assumption of defined health states will be performed, and the primary outcome will be cost per quality-adjusted life year. Deterministic and probabilistic sensitivity analyses of the main model hypothesis and variables will be performed in addition to the main analyses.

Ethics, registration, data sharing plan, funding, potential conflicts of interests, and dissemination The study is performed in accordance with the published study protocol, with the latest version of the Declaration of Helsinki, and applicable regulatory requirements. The study was approved by the Swedish Ethical Review Authority (Approval No: 2019-01137). Patients will be required to give written informed consent to participate.

The trial is registered at clinicaltrials.gov (NCT03909815; December 12, 2019).

Datasets derived from the current study that are needed to replicate main findings will be made available by the principal investigators upon reasonable request.

The trial is supported by a grant from the Swedish Research Council (VR 2019-00436). The funding body has no authority over study design, data collection management, interpretation of data, analysis, or writing of manuscripts. The formal sponsor following the definition of clinicaltrials.gov is Uppsala University, Sweden. Open access funding is provided by Uppsala University.

NPH reports both institutional support and lecturer’s fees from Waldemar Link GmbH and Zimmer Biomet, 2 manufacturers of DMCs used in this study. OW reports lecturer’s fees from Waldemar Link GmbH, Smith & Nephew and DePuy Synthes. MM reports lecturer’s fees from DePuy Synthes.

None of the other authors declare any conflict of interest.

The results from the study the will be distributed through presentions and publication in a scientific peer-review medical journal.

Study start and duration The first patient was recruited on January 9, 2020. We expect to recruit patients for 5 years.

Acta Orthopaedica 2020; 91 (5): 506–513

Discussion The purpose of the DUALITY trial is to provide evidence to support or refute the use of DMC in patients with a displaced FNF treated with THA. The potentially reduced risk of dislocation after the use of DMC in FNF patients is described in several observational studies (Tarasevicius et al. 2013, Bensen et al. 2014, Tabori-Jensen et al. 2019), but no highlevel evidence study has yet been conducted (Griffin et al.

2016), and there may be an increased risk of other adverse events such as loosening or PJI (Kreipke et al. 2019). The number of elderly patients with displaced FNF treated with THA is increasing in most developed countries, but prior to the broad introduction of the costlier DMC concept its safety and efficacy, including cost-effectiveness from a health-economic perspective, must be evaluated (Horriat and Haddad 2018, Bernstein et al. 2019).

Strengths and limitations The obvious weakness of previous observational studies is the presence of residual confounding, and confounding by indication may be introduced by the fact that DMC may have been preferentially used in patients who were at higher risk of dislocation. Other limitations to previous studies include small sample sizes or the lack of comparison groups (Adam et al. 2012, Tabori-Jensen et al. 2019). Thus, by conducting a large-scale RCT, we investigate a sufficiently powered sample of patients randomized to intervention or control treatment, thereby reducing problems related to residual confounding or insufficient sample size. Additionally, the lack of external validity inherent in classical RCT designs may be improved by the pragmatic study design of our trial: broad inclusion criteria, few exclusion criteria, freedom to choose locally established implant brands, surgical approach, and postoperative restrictions contribute to the generalizability of our future findings.

The SFR, supplying the platform used for screening, inclusion, and randomization of our study cohort, is a populationbased register of all fractures in adults and long-bone fractures in children, regardless of treatment (Wennergren et al. 2015).

Linkage of the study cohort with the SHAR and the NPR is performed to gain access to additional baseline data, procedural details, and primary and most secondary outcomes. The SHAR has completeness of 96–98% and 100% coverage (Swedish Hip Arthroplasty Register 2018), completeness for the NPR is above 99%, and its positive predictive value is 85–95% (Ludvigsson et al. 2011). Thus, we believe our data sources to be valid and reliable. Nonetheless, optimal control over primary outcome necessitates individual medical chart assessment of all study participants who are registered with a dislocation in order to ascertain correct laterality and diagnosis.

There are numerous potential limitations: 1. The assumptions underlying the sample size calculation are key to every RCT, and we have attempted at calculat-

511

ing 2 realistic scenarios. Importantly, the dislocation rate of 8% after conventional THA in patients with FNF is based on a recent Swedish study (Jobory 2020). The risk reduction of 0.5 that we assume is associated with the use of DMC cups in the main scenario 1 is based on several observational studies, and this number is at the upper end of the range of reported risk reductions, thus pessimistic. The alternative scenario 2 is based on a recent Danish study (Tabori-Jensen et al. 2019) reporting a higher dislocation rate, which may be explained by the following factors: (a) The longer follow-up period in that study may lead to a higher incidence of dislocations when compared with the 1-year incidence of dislocations that our main scenario is based upon. (b) All patients in the Danish study were operated on via a posterior approach that is known to be associated with an increased risk of dislocation (Hailer et al. 2012), whereas two-thirds of Swedish patients receiving a THA due to a femoral neck fracture are operated on via direct lateral approaches (Swedish Hip Arthroplasty Register 2018). (c) More than half of the Danish cohort were treated with cementless implants, a choice that is also associated with an increased risk of dislocation (Chammout et al. 2017). Nonetheless, our sample size lends above 80% power to detect a relevant difference between the intervention and control group, even in this alternative scenario.

2. We believe that we can include a sample of 1,600 patients within reasonable time, mainly based on the participation of at least 14 orthopedic units that together performed about 2,200 THA procedures on patients with FNF during the period 2016–2018, and more units are expected to be enrolled in the near future. However, several factors can delay inclusion: (a) There is a national recommendation to treat patients with FNF within 24 hours of admission to hospital, thus the time window for screening, inclusion, and randomization is limited and may be too short. (2) Surgical expertise to perform either intervention or control treatment is required but not always available, resulting in failure to include patients, or in cross-over if patients allocated to one treatment receive the other. (3) Cognitive impairment is present in a large proportion of FNF patients. At some participating units such patients receive THA, but these will not be included because the ethical approval was restricted to cognitively intact patients. (4) Last but not least, acute and unforeseen events with a large impact on the available resources in healthcare systems, such as the COVID-19 pandemic, can jeopardize any prospective study, and the effects of the current situation on the inclusion of patients in our trial are already dramatic.

3. Confounding factors such as sex, age, ASA class, BMI, femoral neck length, cup diameter, femoral head diameter, type of cup, type of stem, type of cement, and surgical approach are not stratified for in our trial, and, in a worstcase scenario, these confounders may be unevenly distrib-

512

uted across trial arms. We attempt to address this potential issue by adjusting for the main effect mediators sex and surgical approach, and will undertake subgroup and interaction analyses for all variables mentioned.

In summary, the proposed RCT with its register-nested, pragmatic design will, it is hoped, provide high-level evidence on the topic of DMC in patients with displaced FNF.

NPH, OW, MM, and SM designed the trial and shared in reviewing the manuscript. Ethical applications were handled by NPH. MN drafted the manuscript. All authors have given their final approval of the version to be published and agree to be accountable for all aspects of the work.

The authors thank Krister Arlinger for his valuable contributions to the study protocol, and Krister Arlinger and Fredrik Heidgert for the technical solutions behind the study platform in the SFR. They gratefully acknowledge the financial support of the Swedish Research Council and logistic support from Uppsala Clinical Research Center, especially statistician Ollie Östlund, who performed power calculations and drafted the statistical analysis plan. They thank study monitor Monica Sjöholm for invaluable help with all practical issues of this trial. The support of Gothia Forum, the Center of Registers at the Western Healthcare Region, and the authors’ respective institutions and departments is gratefully acknowledged.

Adam P, Philippe R, Ehlinger M, Roche O, Bonnomet F, Mole D, Fessy M H, French Society of Orthopaedic S Traumatology. Dual mobility cups hip arthroplasty as a treatment for displaced fracture of the femoral neck in the elderly: a prospective, systematic, multicenter study with specific focus on postoperative dislocation. Orthop Traumatol Surg Res 2012; 98(3): 296-300.

Bensen A S, Jakobsen T, Krarup N. Dual mobility cup reduces dislocation and re-operation when used to treat displaced femoral neck fractures. Int Orthop 2014; 38(6): 1241-5.

Bernstein J, Weintraub S, Morris T, Ahn J. Randomized controlled trials for geriatric hip fracture are rare and underpowered: a systematic review and a call for greater collaboration. J Bone Joint Surg Am 2019; 101(24): e132.

Bhandari M, Swiontkowski M. Management of acute hip fracture. N Engl J Med 2017; 377(21): 2053-62.

Caton J H, Ferreira A. Dual-mobility cup: a new French revolution. Int Orthop 2017; 41(3): 433-7.

Caton J H, Prudhon J L, Ferreira A, Aslanian T, Verdier R. A comparative and retrospective study of three hundred and twenty primary Charnley type hip replacements with a minimum follow up of ten years to assess whether a dual mobility cup has a decreased dislocation risk. Int Orthop 2014; 38(6): 1125-9.

Chammout G, Muren O, Laurencikas E, Boden H, Kelly-Pettersson P, Sjoo H, Stark A, Skoldenberg O. More complications with uncemented than cemented femoral stems in total hip replacement for displaced femoral neck fractures in the elderly. Acta Orthop 2017; 88(2): 145-51.

Conroy J L, Whitehouse S L, Graves S E, Pratt N L, Ryan P, Crawford R W. Risk factors for revision for early dislocation in total hip arthroplasty. J Arthroplasty 2008; 23(6): 867-72.

Cuthbert R, Wong J, Mitchell P, Kumar Jaiswal P. Dual mobility in primary total hip arthroplasty: current concepts. EFORT Open Rev 2019; 4(11): 640-6.

Darrith B, Courtney P M, Della Valle C J. Outcomes of dual mobility components in total hip arthroplasty: a systematic review of the literature. Bone Joint J 2018; 100-B(1): 11-9.

Devane P A, Wraighte P J, Ong D C, Horne J G. Do joint registries report true rates of hip dislocation? Clin Orthop Relat Res 2012; 470(11): 3003-6.

Enocson A, Pettersson H, Ponzer S, Tornkvist H, Dalen N, Tidermark J. Quality of life after dislocation of hip arthroplasty: a prospective cohort study on 319 patients with femoral neck fractures with a one-year follow-up. Qual Life Res 2009a; 18(9): 1177-84.

Acta Orthopaedica 2020; 91 (5): 506–513

Enocson A, Hedbeck C J, Tidermark J, Pettersson H, Ponzer S, Lapidus L J. Dislocation of total hip replacement in patients with fractures of the femoral neck. Acta Orthop 2009b; 80(2): 184-9.

Gammer W. A modified lateroanterior approach in operations for hip arthroplasty. Clin Orthop Relat Res 1985; (199): 169-72.

Griffin X L, Parsons N, Achten J, Costa M L. A randomised feasibility study comparing total hip arthroplasty with and without dual mobility acetabular component in the treatment of displaced intracapsular fractures of the proximal femur: the Warwick Hip Trauma Evaluation Two: WHiTE Two. Bone Joint J 2016; 98-B(11): 1431-5.

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.

Hailer N P, Garland A, Rogmark C, Garellick G, Kärrholm J. Early mortality and morbidity after total hip arthroplasty in patients with femoral neck fracture. Acta Orthop 2016; 87(6): 560-6.

Hardinge K. The direct lateral approach to the hip. J Bone Joint Surg Br 1982; 64(1): 17-19.

Horriat S, Haddad F S. Dual mobility in hip arthroplasty: what evidence do we need? Bone Joint Res 2018; 7(8): 508-10.

James S, Rao S V, Granger C B. Registry-based randomized clinical trials: a new clinical trial paradigm. Nat Rev Cardiol 2015; 12(5): 312-6.

Jobory A. Dislocation after hip fracture related arthroplasty: incidence, risk factors and prevention. Thesis, Lund University; 2020.

Jobory A, Kärrholm J, Overgaard S, Becic Pedersen A, Hallan G, Gjertsen J E, Makela K, Rogmark C. Reduced revision risk for dual-mobility cup in total hip replacement due to hip fracture: a matched-pair analysis of 9,040 cases from the Nordic Arthroplasty Register Association (NARA). J Bone Joint Surg Am 2019; 101(14): 1278-85.

Johansson T, Jacobsson S A, Ivarsson I, Knutsson A, Wahlstrom O. Internal fixation versus total hip arthroplasty in the treatment of displaced femoral neck fractures: a prospective randomized study of 100 hips. Acta Orthop Scand 2000; 71(6): 597-602.

Jolles B M, Zangger P, Leyvraz P F. Factors predisposing to dislocation after primary total hip arthroplasty: a multivariate analysis. J Arthroplasty 2002; 17(3): 282-8.

Kazley J M, Banerjee S, Abousayed M M, Rosenbaum A J. Classifications in brief: Garden classification of femoral neck fractures. Clin Orthop Relat Res 2018; 476(2): 441-5.

Kim Y H, Choi Y, Kim J S. Influence of patient-, design-, and surgery-related factors on rate of dislocation after primary cementless total hip arthroplasty. J Arthroplasty 2009; 24(8): 1258-63.

Ko L M, Hozack W J. The dual mobility cup: what problems does it solve?

Bone Joint J 2016; 98-B(1 Suppl. A): 60-3.

Kreipke R, Rogmark C, Pedersen A B, Kärrholm J, Hallan G, Havelin L I, Makela K, Overgaard S. Dual mobility cups: effect on risk of revision of primary total hip arthroplasty due to osteoarthritis: a matched populationbased study using the Nordic Arthroplasty Register Association database. J Bone Joint Surg Am 2019; 101(2): 169-76.

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.

Meek R M, Allan D B, McPhillips G, Kerr L, Howie C R. Epidemiology of dislocation after total hip arthroplasty. Clin Orthop Relat Res 2006; 447: 9-18.

Philippot R, Boyer B, Farizon F. Intraprosthetic dislocation: a specific complication of the dual-mobility system. Clin Orthop Relat Res 2013; 471(3): 965-70.

Phillips C B, Barrett J A, Losina E, Mahomed N N, Lingard E A, Guadagnoli E, Baron J A, Harris W H, Poss R, Katz J N. Incidence rates of dislocation, pulmonary embolism, and deep infection during the first six months after elective total hip replacement. J Bone Joint Surg Am 2003; 85(1): 20-6.

Skoldenberg O, Ekman A, Salemyr M, Boden H. Reduced dislocation rate after hip arthroplasty for femoral neck fractures when changing from posterolateral to anterolateral approach. Acta Orthop 2010; 81(5): 583-7.

Acta Orthopaedica 2020; 91 (5): 506–513

Swedish Hip Arthroplasty Register. Annual Report 2017; 2018. Available from: registercentrum.blob.core.windows.net/shpr/r/Eng_Arsrapport_2017 _Hoftprotes_final-Syx2fJPhMN.pdf.

Tabori-Jensen S, Frolich C, Hansen T B, Bovling S, Homilius M, Stilling M. Higher UHMWPE wear-rate in cementless compared with cemented cups with the Saturne(R) Dual-Mobility acetabular system. Hip Int 2018; 28(2): 125-32.

Tabori-Jensen S, Hansen T B, Stilling M. Low dislocation rate of Saturne((R))/ Avantage((R)) dual-mobility THA after displaced femoral neck fracture: a cohort study of 966 hips with a minimum 1.6-year follow-up. Arch Orthop Trauma Surg 2019; 139(5): 605-12.

513

Tarasevicius S, Robertsson O, Dobozinskas P, Wingstrand H. A comparison of outcomes and dislocation rates using dual articulation cups and THA for intracapsular femoral neck fractures. Hip Int 2013; 23(1): 22-6.

Wennergren D, Ekholm C, Sandelin A, Moller M. The Swedish Fracture Register: 103,000 fractures registered. BMC Musculoskelet Disord 2015; 16: 338.

Wolf O, Sjoholm P, Hailer N P, Moller M, Mukka S. Study protocol: HipSTHeR—a register-based randomised controlled trial—hip screws or (total) hip replacement for undisplaced femoral neck fractures in older patients. BMC Geriatr 2020; 20(1): 19.

514

Acta Orthopaedica 2020; 91 (5): 514–519

Study protocol: Effectiveness of dual-mobility cups compared with uni-polar cups for preventing dislocation after primary total hip arthroplasty in elderly patients — design of a randomized controlled trial nested in the Dutch Arthroplasty Registry Loes W A H VAN BEERS 1, Bart C H VAN DER WAL 2, Tess Glastra VAN LOON 1, Dirk Jan F MOOJEN 1, Marieke F VAN WIER 1, Amanda D KLAASSEN 1, Nienke W WILLIGENBURG 1, and Rudolf W POOLMAN 1,3 1 OLVG, Amsterdam; 2 University Medical Center Utrecht, Utrecht; 3 LUMC, Leiden, the Netherlands Collaborator group: E Scheijbeler, OLVG, Amsterdam; D H R Kempen, OLVG, Amsterdam; W P Zijlstra, Medisch Centrum Leeuwarden, Leeuwarden; B Dijkstra, Medisch Centrum Leeuwarden, Leeuwarden; H B Ettema, Isala, Zwolle; M J Q Steinweg, Isala, Zwolle; J L C van Susante, Rijnstate, Arnhem; L D de Jong, Rijnstate, Arnhem; A de Gast, Diakonessenhuis, Utrecht; J Bekkers, Diakonessenhuis, Utrecht; M van Dijk, St Antonius Ziekenhuis, Utrecht; N Wolterbeek, St Antonius Ziekenhuis, Utrecht; T D Berendes, Meander MC, Amersfoort; T Gosens, ETZ, Tilburg; W van der Weegen, St Anna Ziekenhuis, Geldrop, The Netherlands Correspondence: R.W.Poolman@lumc.nl Submitted 2020-05-01. Accepted 2020-05-20.

Background and purpose — Dislocation is the leading reason for early revision surgery after total hip arthroplasty (THA). The dual-mobility (DM) cup was developed to provide more stability and mechanically reduce the risk of dislocation. Despite the increased use of DM cups, high-quality evidence of their (cost-)effectiveness is lacking. The primary objective of this randomized controlled trial (RCT) is to investigate whether there is a difference in the number of hip dislocations following primary THA, using the posterolateral approach, with a DM cup compared with a unipolar (UP) cup in elderly patients 1 year after surgery. Secondary outcomes include the number of revision surgeries, patient-reported outcome measures (PROMs), and cost-effectiveness.

Methods and analysis — This is a prospective multicenter nationwide, single-blinded RCT nested in the Dutch Arthroplasty Registry. Patients ≥ 70 years old, undergoing elective primary THA using the posterolateral approach, will be eligible. After written informed consent, 1,100 participants will be randomly allocated to the intervention or control group. The intervention group receives a THA with a DM cup and the control group a THA with a UP cup.

PROMs are collected preoperatively, and 3 months, 1 and 2 years postoperatively. Primary outcome is the difference in number of dislocations between the UP and DM cup within 1 year, reported in the registry (revisions), or by the patients (closed or open reduction). Data will be analyzed using multilevel models as appropriate for each outcome (linear/ logistic/survival). An economic evaluation will be performed from the healthcare and societal perspective, for dislocation and quality adjusted life years (QALYs).

Trial registration — This RCT is registered at www. clinicaltrials.gov with identification number NCT04031820.

Dislocation after total hip arthroplasty (THA) is the leading reason for early revision surgery (Bozic et al. 2009, Gwam et al. 2017). Most dislocations occur during the first year after surgery, of which approximately half within the first 3 months (Woo and Morrey 1982, Phillips et al. 2003, Meek et al. 2006, Hailer et al. 2012). Especially in patients with recurrent dislocation and the need for revision surgery, this leads to reduced physical functioning and quality of life (Enocson et al. 2009).

Dislocations also increase healthcare costs (Sanchez-Sotelo et al. 2006, Abdel et al. 2015). A single dislocation adds 19% to the hospital costs of an uncomplicated THA, and of a revision surgery up to 148% (Sanchez-Sotelo et al. 2006).

Despite the increased and, in some countries, broad use of DM cups, high-quality evidence of their effectiveness is lacking (Darrith et al. 2018). Recent reviews did not identify any randomized controlled trials (RCT) comparing DM cups with UP cups (De Martino et al. 2017a, 2017b, Darrith et al.

2018, Jonker et al. 2020) and the existing studies are of low methodological quality and at high risk of bias due to the lack of experimental design. So far only one—non randomized— cost-effectiveness study has been performed, suggesting that the DM cup may result in cost savings compared with a UP cup (Epinette et al. 2016). Although promising, the results of this cost-effectiveness database study are not transferrable outside France.

Therefore we initiated an RCT to establish the effectiveness of DM cups for primary THA. The primary objective is to investigate whether there is a difference in the number of hip dislocations following primary total hip arthroplasty (THA), using the posterolateral approach, for a DM cup compared with a UP cup in elderly patients within 1 year after surgery.

Several secondary outcomes will be specified in the methods

© 2020 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 10.1080/17453674.2020.1798658

Acta Orthopaedica 2020; 91 (5): 514–519

section. The registry-nested design will facilitate long-term follow-up for all study participants.

Methods and analysis Study design This is a prospective registry-nested multicenter singleblinded RCT, which will be conducted in 10 general and academic hospitals in the Netherlands. This RCT compares the number of hip dislocations following primary THA with a DM cup compared with a UP cup and is nested in the Dutch Arthroplasty Registry (LROI).

All patients will be followed up until 2 years after surgery.

The recruitment phase started in April 2019 and was anticipated to last 2.5 years. After the first year of recruitment, we have experienced a slight delay. After final study follow-up, participants remain traceable in the LROI for evaluation of long-term survival and mortality.

Participants All patients at the orthopedic outpatient clinics of participating centers that meet the criteria to undergo an elective primary THA will be screened for the in- and exclusion criteria.

Patients can be included when they are 70 years or older; have adequate comprehension of written and spoken Dutch; and are eligible for elective primary THA with a cup large enough for a 32 or 36 millimeter head diameter, by a surgeon who is comfortable using the posterolateral approach. A previous contralateral THA is not a reason for exclusion, but patients who undergo bilateral hip arthroplasty can only participate in the trial with 1 of the hips. Patients will be excluded when they: are not able to complete PROMs; are not eligible for either a UP or DM cup; have epilepsy, spasticity, dementia, mental retardation, or alcoholism. If dementia or mental retardation is not already mentioned in the medical chart, this can be determined by doctor’s opinion.

Characteristics that will be collected are: age; sex; BMI; smoking; diagnosis; ASA classification; Charnley score; education level according to the Statistics Netherlands classification; surgical details (e.g., side, any complications); implant details (e.g., brand, size); type of fixation (cemented or uncemented); type of stem.

Interventions All patients participating in the RCT will be treated with a THA using the posterolateral approach. Patients are randomly allocated to a DM cup or to a UP cup with a 1:1 allocation ratio. It is a requirement for participating surgeons to feel confident with both procedures. The Dutch guidelines recommend reconstruction of the capsule and external rotators when using the posterolateral approach. There are no restrictions to a specific brand of implant, and participating hospitals can use the implants of the companies they usually work

515

with. This study does not investigate any specific implant, but rather pragmatically the concept of DM cups. The Avantage (Zimmer Biomet, Warsaw, IN, USA) and POLAR (Smith & Nephew, London, UK) cups are examples of commonly used DM cups. The IP (Link, Hamburg, Germany), FAL (Link), Exeter (Stryker, Kalamazoo, MI, USA) and Pinnacle (Johnson & Johnson, New Brunswick, NJ, USA) cups are commonly used UP cups. Cemented DM and UP cups have 5-year survival rates of ≥ 96%, with cumulative revision rates ranging from 1.9% to 4.0% when revision was defined as any change (insertion, replacement, and/or removal) of one or more components of the prosthesis, for any reason (LROI 2017b). Lubinus SP2 (Link), Exeter (Stryker), and Corail (Johnson & Johnson) are the commonly used stems.

All patients receive the same standard pre- and postoperative care for both DM and UP cups according to their hospital’s standard.

Sample size calculation Exact dislocation rates in the Netherlands are unknown, as only those dislocations that result in revision surgery are registered. Based on previous studies and reviews, we assume that the current dislocation rate for UP cups is 4% whereas DM cups result in 1% dislocation (Philippot et al. 2009b, Boyer et al. 2012, Fresard et al. 2013, Prudhon et al. 2013, Caton et al. 2014, Batailler et al. 2017, De Martino 2017a).

Power analysis indicates that a total sample of 976 (488 in each group) is needed to detect a difference in dislocations between 4% in the UP cup group and 1% in the DM cup group, using the chi-square test with 80% power and α = 0.05. To account for loss to follow-up, 550 patients will be included in each group.

Outcomes Primary outcome The primary outcome is the number of hip dislocations, regardless of type of treatment. This information is collected from both the LROI and the patient. Since the LROI registers only revisions, open and closed reductions would be missed. Therefore, patients are asked with a questionnaire at 3 months, 1- and 2-year follow-up whether they have had a hip dislocation.

Secondary outcomes Secondary outcomes are any unplanned hip procedures, including revision surgery of any component, for any reason; cost-effectiveness; and PROMs. The following PROMs are collected preoperatively, and 3 months, 1 and 2 years postoperatively: Physical functioning of the hip measured with the Hip disability and Osteoarthritis Outcome Score Physical Short form (HOOS-PS) (de Groot et al. 2007); Quality of life measured with the EuroQol 5 Dimensions (EQ-5D) (EuroQol 1990); pain measured with a numeric rating scale (NRS) ranging from 0 to 10 for pain at

516

rest and during weight-bearing; change in physical functioning measured with an anchor question; fear of hip dislocation measured on a five-point Likert scale. At all postoperative moments, the awareness of type of cup that was placed is asked.

At 3 months and 1 year postoperatively healthcare and societal costs related to hip dislocation or surgery are measured with a retrospective 4-week cost evaluation questionnaire, which is filled out by the patient. We will obtain information on health care utilization, (pain) medication used, patient costs, use of domiciliary care, use of informal care, and sickness absenteeism from paid or unpaid work.

Healthcare utilization consists of general practitioner care, allied healthcare, medical specialist care, imaging tests, admission to a hospital, rehabilitation center, nursing home or care home, and mobility aids. Participants’ costs concern the patient contribution towards costs for mobility aids and travel. Domiciliary care consists of home nursing care and home help. Healthcare utilization, domiciliary care, informal care, and sickness absenteeism will be valued at Dutch standard costs (Hakkaart-van Roijen et al. 2015). If these are not available, prices reported by professional associations will be used. The costs of prescribed medications will be calculated using prices charged by the Royal Dutch Society for Pharmacy.

Study procedures Informed consent During the preoperative visit at the outpatient clinic, patients who are potential candidates for this study will be screened to determine whether they meet the in- and exclusion criteria. If the patient is eligible, the investigator (or his designated representative) will propose participation in the study to the patient, according to GCP guidelines. Patients must sign an informed consent form approved by the ethical committee, prior to participating in any study-specific related activities.

Randomization After signing informed consent, 1,100 patients will be randomized to either the intervention group (DM cup) or the control group (UP cup). Each group will consist of 550 patients. The investigator (or his designated representative) will perform the randomization using the platform CASTOR Electronic Data Capture (www.castoredc.com). Variable randomization blocks of 2, 4, and 6 patients will be used, and we shall stratify for center. Patients will be blinded for treatment allocation. The participating surgeons may divert from the randomization scheme based on intraoperative findings. Any deviation from the assigned treatment group will be reported as a deviation from the protocol.

Follow-up Patients are evaluated at 3 months, 1 year and 2 years after surgery.

Acta Orthopaedica 2020; 91 (5): 514–519

Data analysis plan Interim analysis Interim analysis for the primary study outcome will be performed when 200 patients have reached the 3 months postoperative PROM evaluation point. In the interim analysis the number of dislocations in each group will be compared.

A chi-square test will be used and in any case where the assumptions of this test are not met, Fisher’s exact test will be applied. To guard against a type 1 error, we will use the O’Brien–Fleming approach. As only 1 interim analysis will be performed, the alpha for this analysis is set at 0.005. Testing will be done 2-sided. Furthermore, we will consider the number of revisions and SAEs in each group, but not formally test for differences in these. Results of the interim analysis will be discussed with the study team, the Van Rens Foundation (funder of this study), and the ethical committee. In the case of a statically significant and relevant higher number of dislocations in the DM group, or more revisions or SAEs, appropriate actions will be taken (such as an early termination of the study).

Primary outcome analysis The primary outcome, the difference in number of dislocations in both groups, will be analyzed using chi-square analysis. Additional exploratory multivariable logistic regression analyses will adjust for clustering of data (e.g., at the hospital level), and possible confounding or effect modification of patient and surgical characteristics (e.g., age; sex; BMI; smoking; diagnosis; ASA classification; Charnley score; education level according to the Statistics Netherlands classification; surgical details; implant details; type of fixation; type of stem). A multilevel survival model will be used to analyze the survival of the implant, corrected for covariates.

Analyses will be performed using both intention-to-treat as well as per-protocol analysis.

Missing values Efforts will be made to prevent missing data by sending reminders and making phone calls when appropriate. A reasonable amount of dropouts is anticipated, and mixed-model analyses will account for missing data using maximum likelihood estimation. In the event of unforeseen numbers of missing values, a state-of-the-art solution will be sought in consultation with a statistician (e.g., imputation, depending on the nature of the missing data).

Secondary outcomes analyses Secondary study outcomes are any surgical intervention on the affected hip including revision surgery, healthcare costs, societal costs, patient-reported physical functioning, quality of life, pain, satisfaction, fear of hip dislocation and devicerelated complications and reoperations. The secondary outcomes will be analyzed using similar multilevel models as appropriate for each outcome (linear/logistic/survival).

Acta Orthopaedica 2020; 91 (5): 514–519

An economic evaluation will be performed from the healthcare and societal perspective, for dislocation and quality adjusted life years (QALYs). Prevailing guidelines of Zorginstituut Nederland will be observed. All costs and consequences relevant to THA, hip dislocation, and hip revision will be accounted for.

To compare costs between groups, confidence intervals around the mean differences in costs at one year after THA will be estimated using the bias-corrected and accelerated bootstrap method. To account for possible clustering of data and to adjust for possible confounders, multi-level analyses will be performed. To present the incremental cost-effectiveness ratios and uncertainty around them graphically, bootstrapped cost-effect pairs will be plotted on cost-effectiveness planes. Cost-effectiveness acceptability curves will present the probability that the DM cup is more cost-effective than the UP cup for a range of willingness-to-pay thresholds. To study the robustness of these results, sensitivity analyses will be performed.

Discussion To the authors’ knowledge, this is the first RCT comparing UP and DM cups for primary THA. In contrast to the observational nature of all (registry) studies to date, this study will be able to draw causal inferences. Previous literature is mostly from France, where DM cups are already used in approximately 30% of all primary THAs (Epinette et al. 2016). Dislocation rates seem lower for dual mobility (DM) cups (range 0% to 3.6%) than for unipolar cups (range 0.5% to 6%) (van der Grinten and Verhaar 2003, Bourne and Mehin 2004, Jolles and Bogoch 2004, Malkani et al. 2010, Lachiewicz and Soileau 2013, Dargel et al. 2014). Good results are also shown when DM cups are used in revision surgery for patients with recurrent dislocation (Langlais et al. 2008, Philippot et al. 2009a, Hailer et al. 2012). The Dutch Arthroplasty Registry shows that 3.9% of all cemented cups in 2015 were DM cups (LROI 2017a).

The proportion of DM cups in all primary THA increased from 0.8% in 2010 to 2.6% in 2016 (Bloemheuvel et al. 2019). In the Netherlands and other countries, DM cups are typically used for primary THA in patients with specific characteristics, such as cognitive impairment (not able to follow restrictions after surgery), neuromuscular diseases (spasms), or alcohol abuse, or as a standard procedure for revision surgeries due to recurrent dislocations (De Martino et al. 2017a, Bloemheuvel et al.

2019). These patient characteristics might negatively influence the risk for dislocation and revision surgery, so data of these specific patient groups cannot be generalized to the regular primary THA population.

Our registry-nested randomized design is an efficient way to obtain an unbiased comparison between DM and UP cups, both in the short term and long term. Currently, dislocations

517

are only reported in the registry if they result in implant revision. Therefore, the primary—relevant to patients—outcome of this study is a composite measure of revisions due to dislocation reported in the registry and patient-reported dislocations that were treated with closed or open reduction. Not many studies used such a composite outcome, which complicated our sample size calculation. The current group sizes are based on informed assumptions, and considered large enough to detect substantial differences between groups. However, regarding this limitation we believe it is fair to compare groups in terms of dislocation rates with corresponding confidence intervals rather than strictly focusing on p-values (Wasserstein and Lazar 2016). Also, the registry-nested design does allow for comparison with large groups of patients who underwent similar hip replacement surgery outside the study. Another limitation is that we do not collect radiographic outcomes for each participant.

The literature shows good survival rates up to 10 years for DM cups, ranging from 90.4% to 100% (Clave et al. 2016, Martz et al. 2017, Puch et al. 2017, Tarasevicius et al. 2017, Laurendon et al. 2018, Spaans et al. 2018, Cypres et al. 2019, Fessy et al. 2019, de l’Escalopier et al. 2020). Nevertheless, our population includes only patients aged 70 and older to minimize risk of revision for other indications such as loosening and wear. The study results may therefore promote additional research with a younger study population that is generally more active.

Important strengths of this study are that we will keep track of complications (serious adverse events) other than dislocations as well. In the long term, we shall be able to study survival of the implants as well as mortality in both study groups, as these remain available in the LROI. Finally, this trial not only evaluates effectiveness, but also the costs associated with both interventions. Such a trial-based economic evaluation is important to determine whether DM cups, which are typically more expensive, are worthwhile in a population undergoing primary THP. Ethics, registration, funding, and potential conflicts of interest This study (NL64819.100.18) is approved by the Medical research Ethics Committees United, the Netherlands, and will be conducted according to the principles of the Declaration of Helsinki (2013) and in accordance with the Medical Research Involving Human Subjects Act (WMO) and Good Clinical Practice guidelines.

The protocol of this trial is registered at clinicaltrials.gov (NCT04031820) and will be published. The main and secondary results of this study will be reported in international peerreviewed journals.

This study is funded with a grant by the Van Rens Foundation, the Netherlands, identification number VRF2018-003.

No competing interests declared.

518

All co-authors (LvB, BvdW, TGvL, DJM, MvW, AK, NW, and RP) have contributed to the concept and design of this study. LvB, BvdW, DJM, NW, and RP have contributed to the writing process of this manuscript. All authors have revised and approved this manuscript.

All members of the collaborator group have contributed to the concept of the initial protocol and agreed with this manuscript.

Abdel M P, Cross M B, Yasen A T, Haddad F S. The functional and financial impact of isolated and recurrent dislocation after total hip arthroplasty.

Bone Joint J 2015; 97-B(8): 1046-9.

Batailler C, Fary C, Verdier R, Aslanian T, Caton J, Lustig S. The evolution of outcomes and indications for the dual-mobility cup: a systematic review.

Int Orthop 2017; 41(3): 645-59.

Bloemheuvel E M, van Steenbergen L N, Swierstra B A. Dual mobility cups in primary total hip arthroplasties: trend over time in use, patient characteristics, and mid-term revision in 3,038 cases in the Dutch Arthroplasty Register (2007–2016). Acta Orthop 2019; 90(1): 11-14. doi: 10.1080/17453674.2018.1542210.

Bourne R B, Mehin R. The dislocating hip: what to do, what to do. J Arthroplasty 2004; 19(4 Suppl. 1): 111-4.

Boyer B, Philippot R, Geringer J, Farizon F. Primary total hip arthroplasty with dual mobility socket to prevent dislocation: a 22-year follow-up of 240 hips. Int Orthop 2012; 36(3): 511-8.

Bozic K J, Kurtz S M, Lau E, Ong K, Vail T P, Berry D J. The epidemiology of revision total hip arthroplasty in the United States. J Bone Joint Surg Am 2009; 91(1): 128-33. doi: 10.2106/JBJS.H.00155.

Caton J H, Prudhon J L, Ferreira A, Aslanian T, Verdier R. A comparative and retrospective study of three hundred and twenty primary Charnley type hip replacements with a minimum follow up of ten years to assess whether a dual mobility cup has a decreased dislocation risk. Int Orthop 2014; 38(6): 1125-9.

Clave A, Maurer D, Tristan L, Dubrana F, Lefevre C, Pandit H. Midterm survivorship of the Lefevre Constrained Liner: a consecutive multisurgeon series of 166 cases. J Arthroplasty 2016; 31(9): 1970-8. doi: 10.1016/j. arth.2016.02.031.

Cypres A, Fiquet A, Girardin P, Fitch D, Bauchu P, Bonnard O, Noyer D, Roy C. Long-term outcomes of a dual-mobility cup and cementless tripletaper femoral stem combination in total hip replacement: a multicenter retrospective analysis. J Orthop Surg Res 2019; 14(1): 376. doi: 10.1186/ s13018-019-1436-y.

Dargel J, Oppermann J, Bruggemann G P, Eysel P. Dislocation following total hip replacement. Dtsch Arztebl Int 2014; 111(51-52): 884-90.

Darrith B, Courtney P M, Della Valle C J. Outcomes of dual mobility components in total hip arthroplasty: a systematic review of the literature. Bone Joint J 2018; 100-B(1): 11-19. de Groot I B, Reijman M, Terwee C B, Bierma-Zeinstra S M, Favejee M, Roos E M, Verhaar J A. Validation of the Dutch version of the Hip disability and Osteoarthritis Outcome Score. Osteoarthritis Cartilage 2007; 15(1): 104-9. de l’Escalopier N, Dumaine V, Auberger G, Babinet A, Courpied J P, Anract P, Hamadouche M. Dual mobility constructs in revision total hip arthroplasty: survivorship analysis in recurrent dislocation versus other indications at three to twelve-year follow-up. Int Orthop 2020; 44(2): 253-60. doi: 10.1007/s00264-019-04445-4.

De Martino I, D’Apolito R, Soranoglou V G, Poultsides L A, Sculco P K, Sculco T P. Dislocation following total hip arthroplasty using dual mobility acetabular components: a systematic review. Bone Joint J 2017a; 99-B(ASuppl. 1): 18-24.

De Martino I, D’Apolito R, Waddell B S, McLawhorn A S, Sculco P K, Sculco T P. Early intraprosthetic dislocation in dual-mobility implants: a systematic review. Arthroplast Today 2017b; 3(3): 197-202.

Enocson A, Pettersson H, Ponzer S, Tornkvist H, Dalen N, Tidermark J. Quality of life after dislocation of hip arthroplasty: a prospective cohort study on 319 patients with femoral neck fractures with a one-year follow-up. Qual Life Res 2009; 18(9): 1177-84.

Acta Orthopaedica 2020; 91 (5): 514–519

Epinette J A, Lafuma A, Robert J, Doz M. Cost-effectiveness model comparing dual-mobility to fixed-bearing designs for total hip replacement in France. Orthop Traumatol Surg Res 2016; 102(2): 143-8.

EuroQol. A new facility for the measurement of health-related quality of life.

Health Policy 1990; 16(3): 199-208.

Fessy M H, Jacquot L, Rollier J C, Chouteau J, Ait-Si-Selmi T, Bothorel H, Chatelet J C. Midterm clinical and radiographic outcomes of a contemporary monoblock dual-mobility cup in uncemented total hip arthroplasty. J Arthroplasty 2019; 34(12): 2983-91. doi: 10.1016/j.arth.2019.07.026.

Fresard P L, Alvherne C, Cartier J L, Cuinet P, Lantuejoul J P. Seven-year results of a press-fit, hydroxyapatite-coated double mobility acetabular component in patients aged 65 years or older. Eur J Orthop Surg Traumatol 2013; 23(4): 425-9.

Gwam C U, Mistry J B, Mohamed N S, Thomas M, Bigart K C, Mont M A, Delanois R E. Current epidemiology of revision total hip arthroplasty in the United States: national inpatient sample 2009 to 2013. J Arthroplasty 2017; 32(7): 2088-92. doi: 10.1016/j.arth.2017.02.046.

Hailer N P, Weiss R J, Stark A, Karrholm 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.

Hakkaart-van Roijen R L, van der Linden N, Bouwmans C A, Kanters T, Tan S S. Kostenhandleiding: Methodologie van kostenonderzoek en referentieprijzen voor economische evaluaties in de gezondsheidszorg (Bijlage 1).

Zorginstituut Nederland; 2015.

Jolles B M, Bogoch E R. Posterior versus lateral surgical approach for total hip arthroplasty in adults with osteoarthritis. Cochrane Database Syst Rev 2004(1): CD003828.

Jonker R C, van Beers L W A H, van der Wal B C H, Vogely H C, Parratte S, Castelein R M, Poolman R W. Can dual mobility cups prevent dislocation without increasing revision rates in primary total hip arthroplasty? A systematic review. Orthop Traumatol Surg Res 2020; 106(3): 509-17.

Lachiewicz P F, Soileau E S. Low early and late dislocation rates with 36- and 40-mm heads in patients at high risk for dislocation. Clin Orthop Relat Res 2013; 471(2): 439-43.

Langlais F L, Ropars M, Gaucher F, Musset T, Chaix O. Dual mobility cemented cups have low dislocation rates in THA revisions. Clin Orthop Relat Res 2008; 466(2): 389-95.

Laurendon L, Philippot R, Neri T, Boyer B, Farizon F. Ten-year clinical and radiological outcomes of 100 total hip arthroplasty cases with a modern cementless dual mobility cup. Surg Technol Int 2018; 32: 331-6.

LROI. http://www.lroi-rapportage.nl/heup-meest-geplaatste-componenten; 2017a.

LROI. http://www.lroi-rapportage.nl/hip-survival-revision-within-1-3-and-5years-per-tha-component-cemented-acetabular-component; 2017b.

Malkani A L, Ong K L, Lau E, Kurtz S M, Justice B J, Manley M T. Early- and late-term dislocation risk after primary hip arthroplasty in the Medicare population. J Arthroplasty 2010; 25(6 Suppl.): 21-5.

Martz P, Maczynski A, Elsair S, Labattut L, Viard B, Baulot E. Total hip arthroplasty with dual mobility cup in osteonecrosis of the femoral head in young patients: over ten years of follow-up. Int Orthop 2017; 41(3): 60510. doi: 10.1007/s00264-016-3344-7.

Meek R M, Allan D B, McPhillips G, Kerr L, Howie C R. Epidemiology of dislocation after total hip arthroplasty. Clin Orthop Relat Res 2006; 447: 9-18.

Philippot R, Adam P, Reckhaus M, Delangle F, Verdot F, Curvale G, Farizon F. Prevention of dislocation in total hip revision surgery using a dual mobility design. Orthop Traumatol Surg Res 2009a; 95(6): 407-13.

Philippot R, Camilleri J P, Boyer B, Adam P, Farizon F. The use of a dualarticulation acetabular cup system to prevent dislocation after primary total hip arthroplasty: analysis of 384 cases at a mean follow-up of 15 years. Int Orthop 2009b; 33(4): 927-32.

Phillips C B, Barrett J A, Losina E, Mahomed N N, Lingard E A, Guadagnoli E, Baron J A, Harris W H, Poss R, Katz J N. Incidence rates of dislocation, pulmonary embolism, and deep infection during the first six months after elective total hip replacement. J Bone Joint Surg Am 2003; 85-A(1): 20-6.

Prudhon J L, Ferreira A, Verdier R. Dual mobility cup: dislocation rate and survivorship at ten years of follow-up. Int Orthop 2013; 37(12): 2345-50.

Acta Orthopaedica 2020; 91 (5): 514â&#x20AC;&#x201C;519

Puch J M, Derhi G, Descamps L, Verdier R, Caton J H. Dual-mobility cup in total hip arthroplasty in patients less than fifty five years and over ten years of follow-up: a prospective and comparative series. Int Orthop 2017; 41(3): 475-80. doi: 10.1007/s00264-016-3325-x.

Sanchez-Sotelo J, Haidukewych G J, Boberg C J. Hospital cost of dislocation after primary total hip arthroplasty. J Bone Joint Surg Am 2006; 88(2): 290-4.

Spaans E A, Koenraadt K L M, Wagenmakers R, van den Hout J, Te Stroet M A J, Bolder S B T. Midterm survival analysis of a cemented dual-mobility cup combined with bone impaction grafting in 102 revision hip arthroplasties. Hip Int 2018; 28(2): 161-7. doi: 10.5301/hipint.5000548.

519

Tarasevicius S, Smailys A, Grigaitis K, Robertsson O, Stucinskas J. Shortterm outcome after total hip arthroplasty using dual-mobility cup: report from Lithuanian Arthroplasty Register. Int Orthop 2017; 41(3): 595-8. doi: 10.1007/s00264-016-3389-7. van der Grinten M, Verhaar J A. [Dislocation of total hip prostheses; risk factors and treatment]. Ned Tijdschr Geneeskd 2003; 147(7): 286-90.

Wasserstein R L, Lazar N A. The ASA statement on p-values: context, process, and purpose. American Statistician 2016; 70(2): 129-33. doi: 10.1080/00031305.2016.1154108 Woo R Y, Morrey B F. Dislocations after total hip arthroplasty. J Bone Joint Surg Am 1982; 64(9): 1295-306.

520

Acta Orthopaedica 2020; 91 (5): 520–522

Perspective

Simulation-based skills training in non-performing orthopedic surgeons: skills acquisition, motivation, and flow during the COVID-19 pandemic Leif Rune HEDMAN 1,2, and Li FELLÄNDER-TSAI 1 1 Department of Clinical Science, Intervention and Technology (CLINTEC), Division of Orthopedics and Biotechnology, Karolinska Institutet, Stockholm, Sweden, and Center for Advanced Medical Simulation and Training (CAMST), Karolinska University Hospital, Stockholm; 2 Department of Psychology, Umeå University, Umeå, Sweden Correspondence: Leif.hedman@ki.se Submitted 2020-05-14. Accepted 2020-06-01.

The COVID-19 pandemic will continue to have a profound effect on society, including orthopedic surgery (Felländer-Tsai 2020, Wright et al. 2020). Educational crisis management is mandated. As elective surgeries are being postponed for many orthopedic residents they must change their daily clinical routines. Increased opportunities for virtual and simulated surgical training can facilitate trainees/residents/experts retaining basic competence. A way to maintain skills acquisition and to prevent skill decay is to provide increased possibilities to practice simulated surgical tasks, i.e., psychomotor training including virtual learning such as 3D visualization.

For a growing number of minimally invasive and technically challenging orthopedic procedures, there is a movement to improve surgical skills training outside the operating room because of, e.g., patient safety concerns (Atesok et al. 2016).

Surgical simulation is a powerful tool that can help meet these training demands (Felländer-Tsai et al. 2004, Gallagher et al.

2005, Johnston et al. 2016). Kogan et al. (2020) pointed out that quarantine away from patient care, active social distancing, and diminished case volume during the pandemic will increase the need for surgical simulation training.

Motivation in simulation training Simulation-based skills training including 3D visualizations is becoming increasingly integrated into surgical education as an important teaching method across most surgical specialties. Proficiency and progression-based simulation training is today an influential approach to surgical simulation-based skills training and medical education (Gallagher et al. 2005, Ahlberg et al. 2007, Stefanidis et al. 2010). This kind of training is considered to improve general motivation, skills training, and skill retention.

Wallace et al. (2017) have remarked that most of the studies in the field of cognitive training are without underlying theory.

The psychological mechanisms underlying the efficacy of cognitive training programs are not well known. There is also

a lack of theoretically based studies of motivational processes in surgical simulation training.

Self-determination theory (SDT) claims that when the basic needs for competence, autonomy, and relatedness are satisfied, personal well-being and social development are optimized (Deci and Ryan 2000). In this condition individuals are intrinsically motivated, able to fulfill their potential, and able to seek out progressively greater challenges. The intrinsic types of motivation are the most self-determined and are performed for the satisfaction gained from the activity. Extrinsic motivation lies at the lower end of the self-determination scale. The extrinsic motivation types ranging from most self-determined to least selfdetermined are: identified regulation; introjected regulation; and external regulation. Amotivation is characterized by the absence of motivation. The different types of motivation postulated by SDT have proved to be meaningful in order to predict the level of engagement in a variety of life domains. Research has shown that there is a strong positive correlation between intrinsic motivation and good work performance (Deci and Ryan op.cit).

Csikszentmihályi et al. (2005) defined flow as a mental experience of intensive enjoyment characterized by complete concentration, heightened sense of control, merging of action and attention, loss of self-consciousness, distortion of time perception, and internally driven. Flow may be described as a temporary psychological state of arousal/ attention involving positive feelings during a specific moment-to-moment activity. Flow treated as a state, assessed by its intensity, can foster positive affect and stimulate positive educational outcomes (Cerasoli et al. 2014, Csikszentmihályi 2014). According to flow theory, flow can increase intrinsic motivation toward a specific activity (Moneta 2012, De Fraga and Moneta 2016).

Flow experience has been studied in fields such as performing arts and sport, education, neuropsychology, educational psychology, social psychology work, and everyday activities (De Fraga and Moneta op.cit).

© 2020 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 10.1080/17453674.2020.1781413

Acta Orthopaedica 2020; 91 (5): 520–522

The extensively validated simulator for generic minimally invasive navigational skills, MIST VR (Gallagher et al. 2005), and also newer and more advanced simulators, have long proved to be very suitable training environments. We have identified state motivation as an important variable for minimally invasive surgical performance both when training in surgical simulators and during acute team training (Schlickum et al. 2013). Moreover, a state of flow appears to contribute to both real-life and simulated learning of minimal invasive skills (Ahlborg et al. 2015).

Engagement in a simulated basic skill task: example from a pilot experiment There is still a lack of detailed knowledge of different motivational states while reaching a predetermined criterion level of performance. As part of the previously published investigation (Schlickum et al. 2016) the present authors collected and analyzed data on flow experience and situational motivation from 30 surgical novices. All received a standardized verbal introduction and a standardized video-demonstration of a well-studied basic task (manipulative diathermy at medium difficulty) in a validated minimally invasive surgery trainingvirtual reality simulator (MIST-VR). Training was then performed for approximately 1 hour until a pre-set criterion level was reached following measurement of performance. Motivation was then self-assessed by validated instruments: a short and long flow state scale (Jackson et al. 2008), and a situational motivation scale (Guay et al. 2000). We found no statistically significant correlations between types of situational motivation and MIST-VR performance. Presumably, the manipulation task in this low-fidelity simulator was too simple. However, flow state scores for the short and long version of the flow scales were positively correlated, and flow state assessed by the short flow state scale was correlated with the best performance (data not shown). Notably, flow state assessed by the short state flow scale were positively correlated with intrinsic motivation (data not shown). The pilot experiment indicates that surgical novices’ flow state was correlated to performance in the simulated basic skills task.

The future Findings in training research should be considered when new training programs have to be developed, delivered, and monitored. Salas et al. (2012) have provided evidence-based recommendations for maximizing training effectiveness before, during, and after training. We recommend using their checklists. They provide insight into best practices for training skills that are durable and resistant to skill decay, assessing skills over time, and repeating training at appropriate intervals.

Most of the proposed actions will promote motivation to learn and increase flow. Moreover, to foster motivation during skill acquisition in simulation training, feedback of performance is critical and given: (1) automatically and immediately by the simulator itself, (2) by the instructor giving verbal feedback

521

during training (technical and pedagogical support), and (3) by the instructor and other team members after the training sessions (debriefing). All kinds of feedback can offer opportunities to tailor the trainee’s abilities and feelings during the process of skill acquisition. As in ideomotoric training (Immenroth et al. 2007), where the surgeon imagines him- or herself performing a procedure, the sensory aspects of the procedure should be included as far as possible in order to elicit motivational and arousal components of cognitive training.

Atesok et al. (2016) have reviewed several barriers to overcome when integrating the simulation of surgical skill training into orthopedic surgical education. They discuss the value of proficiency and progression-based simulator training as the next-generation approach, and using measurable criteria for technical performance. The outcome measures of technical skill performance should be quantitative metric. The model presented by Khamis et al. (2016) integrates principles of curriculum development and simulation design. It involves best practices for, e.g., individualized learning, providing formative and summative feedback, deliberate practice with formative feedback, and mastery learning progressing from novice, competent, proficient, expert to master.

Conclusion When used appropriately, extended simulation training can be a highly effective additional training tool in the development and maintenance of technical skills and combatting skills decay, taking into account motivation and flow. This is relevant also for temporarily non-performing orthopedic surgeons during a crisis affecting the organization of healthcare such as the COVID-19 pandemic.

Funding and conflict of interests Funds were received to conduct the study from an independent agency. The authors report no conflicts of interest.

LRH: Investigation, original draft preparation, final draft writing. LF-T: Final draft writing.

Ahlberg G, Enochsson L, Gallagher A G, Hedman L, Hogman C, McClusky DA 3rd, Ramel S, Smith C D, Arvidsson, D. Proficiency-based virtual reality training significantly reduces the error rate for residents during their first 10 laparoscopic cholecystectomies. Am J Surg 2007; 193(6): 797-804.

Ahlborg L, Weurlander M, Hedman L, Nisell H, Lindqvist P G, FelländerTsai L, Enochsson L. Individualized feedback during simulated laparoscopic training: a mixed methods study. Int J Med Educ 2015; 6: 93-100. doi:10.5116/ijme.55a2.218b.

Atesok K, Satava R M, Van Heest A, Hogan M V, Pedovits R A, Fu F H, Sitnikov I, Marsh J, Hurwitz S L. Retention of skills after simulation-based training in orthopaedic surgery. J Am Acad Orthop Surg 2016; 24 (8): 50514. doi: 10.5435/JAAOS-D-15-00440.

Cerasoli C P, Nicklin J M, Ford M T. Intrinsic motivation and extrinsic incentives jointly predict performance: a 40-year meta-analysis. Psychol Bulletin 2014; 140 (4): 980-1008. doi: 10.1037/a0035661.

Csikszentmihályi M. Intrinsic motivation and effective teaching. In: Csikszentmihályi M, editor. Applications of flow in human development and

522

education: the collected works of Mihaly Csikszentmihalyi, Chapter 8.

Dordrecht: Springer; 2014. p. 173-86. doi: 10.1007/978-94-017-9094-9.

Csikszentmihályi M, Abuhamdeh S, Nakamura J. Flow. In: Elliot AJ, Dweck C S, editors. Handbook of competence and motivation. New York: Guilford Press; 2005. p. 598-608.

Deci E L, Ryan R M. The ‘what’ and ‘why’ of goal pursuits: human needs and the self-determination of behavior. Psychol Inq 2000; 11(4): 227-68. doi: 10.1207/S15327965PLI1104_01.

De Fraga D, Moneta B. Flow in work as a moderator of the self-determination model of work engagement. In Harmat L, Ørsted Andersen F, Ullén F, Wright F J, Sadlo G, editors. The flow experience: empirical research and applications. New York: Springer; 2016. p. 105–23. doi: 10.1007/9783-319-28634-1.

Felländer-Tsai L. Pandemic pressure: policy, politics, profession, and rapid publication, Acta Orthop 2020. doi: 10.1080/17453674.2020.1753162.

Felländer-Tsai L, Kjellin A, Wredmark T, Ahlberg G, Anderberg B, Enochsson L, Hedman L, Johnson E, Mäkinen K, Ramel S, Ström P, Särnå L, Westman B. Basic accreditation for invasive image-guided intervention: a shift of paradigm in high technology education, embedding performance criterion levels in advanced medical simulators in a modern educational curriculum. J Inf Techn Healthc 2004; 3(2): 165-73.

Gallagher A G, Ritter E M, Champion H, Higgins G, Fried M P, Moses G, Smith C D, Satava R M. Virtual reality simulation for the operating room: proficiency-based training as a paradigm shift in surgical skills training.

Ann Surg 2005; 241(2): 364-72. doi:10.1097/01.sla.0000151982.85062.80.

Guay F, Vallerand R J, Blanchard C. On the assessment of situational intrinsic and extrinsic motivation: the Situational Motivation Scale. Motivation and Emotion 2000 24(3): 175-213.

Immenroth M, Bürger T, Brenner J, Nagelschmidt M, Eberspächer H, Troidl H. Mental training in surgical education: a randomized controlled trial.

Ann Surg 2007; 245(3): 385-91.

Jackson S A, Martin A J, Eklund R C. Long and short measures of flow: the construct validity of the FSS-2, DFS-2, and new brief counterparts. J Sport Exerc Psychol 2008; 30(5): 561-87. doi: 10.1123/jsep.30.5.561.

Acta Orthopaedica 2020; 91 (5): 520–522

Johnston M J, Paige J T, Aggarwal R, Stefanidis D, Tsuda S, Aora S. An overview of research priorities in surgical simulation: what the literature has achieved during the 21st century and what remains. Am J Surg 2016; 211(1): 214-25. doi: 10.1016/j.amjsurg.2015.06.014.

Khamis N N, Satava R M, Alnassar S, Kern D E. A stepwise model for simulation-based curriculum development for clinical skills, a modification of the six-step approach. Surg Endosc 2016; 30: 279-87. doi: 10.1007/ s00464-015-4206-x.

Kogan M, Klein S E, Hannon C P, Nolte M T. Orthopaedic education during the COVID-19 pandemic. J Am Acad Orthop Surg 2020. doi: 10.5435/ JAAOS-D-20-00292. [Published online ahead of print May 7, 2020].

Moneta G B. Opportunity for creativity in the job as a moderator of the relation between trait intrinsic motivation and flow in work. Motivation and Emotion 2012; 36; 491-503. doi 10.1007/s11031-012-9278-5.

Salas E, Tannenbaum SI, Kraiger K, Smith-Jentsch K A. The science of training and development in organizations: what matters in practice. Psychol Science Public Interest 2012; 13(2): 74-101. doi: 10.1177/1529100612436661.

Schlickum M, Felländer-Tsai L, Hedman L, Henningsohn L. Endourological simulator performance in female but not male medical students predicts written examination results in basic surgery. Scand J Urol 2013; 47(1): 38-42. doi:10.1007/s00464-007-9287-8.

Schlickum M, Hedman L, Felländer-Tsai L. Visual-spatial ability is more important than motivation for novices in surgical simulator training: a preliminary study. Int J Med Educ 2016; 7: 56-61. doi: 10.5116/ ijme.56b1.1691.

Stefanidis D, Acker C E, Greene F L. Performance goals on simulators boost resident motivation and skills laboratory attendance. J Surg Educ 2010; 67(2): 66-70. doi: 10.1016/j.jsurg.2010.02.002.

Wallace L, Raison N, Ghumman F, Moran A, Dasgupta P, Ahmed K. Cognitive training: how can it be adapted for surgical education? Surgeon 2017; 15(4): 231-9. doi: 10.1016/j.surge.2016.08.003. Epub 2016 Sep 19.

Wright R W, Armstrong A D, Azar F M, et al. The American Board of Orthopaedic Surgery response to COVID-19. J Am Acad Orthop Surgeons 2020. doi: 10.5435/JAAOS-D-20-00392 [published online ahead of print May 7].

Acta Orthopaedica 2020; 91 (5): 523–526

523

Perspective

Cognitive training for the prevention of skill decay in temporarily non-performing orthopedic surgeons Robi KELC 1,2, Matjaz VOGRIN 1,2, and Janja KELC 3 1 Department of Orthopedic Surgery, University Medical Center Maribor; 2 Institute of Sports Medicine, FIFA Medical Center of Excellence, Faculty of Medicine, University of Maribor; 3 Department of Psychiatry, University Medical Center Maribor, Slovenia Correspondence: robi.kelc@ukc-mb.si Submitted 2020-04-09. Accepted 2020-05-08.

This paper has previously been shared with the Slovenian Orthopedic Association as a blog post.

Abstract — Surgical tasks are prone to skill decay. During unprecedented circumstances, such as an epidemic, personal illness, or injury, orthopedic surgeons may not be performing surgical procedures for an uncertain period of time. While not being able to execute regular surgical tasks or use surgical simulators, skill decay can be prevented with regular mental practice, using a scientifically proven skill acquisition and retaining tool. This paper describes different theories on cognitive training answering the question on how it works and offers a brief review of its application in surgery.

Additionally, practical recommendations are proposed for performing mental training while not performing surgical procedures.

Surgical tasks combine many gross and fine motor actions in a strict time frame, demanding a high degree of accuracy. As such, surgical skills usually present with a flat learning curve, especially in the case of minimally invasive operations. However, once acquired, surgical skills are prone to decay, especially after a period of not practicing (Sonnadara et al. 2012, Routt et al. 2015). Other important risk factors affecting decay are time pressure and the quality of the job performed (Wisher et al. 1999), both being vital components of surgical performance.

During special circumstances, such as an epidemic (e.g., COVID-19), surgeons may be relocated to working environments away from their field of expertise, thereby not performing surgical procedures for a period of time. Similarly, an interval of not practicing is inevitable in the case of a surgeon’s illness or injury. While not being able to perform regular sur-

gical tasks or use surgical simulators, the prevention of skill decay can be achieved through regular mental practice, using a scientifically proven skill acquisition and retaining tool.

Mental practice or imagery, a form of cognitive training, is a symbolic rehearsal of a physical activity in the absence of any gross muscular movements (Richardson 1967). It is widely implemented in sports where it has long been used with success in enhancing the performance of elite athletes (Martin et al. 1999) and in certain other areas, such as aviation (Fornette et al. 2012), professional music (Bernardi et al. 2013), and surgery (Wallace et al. 2017). All these fields share crucial similarities, such as the importance of technical skills, performance under stressful conditions, and aiming for perfection without making mistakes. Mental practice improves a variety of different motor skills in sports, as well as acquisition, physical strength (Sevdalis et al. 2013), and technique performance (Surburg 1968), hence its application in surgery is not only scientifically justified but also common sense.

It is believed that experienced subjects may benefit more from mental practice on physical tasks because of the requisite schematic knowledge to imagine an accurate and a precise outcome associated with the imagined performance (Posner 1989). For example, experienced athletes have better visualizing abilities and employ more structured mental practice sessions in comparison with novices (Feltz and Landers 2007).

However, mental practice has also recently gained in popularity with novice surgeons, surgical trainees, and medical students learning new surgical techniques (Hall 2002, Sanders et al. 2004). Cognition, integration, and automation are typical steps in learning new surgical skills (Hamdorf and Hall 2000), the first two being most decay-susceptible and therefore a subject of interest for mental practice as a retaining tool.

© 2020 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 10.1080/17453674.2020.1771520

524

This article provides a brief theoretical background on cognitive training, followed by its application in surgery and surgical education. Finally, recommendations are provided for novice and expert surgeons for performing cognitive training while not performing surgery for a long or uncertain period of time.

Theoretical basis for cognitive training The motor system has been hypothesized to be part of a cognitive network including a variety of psychological activities.

During cognitive training and real-life motor tasks, similar neural paths are being activated. In musicians, a close relationship between motor imagery and motor action has been described, for example changes in corticospinal activity with the same muscles involved in both circumstances (Fadiga et al. 1999).

There are different theories on why mental practice improves motor skills (Vealey and Walter 1993). The psychoneuromuscular theory proposes that mental training causes activation pattern of muscles similar to actual movements (Jacobson 1931). The symbolic learning theory postulates that symbols are coding the sequence of movements (Sackett 1934). The repetition of symbolic components of the movement pattern facilitates execution of an actual motor pattern (Frank et al.

2014). A more recent theory suggests that motor imagery and motor performance are functionally equivalent, thereby suggesting that in both the same underlying neural structures and mechanisms are involved (Jeannerod 1994).

Cognitive training in surgery Surgery as a medical specialty containing complex psychomotor and cognitive tasks is without any doubt subject to skill decay when tasks are not being performed for a certain period (Arthur et al. 1998). It is not only a question of one’s interest but is also a surgeon’s duty to master the skills and retain them. Novice surgeons and surgical trainees usually find themselves in cognitive and integrational phases of learning, thus being especially vulnerable to the decay of their skills, which can happen after a short retention interval (Sugihara et al. 2018). On such occasions, cognitive training can be of critical importance as it has been proven to enhance knowledge of a procedure, flow of an operation, and preparedness for the task (Komesu et al. 2009).

There are numerous cognitive training techniques. At a novice level, cognitive task analysis (CTA) training has been shown to be the most effective, a method by which experts are used to construct a teaching program for novices through intuitive knowledge and thought processes (Tofel-Grehl and Feldon 2013, Wingfield et al. 2015). CTA has recently been proven to be an effective technique in hip arthroplasty, where its use resulted in shorter procedure time, decrease in the number of errors, and increase in accuracy of acetabular cup orientation (Logishetty et al. 2020). However, CTA requires a mentor to be present, which is potentially an inaccessible

Acta Orthopaedica 2020; 91 (5): 523–526

option in some specific situations. In this case, other techniques of cognitive training, such as external observative and subvocal training, are also potentially useful. In the former, a surgeon is an observer of a skill that is to be learned whereas in the latter a visual image is being called up by a surgeon through external or self-talk (Immenroth et al. 2007).

Cognitive training is not beneficial for novice surgeons only.

The most experienced surgeons report going over the procedures in “their mind’s eye” and consider mental readiness the most important type of preparation, followed by technical and physical readiness (Sanders et al. 2004). It has been shown that in experts not only are the same regions of the brain being engaged during visual imagery as in novices, but also additional regions are recruited suggesting that the pattern of activation moves from frontal parts at the beginning of the process to posterior parts responsible for retrieval of domainspecific knowledge around the final expertise stage (Bilalić et al. 2012). While cognitive specific skills tend to be the focus of novice subjects who are learning specific movements, cognitive general skills are usually used more by experts who link the skills together. Additionally, experts often use motivational and arousal techniques to enhance overall performance by setting specific goals and managing stress and relaxation (Mace et al. 1986, MacIntyre et al. 2002).

In a review by Wallace et al. (2017) limitations of some studies evaluating cognitive training in surgical education have been identified, like small sample sizes, being focused only on short-term effectiveness and lacking psychological mechanisms that underlie their efficacy. However, they concluded that cognitive training is to be integrated as a training tool for surgeons.

Recommendations for surgeons Cognitive training in combination with physical training impacts performance to a greater extent than physical training only. In the absence of specific elective surgical procedures, cognitive training is the only skill acquisition and retaining tool that can and should be used constantly. The literature suggests that mental practice should be brief and focused and should optimally be carried out for 20 minutes in a single session, since extended mental practice may lead to loss of concentration (Corbin 1972, Driskell et al. 1994).

Training by observation For mental practice to be effective, the subject must be familiar with the surgical procedure prior to the imagery session (Driskell et al. 1994). It is not useful for one to mentally practice a complex procedure if the individual is not familiar with the procedure and its flow. In such a situation, external observational training should be performed first.

This could be realized through observation of recordings from trusted sources (i.e., VuMedi, clips from industrial educational websites, YouTube with well-known surgeons performing, etc.), surgical technique guides, and textbooks.

Acta Orthopaedica 2020; 91 (5): 523–526

Observing a recording can be followed only by listening to its audio component and visualizing the procedure rather than watching it.

Subvocal imagery Subvocalization is a natural process of internal speech typically undertaken during reading. In terms of surgical training, visual images are recalled by an internal self-talk (Immenroth et al. 2007). This type of training represents not only a possible next step in external observative training but also a practical way of performing mental imagery training for more experienced surgeons. Novice surgeons especially should pay attention to every detail in a systematic manner (i.e., patient positioning before shoulder arthroscopy, trocar insertion, opening of the water inflow valve, rotating the optics, identify intra-articular structures, etc.), whereas more experienced surgeons can focus more on a specific difficult step (i.e., placing stitches in a cuff tear and placing a screw in the correct position).

Ideomotoric training The ideomotoric training technique implies the movement patterns imagined several times, visualized and verbalized by a highly trained expert to facilitate a procedure execution (Khalmanskikh 2016). The subject visualizes movements from an inner perspective, imagining him- or herself performing the procedure (Immenroth et al. 2007). To include motivational and arousal components of cognitive training, the sensory aspects of the procedures should be included as far as possible. One should recall the atmosphere in the operating theatre such as the team, temperature, light, and even background music and the sounds produced by the anesthetic device. Furthermore, the trainee should try to feel the tactile properties of instruments and tissues (i.e., visualize the polydioxanone suture and how tricky it is to catch it with a tissue grasper in a narrow space, or the challenging passage of the endobutton in cortical ACL fixation). Additionally, potential intraoperative complications and their management (i.e., displacement of implants after using all-internal meniscal devices or suture anchor pull-out after tying a knot in rotator cuff repair) should not be left out.

Conclusion In extraordinary conditions, specific orthopedic surgical skills are prone to decay in the case of nonperformance. Cognitive training has been shown to be an effective tool for performance improvement when performed alone or in combination with motor training under ordinary circumstances. In the case of limited access to actual performance, mental practice seems to be reasonable if not mandatory for both novice and experienced surgeons. Not only would such training prevent surgical skill decay but it would probably also lower the anticipatory anxiety level on returning to the operating theatre.

525

RK: conceptualization, investigation, writing, and original draft preparation.

MV: supervision, writing, and original draft preparation. JK: investigation, writing, and original draft preparation.

Acta thanks Pelle Gustafson, Leif Rune Hedman and Antti P Loosi for help with peer review of this study

Arthur J, Bennett W, Stanush P L, McNelly T L. Factors that influence skill decay and retention: a quantitative review and analysis. Human Performance 1998; 11(1): 57-101.

Bernardi N F, De Buglio M, Trimarchi P D, Chielli A, Bricolo E. Mental practice promotes motor anticipation: evidence from skilled music performance. Front Hum Neurosci 2013; 7.

Bilalić M, Turella L, Campitelli G, Erb M, Grodd W. Expertise modulates the neural basis of context dependent recognition of objects and their relations.

Hum Brain Mapp 2012; 33(11): 2728-40.

Corbin C B. Mental practice. In: WP Morgan, editor. Erogenic aids and muscular performance. San Diego, CA: Academic Press; 1972. p. 93-118.

Driskell J E, Copper C, Moran A. Does mental practice enhance performance?

J Appl Psychol 1994; 79(4): 481-92.

Fadiga L, Buccino G, Craighero L, Fogassi L, Gallese V, Pavesi G. Corticospinal excitability is specifically modulated by motor imagery: a magnetic stimulation study. Neuropsychologia 1999; 37(2): 147-58.

Feltz D, Landers D. The effects of mental practice on motor skill learning and performance: a meta-analysis. J Sports Psychol 2007; 5: 25-57.

Fornette M-P, Bardel M-H, Lefrançois C, Fradin J, Massioui F E, Amalberti R. Cognitive-adaptation training for improving performance and stress management of Air Force pilots. Int J Aviat Psychol 2012; 22(3): 203-23.

Frank C, Land W M, Popp C, Schack T. Mental representation and mental practice: experimental investigation on the functional links between motor memory and motor imagery. PLoS ONE 2014; 9(4): e95175.

Hall J C. Imagery practice and the development of surgical skills. Am J Surg 2002; 184(5): 465-70.

Hamdorf J M, Hall J C. Acquiring surgical skills. Br J Surg 2000; 87(1): 28-37.

Immenroth M, Bürger T, Brenner J, Nagelschmidt M, Eberspächer H, Troidl H. Mental training in surgical education: a randomized controlled trial.

Ann Surg 2007; 245(3): 385-91.

Jacobson E. Electrical measurements of neuromuscular states during mental activities, V: Variation of specific muscles contracting during imagination.

Am J Physiol 1931; 96: 115-21.

Jeannerod M. The representing brain: neural correlates of motor intention and imagery. Behav Brain Sci 1994; 17(2): 187-245.

Khalmanskikh A V. Ideomotor training to improve shooting skills in elite biathlon 2016. Available at: http://www.teoriya.ru/ru/node/5815. Accessed March 29, 2020.

Komesu Y, Urwitz-Lane R, Ozel B, Lukban J, Kahn M, Muir T, Fenner D, Rogers R. Does mental imagery prior to cystoscopy make a difference? A randomized controlled trial. Am J Obstet Gynecol 2009; 201(2): 218.e1-9.

Logishetty K, Gofton W T, Rudran B, Beaulé P E, Gupte C M, Cobb J P. A multicenter randomized controlled trial evaluating the effectiveness of cognitive training for anterior approach total hip arthroplasty. J Bone Joint Surg Am 2020; 102(2): e7.

Mace R D, Carroll D, Eastman C. Effects of stress inoculation training on self-report, behavioral and psychophysiological reactions to abseiling. J Sports Sci 1986; 4(3): 229-36.

MacIntyre T, Moran A, Jennings D J. Is controllability of imagery related to canoe-slalom performance? Percept Mot Skills 2002; 94(3 Pt 2): 1245-50.

Martin K A, Moritz S E, Hall C R. Imagery use in sport: a literature review and applied model. Sport Psychol 1999; 13(3): 245-68.

526

Posner M. Foundations of cognitive science. J Hist Behav Sci 1989; 27(3): 242-7.

Richardson A. Mental practice: a review and discussion. Res Q 1967; 38: 95-107.

Routt E, Mansouri Y, de Moll E H, Bernstein D M, Bernardo S G, Levitt J. Teaching the simple suture to medical students for long-term retention of skill. JAMA Dermatol 2015; 151(7): 761-5.

Sackett R S. The influence of symbolic rehearsal upon the retention of a maze habit. J Gen Psychol 1934; 10: 376-98.

Sanders C W, Sadoski M, Bramson R, Wiprud R, Van Walsum K. Comparing the effects of physical practice and mental imagery rehearsal on learning basic surgical skills by medical students. Am J Obstet Gynecol 2004; 191(5): 1811-14.

Sevdalis N, Moran A, Arora S. Mental imagery and mental practice applications in surgery: state of the art and future directions. In: Lacey S, Lawson R, editors. Multisensory imagery. New York: Springer-Verlag; 2013.

Sonnadara R R, Garbedian S, Safir O, Nousiainen M, Alman B, Ferguson P, Kraemer W, Reznick R. Orthopaedic Boot Camp II: examining the retention rates of an intensive surgical skills course. Surgery 2012; 151(6): 803-7.

Acta Orthopaedica 2020; 91 (5): 523â&#x20AC;&#x201C;526

Sugihara T, Yasunaga M, Matsui H, Ishikawa A, Fujimura T, Fukuhara H, Fushimi K, Homma Y, Kume H. A skill degradation in laparoscopic surgery after a long absence: assessment based on nephrectomy case. Miniinvasive Surg 2018; 2(11).

Surburg P R. Audio, visual, and audio-visual instruction with mental practice in developing the forehand tennis drive. Res Q 1968; 39(3): 728-34.

Tofel-Grehl C, Feldon D. Cognitive task analysis-based training: a metaanalysis of studies. J Cogn Eng Decis Mak 2013; 7(3): 293-304.

Vealey R S, Walter S M. Imagery training for performance enhancement and personal development. In: Williams JM, editor. Applied sport psychology: personal growth to peak performance. 2nd ed. Mountain View, CA: McGraw Hill; 1993. p 200-24.

Wallace L, Raison N, Ghumman F, Moran A, Dasgupta P, Ahmed K. Cognitive training: how can it be adapted for surgical education? Surgeon 2017; 15(4): 231-9.

Wingfield L R, Kulendran M, Chow A, Nehme J, Purkayastha S. Cognitive task analysis: bringing olympic athlete style training to surgical education.

Surg Innov 2015; 22(4): 406-17.

Wisher R, Sabol M A, Ellis J, Ellis K. Staying sharp: retention of military knowledge and skills. For Belvoir, VA: US Army Research Institute; 1999.

Acta Orthopaedica 2020; 91 (5): 527–533

527

Physical child abuse demands increased awareness during health and socioeconomic crises like COVID-19 A review and education material Polina MARTINKEVICH 1,2a, Lise Langeland LARSEN 1,2a, Troels GRÆSHOLT-KNUDSEN 3a, Gitte HESTHAVEN 4, Michel Bach HELLFRITZSCH 2,5, Karin Kastberg PETERSEN 5, Bjarne MØLLER-MADSEN 1,2,6, and Jan Duedal RÖLFING 1,2,6,7 1 Department

of Orthopaedics, Aarhus University Hospital; 2 Danish Paediatric Orthopaedic Research; 3 Research Unit for Mental Public Health, Department of Public Health, Aarhus University; 4 Department of Paediatrics, Aarhus University Hospital; 5 Department of Radiology, Aarhus University Hospital; 6 Department of Clinical Medicine, Aarhus University; 7 Corporate HR, MidtSim, Central Denmark Region, Denmark a Shared first authorship Correspondence: jan.roelfing@clin.au.dk Submitted 2020-05-10. Accepted 2020-06-04.

Background and purpose — Physical abuse of children, i.e., nonaccidental injury (NAI) including abusive head trauma (AHT) is experienced by up to 20% of children; however, only 0.1% are diagnosed. Healthcare professionals issue less than 20% of all reports suspecting NAI to the responsible authorities. Insufficient knowledge concerning NAI may partly explain this low percentage. The risk of NAI is heightened during health and socioeconomic crises such as COVID-19 and thus demands increased awareness. This review provides an overview and educational material on NAI and its clinical presentation.

Methods — We combined a literature review with expert opinions of the senior authors into an educational paper aiming to help clinicians to recognize NAI and act appropriately by referral to multidisciplinary child protection teams and local authorities.

Results — Despite the increased risk of NAI during the current COVID-19 crisis, the number of reports suspecting NAI decreased by 42% during the lockdown of the Danish society. Healthcare professionals filed only 17% of all reports of suspected child abuse in 2016.

Interpretation — The key to recognizing and suspecting NAI upon clinical presentation is to be aware of inconsistencies in the medical history and suspicious findings on physical and paraclinical examination. During health and socioeconomic crises the incidence of NAI is likely to peak.

Recognition of NAI, adequate handling by referral to child protection teams, and reporting to local authorities are of paramount importance to prevent mortality and physical and mental morbidity.

Physical abuse of children, i.e., non-accidental injury (NAI) including abusive head trauma (AHT), is experienced by up to 20% of children; however, only 0.1% are diagnosed with the ICD-10 code: T74.1 physical abuse (Christoffersen 2010, Stoltenborgh et al. 2013, Oldrup et al. 2016).

During the current COVID-19 crisis some European countries have reported an alarming increase in domestic violence by one-third (Delaleu 2020). Likewise, the risk of NAI is heightened during health and socioeconomic crises (Baird 2020, Peterman et al. 2020). Therefore, a Joint Leaders’ statement by the World Health Organization, UNICEF, Save the Children International, and SOS Children’s Villages International among others, highlights the acute risk of violence against children due to COVID-19 and calls for increased awareness (World Health Organization 2020).

The vast majority of NAI is reported by staff working at institutions (daycares, kindergartens, schools), which are temporarily closed during the COVID-19 pandemic. Healthcare professionals issue less than 20% of reports regarding suspected maltreatment to the responsible child protection authorities (Christoffersen 2010, Oldrup et al. 2016). Failure to recognize NAI due to insufficient knowledge among healthcare professionals may partly explain this low percentage (­Villadsen et al. 2015).

Healthcare professionals need to be aware of the increased risk of NAI during COVID-19 and future health and socioeconomic crises in order to act appropriately based on current knowledge of the issue. Only then can they begin to recognize patterns of NAI from the medical history and objective findings, and act appropriately through immediate consultation and referral to multidisciplinary child protection teams, who can clarify the suspicion and ensure child protection.

© 2020 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 10.1080/17453674.2020.1782012

528

Methods Data were synthesized from a literature review, scholarly reports from Danish national authorities, and expert opinions on the clinical presentations of physical abuse, i.e., NAI including AHT, with the aim to provide educational material to encourage peer-to-peer teaching and facilitate a visual reference.

Ethics and conflict of interests This review complies with the Helsinki Declaration. The authors declare no conflicts of interest. Figures and Table are available (see also Supplementary data) for usage or modification according to a Creative Commons license (CC BY-SA 4.0) as long as this paper is attributed. The original, editable files can also be requested from the corresponding author.

Results NAI clinical presentation and reporting Physical abuse of infants and toddlers often presents to healthcare professionals as injury or illness such as seizures or respiratory distress (Table 1). In general, any injury in nonambulatory children should raise suspicion and be discussed with a child protection team in order to plan the appropriate measures.

If the suspicion is raised, immediate referral and hospitalization is required to clarify the potentially lethal suspicion and ensure child protection.

In older children who are able to walk and talk, institutions (daycare, schools, sport clubs) report the majority of suspicions regarding child maltreatment to the local authorities and the police. However, healthcare professionals also need to stay alert regarding this age group.

Red flags in the medical history and clinical presentations of NAI It is important to consider the age and the developmental status of the child when distinguishing accidental injuries from NAI. Any injuries in an infant < 6 months are suspicious (Sugar et al. 1999, Maguire et al. 2005). Also, there is a need to be aware that NAI may present as part of the presentation of poly-victimization, a term referring to having experienced multiple victimizations of different kinds, e.g., sexual abuse, physical abuse, and neglect (Finkelhor et al. 2007). Children with any history of maltreatment should therefore be scrutinized for signs of NAI. The key to increasing the relatively low percentage of healthcare-reported cases is to recognize NAI upon clinical presentation by considering suspicious findings revealed through the medical history, plus physical and paraclinical examinations. Risk indicators concerning the child, caregiver,

Acta Orthopaedica 2020; 91 (5): 527â&#x20AC;&#x201C;533

and environment may be used as a supplement to guide clinical attention, although one should be aware of detection bias when judging already struggling families without sufficient clinical evidence (Widom et al. 2015) (Figure 1, see Supplementary data).

Child protection is paramount and barriers that hamper reporting must be overcome, even if this might stigmatize families until the suspicion has been fully investigated. For instance, the most common orthopedic injury in abused children is a fracture of the femur or humerus (Figure 2, see Supplementary data) (Loder and Feinberg 2007). NAI is diagnosed in about 25% of these fractures, when occurring under the age of 2 years. Hence, in the majority of cases, highlyspecific fractures/injuries are absent and the combination of history and injuries should alarm the physician (Loder and Feinberg 2007, Kemp et al. 2008).

Maltreatment may present as disturbed sleep, unusual anger and irritability, withdrawal, attention and concentration difficulties, repeated and intrusive thoughts, helplessness, insecure relationship with caregivers, and intense emotional distress, particularly when confronted by trauma reminders (Al Odhayani et al. 2013). Also, it is important to recognize that children can exhibit problematic emotional and behavioral reactions long after the abuse or neglect has ended (Sege and AmayaJackson 2017). However, these potentially subtle changes in behavior may be difficult to detect during brief encounters such as emergency department visits.

Skin manifestations Skin manifestations are the most common findings in NAI, and present in up to 90% of victims of physical abuse (Kos and Shwayder 2006). Hence, physical examination of the entire body is mandatory. In general, skin manifestations are most frequently classified into bruises (Figure 2, see Supplementary data), bite marks, and thermal injuries. Oral injuries are often considered as a separate entity, and lack highly specific signs of NAI, thus warranting an odontological opinion.

It can be difficult to distinguish accidental injuries from NAI, but considering the location, pattern, and the age of the child at the time of injury will provide a clue to the suspected type of injury. Equally important is to consider underlying skin diseases, either as a coexisting condition or as a differential diagnosis.

Bruises Bruising is the most common skin manifestation, and it is often inflicted by blunt trauma leaving a pattern marking (Swerdlin et al. 2007). In NAI bruises are often clustered on protected areas such as the Torso (e.g., chest, abdomen, back, buttocks, genitourinary region, and hip), Ears and Neck (TEN) (­Maguire et al. 2005, Pierce et al. 2010). TEN bruising in a child < 4 years or any bruising in a child < 4 months has a sensitivity of 95% and specificity of 84% for prediction of abuse (Pierce et al. 2010). Abdominal bruising is rare, but warrants

Acta Orthopaedica 2020; 91 (5): 527â&#x20AC;&#x201C;533

investigation of the internal organs, as 10% of victims will have intra-abdominal injury (Sheybani et al. 2014).

Bruises from accidents are located on bony prominences and occur more frequently in ambulatory children.

Bite marks All bite marks are suspicious of abuse, and considered dangerous, as they can be a source of infection (Kos and Shwayder 2006).

Consulting a forensic odontologist regarding the bite marks can be valuable, as such professionals possess various techniques to identify the perpetrator. Adult bites have an intercanine distance of > 3 cm, which can distinguish them from child and animal bites (Kos and Shwayder 2006).

Thermal injuries Thermal injuries account for up to 20% of all child abuse cases and data from burn centers suggest that nearly 14% of burn injuries in children are due to abuse, with increased hospitalization time and mortality rate compared with cases of accidental injury (Peck and Priolo-Kapel 2002, Thombs 2008, Royal College of Paediatrics and Child Health 2018). The victims are often < 3 years. Distinction is made between immersion and contact injuries. Often the inflicted thermal injuries leave characteristic patterns that are highly suspicious of child abuse.

Immersion such as scalding with hot tap water is most common. This tends to create symmetrical and distinct lines of demarcation. Frequent mechanisms include holding the childâ&#x20AC;&#x2122;s hands and feet under water (glove-and-stocking pattern, sparing of the palm), or submerging the child in hot water in a flexed position, creating a so-called zebra pattern with sparing of the flexural creases including palms. In general, inflicted burn injuries cover a wider and deeper surface area, and tend to include rather the back, buttocks, perineum, and lower extremities with symmetrical and clear demarcation lines as compared with accidental burn injuries (Thombs 2008). The most frequently reported contact burns include those inflicted by cigarettes, while other instruments include iron, hairdryer, cigarette lighters, oil, flame and chemical burns (Royal College of Paediatrics and Child Health 2018).

Radiological red flags NAI represents a small proportion of all childhood fractures, but all healthcare professionals should be able to recognize the characteristics of fractures resulting from abuse (Figure 3). In infants and toddlers, physical abuse accounts for 12â&#x20AC;&#x201C;20% of all fractures (Leventhal et al. 2008), thus indicating a skeletal survey in children < 24 months (Wootton-Gorges et al. 2017).

Approximately 80% of all fractures caused by NAI occur in children < 18 months, and the proportion of fractures caused by child abuse declines rapidly as ambulatory function develops (Leventhal et al. 2008).

Importantly, the majority of physically abused children present with bruises and a simple fracture, which also can occur

529

after accidents and thus has a lower specificity. Highly specific fractures for NAI and AHT are less common, but should be recognized upon presentation.

Multiple fractures and fractures of different ages should alert the clinician and any concern by the treating healthcare professional based on any anamnestic, objective, or radiological red flags and risk indicators should be considered in consultation with a child protection team.

Common and highly specific fractures Classic metaphyseal lesions (CML, i.e., metaphyseal corner and bucket-handle fractures) are the most specific and common signs of NAI in children < 18 months (Figure 3). They occur most commonly in the lower extremities around the knee and ankle, but are also seen in the upper limbs. Diaphyseal fractures are more common in ambulant children, but less specific (Kemp et al. 2008).

Multiple rib fractures can either be detected incidentally in a child presenting with respiratory compromise or on a skeletal survey on suspicion of NAI. Multiple rib fractures, especially of posterior location, have a high positive predictive value (Barsness et al. 2003, Kemp et al. 2008).

Less common, but highly specific fractures for NAI include epiphyseal separations, fractures of the digits, as well as complex skull fractures, while fractures of the scapula, sternum, and spinous processes are rare.

Intracranial lesion Subdural hematomas have been reported in up to 90% of young infants with AHT. Albeit not pathognomonic for AHT, they do become strongly suggestive of AHT when several SDH of different dates are observed, or the claimed injury mechanism is incompatible with, for instance, a simple fall from less than 1.5 m. Other types of intracranial hemorrhages may also be suggestive of AHT but are also common after accidents. Parenchymal injury is the most significant cause of morbidity and mortality (Choudhary et al. 2018) Abusive head trauma (AHT) AHT accounts for around 50% of severe traumatic head injury cases (Keenan et al. 2003), and it is the leading cause of the death in children < 2 years with a peak before 5 months (Maguire et al. 2011). It is important to remember that no single AHT injury has intrinsic diagnostic value, therefore all findings and history should be considered together. Kelly et al. (2015) found that in children < 2 years the characteristics of AHT included hypoxic-ischemic injury (97%), no history of trauma (90%), no external evidence of impact to the head (90%), subdural hemorrhage (89%), and complex skull fractures with intracranial injury (79%).

In recent years, several clinical prediction or decision rules for AHT have been developed to differentiate between AHT and other reasons for a validated intracranial injury and likewise have a high sensitivity and positive predictive value (PPV)

530

Acta Orthopaedica 2020; 91 (5): 527â&#x20AC;&#x201C;533

retinal hemorrhage

posterior rib fractures

subdural hematoma / hygroma

intra- / extracranial lesions

complex skull fracture

anterior rib fractures

clavicular fracture(s)

skull fracture

classic metaphyseal lesions

2 weeks after with callus

classic metaphyseal lesion with callus

transvers femoral fracture 3-month-old infant

spiral femoral fracture 2-year-old child -

bruises multiple fractures of different ages

abdominal bruises + elevated ALT -> abd. + pelvic CT pelvic fracture suspect abuse often accident

Do not hesitate!

Report your suspicion to the local child protection team it may save a life

Figure 3. Radiological findings associated with NAI. While some findings are highly specific for NAI, the less specific findings are common in both NAI and accidents. Thus, we refrained from subdivision as any of these findings without an appropriate accident should result in involvement of a child protection team.

Acta Orthopaedica 2020; 91 (5): 527–533

for AHT (Pfeiffer et al. 2018). These comprise tools such as the Pittsburgh Infant Brain Injury Score (PIBIS), which aids the decision to perform head CT scans in well-appearing children under 1 year in the emergency department (sensitivity 93%), while the Pediatric Brain Injury Research Network’s (PediBIRN) and Predicting Abusive Head Trauma (PredAHT) have been developed to differentiate between AHT and other reasons for a validated intracranial injury and have a high sensitivity and PPV for AHT. Predicting Abusive Head Trauma (PredAHT) found a positive predictive value (PPV) of 85% for AHT when intracranial hemorrhage in children < 3 years was accompanied by at least 3 of 6 key features (head/neck bruising, seizure, apnea, ribor long-bone fractures, and retinal hemorrhage). PediBIRN’s AHT probability calculator uses 4 clinical parameters: (1) “clinically significant respiratory compromise at the scene of injury, during transport, in the emergency department, or prior to hospital admission”, (2) “bruising of the torso, ear(s), or neck,” (3) “subdural hemorrhage or fluid collection that is bilateral or that involves the interhemispheric space,” (4) “any skull fracture(s) other than an isolated, unilateral, non-diastatic, linear, parietal, skull fracture.” The clinical feasibility of these tools and the perceived disadvantages, i.e., possible over-reliance and false reassurance are going to be investigated (Pfeiffer et al. 2018). Nonetheless, clinicians may benefit from applying these tools or at least being aware of the key clinical findings that these tools take into consideration.

Discussion The initiative for this study was triggered by the alarming surge of domestic violence precipitated by the restrictions imposed to contain the COVID-19 pandemic (Human Rights Watch 2020).

Peterman et al. (2020) identified distinct pathways for how pandemics might increase violence against intimate partners and children: “(1) economic insecurity and poverty-related stress, (2) quarantines and social isolation, (3) disaster and conflict-related unrest and instability, (4) exposure to exploitative relationships due to changing demographics, (5) reduced health service availability and access to first responders, (6) inability to temporarily escape the abuser, and (7) virus-specific sources of violence.”

In attempting to understand the surge in risk, it is widely accepted that stressors overcoming supportive factors comprise the underlying etiology explaining the majority of physical abuse (Belsky 1993). Earlier studies have shown increased risk of physical child abuse as a consequence of society-wide stress exemplified by natural disasters (Keenan et al. 2004, Melissa 2012). Stressors such as poverty (Berger et al. 2011, Doidge et al. 2017), unemployment (Krugman et al. 1986), and parental physical and mental health (Chang et al. 2018) have been shown to increase the risk of physical child abuse—

531

all of which are likely to be affected by the current situation. In addition, the inability to escape the perpetrator due to restrictions of movement might further aggravate the risk (Peterman et al. 2020).

However, the scientific evidence regarding the association of socioeconomic health crises and NAI varies in methods and their applicability to the current situation, and the incidence of NAI is difficult to access objectively during times of crises.

During previous Ebola epidemics an increase in NAI and child abuse was reported in the affected countries (Kostelny et al. 2016). The lessons learned from these crises regarding child protection (in developing countries) were published by UNICEF (UNICEF 2016).

During the first month of the confinement in France, the police received more reports of domestic violence and intervened in 92 child abuse cases; helplines received around 20% more reports of child abuse, via either the victims, relatives, or their network (Innocence En Danger 2020). Conversely, the Canadian regional social services received 75% fewer daily notifications of suspected child abuse (ICI.Radio-Canada.ca 2020). A similar tendency was observed by the Danish authorities, who reported a 42% decline in notifications regarding suspected child abuse immediately after the “lockdown” of society that included daycares and schools (Scheel et al. 2020).

Nearly all of the prevailing tools to mitigate the impact of domestic violence are based on the social environment, assistance, and access to healthcare. Due to confinement measures, closing of schools, kindergartens, and daycares, and the accumulating pressure on healthcare systems these tools are no longer readily available, and thus the risk of NAI will increase.

Notably, the stress exhibited on the healthcare system challenges frontline healthcare professionals’ ability to maintain the usual standards of care. This raises serious concerns that NAI, which is already a hidden but frequent problem, risks being left unnoticed and unsuspected, leading to higher child morbidity and mortality as well as long-term negative developmental consequences (Buckingham and Daniolos 2013).

NAI has been studied as one of several adverse childhood experiences (Felitti et al. 2019), and has been shown to carry increased risks of future ischemic heart disease (Gilbert et al.

2015), cancer, alcoholism, depression, suicide attempts (Felitti 1998), post-traumatic stress disorders (Cross et al. 2018), and social phobia (Scott et al. 2010). In addition to the direct consequences for the victim, child abuse bears high economic costs (Peterson et al. 2018).

The solution to the hidden problem of NAI requires a multidisciplinary, multilateral, and multistep approach (prevention, detection, and intervention). The complexity and universal pertinence of the problem is so extensive that it has been listed on the EU’s 2030 Agenda for the 17 Sustainable Development Goals proposed by the UN (UN 2015).

Medical professionals are often presented for the victims of NAI in an emergency setting, leaving a narrow window for detecting and facilitating appropriate follow-up of NAI.

532

A global meta-analysis (Stoltenborgh et al. 2013) concludes that physical abuse is 75 times higher than suggested by official reports. Consulting different hospitals after trauma instead of making recurrent visits to the same hospital, and frequent changes of primary healthcare provider (Friedlaender et al.

2005), as well as any disabilities of the child, might either conceal or delay the diagnosis (Nowak 2015). Furthermore, numerous barriers to reporting NAI exist, which partly explain the underreporting: insufficient knowledge concerning NAI and thus failure to recognize it, and delayed or inappropriate decision-making during the diagnostic process.

Many definitions of NAI exist in the legal and scientific literature, but there is no consensus on an absolute definition.

For further information on terminology, Medical Subject Headings and ICD-10 codes please refer to Supplementary data, Table 1. We advocate the use of ICD-10 codes: physical abuse in conjunction with non-accidental injury and abusive head trauma (Choudhary et al. 2018).

In conclusion, healthcare professionals should be aware of the heightened risk of physical abuse of children during health and socioeconomic crises. Only if healthcare professionals are familiar with NAI—through the anamnestic, objective, and radiological red flags and risk indicators—can they recognize NAI and act appropriately by consulting with the multidisciplinary child protection team who can verify or reject this devastating diagnosis. In addition, any child considered in immediate danger of further harm, and all children less than 2 years old, should be hospitalized to provide shelter and enable further investigations. We hope that this review and its illustrations can help frontline healthcare staff to achieve this aim and potentially save lives or minimize the long-term effects of adverse childhood experiences.

Supplementary data Supplementary Figures: “All figures from the current manuscript for free usage and modification,” and Supplementary Table: “Terminology, Medical Subject Heading of NAI” are available as supplementary data in the online version of this article, http://dx.doi.org/10.1080/17453674.2020.1782012

PM, LLL, and TGK contributed equally to the study. They conducted the literature search and drafted the first version of the manuscript. GH contributed with clinical cases, photos, and critical revision of the manuscript.

MBH and KKP provided radiological images and radiological red flags.

BMM contributed in drafting the manuscript and critical revision. JDR contributed to all parts of the study and layout of the fig The authors would like to express their gratitude to Rikke Holm Bramsen, Børnehus Midt, Muncipality of Aarhus, Denmark and Ole Ingemann Hansen, Department of Forensic Medicine, Aarhus University and Hans Ulrik Møller, Department of Ophtalmology, Aarhus University Hospital for their invaluable contributions.

Acta thanks Nanni Allington and Hanne Hedin for help with peer review of this study.

Acta Orthopaedica 2020; 91 (5): 527–533

Baird E. Non-accidental injury in children in the time of Covid-19 pandemic.

Loondon: British Orthopaedic Association; 2020.

Barsness K A, Cha E, Bensard D D, Clakins C M, Partrick D A, Karrer F M, Strain J D. The positive predictive value of rib fractures as an indicator of nonaccidental trauma in children. J Trauma 2003; 54(6): 1107-10. doi: 10.1097/01.TA.0000068992.01030.

Belsky J. Etiology of child maltreatment: a developmental-ecological analysis. Psychol Bull 1993; 114(3): 413-34. doi: 10.1037/0033-2909.114.3.413.

Berger R P, Fromkin J B, Sturz H, Makoroff K, Scribano P V, Feldman K, Tu L C, Fabio A. Abusive head trauma during a time of increased unemployment: a multicenter analysis. Pediatrics 2011; 128(4): 637-43. doi: 10.1542/peds.2010-2185.

Buckingham E, Daniolos, P. Longitudinal outcomes for victims of child abuse. Curr Psychiatry Rep 2013; 15(2): 342.

Berger R P, Fromkin J B, Sturz H, Makoroff K, Scribano P V, Feldman K, Tu L C, Fabio A. Multiple factors associated with child abuse perpetration: a nationwide population-based retrospective study, J Interpersonal Violence 2018; 088626051880510. doi: 10.1177/0886260518805100.

Choudhary A K, Servaes S, Slovis T L, Palusci V J, Hedlund G L, Narang S K, Moreno J A, Dias M S, Christian C W, Nelson Jr M D, Silvera V M, Palasis S, Raissaki M, Rossi A, Offiah A C. Consensus statement on abusive head trauma in infants and young children. Pediatr Radiol 2018; 48(8): 1048-65. doi: 10.1007/s00247-018-4149-1.

Christoffersen M N. Børnemishandling i hjemmet; 2010. SFI - Det Nationale Forskningscenter for Velfærd. Available at: https://www.vive.dk/da/udgivelser/boernemishandling-i-hjemmet-4360/. May 1, 2020.

Cross D, Vance L A, Kim Y J, Ruchard A L, Fox N, Jovanovic T, Bradley B. Trauma exposure, PTSD, and parenting in a community sample of lowincome, predominantly African American mothers and children. Psychol Trauma 2018; 10(3): 327-35. doi: 10.1037/tra0000264.

Delaleu N. COVID-19: Stopping the rise in domestic violence during lockdown. European Parliament; 2020. Available at: https://www.europarl. europa.eu/news/en/press-room/20200406IPR76610/covid-19-stoppingthe-rise-in-domestic-violence-during-lockdown (accessed April 7, 2020).

Doidge J C, Higgins D J, Delfabbro P, Edwards B, Vassallo S, Toumbourou J W, Segal L. Economic predictors of child maltreatment in an Australian population-based birth cohort, Children and Youth Services Review 2017. doi: 10.1016/j.childyouth.2016.10.012.

Felitti V J, Anda R F, Nordenberg D, Williamson D F, Spitz A M, Edwards V, Koss M P, Marks J S. Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults: the Adverse Childhood Experience (ACE) study. Am J Prev Med 1998; 14(4): P245258. doi:10.1016/S0749-3797(98)00017-8.

Finkelhor D, Ormrod R K, Turner H A. Poly-victimization: a neglected component in child victimization. Child Abuse and Neglect 2007; 31: 7-26. doi: 10.1016/j.chiabu.2006.06.008.

Friedlaender E Y, Rubin D M, Alpern E R, Mandell D S, Christian C W, Alessandrini E A. Patterns of health care use that may identify young children who are at risk for maltreatment. Paediatrics 2005; 116(6): 1303-8. doi: 10.1542/peds.2004-1988 Gilbert L K, Breiding M J, Merrick M T, Thompson W W, Ford D C, Dhingra S S, Parks S E. Childhood adversity and adult chronic disease: an update from ten states and the District of Columbia, 2010. Am J Prev Med 2015 48(3): 345-9. doi: 10.1016/j.amepre.2014.09.006.

Human Rights Watch.) COVID-19 and children’s rights. Human Rights Watch April 9, 2020.

ICI.Radio-Canada.ca, Z. S. COVID-19: la baisse des signalements de maltraitance d’enfant ne reflète pas la réalité. Coronavirus. Ontario: RadioCanada.ca; 2020.

Innocence En Danger. Conséquences du confinement sur les violences familiales. 2020. Available at: https://innocenceendanger.org/2020/04/27/consequences-du-confinement-sur-les-violences-familiales/ (accessed May 9, 2020).

Keenan H T, Runyan D K, Marshall S W, Nocera M A, Merten D F, Sinal S H. A population-based study of inflicted traumatic brain injury in young children. JAMA 2003; 290(5): 621-6. doi: 10.1001/jama.290.5.621.

Acta Orthopaedica 2020; 91 (5): 527–533

Keenan H T, Marshall S W, Nocera M A, Runyan D K. Increased incidence of inflicted traumatic brain injury in children after a natural disaster. Am J Prev Med 2004; 26(3): 189-93. doi: 10.1016/j.amepre.2003.10.023.

Kelly P, John S, Vincent A L, Reed P. Abusive head trauma and accidental head injury: A 20-year comparative study of referrals to a hospital child protection team, Arch Dis Child 2015; 100(12): 1123-30. doi: 10.1136/ archdischild-2014-306960.

Kemp A M, Dunstan F, Harrison S, Morris S, Mann M, Rolfe K, Datta S, Thomas D P, Sibert J R, Maguire S. Patterns of skeletal fractures in child abuse: systematic review. BMJ 2008; 337: a1518. doi: 10.1136/bmj.a1518.

Kos L, Shwayder T. Cutaneous manifestations of child abuse. Pediatr Dermatol 2006; 23(4): 311-20. doi: 10.1111/j.1525-1470.2006.00266.x.

Kemp A M, Dunstan F, Harrison S, Morris S, Mann M, Rolfe K, Datta S, Thomas D P, Sibert J R, Maguire S. Worse than the war: an ethnographic study of the impact of the Ebola crisis on life, sex, teenage pregnancy, and a community-driven intervention in rural Sierra Leone. Resource Cengtre, Save the Children; 2016.

Krugman R D, Lenherr M, Betz L, Fryer G E. The relationship between unemployment and physical abuse of children. Child Abuse & Neglect 1986; 10(3): 415-18. doi: 10.1016/0145-2134(86)90018-9.

Leventhal J M, Martin K, Asnes A. Incidence of fractures attributable to abuse in young hospitalized children: results from analysis of a United States database. Pediatrics 2008; 122(3): 599-604..

Loder R T, Feinberg J R. Orthopaedic injuries in children with nonaccidental trauma: demographics and incidence from the 2000 Kids’ Inpatient Database. J Pediatr Orthop 2007; 27(4): 421-6. doi: 10.1097/01. bpb.0000271328.79481.07.

Maguire S, Mann M K, Sibert J, Kemp A. Are there patterns of bruising in childhood which are diagnostic or suggestive of abuse? A systematic review. Arch Dis Child 2005; 90(2): 182-6. doi: 10.1136/adc.2003.044065.

Maguire S A, Kemp A M, Lumb R C, Farewell D M. Estimating the probability of abusive head trauma: a pooled analysis. Pediatrics 2011; 128(3): e550-64. doi: 10.1542/peds.2010-2949.

Melissa K. The incidence of childhood injury following an “Inland Tsunami”: the experience of Toowoomba. Injury Prevention 2012(Suppl. 1): A104-5. doi: 10.1136/injuryprev-2012-040590d.27.

Nowak C B. Recognition and prevention of child abuse in the child with disability. Am J Med Genet C Semin Med Genet 2015; 169(4): 293-301.

Al Odhayani A, Watson W J. Behavioural consequences of child abuse. Can Fam Physician 2013; 59(8): 831-6.

Oldrup H, Christoffersen M N, Kristiansen I L, Østergaard S V. Vold og seksuelle overgreb mod børn og unge i Danmark 2016. 2016. https://pure.vive. dk/ws/files/748603/1616_Vold_og_seksuelle_overgreb_mod_boern_og_ unge.pdf Accessed: 28 April 2020.

Peck M D, Priolo-Kapel D. Child abuse by burning: a review of the literature and an algorithm for medical investigations. J Trauma 2002; 53(5): 101322.doi: 10.1097/00005373-200211000-00036.

Peterman A, Potts A, O’Donnel M, Thomson K, Shah N, Oertelt-Prigione S, van Gelder N. Pandemics and violence against women and children. Center for Global Development, Washington DC, USA. 2020.

Peterson C, Florence C, Klevens J. The economic burden of child maltreatment in the United States, 2015. Child Abuse & Neglect 2018; 86: 178-83. doi: 10.1016/J.CHIABU.2018.09.018.

Pfeiffer H, Crowe L, Kemp A M, Cowley L E, Smith A S, Babl F E. Clinical prediction rules for abusive head trauma: A systematic review, Arch Dis Child 2018; 103(8): 776-83. doi: 10.1136/archdischild-2017-313748.

533

Pierce M C, Kaczor K, Aldridge S, O’Flynn J, Lorenz D J. Bruising characteristics discriminating physical child abuse from accidental trauma. Pediatrics 2010; 125(1): 67-74. doi: 10.1542/peds.2008-3632.

Royal College of Paediatrics and Child Health. (2018) Child protection evidence: systematic review on fractures (February 2017). London: RCPCH; 2018. p. 1-67.

Scheel A F, Lytken L H, Sander N, Clante C. Coronakrisen har efterladt udsatte børn alene: Markant fald i underretninger. 2020. Available at: https://www.dr.dk/nyheder/indland/coronakrisen-har-efterladt-udsatteboern-alene-markant-fald-i-underretninger. Assessed: April 25, 2020 Scott K M, Smith D R, Ellis P M. Prospectively ascertained child maltreatment and its association with DSM-IV mental disorders in young adults.

Arch Gen Psychiatry 2010; 67(7): 712-19. doi: 10.1001/archgenpsychiatry.2010.71.

Sege R D, Amaya-Jackson L. Clinical considerations related to the behavioral manifestations of child maltreatment. Pediatrics 2017; 139(4): e20170100. doi: 10.1542/peds.2017-0100.

Sheybani E F, Gonzales-Araiza G, Kousari Y M, Hulett R L, Menias C O. Pediatric nonaccidental abdominal trauma: what the radiologist should know. RadioGraphics 2014; 34(1): 139-53. doi: 10.1148/rg.341135013.

Stoltenborgh M, Bakermans-Kranenburg M J, van Ijzendoorn M H, Alink L R A. Cultural-geographical differences in the occurrence of child physical abuse? A meta-analysis of global prevalence. Int J Psychol 2013; 48(2): 81-94.

Sugar N F, Taylor J A, Feldman K W, Puget Sound Pediatric Research Network Bruises in infants and toddlers: those who don’t cruise rarely bruise.

Arch Pediatr Adolesc Med 1999; 153(4): 399-403. doi: 10.1001/archpedi.153.4.399.

Swerdlin A, Berkowitz C, Craft N. Cutaneous signs of child abuse, J Am Acad Dermatol 2007; 57(3): 371-92. doi: 10.1016/j.jaad.2007.06.001.

Thombs B D. Patient and injury characteristics, mortality risk, and length of stay related to child abuse by burning evidence from a national sample of 15,802 pediatric admissions. Ann Surg 2008; 247(3): 519-23. doi: 10.1097/ SLA.0b013e31815b4480.

UN. Transforming our world: the 2030 Agenda for sustainable development.

Sustainable Development Knowledge Platform; 2015. Available at: https:// sustainabledevelopment.un.org/post2015/transformingourworld (accessed May 9, 2020).

UNICEF. Care and protection of children in the West African Ebola virus disease epidemic: lessons learned for future emergencies. New York: UNICEF; 2016. p. 73.

Villadsen J K, Bersang A B, Thorninger R, Møller-Madsen B, Rahbek O. Skadestuelægers kendskab til battered child syndrome er mangelfuldt.

Ugeskrift for Laeger 2015; 177(8): 749-51.

Widom C S, Czaja S J, DuMont K A. Intergenerational transmission of child abuse and neglect: Real or detection bias? Science 2015; 347(6229): 14805. doi: 10.1126/science.1259917.

Wootton-Gorges S L, Soares B P, Alazraki A L, Anupindi S A, Blount J P, Booth T N, Dempsey M E, Falcone Jr R A, Hayes L L, Kulkarni A V, Partap S, Rigsby C K, Ryan M E, Safdar N M, Trout A T, Widmann R F, Karmazyn B K, Palasis S (Expert Panel on Pediatric Imaging). ACR appropriateness criteria® suspected physical abuse—child. J Am Coll Radiology 2017; 14(5S): S338-49. doi: 10.1016/j.jacr.2017.01.036.

World Health Organization. Joint Leaders statement. Violence against children: a hidden crisis of the COVID-19 pandemic. Geneva: WHO; 2020. p. 2.

534

Acta Orthopaedica 2020; 91 (5): 534–537

Virus transmission during orthopedic surgery on patients with COVID-19 – a brief narrative review Trude BASSO 1, Håvard DALE 2,3, Håkon LANGVATN 1,3, Greger LØNNE 4,5, Inge SKRÅMM 7, Marianne WESTBERG 6, Tina S WIK 1,5, and Eivind WITSØ 1 1 Department of Orthopedics, St Olavs University Hospital; 2 Department of Orthopedics, Haukeland University Hospital; 3 Faculty of Medicine, of Bergen; 4 Department of Orthopedics, Innlandet Hospital Trust; 5 Faculty of Medicine and Health Sciences, Norwegian University of Science Technology; 6 Department of Orthopedics, Oslo University Hospital; 7 Clinic of Orthopedic surgery, Akershus University Hospital, Norway

University and

Correspondence: trude.basso@stolav.no Submitted 2020-04-16. Accepted 2020-04-29.

Background and purpose — COVID-19 is among the most impactful pandemics that the society has experienced.

Orthopedic surgery involves procedures generating droplets and aerosols and there is concern amongst surgeons that otherwise rational precautionary principles are being set aside due to lack of scientific evidence and a shortage of personal protective equipment (PPE). This narrative review attempts to translate relevant knowledge into practical recommendations for healthcare workers involved in orthopedic surgery on patients with known or suspected COVID-19.

Patients and methods — We unsystematically searched in PubMed, reference lists, and the WHO’s web page for relevant publications concerning problems associated with the PPE used in perioperative practice when a patient is COVID19 positive or suspected to be. A specific search for literature regarding COVID-19 was extended to include publications from the SARS epidemic in 2002/3.

Results — Transmission of infectious viruses from patient to surgeon during surgery is possible, but does not appear to be a considerable problem in clinical practice.

Seal-leakage is a problem with surgical masks. Due to the lack of studies and reports, the possibility of transmission of SARS-CoV-2 from patient to surgeon during droplet- and aerosol-generating procedures is unknown.

Interpretation — Surgical masks should be used only in combination with a widely covering visor and when a respirator (N95, FFP2, P3) is not made available. Furthermore, basic measures to reduce shedding of droplets and aerosols during surgery and correct and consistent use of personal protective equipment is important.

Due to the COVID-19 pandemic, elective orthopedic procedures are currently, to a great extent, postponed (CDC 2020, ECDC 2020). However, patients with and without COVID-19, with cancer, infections in bones, joints, and soft tissues, critical ischemia, open and unstable fractures, and other urgent diagnoses will still be in need of orthopedic surgery.

It is widely accepted that healthcare workers (HCWs) performing procedures involving the respiratory tract face a high risk of contracting Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2). This is reflected in the WHO’s recommendation (WHO 2020b) for the highest standard of personal protective equipment (PPE) in these cases. The risk for HCWs to contract SARS-CoV-2 during orthopedic surgery, or most other surgeries for that matter, has not been studied and is consequently less clear.

Orthopedic surgery often involves the use of high-speed saws, power drills, pulsed lavage, suction, and electrocauterization. Shedding of droplets from the wound is reflected in the extensive use of protective eyewear, such as goggles and visor, in everyday practice. Concern amongst surgeons and other HCWs that otherwise rational precautionary principles are being set aside due to lack of scientific evidence and a shortage of PPE is obvious on social media platforms and amongst colleagues.

This short review is an attempt to translate relevant knowledge into practical recommendations for HCWs involved in orthopedic surgery on patients with known or suspected COVID-19.

Transmission of SARS-CoV-2 Virus shedding through droplets that rapidly fall to the ground requires a different PPE approach to prevent transmission than

© 2020 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 10.1080/17453674.2020.1764234

Acta Orthopaedica 2020; 91 (5): 534â&#x20AC;&#x201C;537

infectious aerosols that can remain airborne for a longer period.

According to the WHO, the transmission of SARS-CoV-2 mainly occurs through droplets (WHO 2020a). The organization concludes that transfer through aerosols is unlikely except during specific aerosol-generating procedures. As a response, Nature problematized existing controversies regarding possible air-transmission in a news story, shedding light on the complexity of the subject and that several uncertainties cannot be clarified in a long time (Anon 2020).

SARS-CoV-2 RNA (vRNA) has been found in aerosols (Ong et al. 2020, Santarpia et al. 2020). However, it is not clear whether aerosols contain infectious SARS-CoV-2, or enough viable virus to transmit the disease. We are not aware of any studies that have investigated aerosols produced during surgery on patients with SARS-CoV-2 viremia or disseminated disease.

Presence of SARS-CoV-2 in the musculoskeletal system SARS-CoV-2 was first identified in December 2019 and, consequently, is not fully understood. Like SARS-CoV, the coronavirus causing the 2002/3 SARS epidemic, SARS-CoV-2 binds to ACE2 receptors on human cells (Shang et al. 2020).

ACE2 receptors are present on cells in the lungs and small intestines, but also on cells in a variety of other tissues, including veins, arteries, and skeletal muscle, throughout the body (Hamming et al. 2004, Riquelme et al. 2014).

Most current tests used to confirm the presence of SARSCoV-2 use PCR technology to detect vRNA. A vRNA test will return positive with viable virus, but also with non-viable virus and virus debris. Only a viable virus can infect new individuals. To our knowledge infectious virus has been found only in respiratory tract tissue and in 2 fecal samples from 8 patients (Wang et al. 2020, Wolfel et al. 2020). We are not aware of any published studies that have aimed to find viable SARS-CoV-2 in blood, bone, bone marrow, or skeletal muscle in COVID-19 patients. An in vitro study showed replication of SARS-CoV-2 within blood-vessel organoids (Monteil et al. 2020).

Several studies have identified vRNA in blood and serum (Shang et al. 2020, Wang et al. 2020, Young et al. 2020). vRNA has been found in both the severely ill and in patients with mild symptoms. Amongst 15 patients with multiple site samples, vRNA in blood was detected in 6 patients with negative swabs from the upper respiratory system (Zhang et al. 2020).

An autopsy study including 8 deceased patients from the SARS epidemic in 2002/3 showed widespread virus dissemination in immune cells of the blood, spleen, and lymph nodes and in cells of the respiratory tract, renal tubules, intestines, and brain (Gu et al. 2005). Virus was not found within skeletal muscle cells.

Aerosol formation during orthopedic surgery High-speed saws, power-drills, pulsed lavage, suction, and electrocauterization are all droplet- and aerosol-generating

535

procedures. Infected fluids, such as blood and irrigation fluid, can aerosolize during surgery and shed bacteria and viruses and have the potential to transmit disease (Heinsohn and Jewett 1993).

During experimental set-ups in vivo, infectious HIV-1 particles have been found in aerosols produced using an oscillating saw on a known infected individual (Johnson and Robinson 1991) and aerosols formed in laser fume transmitted disease in a bovine Papillomavirus model (Garden et al. 2002). Both the use of a high-speed cutter and pulsed lavage showed shedding of Staphylococcus aureus several meters from the operating field. The shedding was reduced, but not eliminated, when a drape was used as an overlying protective barrier (Nogler et al.

2001, Putzer et al. 2017). Literature is sparse, and we could not find evidence of disease transmission from patient to surgeon through aerosolized virus-infected fluids from orthopedic-like procedures in clinical practice.

Surgical masks and particulate respirators (N95, FFP2/P3) Originally, surgical masks were made to protect the patient from infectious pathogens in HCWs. Respirators, the somewhat confusing technical term for face masks with the standards N95, FFP2/P3, were designed to protect the user from airborne particles.

The WHOâ&#x20AC;&#x2122;s recommendations regarding PPE do not discuss aerosol-generating surgical procedures on infected patients (WHO 2020b). A review from the Norwegian Institute of Public Health (2020) concludes that evidence regarding the risk of aerosol transmission through aerosol-generating procedures, other than those directly or indirectly affecting the airways, is low. Respirators (N95, FFP2/P3), are consequently not recommended for open surgeries elsewhere in the body (WHO 2020b, FHI 2020).

A randomized controlled clinical trial including 446 nurses concluded non-inferiority of surgical masks when compared with N95 respirators in preventing transmission of influenza and other respiratory viruses (coronavirus included) from patients to HCWs (Loeb et al. 2009). This finding from clinical practice has been supported by 3 later meta-analyses including approximately 9,000 subjects (Smith et al. 2016, Bartoszko et al. 2020, Long et al. 2020). All the included studies were performed in non-aerosol-generating settings. N95 respirators were found to be superior to surgical masks under laboratory settings regarding filter penetration and face-seal leakage (Smith et al. 2016). Experience from the SARS epidemic stresses the importance of correct and consistent use of PPE and that this might be just as important as type of airway protection to prevent nosocomial disease transmission (Seto et al. 2003, Loeb et al. 2004). It must still be emphasized that data from the SARS outbreak in Toronto showed a trend in favor of N95 respirators over surgical masks for HCWs involved in respiratory tract procedures. The difference did not reach statistical significance, but the number of nurses

536

included was low for this sub-analysis (n = 20, 3 infected) (Loeb et al. 2004).

Can virus transmission occur during orthopedic surgery on patients with Covid-19?

COVID-19 is a new, harmful and rapidly spreading disease that first occurred less than 4 months ago, i.e., in December 2019. The knowledge regarding the potential of SARS-CoV-2, and the previous SARS-CoV, to spread via droplets and aerosols produced during surgery is very sparse.

Some patients with both mild and severe COVID-19 have vRNA in their blood indicative of viremia. Infectious disease transmission through both droplets and aerosols produced during orthopedic surgery is possible. In the case of SARSCoV-2, the risk naturally depends on the virus’s capability of transmission through tissues other than respiratory tract tissues and feces. Results from possible investigations of such a capability have not been published at the time of writing (April 2020).

Conclusions The presence of viable virus in blood, bone marrow, or soft tissues has to our knowledge not yet been studied and should be addressed as soon as possible during this pandemic.

Our following recommendations are based on available literature at the time of writing. The absence of evidence is, in situations like this, not a sufficient reason to restrict HCWs’ access to the PPE necessary to protect against transmission of a virus with the potential to cause a severe infection.

The inferiority of surgical masks to respirators (P95, FFP2/ P3) during surgery on a COVID-19 patient is not so clear.

Seal-leakage due to poor fit is a problem with surgical masks.

Hence, surgical masks should only be used in combination with a widely covering visor and if a respirator (P95, FFP2, P3) is not made available. Furthermore, basic protective measures should be implemented, such as avoiding pulsed lavage and the use of power tools when possible, and covering the surgical field with a waterproof drape while performing droplet- and aerosol-generating procedures. Finally, correct and consistent use of PPE is of outmost importance for orthopedic surgeons in general, and, based on current literature, probably more important than the use of respirators (P95, FFP2, P3) per se.

TB wrote the first draft in collaboration with EW. TB coordinated discussions and preparation of the final manuscript. TB, HD, GL, MW, TSW, and EW evaluated scientific papers and public recommendations, and all contributed substantially to the contents of the article.

The authors would like to thank Andreas Radtke (MD, PhD), the senior Infection Control Physician at St Olavs University Hospital, for highly appreciated feedback during the process.

Acta thanks Yan Li and Javad Parvizi for help with peer review of this study.

Acta Orthopaedica 2020; 91 (5): 534–537

Anon. Is the coronavirus airborne? Experts can’t agree. https://www.nature. com/articles/d41586-020-00974-w: Nature; 2020.

Bartoszko J J, Farooqi M A M, Alhazzani W, Loeb M. Medical masks vs N95 respirators for preventing COVID-19 in health care workers: a systematic review and meta-analysis of randomized trials. Influenza Other Respir Viruses 2020. doi: 10.1111/irv.12745.

Centers for Disease Control and Prevention. Strategies to optimize the supply of PPE and equipment. https://www.cdc.gov/coronavirus/2019-ncov/hcp/ ppe-strategy/index.html; 2020.

European Centre for Disease Prevention and Control. Novel coronavirus disease 2019 (COVID-19) pandemic: increased transmission in the EU/EEA and the UK—sixth update. https://www.ecdc.europa.eu/sites/default/files/ documents/RRA-sixth-update-Outbreak-of-novel-coronavirus-disease2019-COVID-19.pdf; 2020.

FHI. Aerosol generating procedures in health care, and COVID-19. Rapid review. https://www.fhi.no/globalassets/dokumenterfiler/rapporter/2020/ aerosol-generating-procedures-in-health-care-and-covid19-rapport-2020. pdf; 2020.

Garden J M, O’Banion M K, Bakus A D, Olson C. Viral disease transmitted by laser-generated plume (aerosol). Arch Dermatol 2002; 138(10): 1303-7. doi: 10.1001/archderm.138.10.1303.

Gu J, Gong E, Zhang B, Zheng J, Gao Z, Zhong Y, Zou W, Zhan J, Wang S, Xie Z, Zhuang H, Wu B, Zhong H, Shao H, Fang W, Gao D, Pei F, Li X, He Z, Xu D, Shi X, Anderson V M, Leong AS. Multiple organ infection and the pathogenesis of SARS. J Exp Med 2005; 202(3): 415-24. doi: 10.1084/ jem.20050828.

Hamming I, Timens W, Bulthuis M L, Lely A T, Navis G, van Goor H. Tissue distribution of ACE2 protein, the functional receptor for SARS coronavirus: a first step in understanding SARS pathogenesis. J Pathol 2004; 203(2): 631-7. doi: 10.1002/path.1570.

Heinsohn P, Jewett D L. Exposure to blood-containing aerosols in the operating room: a preliminary study. Am Ind Hyg Assoc J 1993; 54(8): 446-53. doi: 10.1080/15298669391354946.

Johnson G K, Robinson W S. Human immunodeficiency virus-1 (HIV-1) in the vapors of surgical power instruments. J Med Virol 1991; 33(1): 47-50. doi: 10.1002/jmv.1890330110.

Loeb M, McGeer A, Henry B, Ofner M, Rose D, Hlywka T, Levie J, McQueen J, Smith S, Moss L, Smith A, Green K, Walter S D. SARS among critical care nurses, Toronto. Emerg Infect Dis 2004; 10(2): 251-5. doi: 10.3201/ eid1002.030838.

Loeb M, Dafoe N, Mahony J, John M, Sarabia A, Glavin V, Webby R, Smieja M, Earn D J, Chong S, Webb A, Walter S D. Surgical mask vs N95 respirator for preventing influenza among health care workers: a randomized trial.

JAMA 2009; 302(17): 1865-71. doi: 10.1001/jama.2009.1466.

Long Y, Hu T, Liu L, Chen R, Guo Q, Yang L, Cheng Y, Huang J, Du L. Effectiveness of N95 respirators versus surgical masks against influenza: a systematic review and meta-analysis. J Evid Based Med 2020. doi: 10.1111/ jebm.12381.

Monteil V, Kwon H, Prado P, Hagelkrüys A, Wimmer R A, Stahl M, Leopoldi A, Garreta E, Hurtado Del Pozo C, Prosper F, Romero J P, Wirnsberger G, Zhang H, Slutsky A S, Conder R, Montserrat N, Mirazimi A, Penninger J M. Inhibition of SARS-CoV-2 infections in engineered human tissues using clinical-grade soluble human ACE2. Cell 2020; Apr 17. doi: 10.1016/j.cell.2020.04.004.

Nogler M, Lass-Flörl C, Ogon M, Mayr E, Bach C, Wimmer C. Environmental and body contamination through aerosols produced by highspeed cutters in lumbar spine surgery. Spine 2001; 26(19): 2156-9. doi: 10.1097/00007632-200110010-00023.

Norwegian Institute of Public Health. Råd til helsepersonell i spesialisthelsetjenesten om covid-19. https://www.fhi.no/nettpub/coronavirus/ helsepersonell/tiltak-i-spesialisthelsetjenesten-ved-mistenkt-og-bekreftetsmitte-med-nytt/#smitteregime-ogbeskyttelsesutstyr. Norwegian Institute of Public Health; 2020.

Ong S W X, Tan Y K, Chia P Y, Lee T H, Ng O T, Wong M S Y, Marimuthu K. Air, surface environmental, and personal protective equipment contamination by Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) from a symptomatic patient. JAMA 2020. doi: 10.1001/jama.2020.3227.

Acta Orthopaedica 2020; 91 (5): 534–537

Putzer D, Lechner R, Coraca-Huber D, Mayr A, Nogler M, Thaler M. The extent of environmental and body contamination through aerosols by hydro-surgical debridement in the lumbar spine. Arch Orthop Trauma Surg 2017; 137(6): 743-7. doi: 10.1007/s00402-017-2668-0.

Riquelme C, Acuña M J, Torrejón J, Rebolledo D, Cabrera D, Santos R A, Brandan E. ACE2 is augmented in dystrophic skeletal muscle and plays a role in decreasing associated fibrosis. PloS One 2014; 9(4): e93449. doi: 10.1371/journal.pone.0093449.

Santarpia J L, Rivera D N, Herrera V, Morwitzer M J, Creager H, Santarpia G W, Crown K K, Brett-Major D, Schnaubelt E, Broadhurst M J, Lawler J V, Reid S P, Lowe J J. Transmission potential of SARS-CoV-2 in viral shedding observed at the University of Nebraska Medical Center. bioRxiv 2020; 2020.03.23.20039446. doi: 10.1101/2020.03.23.20039446 %J medRxiv.

Seto W H, Tsang D, Yung R W, Ching T Y, Ng T K, Ho M, Ho L M, Peiris J S. Effectiveness of precautions against droplets and contact in prevention of nosocomial transmission of severe acute respiratory syndrome (SARS).

Lancet 2003; 361(9368): 1519-20. doi: 10.1016/s0140-6736(03)13168-6.

Shang J, Ye G, Shi K, Wan Y, Luo C, Aihara H, Geng Q, Auerbach A, Li F. Structural basis of receptor recognition by SARS-CoV-2. Nature 2020. doi: 10.1038/s41586-020-2179-y.

Smith J D, MacDougall C C, Johnstone J, Copes R A, Schwartz B, Garber G E. Effectiveness of N95 respirators versus surgical masks in protecting health care workers from acute respiratory infection: a systematic review and meta-analysis. CMAJ 2016; 188(8): 567-74. doi: 10.1503/cmaj.150835.

Wang W, Xu Y, Gao R, Lu R, Han K, Wu G, Tan W. Detection of SARS-

537

CoV-2 in different types of clinical specimens. JAMA 2020. doi: 10.1001/ jama.2020.3786.

WHO. Coronavirus disease 2019 (COVID-19): Situation report—66. https://www.who.int/docs/default-source/coronaviruse/situationreports/20200326-sitrep-66-covid-19.pdf?sfvrsn=81b94e61_2; 2020a.

WHO. Rational use of personal protective equipment (PPE) for coronavirus disease (COVID-19). Interim guidance. https://www.who.int/publicationsdetail/rational-use-of-personal-protective-equipment-for-coronavirus-disease-(covid-19)-and-considerations-during-severe-shortages; 2020b.

Wolfel R, Corman V M, Guggemos W, Seilmaier M, Zange S, Muller M A, Niemeyer D, Jones T C, Vollmar P, Rothe C, Hoelscher M, Bleicker T, Brunink S, Schneider J, Ehmann R, Zwirglmaier K, Drosten C, Wendtner C. Virological assessment of hospitalized patients with COVID-2019.

Nature 2020. doi: 10.1038/s41586-020-2196-x.

Young B E, Ong S W X, Kalimuddin S, Low J G, Tan S Y, Loh J, Ng O T, Marimuthu K, Ang LW, Mak T M, Lau S K, Anderson D E, Chan K S, Tan T Y, Ng T Y, Cui L, Said Z, Kurupatham L, Chen M I, Chan M, Vasoo S, Wang L F, Tan B H, Lin R T P, Lee V J M, Leo Y S, Lye D C, Singapore Novel Coronavirus Outbreak Research T. Epidemiologic features and clinical course of patients infected with SARS-CoV-2 in Singapore. JAMA 2020. doi: 10.1001/jama.2020.3204.

Zhang W, Du R H, Li B, Zheng X S, Yang X L, Hu B, Wang Y Y, Xiao G F, Yan B, Shi Z L, Zhou P. Molecular and serological investigation of 2019nCoV infected patients: implication of multiple shedding routes. Emerg Microbes Infect 2020; 9(1): 386-9. doi: 10.1080/22221751.2020.1729071.

538

Acta Orthopaedica 2020; 91 (5): 538–542

Does a surgical helmet provide protection against aerosol transmitted disease?

Max Joachim TEMMESFELD 1a, Rune Bruhn JAKOBSEN 1,2a and Peter GRANT 3,4 1 Department

of Orthopedic Surgery, Akershus University Hospital, Lørenskog, Norway; 2 Department of Health Management and Health Economics, Institute of Health and Society, University of Oslo, Norway; 3 Department of Orthopaedics, Institute of Clinical Sciences Sahlgrenska Academy, University of Gothenburg, Sweden; 4 Lovisenberg Diaconal Hospital, Oslo, Norway a Shared first authorship.

Correspondence: r.b.jakobsen@medisin.uio.no Submitted 2020-05-05. Accepted 2020-05-13.

Background and purpose — The COVID-19 pandemic caused by infection with SARS-CoV-2 has led to a global shortage of personal protective equipment (PPE). Various alternatives to ordinary PPE have been suggested to reduce transmission, which is primarily through droplets and aerosols. For many years orthopedic surgeons have been using surgical helmets as personal protection against blood-borne pathogens during arthroplasty surgery. We have investigated the possibility of using the Stryker Flyte surgical helmet as a respiratory protective device against airborne- and droplettransmitted disease, since the helmet shares many features with powered air-purifying respirators.

Materials and methods — Using an aerosol particle generator, we determined the filtration capacity of the Stryker Flyte helmet by placing particle counters measuring the concentrations of 0.3, 0.5, and 5 µm particles inside and outside of the helmet.

Results — We found that the helmet has insufficient capacity for filtrating aerosol particles, and, for 0.3 µm sized particles, we even recorded an accumulation of particles inside the helmet.

Interpretation — We conclude that the Stryker Flyte surgical helmet should not be used as a respiratory protective device when there is a risk for exposure to aerosol containing SARS-CoV-2, the virus causing COVID-19, in accordance with the recommendation from the manufacturer

The rapid development of the COVID-19 pandemic has led to severe shortages around the globe of personal protective equipment (PPE) for healthcare personnel such as regular surgical masks, tight-fitting masks (filtering facepieces [FFP]), protective eyeglasses, and face shields (Kamerow 2020, World Health Organization 2020). The virus causing COVID-19, SARS-CoV-2, is believed to spread primarily through droplets and aerosol in the immediate vicinity of an infected person (Bahl et al. 2020). A recent study showed that SARS-CoV-2 aerosols remain airborne and viable for at least 3 hours in closed spaces, thus raising the concern of airborne transmission (van Doremalen et al. 2020). A recent review also discussed the transmission of viral particles from aerosolized body fluids by using power drills, pulsed lavage, and other equipment during surgery (Basso et al. 2020). This has not been reported for SARS-CoV-2, but it is conceivable and has been shown in vitro for other viruses (Johnson and Robinson 1991, Garden et al. 2002).

Numerous alternative concepts of respiratory PPE have been suggested, for example the use of powered air-purifying respirators (PAPR). In these devices, filtered air is drawn by an electric fan into a closed helmet. Even though PAPRs offer superior protection compared with standard FFPs, hospitals would have to pay a lot to commercially acquire a sufficient number of PAPRs to equip their healthcare personnel.

Additionally, a shortage of PAPRs is to be expected during a pandemic.

Surgical helmets with internal electric fans share many features of a PAPR and were suggested as an alternative during the SARS epidemic in China in 2003 (Ahmed et al. 2005).

Such helmets are regularly used in orthopedic arthroplasty surgery. The hood of the surgical helmet is air-permeable over the fan intake, while the rest of the hood material is practically

© 2020 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 10.1080/17453674.2020.1771525

Acta Orthopaedica 2020; 91 (5): 538–542

539

Figure 1. Test setup. The dummy with helmet, hood, and gown in the test setup. Arrow indicating particle counter inside helmet. (White probe is a passive pressure probe not used for tests reported in this paper.)

Figure 2. Test setup. Arrow indicating particle counter outside helmet.

air impermeable. The original purpose of surgical helmets was to protect patients from particles that the surgical team might emit into the wound. Additionally, surgical helmets will protect the surgeon from direct body fluid contamination.

The Flyte model (Stryker Instruments, Kalamazoo, MI, USA) is commonly used in Scandinavia, continental Europe, and the United States and thus available in many Western hospitals. We investigated the protective abilities of this helmet in the context of the ongoing COVID-19 pandemic as part of an ongoing research project, which aims to convert surgical helmets into PAPRs with the help of specialized filters.

Another identical particle counter was positioned approximately 20 cm adjacent to the fan intake outside the hood (Figure 2). The concentrations of 0.3, 0.5, and 5µm particles per cubic foot inside and outside the surgical helmet were continuously recorded with synchronized dataloggers.

At first, the helmet fan was set to maximum speed, and the dummy was left for 30 minutes to establish a steady-state of ambient particles and to minimize the risk of false readings due to particles released from the hood and dummy.

We performed tests with the fan at the minimum and the maximum speed. Each test lasted between 15 and 30 minutes.

The total inward leakage (TIL) was calculated with the following formula:

Materials and methods The investigation was performed in an orthopedic operation theatre at Sahlgrenska University Hospital, Gothenburg, Sweden. The theatre is equipped with a mixed ventilation system that meets SIS-TS 39:2015 requirements and is audited annually (Swedish Standards Institute 2015). A Stryker Flyte helmet (Stryker Instruments, Kalamazoo, MI, USA) with the standard hood (Flyte Hood, product no. 0408-800-000) was mounted onto a dummy with head and torso. A standard surgical gown (Barrier Surgical Gown Classic, Mölnlycke, Sweden) was tightened around the neck. A 6D Laskin nozzle aerosol generator (Air Techniques International, Owings Mills, MD, USA) generated an oil-based hydrogenated 1-Decene homopolymer (PAO-4) test aerosol. The generator was active for approximately 15 seconds at the start of each test. A particle counter (Solair 3100, Lighthouse, Fremont, CA, USA) detector probe was fastened to the nose of the dummy (Figure 1).

TIL = (Particle concentration inside helmet)/(Particle concentration outside helmet (ambient)) Reciprocally, the filtration efficiency (FE) was calculated as follows: FE = 1–TIL Particle concentrations were converted to metric units and the data were summarized and visualized with descriptive statistics (area under the curve [AUC], mean, and 95% confidence intervals [CI]) using Prism 8 (Graphpad Software, San Diego, CA, USA).

Funding and potential conflicts of interest This work was performed as an ongoing project to research whether surgical helmets can be modified safely for use during the ongoing pandemic, with a patent pending for a retrofitted adapter. The project did not receive any specific funding.

540

A

Acta Orthopaedica 2020; 91 (5): 538–542

Total inward leakage ( ; %)

Total particles/m3

120

109

Inside total Outside total

100

108

80

107

60

0.3 µm

n × 107

Particle size 0.5 µm n × 107

5 µm n

Total n × 107

106

40

105

20 0 0

B

Particle counts inside and outside and filtration efficiency of the Stryker Flyte helmet

5

10

15

20

25

30

104

Time /min

Total inward leakage (%) 120

109

100

Inside helmet 1.67 0.60 439 2.27 (95% CI) (1.60–1.75) (0.52–0.68) (214–664) (2.12–2.25) Outside helmet 1.73 1.08 2,823 2.81 (95% CI) (1.58–1.89) (0.93–1.24) (1,180–4,467) (2.51–3.12) FE (%) 3.3 44 84 19 (95% CI) (–7.6 to 14) (31–57) (80–89) (7.2–31) CI, confidence interval, FE, filtration efficiency

108

80

107

60 106

40

105

20

104

0 0

C

5

10

15

20

25

Time /min

Total inward leakage (%) 120

109

100

108

80

107

60 106

40

105

20

104

0 0

5

10

15

Time /min

Figure 3. Line plots showing the total particles per m3 for the outside counter (green) and inside counter (red) versus time in minutes with bars showing the calculated percentage total inward leakage (TIL) at every time point. Each panel represents an individual experiment. Top panel with fan at maximum speed and the other 2 with fan at minimum speed.

Results The filtration efficiency (FE) for the total number of particles of all measured sizes was 19% (CI 7.2–31), corresponding to a total inward leakage (TIL) of 81%. The FE was statistically significantly lower for smaller sized particles and for the smallest particles of 0.3 µm the FE was particularly poor at 3.3% (CI –7.6 to 14) (Table 1). At declining ambient particle concentrations outside the helmet, the TIL also decreased in all experiments (Figure 3).

Nevertheless, the TIL for the 0.3 µm size aerosols regularly exceeded 100%, indicating an accumulation of particles inside the helmet when the outside particle concentration exceeded approximately 7×106/m3. This was not the case for the 0.5 µm size aerosols, yet the TIL did approach 100% at the highest concentrations of particles outside the helmet (Figure 4).

The filtration efficiency for the largest particles (5.0 µm) was markedly higher than for smaller particles (Table). The absolute numbers of large particles were few compared with the smaller sizes and did not markedly influence the TIL for the total number of particles (Figure 4).

Total inward leakage, 0.3 µm particles (%)

Total inward leakage, 0.5 µm particles (%)

Total inward leakage, all particles (%)

150

150

150

100

100

100

50

50

50

Fan at low speed Fan at high speed

0

0

0 104

105

106

107

108

Total count of 0.3-µm particles/m3

104

105

106

107

108

Total count of 0.5-µm particles/m3

105

106

107

108

109

Total count of all particles/m3

Figure 4. Scatterplots of percentage total inward leakage versus total count of 0.3 µm, 0.5 µm, and all particle sizes per m3.

Acta Orthopaedica 2020; 91 (5): 538–542

Discussion This study supports the recommendation by the manufacturer not to use the regular Stryker Flyte helmet as respiratory protection equipment against SARS-CoV-2 (Stryker Corporation, 2020). The filtration efficiency for particles of all measured sizes is low (19%). Our data indicate an accumulation of 0.3 µm particles inside the helmet, when ambient concentrations are high. In the light of this, wearing the Stryker Flyte helmet as the only respiratory PPE is not advisable.

Comparing the TIL for 0.3 µm and 0.5 µm (97% and 56%) in the present study with regular respiratory protective equipment such as FFPs further emphasizes that the Flyte helmet is by no means effective as protective equipment against aerosoltransmitted disease. Maximum permitted TIL for FFPs ranges from maximum 22% for an FFP-1 to 2% for FFP-3 (EUstandard EN149:2001), the latter being the recommended FFP to use for COVID-19 patients during aerosol-generating procedures. However, also FFPs have inherent problems and depend on a tight fit to the face of the user to reduce air bypassing the filter. Maximum protection is achieved using a PAPR equipped with an approved filter. It is imaginable that regular surgical helmets could be modified with proper filters to achieve sufficient FE. The protective ability of surgical hoods to safeguard the surgeon against exposure to infectious bodily fluids and direct transfer of microorganisms or particulate matter has been verified in vitro (Wendlandt et al.

2016). The hood on the Flyte Personal Protection System provides leading-class AAMI/ANSI Level 4 protection (Association for Advancement of Medical Instrumentation/American National Standards Institute); nevertheless, the top of the hood that the air passes through is not designed to filter aerosols.

Our findings are in line with a previous study from 2004 that evaluated the respiratory protective properties of 2 types of surgical helmets compared with an N100 filtering facepiece respirator combined with a surgical mask and full face shield (Derrick and Gomersall 2004). In that study they found ratios of ambient particle concentration to particle concentration inside the helmet to fall between 2 and 5, which corresponds to a TIL of 20–50%.

There are several limitations to this investigation. First, the experimental setup in this study did not comply with any formal regulatory standards. Second, we tested artificially produced aerosol particles of 3 predefined sizes and not viruscontaining particles as is sometimes performed (Fabian et al. 2008, Makison Booth et al. 2013). Third, it is likely that particles from a sneeze may be substantially larger than the particles tested in our setup (Han et al. 2013). However, the setup we constructed was deliberately similar to real-life situations with COVID-19 patients in the operation theatre. We used a PAO-4 test aerosol, which is FDA approved for regulated filter leakage testing. The particles we tested were of the same size as found in aerosols of healthy patients and patients

541

with influenza during coughing, assisted and regular breathing where a majority of particles have been found to be less than 1 µm (Papineni and Rosenthal 1997, Yang et al. 2007, Fabian et al. 2008, Wan et al. 2014). Most testing and certification protocols for respiratory protective equipment use very high concentrations of particles in the range of 7–10×1011/m3 (Derrick and Gomersall 2004, Gawn et al. 2008, Makison Booth et al. 2013). Even the highest concentrations of particles generated in our study were markedly lower than in other published studies. We consider it a strength that we used a maximal concentration of particles several orders of magnitude lower (~1,6×109/m3) and could still demonstrate substantial inward leakage.

We conclude that the Stryker Flyte surgical helmet does not provide sufficient protection against aerosol transmitted diseases. It is important to comply with the instructions of the manufacturer that the Stryker Flyte surgical helmet should not be used as a respiratory protective device against COVID-19.

MJT and PG conceived the study; PG performed the experiments with support as stated in the acknowledgments; RBJ and MJT performed the analysis; PG, MJT, and RBJ together wrote the first draft, revised, and approved the final version of the manuscript.

The authors would like to thank Anders Rehn and Carl Christiansen at CRC Medical for supplying equipment and help performing particle testing and Jan Gustén, Professor Emeritus, Chalmers University of Technology, for help in designing the test setup.

Acta thanks Maziar Mohaddes and Petri Virolainen for help with peer review of this study.

Ahmed N, Hare G M, Merkley J, Devlin R, Baker A. Open tracheostomy in a suspect severe acute respiratory syndrome (SARS) patient: brief technical communication. Can J Surg 2005; 48(1): 68-71.

Bahl P, Doolan C, de Silva C, Chughtai A A, Bourouiba L, MacIntyre C R. Airborne or droplet precautions for health workers treating COVID-19? J Infect Dis 2020. doi: 10.1093/infdis/jiaa189.

Basso T, Dale H, Langvatn H, Lønne G, Skråmm I, Westberg M, Wik T, Witsø E. Virus transmission during orthopedic surgery on patients with COVID19: a brief narrative review. Acta Orthop 2020; 91: [Ahead of print] Derrick J L, Gomersall C D. Surgical helmets and SARS infection. Emerg Infect Dis 2004; 10(2): 277-9. doi: 10.3201/eid1002.030764.

Fabian P, McDevitt J J, DeHaan W H, Fung R O, Cowling B J, Chan K H, Leung G M, Milton D K. Influenza virus in human exhaled breath: an observational study. PLoS One 2008; 3(7): e2691. doi: 10.1371/journal. pone.0002691.

Garden J M, O’Banion M K, Bakus A D, Olson C. Viral disease transmitted by laser-generated plume (aerosol). Arch Dermatol 2002; 138(10): 1303-7. doi: 10.1001/archderm.138.10.1303.

Gawn J, Clayton M, Makison C, Crook B. Evaluating the protection afforded by surgical masks against influenza bioaerosols 2008. Research Report 619 of the Health and Safety Executive. Available at https://www.hse.gov.uk/ research/rrhtm/rr619.htm. Accessed April 5, 2020.

Han Z Y, Weng W G, Huang Q Y. Characterizations of particle size distribution of the droplets exhaled by sneeze. J R Soc Interface 2013; 10(88): 20130560. doi: 10.1098/rsif.2013.0560.

542

Johnson G K, Robinson W S. Human immunodeficiency virus-1 (HIV-1) in the vapors of surgical power instruments. J Med Virol 1991; 33(1): 47-50. doi: 10.1002/jmv.1890330110.

Kamerow D. Covid-19: the crisis of personal protective equipment in the US. BMJ 2020; 369:m1367. doi: 10.1136/bmj.m1367.

Makison Booth C, Clayton M, Crook B, Gawn J M. Effectiveness of surgical masks against influenza bioaerosols. J Hosp Infect 2013; 84(1): 22-6. doi: 10.1016/j.jhin.2013.02.007.

Papineni R S, Rosenthal F S. The size distribution of droplets in the exhaled breath of healthy human subjects. J Aerosol Med 1997; 10(2): 105-16. doi: 10.1089/jam.1997.10.105.

Stryker Corporation. Stryker Flyte Personal Protection System COVID19 update April 10, 2020. Available at https://www.stryker.com/content/dam/stryker/orthopaedic-instruments/resources/Flyte%20Personal%20Protection%20System%20-%20COVID-19.pdf.

Accessed June 6, 2020.

Swedish Standards Institute. Microbiological cleanliness in the operating room – Preventing airborne contamination – Guidance and fundamental requirements. 2nd ed. Stockholm: Swedish Standards Institute; 2015.

Acta Orthopaedica 2020; 91 (5): 538–542

van Doremalen N, Bushmaker T, Morris D H, Holbrook M G, Gamble A, Williamson B N, Tamin A, Harcourt J L, Thornburg N J, Gerber S I, LloydSmith J O, de Wit E, Munster V J. Aerosol and surface stability of SARSCoV-2 as compared with SARS-CoV-1. N Engl J Med 2020; 382(16): 1564-7. doi: 10.1056/NEJMc2004973.

Wan G H, Wu C L, Chen Y F, Huang S H, Wang Y L, Chen C W. Particle size concentration distribution and influences on exhaled breath particles in mechanically ventilated patients. PLoS One 2014; 9(1): e87088. doi: 10.1371/journal.pone.0087088.

Wendlandt R, Thomas M, Kienast B, Schulz A P. In-vitro evaluation of surgical helmet systems for protecting surgeons from droplets generated during orthopaedic procedures. J Hosp Infect 2016; 94(1): 75-9. doi: 10.1016/j.jhin.2016.05.002.

World Health Organization. Rational use of personal protective equipment (PPE) for coronavirus disease (COVID-19) 2020. Available at https://apps. who.int/iris/bitstream/handle/10665/331498/WHO-2019-nCoV-IPCPPE_ use-2020.2-eng.pdf. Accessed April 1, 2020.

Yang S, Lee G W, Chen C M, Wu C C, Yu K P. The size and concentration of droplets generated by coughing in human subjects. J Aerosol Med 2007; 20(4): 484-94. doi: 10.1089/jam.2007.0610.

Acta Orthopaedica 2020; 91 (5): 543–546

543

Orthopedic surgery residents’ perception of online education in their programs during the COVID-19 pandemic: should it be maintained after the crisis?

Francisco FIGUEROA 1,2, David FIGUEROA 1, Rafael CALVO-MENA 1, Felipe NARVAEZ 3, Natalia MEDINA 4, and Juan PRIETO 5 1 Clinica

Alemana-Universidad del Desarrollo, Santiago; 2 Hospital Sotero del Rio, Santiago; 3 Pontificia Universidad Catolica, Santiago; 4 Universidad de Santiago, Santiago; 5 Universidad de Chile, Santiago, Chile Correspondence: franciscofigueroab@gmail.com Submitted 2020-05-11. Accepted 2020-05-26.

Background and purpose — During the COVID-19 pandemic, most of the teaching centers in Chile have shifted to online resources. We decided to do a survey on orthopedic residents regarding this type of education to assess for strengths and weaknesses of digital education in orthopedic programs.

Methods — A survey was performed targeting 110 orthopedic residents belonging to different training programs around the country. 100 residents completed the survey.

Results — 86% stated that their programs are using online education. When asked in detail, 86% had been involved in webinars, 28% had received online presentations, 12% had participated in online tests, and 7% had evaluated patients.

Webinars were rated (1 = very unsatisfactory, 10 = very satisfactory) with a mean grade of 8.1 (1–10), online presentations 7.3 (1–10), online tests 3.8 (1–8), and online patient evaluations 2.9 (1–9). When asked if, after the end of the pandemic, they would continue using the online modalities, 82% would continue attending webinars, 72% would continue watching online presentations, 27% would continue performing online tests, and 33% of the residents would continue performing online evaluations of patients.

Interpretation — Even though resident evaluation of online activities is positive, face-to-face theoretical activities are still valued as a necessary complement for orthopedic residency education.

The rapidly unfolding pandemic COVID-19 implies a surge in inpatient demand that may exceed the capacity of health systems even in developed economies (Ministry of Health Chile 2020). In response, local health governances have adopted drastic changes to care structures, particularly the operating theatre, deferring and cancelling elective surgeries, and limiting their functioning to urgent cases (Ministry of Health Chile 2020). On the other side, the population has been recommended to stay at home, with some locations including government-regulated quarantine in an attempt to limit the propagation of the disease. One of the results of these measures has been a dramatic decrease in outpatient and inpatient activity for orthopedic surgeons and their trainees, including a drastic reduction in the amount of surgeries. Restrictions of academic meetings and departmental gatherings among others have also been part of the efforts made to limit the spread of the virus and its impact on health workers. A regular day in orthopedic training includes patient rounds, surgeries, seminars, and research activities, all of which have been affected by the health ministry restrictions.

In a response to this situation, and in an attempt to maintain a reasonable number of educational instances for residents, our department and most of the teaching centers in Chile have shifted to online resources to continue the education of our residents. Online learning or e-learning is the use of internetbased resources for teaching and learning (Jayakumar et al.

2015). The advantages of e-learning in the surgical setting are well established. In addition to being easily accessed and updated, e-learning platforms accommodate a wide variety of learning styles and can effectively teach a broad array of surgically relevant information (Tarpada et al. 2016).

© 2020 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 10.1080/17453674.2020.1776461

544

A number of studies indicate that both surgical residents and medical students who use e-learning programs are more satisfied with their learning experience (time invested for learning in relation to learning they obtained, organization and clarity of the study subjects, and better performance on simulated cases, among others) compared with those who use traditional teaching methods (Citak et al. 2009, Horstmann et al.

2009, Funke et al. 2013). While use of e-learning platforms in surgical training has expanded over recent years, the role of this technology within orthopedic surgery residents remains scarcely studied. Tarpada et al. (2016) in a recent systematic review found 3 studies (Ceponis et al. 2007, Hearty et al. 2013, Heiland et al. 2014) regarding orthopedic residents’ online training for different surgical skills, all with promising results including better performance at diagnostic shoulder arthroscopy, better knowledge of the technique for closed reduction and pinning of supracondylar humeral fractures, and better surgical skills in spine surgery.

Our personal thought is that this modality of education has arrived abruptly in the different orthopedic residency programs, but has arrived to stay. Therefore, we decided to conduct a survey among orthopedic residents on a broad range of teaching programs in Chile to identify the strengths and weaknesses of this form of transmitting knowledge for the future of the orthopedic surgery.

Methods A survey was sent to 110 orthopedic residents in the first, second, and third year of education, who were on 7 different training programs around Chile on April 13, 2020. The survey was developed using the SurveyMonkey software (Palo Alto, CA, USA) and sent via WhatsApp (Mountain View, CA, USA). Residents were informed about the nature of the survey and that the results were to be used as part of a study. All the participants gave formal consent to be included in the analysis. The results were collected 5 days after the survey was sent. 100 residents answered the survey (80 male) with an age range from 25 to 34 years, representing around 50% of the total number of residents in Chile (Table).

Conflicts of interest The authors declare no conflicts of interest related to the subject of this study.

Acta Orthopaedica 2020; 91 (5): 543–546

Full questionnaire with residents’ answers

Rating

n mean (range)

1. Has your program started any internet-based education modalities because of the COVID-19 pandemic? (If your answer is yes go to question number 2, if your answer is no go to question number 7) • Yes 86 • No 14 2. In what type of online modalities have you participated? ( You can mark more than 1 answer) • Webinars (seminars with discussion

and participation of the audience) 77 • Presentations 25 • Surgical videos 0 • Exams and tests 11 • Video consultation on real patients 6 • Other 0 3. What web apps have you used for these instances?

• Zoom 78 • Microsoft Teams 8 4. How do you rate the online education modalities received during the COVID-19 pandemic? (Grades from 1 to 10, 1 is very unsatisfactory, 10 is very satisfactory) • Webinars 8.1 (1–10) • Presentations 7.3 (1–10) • Surgical videos Not rated • Exams and tests 3.8 (1–8) • Video consultation on real patients 2.9 (1–9) • Other Not rated 5. What difficulties do you see with this form of education?

(Free text question) • Technical aspects (slow internet

connection, audio problems) 36 • Absence of practical education 11 • Lack of concentration due to home

distractions 8 • Difficulties with scheduling and overload

of seminars/presentations 8 6. After the pandemic is over, which online-based educational instances would you continue to use?

(You can mark more than 1 answer) • Webinars 63 • Presentations 18 • Surgical videos 0 • Exams and tests 3 • Video consultation on real patients 2 • Other 0 7. Do you believe that after the pandemic is over, all the theoretical education of orthopedic residencies should be done using webbased platforms?

• Yes 30 • No 70 8. Do you consider that educational instances with groups smaller than 100 people (orthopedic department meetings, small subspecialist meetings) should be done using web-based platforms after the pandemic is over?

• Yes 40 • No 60

Results Regarding the utilization of online instances to compensate for the decrease in formal face-to-face training, 86% of the residents stated that their programs are using this type of education. When asked in detail about the specific modalities used, the most common was online seminars with the opportunity to

discuss topics from the audience (webinars) (Figure 1). Other options mentioned by the residents were: online presentations, online evaluations or tests, and video patient consultations.

The main app that has been used is “Zoom” (Zoom Video Communications, San Jose, CA, USA), in 88% of cases.

Acta Orthopaedica 2020; 91 (5): 543â&#x20AC;&#x201C;546

545

Number involved

Rating (0â&#x20AC;&#x201C;10)

Percentage of residents

100

10

100

80

8

80

60

6

60

40

4

40

20

2

20

0

Webinars Presentations Tests

Patient consultations

Figure 1. Number of residents involved in different online modalities.

0

Webinars Presentations Tests

Patient consultations

Figure 2. Evaluation of the different online modalities.

Regarding how the residents rate the different online learning experiences grading from 1 to 10 (with 1 being very unsatisfactory and 10 being very satisfactory), webinars were rated highest with a mean grade of 8.1 (Figure 2).

The most important difficulties of the different modalities for online education mentioned by the residents were: technical aspects (slow internet connection, audio problems) for 42% of the residents, absence of practical education in surgical training for 13%, lack of concentration due to home distractions for 9%, and difficulties with scheduling and overload of seminars/presentations for 9%.

When asked if, after the end of the pandemic, they would continue using the online modalities performed in the last weeks, most of the residents answered that they would continue attending webinars and would continue to perform online presentations. On the other hand, only a minority would continue performing online evaluations and tests and performing video consultations with patients (Figure 3).

30% of the residents believed that, when the COVID-19 pandemic is over, all non-practical medical education should be performed using internet-based tools. Similarly, 40% of the residents responded that academic meetings (orthopedic department meetings for example) or meetings involving less than 100 participants, should be done in a non-face-to-face fashion.

Discussion As the COVID-19 pandemic spreads around the world (World Health Organization 2020), and with the future of social interaction already unknown, orthopedic residency programs must consider different paths of learning. Meetings and conferences that previously took place face-to-face will now forcibly be done using internet-based spaces for an unknown period of time. However, despite the difficulties associated

0

Webinars Presentations Tests

Patient consultations

Figure 3. Percentage of residents who would continue with different online modalities.

with this change of paradigm, human minds always flourish and solutions are found to new problems. Weeks ago, we would have never thought that clinical meetings or even complete congresses would be completely undertaken using online resources, without the need to move from oneâ&#x20AC;&#x2122;s home or office. As this is new to residents in Chile, and they are the primary objective of these measures, we decided to survey them to learn about their preferences and the strengths and weakness of the different teaching methods during this pandemic, considering which should survive after the pandemic is over, and which should disappear.

We found notable preferences; for example, it seems that webinars and any kind of online presentation with the possibility for the audience to participate have arrived to stay with excellent ratings among residents. In addition, online presentations, a resource that was widely used around the world before the pandemic, are reinforced as an excellent teaching method because of the chance of being widely distributed with the resources already available today. Conversely, online tests or exams and video patient consultations seem to be only contingent solutions for a momentary problem rather than a permanent modality of teaching, with both being widely rejected by orthopedic residents around the country.

Despite belonging to the OECD (Organization for Economic Cooperation and Development) (OECD 2020) and having the 4th fastest internet speed in America (Fastmetrics 2020), technical problems are still important during online activities that require more data transmission than the regular amount, with loss of connection or delay and decrease of fluency reported.

Technical problems are frequently reported in articles evaluating online education methods in medicine, with one-quarter of the participants in the study by Horstmann et al. (2009) reporting this issue. The same problem is listed in the Jayakumar et al. (2015) systematic review on e-learning for surgical education. Eventually, with the dramatic increase in virtual meeting applications (Digital Trends 2020) because of the pandemic,

546

there will be advances in the consumption of data required, and this problem will be solved.

Another challenge noted was the necessity to obtain an adequate environment in the resident’s home, avoiding the many distractions that usually are not present in a more formal situation. Additionally, scheduling and overload of online activities were mentioned as a recurrent problem for residents. Regular training activities in Chile during normal conditions are a mixture of more practical activities (patient rounds, surgeries) than academic theoretical training. However, because of the COVID-19 pandemic, and the decrease in orthopedic surgery activity, an important number of trainers now have more time to perform seminars, webinars, or other modalities of online training. This produces a sense of overload in the residents and scheduling issues (more than 1 activity at the same time).

The solution to this problem involves better coordination and communication between trainers and the ability to provide a formal schedule of activities at reasonable times.

Finally, in spite of the good evaluation of some of the online modalities, as noted in the final 2 questions, most of the orthopedic residents believe that after the pandemic is over nonpractical medical education should be at least in part face-toface instead of being completely online. This corresponds to the results presented by Funke et al. (2013) in which after a questionnaire to evaluate an online teaching method, teaching in small groups and face-to-face learning obtained the best scores compared with online modalities. This creates a challenge for educators, because they need to balance online and in-person modalities, as face-to-face theoretical activities are still valued by most residents as a necessary complement to their education as orthopedic surgeons.

In summary, in an attempt to maintain a reasonable number of educational instances for residents, most of the teaching centers in Chile have shifted to online resources. Even though the evaluation of some of the activities is good (webinars, presentations), face-to-face theoretical activities are still valued as a necessary complement for their education as orthopedic surgeons.

Acta Orthopaedica 2020; 91 (5): 543–546

FF and DF: planning and writing, RC, FN, NM, JP: conducting the study and writing.

Acta thanks Bart Burger, Wilhelmina Henrietta Georgette Ekström for help with peer review of this study.

Ceponis P J, Chan D, Boorman R S, Hutchison C, Mohtadi N G. A randomized pilot validation of educational measures in teaching shoulder arthroscopy to surgical residents. Can J Surg 2007; 50(5): 387-93.

Citak M, Calafi A, Kendoff D, et al. An internet based learning tool in orthopaedic surgery: preliminary experiences and results. Technol Health Care 2009; 17(2): 141-8.

Digital Trends. Amid Zoom’s rise, Microsoft Teams is hosting 2.7 billion meeting minutes per day. https://www.digitaltrends.com/computing/microsoft-teams-surge-in-use/ 2020. Accessed April 14, 2020.

Fastmetrics. Average internet speeds by country. https://www.fastmetrics. com/internet-connection-speed-by-country.php 2020. Accessed April 14, 2020.

Funke K, Bonrath E, Mardin W A, et al. Blended learning in surgery using the Inmedea Simulator. Langenbecks Arch Surg 2013; 398(2): 335-40.

Heiland D H, Petridis A K, Maslehaty H, et al. Efficacy of a new video-based training model in spinal surgery. Surg Neurol Int 2014; 5: 1.

Hearty T, Maizels M, Pring M, et al. Orthopaedic resident preparedness for closed reduction and pinning of pediatric supracondylar fractures is improved by e-learning: a multisite randomized controlled study. J Bone Joint Surg Am 2013; 95(17): e1261-7.

Horstmann M, Renninger M, Hennenlotter J, Horstmann C C, Stenzl A. Blended e-learning in a web-based virtual hospital: a useful tool for undergraduate education in urology. Educ Health (Abingdon) 2009; 22(2): 269.

Jayakumar N, Brunckhorst O, Dasgupta P, Khan M S, Ahmed K. e-Learning in surgical education: a systematic review. J Surg Educ 2015; 72(6): 114557.

Ministry of Health Chile [Ministerio de Salud de Chile]. https://www.minsal. cl/wp-content/uploads/2020/03/1745861_web.pdf 2020. Accessed April 14, 2020.

Organization for Economic Cooperation and Development (OECD). https:// www.oecd.org 2020. Accessed April 14, 2020.

Tarpada S P, Morris M T, Burton D A. E-learning in orthopedic surgery training: a systematic review. J Orthop 2016; 13(4): 425-30.

World Health Organization (WHO). Coronavirus disease (COVID-19) pandemic. https://www.who.int/emergencies/diseases/novel-coronavirus-2019 2020. Accessed April 14, 2020.

Acta Orthopaedica 2020; 91 (5): 547–550

547

Reflections The personal and professional impact of COVID-19 on orthopedic surgery trainees: reflections from an incoming intern, current intern, and chief resident David N BERNSTEIN, Nattaly GREENE, and Ishaq O IBRAHIM Harvard Combined Orthopaedic Residency Program, Massachusetts General Hospital, Boston, USA Correspondence: dbernstein4@mgh.harvard.edu Submitted 2020-06-10. Accepted 2020-06-22.

On December 31, 2019, Chinese health officials reported to the World Health Organization (WHO) Country Office (World Health Organization 2020) growing concern regarding an increasing number of cases of pneumonia of unknown cause. Since that time, numerous public health interventions have been introduced (Patel and Jernigan 2020), a global pandemic was declared (Cucinotta and Vanelli 2020), and much was learned about the novel coronavirus (i.e., SARS-CoV-2), the microscopic culprit behind the world’s ongoing public health and economic crises. In addition, the COVID-19 global pandemic has had a notable impact on health systems globally, placing a strain on resources and reducing or halting nonemergent medical and operative care.

Much has been written about COVID-19 and its impact on orthopedic surgery (Jain et al. 2020, Ranuccio et al. 2020), including on orthopedic surgery trainee education (Kogan et al. 2020, Schwartz et al. 2020, Stambough et al. 2020). Further, there is a growing focus and commentary on physician well-being—both physical and mental (Massey et al. 2020).

However, transparent COVID-19 perspectives by orthopedic surgery trainees—on personal and professional levels—have been limited. While the delivery of high-quality orthopedic care and education remains central to the musculoskeletal health of patients globally, we must also pause for reflection, a crucial element of narrative medicine that has a number of benefits for patient care and provider well-being (Charon 2001).

In this perspective, we share the reflections of 3 orthopedic surgery trainees at a single academic training program in a large city, Boston, in the United States: 1 incoming intern; 1 current intern; and 1 chief resident. While all are trainees, the difference in training year is crucial to illuminate the different professional stressors. Additionally, each individual has their own personal stressors and reactions to the ongoing crisis. We conclude this perspective by summarizing key themes and urging our orthopedic surgery trainee colleagues globally to engage in a similar exercise.

Professional reflections Chief resident (IOI) As a burgeoning surgeon in my final year of residency, the pandemic has appeared to strike at a particularly inopportune time. In many ways, I viewed my last year of residency as my “final act” as a trainee prior to moving on to the increased responsibility and autonomy of fellowship and, subsequently, my attending career. Now several weeks into the pandemic, personal excitement for a strong finish has been replaced by unease and discontentment.

When social distancing policies began to take hold in March, I was settling into my role as service chief for 1 of our busy arthroplasty services. The high-volume rotation would have afforded me plenty of additional “reps” to maximize my proficiency in arthroplasties prior to moving on to trauma fellowship in the summer. The COVID-19 pandemic swiftly brought that opportunity to an end. The lost training and uncertainty regarding our return to familiar affairs has left me with an odd sense of unfulfillment as I prepare to take the next steps in my career.

At the clinical level, COVID-19 has entirely altered our workflow. As our resident pool is thinned by redeployments to the emergency department (ED) and surgical intensive care unit (SICU), coverage of the orthopedic trauma service and other emergent or urgent cases has become a collective effort amongst the remaining residents. Persistent limitations of COVID-19 testing and ever-evolving perioperative protocols secondary to the virus have been frequent sources of frustration. Additionally, loss of the dedicated “trauma room” has made the workday less predictable.

Overall, support from our program leadership has been thoughtful, compassionate, and frequent. As an example, our residency program director and hospital department chairs updated residents on a daily basis. One of the core goals of this communication was to keep us abreast of important hospital trends and the possible implications for us as trainees (e.g., redeployment to another service/area of the hospital or sched-

© 2020 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 10.1080/17453674.2020.1789310

548

ule restructuring). This daily insight was critical in maintaining camaraderie and morale in an unprecedented time.

Current intern (NG) COVID-19 was deemed to be a global pandemic 2 weeks prior to the start of our intern bootcamp month—a month dedicated to teaching first-year residents critical orthopedic knowledge and skills. At the time, I was on a general surgery rotation at 1 of our community hospitals, and I witnessed notable changes daily as we prepared to face the crisis. Elective cases began dropping off our operating room (OR) boards, emails began flooding our inboxes, and schedules were rearranged in attempt to decrease the number of people who might be exposed to the virus. Our general surgery team, which typically includes 2 interns, a postgraduate year (PGY)-2, PGY-3, and a PGY-5, was distilled to me and the PGY-5.

As time progressed, it became clear that a virtual orthopedic surgery intern bootcamp would take the place of the traditional in-person month of orthopedic learning and co-intern bonding. To make this transition successful, a box of orthopedic surgery instruments was delivered to my home and a calendar of remote lectures to be delivered via web conference was created. Instead of developing my knowledge and skills alongside my co-interns in person, I have been glued to my computer screen, learning orthopedic surgery from afar. In addition, the hands-on portion of skill development has led to my small studio apartment being filled with a Black & Decker work bench, sawbones, drills, plates, sutures, and other essential instruments and equipment required to learn the basics of orthopedic surgery. Overall, I feel that I have been able to adapt and dedicate time to learning and developing my orthopedic surgery skills remotely.

My professional growth via a virtual bootcamp was only made possible because of the incredible support of program leadership, faculty, and the tireless work of our senior trainees.

Without their remarkable effort to adapt teaching and learning opportunities to online platforms, I would not have had such a transformative learning experience.

Incoming intern (DNB) The final year of medical school in the United States is a rollercoaster ride. While the stress of visiting clinical clerkships and applying to residency begins the year, celebrations and extended travel typically conclude the year. Most medical students long for the final stretch of medical school, a sign that their years of hard work and dedication has paid off.

However, 2020 is not a typical year. As the global pandemic raged on, it became clear by early March that these traditional celebrations, including Match Day (i.e., the day when medical students who apply to complete their medical training in the United States learn where they will attend residency) and graduation, were not going to be held in person. In lieu of the pomp and circumstance, short virtual “celebrations” were planned and carried out. Despite the efforts and hard work of

Acta Orthopaedica 2020; 91 (5): 547–550

medical school leaders, I felt, and continue to feel, a gaping hole in our medical school experience—a lack of closure. Yet, I still feel a sense of professional pride and accomplishment, especially having matched into my desired field. Once residency begins, I firmly believe I will experience closure to my medical school journey.

Overall, the ongoing global pandemic has caused a notable disruption of the “normal” conclusion of medical school.

Fortunately, however, even a global pandemic could not take away my remarkable mentors and their dedication to my professional growth; indeed, they have spent innumerable hours engaging with me via email and virtual meetings. Further, the clear messaging by my incoming orthopedic surgery residency program’s leadership has eased my anxiety and prepared me to enter the orthopedic surgery workforce during this unusual time in history.

Personal reflections Chief resident During the COVID-19 pandemic, concern over the well-being of my family, friends, and peers has been a constant source of stress. Increased downtime, however, has allowed me time to frequently connect with my family and even reconnect with friends from earlier stages of my life. During more “normal” times, this might not have been possible. Additionally, closure of “nonessential” services, including the gym and barbershop, has forced us to be resourceful and explore new ways to exercise, cook, and groom at home. These endeavors have been mostly successful and have helped us to maintain our sanity during these difficult times.

Current intern Throughout this crisis, I have felt more anxious and preoccupied; however, these feelings have often been mixed with a sense of overwhelming gratitude for my health and for the health of my family. My core concerns have revolved around my family members, including their health and their financial well-being. As the pandemic evolved, it became clear that its impact started to affect communities of color disproportionally. Thus, additional concerns included trying to understand the root cause of this inequity and how I could best help my community. I believe that the best antidote for my anxiety has been going into work and feeling useful. We went into medicine with the intention of helping others, and as orthopedic surgeons we find particular joy in fixing problems. Therefore, being at the hospital, doing meaningful work, and caring for others has been energizing.

Additionally, I have developed a workout routine, which has been an important way for me to look after myself. Reincorporating exercise has been very therapeutic, and it has helped to break up each day. Also, I have resumed journaling and writing to friends and family as a way to reflect on this unique experience. Lastly, I have noticed that I am frequently using the FaceTime application on my iPhone to chat face-to-face

Acta Orthopaedica 2020; 91 (5): 547–550

with family and friends. I have reconnected as well with those with whom I had previously lost contact. This has helped me fight the feeling of isolation.

Incoming intern Throughout the COVID-19 global pandemic, I have had a persistent level of uneasiness—a pit in my stomach. While many of my colleagues across the country graduated medical school early to join the frontlines, my training program did not require me to begin residency early. At times, I felt a sense of guilt for not being directly engaged in patient care, especially because I felt as if I could help—even if that meant simply handing out masks or organizing documents.

In addition to my own uneasiness, I had a heightened level of angst surrounding the health of my family. During the pandemic, I have been fortunate to spend time in person with my family. However, my father, a pediatrician, has continued to deliver in-person care to children. Further, my parents are in an age range that increases their risk of COVID-19 complications, and my sister is pregnant. Despite each member of my family taking all the necessary safety precautions, there remains a possibility of infection in 1 or more of them. However, because my family has remained healthy to date, I have had a constant sense of gratitude, especially as I witness so many around the world suffering. Further, I am thankful that I was able to utilize the “Stay At Home” order to spend highquality time with those who matter most to me.

What one word would you use to describe your feelings about the current COVID-19 global pandemic?

Chief resident: Ambiguity.

Current intern: Uncertainty.

Incoming intern: Grit.

When I look back on this public health crisis in 10 years, I will remember…

Chief resident: … the leadership, strength, and resilience of our healthcare communities.

Current intern: … my work with the Latinx community as part of the Spanish Language Care Group at our tertiary care hospital. I will never forget the sense of relief that I witnessed on patients’ faces or heard in patients’ voices as they first saw me or heard me over the phone communicating in their native language. Additionally, I will always remember how rewarding and gratifying it was for me to be able to contribute in a meaningful way and to ensure that Spanish-speaking patients had a voice during this crisis.

Incoming intern: … the stories of inspirational everyday heroism, which reinforced my passion for helping others through a career in orthopedic surgery.

Final thoughts Three key themes transcend our individual reflections. A common theme shared by all of us is a heightened sense of

549

anxiety. While similar reasons underlie this feeling from a personal standpoint (i.e., concern about the well-being of family and friends), there are differences in the cause of uneasiness from a professional standpoint. This is likely secondary to the different levels of orthopedic surgery training we have each completed to this point in time. Another theme in our responses focused on family, friends, and community. Each of us was, and continues to be, most worried not about ourselves but about those around us. From a community standpoint, COVID-19 magnified the presence of systemic issues in our society, including race-related inequity and healthcare disparities such as burden of illness and death (US Centers for Disease Control and Prevention 2020). For example, many in minority communities often hold essential jobs that do not allow for them to work from home. Unfortunately, there is no easy fix to this inequity; however, discussion, education, and pledging to take action is crucial to make substantive change.

Further, we took the additional time available to us, given the pandemic, to connect with family and friends more consistently. Another consistent theme is the deep appreciation of support and leadership in a challenging time. Indeed, despite the uncertainties around COVID-19, we each appreciated the dedication of faculty; they have sacrificed much to ensure our continued professional development and personal well-being.

Further, leadership provided and encouraged communication, making it easy and acceptable to communicate with them regarding personal matters and struggles, as well as seek out resources for help, if needed.

By reflecting on COVID-19 and its impact on each of us— both professionally and personally—we hope to normalize many of the feelings experienced by other orthopedic surgery trainees. We urge our fellow trainees in the United States and around the globe to share their reflections and insights on their own experiences as well. This can be done via social media, peer-reviewed journal articles, or other means. Not only may “best practices” be able to be appreciated, but the healing process can begin. The future of health care and musculoskeletal care is bright. If all of us in orthopedic surgery—from attending surgeons to trainees—fully support one another, we will move beyond the COVID-19 global pandemic physically and mentally stronger than ever.

Funding, potential conflicts of interest No funding was received in support of this perspective. All authors (DNB, NG, IOI) declare no related potential conflicts of interest.

Acta thanks Roger Skogman for help with peer review of this study.

Charon R. Narrative medicinea model for empathy, reflection, profession, and trust. JAMA 2001; 286: 1897-902.

Cucinotta D, Vanelli M. WHO declares COVID-19 a pandemic. Acta Biomed 2020; 91: 157-60.

550

Jain A, Jain P, Aggarwal S. Sars-Cov-2 impact on elective orthopaedic surgery: implications for post-pandemic recovery. J Bone Joint Surg Am 2020. doi: 10.2106/JBJS.20.00602. Online ahead of print.

Kogan M, Klein S E, Hannon C P, Nolte M T. Orthopaedic education during the COVID-19 pandemic. J Am Acad Orthop Surg 2020; 28: e456-e64.

Massey P A, McClary K, Zhang A S, Savoie F H, Barton R S. Orthopaedic surgical selection and inpatient paradigms during the coronavirus (COVID19) pandemic. J Am Acad Orthop Surg 2020; 28: 436-50.

Patel A, Jernigan D B. Initial public health response and interim clinical guidance for the 2019 novel coronavirus outbreak: United States, December 31, 2019–February 4, 2020. MMWR Morb Mortal Wkly Rep 2020; 69: 140-6.

Ranuccio F, Tarducci L, Familiari F, Mastroianni V, Giuzio E. Disruptive effect of COVID-19 on orthopaedic daily practice: a cross-sectional survey. J Bone Joint Surg Am 2020. doi: 10.2106/JBJS.20.00604. Online ahead of print.

Acta Orthopaedica 2020; 91 (5): 547–550

Schwartz A M, Wilson J M, Boden S D, Moore T J Jr, Bradbury T L Jr, Fletcher N D. Managing resident workforce and education during the COVID-19 pandemic: evolving strategies and lessons learned. JBJS Open Access 2020; 5.

Stambough J B, Curtin B M, Gililland J M, et al. The past, present, and future of orthopedic education: lessons learned from the COVID-19 pandemic. J Arthroplasty 2020; 35(7): S60–S64.

US Centers for Disease Control and Prevention (CDC). COVID-19 in racial and ethnic minority groups 2020. https://www.cdc.gov/coronavirus/2019ncov/need-extra-precautions/racial-ethnic-minorities.html (accessed June 17, 2020).

World Health Organization. Rolling updates on coronavirus disease (COVID19) 2020. https://www.who.int/emergencies/diseases/novel-coronavirus2019/events-as-they-happen (accessed June 1, 2020).

Acta Orthopaedica 2020; 91 (5): 551–555

551

The COVID-19 pandemic in Singapore: what does it mean for arthroplasty?

Joshua DECRUZ, Sumanth PRABHAKAR, Benjamin Tze Kiong DING, and Remesh KUNNASEGARAN

Department of Orthopaedic Surgery, Tan Tock Seng Hospital, Singapore Correspondence: joshua.decruz@mohh.com.sg Submitted 2020-04-28. Accepted 2020-05-13.

Background and purpose — The ongoing Coronavirus Disease-19 (COVID-19) pandemic has taken a toll on healthcare systems around the world. This has led to guidelines advising against elective procedures, which includes elective arthroplasty. Despite arthroplasty being an elective procedure, some arthroplasties are arguably essential, as pain or functional impairment maybe devastating for patients, especially during this difficult period. We describe our experience as the Division of Arthroplasty in the hospital at the epicenter of the COVID-19 pandemic in Singapore.

Patients and methods — The number of COVID-19 cases reported both nationwide and at our institution from February 2020 to date were reviewed. We then collated the number of arthroplasties that we were able to cope with on a weekly basis and charted it against the number of new COVID-19 cases admitted to our institution and the prevalence of COVID-19 within the Singapore population.

Results — During the COVID-19 pandemic period, a significant decrease in the volume of arthroplasties was seen. 47 arthroplasties were performed during the pandemic period from February to April, with a weekly average of 5 cases.

This was a 74% reduction compared with our institutional baseline. The least number of surgeries were performed during early periods of the pandemic. This eventually rose to a maximum of 47% of our baseline numbers. Throughout this period, no cases of COVID-19 infection were reported amongst the orthopedic inpatients at our institution.

Interpretation — During the early periods of the pandemic, careful planning was required to evaluate the pandemic situation and gauge our resources and manpower.

Our study illustrates the number of arthroplasties that can potentially be done relative to the disease curve. This could serve as a guide to reinstating arthroplasty as the pandemic dies down. However, it is prudent to note that these situations are widely dynamic and frequent re-evaluation is required to secure patient and healthcare personnel safety, while ensuring appropriate care is delivered.

Tan Tock Seng Hospital (TTSH) in Singapore is a unique institution, functioning almost as a single unit with the National Centre of Infectious Disease (NCID), which has been at the epicenter of the Coronavirus Disease-19 (COVID-19) pandemic in the nation. Following the 2002–2004 Severe Acute Respiratory Syndrome (SARS) epidemic, the NCID was built in an effort to prepare Singapore for future outbreaks and to allow for continued management of patients with chronic conditions (Kurohi 2019).

The NCID handled about 70% of the screening load at the start of the COVID-19 pandemic in Singapore, a country with a population of around 5.8 million. Operational manpower required for the running of the screening centers, wards, and intensive care units in NCID had been solely derived from TTSH until April 1, 2020, after which other hospitals in Singapore began contributing manpower to aid with the crisis.

The Division of Arthroplasty, within the Department of Orthopaedic Surgery in TTSH, consists of 8 surgeons who performed an average of 84 arthroplasties, including knee replacements, hip replacements, and elective revision arthroplasties, per month prior to the COVID-19 outbreak. When evidence of community spread of the disease first surfaced on February 7, 2020, the Ministry of Health (MOH) of Singapore responded by declaring the Disease Outbreak Response System Condition (DORSCON) alert level of Orange (Ministry of Health, Singapore 2020c [last accessed April 19, 2020]).

At the epicenter of the outbreak, TTSH began reassigning manpower to ensure that NCID was adequately staffed to handle the situation. At that point, one-quarter of the manpower from the Department of Orthopaedic Surgery was deployed to assist with the screening effort in NCID. Considering the available resources, the hospital management advised a reduction in the “Business As Usual (BAU)” activities to half its capacity, which included elective surgeries like arthroplasty.

Throughout this period, the Ministry of Health remained in close contact with all public health institutions to provide directives regarding the deferment of elective surgeries, with

© 2020 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 10.1080/17453674.2020.1774138

552

Acta Orthopaedica 2020; 91 (5): 551–555

Number of COVID-19 cases/week

Total number of COVID-19 cases

450

1,200

30

NCID/TTSH Singapore

400

Arthroplasties/week

Weekly number of new COVID-19 cases

Arthroplasties/week

300

30

COVID-19 cases (Singapore) Arthroplasties/week (TTSH)

New COVID-19 cases/week (TTSH/NCID) Arthroplasties/week (TTSH) (baseline 19/week)

1,000

25

250

800

20

200

20

600

15

150

15

400

10

100

10

200

5

50

5

0

0

25

350 300 250 200 150 100 50 0

0 2–8

9–15

16–22 23–29

February

1–7

8–14

15–21 22–28

March

29–4

April

Figure 1. Weekly COVID-19 caseload at NCID/TTSH in comparison with cases nationwide.

2–8

9–15

16–22 23–29

February

1–7

8–14

15–21 22–28

March

29–4

April

Figure 2. Weekly comparison of total COVID19 cases in Singapore (cumulative numbers), and the number of arthroplasties performed at TTSH.

an eye on the rising incidence of COVID-19 in Singapore, and in anticipation of the potential increase in the number of patients requiring critical care (Ministry of Health, Singapore 2020a, b [last accessed April 19, 2020]).

We describe our experience in the Division of Arthroplasty at the epicenter of the COVID-19 pandemic in Singapore.

Patients and methods This study is a retrospective review of all elective arthroplasties that were performed since the start of the COVID-19 pandemic. The number of COVID-19 cases reported both nationwide and at our institution from February 2020 to date were also reviewed. We collated the number of arthroplasties that were performed on a weekly basis and charted it against the number of new COVID-19 cases admitted to our institution and the prevalence of COVID-19 within the Singapore population.

Surgeries were performed in concordance with hospital policy: only patients with severe arthropathy with significant impairment of function were permitted to proceed with their planned surgery during this period. Surgeries were performed without jeopardizing the stock of essential items while maintaining the safety of healthcare personnel and patients. All other planned arthroplasties were cancelled.

The months of October/November 2019 were selected as a baseline reference for the volume of arthroplasties performed weekly.

Funding and potential conflicts of interest We did not receive funding of any kind. We declare that there is no conflict, disclosure of relationship, or interests that could have a direct or potential influence or impart bias on the work.

0 2–8

9–15

16–22 23–29

February

1–7

8–14

15–21 22–28

March

29–4

April

Figure 3. Weekly comparison of number of new COVID-19 cases reported at TTSH/ NCID and number of arthroplasties performed at TTSH.

Table 1. Demographic data on 47 arthroplasties Demographic

mean (SD)

n

Female sex Age 68 (8.1) 50–59 60–69 70–79 80–89 BMI 27 (5.5) 15–19 20–24 25–29 30–34 35–39 40–44 45–49 ASA Class: 1 2 3 Operation performed TKR THR UKA Indication Osteoarthritis Rheumatoid arthritis Avascular necrosis Revision surgery Length of stay (days) 4.7 (2.3) < 5 5–10 > 10

35 11 13 19 4 2 15 19 6 3 1 1 1 32 14 32 5 10 39 2 4 2 29 16 2

Results The majority of COVID-19 cases in Singapore were admitted to our institution (Figure 1). Of the 1,189 cases reported by April 4, 2020, 757 (64%) were admitted to both NCID and

Acta Orthopaedica 2020; 91 (5): 551–555

553

Table 2. Management of TTSH Arthroplasty Division against the World Health Organization (WHO) pandemic phases WHO Manpower pandemic Arthroplasty unit Arthroplasty contribution phase Definition management goals caseload limit for pandemic Phase 3 Human infection

(transmission in

close contacts) Phase 4 Small cluster

(< 25 cases

lasting < 2 weeks) Phase 5 Large cluster

(25–50 cases

over 2–4 weeks)

Phase 6 Widespread in

general population

Post-peak Levels of infections period have dropped below

peak levels in most

countries

Assessment of stockpile and resources

0%

All cases

Consider postponing future 100% 0% cases that are high risk (ASA 3 and above) or less severe Minimize use of intensive 50% 25% care beds/blood products Limit stockpile and manpower usage Preserve intensive care 0% 25–75% beds/blood products Stockpile for pandemic usage solely Minimize use of intensive care 25% 25% beds/blood products Limit stockpile and manpower usage

All cases

TTSH. The number of COVID-19 increased substantially after March 14, from 212 on March 14, to 1,189 on April 4 (Figure 2).

167 arthroplasties were performed during the baseline reference months of October–November 2019, which translates to an average weekly volume of 19 surgeries.

During the COVID-19 pandemic period, the volume of arthroplasties decreased (Figures 2 and 3). 47 arthroplasties were performed during the pandemic period from February to April, with a weekly average of 5 cases. This was a 74% reduction compared with our institutional baseline. The patients’ mean age was 68 years, with a mean BMI of 27. Most of them were ASA Class II (Table 1).

A significant reduction in arthroplasties was initially seen during the early periods of the pandemic with the least number of cases being performed (Figures 2 and 3). Only 2 arthroplasties were performed between February 2 and February 15, 2020, which was a mere 5% of our usual volume. There was an increase in the number of elective arthroplasties between February 15 and April 4, 2020 with a peak volume of 9 cases weekly (47% of baseline) during the March 8 to March 14.

During the pandemic period, a review of medical records and readmissions showed that none of the orthopedic inpatients in our institution were diagnosed with COVID-19 infection. The diagnosis of COVID-19 infection was based on the results of the RT-PCR based nasopharyngeal swab.

Discussion The ongoing Coronavirus Disease-19 (COVID-19) pandemic has taken a toll on healthcare systems around the world.

100%

Elective cases selection criteria

ASA 1 or 2 patients Unlikely to require intensive care/blood products Severe arthropathy with limited mobility Cancel all cases Same as Phase 5

Shortage of resources, such as personal protective equipment (PPE), masks, ventilators, ICU beds, and even adequately trained healthcare workers has made this pandemic a global crisis (Vannabouathong et al. 2020). Despite being a purposebuilt center, the National Centre of Infectious Disease (NCID) in Singapore has now reached its full capacity. We now cope with a progressively lower proportion of the total number of patients in Singapore from a peak of 78% in mid-February to 64% as of April 4 (Figure 1). The other hospitals and community isolation facilities in Singapore have ramped up their efforts in handling the pandemic.

In our institution, the initial reduction in the number of arthroplasties performed between February 2 and February 15, 2020 coincided with the referral of all potential COVID19 cases from primary healthcare to NCID and a resultant diversion of manpower from TTSH to NCID. This planned reduction of caseload also allowed the institution to plan effectively for the evolving pandemic. We worked with the hospital administrators and anesthetists to evaluate the pandemic and gauge our resources and manpower.

The number of surgeries performed increased marginally after February 15, 2020 with the hospital administration allowing for arthroplasties in patients with severe arthropathy to proceed because of the initial low and controlled numbers of patients with COVID-19. A summary of the management goals, manpower reorganization, and elective caseload limits of our Arthroplasty Division is shown against the World Health Organization (WHO) pandemic phases in Table 2 (World Health Organization 2010).

With efforts concentrated on the control and eradication of this infection, the management of patients with unrelated chronic diseases, including severe arthropathies, has taken a

554

step back and rightfully so. Guidelines have been developed by institutions to aid surgeons with their practice and with prioritization of elective surgeries. The main aims of these guidelines are: (1) to minimize use of essential items (PPE, cleaning supplies, ventilators, bed, blood products, drugs, and personnel); (2) protect healthcare personnel and patients from additional risks during this period; and (3) ensure appropriate care is delivered (Ding et al. 2020, Guy et al. 2020, Liang et al.

2020b, Soh et al. 2020, Vannabouathong et. al 2020).

The American College of Surgeons (ACS) Guidance for Elective Surgery recommended that all surgeries for chronic hip and knee pain be rescheduled in the early stages of the pandemic. In the later stages, even surgeries for acute hip and knee pain were to be rescheduled (American College of Surgeons [last accessed April 19, 2020]). On the other hand, the National Health Service (NHS) of the United Kingdom suggested that elective surgery can proceed during low COVID19 prevalence, especially for ASA class 1 patients (National Health Service, United Kingdom [last accessed April 19, 2020]). Similarly, the algorithm from Piedmont Orthopedics in Georgia, USA suggested that elective surgeries can proceed after careful consideration of all factors (Schmidt 2020). The American Academy of Orthopaedic Surgeons (AAOS) also suggests the application of guidelines depending on the severity of the pandemic in each country or area and their available resources (Guy et al. 2020). The WHO recommends that routine and elective services be deferred immediately or displaced to other settings or non-affected areas. However, this is to be done while minimizing adverse effects of such interruptions, such as decreased quality of life, increased burden on caregivers, and poor self-management as a result of an exacerbation of existing conditions (Regional Office for Europe, World Health Organization [last accessed May 11, 2020]).

However, proceeding with elective surgeries that use essential items has been said to be shortsighted and negatively looked upon in the media, especially when the availability of these items may be unequal between institutions and regions (Guy et al. 2020, O’Donnell 2020). Only 5 of the 50 states in the United States of America provided guidance on elective orthopedic procedures, and 4 out of these 5 states advised against all arthroplasty procedures during this period (Sarac et al. 2020).

Subsequently, Liang et al. (2020a) recommend not exceeding 25% of the usual caseload during the post-peak period of the pandemic, as there will still be a need for evaluation of response, recovery, and preparation for a possible second wave.

Arthroplasty has been one of the most commonly performed elective surgical procedures and has been increasing in incidence in a linear fashion in recent times (Sloan et al. 2018).

The average number of arthroplasties in the Organisation for Economic Co-operation and Development (OECD) countries in 2015 was 292 per 100,000 population (OECD 2017). This is because osteoarthritis is recognized as 1 of the 10 most disabling diseases in developed countries. Despite being an elective procedure, arguments can be made for the need for

Acta Orthopaedica 2020; 91 (5): 551–555

some arthroplasties, as pain or functional impairment could prove devastating for patients, especially during this difficult period (Ding et al. 2020). The 47 arthroplasties performed in our institution were selected on that basis in accordance with our institutional guidelines.

Our study illustrates the number of arthroplasties that can be done relative to the disease curve. As an institution, it may be possible for us to reinstate arthroplasties as the pandemic dies down based on these numbers, provided that the resources and manpower remain available. In case of subsequent waves of this pandemic, we may also use these numbers as a guide to cut down on elective arthroplasties in a timely manner.

Acta thanks Anne Garland and Katre Maasalu for help with peer review of this study.

American College of Surgeons, USA. COVID-19 guidelines for triage of orthopaedic patients. March 24, 2020. https://www.facs.org/covid-19/clinical-guidance/elective-case/orthopaedics (accessed April 19, 2020).

Ding B T K, Soh T, Tan B Y, Oh J Y, Mohd Fadhil M F B, Rasappan K, Lee K T. Operating in a pandemic. Bone Joint Surg Am 2020: 00: 1-8.

Guy D K, Bosco III J A, Savoie II F H. American Academy of Orthopaedic Surgeons. AAOS guidelines on elective surgery during the COVID-19 pandemic: March 31. https://www.aaos.org/globalassets/about/covid-19/aaosguidelines-on-elective-surgery.pdf (accessed April 19, 2020).

Liang Z C, Chong M S Y, Liu G K P, Della Valle A G, Wang D, Lyu X, Chang C H, Cho T J, Haas S B, Fisher D, Murphy D, Hui J H P. COVID-19 and elective surgery: 7 practical tips for a safe, successful and sustainable reboot.

Ann Surg 2020a. https://journals.lww.com/annalsofsurgery/Documents/ COVID-19%20and%20Elective%20Surgery.pdf (accessed May 14, 2020) Liang Z C, Wang W, Murphy D, Hui J H P. Novel coronavirus and orthopaedic surgery. J Bone Joint Surg Am 2020b; 00: e1(1-5).

Kurohi R. Centre to boost infectious disease management. Straits Times, January 17, 2019; Sect. B3 (col. 1).

Ministry of Health, Singapore. Circuit breaker to minimise further spread of COVID-19. 2020a, Apr 3. https://www.moh.gov.sg/news-highlights/ details/circuit-breaker-to-minimise-further-spread-of-covid-19 (accessed April 19, 2020).

Ministry of Health, Singapore. Continuation of essential healthcare services during period of heightened safe distancing measures. 2020b, April 4. https://www.moh.gov.sg/news-highlights/details/continuation-of-essential-healthcare-services-during-period-of-heightened-safe-distancing-measures (accessed April 19, 2020).

Ministry of Health, Singapore. Risk assessment raised to DORSCON Orange.

2020c, February 7. https://www.moh.gov.sg/news-highlights/details/riskassessment-raised-to-dorscon-orange (accessed April 19, 2020).

National Health Service, United Kingdom. Clinical guide for the management of trauma and orthopaedic patients during the coronavirus pandemic. April 14, 2020 Version 2. https://www.england.nhs.uk/coronavirus/wp-content/ uploads/sites/52/2020/03/C0274-Specialty-guide-Orthopaedic-traumav2-14-April.pdf (accessed April 19, 2020).

O’Donnell J. Elective surgeries continue at some US hospitals during coronavirus outbreak despite supply and safety worries. USA Today (Internet).

March 21, 2020. https://www.usatoday.com/story/news/health/2020/03/21/ hospitals-doing-elective-surgery-despite-covid-19-risk-short-supplies/2881141001/ (accessed April 19, 2020).

OECD. Hip and knee replacement. In: Health at a glance 2017: OECD indicators. Paris: OECD Publishing; 2017.

Regional Office for Europe, World Health Organization. Strengthening the health systems response to COVID-19. Technical working guidance

Acta Orthopaedica 2020; 91 (5): 551â&#x20AC;&#x201C;555

#1: Maintaining the delivery of essential health care services freeing up resources for the COVID-19 response while mobilizing the health workforce for the COVID-19 response. April 18, 2020. http://www.euro.who. int/__data/assets/pdf_file/0007/436354/strengthening-health-systemsresponse-COVID-19-technical-guidance-1.pdf (accessed May 11, 2020).

Sarac N J, Sarac B A, Schoenbrunner A R, Janis J E, Harrison R K, Phieffer L S, Quatman C E, Ly T V. A review of state guidelines for elective orthopaedic procedures during the COVID-19 outbreak. J Bone Joint Surg Am April 13, 2020. [Epub ahead of print] Schmidt T. American Academy of Orthopaedic Surgeons. Piedmont elective surgery algorithm: March 15, 2020 (accessed April 19, 2020).

555

Sloan M, Premkumar A, Sheth N P. Projected volume of primary total joint arthroplasty in the U.S., 2014 to 2030. J Bone Joint Surg Am 2018; 100: 1455-60.

Soh T L T, Ho S W L, Yap W M Q, Oh J Y. Spine surgery and COVID-19: challenges and strategies from the front lines. J Bone Joint Surg Am 2020; 00 e1(1-3).

Vannabouathong C, Devji T, Ekhtiari S, Chang Y, Phillips S A, Zhu M, Chagla Z, Main C, Bhandari M. Novel Coronavirus COVID-19: current evidence and evolving strategies. J Bone Joint Surg Am 2020: 00: e1(1-11).

World Health Organization. Pandemic influenza preparedness and response: a WHO guidance document. World Health Organization; 2010.

556

Acta Orthopaedica 2020; 91 (5): 556–561

Impact of the COVID-19 pandemic on orthopedic trauma workload in a London level 1 trauma center: the “golden month” The COVid Emergency Related Trauma and orthopaedics (COVERT) Collaborative Chang PARK a, Kapil SUGAND a, Dinesh NATHWANI, Rajarshi BHATTACHARYA, and Khaled M SARRAF

Imperial College Healthcare NHS Trust & North West London Major Trauma Centre, London, UK a Shared first authorship Correspondence: ks704@ic.ac.uk Submitted 2020-05-15. Accepted 2020-06-02

Background and purpose — The COVID-19 pandemic has been recognized as an unprecedented global health crisis.

This is the first observational study to evaluate its impact on the orthopedic workload in a London level 1 trauma center (i.e., a major trauma center [MTC]) before (2019) and during (2020) the “golden month” post-COVID-19 lockdown.

Patients and methods — We performed a longitudinal observational prevalence study of both acute orthopedic trauma referrals, operative and anesthetic casemix for the first “golden” month from March 17, 2020. We compared the data with the same period in 2019. Statistical analyses included median (median absolute deviation), risk and odds ratios, as well as Fisher’s exact test to calculate the statistical significance, set at p ≤ 0.05.

Results — Acute trauma referrals in the post-COVID period were almost halved compared with 2019, with similar distribution between pediatric and adult patients, requiring a significant 19% more admissions (RR 1.3, OR 2.6, p = 0.003). Hip fractures and polytrauma cases accounted for an additional 11% of the modal number of injuries in 2020, but with 19% reduction in isolated limb injuries that were modal in 2019. Total operative cases fell by a third during the COVID-19 outbreak. There was a decrease of 14% (RR 0.85, OR 0.20, p = 0.006) in aerosol-generating anesthetic techniques used.

Interpretation — The impact of the COVID-19 pandemic has led to a decline in the number of acute trauma referrals, admissions (but increased risk and odds ratio), operations, and aerosolizing anesthetic procedures since implementing social distancing and lockdown measures during the “golden month.”

The global impact of COVID-19 The novel coronavirus SARS-COV-2 (COVID-19) was first reported in December 2019 with the first patient hospitalized in the city of Wuhan, China (Wu et al. 2020). By mid-March 2020 the outbreak affected over 190 countries with over 450,000 cases and over 20,000 deaths, thus being declared a pandemic and a global public health emergency by the World Health Organization (2020). On January 24, 2020 Europe reported its first case followed by a case in the United Kingdom (UK) 5 days later (Spiteri et al. 2020). Such a pandemic is an unprecedented event, and governments have had to enact firm social distancing and lockdown measures in an attempt to mitigate further viral transmission (Anderson et al. 2020) in order to reduce morbidity and mortality.

British response to the pandemic The English government responded by implementing social distancing measures on the March 16, 2020 in an attempt to reduce the rate of transmission and therefore the demands on the National Health Service (UK Government 2020a). This was followed a week later by more stringent measures, commonly referred to as a societal “lockdown” (UK Government 2020b). As of March 23, 2020, all members of the public were required to stay at home except for limited purposes and this ruling received Royal Assent by March 26 within the rest of the UK. Furthermore, all public gatherings of more than 2 people and non-essential businesses were suspended. In response to the NHS emergency declaration (National Health Service England 2020), the Royal College of Surgeons (2020) and the British Orthopaedic Association (2020) both issued statements and guidelines for delivering emergency trauma and orthopedic care during the COVID-19 outbreak. The phenomenon of a reduction in trauma burden due to such social distancing measures has been described by Stinner et al.

© 2020 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 10.1080/17453674.2020.1783621

Acta Orthopaedica 2020; 91 (5): 556–561

(2020), as well as the potential impact of COVID-19 on operative capacity and pathways. There has been little to explore on how COVID-19 affects the etiology of trauma referral workloads and the operative casemix.

We evaluated the impact of the COVID-19 pandemic at a central London level 1 trauma center, also known as a Major Trauma Centre (MTC), evaluating the trends of acute orthopedic trauma referral caseload and operative casemix before (2019) and during (2020) the COVID-19 lockdown (i.e., the “golden” month period starting from March 17).

Methods Patient sampling All acute referrals, operative notes, inpatient medical records, and discharge summaries were accessed using the electronic medical system.

Study period The study period was from the start of social distancing on the morning of March 17, 2020 to April 15, 2020, which also encompasses the morning following more firm “lockdown” measures on March 24, 2020. This was compared with the same 4-week interval in March–April 2019 prior to any COVID-19 related measures to compare its impact 1 year apart.

Inclusion criteria The study criteria comprised all acute orthopedic trauma referrals presenting to the Emergency Department of a busy level 1 trauma center (North West London Major Trauma Centre, UK) during the stated intervals 1 year apart, and all orthopedic trauma cases that required an operation, including those from acute orthopedic trauma referrals, within the stated intervals 1 year apart. Those patients listed for an operation prior to the period of data collection were included in the final analysis. We adhered to the STROBE guidelines.

Exclusion criteria Any cases referred internally from other specialties for trauma and orthopedic advice and input, as well as referrals from any external center asking for tertiary advice, were excluded from further analysis. Any patient with a postoperative complication arising in the period prior to the data collection were excluded. For operative trauma cases, those undergoing spinal procedures were excluded as the service is delivered jointly by the neurosurgery service. With respect to infections, all acute and chronic surgical site infections (SSI) and non-SSIs were excluded from the final analysis. All non-urgent semi-elective procedures were excluded from analysis as well, to avoid inaccurate assessment of the impact of any social distancing measures on trauma workloads. Routine elective orthopedic cases were excluded.

557

Pre-COVID referrals n = 193

Post-COVID referrals n = 94

Excluded n = 31

Excluded n=7 Included n = 87

Included n = 162 Age (n = 162): – child, 27 – adult, 135

Sex (n = 162): – female, 72 – male, 90

Age (n = 87): – child, 12 – adult, 75

Sex (n = 87): – female, 43 – male, 44

Demographic data pre- and post-COVID for acute referrals.

Data points Demographics including age, sex, and ASA grades were recorded for all patients. Injury characteristics were recorded, including the anatomical location and whether the injury was open or closed. The mechanism of injury was categorized and whether the patient presented as a trauma call. The nature of the operative procedures and the anesthetic techniques were recorded. Patients undergoing multiple procedures were recorded for every episode when they were taken to theatre.

Statistics All the data were recorded, anonymized, and verified by 2 authors for their accuracy. The median (median absolute deviation) was calculated for both age and ASA grade. Both risk and odds ratios were calculated as well as a Fisher’s exact test for statistical significance, defined as p ≤ 0.05.

Ethics, funding, and potential conflicts of interest No formal ethical approval was required as these were audit data. The study was registered and approved with the Trust’s audit department. No identifiable patient data have been kept or reported. This study required no internal or external funding. The authors have no conflict of interests to declare.

Results There were no missing data, as all data points were extracted from electronic patient records (Figure).

Pre-COVID era For the pre-COVID period in 2019 there were 193 new referrals. 106 (55%) were male. 31 patients were excluded, which left 162 patients in the pre-COVID period who were acutely referred. 90 (56%) were male. 135 (83%) patients were adults (> 18 years old) (Figure).

Post-COVID era For the post-COVID period in 2020 there were 94 referrals (53% of those in 2019). Sex was split equally.

558

Acta Orthopaedica 2020; 91 (5): 556–561

Table 1. Referrals between pre- and post-COVID

Acute trauma referrals Pre-COVID Post-COVID n = 162 n = 87

Adults Paediatric Pre-COVID Post-COVID Pre-COVID Post-COVID n = 135 n = 75 n = 27 n = 12

Demographic Male 90 43 73 38 17 6 Female 72 43 62 37 10 6 Age a 47 (26) 50 (24) 54 (21) 56 (23) 9 (4) 9.5 (5.5) ASA a 1 (0) 2 (1) 2 (1) 2 (1) 1 (0) 1 (0) Injury Upper limb 49 22 34 14 15 8 Lower limb 54 22 47 20 7 2 Hip 14 17 14 17 0 0

Pelvis 8 2 7 2 1 0 Polytrauma 18 17 18 1 0 0 Infection 15 7 12 5 3 2

Other 4 0 7 0 1 0 Mechanism of injury Sporting 18 2 11 2 7 0 Fall 80 50 67 44 14 6 Fall from height > 1.5 m 8 9 8 8 0 1 Road traffic collision 25 13 24 12 1 1

Pathological 6 1 3 0 0 1 Other 25 12 20 9 5 3 Open injury 23 17 21 15 2 2 Trauma call 44 25 42 24 2 1 Operative requirement 76 48 66 43 10 5 a Median

and (median absolute deviation).

ASA: American Society of Anesthesiologists.

Demographics 7 patients were excluded, which left 87 new acute trauma referrals in the post-COVID period (Figure). 44 (51%) patients were male. 75 (86%) patients were adults.

Results have been tabulated as acute referrals, categorized as all referrals, adult referrals, and pediatric referrals between the 2 years (Table 1). Table 2 reflects the operative casemix.

Prevalence, risk, and odds ratios Table 3 outlines the prevalence and prevalence odds ratios alongside their 95% confidence intervals (CI) and statistical significance. The risk ratio is synonymous with the prevalence ratio. There was no statistically significant difference in the number of trauma calls and adult versus pediatric acute trauma referrals between the 2 years. On closer inspection, even though just over half the acute orthopedic trauma referrals were made in the post-COVID period, there was a greater proportion of acute presentations referred, a 30% (RR 1.3, CI 1.1–1.5) increased prevalence of admission with the odds of admission increased by 156% (OR 2.6, CI 1.4– 4.7). Hence, the threshold for referral was much lower and these patients were more in need of inpatient medical care in spite of admission of trauma patients being discouraged to reduce viral transmission and to uphold patient safety.

Although a greater number of the acute referrals required surgery, this was ultimately not statistically significant (RR 1.2, CI 0.9–1.5, OR 1.4, CI 0.8–2.4). Nevertheless, if a patient

did require surgery during the COVID outbreak, there was a 131% (OR 2.3, CI 0.99–5.4) increased odds that the operation would be consultant-led (either as primary surgeon or scrubbed in to supervise, as opposed to being unscrubbed) with a 19% increased prevalence of personal involvement compared with 2019. As expected, every attempt was made to minimize reliance on aerosolizing anesthetic procedures wherever possible in order to reduce viral transmission, load, and exposure to the healthcare staff. In 2020, COVID significantly decreased the prevalence of GA (±spinal) by 15% (RR 0.85, CI 0.75–0.96) as well as decreasing the odds of receiving an aerosolized anesthetic procedure by 80% (OR 0.20, CI 0.06–0.65).

Discussion A shift in clinical practice There was a notable difference between the number of acute referrals and the operative casemix between the time intervals 1 year apart pre- and post-COVID in a London level 1 trauma center. There was a substantial decrease in the number of relevant acute trauma referrals (without a statistically significant difference between age and sex but significantly fewer sporting injuries), number of operations (without significant difference between mechanisms of injury and type of surgery or technique) with a lower number of aerosolizing anesthetic procedures (with significantly less risk and odds ratios). This

Acta Orthopaedica 2020; 91 (5): 556–561

559

Effect of local, regional, and nationwide service reconfiguration Post-COVID there has also been a reduction in external referrals as compared with 2019, representing the effects of the major service reconfiguration with disbanding of the elective, private, and inpatient practice to cater for the increased space requirement to host acute COVID patients pre- and post-ITU treatment. To reduce the risk of virus transmission and reduce the demands on services, some injuries previously treated at the level 1 trauma center were being treated at level 2 trauma units, a change of practice supported by the British Orthopaedic Association (2020). There are 22 equivalent level 1 trauma centers in England out of 152 acute specialist trusts with a further 5 pediatric specific units (National Health Service England 2016). Within London there are 4 level 1 trauma centers of which our center is 1 of the largest. In the pre-COVID period, level 1 trauma centers would exclusively provide care for polytrauma, complex and open injuries. However, these may now be expected to be managed at smaller level 2–4 trauma units (i.e., district general hospitals [DGH]) as highlighted by Morgan et al. (2020). Some injuries such as Gustilo-Anderson type 3 injuries, polytrauma, complex intra-articular fractures, and those requiring cross-specialty expertise from plastic and vascular surgeons, will continue to require treatment at a level 1 trauma center as the specialist skill may be unavailable at many level 2 trauma units. However, as level 1 trauma centers are present in less than a fifth of all English acute NHS trusts, the COVID pandemic may alter the expectations and the role of level 1 trauma centers in the future, especially in the treatment of those injuries not requiring cross-specialist input or complex management in the first instance.

Table 2. Operative trauma casemix between pre- and post-COVID

Operative trauma cases only Pre-COVID Post-COVID n = 90 n = 63

Demographic Male 49 Female 41 Age a 43.5 (19) ASA a 1 (0) Injury Upper limb 21 Lower limb 30 Hip 16 Pelvis 2 Polytrauma 17 Infection 2 Other 2 Mechanism of injury Sporting 7 Fall 36 Fall from height > 1.5 m 4 Road traffic collision 25 Pathological 2 Other 6 Open injury 28 Trauma call 36 Operation Total 91 MUA 4 External fixator 7 Frame 1 Removal of metal 2 Soft tissues/other 12 Percutaneous wiring 2 ORIF 34 Intramedullary device 16 Dynamic hip screw 8 Hemiarthroplasty 5 Anaesthetic method General anaesthesi (GA) 78 Spinal 3 GA + spinal 7 Block 0 Local 2

39 24 50 (20) 2 (1) 12 12 14 1 16 5 3 0 33 6 15 0 9 20 28 67 3 5 2 1 10 2 23 13 4 4 46 10 5 2 0

a Median

and (median absolute deviation), ASA: American Society of Anesthesiologists, MUA: manipulation under anaesthesia.

reduction is likely to have been a direct consequence of the social distancing measures implemented on a national scale.

Demographic and injury pattern The comparison of demographics of acute trauma referrals is mostly similar between the 2 periods, as seen in Table 1. There is a near equal split in sex in 2019 and an exact split in 2020.

Similarly, in 2019, 83% of referrals were adults compared with 86% in 2020, with a higher median ASA grade (2) to signify sicker patients.

Pre-COVID in 2019, the most common injury pattern for acute referrals was lower limb injuries at 33%. This was followed by upper limb injuries (30%) and together they

Table 3. Risk and Odds ratios (95% CI)

Acute referrals requiring admission Acute referrals requiring surgery Consultant-led operations Operations requiring GA (± spinal) Adult vs paediatric acute referrals Trauma calls Sporting injuries from acute referrals

Pre vs Post COVID RR OR 1.3 (1.1–1.5) 1.2 (0.9–1.5) 1.2 (1.0–1.4) 0.9 (0.8–1.0) 1.0 (0.9–1.2) 1.0 (0.6–1.5) 0.2 (0.1–0.9)

2.6 (1.4–4.7) 1.4 (0.8–2.4) 2.3 (1.0–5.4) 0.2 (0.1–0.7) 1.3 (0.6–2.6) 1.0 (0.5–1.7) 0.2 (0.0–0.8)

Fisher’s p-value 0.003 0.2 0.05 0.006 0.6 1 0.01

560

accounted for nearly two-thirds of all referrals. Post-COVID, both upper and lower limb injuries are still the most common injury but combined they accounted for just 50% of all referrals. Hip fractures and polytrauma (often from road traffic collisions and high-energy injuries) patients, however, accounted for an increased proportion of acute referrals, each accounting for 19% of cases.

Hip fractures (HF) Usually the result of low-energy falls, HFs often occur indoors, in the garden, and within the property, and the incidence may not be directly impacted by the social distancing measures that were implemented. Therefore, it would be expected that the number of HF referrals would be consistent, as demonstrated by 17 HF referrals in 2020, an increase from the 14 in 2019.

HF referrals reflected a higher percentage of all acute referrals during the pandemic study period when compared with 2019 (i.e., 19% vs. 9%). Yet there was no significant change in risk or odds ratios of HF between the 2 years.

Polytrauma Conversely, polytrauma occurs as a consequence of highenergy injuries. This may often be as a result of a road traffic collision (RTC) or fall from a height greater than 1.5 m. RTC rates have not changed significantly following social isolation advice and may account for some of the greater proportion of polytrauma observed. There has been an increase in those falling from a height greater than 1.5 m in 2020 of 10% as compared with 5% in 2019. This could be due to the construction industry being exempt from the lockdown or people spending more time at home committing to home improvements or doit-yourself tasks.

Mechanism of injury (MOI) The breakdown of the mechanism of injury has remained largely the same pre- and post-COVID as seen in Tables 1–2. A fall from less than a 1.5 m height still accounts for the majority of cases in 2020 at 58%, compared with 49% in 2019. Nevertheless patients, especially the geriatric community, will continue to suffer from low-energy falls despite the social isolation, be it within their homes, from simple falls and trips, and this may explain the overall consistency.

Sporting injuries There has been an 89% reduction in acute referrals due to sporting injuries and a 100% drop in the MOI in the operative casemix 1 year apart (Table 2). This would be expected, as all group activities have been banned following social isolation and gyms are closed to reduce the risk of viral transmission. This significant decrease may represent the main etiology of the reduction in trauma referrals seen between the 2 periods (RR 0.21, CI 0.05–0.87; OR 0.2, CI 0.04–0.83), which would correlate closely with the government’s advice post-COVID.

Acta Orthopaedica 2020; 91 (5): 556–561

Road traffic collisions (RTCs) Despite the significant reduction in personal vehicle use, there has been a consistent proportion of injuries seen following social isolation attributable to RTCs (Table 1) at 15% (n = 13) post-COVID compared with 15% in 2019. Although the roads are quieter, there have been concerns that, paradoxically, fewer vehicles may result in more RTCs due to speeding. The Metropolitan Police have described an increase incidence of speeding, with average speeds of 37 mph in some 20 mph zones, following social isolation (British Broadcasting Corporation 2020). Although the data represent a 40% decrease in the number of patients being admitted following an RTC, nevertheless this is some way from the total reduction of road use estimate of 70% (Carrington 2020).

Operative cases Accounting for all exclusions, the total number of operative cases has fallen by a third following COVID. A proportion of this is due to the ceasing of semi-elective operating. However, the reality of practice at our center is that the management of orthopedic injuries during the COVID era has not changed significantly. Non-urgent and elective procedures have been cancelled or postponed following national advice (National Health Service England 2020), but the decision to offer operative intervention is still, first and foremost, a decision based on clinical need, balancing risk and benefit to the patient. The key driving force behind the overall reduction in operative procedures performed in 2020 is the reduction in referral volume (Table 2) and not due to an altered threshold of operative intervention. Indeed, we currently do not anticipate any fracture complication or secondary intervention required as being directly due to any altered management decision during the COVID period.

Anesthetic choice There was a preference for performing non-aerosol generating procedures (AGPs) as the anesthetic methods in 2020 (19%) compared with 2019 (6%) as seen in Tables 2 and 3. There has been evidence that AGPs such as intubation for a GA increases the risk of healthcare work transmission with an increased viral load (Vannabouathong et al. 2020). As such, this change may represent a shift in an attempt to mitigate this risk. In order to avoid AGPs in the theatre setting, patients are encouraged to consent and agree to regional blocks including spinals, which in themselves also take longer to perform and have an effect compared with intubation for GA. Whereas 87% of total patients operated on pre-COVID had a GA, this was reduced to 73% post-COVID with an increase in regional blocks from 3% to 16% among all patients during the COVID period.

Limitations and future studies The limitations of this longitudinal observational study include analyzing 2x4-week snapshots 1 year apart, at a single-center London level 1 trauma center. This may not be representative of the national profile. There has been much literature on the

Acta Orthopaedica 2020; 91 (5): 556–561

benefits of large-volume centralization of orthopedic trauma to level 1 trauma centers, but this might need to be re-analyzed if smaller peripheral trauma units return to managing grade 1–2 Gustilo Anderson open fractures and simpler closed polytrauma. Further work is required to observe for trends in acute orthopedic referrals and orthopedic trauma surgical casemix as a result of the structural reconfiguration due to COVID. Bias was kept to a minimum and the date range between the two years was dictated by the evolution of the pandemic.

Conclusion The COVID-19 pandemic has had a unique impact on trauma and orthopedic care. Acute trauma referral rates have fallen, with fewer trauma procedures being performed since the implementation of the UK social distancing measures indicating a change in prevalence pre- and post-COVID. Every attempt has been made to substantially reduce the prevalence of aerosol-generating anesthetic procedures with general anesthesia and intubation. We recommend more work to investigate the phenomenon further and whether a similar pattern is seen across the UK. Acta thanks Karin Bernhoff and Minna K Laitinen for help with peer review of this study.

Anderson R M, Heesterbeek H, Klinkenberg D, Hollingsworth T D. How will country-based mitigation measures influence the course of the COVID-19 epidemic? Lancet 2020; 395: 931-4.

British Broadcasting Corporation. Coronavirus: Empty-roads speeding may impact NHS, drivers warned. April 2, 2020. www.bbc.co.uk/news/av/ uk-england-london-52141678/coronavirus-empty-roads-speeding-mayimpact-nhs-drivers-warned.

British Orthopaedic Association. British Orthopaedic Association for Standards of Trauma (BOAST). Management of patients with urgent orthopaedic conditions and trauma during the coronavirus pandemic. March 24, 2020. www.boa.ac.uk/resources/statement-for-boa-members-on-traumaand-orthopaedic-care-in-the-uk-during-coronavirus-pandemic.html.

561

Carrington D. UK road travel falls to 1955 levels as Covid-19 lockdown takes hold. Guardian, April 3, 2020. www.theguardian.com/uk-news/2020/ apr/03/uk-road-travel-falls-to-1955-levels-as-covid-19-lockdowntakes-hold-coronavirus-traffic.

Morgan C, Ahluwalia A K, Aframian A, Li L, Sun S N M. The impact of the novel coronavirus on trauma and orthopaedics in the UK. Br J Hosp Med (Lond) 2020; 81(4): 1-6.

National Health Service England. Major trauma centres in England. NHS, October 2016. www.nhs.uk/NHSEngland/AboutNHSservices/Emergencyandurgentcareservices/Documents/2016/MTS-map.pdf.

National Health Service England. Next steps on NHS response to COVID-19: Letter from Sir Simon Stevens and Amanda Pritchard. March 17, 2020. www.england.nhs.uk/coronavirus/publication/next-steps-on-nhs-responseto-covid-19-letter-from-simon-stevens-and-amanda-pritchard.

Royal College of Surgeons. Guidance for surgeons working during the COVID-19 pandemic from the Surgical Royal Colleges of the United Kingdom and Ireland. March 20, 2020. www.rcseng.ac.uk/coronavirus/ joint-guidance-for-surgeons-v1.

Spiteri G, Fielding J, Diercke M, Campese C, Enouf V, Gaymard A, et al.

First cases of coronavirus disease 2019 (COVID-19) in the WHO European Region, 24 January to 21 February 2020. Eurosurveillance 2020 (25): 2000178.

Stinner D J, Lebrun C, Hsu J R, Jahangir A A, Mir H R. The orthopaedic trauma service and COVID-19: practice considerations to optimize outcomes and limit exposure. J Orthop Trauma 2020; April 13: 10. 1097/ BOT.0000000000001782.

UK Government. Guidance: Oral statement to and away from others (social distancing). March 23, 2020a. www.gov.uk/government/publications/fullguidance-on-staying-at-home-and-away-from-others.

UK Government. Oral statement to Parliament. Controlling the spread of COVID-19: Health Secretary’s statement to Parliament. March 16, 2020b. www.gov.uk/government/speeches/controlling-the-spread-of-covid19-health-secretarys-statement-to-parliament.

Vannabouathong C, Devji T, Ekhtiari S, Chang Y, Phillips S A, Zhu M, et al.

Novel coronavirus COVID-19: current evidence and evolving strategies. J Bone Joint Surg 2020; April 2.

World Health Organization. Coronavirus disease 2019 (COVID-19): situation report, 66. 26 March 2020. https://www.who.int/docs/defaultsource/coronaviruse/situation-reports/20200326-sitrep-66-covid-19. pdf?sfvrsn=9e5b8b48_2.

Wu F, Zhao S, Yu B, Chen Y M, Wang W, Song Z G, et al. A new coronavirus associated with human respiratory disease in China. Nature 2020; 579: 265-9.

562

Acta Orthopaedica 2020; 91 (5): 562–566

Perspective

Challenges and adaptations in training during pandemic COVID-19: observations by an orthopedic resident in Singapore Wei-Sheng FOONG, H L Terry TEO, D H Bryan WANG, and S Y James LOH

Department of Orthopaedic Surgery, Changi General Hospital, Singapore Correspondence: weisheng.foong@mohh.com.sg Submitted 2020-05-06. Accepted 2020-06-16.

Orthopedic surgery training in Singapore is provided by 3 Accreditation Council for Graduate Medical Education International (ACGME-I) accredited residency programs. Mandatory clinical rotations in general surgery, plastic surgery, and intensive care in a foundation year (R-1) and followed by subspecialties within a 5-year program are requirements for graduation.

In January 2020, China confirmed its first case of COVID-19 coronavirus case in Wuhan (Chen et al. 2020). This occurred 2 months before the World Health Organization (WHO) announced the disease as a global health crisis. The pandemic has affected the healthcare industry in Singapore in a myriad of ways. There was an exponential increase in the usage of personal protection devices such as masks and gloves. Measures were taken to triage surgeries to control bed usage, such as high-dependency beds, as well as to stockpile crucial anesthetic drugs in the event of a tsunami of patients who needed intensive care.

The impact of this fast-evolving and widespread pandemic has since affected all continents with the exception of Antarctica. The various control measures (ECDC 2020) implemented have changed the routines of healthcare provision.

This has inadvertently affected residency training in orthopedic surgery. Several programs (Amparore et al. 2020, Nassar et al. 2020) worldwide have undertaken measures to ensure the continuity of residency training with minimal disruption.

Similarly, my (“my/I” refers from now on to author W-S F) training program has made innovative changes (Schwartz et al. 2020) and adaptations with utmost emphasis on training with all the necessary safety precautions.

The following account is based on the views and observations of an individual and does not reflect that of any specific training program or organization.

ACGME-I Orthopedic Surgery Residency Program: components and competencies ACGME-I is guided by 3 main administrative components inclusive of clinical experience, didactic teaching, and duty

hours. 6 core competencies have been identified to guide the design of a well-balanced surgical training program. These competencies include patient care, knowledge, systems-based practice, professionalism, practice-based learning, and communication skills (ACGME 2020a, 2020b).

The 3 core components Clinical experience COVID-19 spreads predominantly via aerosol droplets and contaminated surfaces (Viswanath and Monga 2020). This dictates that the healthcare provider should use various personal protection devices such as mask and gloves, coupled with strict hand hygiene. The mitigation measures (Sahu et al. 2020) also advised reduction of doctor and patient contact.

A direct consequence of this measure is the postponement of non-essential surgeries and clinic appointments, resulting in a decline in clinical exposure in these areas. In spite of these measures, essential orthopedic surgeries such as fractures and spine conditions with neurological compromise continued to proceed.

Surgery The orthopedic resident is expected to participate in 200 (ACGME 2020a) surgeries in a year. Over a period of 3 months, from February to April 2020, there was a steady and significant decline in the number of elective arthroscopic and arthroplasty surgeries. The decline in the number of listed surgeries is possibly the result of a lower demand in non-emergency conditions and advisories to postpone non-essential elective surgeries. This decline correlates inversely with the sharp climb in the number of positive COVID-19 cases reported over the same period. Figure 1 details the difference in total number of cases for the same period of 3 months from February through April in the years 2019 and 2020. It compares my experience during this period with a fellow resident under the same supervising attending consultant in 2019. There was an overall drop of 9% in the total number of surgeries.

© 2020 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 10.1080/17453674.2020.1786641

Acta Orthopaedica 2020; 91 (5): 562â&#x20AC;&#x201C;566

563

Trauma/infection surgeries

Elective surgeries

New referral attendances

40

8

2,000

35

7

30

6

25

5

20

4

15

3

10

2

5 0

2019 2020

February

2019 2020

5,000 1,500 4,000

April

Figure 1. Number of trauma/infection surgeries from February to April, 2019 and 2020.

0

3,000

1,000

2,000 500 1,000

2019 2020

1

March

Follow-up attendances 6,000

February

March

April

Figure 2. Number of elective surgeries from February to April, 2019 and 2020.

The surgeries were categorized into tiers 1, 2. and 3.

These are vetted by the Operating Theatre Management Unit (OTMU). Tier 1 is for life-threatening conditions such as tumor surgery. Fractures and musculoskeletal infections are in tier 2. Elective orthopedic surgeries such as knee replacement and knee arthroscopy are categorized as tier 3.

There was a substantial decrease in elective surgeries by 93% and a corresponding 5% for trauma/infection surgeries (Figure 2). This is a great reduction in the exposure to elective surgery. However, the trauma and infection surgeries continued at a similar rate compared with the same period in 2019 (Figure 3). This enabled my continued training in these subspecialties. For a junior resident, this exposure is suitable.

However, the same situation would not benefit a senior resident who had already completed his or her rotation in these subspecialties. The time needed to observe properly the gowning process, personnel movement before and after patient intubation, and the actual surgery itself using N95 masks, and a powered air purifying respirator (PAPR) are new challenges to the surgical team. Specialists were expected to perform surgery on positively identified or high-risk patients to minimize operative time and reduction of disease transmission. Hence in spite of the lower surgical caseload and reduced time constraints, proper patient care and appropriate risk mitigation at this juncture still supersedes my opportunity for hands-on surgical training. When an appropriate surgery for training is identified and this is triaged as a low-risk case with no history or signs of the disease, I took a meticulous approach to preoperative planning and postoperative management. These were aspects of surgery that were not emphasized as strongly in busier times.

Specialist outpatient clinic The trend in clinic attendance reflects a similar decline to that of elective surgeries. There is an overall decline in the number of new referrals and follow-up patients across the same period in 2019 and 2020. Long-term follow-up of patients at 6 months to 1 year were deferred after the diligent screening of their medical histories. There has also been a significant number of no-shows. New referrals were further categorized

0

0

February

March

April

Figure 3. Number of new referral clinic attendances from February to April, 2019 and 2020.

2019 2020

February

March

April

Figure 4. Number of follow-up clinic attendance from February to April, 2019 and 2020.

into trauma/infection and elective conditions. The numbers of trauma/infectious conditions requiring surgery, conversely, did not mirror the decline in clinic attendances (Figures 3 and 4) and continued to serve the needs of my residency training.

This is aligned with the fact that orthopedic surgery continues to provide an essential service during a pandemic, dealing with fractures and infectious conditions.

Should this pandemic persist, it could lead to an unavoidable skew in my clinical exposure towards trauma and musculoskeletal infections and away from arthroplasty and arthroscopy conditions.

Inpatient When attending to a new admission, it would be necessary to exclude the possibility of a concomitant COVID-positive infection. There exists the need to consistently update oneself on the evolving screening criteria such as travel history, contact history, symptoms of acute respiratory infection, and fever. The advised level of precautionary measure needs to be observed as this prevents transmission of infection between patients and healthcare workers. The use of N95 masks and PAPR is reinforced for suspected and confirmed cases. A high index of suspicion is needed as the history of the patient might be unreliable and there are reported cases of â&#x20AC;&#x153;cleanâ&#x20AC;? surgical patients who were diagnosed COVID-positive after surgery was performed.

Team segregation entailed reorganizing the department staff into more teams. This change in manpower structure stretched the work capacity of junior doctors. Segregation of doctors into smaller teams to minimize physical contact mitigates risk of transmission. Advantages include facilitation of contact tracing and ability to quarantine smaller groups of healthcare providers in the event of disease transmission. Adapting to these changes, I personally made multiple check-backs on treatment instructions to ensure that patient care is not compromised.

Didactic teaching Orthopedic residency requirements (ACGME 2020a) dictate 4 hours a week of didactic teaching sessions on various topics

564

Acta Orthopaedica 2020; 91 (5): 562â&#x20AC;&#x201C;566

and enabled robust discussions. The schedule is published in advance to enable preparation and subsequent effective discussion.

However, there is a need to circumvent challenges such as speaker inaudibility caused by speaking distance from microphones, statics, and muffling by masks. Having participated in these early sessions, I feel that an interactive educational platform enhances the learning process as opposed to broadcast didactic lectures.

Figure 5. Resident conducting an interactive tutorial via teleconferencing.

Figure 6. Resident conducting interactive session with supervising consultant.

Residency scholarly activities The orthopedic resident is required to complete a series of mandatory courses locally.

Training resources such as cadaveric arthroplasty (James et al. 2020) and arthroscopic surgery courses that require physical participation had to cease. Conferences worldwide have been postponed and the opportunities to make presentations are now void. Various conference organizers and scientific groups have made learning possible via alternative options such as webinars (AO Foundation 2020). Research not involving patient contact such as cadaver-based studies is an option if the resource is available. The reduction in workload potentially frees up the faculty to mentor activities such as research and fellowship applications.

Duty hours The manpower demand is erratic and thus carries the risk of exceeding the stipulated limit on duty hours in a working week. This Figure 7. Residentâ&#x20AC;&#x2122;s work week schedule in April 2019 (left panel) and 2020 (right paenl). situation was avoided due to close monitoring of the duty roster and strict adherence to such as operative techniques, anatomy, pathology, biome- the residency guidelines by a dedicated program director.

Figure 7 illustrate an increase in overnight duties (shown in chanics, and radiography. Conferences, workshops, didactic lectures, trauma rounds, and peer-reviewed learning sessions red) to compensate for colleagues deployed to the frontline. contribute to this weekly requirement. These teaching ses- There is an overall decrease in the number of teaching sessions involve a robust interaction between the residents and sions (in yellow). Scheduled teaching hours once a week are faculty. The need for social distancing has curtailed these still maintained, albeit with online teaching resources. Allomeetings. Teleconferencing (Denstadli et al. 2012) in medi- cated time for surgery (in magenta) remained unaffected.â&#x20AC;&#x192; cine (Lamba 2019) is one such measure that enables interactive sessions to be conducted (Figure 5). Third-party software The 6 competencies enables physical distancing and the broadcasting of clinical 1. Patient care cases for discussion (Figure 6). There are many teething prob- New guidelines (WHO 2020) for personal protection and surlems in the setting up of this form of communication to meet veillance of patients in both emergency and elective settings the demands of a large department on a daily basis. One of the are being implemented following WHO recommendations. main concerns is ensuring security of information and abid- Platforms for telemedicine consultation for non-urgent cases ing by the law in terms of the Personal Data Protection Act have received unprecedented attention and resources to boost (PDPA), which came into full effect in July 2014 in Singapore. their capability and capacity. This form of treatment needs to Such sessions have shown effective delivery of information be adopted promptly to provide an acceptable level of con-

Acta Orthopaedica 2020; 91 (5): 562–566

sultation without physical physician–patient contact, while bearing in mind the importance of personal data security and effective delivery of patient care.

The physical and mental health of the resident is potentially at risk (Kim et al. 2019). There is uncertainty in light of the tightening social constraints and no end in sight. The gnawing concern in relation to contracting COVID-19 in daily activities poses additional strain on pre-existing work stress. The Trauma Recovery and Counselling Services (TRACS) had raised awareness of burnout and provided counselling and stress-coping tools.

A recent survey conducted in a separate Singapore healthcare institution studied the experiences of its medical and surgical residents during the initial phase of this outbreak.

Residents reported adverse effects on their medical training and career. They also reported an increase in the level of stress and burnout, citing an average of 4.7 on a scale of 0 (no stress at all) to 10 (extreme level of stress) (Wong et al. 2020). Factors such as long hours away from family and partners, freezing of non-essential annual leave entitlement, and frequent shifts at work are major contributory factors.

Rest and recreation have always been crucial in preserving the work–life balance. Team-based (Brinkley et al. 2016) physical activities amongst colleagues have been shown to benefit cohesion and organizational performance. Weekly events such as football with fellow residents have now stopped. I maintain fitness and health via individual-based exercises such as running. The study on social distancing during running (Blocken et al. 2020) appears to be common knowledge to most runners.

2 and 4. Knowledge, practice-based learning Self-directed learning (Merriam 2001) is always an essential part of education. The resident needs to keep abreast of the evolving knowledge of COVID-19 disease such as its clinical course as well as management (Balla et al. 2020, Lauer et al. 2020). The intelligent sourcing and sharing of learning resources is beneficial to all involved. The same applies to orthopedic training. Resources such as the surgical skills laboratory (Sonnadara et al. 2011), cadaveric dissection, and arthroscopic simulation Gomoll et al. 2007) need to be explored. I have spent time in the surgical skills laboratory honing skills such as knot tying in shoulder arthroscopy.

3 and 5. Systems-based practice, interpersonal and communication skills Residents needs to work effectively in a multidisciplinary team approach such as deployment to the intensive care unit or the emergency medicine department. The grounding in anesthesia and emergency medicine in R-1 makes them more prepared than a colleague without such prior exposure. The need to communicate effectively is tested. I take this as an opportunity to assess and improve on interpersonal and communication skills, as these are definitely applicable in future.

565

6. Professionalism The core value of professionalism (Stern 2005) derives from the universality of disease and begins with caring and compassion. Residents make up a significant proportion of the medical community and are now called to action to support their colleagues at the frontline such as attending to intensive care patients. I need to handle the anxiety and uncertainty prevailing in the current environment and execute a treatment plan effectively, alongside compassion for the patient. Such a situation requires true professionalism and I gain insight into what it takes to do so.

The future Looking ahead to the near future, ramifications of the pandemic have started to impact on my career progression towards obtaining qualifications as a board-certified specialist. At the point of writing, the annually held qualification examination (Fellowship of the Royal College of Surgeons—Orthopaedic Surgery) had been announced to be postponed to a later date.

A direct result is a delay in career progression with repercussions of delayed employment in my institution and associated monetary and opportunity costs. The mid-term effects of COVID-19 in the year ahead have manifested as an extension in the total duration of surgical training.

The extension of training has significant psychological bearing on my outlook and morale, and several prior personal and professional engagements have suffered from this disruption.

On an optimistic note, the training program directives, mandatory competencies, and courses have come under urgent review to lessen the impact of these disruptions. A proportion of courses have been converted to web-based sessions. Utilizing a different delivery vehicle, these sessions continue to enable residents to participate in mandatory courses. A pragmatic approach taken via a two-way communication model enables residents and faculty to identify early potential obstacles and institute appropriate changes. There is no suggestion of relaxation in these segregation measures in the near future, thus it is hopeful that these changes will continue to evolve and serve the needs of my surgical training.

Conclusion The COVID-19 pandemic is undoubtedly challenging in many aspects and the measures taken to facilitate training will definitely find their place in a less turbulent time in the future.

There are learning points from my own experience. The ability of the program to navigate this pandemic is a credit to its robust foundation and more importantly its people. The determination of the faculty that adapted and innovated to continue the training is a reflection of their passion for education. The values of the six competencies are truly learnt and appreciated in such a trying situation. The pandemic is testing the resilience and fortitude of the orthopedic surgeon who rises to the occasion. I also see the part he can play as an orthopedic surgeon in the care of any patient, regardless of medical

566

pathology, and hopes to look back on this pandemic with a wiser perspective on medicine. As a senior once shared with me: “First a doctor, then a surgeon.”

Disclosure The authors have no conflicts of interest to declare. All authors received no form of funding.

Dr Dalun Leong (Resident year 2) posed for the photographs in Figures 5 and 6.

Acta thanks Anne Garland and Stefan Nehrer for help with peer review of this study.

Accreditation Council for Graduate Medical Education. Orthopaedic surgery.

2020a. https://www.acgme-i.org/Specialties/Landing-Page/pfcatid/16/ Orthopaedic-Surgery (accessed April 16, 2020).

Accreditation Council for Graduate Medical Education. ACGME common program requirement (residency). 2020b. https://www.acgme.org/Portals/0/PFAssets/ProgramRequirements/CPRResidency2019.pdf (accessed April 16, 2020).

Amparore D, Claps F, Cacciamani G, Esperto F, Fiori C, Liguori G, Serni S, Trombetta C, Carini M, Porpiglia F, Checcucci E, Campi R. Impact of the COVID-19 pandemic on urology residency training in Italy. Minerva Urol Nefrol 2020. [Ahead of print]doi: 10.23736/S0393-2249.20.03868-0.

AO Foundation. Trauma clinical library and tools. Videos and webinars. 2020. https://aotrauma.aofoundation.org/clinical-library-and-tools/videos-andwebinars (accessed April 13, 2020).

Blocken B, Malizia F, van Druenen T, Marchal T. Towards aerodynamically equivalent COVID-19 1.5m social distancing for walking and running.

2020. http://www.urbanphysics.net/Social%20Distancing%20v20_White_ Paper.pdf (accessed April 16, 2020). [Epub ahead of print] Balla M, Merugu G, Pokal M, Gayam Balla M, Merugu G, Pokal M, Gayam, et al. A comprehensive approach is vital for diagnosing COVID-19: a case of false negative. J Clini Med Res 2020; 12: 315-19. doi: 10.14740/ jocmr4173.

Brinkley A, McDermott H, Munir F. What benefits does team sport hold for the workplace? A systematic review. J Sports Sci 2016; 35: 1-13. doi: 10.1080/02640414.2016.1158852.

Chen N, Zhou MD, Dong X, et al. (2020). Epidemiological and clinical characteristics of 99 cases of 2019 novel coronavirus pneumonia in Wuhan, China: a descriptive study. Lancet 2020; 395(10223): 507-13. doi: 10.1016/ S0140-6736(20)30211-7.

Denstadli J, Julsrud T, Hjorthol R. Videoconferencing as a mode of communication: a comparative study of the use of videoconferencing and face-

Acta Orthopaedica 2020; 91 (5): 562–566

to-face meetings. J Bus Tech Commun 2012; 26: 65-91. doi: 10.1177/ 1050651911421125.

ECDC. Coronavirus disease 2019 (COVID-19) pandemic: increased transmission in the EU/EEA and the UK—seventh update, 25 March 2020.

Stockholm: ECDC; 2020. http://www.ecdc.europa.eu/sites/default/files/ documents/RRA-seventh-update-Outbreak-of-coronavirus-diseaseCOVID-19.pdf (accessed April 16, 2020).

Gomoll A H, O’Toole R V, Czarnecki J, Warner J J. Surgical experience correlates with performance on a virtual reality simulator for shoulder arthroscopy. Am J Sports Med 2007; 35: 883-8.

James H, Pattison G, Griffin D, Fisher J. How does cadaveric simulation influence learning in orthopedic residents. J Surg Educ 2020. doi: 10.1016/j. jsurg.2019.12.006.

Kim E, Mallett R, Hrabok M, Yang A Y, et al. Reducing burnout and promoting health and wellness amongst medical students, residents and physicians in Alberta: Study Protocol (Preprint). JMIR Res Protoc 2019; 9(4): e16285. doi: 10.2196/16285.

Lamba P. Teleconferencing in medical education: a useful tool. Australasian Med J 2019; 4: 442-7. doi: 10.4066/AMJ.2011.823.

Lauer S A, Grantz K H, Bi Q, Jones F K, Zheng Q, Meredith H R, et al. The incubation period of coronavirus disease 2019 (COVID-19) from publicly reported confirmed cases: estimation and application. Ann Intern Med 2020; 172(9): 577-82.

Merriam S. Andragogy and self–directed learning: pillars of adult learning theory. New Directions for Adult and Continuing Education 2001; 3-14. doi: 10.1002/ace.3.

Nassar A, Zern N K, McIntyre L K, Lynge, D et al. Emergency restructuring of a general surgery residency program during the coronavirus disease 2019 pandemic: the University of Washington experience. JAMA Surg 2020. [Ahead of print] doi: 10.1001/jamasurg.2020.1219.

Sahu K, Mishra A, Lal A. COVID-2019: update on epidemiology, disease spread and management. Monaldi Arch Chest Dis. 2020; 90(1). doi: 10.4081/monaldi.2020.1292.

Schwartz A M, Wilson J, Boden S D, Moore T J, et al. Managing resident workforce and education during the COVID-19 pandemic: evolving strategies and lessons learned. JBJS Open Access 2020; 5: e0045. doi: 10.2106/ JBJS.OA.20.00045.

Sonnadara R R, Van Vilet A, Safir O, et al. Orthopedic boot camp: examining the effectiveness of an intensive surgical skills course. Surgery 2011; 149: 745-9.

Stern D T, editor. Measuring medical professionalism. New York: Oxford University Press; 2005.

Viswanath A, Monga P. Working through the COVID-19 outbreak: rapid review and recommendations for MSK and allied heath personnel. J Clin Orthop Trauma 2020. doi: 10.1016/j.jcot.2020.03.014.

Wong C, Tay W, Yap J, et al. Love in the time of coronavirus: training and service during COVID-19. Singapore Med J 2020. doi: 10.11622/ smedj.2020053 World Health Organization. Country & technical guidance—coronavirus disease (COVID-19). 2020. https://www.who.int/emergencies/diseases/novelcoronavirus-2019/technical-guidance (accessed April 12, 2020).

Acta Orthopaedica 2020; 91 (5): 567–570

567

Surgical intervention in patients with proximal femoral fractures confirmed positive for COVID-19—a report of 2 cases Suk Kyoon SONG, Won Kee CHOI, and Myung Rae CHO Daegu Catholic University Medical Center, Daegu, South Korea Correspondence: Myung Rae Cho: cmr0426@cu.ac.kr Submitted 2020-04-19. Accepted 2020-05-08.

Case 1 (Figure 1): An 81-year-old woman presented to a local clinic because of hip pain after a fall on March 17, 2020. Her chest radiographs revealed pneumonia and she was referred to our hospital. At our emergency department, a diagnosis of COVID19 was confirmed by chest imaging studies and positive viral nucleic acid tests on throat swab specimens (Hong et al. 2020).

The patient used a cane for walking and had diabetes mellitus (DM), chronic kidney disease (CKD), chronic obstructive pulmonary disease (COPD), atrial fibrillation, and was on apixaban. A pacemaker was implanted in 2007 to treat sick sinus syndrome, and she had undergone mitral valve replacement surgery because of mitral valve stenosis. Plain radiographs of the pelvis showed an intertrochanteric fracture (AO 31A2.2) of the right femur (Müller et al. 2012). On presentation, there were no COVID-19 symptoms, such as fever, cough, or fatigue. She also showed no dyspnea on room air. She was admitted to the COVID-19 isolation ward, and antibiotic treatment (lopinavir/ritonavir twice daily, levofloxacin 750 mg IV every 24 hours, ceftriaxone 2 g IV every 24 hours) was initiated.

During the admission her symptoms did not worsen, and she did not require oxygen therapy. She had to omit apixaban for 2 days in order to reduce bleeding risk. We were able to perform surgery within 72 hours of admission. Due to poor visibility from wearing protective devices, a videoscope was used for endotracheal intubation. The patient was put under general anesthesia and placed in the supine position on a fracture table. After closed reduction under fluoroscopic guidance, we used the direct lateral approach to the hip and fixated the fracture with a compression hip screw and a trochanteric stabi-

Figure 1. Radiographic images of the chest and the hip. Plain radiographs (Panels A and B) obtained at our emergency department show prominently increased interstitial opacities at the peripheral area in both lungs (A) and right femur intertrochanteric fracture (B). Axial and coronal CT images (Panels C and D) show nodular opacities based in the subpleural area (arrows) and a small amount of bilateral pleural effusion. Panels E and F are plain radiographs taken the day after surgery.

© 2020 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 10.1080/17453674.2020.1770420

568

Acta Orthopaedica 2020; 91 (5): 567â&#x20AC;&#x201C;570

Figure 2. Chest radiographs (Panels A and C) obtained in a local clinic (A) and after presentation to our hospital (C) showing no active pulmonary lesions. Panel B shows a left femur neck fracture. Panels D and E are plain radiographs taken the day after surgery. Panel F shows the surgical team performing surgery while wearing level D PPE.

lizing plate (Solco, Pyeongtaek, Korea). The operative time (from skin incision to the end of skin closure) was 45 minutes. Since droplets frequently occur during extubation, only an anesthesiologist and a nurse anesthetist remained in the operating room during extubation after surgery. Blood loss during surgery was estimated to be 450 milliliters. From the day after surgery, the patient was encouraged to ambulate with a walker. Antibiotic treatment was continued during her stay on the isolation ward. She showed no symptoms of COVID-19 and there was no worsening of chest radiograph findings at 2 weeks after the operation.

Case 2 (Figure 2): An 83-year-old woman presented to a local clinic with dyspnea on March 20, 2020. Her chest radiograph showed no active pulmonary lesions but she was confirmed positive for COVID-19 by viral nucleic acid tests on throat swab specimens. She resided in a nursing-home facility before presentation to the local clinic and had hypertension and dementia. On March 31, while in the isolation ward in the local clinic, she fell off of the bed. Plain radiographs of the pelvis showed a transcervical fracture (Garden type 4) of the left femoral neck. She underwent intramedullary nailing surgery of the ipsilateral femur 3 years prior due to a femoral shaft fracture.

She was referred to our hospital for surgical care. No symptoms, such as fever, cough, or fatigue, were seen on

presentation. She also showed no dyspnea on room air, and there were no radiographic findings of pneumonia on chest radiographs. She was admitted to the COVID-19 isolation ward, and antibiotic treatment (hydroxyquinoline 400 mg PO daily, ceftriaxone 2 g IV every 24 hours) was initiated. During admission, her symptoms did not worsen and she did not require oxygen therapy. We were able to perform surgical treatment within 43 hours of admission. The patient underwent surgery while wearing a Korean Filter 94 (KF 94) facemask and with supplemental oxygen of 2 liters per minute via nasal cannula. Under spinal anesthesia, the patient was positioned laterally, and the intramedullary nail was removed initially. We used the modified Hardinge approach and performed a bipolar hemiarthroplasty. A crack occurred in the greater trochanteric region during removal of the intramedullary nail, and was fixated with 2 cannulated screws, 18-gauge wire, and a multifilament cable. The operative time (from skin incision to the end of skin closure) was 59 minutes. Blood loss during surgery was estimated to be 600 milliliters. From the day after surgery, the patient was encouraged to ambulate with a walker. Antibiotic treatment was continued during her stay on the isolation ward.

She showed no symptoms of COVID-19 and there was no worsening of chest radiograph findings at 2 weeks after the operation.

Acta Orthopaedica 2020; 91 (5): 567–570

569

rated considering the clinical condition of the patient. For example, if the patient suffers from severe pneumonia due to COVID-19, it Patient factors Patient 1 Patient 2

may be safer to delay the surgery until after Age 81 83 the pneumonia resolves. However, in patients Sex Female Female with mild COVID-19-related pneumonia, if BMI 29.5 17.9 the timing of surgery is delayed, complicaLab findings Initial Pre-op. Post-op.

Initial Pre-op. Post-op. tions of the fracture, rather than COVID-19, may put the patients at a greater risk for morWhite-cell count (per mm³) 11.4K 11.6K 11.0K 9.1K 6.6K 7.2K Hemoglobin (g/L) 9.2 8.7 10.1 13.7 11.8 8.4 tality. Therefore, early surgical intervention Platelet (per mm³) 267K 297K 270K 141K 98K 94K for fractures may contribute to better progPT (sec) 16 17 15 nosis. Of our 2 patients, 1 patient developed APTT (sec) 41 41 31 Creatinine (µmol/L) 2.1 1.6 1.6 0.8 0.8 0.6 mild pneumonia, while the other did not show EGFR (mL/min/ 1.73m²) 22 30 30 65 71 85 radiographic features or signs of pneumonia.

Albumin (g/L) 3.1 3.2 4 The patients were clinically stable after surCRP (mg/L) 163 85 37 106 109 Procalcitonin (ng/mL) 0.22 0.09 0.24 gery, with no complications.

D-dimer (mg/L) 3.8 14 All surgical procedures were conducted in a negative-pressure operating room dediAbbreviations: PT, prothrombin time; APTT, activated partial thromboplastin time; EGFR, estimated glomerular filtration rate; CRP, C-reactive protein. cated to patients with infectious diseases, and located in a remote corner of the operating complex, with separate access (Ti et al. 2020). All equipment for surgery and anesthesia was prepared and covered with sterile drapes in advance of the patient entering the operating room. Biological isolation chambers Discussion with negative-pressure filtration systems were used to move Over 1.3 million people from 213 countries had been confirmed the patients from the isolation ward to the operating room. positive for COVID-19 by April 7, 2020, and 73,497 related Patients wore KF 94 face masks throughout the entire surgideaths occurred (6% case fatality rate [CFR]). While the CFR cal procedure, except when they were intubated under general varies by country and by patient characteristics, CFR generally anesthesia. KF 94 is an abbreviation for “Korean Filter 94.” increases with age. According to a study by the Chinese Center The number 94 indicates a 94% ability to protect against fine for Disease Control and Prevention, the CFR was: age 40–49, particles 0.4 µm in size. It blocks the passage of fine droplets 0.4%; 50–59, 1.3%; 60–69, 3.6%; 70–79, 8.0%; and 80 or older, of saliva or sputum from an unpredictable cough.

The traffic in an operating room is a risk factor for increased 14.8% (Onder et al. 2020). During this pandemic, as millions of patients are testing positive for COVID-19, the diagnosis and postoperative infection and exposure to COVID-19. We treatment of conditions other than COVID-19 may be delayed, attempted to minimize the number of staff members in the causing additional morbidity and mortality. Patients with proxi- operating room at any one time. Only 6 staff members (4 surmal femoral fractures are mostly elderly and many have several geons and 2 nurses) attended the operation. An experienced comorbidities. Overall mortality rate following proximal fem- surgeon is important to decrease surgical time and potential oral fracture is above 25%, and the rate is higher when surgi- preoperative complications. The surgeon who operated on cal intervention is delayed. If a patient with proximal femoral these patients was an expert in hip surgery, who had performed fracture is confirmed positive for COVID-19, the response to more than 100 proximal femoral fracture surgeries annually COVID-19 may be prioritized and management of the fracture since 2004. All surgeons and scrub nurses donned level D permay be delayed. Clinically, COVID-19 has a reported incubation sonal protective equipment (PPE) with N-95 facemasks (Forperiod of approximately 5–6 days, and symptoms last for about rester et al. 2020) (Figure 2F) and the anesthesiologist and 2 weeks on average (Onder et al. 2020). However, the clinical nurse anesthetist donned level D PPE with N-95 facemasks course may vary in different patient populations, and it may take and powered air-purifying respirators (PAPR). Surgeons and several weeks for some patients to be completely cured. Even scrub nurses cannot use powered air-purifying respirators though proximal femoral fracture is not an emergency condi- because the equipment can prevent the wearing of a sterile tion, early surgical intervention contributes to better prognosis operation gown and restrict the surgeon’s activity. and puts the patient at lower risk of mortality. If proximal femoThe first patient was intubated because of the prolongation ral fracture surgery is delayed for too long, complications of the of prothrombin time from the use of apixaban due to atrial fracture, rather than the COVID-19 itself, may put the patient fibrillation. The prolongation of coagulation time is a contrainat greater risk of morbidity (Moja et al. 2012, Vrahas and Sax dication for spinal anesthesia because damage to the epidural 2017, Seong et al. 2020). Thus, this decision needs to be elabo- vein during spinal anesthesia may trigger epidural hematoma Demographics, clinical characteristics, and laboratory findings of the two patients

570

and cause spinal cord compression (Olawin and Das 2019).

We chose the surgical approach in the usual manner. Since COVID-19 patients are more prone to developing pulmonary symptoms, we used a tapered non-cemented stem (Zimmer, Warsaw, IN, USA) in Case 2 as it can reduce surgical time and has been associated with a lower incidence of pulmonary embolism than the cemented type.

Existence of viral particles in the blood is still a concern.

Therefore, the use of power tools such as bone saws and bone drills during surgery might accompany a risk of transmitting viral particles in body fluids and tissues; SARS-CoV-2 is known to be present in all bodily fluids. We also used bulb syringes rather than pulsatile irrigation to minimize the spraying of blood and bodily fluids. Electrocautery for cutting and coagulation may increase the risk of infection via aerosol generation. In order to minimize exposure to aerosols we lowered the power settings used for electrocautery and minimized its use. Electrocautery smoke was evacuated exhaustively using a suction device.

Surgical staff members were kept out of the operating room during the intubation and extubation procedures in order to minimize transmission via droplets from unpredictable coughing; they stood by in a sterile corridor at the ready for an emergency situation. The routine duration of postoperative antibiotic use in our orthopedic department is 72 hours. Since COVID-19 patients are more prone to developing viral pneumonitis, preventive antibiotics were used to prevent bacterial super-infection in these at-risk patients. Combination antiviral therapy should be added in patients with comorbidities and elderly patients (aged over 65 years).

Both patients had a considerably long waiting time before surgery. The procedures for diagnosing COVID-19 and preoperative cardiopulmonary risk assessment were the main causes of delay. In particular, the first patient had to halt apixaban for 2 days in order to reduce bleeding risk. Apixaban is recommended to be discontinued 48 hours before high-bleeding-risk procedures such as major orthopedic surgery (Mandernach et al. 2015). (We discussed this matter with the department of cardiology.) The second patient initially was admitted to a local clinic for other causes and later referred to our hospital after the fracture event. It took several days to assess comorbidities and preoperative risks. We would have postponed surgery if the risks of COVID-19 were more severe than the risks of surgical delay (e.g., severe or bilateral pneumonia, high oxygen demand).

The still-unknown characteristics of the disease and the high mortality rate in patients with risk factors may cause fear in healthcare workers managing patients confirmed positive for COVID-19. Medical personnel must approach such situations cautiously until vaccines or therapeutic agents for COVID-19 are developed. More caution is necessary for patients who are in higher risk categories. However, as in our study, in COVID19-positive patients with no or mild symptoms, early surgical intervention for comorbid conditions may be carried out if active precautions are taken. If protective measures are taken, doctors and nurses may safely perform surgery, especially

Acta Orthopaedica 2020; 91 (5): 567–570

because compared with SARS or MERS, which were more often fatal, COVID-19 is reported to show a lower fatality (Lin et al. 2020, Wang et al. 2020).

We followed the treatment guidelines of our hospital at all times when managing these 2 patients. There was no exacerbation of COVID-19 symptoms or radiographic findings; it is unclear whether this fact is attributable to antiviral therapy, or whether it is because COVID-19 is in many cases a self-limiting disease that requires only supportive care. This remains a limitation of our study, and further research is needed to decide the matter. Neither of the participants displayed any symptoms of COVID-19 in the observation period, which lasted 14 days after surgery.

In conclusion, based on the fact that early surgical intervention for proximal femoral fractures in elderly patients may result in superior surgical outcomes and lower mortality rates (Moja et al. 2012, Vrahas and Sax 2017, Seong et al. 2020), we recommend early surgical intervention for patients with proximal femoral fractures who are confirmed positive for COVID-19 when symptoms of the illness are tolerable.

Acta thanks Harald Brismar and Maziar Mohaddes for help with peer review of this study.

Forrester J D, Nassar A K, Maggio P M, Hawn M T. Precautions for operating room team members during the COVID-19 pandemic. J Am Coll Surg 2020; Apr 2. pii: S1072-7515(20)30303-3. doi: 10.1016/j.jamcollsurg.2020.03.030. [Epub ahead of print] Hong K, Lee S, Kim T, Huh H, Lee J, Kim S, Park J, Kim G, Sung H, Roh K. Guidelines for laboratory diagnosis of coronavirus disease 2019 (COVID19) in Korea. Ann Lab Med 2020; 40(5): 351-60.

Lin J, Ouyang J, Peng X-R, Isnard S, Fombuena B, Routy J-P, Chen Y-K. Potential therapeutic options for COVID-19: using knowledge of past outbreaks to guide future treatment 2020. Chin Med J 2020; March 17. doi: 10.1097/CM9.0000000000000816 Mandernach M W, Beyth R J, Rajasekhar A. Apixaban for the prophylaxis and treatment of deep vein thrombosis and pulmonary embolism: an evidence-based review. Ther Clin Risk Manag 2015; 11: 1273-82.

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.

Müller M E, Nazarian S, Koch P, Schatzker J. The comprehensive classification of fractures of long bones. Berlin: Springer-Verlag; 2012/1990.

Olawin A M, Das J M. Spinal anesthesia. StatPearls [Internet]. Treasure Island, FL: StatPearls Publishing; 2019.

Onder G, Rezza G, Brusaferro S. Case-fatality rate and characteristics of patients dying in relation to COVID-19 in Italy. JAMA 2020; Mar 23. doi: 10.1001/jama.2020.4683. [Epub ahead of print] Seong Y J, Shin W C, Moon N H, Suh K T. Timing of hip-fracture surgery in elderly patients: literature review and recommendations. Hip Pelvis 2020; 32(1): 11-16.

Ti L K, Ang L S, Foong T W, Ng B S W. What we do when a COVID-19 patient needs an operation: operating room preparation and guidance. Can J Anaesth 2020: 1-3.

Vrahas M S, Sax H C. Timing of operations and outcomes for patients with hip fracture: it’s probably not worth the wait. JAMA 2017; 318(20): 1981-2.

Wang C, Horby P W, Hayden F G, Gao G F. A novel coronavirus outbreak of global health concern. Lancet 2020; 395(10223): 470-3.

Acta Orthopaedica 2020; 91 (5): 571–575

571

Implant migration and bone mineral density measured simultaneously by low-dose CT scans: a 2-year study on 17 acetabular revisions with impaction bone grafting Hampus STIGBRAND 1,2 , Keenan BROWN 3, Henrik OLIVECRONA 4, and Gösta ULLMARK 1,2 1 Department 2 Department

of Orthopedics Gävle Hospital, Center for Research & Development, Uppsala University/County Council of Gävleborg, Sweden; of Surgical Sciences/Orthopedics, Uppsala University, Sweden; 3 Mindways Software Inc., Austin, TX, USA; 4 Department of Molecular Medicine and Surgery, Karolinska Institute, Stockholm, Sweden Correspondence: hampus.stigbrand@regiongavleborg.se Submitted 2019-12-18. Accepted 2020-04-16.

Background and purpose — Early postoperative implant migration predicts failure of joint replacements. Bone mineral density reflects bone quality and bone-graft incorporation. Implant migration and bone densitometry analysis usually require special equipment. We investigated cup migration and bone mineral density changes simultaneously with low-dose CT scans after acetabular revision hip arthroplasty using impaction bone grafting.

Patients and methods — We performed a low-dose CT postoperatively, after 6 weeks, and after 2 years in 17 patients, all revised using impaction bone grafting and a graft-compressing titanium shell in the acetabulum. 6 patients had combined segmental and cavitary acetabular defects. Cup migration was analyzed using CT-based micromotion analysis (CTMA). Bone mineral density was determined in the graft and in surrounding native bone using volumetric quantitative computed tomography (QCT). The bone graft volume was calculated from 3D reconstructions.

Results — At 2 years, the translations were 1.5 (95% CI 0.4–2.6) mm in proximal direction, -0.6 (CI –1.6 to 0.4) in the medial direction and 0.3 (CI 0.0–0.6) in the anterior direction. The mean volume of impacted bone graft was 40 cm³ (CI 28–52). In the graft bone mineral density increased 14% after 6 weeks and 23% after 2 years. There was 1 mechanical failure.

Interpretation — Proximal migration of the acetabular component was low and comparable to previous reports.

There was a rapid increase of bone mineral density in the bone graft. Low-dose CT scans make migration analysis and bone densitometry measurements possible in the same setting, offering great diagnostic potential for hip arthroplasty patients.

There is a strong relationship between early prosthetic migration and long-term prosthetic survival for both knee and hip replacements (Pijls et al. 2012). Radiostereometric analysis is the traditional method for exact measurement of prosthetic micromotion. Computed tomography is an alternative that offers comparable precision when using a new software called CT-based micromotion analysis (CTMA, SECTRA AB, Linköping, Sweden). Migration analysis is much easier to perform without specialized equipment. This software defines the surface of the pelvic cortical bone in 2 CT scans taken on 2 different occasions. The software will overlap and match these digital 3D reconstructions in a precise manner (pelvic rigid body). The implant is then defined in the same way in the same examinations. By defining the pelvic bone as the reference, the migration of the implant between the 2 examinations is calculated along the x-, y-, and z-axis. By using the pelvic cortical surface as reference, tantalum markers are no longer a prerequisite for precise definition of the pelvic bone reference (Brodén et al. 2020). This CT-based motion analysis had a precision of 0.07–0.16 mm for translations and 0.10°–0.32° for rotations of acetabular components in a recent study of 24 double examinations with different patient cohorts from 3 Swedish hospitals. 10 patients, with double examinations, from this study were also included in the precision study by Brodén et al. (2020). Prosthetic implants alter the load distribution in peri-prosthetic bone and cause osteopenia—a phenomenon known as stress-shielding (Wright et al. 2001, Bodén et al. 2006) Bone mineral density (BMD) measures only mass of mineral per volume bone tissue. A low BMD is associated with fragility fractures and BMD increases when morselized bone graft is incorporated (Gerhardt et al. 2018). However, the mechanical strength of bone is also dependent on mineral-

© 2020 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 10.1080/17453674.2020.1769295

572

Acta Orthopaedica 2020; 91 (5): 571–575

Figure 1. Postoperative images. Note the greater density in the bone graft compared with native bone. The 3 red circles indicate the ROIs of the bone mineral density measurements.

ization degree, trabecular architecture, hydroxyapatite crystal size, and collagen properties, therefore BMD is a proxy measurement for bone quality (Fonseca et al. 2014). Dual energy X-ray absorptiometry (DEXA) is currently the most widely used method for clinical measurement of bone mineral density in orthopedics (DeSapri and Brook 2020). However, DEXA screens both cortices and adjacent tissue, whereas quantitative computed tomography (QCT) can study the bone mineral density in a specific region of interest.

We measured postoperative cup migration and bone mineral density simultaneously with serial low-dose CT scans in 17 patients after revision total hip arthroplasty with impaction bone grafting.

Patients and methods Patients and surgery 17 consecutive patients (9 men) scheduled for revision total hip arthroplasty were recruited at the Orthopedic Department, Gävle Hospital from July 2015 to November 2016.

The inclusion criterion was acetabular bone loss. Patients with rheumatoid disease, dementia, or patients treated with drugs affecting bone metabolism were excluded. Mean BMI was 28 (23–37) and mean age 73 years (49–87). All patients were revised through a posterolateral approach. Bone defects were classified intraoperatively. 6 patients had combined segmental and cavitary acetabular bone defects according to the AAOS classification. In 5 patients, an acetabular titanium rim plate (Waldemar Link GmbH & Co, Hamburg, Germany) was inserted to repair a segmental defect. A posterior or posteriorsuperior supporting rim plate was used when the segmental defect compromised cup stability and/or adequate coverage of the implant in this load-bearing area. Medial or anterior acetabular rim reinforcement was not used in this material.

Acetabular impaction bone grafting was performed in all

Figure 2. 3-D rendering of bone graft volume.

cases. A thin, graft-compressing titanium shell, from the same company (Waldemar Link GmbH & Co, Hamburg, Germany), was inserted on top of the graft to enhance compression. A polyethylene cup, size 46–52 mm, with a prosthetic head of 32 or 36 mm (Lubinus Eccentric X-linked acetabular cup, Waldemar Link GmbH & Co, Hamburg, Germany) was cemented inside the titanium shell, for details see Stigbrand et al. (2018).

Weight-bearing was allowed immediately postoperatively.

Radiographic analysis Low-dose CT scans were performed postoperatively at 6 weeks and 2 years. The polyethylene cup was marked with 1.0-mm tantalum markers. CT-based migration analysis was performed using the CTMA software (SECTRA AB, Linköping, Sweden). Based on 12 double examinations a precision of 0.11–0.14 mm migration was calculated (assuming zero migration). Double examinations were performed within minutes with the patient standing up in between. The surface of the pelvic bone was defined as the skeletal reference body, without tantalum markers in the bone. Precision was defined as repeatability, i.e., the variation in 2 repeated measurements on the same subject under identical conditions over a short period of time.

To get the precision estimates we followed the procedure commonly used for RSA. With n representing the number of patients, we estimated the precision of the method by extracting from a Student’s t-distribution chart of n–1 degrees of freedom the critical value encompassing 95% of the distribution. We then multiplied this by the standard deviation of our double measurements.

The resulting value is our precision. QCT analysis was performed with Mindways Software (Mindways Software Inc, Austin TX, USA) by defining a volume of 0.4–1.0 cm3 in 3 regions of interests: (a) bone graft, (b) native bone cranial to the acetabular bone graft, and (c) native bone in the inferior ramus (Figure 1).

Acta Orthopaedica 2020; 91 (5): 571–575

573

Table 1. Postoperative translations of the acetabular component. Values are mean (mm) with 95% confidence intervals Time-point Axis 6 weeks

2 years

X (med. +) –0.16 (–0.46 to 0.14) –0.56 (–1.56 to 0.44) Y (prox. +) 0.31 (0.11 to 0.51) 1.52 (0.42 to 2.62) Z (ant. +) 0.00 (–0.10 to 0.10) 0.27 (–0.03 to 0.57)

Table 2. Postoperative measurements of bone mineral density in the ROIs. Values are mean (mg/cm³) with 95% confidence intervals Region Time-point of interest Postop., day 1 6 weeks Native bone cranial caudal Bone graft

To ensure consistency of sampled volumes over time, the baseline and follow-up studies were analyzed in parallel by the same observer (KB). Distinctive landmarks were viewed on screen simultaneously in each CT study in a set of serial exams, and sample volumes were positioned in each exam with respect to the landmarks.

Care was taken not to include bone cement or screws in the analyzed volumes. A calibration phantom was used to calibrate the CT scan for density measurements. All CT scans were performed on the same Toshiba Aquilion One CT scanner (Canon Medical Systems, Tustin, CA, USA) and radiation dose was approximately 2.3 mSv per examination. For technical reasons, 2 CT scans in 2 cases were excluded. 1 patient had the contralateral hip examined at 2 years and another patient had image disturbances in the QCT analysis.

Volume measurement The bone graft volume was measured by defining the bone graft in 3 dimensions. To do this, the border was marked between bone graft and native bone and between bone graft and cement/titanium shell on every 5th CT slide (i.e., every 2.5 mm) (Figure 1). The software Vital Imaging (Miami, FL, USA) approximated the remaining slides. Finally, the grafted volume was controlled and corrected in all 3 dimensions on every slide (Figure 2).

Clinical follow-up Clinical examination was performed preoperatively and at 2 years with hip scoring according to Merle D’Aubigne and Postel (Charnley 1979). No patient was lost to follow-up.

Statistics 17 patients were included. Based on previous studies regarding bone mineral changes and proximal migration of acetabular components this number of patients was considered adequate (Ornstein et al. 1999, Saari et al. 2014, Gerhardt et al. 2018, Zampelis et al. 2018). Multiple measurements were performed on each of them, so a linear model for repeated measurements was used to investigate factors influencing migration and bone mineral density changes. Within the linear model concept the fixed effects-only model with repeated statement was used.

The functional form of the residuals was specified as compound symmetry for the covariance pattern. The degrees of

137 (88–186) 79 (18–140) 378 (355–401)

145 (94–196) 93 (33–153) 430 (412–448)

2 years 132 (72 to 192) 35 (–38 to 108) 466 (390 to 542)

p-value 0.2 0.05 0.05

freedom using the maximum likelihood method was obtained

by Satterthwait approximation. P-values in Table 2 are a result of repeated measure ANOVAs. Statistical analyses were performed in SPSS Statistics 25 (IBM Corp, Armonk, NY, USA).

A confidence interval (CI) of 95% was used.

Ethics, funding, and potential conflicts of interest The study was approved by the Regional Ethics Committee, Uppsala, Sweden, Registration number 2015/228. The study received grants from the foundation Skobranschens utvecklingsfond and from the Center of Research and Development, Uppsala University, and County Council of Gävleborg. KB has a financial relationship with Mindways Software Inc. (Texas, USA). HO has a financial relationship with SECTRA AB (Linköping, Sweden). Other authors have no disclosures to make.

Results CT-based migration At 2 years, the translations were 1.5 mm (CI 0.4–2.6) in the proximal direction, –0.6 (CI –1.6 to 0.4) in the medial direction and 0.3 (CI 0.0–0.6) in the anterior direction (Table 1).

Bone mineral density Immediately postoperatively, the bone mineral densities in the 3 different regions of interest (ROIs) were: 378 mg/cm³ (CI 357–401) in the bone graft; 137 (CI 88–186) cranial to the graft; and 79 (CI 18–140) in the inferior ramus (baseline) (Table 2). Already at 6 weeks, density had increased 14% in the bone graft, continuing to 23% at 2 years (p = 0.05).

Volume measurement The mean bone grafted volume was 40 cm³ (CI 28–52) (Figure 2).

Clinical results Hip score according to Merle, d’Aubigne and Postel increased from 14 (CI 12–15) preoperatively to 18 (CI 18–18) postoperatively. There was 1 mechanical failure, which was excluded from the follow-up. The failure case had an AAOS type IV acetabular bone defect (pelvic dissociation), unknown at preoperative planning. The failure was

574

aseptic with a proximal migration of 6 mm and a decrease in BMD in the bone graft, from 358 mg/cm³ postoperatively to 265 mg/cm³ at 2 years.

Discussion This study describes 2 new aspects: 1st, implant migration and bone mineral density can be measured simultaneously by low-dose CT scans. Instead of using 2 different labs, and 2 different examinations with 2 different machines, a low-dose standard CT scan can measure prosthetic migration and bone mineral density at the same time. 2nd, implant migration can be measured with high precision without tantalum markers in bone. In both primary and revision hip arthroplasty, a high proximal migration of acetabular components at 2 years predicts late loosening regardless of surgical technique (Klerken et al. 2015, Mohaddes et al. 2017b). Proximal migration and its predictive value is complex in revision surgery with bone grafting (Klerken et al. 2015, Mohaddes et al. 2017a). Comparisons are difficult because the severity of acetabular bone defects, graft preparation, impaction technique, and implants differs between studies. Our results of a proximal migration of 1.5 mm is comparable with previous reports (Ornstein et al.

1999, Mohaddes et al. 2017b), although 6 of our 17 cases had combined acetabular defects that required rim reinforcement.

Zampelis et al. (2018) reported a proximal migration of only 0.22 mm—compared with 0.59 for their control group—at 2 years for cavitary defects when clodronate was added to the bone graft intraoperatively (Zampelis et al. 2018). In the same study, there was no statistically significant difference for the 2 groups in the bone mineral density (measured in the graft proximal to the cup) at 2 years.

Few studies exist on changes in bone mineral density during bone allograft incorporation (Gerhardt et al. 2018, Zampelis et al. 2018). DEXA measurements after hip revision with impaction bone grafting found an increase of 14% in the cranial graft bed after 2 years (Gerhardt et al. 2018).

Our result of 23% at 2 years is rather close, and the difference may be because we measured with QCT instead of DEXA. The increase in actual trabecular bone mineral density in their study might also have been influenced by the fact that they included cortices in their measurement whereas we did not.

At 6 weeks in our study, density increased in native bone cranial and caudal to the cup, although without statistical significance. In animal models, distant metaphyseal trauma affects mineralization and cellular expression in unrelated bones (Tatting et al. 2017). This effect plus increased load, mobilization, and metabolic activity may explain this unexpected finding in our study at 6 weeks. Postoperatively the graft bed was 3–5 times denser than the surrounding bone. Despite this, the density continued to increase in the graft during the follow-up period, both in our study and that of Gerhardt et al. (2018). This could be explained by further compression

Acta Orthopaedica 2020; 91 (5): 571–575

of dead bone graft rather than graft incorporation. However, of the total BMD increase in the bone graft, 59% occurred during the first 6 weeks but only 20% of the proximal migration occurred during the same period, indicating a biological response. Several studies of primary total hip arthroplasty show that cementless cup fixation decreases retro-acetabular bone stock more than cemented fixation (Wright et al. 2001, Digas et al. 2006). Differences in load distribution to the acetabulum may explain this, but fluid exchange and wear debris circulating through screw holes is also a possible cause (Fahlgren et al. 2010). Wright et al. (2001) have analyzed retro-acetabular bone mineral density changes with QCT 14 months after primary cementless cup fixation and found a decrease of 26% immediately cranial to the cup.

Goldvasser et al. (2012, 2014) have studied the accuracy of CT-based polyethylene wear measurements. They compared femoral head penetration on CT scans with measurements from a micrometer on the explanted liner after revision surgery. They found no statistically significant difference between the 2 methods, which emphasizes the accuracy of CT for wear measurements in vivo.

In our study only age affected the increase in bone mineral density and proximal migration (model linear regression).

Bone graft volume, sex, and rim reinforcement did not.

Our study had some limitations. There were few patients, no inter- and intra-observer analysis of the measurements, and only 12 of the patients had double examinations performed.

However, Schmidt et al. (2005) reported an inter- and intraobserver correlation coefficient over 0.99 in periacetabular osteo-densitometry, indicating a high reproducibility of QCT measurements around acetabular implants. Wright et al.

(2001) also conducted intra-observer testing with a correlation coefficient of ≥ 0.89, although in a small group of 6 patients.

Our results also confirm previous results regarding BMD changes in bone graft (Gerhardt et al. 2018). The single failure case in our study had a pelvic discontinuity. He experienced a migration of 6 mm at 2 years and the second-largest bone grafted volume. We do not recommend the present method for patients with acetabular bone loss type IV (pelvic discontinuity).

Compared with plain radiographs a CT scan will enable: 1. prognosis prediction by migration analysis of the prosthetic implant; 2. wear measurements of implanted polyethylene cups or liners; 3. more accurate assessment of periprosthetic osteolysis and bone mineral density changes.

In conclusion, a postoperative, low-dose CT scan offers diagnostic potential. A radiation dose of 2.3 mSv is acceptable in this context.

The authors would like to thank Krister Ågren and Hans Högberg for statistical assistance.

Acta Orthopaedica 2020; 91 (5): 571–575

GU, HS and HO designed the study. All patients were recruited and operated

by HS or GU. All authors have reviewed and contributed to the manuscript.

HS performed the migration analysis, volume measurement, clinical followup, and drafted the manuscript. KB performed the measurements of bone mineral density. HO assisted HS in the migration analysis.

Acta thanks Davey Gerhardt and Wim Schreurs for help with peer review of this study.

Bodén H S, Sköldenberg O G, Salemyr M O, Lundberg H J, Adolphson P Y. Continuous bone loss around a tapered uncemented femoral stem: a longterm evaluation with DEXA. Acta Orthop 2006; 77(6): 877-85.

Brodén C, Sandberg O, Sköldenberg O, Stigbrand H, Hänni M, Giles J W, Emery R, Lazarinis S, Nyström A, Olivecrona H. Low-dose CT-based implant motion analysis is a precise tool for early migration measurements of hip cups: a clinical study of 24 patients. Acta Orthop 2020; 14: 1-6.

Charnley J. Numerical grading of clinical results. In: Low friction arthroplasty of the hip: theory and practice. Heidelberg: Springer-Verlag; 1979. p. 20-4.

DeSapri K T, Brook R. To scan or not to scan? DXA in postmenopausal women. Cleve Clin J Med 2020; 87(4): 205-10.

Digas G, Kärrholm J, Thanner J. Different loss of BMD using uncemented press-fit and whole polyethylene cups fixed with cement: repeated DXA studies in 96 hips randomized to 3 types of fixation. Acta Orthop 2006; 77(2): 218-26.

Fahlgren A, Boström M P, Yang X, Johansson L, Edlund U, Agholme F, Aspenberg P. Fluid pressure and flow as a cause of bone resorption. Acta Orthop 2010; 81(4): 508-16.

Fonseca H, Moreira-Goncalves D, Coriolano H J, Duarte J A. Bone quality: the determinants of bone strength and fragility. Sports Med 2014; 44(1): 37-53.

Gerhardt D, De Visser E, Hendrickx B W, Schreurs B W, Van Susante J L C. Bone mineral density changes in the graft after acetabular impaction bone grafting in primary and revision hip surgery. Acta Orthop 2018; 89(3): 302-7.

Goldvasser D, Noz M E, Maguire G Q, Jr, Olivecrona H, Bragdon C R, Malchau H. A new technique for measuring wear in total hip arthroplasty using computed tomography. J Arthroplasty 2012; 27(9): 1636-40.e1.

Goldvasser D, Hansen V J, Noz M E, Maguire G Q, Jr, Zeleznik M P, Olivecrona H, Bragdon C R, Weidenhielm L, Malchau H. In vivo and ex vivo measurement of polyethylene wear in total hip arthroplasty: com-

575

parison of measurements using a CT algorithm, a coordinate-measuring machine, and a micrometer. Acta Orthop 2014; 85(3): 271-5.

Klerken T, Mohaddes M, Nemes S, Kärrholm J. High early migration of the revised acetabular component is a predictor of late cup loosening: 312 cup revisions followed with radiostereometric analysis for 2–20 years. Hip Int 2015; 25(5): 471-6.

Mohaddes M, Shareghi B, Kärrholm J. Promising early results for trabecular metal acetabular components used at revision total hip arthroplasty: 42 acetabular revisions followed with radiostereometry in a prospective randomised trial. Bone Joint J 2017a; 99-b(7): 880-6.

Mohaddes M, Herberts P, Malchau H, Johanson P E, Kärrholm J. High proximal migration in cemented acetabular revisions operated with bone impaction grafting; 47 revision cups followed with RSA for 17 years. Hip Int 2017b; 27(3): 251-8.

Ornstein E, Franzen H, Johnsson R, Sandquist P, Stefansdottir A, Sundberg M. Migration of the acetabular component after revision with impacted morselized allografts: a radiostereometric 2-year follow-up analysis of 21 cases. Acta Orthop Scand 1999; 70(4): 338-42.

Pijls B G, Nieuwenhuijse M J, Fiocco M, Plevier J W, Middeldorp S, Nelissen R G, Valstar E R. Early proximal migration of cups is associated with late revision in THA: a systematic review and meta-analysis of 26 RSA studies and 49 survival studies. Acta Orthop 2012; 83(6): 583-91.

Saari T M, Digas G, Kärrholm J N. Risedronate does not enhance fixation or BMD in revision cups: randomised study with three years follow-up. Hip Int 2014; 24(1): 49-55.

Schmidt R, Pitto RP, Kress A, Ehremann C, Nowak TE, Reulbach U, Forst R, Muller L. Inter- and intraobserver assessment of periacetabular osteodensitometry after cemented and uncemented total hip arthroplasty using computed tomography. Arch Orthop Trauma Surg. 2005 Jun;125(5):291-7.

Stigbrand H, Gustafsson O, Ullmark G. A 2- to 16-year clinical follow-up of revision total hip arthroplasty using a new acetabular implant combined with impacted bone allografts and a cemented cup. J Arthroplasty 2018; 33(3): 815-22.

Tatting L, Sandberg O, Bernhardsson M, Ernerudh J, Aspenberg P. Isolated metaphyseal injury influences unrelated bones. Acta Orthop 2017; 88(2): 223-30.

Wright J M, Pellicci P M, Salvati E A, Ghelman B, Roberts M M, Koh J L. Bone density adjacent to press-fit acetabular components: a prospective analysis with quantitative computed tomography. J Bone Joint Surg Am 2001; 83(4): 529-36.

Zampelis V, Belfrage O, Tägil M, Sundberg M, Flivik G. Decreased migration with locally administered bisphosphonate in cemented cup revisions using impaction bone grafting technique. Acta Orthop 2018; 89(1): 17-22.

576

Acta Orthopaedica 2020; 91 (5): 576–580

Patients with hip resurfacing arthroplasty are not physically more active than those with a stemmed total hip Jetse JELSMA 1, Martijn G M SCHOTANUS 1, Ivo T A F BUIL 2, Sander M J VAN KUIJK 3, Ide C HEYLIGERS 1,4, and Bernd GRIMM 5 1 Department

of Orthopedic Surgery and Traumatology, Zuyderland Medical Centre, Sittard-Geleen, The Netherlands; 2 Department of Innovation and Funding, Zuyderland Medical Centre, Sittard-Geleen, The Netherlands; 3 Department of Clinical Epidemiology and Medical Technology Assessment (KEMTA), Maastricht, The Netherlands; 4 School of Health Professions Education, Maastricht University, Maastricht, The Netherlands; 5 Luxembourg Institute of Health, Human Motion, Orthopedics, Sports Medicine, Digital Methods (HOSD), Luxembourg, Luxembourg Correspondence: jetsejelsma@gmail.com Submitted 2020-02-18. Accepted 2020-04-22.

Background and purpose — Hip resurfacing arthroplasty (HRA) was designed for the highly active patient because of the various theoretical advantages compared with stemmed total hip arthroplasty (THA), but has shown high failure rates. Physical activity (PA) after arthroplasty is frequently determined with the use of questionnaires, which are known for their subjective nature, recall bias, and ceiling effect.

These disadvantages are not applicable to physical activity monitoring (AM) using sensors. We compared objectively measured PA at long-term follow-up in a matched cohort of HRA and stemmed THA subjects.

Patients and methods — We compared 2 groups of 16 patients (12 males) in each group, one having received unilateral HRA (median age 56 years at surgery) and a matched group having received unilateral stemmed THA with a small diameter femoral head (28 mm) on conventional polyethylene (median age 60 years at surgery) with osteoarthritis as indication for surgery, 10 years after surgery. Groups were matched by sex, age at surgery, and BMI. The daily habitual PA was measured over 4 consecutive days in daily living using a 3-axis accelerometer, gyroscope, and magnetometer.

Both quantitative parameters (time standing, sitting, walking, number of steps, and sit–stand transfers) and qualitative parameters (walking cadence) were determined.

Results — The AM was worn for a median 13 (11–16) hours per day. The median daily step count was 5,546 (2,274–9,966) for the HRA group and 4,583 (1,567–11,749) for the stemmed THA-group with 39 (21–74) versus 37 (24–62) daily sit–stand transfers respectively. The other PA parameters were also similar in both groups.

Interpretation — We found similar median PA levels and also identical ranges. While short-term effects may exist, ageing and related behavioral adaptations or other effects seem to render the theoretical activity benefits from HRA irrelevant at longer follow-up.

Metal-on-metal (MoM) hip resurfacing arthroplasty (HRA) was designed for the highly active patient because of various theoretical advantages compared to conventional stemmed metal-on-polyethylene (MoP) THA: low volumetric wear, large physiological diameter femoral heads offering stability, near-natural joint kinematics, and increased range of motion compared with small-diameter THA and preservation of the femoral bone (Corten et al. 2011). HRA was commonly advertised as “the sporting hip” and publicity was created with subjects participating in triathlons after HRA (Girard et al. 2017).

Gait analysis studies showed that HRA subjects returned to a more normative gait pattern with a higher walking speed when compared with THA (Nantel et al. 2009, Gerhardt et al. 2019).

HRA showed initially promising results; however, since 2004 concerns have been raised because of high failure rates. In the Netherlands the use of HRA has been forbidden by the Dutch Orthopaedic Society (NOV) and Government since January 2012 (Verhaar 2012).

It can be expected that patients who received an HRA would be more physically active after surgery when compared with patients who received a stemmed THA with a small-diameter MoP or ceramic-on-polyethylene (CoP) bearing for multiple reasons: the theoretically better implant design features of HRA listed above claiming to support a more active lifestyle, the selection of patients for this particular implant (young and active), and the related patient expectations (high preoperative demands on postoperative activity). A few studies have investigated return to sports after HRA and showed that patients were able to return to high activity levels and sporting activities postoperatively, at least at short-time follow-up (Narvani et al. 2006, Naal et al. 2007). Some studies comparing an HRA with a stemmed MoP or CoP THA showed higher postoperative activity levels after HRA at 3–4 years’ follow-up. However, activity levels were determined using a (weighted) self-reported activity questionnaire (Zywiel et al.

© 2020 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 10.1080/17453674.2020.1771652

Acta Orthopaedica 2020; 91 (5): 576–580

577

2009, Mont et al. 2009). The highly subjective Unilateral Hip Resurfacing with 10 ± 2 years follow-up nature, strong recall bias, and possible ceiling n = 40 effect are known disadvantages of such questionnaires, especially for quantifying activity Excluded (n = 24): – revision, 9 levels, which is in contrast to activity monitor– bilateral arthroplasty, 8 ing (AM) using sensors (Terwee et al. 2011).

– declined participation, 5 – not reached, 2 Wearable AMs measure a patient’s habitual – dead, 1 physical activity (PA) objectively and continuously in the free-living environment and different physical activities can be differentiated (Lipperts et al. 2017, Jelsma et al. 2019).

We objectively measured PA at long-term follow-up in an ageand sex-matched cohort of HRA and stemmed THA subjects.

The hypothesis was that subjects with a unilateral HRA are physically more active in habitual daily life, measured by AM.

Stemmed THA with 10 ± 2 years follow-up n = 371 Excluded (n = 311): – other fixation type, 141 – bilateral arthroplasty, 87 – participating in other studies, 19 – major surgery or PA influencing disease, 18 – dead, 15 – dementia, 11 – no surgical documentation, 9 – revision, 8 – previous surgery, 2 – moved to other region, 1

a

a

Eligible n = 60 Excluded after first contact (n = 14): – bilateral arthroplasty, 4 – declined participation, 3 – major surgery or PA influencing disease, 2 – moved to other region, 2 – not reached, 2 – measurement failure, 1

Patients and methods We conducted a cohort study at the Zuyderland Medical Centre, Sittard-Geleen, The Netherlands between August and November 2017 (recruitment of HRA group) and from February to June 2018 (recruitment of stemmed THA group). We compared 2 groups, one having received unilateral HRA and a matched group having received unilateral stemmed THA with a small diameter metal or ceramic femoral head (28 mm) on conventional polyethylene with osteoarthritis as indication for surgery. The HRA group with a median follow-up of 10 (9–11) years was initially recruited for another study, the methods of which are described in detail elsewhere (Jelsma et al. 2020). The stemmed THA group was matched by sex, age at surgery, follow-up since surgery, and BMI. Patients with an uncemented, unilateral stemmed THA with a MoP or CoP bearing were included. The follow-up was set at 8–12 years to optimize the chances of a matched cohort. Finally, 16 patients consented to the study and were included as a matched cohort (Figure). There were no statistically significant differences between group characteristics at baseline (Table 1). The use of the AM has been described in detail elsewhere (Lipperts et al. 2017). The daily habitual PA was measured during waking hours for 4 consecutive days in daily living. The AM used to collect the raw signal was a 3-axis accelerometer, gyroscope, and magnetometer (HAM-IMU+alt, Gulf Coast Data Concepts LLC, Waveland, MI, USA). The data received with this AM were analyzed with MATLAB (MATLAB R2017a, The Mathworks Inc., Natick, MA, USA) with the use of previously validated algorithms with excellent accuracy (> 97%) in determining PA levels in a semi-free setting (Lipperts et al. 2017).

With the AM, various quantitative parameters of PA can be obtained and in this study we assessed the following metrics: the time in hours standing, sitting, walking, and cycling and the amount of steps and sit–stand transfers. Walking cadence, defined as the number of steps per minute and a proxy of walking speed, was calculated as a qualitative parameter.

Eligible n = 16

Eligible n = 46 Matching

Analyzed n = 16

Analyzed n = 16

Number of patients enrolled and analyzed in this study. a One patient underwent revision surgery, but died 6 months later due to a non-surgical reason. 1 patient underwent revision surgery and a primary arthroplasty contralaterally a few years after.

We also assessed outcome by 3 commonly used patientreported outcome measures (PROMs): the 12-item Forgotten Joint Score (FJS-12) and the Hip disability and Osteoarthritis Outcome Score Physical Function Short Form (HOOS-PS), both with “100” as the best possible score, and the Short Questionnaire to Assess Health-enhancing physical activity (SQUASH) to determine a total activity score (Davis et al.

2008, Behrend et al. 2012, Wendel-Vos et al. 2003).

Statistics Group comparison (e.g., patient characteristics) and parameters of PA between the groups were performed using the Mann–Whitney U test, because of the small groups, Pearson’s chi-square test and, in the case of expected cell-counts, Fisher’s exact test was used to test for differences between groups present at baseline. For all analyses, a p-value was considered to be statistically significant at p ≤ 0.05. Results are presented as median (range). IBM SPSS Statistics 22 (IBM Corp, Armonk, NY, USA) was used for statistical analysis.

Ethics, funding, and potential conflicts of interest This study was performed in compliance with the 1975 Declaration of Helsinki, as revised in 2013, and was studied and

578

Acta Orthopaedica 2020; 91 (5): 576–580

Table 1. Baseline characteristics. Values are median (range) unless otherwise specified Factor

HRA group

Stemmed THA group

Male sex, n 12 12 Age at surgery 56 (43–67) 60 (53–68) Follow-up (years) 10 (9–11) 10 (8–12) BMI 26 (22–37) 29 (20–40) Approach, n Posterolateral 7 7 Straight lateral 9 9 Bearing, n MoM 16 – MoP – 4 CoP – 12

p-value 1.0 0.1 0.4 0.1 1.0

– – –

MoM = metal-on-metal, MoP = metal-on-polyethylene, CoP = ceramic-on-polyethylene.

Table 2. Parameters of physical activity monitoring. Values are median (range)

Factor

AM wearing time (h) 12 (11–16) Time sitting (h) 7.6 (4.6–12) Time standing (h) 3.0 (1.8–5.7) Time walked (h) 1.3 (0.5–1.9) Time cycled (h) 0.05 (0.0–0.48) Steps taken (n×103) 5.5 (2.3–10) Sit–stand transfers (n) 39 (21–74) Cadence (steps/min) 102 (81–112)

Results Both groups had similar baseline characteristics (Table 1) and showed similar PA monitor parameters (Table 2) and PROMs (Table 3).

Discussion This observational matched-cohort study showed that patients with a unilateral HRA are not physically more active when compared with subjects with a unilateral stemmed MoP or CoP THA at 10 years’ follow-up. This is counterintuitive to expectations, as HRA patients received the theoretical advantages of the implant design (large, near physiological head diameter) and surgical procedure (anatomical preservation), and represent a selection bias towards subjects presenting, being perceived by the surgeon, or themselves expecting to be more physically active and demanding than patients in the stemmed conventional MoP/CoP small-diameter head THA group. In addition, in this matched-cohort study, the median age and BMI was higher, but not significant, in the stemmed THA group, both factors established to be related with a less active lifestyle. Patients with both types of implants did not only have comparable mean PA levels but also showed identical ranges. Thus it seems that both implant types enable the same level of PA and that activity levels depend on individual lifestyle rather than on implant type, at least at 10 years’ follow-up.

14 (11–16) 9.6 (3.8–13) 3.0 (1.6–6.2) 1.1 (0.4–1.8) 0.01 (0.0–1.2) 4.6 (1.6–11) 37 (24–65) 98 (80–110)

0.1 0.1 0.9 0.6 0.5 0.6 0.7 0.3

Table 3. Patient reported outcome measures. Values are median (range) Factor

approved by the IRB (METC Zuyd, Heerlen, The Netherlands, IRB nr: 10N72 + amendment) and conducted in accordance with the guidelines for Good Clinical Practice (GCP). All patients signed informed consent. The authors declare they do not have any kind of conflict of interest.

HRA group Stemmed THA group p-value

HOOS-PS FJS-12 SQUASH

HRA group

Stemmed THA group

87 (49–100) 63 (2–100) 6,150 (1,110–18,480)

67 (49–100) 56 (4–100) 4,560 (1,050–9,300)

p-value 0.3 0.5 0.2

PA is considered a major risk factor for a number of adverse health outcomes. Reaching a daily step count > 8,000 has been associated with a lower risk of all-cause mortality (Saint-Maurice et al. 2020). In our study 13 subjects (5 HRA, 8 stemmed THA) made < 5,000 steps/day, 5 subjects ≥ 8,000 steps/day (2 HRA, 3 stemmed THA), and only 1 subject > 10,000 steps/ day (HRA). This suggest that almost half of the patients in this study would be considered to be living sedentary lifestyles with the associated risks of developing non-communicable diseases (Tudor-Locke et al. 2011). Sedentary time as a parameter is related to but largely independent of PA levels and was numerically higher in THA than HRA (p = 0.05), but this absolute time difference corresponds almost completely to the difference in total wear time between both groups, indicating a difference in instructions or compliance with it for wear time (waking hours) more than in activity behavior.

The groups showed no statistically significant differences in the HOOS-PS and FJS-12, suggesting that the groups are comparable according to the assessed domains such as pain, patient satisfaction, or perceived function. A 20-points mean difference was seen in the HOOS-PS in favor of HRA. This is in accordance with a recent publication of Oxblom et al.

(2019). They studied 726 subjects 7 years after primary HRA or conventional THA showing a significant difference in HOOS subscales of function of daily living and function in sport and recreation, although HOOS subscales of symptoms, pain, and quality of life, EQ-5D index, and visual analog scores for pain and satisfaction did not differ.

It has been shown that patients 1 year after receiving stemmed THA show only a few changes in objectively measured free-living PA compared with preoperative levels (Jeldi

Acta Orthopaedica 2020; 91 (5): 576–580

et al. 2017, Thewlis et al. 2019). While the reason not to use a pain-free hip and improved functional capacity towards higher PA levels is multifactorial, one possible explanation is that, as PA levels are known to be related to wear of MoP bearings, long-term participation in high-impact activities is usually not recommended (Schmalzried et al. 2000). However, there is little prospective evidence reporting a poor clinical outcome with higher levels of activity (Meira and Zeni 2014).

In Danish and American guidelines and a Dutch survey most low-impact activities only were allowed, though not necessarily promoted post-THA (Meester et al. 2018). This is in contrast to the advice given by orthopedic surgeons to HRA patients and the publicity of HRA manufacturers calling it a “sporting hip.” Multiple studies have shown a high return to sport, including high-intensity activities such as long-distance triathlon, after HRA (Fouilleron et al. 2012, Girard et al.

2017). For subjects with a conventional THA this has not been advised but may be possible.

The studies in current literature comparing PA or sports participation in HRA and stemmed THA have all been performed using (weighted) PA questionnaires (Mont et al. 2009, Zywiel et al. 2009). Our study is one of the first of its kind to evaluate habitual PA in the free-living environment using a wearable AM. Questionnaires are limited by the highly subjective nature and ceiling effect, which is in contrast to the objective results obtained by AM (Terwee et al. 2011). Zywiel et al. (2009) performed a study comparing PA, measured by a weighted questionnaire (comparable with the SQUASH), in HRA and a matched cohort of patients with a stemmed THA. At final follow-up (3–4 years) the HRA group had a higher mean weighted activity score than the stemmed THA group (p < 0.01), while activities preoperatively were similar. It was not stated whether there were differences regarding the instructions for postoperatively approved (high-intensity) PA. Other comparison studies by Pollard et al. (2006), Vail et al. (2006) and Mont et al. (2009) used UCLA activity scores and weighted activity scores and found a higher degree of PA in HRA, although these studies have numerous limitations, mainly related to the uncontrolled bias of HRA towards very high preoperative PA levels.

Our study has limitations. The number of subjects was rather low. The main cause for this was the initially strict inclusion criteria for the initial HRA study (Jelsma et al. 2020). Another limitation was that no objective information, e.g., PA monitoring data, were available for the preoperative setting of the subjects. This might have influenced our results, because the physically more active patients could have been designated for hip resurfacing at the time of surgery, and therefore selection bias may have occurred. However, such a possible selection bias would further support the findings of this study.

Conclusion This is the first study comparing postoperative PA levels between HRA and stemmed THA using wearable sensors for

579

objective PA measures. HRA theoretically supports high PA levels, by design and surgery, which should result in a difference at 10 years, although this study found no differences in PA and ranges are also comparable. Even well-reasoned theoretical advantages concerning functional advantages of any implant design require clinical validation and should not be assumed as an indication (especially at the risk of a disadvantage). While short-term effects may exist, ageing and related behavioral adaptations or other effects seem to render the theoretical activity benefits from HRA irrelevant at longer follow-up. PA levels at long follow-up seem to depend less on implant type but rather on other factors, warranting further research to ensure the related health benefits in THA patients.

JJ: design of the study, patient inclusion, statistical analysis, manuscript writing. MS: design of the study, patient inclusion, cohort-matching, manuscript review. IB: data analysis. SvK: statistical analysis, manuscript review.

IH, BG: design of the study, manuscript review Acta thanks Anne Garland and Keijo T Mäkelä 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.

Corten K, Ganz R, Simon J P, Leunig M. Hip resurfacing arthroplasty: current status and future perspectives. Eur Cell Mater 2011; 21: 243-58.

Davis A M, Perruccio A V, Canizares M, Tennant A, Hawker G A, Conaghan P G, Roos E M, Jordan J M, Maillefert J F, Dougados M, Lohmander L S. The development of a short measure of physical function for hip OA HOOS-Physical Function Shortform (HOOS-PS): an OARSI/OMERACT initiative. Osteoarthritis Cartilage 2008; 16(5): 551-9.

Fouilleron N, Wavreille G, Endjah N, Girard J. Running activity after hip resurfacing arthroplasty: a prospective study. Am J Sports Med 2012; 40(4): 889-94.

Gerhardt D M J M, Mors T G T, Hannink G, Van Susante J L C. Resurfacing hip arthroplasty better preserves a normal gait pattern at increasing walking speeds compared to total hip arthroplasty. Acta Orthop 2019; 90(3): 231-3.

Girard J, Lons A, Pommepuy T, Isida R, Benad K, Putman S. High-impact sport after hip resurfacing: The Ironman triathlon. Orthop Traumatol Surg Res 2017; 103(5): 675-8.

Jeldi A J, Deakin A H, Allen D J, Granat M H, Grant M, Stansfield B W. Total hip arthroplasty improves pain and function but not physical activity.

J Arthroplasty 2017; 32(7): 2191-9.

Jelsma J, Schotanus M G, Senden R, Heyligers I C, Grimm B. Metal ion concentrations after metal-on-metal hip arthroplasty are not correlated with habitual physical activity. Hip Int 2019; 29(6): 638-46.

Jelsma J, Schotanus M G M, van Kuijk S M J, Buil I T A F, Heyligers I C, Grimm B. Quality, but not quantity of physical activity is associated with metal ion concentrations in unilateral hip resurfacing. J Orthop Res 2020; Feb 22. doi: 10.1002/jor.24637.

Lipperts M, van Laarhoven S, Senden R, Heyligers I, Grimm B. Clinical validation of a body-fixed 3D accelerometer and algorithm for activity monitoring in orthopaedic patients. J Orthop Translat 2017; 11: 19-29.

Meester S B, Wagenmakers R, van den Akker-Scheek I, Stevens M. Sport advice given by Dutch orthopaedic surgeons to patients after a total hip arthroplasty or total knee arthroplasty. PLoS One 2018; 13(8): e0202494.

Meira E P, Zeni J Jr. Sports participation following total hip arthroplasty. Int J Sports Phys Ther 2014; 9(6): 839-50.

580

Mont M A, Marker D R, Smith J M, Ulrich S D, McGrath M S. Resurfacing is comparable to total hip arthroplasty at short-term follow-up. Clin Orthop Relat Res 2009; 467(1): 66-71.

Naal F D, Maffiuletti N A, Munzinger U, Hersche O. Sports after hip resurfacing arthroplasty. Am J Sports Med 2007; 35(5): 705-11.

Nantel J, Termoz N, Vendittoli P A, Lavigne M, Prince F. Gait patterns after total hip arthroplasty and surface replacement arthroplasty. Arch Phys Med Rehabil 2009; 90(3): 463-9.

Narvani A A, Tsiridis E, Nwaboku H C, Bajekal R A. Sporting activity following Birmingham hip resurfacing. Int J Sports Med 2006; 27(6): 505-7.

Oxblom A, Hedlund H, Nemes S, Brismar H, Felländer-Tsai L, Rolfson O. Patient-reported outcomes in hip resurfacing versus conventional total hip arthroplasty: a register-based matched cohort study of 726 patients. Acta Orthop 2019; 90(4): 318-23.

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.

Saint-Maurice P F, Troiano R P, Bassett D R Jr, Graubard B I, Carlson S A, Shiroma E J, Fulton J E, Matthews C E. Association of daily step count and step intensity with mortality among US adults. JAMA 2020; 323(12): 1151-60.

Schmalzried T P, Shepherd E F, Dorey F J, Jackson W O, dela Rosa M, Fa’vae F, McKellop H A, McClung C D, Martell J, Moreland J R, Amstutz H C. The John Charnley Award: Wear is a function of use, not

Acta Orthopaedica 2020; 91 (5): 576–580

time. Clin Orthop Relat Res 2000; (381): 36-46.

Terwee C B, Bouwmeester W, van Elsland S L, de Vet H C, Dekker J. Instruments to assess physical activity in patients with osteoarthritis of the hip or knee: a systematic review of measurement properties. Osteoarthritis Cartilage 2011; 19(6): 620-33.

Thewlis D, Bahl J S, Fraysse F, Curness K, Arnold J B, Taylor M, Callary S, Solomon L B. Objectively measured 24-hour activity profiles before and after total hip arthroplasty. Bone Joint J 2019; 101-B(4): 415-25.

Tudor-Locke C, Craig C L, Aoyagi Y, Bell R C, Croteau K A, De Bourdeaudhuij I, Ewald B, Gardner A W, Hatano Y, Lutes L D, Matsudo S M, Ramirez-Marrero F A, Rogers L Q, Rowe D A, Schmidt M D, Tully M A, Blair S N. How many steps/day are enough? For older adults and special populations. Int J Behav Nutr Phys Act 2011; 8: 80.

Vail T P, Mina C A, Yergler J D, Pietrobon R. Metal-on-metal hip resurfacing compares favorably with THA at 2 years followup. Clin Orthop Relat Res 2006; 453: 123-31.

Verhaar J A. [The hard lesson of metal-on-metal hip implants]. Ned Tijdschr Geneeskd 2012 156(42): A5564.

Wendel-Vos G C, Schuit A J, Saris W H, Kromhout D. Reproducibility and relative validity of the short questionnaire to assess health-enhancing physical activity. J Clin Epidemiol 2003; 56(12): 1163-9.

Zywiel M G, Marker D R, McGrath M S, Delanois R E, Mont M A. Resurfacing matched to standard total hip arthroplasty by preoperative activity levels: a comparison of postoperative outcomes. Bull NYU Hosp Jt Dis 2009; 67(2): 116-19.

Acta Orthopaedica 2020; 91 (5): 581–586

581

A small number of surgeons outside the control-limit: an observational study based on 9,482 cases and 208 surgeons performing primary total hip arthroplasties in western Sweden Per JOLBÄCK 1–4, Emma NAUCLÉR 3, Erik BÜLOW 1,3, Hans LINDAHL 1,2, and Maziar MOHADDES 1,3 1 Department

of Orthopaedics, Institute of Clinical Sciences, The Sahlgrenska Academy, University of Gothenburg, Gothenburg; 2 Department of Orthopaedics, Skaraborg Hospital, Lidköping; 3 Swedish Hip Arthroplasty Register, Gothenburg; 4 Research and Development Centre, Skaraborg Hospital, Skövde, Sweden Correspondence: per.jolback@vgregion.se Submitted 2019-12-03. Accepted 2020-05-06.

Background and purpose — Feedback programs relating to surgeon levels have been introduced in some orthopedic quality registers around the globe. The aim of an established surgeon feedback program is to help surgeons understand their practice and enable an analysis of their own results.

There is no surgeon feedback program in Sweden in the orthopedic quality registers and there is a fear that a feedback system might pinpoint surgeons as poor performers, partly due to patient case mix. As a step prior to the introduction of a future possible feedback program in Sweden, we assessed the variation in the occurrence of adverse events (AE) within 90 days and reoperations within 2 years between surgeons in western Sweden and explored the number of surgeons outside the control-limit following primary total hip arthroplasties (THAs).

Patients and methods — Patient data, surgical data, and information on the surgeons, relating to surgeries performed in 2011–2016, were retrieved from 9 publicly funded hospitals in western Sweden. Data from medical hospital records, the Swedish Hip Arthroplasty Register (SHAR) and a regional patient register located in western Sweden were linked to a database. Funnel plots with control-limits based on upper 95% and 99.8% confidence intervals (CI) were used to illustrate the variation between surgeons in terms of the outcome and to explore the number of surgeons outside the control-limit. Both observed and standardized proportions are explored. The definition of surgeons outside the controllimit in the study is a surgeon above the upper 95% CI. Results — The study comprised 9,482 primary THAs due to osteoarthritis performed by 208 surgeons, where 91% of the included primary THAs were performed by orthopedic specialists and 9% by trainees. The mean overall annual volume for all surgeons was 27. The observed overall mean rate for AEs within 90 days for all surgeons was 6.2% (5.8– 6.7) and for reoperations within 2 years 1.8% (1.7–2.2).

The proportion of surgeons outside the 95% CI was low for

both AEs (0–5%) and reoperations within 2 years (0–1%) in 2011–2016. The corresponding numbers were even lower for AEs (0–3%) but similar for reoperations (0–1%) after standardization for differences in case mix. In a sub-analysis when the number of surgeries performed was restricted to more than 10 primary THAs annually to being evaluated, almost half or more of all the surgeons were excluded from the annual analysis. The result of this restriction was that all surgeons outside the control-limit disappeared after standardization for both AEs and reoperations for all the years investigated. Considering the complete period of 6 years, less than 1% (1 high-volume surgeon for AEs and 2 highvolume surgeons for reoperations) after risk adjustments were outside the 95% CI, and no surgeons were outside the 99.8% CI. Interpretation — In a Swedish setting, the variation in surgeon performance, as measured by AEs within 90 days and reoperations within 2 years following primary THA, was small and 3% or less of the surgeons were outside the 95% CI for the investigated years after adjustments for case mix.

The risk for an individual surgeon to be regarded as having poor performance when creating surgeon-specific feedback in the SHAR is very low when volume and patient risk factors are considered.

In 1975, the 1st orthopedic quality register, the Swedish Knee Arthroplasty Register (Robertsson et al. 2000, Malchau et al. 2018), was started and, 4 years later, it was followed by the Swedish Hip Arthroplasty Register (SHAR) (Kärrholm 2010). These 2 quality registers have played an important role as models for the fair number of successful registers in other countries (Malchau et al. 2018). Today, almost all orthopedic registers publish an annual report with results aggregated at hospital level. Some of the registers have also developed pro-

© 2020 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 10.1080/17453674.2020.1772584

582

grams for providing surgeon-level feedback and benchmarking data with other surgeons (National Joint Register 2015, Australian Orthopaedic Association National Joint Replacement Registry 2017). The main aim of the feedback programs at surgeon level, hosted by quality registers, is to help surgeons understand their practice.

The models used for visualizing single surgeons and benchmarking between peers in the National Joint Register for England, Wales, Northern Ireland, the Isle of Man and the States of Guernsey (NJR) and the Australian Orthopaedic Association National Joint Replacement Registry (AOANJRR) is funnel plots (Spiegelhalter 2005). The funnel plot has been suggested to be an appropriate statistical technique for reporting surgeon outcomes (Walker et al. 2013). The AOANJRR adjusts surgeon-level data for patients’ age and sex but not for other factors, such as BMI, comorbidities, and smoking, which have been suggested to influence the risk of AE and reoperation following arthroplasty (Mantilla et al. 2002, Thörnqvist et al.

2014, Duchman et al. 2015, Singh et al. 2015, Lübbeke et al.

2016). As yet, none of the Swedish orthopedic registers has started a feedback program to provide individual surgeon data.

Little is known about the surgeon performance in a Swedish setting. Here, we describe the variation in outcomes of AE within 90 days and reoperations within 2 years following primary total hip arthroplasties (THAs) among surgeons in western Sweden and explore the number of surgeons outside the control-limit.

Patients and methods All primary THAs in patients with a diagnosis of osteoarthritis (OA) of the hip, performed in hospitals managed by the county council of western Sweden between 2011 and 2016, were included in the study (Figure 1). Hospital medical records, the SHAR and the regional patient register, Vega (hereafter only namned regional patient register) were used as data sources. A complete list of sources for the variables that were used, including confounders, is shown in the Supplementary data (Appendix). The link between hospital medical records and the SHAR was made using the 10-digit personal identity number (PIN), the name of the hospital, and the date of surgery. If divergent information was obtained from the SHAR and the hospital medical records, the information in the SHAR was regarded as superior. The linked dataset, containing information from hospital medical records and the SHAR, was subsequently forwarded to the regional patient register to add all AEs and the data were made anonymous by replacing the PIN with a unique identifier.

Swedish Hip Arthroplasty Register The aim of the SHAR is to register all primary THAs and reoperations performed in Sweden (Kärrholm 2010). Although participation is voluntary for both hospitals and patients,

Acta Orthopaedica 2020; 91 (5): 581–586

Primary THAs performed in western Sweden 2007–2016 extracted from hospital medical records n = 15,086 Excluded Missing data on operating surgeon n = 29 Primary THAs with data on operating surgeon n = 15,057 Excluded Reason for surgery not OA in SHAR n = 120 Primary THAs for OA with data on operating surgeon n = 14,937 Excluded Surgery outside the investigated period n = 5,455 Primary THA included in the analysis n = 9,482

Figure 1. Flow chart.

the completeness and coverage of the SHAR have been high during the past few decades (Kärrholm et al. 2018). The variables recorded in the SHAR include patient factors such as age, sex, diagnosis for implantation, BMI, ASA classification, and technical details on the surgery, such as fixation technique and type of implant.

Vega­—a regional patient register The regional patient register was initiated in 2000. It is an aggregated database, containing records relating to all healthcare contacts (both publicly and privately funded) for all the residents in western Sweden. The population in western Sweden was approximately 1.6 million people in 2011 and 1.7 million people in 2016, which constitutes approximately 17% of all the residents in Sweden. The regional patient register provides information to the National Patient Register (NPR) (Ludvigsson et al. 2011). The PIN is used as the unique identifier of all entries in the regional patient register. The regional patient register contains details on the depiction of the caregiver at the point of contact, for example, the level of hospital or elective care, diagnoses, interventions, and length of stay in hospital.

AEs and reoperations The definition of AEs we used is the same as in the SHAR and was presented in their 2018 Annual Report (Kärrholm et al. 2018). It has also been used in previous studies (Berg et al.

2018, Jolbäck et al. 2019). The AE code list includes both surgical complications (local complications, secondary fractures,

Acta Orthopaedica 2020; 91 (5): 581–586

583

Table 1. Patient demographics for each year included in the study. Values are number (%) unless otherwise specified

2011 2012 2013 2014 2015 2016 n = 1,299 n = 1,320 n = 1,569 n = 1,774 n = 1,733 n = 1,787

Age, mean (range) 69 (21–95) 69 (22–93) 69 (20–92) 69 (26–97) 69 (17–97) 68 (21–94) Sex Male 538 (41) 561 (43) 637 (41) 730 (41) 743 (43) 733 (41) Female 761 (59) 759 (58) 932 (59) 1,044 (59) 990 (57) 1,054 (59) BMI, mean (range) 27 (16–49) 27 (15–53) 28 (10–63) 28 (16–61) 28 (15–72) 28 (16–68) Missing a 111 (9) 110 (8) 92 (6) 101 (6) 88 (5) 25 (1) ASA classification I 325 (25) 338 (26) 431 (28) 452 (26) 401 (23) 394 (22) II 745 (57) 766 (58) 903 (58) 1,056 (60) 1,043 (60) 1,083 (61) III 182 (14) 171 (13) 202 (13) 223 (13) 225 (13) 305 (17) IV 3 (0.2) 4 (0.3) 3 (0.2) 3 (0.2) 2 (0.1) 1 (0.1) Missing a 44 (3) 41 (3) 30 (2) 40 (2) 62 (4) 4 (0.2) Diagnosis for implantation Primary OA 1,226 (94) 1,249 (95) 1,498 (96) 1,681 (95) 1,651 (95) 1,694 (95) Secondary OA 73 (6) 71 (5) 71 5) 93 (5) 82 (5) 93 (5) Missing 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0)

Table 2. Mean annual surgeon volume during the period investigated Annual surgeon volume Year Mean (SD) 2011 2012 2013 2014 2015 2016 All years

21 (12) 24 (16) 24 (14) 31 (19) 28 (16) 34 (17) 27 (17)

BMI = body mass index, ASA = American Society of Anesthesiologists, OA = osteoarthritis.

1 hospital (with only 2 surgeons operating during the period) did report a low level of ASA classification and BMI. In 2016 it stopped producing primary THAs. Therefore, there is a large reduction in missing values in 2016 for ASA classification and BMI.

a

tendon ruptures in the lower extremities), and medical complications (thromboembolic events, myocardial infarction, pneumonia, gastroduodenal ulcers, acute kidney injury, and urinary retention). Reoperations are defined as any further surgery following the index surgery on the previously operated hip. All diagnoses for computing AEs were retrieved from the regional patient register, while the reoperations were retrieved from the SHAR. Statistics Continuous data are presented as means (SD), while categorical data are presented as proportions. Funnel plots were used to visualize variations between surgeons in the proportion of AEs and reoperations respectively and to explore the number of surgeons outside the control-limits. The control-limits are based on upper 95% and 99.8% confidence intervals (CI).

Wilson’s method suitable for low n was used to construct the CIs (Brown et al. 2001). The control-limits are dependent on the sample size; a small sample size increases the control-limits and a larger sample size reduces the limits (i.e., surgeons undertaking few surgeries will have a wider control-limit).

Both observed and standardized proportions are explored.

The standardized proportion was calculated for each surgeon as the ratio of the number of observed events divided by the number of expected events, multiplied by the overall proportion of events (Spiegelhalter 2005). Logistic regression with adjustments for patient risk factors was used to determine the probability of an event for a patient. The expected number of events for a surgeon was estimated by summing the predicted values of an event for the surgeon’s patients. The variables age, sex, ASA classification, BMI, and diagnosis for implan-

tation were assumed to be related to the outcome (Mantilla et al. 2002, Thörnqvist et al. 2014, Duchman et al. 2015, Singh et al. 2015, Lübbeke et al. 2016). For AEs within 90 days, all 5 predictors were included in the logistic regression model.

For reoperations within 2 years, the best-performing model included sex, ASA classification, and BMI. Only cases with complete data were included in the analysis. Patients undergoing simultaneous bilateral primary THAs were included as 1 surgery in the study; 43 patients underwent simultaneous bilateral primary THAs. Staged bilateral primary THAs were performed on 732 patients. We also performed a sub-analysis including surgeons performing more than 10 THAs annually (Walker et al. 2013).

SPSS version 25 (IBM Corp, Armonk, NY, USA) and R version 3.2.3 (R Foundation for Statistical Computing, Vienna, Austria) (R Core Team 2019) were used for the statistical analysis.

Ethics, funding, and potential conflicts of interest The study was approved by the Central Ethical Review Board in Stockholm (DNR Ö 9-2016). A research grant for the project was received from Skaraborgs Hospital research foundation. There is no conflict of interest.

Results The analysis included 208 surgeons from 9 public hospitals in western Sweden who performed the 9,482 primary THAs due to OA (Table 1). The categorization of the hospitals included was 1 university-regional hospital, 3 county hospitals, and 5

584

Acta Orthopaedica 2020; 91 (5): 581–586

Table 3. Annual number of surgeons outside the control-limit (above the upper 95% CI) in funnel plots due to AE within 90 days and reoperations within 2 years for all surgeons regardless of annual surgeon volume of primary THAs

Factor

2011 2012 2013 2014 2015 2016 n = 116 n = 121 n = 122 n = 116 n = 110 n = 98

Number of surgeons outside the control-limit for adverse events within 90 days Observed 5 2 6 3 1 0 Standardized a 3 1 3 1 1 0 Number of surgeons outside the control-limit for reoperations within 2 years Observed 0 0 0 1 1 1 Standardized b 0 0 0 1 0 1 a Age, sex, ASA classification, BMI, b Sex, ASA classification, and BMI

and diagnosis for implantation

ASA = American Society of Anesthesiologists, BMI = body mass index, CI = confidence interval, THAs = total hip arthroplasties.

Table 4. Annual number of surgeons outside the control-limit (above the upper 95% CI) in funnel plots due to AE within 90 days and reoperations within 2 years when only surgeons with 10 or more primary THAs annually are included in the analysis

Factor

2011 2012 2013 2014 2015 2016 n = 52 n = 41 n = 54 n = 54 n = 62 n = 51

Number of surgeons outside the control-limit for adverse events within 90 days

Observed 1 0 1 1 0 0 Standardized a 0 0 0 0 0 0 Number of surgeons outside the control-limit for reoperations within 2 years

Observed 0 0 0 0 0 1 Standardized b 0 0 0 0 0 0 For footnotes, see Table 3

rural hospitals (based on the SHAR’s categorization of hospitals). Of the 9,482 primary THAs included in the analysis, 8,636 (91%) were performed by orthopedic specialists and 846 (9%) by trainees. The mean annual volume of primary THAs for all surgeons and all years was 27 (SD 17). The mean annual volume of primary THAs varied (Table 2). The annual number of surgeons performing primary THAs decreased in the latter part of the period investigated (Table 3).

The overall mean rate for AEs within 90 days for all surgeons was 6.2% (SD 7.3), with a variation during the years between a minimum of 5.8% (year 2013) and a maximum of 6.7% (year 2011). The corresponding proportion for reoperations within 2 years was 1.7% (2011) to 2.2% (2016), with an overall mean rate of 1.8% (SD 3.9%).

During the years 2011–2016, there were few surgeons outside the upper 95% CI. The year with the highest number of surgeons outside the 95% CI for AEs within 90 days was 2013, there were 6 surgeons outside the control-limit and, after standardization for case mix, only 3 surgeons remained outside the limit. The proportion of surgeons outside the 95% CI during the years investigated varied between 0% and 5%.

Figure 2. Funnel-plots for AE within 90 days (top panel) and reoperation within 2 years (bottom panel) with the observed and standardized proportions overlaying. The green line is the mean value for the outcome of interest. The yellow line is the 95% CI and the red line is the 99.8% CI. Each dot represents one surgeon. Red dots are the observed proportion and blue dots are the standardized proportion.

The proportions of surgeons outside the 95% CI for reoperations within 2 years were also small, with variations between 0% and 1% annually (min–max) when examining both the observed and standardized proportions.

The result of the sub-analysis, when we included surgeons performing more than 10 primary THAs annually, showed that the surgeons who were outside the control-limit for AEs within 90 days were reduced by more than half (observed) and disappeared when standardization were made for case mix (Table 3). For reoperations within 2 years, all the surgeons outside the control-limit disappeared in the sub-analysis, apart from 1 surgeon in 2016, but, after standardization, this remaining surgeon also disappeared (Table 4).

Considering the complete period of 6 years, less than 1% (1 high-volume surgeon for AEs and 2 high-volume surgeons for reoperations) after risk adjustments were outside the 95% CI, and no surgeons were outside the 99.8% CI (Figure 2).

Acta Orthopaedica 2020; 91 (5): 581–586

Discussion Less than 3% of the surgeons were outside the upper 95% CI in this study for both AEs within 90 days and reoperations within 2 years, not only after adjustments for differences in patients’ characteristics but also before any standardization was made. The overall mean rates of both AEs and reoperations in the study are similar to the national average for elective primary THAs in Sweden (Kärrholm et al. 2018).

All the confounders we were able to adjust for are known from earlier studies to influence AEs and the risk of reoperation (Mantilla et al. 2002, Thörnqvist et al. 2014, Duchman et al. 2015, Singh et al. 2015, Lübbeke et al. 2016). However, there could be unknown confounders not available in this study that might affect the outcome following primary THA. The number of surgeons outside the control-limit due to reoperations within 2 years for patients undergoing surgery between 2015 and 2016 needs to be interpreted with some caution, as the follow-up period for these cases is shorter (0.5–1.5 years) than for the surgeries performed in 2011–2014.

The small number of surgeons outside the control-limit for both AEs and reoperations might be an effect of primary THA surgery being a highly standardized procedure in Sweden, but it might also be an effect of the long tradition of quality registers in Sweden providing feedback at hospital level. Primary THA surgeons in Sweden follow the recommendations given by the SHAR and the relatively large proportion of cemented THAs (Mäkelä et al. 2014, Kärrholm et al. 2017), with a fairly small number of different prostheses accounting for the majority of operations that are reported, may also contribute to the excellent outcomes.

We included only primary THAs due to primary and secondary OA. OA is the most common reason for primary THAs, where four-fifths have OA as a reason for implantation (Kärrholm et al. 2016) during the period of the study. Thus, some of the surgeons included in the study might produce a higher annual volume of primary THAs for reasons other than implantation for OA (e.g., fracture, inflammatory arthritis, femoral head necrosis, childhood disease). The experience from these other primary THAs might contribute to improved outcomes following all THAs for these surgeons and the same improvement in outcomes might be seen between surgeons performing both revision THAs and primary THAs. The control-limits in our funnel plots are based on CIs, as is the case in the AOANJRR’s feedback system. The AOANJRR has had a lower limit of 50 performed surgeries since the start of the feedback system. We examined the surgeons’ results every year and a fairly large number of surgeons performed only a few operations every year. The number of surgeons outside the control-limit in the study must be interpreted with caution, as we have included all surgeons, regardless of annual surgical volume, and this might increase the uncertainty.

585

We chose to present individual surgeon variations in funnel plots with control-limits based on CIs using Wilson’s method.

The choice of method for constructing control-limits is sensitive when the volume of annual surgeries is low. Wilson’s method was chosen because of the low annual surgeon volume.

However, the number of surgeons outside the control-limits should be interpreted with caution when exploring the variation between surgeons with a low annual volume.

When we excluded surgeons performing 10 or fewer primary THAs, almost half or more than half of the surgeons were excluded from the analysis. However, despite this halving of the number of surgeons, we executed the sub-analysis and the findings in this sub-analysis not only halved the number of surgeons, it also reduced or removed the number of surgeons outside the control-limit for both AEs and reoperations. Perhaps there is a “lower volume issue” that needs to be considered in order to make a reliable comparison between surgeons, and not only a problem with the case mix in terms of differences.

Only 9% of the primary THAs was operated on by trainees.

This small number of procedures performed by trainees might reflect the trainee education system in Sweden where almost all hospitals educate their own trainees. We can only speculate as to whether trainees are more likely to be outside the control-limit than trained surgeons. The reason for this is that in Sweden a trainee can apply for specialist certification in orthopedics at the Swedish National Board of Health and Welfare after fulfilling the requirements of the orthopedic trainee program at any time of the year. Therefore, a surgeon could have been both a trainee and orthopedic specialist during the same year. However, trainees or newly certified specialists are more probably likely to be low-volume surgeons, and thereby have an increased risk of being regarded as poor performers compared with more experienced surgeons (Ravi et al. 2014, Koltsov et al. 2018).

The small number of surgeons outside the control-limit for both AEs and reoperations in our study might be in conflict with the development of an individual surgeon feedback program following primary THAs in Sweden. However, there might be other aspects and benefits of an individual surgeon feedback program rather than presenting surgeons outside the control-limits, such as general information on individual surgeons’ practice, a substitute for former clinical follow-up visits to the operating surgeon, etc. Further research is needed to explore whether there are other aspects, benefits, or doubts from the surgeons’ point of view on the development of a feedback program.

One strength in this study is that we have been able to adjust for BMI and ASA classification. These 2 confounders are recorded in the SHAR’s standard collection of variables and it is therefore easy to add them to a possible future program for individual surgeon feedback.

One limitation in our study is that only primary THAs performed within the region of western Sweden were included.

586

Some of the surgeons involved in the study might have had temporary or partial employment, having performed primary THAs outside the region investigated. Due to the terms of employment laws in Sweden, it is very uncommon for surgeons to perform surgeries for multiple employers. We anticipated that the limited number of surgeons operating outside the region of western Sweden would not influence our conclusions.

Our study also shares the same limitation as all observational studies using administrative data. Both changes in practice during the study period and local trends, as well as differences in registration, might occur between the included hospitals during the period investigated. The regional patient register we used has not been validated on its own, but it provides data to the NPR. The Swedish National Inpatient Register (IPR) is part of the NPR. The IPR has been validated and contains 99% of all hospital discharges (Ludvigsson et al.

2011). We used a definition of AEs and reoperations requiring hospital admission. We therefore believe that our data are robust and our conclusions are valid.

In summary, the variation in surgeon performance, as measured by AEs within 90 days and reoperations within 2 years following primary THA, was small and 3% or less of the surgeons were outside the 95% control-limit for the years investigated after adjustments for case mix. The risk for an individual surgeon to be regarded as having poor performance when creating surgeon-specific feedback in the SHAR is very low when volume and patient risk factors are considered.

Supplementary data The Appendix is available as supplementary data in the online version of this article, http://dx.doi.org/10.1080/17453674.

2020.1772584 PJ had the original idea for the study and prepared the first version of the manuscript. PJ and EN processed the data. All the authors took part in the planning of the study, the analysis and interpretation of the data, and the writing of the manuscript.

Acta thanks Gary Hooper and Stein Atle Lie for help with peer review of this study.

Australian Orthopaedic Association National Joint Replacement Registry (AOANJRR). Annual Report; 2017. Available from https://aoanjrr.sahmri. com/annual-reports-2017 Berg U, Bülow E, Sundberg M, Rolfson O. No increase in readmissions or adverse events after implementation of fast-track program in total hip and knee replacement at 8 Swedish hospitals: an observational before-and-after study of 14,148 total joint replacements 2011–2015. Acta Orthop 2018; 89(5): 522-7.

Brown D L, Cai T T, DasGupta A. Interval estimation for a binomial proportion. Stat Sci 2001; 16(2): 101-17.

Acta Orthopaedica 2020; 91 (5): 581–586

Duchman K R, Gao Y, Pugely A J, Martin C T, Noiseux N O, Callaghan J J. The effect of smoking on short-term complications following total hip and knee arthroplasty. J Bone Joint Surg Am 201597(13): 1049-58.

Jolbäck P, Rolfson O, Cnudde P, Odin D, Malchau H, Lindahl H, Mohaddes M. High annual surgeon volume reduces the risk of adverse events following primary total hip arthroplasty: a registry-based study of 12,100 cases in Western Sweden. Acta Orthop 2019; 90(2): 153-8.

Koltsov J C B, Marx R G, Bachner E, McLawhorn A S, Lyman S. Risk-based hospital and surgeon-volume categories for total hip arthroplasty. J Bone Joint Surg Am 2018; 100: 1203-8.

Kärrholm J. The Swedish Hip Arthroplasty Register (www.shpr.se). Acta Orthop 2010; 81(1): 3-4.

Kärrholm J, Lindahl H, Malchau H, Mohaddes M, Rogmark C, Rolfson O. The Swedish Hip Arthroplasty Register Annual Report 2015; 2016.

Kärrholm J, Lindahl H, Malchau H, Mohaddes M, Rogmark C, Rolfson O. The Swedish Hip Arthroplasty Register Annual Report 2016; 2017.

Kärrholm J, Mohaddes M, Odin D, Vinblad J, Rogmark C, Rolfson O. The Swedish Hip Arthroplasty Register Annual Report 2017; 2018.

Lübbeke A, Zingg M, Vu D, Miozzari H H, Christofilopoulos P, Uckay I, Harbarth S, Hoffmeyer P. Body mass and weight thresholds for increased prosthetic joint infection rates after primary total joint arthroplasty. Acta Orthop 2016; 87(2): 132-8.

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.

Malchau H, Garellick G, Berry D, Harris W H, Robertson O, Kärrholm J, Lewallen D, Bragdon C R, Lidgren L, Herberts P. Arthroplasty implant registries over the past five decades: development, current, and future impact.

J Orthop Res 2018; 36(9): 2319-30.

Mantilla C B, Horlocker T T, Schroeder D R, Berry D J, Brown D L. Frequency of myocardial infarction, pulmonary embolism, deep venous thrombosis, and death following primary hip or knee arthroplasty. Anesthesiology 2002; 96(5): 1140-6.

Mäkelä K T, Matilainen M, Pulkkinen P, Fenstad A M, Havelin L I, Engesaeter L, Furnes O, Overgaard S, Pedersen A B, Kärrholm J, Malchau H, Garellick G, Ranstam J, Eskelinen A. Countrywise results of total hip replacement: an analysis of 438,733 hips based on the Nordic Arthroplasty Register Association database. Acta Orthop 2014; 85(2): 107-16.

National Joint Register (NJR). NJR Clinician Feedback User Guide; 2015.

Ravi B, Jenkinson R, Austin P C, Croxford R, Wasserstein D, Escott B, Paterson J M, Kreder H, Hawker G A. Relation between surgeon volume and risk of complications after total hip arthroplasty: propensity score matched cohort study. BMJ 2014; 348: g3284.

R Core Team. R: A language and environment for statistical computing.

Vienna, Austria: R Foundation for Statistical Computing; 2019. Available from https://www.R-project.org/.

Robertsson O, Lewold S, Knutson K, Lidgren L. The Swedish Knee Arthroplasty Project. Acta Orthop Scand 2000; 71(1): 7-18.

Singh J A, Schleck C, Harmsen W S, Jacob A K, Warner D O, Lewallen D G. Current tobacco use is associated with higher rates of implant revision and deep infection after total hip or knee arthroplasty: a prospective cohort study. BMC Med 2015; 13: 283.

Spiegelhalter D J. Funnel plots for comparing institutional performance. Stat Med 2005; 24(8): 1185-202.

Thörnqvist C, Gislason G H, Kober L, Jensen P F, Torp-Pedersen C, Andersson C. Body mass index and risk of perioperative cardiovascular adverse events and mortality in 34,744 Danish patients undergoing hip or knee replacement. Acta Orthop 2014; 85(5): 456-62.

Walker K, Neuburger J, Groene O, Cromwell D, van der Meulen J. Public reporting of surgeon outcomes: low numbers of procedures lead to false complacency. Lancet 2013; 382: 1674.77.

Acta Orthopaedica 2020; 91 (5): 587–592

587

The incidence of pelvic fractures and related surgery in the Finnish adult population: a nationwide study of 33,469 patients between 1997 and 2014 Pasi P RINNE 1, Minna K LAITINEN 2,3, Pekka KANNUS,4,5 and Ville M MATTILA 4,5 1 Vaasa

Central Hospital, Vaasa, Finland; 2 Helsinki University Hospital, Helsinki, Finland; 3 University of Helsinki, Helsinki, Finland; 4 School of Medicine, Tampere University, Tampere, Finland; 5 Department of Orthopaedics, Unit of Musculoskeletal Surgery, Tampere University Hospital, Tampere, Finland Correspondence: pasi.rinne@fimnet.fi Submitted 2019-11-04. Accepted 2020-04-25.

Background and purpose — Information on the epidemiological trends of pelvic fractures and fracture surgery in the general population is limited. We therefore determined the incidence of pelvic fractures in the Finnish adult population between 1997 and 2014 and assessed the incidence and trends of fracture surgery.

Patients and methods — We used data from the Finnish National Discharge Register (NHDR) to calculate the incidence of pelvic fractures and fracture surgery. All patients 18 years of age or older were included in the study.

The NHDR covers the whole Finnish population and gives information on health care services and the surgical procedures performed.

Results and interpretation — We found that in Finnish adults the overall incidence of hospitalization for a pelvic fracture increased from 34 to 56/100,000 person-years between 1997 and 2014. This increase was most apparent for the low-energy fragility fractures of the elderly female population. The ageing of the population is likely therefore to partly explain this increase. The annual number and incidence of pelvic fracture surgery also rose between 1997 and 2014, from 118 (number) and 3.0 (incidence) in 1997 to 187 and 4.3 in 2014, respectively. The increasing number and incidence of pelvic fractures in the elderly population will increase the need for social and healthcare services. The main focus should be on fracture prevention.

Pelvic fractures range from minor to major trauma and constitute about 3% to 8% of all fractures treated in hospitals (CourtBrown and Caesar 2006). The incidence of pelvic fractures has varied from 17 to 364/100,000 person-years (Melton et al. 1981, Ragnarsson and Jacobsson 1992, Lüthje et al. 1995, Kannus et al. 2000, Balogh et al. 2007, Andrich et al. 2015, Kannus et al. 2015, Verbeek et al. 2017). This wide range in incidence rates can be explained by different study populations with varying age, and by variations in study designs and follow-up periods. In previous studies, the incidence (n/100,000 person-years) of pelvic fractures was in the United States 37 between 1968 and 1977 (Melton et al. 1981), in Sweden 20 between 1976 and 1985 (Ragnarsson and Jacobsson 1992), in Finland 24 in 1988 (Lüthje et al. 1995), in the Finnish population aged 60 years or older 20 in 1970 and 92 in 1997 (Kannus et al. 2000), in Australia 23 between 2005 and 2006 (Balogh et al. 2007), in the German population aged 60 years or older 22 between 2008 and 2011 (Andrich et al. 2015), in the Finnish population aged 80 years or older 73 in 1971 and 364 in 2013 (Kannus et al. 2015) and in the Netherlands 14 between 2008 and 2012 (Verbeek et al. 2017).

In the 80 years and older population, the incidence of lowenergy pelvic fractures seems to be increasing (Kannus et al.

2015). Indeed, between 1997 and 2014, the incidence of acetabular fractures, especially low-energy acetabular fractures, rose in Finland (Rinne et al. 2017), whereas the incidence of high-energy acetabular fractures remained at the same level.

Notably, since 1997, the incidence of many other fall-related low-energy fractures, such as hip fractures, has decreased in Finland (Korhonen et al. 2013, Kannus et al. 2018).

Most pelvic fracture studies concentrate on surgical treatment, even though the majority of these fractures can be

© 2020 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 10.1080/17453674.2020.1771827

588

Acta Orthopaedica 2020; 91 (5): 587–592

treated nonoperatively (Osterhoff et al. 2019, Tornetta et al.

2019). Unstable and dislocated pelvic fractures often need surgery, while stable, non-displaced, or minimally displaced fractures, mostly occurring in elderly people after a simple fall, can usually be treated nonsurgically. At present, however, there is only limited information available regarding the incidence and trends of pelvic fracture surgery in the general population.

We assessed the incidence of pelvic fractures in the Finnish adult population between 1997 and 2014 and the incidence and trends of pelvic fracture surgery.

Hospitalization of patients with a primary or secondary diagnosis of pelvic fracture in Finland between 1997 and 2014 Sampling from the NHDR by ICD-10 codes S32.1, S32.2, S32.3, S32.4, S32.5, S32.7 and S32.8 Age ≥ 18 years n = 80,047 hospitalizations Study population First hospitalization of a single patient with a pelvic fracture n = 33,469 patients

Figure 1. Flowchart of the study population.

Patients and methods The Finnish National Hospital Discharge Register (NHDR) (THL 2015) is maintained by the National Institute for Health and Welfare, which, in turn, is a research and development institute under the Finnish Ministry of Social Affairs and Health, Helsinki, Finland. The main purpose of the NHDR is to collect data on patients and hospitalization events in Finland. The NHDR covers the entire well-defined Finnish population of 5.5 million (in 2014) people (Statistics Finland 2019). The production of this NHDR information is mandatory for all medical service providers in Finland, and the funding of these institutions is based on this information.

The Finnish NHDR contains data on age, sex, domicile of the patient, length of hospital stay, primary and secondary diagnoses, surgical procedures performed during the stay, and trauma mechanisms. Since 1996, the diagnoses have been coded according to the 10th revision of the International Classification of Diseases (ICD) (World Health Organization 2004). The surgical procedures are coded according to the Finnish Classification of Procedures (FCP), which is based on the Nordic Classification of Surgical Procedures (NCSP) (Committee NM-S 2011, Lehtonen 2013). Data from the NHDR were available until 2014.

Outcome variables The main outcome variable for this study was the number of hospitalized patients with a primary or secondary diagnosis of pelvic fracture (ICD-10 codes S32.1, S32.2, S32.3, S32.4, S32.5, S32.7, or S32.8) in Finland between 1997 and 2014.

All patients aged 18 years or older were included.

A secondary outcome variable was the number of surgical operations performed due to a pelvic fracture in Finland between 1997 and 2014 (FCP codes NEJ50, NEJ60, NEJ70, and NEJ86).

Study population All persons aged 18 years and older who were admitted to a hospital in Finland due to a pelvic fracture between 1997 and 2014 were included (Figure 1). The study population was categorized into 2 age groups: younger patients including adults

under the general retirement age (18–64 years) and elderly patients over the general retirement age (65 years and older).

Statistics All incidences were calculated and expressed annually. To compute the incidence ratios of pelvic fractures leading to hospitalization, the annual mid-populations were obtained from Official Statistics of Finland (Statistics Finland 2019), a computer-based national population register. Crude incidence (later called “incidence”) and the age-standardized incidence of pelvic fractures were calculated for both sexes and were expressed as number of cases per 100,000 person-years. In the calculation of the age-standardized incidence rates, age adjustment was carried out by direct standardization using the mean population of Finland between 1997 and 2014 as the standard population. For the entire study period, 1 person was counted only once. The number of pelvic fracture operations were calculated for the study population.

The number and incidence of fractures were calculated for the entire Finnish adult population (3,988,773 adult inhabitants in 1997, and 4,396,261 in 2014) (Statistics Finland 2019) and expressed by sex and the two age categories (18–64 years and 65 and older). 95% confidence intervals (CI) were calculated for the incidence numbers.

Ethics, funding, and potential conflicts of interest In Finland—by law—register studies without the use of biological material do not require ethics committee approval.

RECORD guidelines were followed (Benchimol et al. 2015).

This research did not receive any funding. None of the authors has any conflicts of interest to declare. Permission to use the NHDR data for this study (THL/1244/5.05.00/2016) was provided by the National Institute for Health and Welfare, Helsinki, Finland.

Results Pelvic fracture incidence In the following, “incidence” refers to n/100,000 person-years.

Acta Orthopaedica 2020; 91 (5): 587–592

589

Frequency

Incidence per 105 persons-years

Frequency

1,200

300

300

Female Male

Women ≥ 65 years Women 18–64 years Men ≥ 65 years Men 18–64 years

All ≥ 65 years All 18–64 years All Standardized

250

1,000

250

800

200

200

600

150

150

400

100

100

200

50

50

0

20

30

40

50

60

70

80

90

100

Age

Figure 2. Age and sex distribution of patients with pelvic fracture in Finland from 1997 to 2014.

0

1998 2000 2002 2004 2006 2008 2010 2012 2014

0

All NEJ50: Open reduction of fracture of pelvis NEJ60: Open reduction of fracture of acetabulum NEJ70: External fixation of fracture of pelvis NEJ86: Reoperation or late surgery of pelvis

1998 2000 2002 2004 2006 2008 2010 2012 2014

Year

Year

Figure 3. Incidence of pelvic fractures in Finland from 1997 to 2014.

Between 1997 and 2014, there were 80,047 hospitalizations in Finland with a pelvic fracture diagnosis. In the case of multiple hospitalizations of a single patient, only the first episode with pelvic fracture diagnosis during the study period was included. Thus, 33,469 patients with pelvic fracture were included in the analysis. 2,755 surgical procedures for pelvic fractures were performed during this time period, which includes 8.2% of the fractures. The annual number of new pelvic fracture hospitalizations was 1,345 in 1997 and 2,460 in 2014. The age distribution of pelvic fracture patients was clearly bimodal: the major mode comprised older patients and the minor mode comprised younger patients (Figure 2).

The incidence of pelvic fracture hospitalization increased from 34 (CI 32–36) to 56 (CI 54–58) between 1997 and 2014 (Figure 3, Table). The age-standardized incidence increased correspondingly from 38 to 49.

Sex and age The frequency and incidence of pelvic fractures was different between the sexes, and this difference was more evident in the elderly population. Of all pelvic fractures, 66% occurred in women, and 52% of fractures occurred in women aged 65 years or older. 68% of all pelvic fractures were in the elderly population.

The age-specific incidence increased from 121 (CI 113– 129) to 169 (CI 161–177) between 1997 and 2014 in persons aged 65 years and older. The incidence in the elderly female population was twice as high as that in the elderly male population. From 1997 to 2014, the incidence increased from 159 (CI 148–171) to 220 (CI 209–232) in the elderly female population and from 57 (CI 49–67) to 100 (CI 92–110) in the elderly male population. In the younger population, including both sexes, there was only a slight increase in the incidence of pelvic fractures from 14 (CI 12–15) to 19 (CI 17–20). The sex-specific increase in incidence in the younger population

Figure 4. Number of surgical treatments due to a pelvic fracture in Finland from 1997 to 2014.

Incidence of pelvic fractures (n/100,000 person-years)

Overall 18–64 years ≥ 65 years Male 18–64 years Male ≥ 65 years Female 18–64 years Female ≥ 65 years

Year 1997 n (95% CI)

Year 2014 n (95% CI)

34 (32–36) 14 (12–15) 121 (113–129) 17 (15–19) 57 (49–67) 10 (8–11) 159 (148–171)

56 (54–58) 19 (17–20) 169 (161–177) 20 (18–22) 100 (92–110) 17 (15–19) 220 (209–232)

was from 17 (CI 15–19) to 20 (CI 18–22) in males and from 10 (CI 8–11) to 17 (CI 15–19) in females during the same period.

Pelvic fracture surgery 2,755 operations for pelvic fractures (8.2% of all pelvic fractures) were performed between the years 1997 and 2014.

During the study period, the annual number and incidence of pelvic fracture operations (FCP classification codes NEJ50, NEJ60, NEJ70, and NEJ86 increased from 118 to 187 operations/year and from 3.0 to 4.3 operations (Figure 4).

The annual number and incidence of internal pelvic fracture fixations (NEJ50 or NEJ86) in both age groups and sexes increased from 30 to 82 operations/year and from 0.8 to 1.9 between 1997 and 2014. The annual number and incidence of internal pelvic fracture fixations was most common in young male patients (from 26 to 32 operations/year and from 1.6 to 1.9). In the younger female population, the number and incidence of internal pelvic fracture fixations was also increasing (from 2 to 16 operations/year and from 0.1 to 1.0). Between 1997 and 2014, the number and incidence of internal pelvic fracture fixations also increased in the elderly population from 2 to 34 operations/year and from 0.3 to 3.1. In the elderly male

590

population, the number and incidence rose from 1 to 10 and from 0.4 to 2.1. In the elderly female population, the number and incidence rose from 1 to 24 operations/year and from 0.2 to 3.9.

The annual number and incidence of acetabular fracture operations (NEJ60) increased from 56 to 100 operation and from 1.4 to 2.3 from 1997 to 2014. In the elderly population, the increase was more evident; the annual number of surgeries increased from 15 to 55 operations/year and the incidence of operations from 2 to 5, respectively. In the younger population, the number and incidence of acetabular fracture surgery remained at the same level (from 41 to 45/year and from 1.3 to 1.4) from 1997 to 2014.

The rate and incidence of external fixation (NEJ70) operations was highest between the years 2000 and 2002 (range 44–42 operations/year, incidence range 1.1–1.0). However, since then, the number of external fixation decreased dramatically. The annual number of external fixations decreased from 32 to 5 operations/year and the incidence of the operations decreased from 0.8 to 0.1 between 1997 and 2014. External fixation was more common in the treatment of pelvic fractures in the younger population. In the elderly population, the number of external fixations has always been low, and no significant changes occurred during the study period. The use of external fixation became sporadic at the end of the study period.

Discussion Pelvic fracture incidence The main finding in this nationwide study was that the overall incidence of pelvic fractures in adult Finns increased by 67% between 1997 and 2014. Pelvic fractures are common in high-energy-induced polytrauma patients. However, they are seen more frequently in elderly populations after a simple fall at ground level (Kannus et al. 2000, Rinne et al. 2017).

Young adults are more at risk for high-energy pelvic fractures, whereas low-energy pelvic fractures mostly occur in elderly patients (Ragnarsson and Jacobsson 1992, Balogh et al. 2007, Andrich et al. 2017).

Our results also show that both the number and the increase in incidence of pelvic fractures were highest in the elderly female population, the latter rising from 159 to 221/100,000 person-years. Verbeek et al. reported similar findings in the Dutch population (Verbeek et al. 2017). In our elderly male population, there was also a substantial increase in pelvic fracture incidence from 57 to 101/100,000 person-years during the study period, although the incidence was lower than that in females.

The exact reasons for the rise in the elderly population’s and especially elderly female population’s age-specific incidence are unclear. The comparison between the crude incidence rate and age-standardized incidence rate shows that the changed age distribution does not explain all of the increase in inci-

Acta Orthopaedica 2020; 91 (5): 587–592

dence. Pelvic fractures in the elderly population are mostly low-energy fragility fractures that are related to falling and osteoporosis. The age structure of the Finnish population is changing and the mean age of the population is becoming older: the mean life expectancy in Finland was rising constantly during the study period (Statistics Finland 2017).

People in Finland are also living longer at home. Impaired muscle strength, balance problems, physical inactivity, and degenerative joint diseases are common in the elderly population, and increase the risk of falling. Osteoporosis increases the fracture risk when falling. As pelvic fractures occur more frequently in the growing elderly population, the increase in the number and the incidence of fracture is expected to keep on rising.

The incidence of pelvic fractures in elderly people in Finland has been increasing for decades (Kannus et al. 2000) and is still increasing. Notably, since 1997, the crude incidence of hip fractures in elderly people in Finland has decreased, whereas it had been increasing for decades (Kannus et al.

2018). Kannus et al. (2000) showed that the incidence of hip fractures in the population aged 50 years or older in Finland increased considerably between 1970 and 1997 (from 160 to 440 fractures/100,000 person-years). Since then, however, the trend has been declining (340 fractures/100,000 person-years in 2015) (Kannus et al. 2006, Korhonen et al. 2013, Kannus et al. 2018). It might be suggested that a low-level fall in the elderly population would nowadays result more frequently in a fragility pelvic fracture instead of a hip fracture and this could partially explain the increase in incidence of pelvic fractures (Sullivan et al. 2014). Kannus et al. in their study in 2000 made a prediction for the annual number of first osteoporotic pelvic fractures in Finland to be about 1,400 by the year 2010 in the population aged 60 or older. In our present study, the number of pelvic fractures in the population aged 65 or older was 1,508 in the year 2010. Thus, the rate of the incidence of pelvic fractures in the elderly population was even higher than expected. The increasing number of pelvic fragility fractures cause challenges for the health and social care systems to be prepared and provide care and help for a rising number of fragility fracture victims. The main focus should be on fracture prevention by minimizing the risk factors of elderly people’s falls.

In our study, the overall incidence of pelvic fractures in adults was higher than previously presented (Melton et al.

1981, Ragnarsson and Jacobsson 1992, Lüthje et al. 1995, Balogh et al. 2007, Andrich et al. 2015, Verbeek et al. 2017).

Based on the study by Lüthje et al., the incidence of pelvic fracture hospitalization in Finland was 24/100,000 personyears in 1988. In a recent study from the Netherlands, the average annual incidence of pelvic fractures was 14/100,000 person-years between 2008 and 2012 (Verbeek et al. 2017).

In Sweden, the overall incidence of pelvic fractures requiring hospitalization was 20/100 et al. 000 person-years between 1976 and 1985 (Ragnarsson and Jacobsson 1992). The dif-

Acta Orthopaedica 2020; 91 (5): 587–592

ferent incidence rates in previous studies (Melton et al. 1981, Ragnarsson and Jacobsson 1992, Lüthje et al. 1995, Kannus et al. 2000, Balogh et al. 2007, Andrich et al. 2015, Kannus et al. 2015, Verbeek et al. 2017) may have been caused by different study populations, variation in treatment protocols, or selection bias. The accuracy and coverage of a trauma registry of a trauma center is based on reported patients covered by the registry. As pelvic fragility fractures are often treated outside trauma centers, the number of patients with pelvic fragility fractures reported in a trauma register might be underestimated. In addition, health insurance registry-based study populations might differ from the general population in socio-economic or occupational status (Andrich et al. 2015).

In nation-to-nation comparisons, the levels of the age-specific incidences of pelvic fracture depend much on selection of the age groups for each study. Also, the age structure of the background population varies. Thus, it is often difficult to compare the results of different studies. The Finnish National Hospital Discharge Register (NHDR) has the advantage of including the whole Finnish population. Moreover, the accuracy and coverage of the NHDR data are reported to be excellent (Mattila et al. 2008, Huttunen et al. 2014). Thus, the rates calculated from the register are not sample-based estimates but actual population results (Sund 2012).

Pelvic fracture surgery The annual number and incidence of pelvic fracture operations in Finland increased from 118 to 187/year and from 3.0 to 4.3 operations/100,000 person-years. This increase is most likely due to the increase in the incidence of pelvic fractures, which is most evident in the elderly population.

After 2008, the surgical treatment of pelvic fractures in Finland increased in the elderly population. This increase might be considered “minor” when compared with the increasing number and incidence of pelvic fractures. Nevertheless, new implants with anatomical design and a locking screw mechanism became available for pelvic fracture surgery during the study period and this might have had an impact on pelvic fracture operation rates.

External fixation (the Slätis frame) was used both as temporary and definitive fixation of pelvic fractures during the beginning of the study period. Since then, the role of external fixation in the treatment of pelvic fractures has diminished.

One reason for this might be the increased use of pelvic binders in emergency situations. The definitive operative treatment of pelvic fractures in Finland has been centralized mainly to our 5 level I trauma centers, whereas primary stabilization and emergency surgery were previously also performed in smaller hospitals. Surgical treatment protocols of pelvic fractures have changed towards performing the definitive fixation of a pelvic fracture as a primary fracture operation after pelvic binder as the primary treatment. Therefore, the role of external fixation as a temporary fixator for the stabilization period and transport to a level I trauma center has diminished.

591

A limitation of this study is related to the ICD-10 coding system, which is not entirely accurate with all pelvic fractures.

Codes S32.7 (multiple fractures of lumbar spine and pelvis) and S32.8 (fracture of other and unspecified parts of lumbar spine and pelvis) include both pelvic and lumbar spinal fractures, which may cause some inaccuracy in the register, and therefore the ICD-10 system is not entirely unambiguous in classifying pelvic fractures. The change in the practice of coding of the fractures might also affect the study results. The ICD-10 coding system has been used in the Finnish NHDR since 1996 while our study started in 1997. Thus, introduction of the ICD-10 coding system may have had some effect on the practice of coding at the beginning of our study period.

Another limitation of this study is related to the FCP coding system, which is not entirely accurate with the codes relating to open reduction of pelvic fracture (NEJ50) and reoperation or late fracture surgery of pelvis (NEJ86). However, as the same ICD-10 and FCP coding was used during the whole study period from 1997 to 2014, this possible problem with the accuracy of the classification did not affect the reported time trends in fracture incidence. Finally, as such, the NHDR cannot separate the hospitalization events of two different pelvic fractures in a single patient. We solved this by counting one patient only once. Also, it was possible that some patients in the early study population might have had a pelvic fracture prior to the study period but been sampled into the study population merely due to re-hospitalization of the fracture. We note that this type of case must have been rare.

In summary, we observed that in Finnish adults the overall incidence of hospitalization for a pelvic fracture increased from 34 to 56/100,000 person-years between 1997 and 2014.

The increase was especially apparent in low-energy fragility fractures among the elderly female population. The increasing number and incidence of pelvic fractures in the elderly population will increase the need for social and healthcare services.

The main focus should be on fracture prevention.

Conception and design: VMM. Collection and assembly of data: PR. Analysis: PR. Interpretation of the data: PR, VMM, ML. Drafting of the manuscript: PR. Critical revision and final approval of the article: PR, VMM, ML, PK. Acta thanks Marianne Hansen Gillam and Alma B Pedersen for help with peer review of this study.

Andrich S, Haastert B, Neuhaus E, Neidert K, Arend W, Ohmann C, Grebe J, Vogt A, Jungbluth P, Rosler G, Windolf J, Icks A. Epidemiology of pelvic fractures in Germany: considerably high incidence rates among older people. PLoS One 2015; 10(9): e0139078.

Andrich S, Haastert B, Neuhaus E, Neidert K, Arend W, Ohmann C, Grebe J, Vogt A, Jungbluth P, Thelen S, Windolf J, Icks A. Excess mortality after pelvic fractures among older people. J Bone Miner Res 2017; 32(9): 1789801.

Balogh Z, King K L, Mackay P, McDougall D, Mackenzie S, Evans J A, Lyons T, Deane S A. The epidemiology of pelvic ring fractures: a population-based study. J Trauma 2007; 63(5): 1066-73.

592

Benchimol E I, Smeeth L, Guttmann A, et al. The REporting of studies Conducted using Observational Routinely-collected health Data (RECORD) statement. PLoS Med 2015; 12(10):e1001885. doi: 10.1371/journal. pmed.1001885 Committee NM-S. NOMESCO Classification of Surgical Procedures (NCSP), version 1.16; 2011.

Court-Brown C M, Caesar B. Epidemiology of adult fractures: a review.

Injury 2006; 37(8): 691-7.

Huttunen T T, Kannus P, Pihlajamaki H, Mattila V M. Pertrochanteric fracture of the femur in the Finnish National Hospital Discharge Register: validity of procedural coding, external cause for injury and diagnosis. BMC Musculoskelet Disord 2014; 15: 98.

Kannus P, Palvanen M, Niemi S, Parkkari J, Jarvinen M. Epidemiology of osteoporotic pelvic fractures in elderly people in Finland: sharp increase in 1970–1997 and alarming projections for the new millennium. Osteoporos Int 2000; 11(5): 443-8.

Kannus P, Niemi S, Parkkari J, Palvanen M, Vuori I, Jarvinen M. Nationwide decline in incidence of hip fracture. J Bone Miner Res 2006; 21(12): 1836-8.

Kannus P, Parkkari J, Niemi S, Sievanen H. Low-trauma pelvic fractures in elderly Finns in 1970–2013. Calcif Tissue Int 2015; 97(6): 577-80.

Kannus P, Niemi S, Parkkari J, Sievanen H. Continuously declining incidence of hip fracture in Finland: analysis of nationwide database in 1970–2016.

Arch Gerontol Geriatr 2018; 77: 64-7.

Korhonen N, Niemi S, Parkkari J, Sievanen H, Palvanen M, Kannus P. Continuous decline in incidence of hip fracture: nationwide statistics from Finland between 1970 and 2010. Osteoporos Int 2013; 24(5): 1599-603.

Lehtonen J, Lehtovirta J, Mäkelä-Bengs P. THL-Toimenpideluokitus—THLÅtgärdskalssifikation—THL-classification of surgical procedures; 2013.

Lüthje P, Nurmi N, Kataja M, Heliövaara M, Santavirta S. Incidence of pelvic fractures in Finland in 1988. Acta Orthop 1995; 66(3): 245-8.

Mattila V M, Sillanpaa P, Iivonen T, Parkkari J, Kannus P, Pihlajamaki H. Coverage and accuracy of diagnosis of cruciate ligament injury in the Finnish National Hospital Discharge Register. Injury 2008; 39(12): 1373-6.

Melton L J 3rd, Sampson J M, Morrey B F, Ilstrup D M. Epidemiologic features of pelvic fractures. Clin Orthop Rel Res 1981; (155): 43-7.

Acta Orthopaedica 2020; 91 (5): 587–592

Osterhoff G, Noser J, Held U, Werner C M L, Pape H C, Dietrich M. Early operative versus non-operative treatment of fragility fractures of the pelvis: a propensity matched multicenter study. J Orthop Trauma 2019; 33(11): e410-5.

Ragnarsson B, Jacobsson B. Epidemiology of pelvic fractures in a Swedish county. Acta Orthop Scand 1992; 63(3): 297-300.

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.

Sullivan M P, Baldwin K D, Donegan DJ, Mehta S, Ahn J. Geriatric fractures about the hip: divergent patterns in the proximal femur, acetabulum, and pelvis. Orthopedics 2014; 37(3): 151-7.

Sund R. Quality of the Finnish Hospital Discharge Register: a systematic review. Scand J Publ Health 2012; 40(6): 505-15.

Statistics Finland. Deaths (e-publication). Officia Statistics of Finland, Helsinki; 2017. Available from http://www.stat.fi/til/kuol/2017/01/ kuol_2017_01_2018-10-26_tie_001_en.html Statistics Finland. Population structure (e-publication). Officia Statistics of Finland, Helsinki; 2019. Available from http://www.stat.fi/til/vaerak/ index_en.html THL. HILMO—Care register for health care (2015)—Definitions and instructions. Helsinki: Finnish Institute for Health and Welfare (THL). Social and Health Data Permit Authority Findata; 2015. www.thl.fi Tornetta P 3rd, Lowe J A, Agel J, Mullis B H, Jones C B, Teague D, Kempton L, Brown K, Friess D, Miller A N, Spitler C A, Kubiak E, Gary J L, Leighton R, Morshed S, Vallier D H. Does operative intervention provide early pain relief for patients with unilateral sacral fractures and minimal or no displacement? J Orthop Trauma 2019; 33(12): 614-18.

Verbeek D O, Ponsen K J, Fiocco M, Amodio S, Leenen L P H, Goslings J C. Pelvic fractures in the Netherlands: epidemiology, characteristics and risk factors for in-hospital mortality in the older and younger population. Eur J Orthop Surg Traumatol 2017; 28(2): 197-205.

World Health Organization. International statistical classification of diseases and related health problems. 10th revision, 2nd ed. ed. Geneva: World Health Organization; 2004.

Acta Orthopaedica 2020; 91 (5): 593–597

593

Development of the annual incidence rate of fracture in children 1980–2018: a population-based study of 32,375 fractures Andreas V LARSEN 1, Esben MUNDBJERG 1, Jens M LAURITSEN 1, and Christian FAERGEMANN 1,2 1 Accident

Analysis Group, Department of Orthopaedics and Traumatology, Odense University Hospital, Odense; 2 Section for Pediatric Orthopaedics, Department of Orthopaedics and Traumatology, Odense University Hospital, Odense, Denmark Correspondence: andreas.larsen@hotmail.com, uag@rsyd.dk Submitted 2019-10-17. Accepted 2020-04-25.

Background and purpose — Pediatric fractures are a common cause of morbidity. So far, no larger Danish study has described the development in the incidence rates. Therefore, we describe the development in the incidence rates of pediatric fractures in the time period 1980–2018 and the frequency of the most common type of fractures.

Patients and methods — This is a retrospective register study of all children aged 0–15 years with a fracture treated in the Emergency Department at Odense University Hospital, Denmark, between 1980 and 2018. For all cases, information on age, sex, date of treatment, diagnosis, and treatment was obtained from the patient registration system.

Based on official public population counts we estimated age and sex-specific annual incidence rates.

Results — 32,375 fractures were included. In the study period the incidence rate decreased by 12%. The incidence increased until the early 1990s. Thereafter incidence rates decreased until 2004–09, from then onward increasing towards the end of the study period. The highest age-specific incidence rate in boys of 522 per 10,000 person-years was at 13 years of age. In girls the age of the highest incidence rate decreased from 11 years in 1980 to 10 years in 2018.

Fracture of the lower end of the forearm, the clavicle, and the lower end of the humerus had the highest single fracture incidence rates.

Interpretation — The incidence rate of pediatric fractures decreased in the study period by 12%. The highest single fracture incidence rates were for fracture of the lower end of the forearm, the clavicle, and the lower end of the humerus. As the first longitudinal Danish study of pediatric fractures this study is a baseline for evaluating future interventions and future studies.

Injuries are one of the leading causes of morbidity in children and are the leading cause of admission to the healthcare system. In 2018 more than 300,000 children and adolescents were treated in Danish Emergency Departments because of injuries (Statistics Denmark 2018).

Previous studies have found that the overall risk of sustaining a fracture during childhood is 10–25% (Sibert et al. 1981, Landin 1983, Cheng and Shen 1993, Landin 1997). Additionally, the lifetime risk of sustaining a fracture is 27% and 42% for girls and boys respectively (Landin 1997). In a study from Hong Kong the lifetime risk of hospitalization due to a pediatric fracture was 7% (Cheng and Shen 1993).

Studies have demonstrated variations in the incidence rate of childhood fractures. In Sweden an increase in the incidence rate of pediatric fractures was found from 1950 to 1979 (Landin 1983, 1997). A similar increase was found in Finland from 1967 to 1983 (Mäyränpää et al. 2011). Conversely, during the 1980s the incidence rate decreased in these countries (Tiderius et al. 1999, Mäyränpää et al. 2011). Since 1993 different trends have been found in incidence rates as an increase was found in Sweden while a continued decrease was found in Finland (Hedström et al. 2010, Mäyränpää et al. 2011).

Only 1 recent study has described the changes in the incidence rates of pediatric fractures (Lempesis et al. 2017). This Swedish study found a decrease in the fracture incidence rate in girls from 1993–1994 to 2005–2006, but not in boys. So far, no other recent longitudinal population-based study of the variation in the incidence rates of pediatric fractures has been published.

We describe the development in the incidence rates of pediatric fractures in the period 1980–2018 and describe the frequency and changes of most common type of fractures.

© 2020 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 10.1080/17453674.2020.1772555

594

Patients and methods The population base for this study was the Odense Municipality in Denmark from January 1980 to December 2018. Odense Municipality is a well-defined geographical area with a population of 202,663 in 2018, mainly consisting of the city of Odense (Statistics Denmark 2020). The midyear population of children 0–15 years of age has decreased from 36,665 in 1980 to 33,751 in 2018 (Statistics Denmark 2019a and b, 2020).

The cases included are all children living in Odense Municipality aged 0–15 years with a bone fracture treated in the Emergency Department (ED) at Odense University Hospital (OUH) from 1980 to 2018. The ED at OUH is the only ED in the municipality. All registration was done by qualified staff and registered in a similar manner in the entire study period. In Denmark all registered inhabitants have a unique civil registration number (CPR number), which follows each individual for their entire life. The CPR number was used to identify individuals with recurrent contacts due to the same fracture.

In case of more than 1 contact for the same fracture only the 1st contact was registered. Trained physicians determined diagnosis according to the ICD system. ICD-8 was used from 1980 to 1993, and from 1994 onward the ICD-10 was used.

Due to differences between ICD-8 and ICD-10 only patients with ICD-10 coding were including for describing changes in fracture types to ensure quality of data.

For all cases information on age, sex, date of treatment, diagnosis, and type of treatment was obtained from the patient registration system.

We defined 4 age groups according to psychosocial and physiological steps of children’s development: 0–1 years were classified as infants, 2–5 years as pre-school children, 6–11 years as schoolchildren, and 12–15 years as adolescents.

Fractures were defined as any bone damage including epiphyseal fractures (Salter–Harris types), complete fractures, incomplete fractures (bowing fractures, greenstick fractures, and torus fractures), avulsions, and Tillaux/triplane fractures.

ICD-8 coding had a specific diagnosis for clinical fractures defined as growth zone tenderness combined with swelling without radiological verification. In the whole study period “clinical fractures” have been considered Salter–Harris type 1 fractures and treated as such. Those “clinical fractures” represented 14% of fractures 1980–1993 in all age groups and were excluded from the study. The ICD-10 coding has no specific diagnosis for “clinical fractures.” However, since the percentages of clinical fractures remained constant through all years in all age groups in the ICD-8 period 1980–93, we assumed that the percentage of “clinical fractures” remained constant in the entire study period. Therefore, we excluded the same proportion of fractures in the ICD-10 period 1994–2018, thereby excluding clinical fractures.

The data were grouped and analyzed in 5-year groups. For unknown reasons data from 1995 were corrupted in relation

Acta Orthopaedica 2020; 91 (5): 593–597

to identifying the patients, rendering them untrustworthy, thus they were excluded entirely from the dataset. Due to the change from ICD-8 to ICD-10 in 1994 we defined the following 4- or 5-year groups with ICD8: 1980–1984, 1985–1989, 1990–1993, and the following groups with ICD10: 1994– 1999 (1995 excluded), 2000–2004, 2005–2009, 2010–2014, 2015–2018.

Statistics Based on population counts we calculated age- and sex-specific annual incidence rates, with the statistics including 95% confidence intervals (CI). Population counts were extracted as population at risk from Statistics Denmark (Statistics Denmark 2019a and b, 2020). The incident rates were estimated as incidence densities in a dynamic cohort allowing subjects to enter and leave the cohort by migration (Rothman et al. 2008).

Children with fractures were not excluded from the population at risk. EpiData Analysis V2.2.2.185 (https://www.epidata. dk/) and Stata 10 (StataCorp, College Station, TX, USA) were used in all statistical analyses. P-values < 0.05 were considered significant.

Ethics, funding, and potential conflicts of interest This work is a retrospective register study, which is completely based on patient data existing beforehand. The study was approved by the National Data Protection Agency and the Danish Patient Safety Authority. This study recieved no funding. No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this work. None of the authors have any conflict of interest to declare.

Results During the 38-year study period the ED at OUH had 253,198 admissions of children due to injuries, of whom 32,375 had fractures. The yearly percentage of all contacts in which a fracture was diagnosed ranged from 12% in 1984 to 17% in 1994. There were 30,621 single fracture incidents. In 1,620 incidents the child had 2 fractures and in 134 incidents the child had 3 fractures. 59% were boys and 41% were girls. In the different age groups the distribution was 0–1 years 49% vs. 51%, 2–5 years 56% vs. 44%, 6–11 years 54% vs. 46%, and 12–15 years 66% vs. 34%. The median age was 11 years for both boys and girls.

The overall annual incidence rate of fractures was 255 (CI 252–258) per 10,000 person-years. The incidence rate was 215 (CI 211–219) for girls and 293 (CI 289–297) for boys. When comparing 1980–1984 with 2015–2018 the overall incidence rate decreased from 284 (CI 276–292) to 249 (CI 240–258) per 10,000 person-years, corresponding to a 12% decrease. The highest incidence rate of 296 (CI 287–304) per 10,000 personyears was found in 1985–1989, which was a 4% increase from

Acta Orthopaedica 2020; 91 (5): 593–597

595

incidence decreased with increasing age.

In girls, a similar increase was found until Age 12–15 years Age 12–15 years Age 6–11 years Age 6–11 years the peak at the age of 11 years. After that Age 2–5 years Age 2–5 years 500 500 the incidence decreased.

Age 0–1 years Age 0–1 years Figure 4 compares the age-specific 400 400 annual incidence rate in girls in the years 1980–1984 to 2015–2018. A similar pat300 300 tern in incidence rates was observed in the 2 time periods. However, in girls the peak 200 200 incidence was at 11 years in 1980–1984; this changed to 10 years of age in 2015– 100 100 2018. In boys the age of peak incidence rate remained unchanged in the study 0 0 period. In girls aged 13–15 we found 1980 1985 1990 1994 2000 2005 2010 2015 1980 1985 1990 1994 2000 2005 2010 2015 – – – – – – – – – – – – – – – – statistically significantly fewer fractures 1984 1989 1993 1999 2004 2009 2014 2018 1984 1989 1993 1999 2004 2009 2014 2018 than during 2015–2018 compared with Figure 1. Annual incidence rate of pediatric Figure 2. Annual incidence rate of pediatric 1980–1984. fractures for boys in different age groups. fractures for girls in different age groups. a 1995 excluded. a 1995 excluded.

The most common fractures were fracture in the lower end of the forearm involving radius and/or ulna (29%), clavIncidence per 10,000 person–years Incidence per 10,000 person–years 600 600 icle (6.4%), the lower end of humerus Boys Girls 1980–1984 (4.9%), the metatarsal bone (4.2%), the Girls Girls 2015–2018 500 500 metacarpal bone (4.1%), finger (3.1%), and the lower end of tibia and/or fibula 400 400 (2.5%). The overall annual incidence rates per 10,000 population/years were 77 300 300 (CI 75–79) for distal forearm fractures, 17 (CI 16–18) for fractures of the clavicle, 13 200 200 (CI 12–14) for fractures of the lower end of humerus, 11 (CI 10–12) for fractures 100 100 of the metatarsal bones, 11 (CI 10–12) for fractures of the metacarpal bones, 8.3 (CI 0 0 7.7–9.0) for finger fractures, and 6.7 (CI 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 6.2–7.3) for fractures of the lower end of Age Age tibia and/or fibula.

Figure 3. Age-specific annual incidence rates Figure 4. Age-specific annual incidence of pediatric fractures stratified by sex. rate of pediatric fractures in girls in the Fracture of the lower end of radius and/ years 1980–84 and 2010–14. or ulna was the most common fracture in all age groups (Table). A statistically sig1980–1984. From 1985–1989 to 2015–2018 the incidence rate nificant decrease was found in the fractures of the metacarpal decreased by 14%. bones in the age groups 6–11 and 12–15 years.

There was a statistically significant overall increase in fracture incidence rates during the late 1980s and then decreases during the late 1990s until 2004–2009, whereafter the inciDiscussion dence increases again (Figures 1 and 2).

It should be noted that the decrease in fracture incidence is During the study period the overall annual incidence rate not evenly spread throughout the different age groups during decreased statistically significantly by 12% from 1980–1984 the study period. The decrease was most pronounced in the to 2015–2018. The highest incidence rates were found in age group 12–15 years. This was in fact the only age group 1990–1993 or 1994–1999. with a statistically significant decrease with no overlapping Compared with studies from Sweden, we found a higher confidence intervals. The decrease was most prominent for fracture incidence rate during the 1980s and 1990s (Landin girls in this age group for whom the incidence rate decreased 1997, Tiderius et al. 1999). However, the overall decreasing by approximately a third from the early 1990s. trend in incidence rates was similar to that in our study. A In the group of boys, the age specific incidence rates recent Swedish study found an increasing fracture incidence increased until the age of 13 years (Figure 3). Thereafter the rate in boys from 1995 to 2006, and a continuously decreasIncidence per 10,000 person–years

Incidence per 10,000 person–years

600

600

a

a

596

Acta Orthopaedica 2020; 91 (5): 593–597

in the years 1998–2011 (Morgen et al. 2013), a period in which our study found both decrease and increase in Age 1994–1999 2015–2018 fracture incidences. Fourth, children’s Fracture location ICD10 IR (95% CI) IR (95% CI) fractures have previously been related to socioeconomic status (Hippisley0–1 Lower end of radius and/or ulna DS52.5–DS52.6 26 (20–33) 23 (16–33) Cox et al. 2002). The Danish BNP has Lower end of tibia DS82.3–DS82.4 9.8 (6.3–15) 18 (12–25) been steadily increasing over the study Clavicle DS42.0 14 (6.6–15) 5.6 (2.7–10) period, which may contribute to the Lower end of humerus DS42.4 4.9 (2.5–8.6) 5.1 (2.3–9.6) 2–5 decrease. Additionally, safety equip Lower end of radius and/or ulna DS52.5–DS52.6 54 (48–62) 61 (53–70) ment has been more common over the Clavicle DS42.0 27 (23–33) 24 (19–29) study period, i.e., wrist and elbow pro Lower end of humerus DS42.4 25 (20–30) 17 (13–22) Lower end of tibia and/or fibula DS82.3–DS82.4 9.9 (7.2–13) 11 (7.9–12) tectors. Notably, comparing the 2004– 6–11 09 with 2015–18 there was a tendency Lower end of radius and/or ulna DS52.5–DS52.6 117 (109–126) 118 (109–128) for increasing incidence. This was Lower end of humerus DS42.4 16 (13–19) 14 (11–18) Clavicle DS42.0 13 (10–16) 12 (9.8–16) statistically significant in both sexes in Metacarpal bone DS62.2–DS62.4 16 (13–19) 6.9 (4.9–9.6) age groups 2–5 years and 6–11 years 12–15 as well as in girls in the age group 0–1 Lower end of radius and/or ulna DS52.5–DS52.6 85 (77–95) 98 (87–110) Metacarpal bone DS62.2–DS62.4 31 (26–36) 17 (13–27) years.

Clavicle DS42.0 17 (13–22) 19 (14–24) The highest incidence rate for boys Finger DS62.5–DS62.6 18 (14–22) 12 (7.7–15) and girls in our study was 13 and 11 years, respectively. This corresponds to findings of other studies (Landin ing incidence in girls (Lempesis et al. 2017). A Danish study 1983, 1997, Hedström et al. 2010, Rennie et al. 2007, Meling found similar results in 1988 concerning distal forearm frac- et al. 2009, Lempesis et al. 2017). We also found that the peak tures with a higher incidence in Denmark compared with incidence rate in girls decreased from 11 years in 1980–1984 to Sweden (Kramhøft and Bødtker 1988).

10 years in 2015–2018. However, this was not statistically sigContrary to our findings, a Northern Swedish study found nificant. The decrease could correlate with the earlier onset of an increase in the overall incidence rate of pediatric fractures puberty in Danish girls, thus strengthening the theory that peak in the early 2000s (Hedström et al. 2010). In another Swed- fracture incidence in children correlates with maximum growth ish study from Malmö the incidence rate of fractures in girls rate and changes in behavior and interests at onset of puberty decreased in the early 2000s whereas the incidence rate in (Aksglaede et al. 2009). However, the decrease in fracture inciboys remained unchanged (Lempesis et al. 2017). dence was statistically significant for girls aged 13 to 15.

The increase in the incidence rates in the 1980s may be due The most common fractures found in this study correspond to changes in leisure activities, as well as more traffic injuries well to findings in previous studies (Landin 1997, Tiderius et (Landin 1983). The decrease in the incidence rate in our study al. 1999, Rennie et al. 2007, Hedström et al. 2010, Mäyränpää can be explained by several factors. First, in the late 1980s et al. 2011). Previous studies found that fracture of the distal nurses began visiting homes in Denmark when a child was radius and/or ulna is the most common fracture, representing born in order to secure the surroundings in an attempt to pre- 23–33% of all fractures, followed by phalangeal fractures of vent home accidents (Accident Analysis Group 1983). Second, the hand, clavicle fractures, fractures of the lower end of the changes in children’s activities and interests during the study humerus, and fractures of the ankle. period may have led to the decreasing incidence rate. The daily The strengths in this study are the long study period with time spend on watching television increased during the 1990s, continuous data in a geographically well-defined municipaland computer and internet usage increased during the early ity, the valid population data, and the good registration prac2000s, which may have led to reduced time involving risky tice in the ED. There are some limitations to the study design. activities (Pilgaard 2008, Bille 2008, Lubans et al. 2010). Third, The annual incidence rates described in our study include only obesity has been suspected to increase the risk of fractures. An cases requiring medical attention in the Emergency DepartItalian study found that obese children have an increased frac- ment at Odense University Hospital. Possible selection bias ture risk at ages 6–12 years. The authors suspect reduced bone arises as we have no information available regarding the total density as a result of reduced physical activity, biomechani- number of cases who seek medical attention via general praccal factors, and vitamin D deficiency to be the reason for the titioners or in neighboring hospitals 45 km away. However, increased fracture risk (Ferro et al. 2018). However, a Danish an earlier study found that very few patients travel for treatstudy found that the prevalence rates of obesity among chil- ment (Lauritsen 1987). Additionally, the number of general dren have plateaued and even shown a tendency for decline practitioners treating fractures is negligible because they do Incidence rate (IR) per 10,000 population–years of the 4 most common fractures in each age group in the study periods 1994–99 and 2015–18

Acta Orthopaedica 2020; 91 (5): 593–597

not have access to radiographic equipment. As this was a register-based study, data may be subject to errors in registration. However, all information was registered by the trained staff and diagnosis determined by trained physicians. In 1994 the registration of fractures changed from ICD-8 to ICD-10 coding system. To ensure the quality of the data, only data from 1994 and onward (ICD-10) were used for evaluation of fracture location. However, regarding the total number of fractures during the entire study period both ICD-8 and ICD-10 were used. We have no reason to believe that the change from IDC-8 to ICD-10 in 1994 has produced changes in the overall coding practice of fractures. Any fracture would be classified as such in both ICD-8 and ICD-10 and the coding differs only regarding the anatomical location; thus, the total numbers of fractures should not be influenced.

The chosen method of excluding the estimated proportion of clinical fractures in the ICD-10 period 1994–2018 may lead to a slightly under- or overestimation of the incidence rates in this period. We consider this method necessary to ensure comparability with neighboring countries and other studies.

However, in the entire study period tenderness of growth zones combined with swelling without radiographic sign of fracture has been treated and coded as a fracture according to the local guideline for pediatric fracture treatment. We have no reason to believe that practice changed in the study period.

Furthermore, there was no systematic change in the incidence rates between the ICD-8 registration in 1990–1993 and the ICD-10 registration in 1994–1999.

This is the first longitudinal Danish study of pediatric fractures. The study provides a baseline for comparing Danish incidence rates with neighboring countries and a baseline for evaluation of future interventions. Furthermore, the study gives important knowledge to emergency departments. However, more detailed studies of injury mechanism in order to prevent pediatric fractures are needed. This study focuses on the total number of fractures among children. However, only a certain percentage of children experience fractures during childhood, and some children experience more than 1 fracture incident. Further studies on the recurrence of fractures in children are needed. Furthermore, studies concerning the decreasing age of puberty among girls correlated to the considerable decrease in fracture incidence are needed.

AVL, EM: data analysis, manuscript writing, writing review, and editing.

JML: methodology, supervision. CF: conceptualization, methodology, project administration, supervision.

Acta thanks Torsten Backteman and Hanne Hedin for help with peer review of this study.

Accident Analysis Group. [Prevention of home accidents in children]. Odense University Hospital, Denmark; 1983 [in Danish].

597

Aksglaede L, Sørensen K, Petersen J H, Skakkebæk NE, Juul A. Recent decline in age at breast development: the Copenhagen Puberty Study. Pediatrics 2009; 123; e932.

Bille T. [The Danes’ cultural and leisure activities 2004]. National Danish Centre for Social Research, Copenhagen, Denmark; 2008 [in Danish].

Cheng J C, Shen W Y. Limb fracture pattern in different pediatric age groups: a study of 3,350 children. J Orthop Trauma 1993; 7(1): 15-22.

Ferro V, Mosca A, Crea F, Mesturino M A, Olita C, Vania A, Reale A, Nobili V, Raucci U. The relationship between body mass index and children’s presentations to a tertiary pediatric emergency department. Ital J Pediatr 2018; 44(1): 38.

Hedström E M, Svensson O, Bergström U, Michno P. Epidemiology of fractures in children and adolescents. Acta Orthop 2010; 81(1): 148-53.

Hippisley-Cox J, Groom L, Kendrick D, Coupland C, Webber E, Savelyich B. Cross-sectional survey of socioeconomic variations in severity and mechanism of childhood injuries in Trent 1992–7. BMJ 2002; 324:1132 Kramhøft M, Bødtker S. Epidemiology of distal forearm fractures in Danish children. Acta Orthop Scand 1988; 59(5): 557-9.

Landin L. Fracture patterns in children: analysis of 8,682 fractures with special reference to incidence and secular changes in a Swedish population 1950–1979. Acta Orthop 1983; 54(Suppl. 202): 3-109.

Landin L. Epidemiology of children’s fractures. J Pediatr Orthop Part B 1997; 6(2): 79-83.

Lauritsen J. [Injury and facility contacts: a random population sample].

Report No 9. Institute for Health Economy and Health Care. Odense Denmark: University of Odense; 1987 [in Danish].

Lempesis V, Rosengren B E, Nilsson J A, Landin L, Tiderius C J, Karlsson M K. Time trends in pediatric fracture incidence in Sweden during the period 1950–2006. Acta Orthop 2017; 88(4): 440-5.

Lubans D R, Morgan P J, Cliff D P, Barnett L M, Okely A D. Fundamental movement skills in children and adolescents: a review of associated health benefits. Sports Med 2010; 40(12): 1019-35.

Mäyränpää M K, Mäkitie O, Kallio P E. Decreasing incidence and changing pattern of childhood fractures: a population-based study. J Bone Miner Res 2011; 26(12): 2752-9.

Meling T, Harboe K, Søreide K. Incidence of traumatic long-bone fractures requiring in-hospital management: a prospective age- and gender-specific analysis of 4890 fractures. Injury 2009; 40(11): 1212-19.

Morgen C S, Rokholm B, Brixval C S, Andersen C S, Andersen L G, Rasmussen M, Andersen A N, Due P, Sørensen T I A. Trends in prevalence of overweight and obesity in Danish infants, children and adolescents: are we still on a plateau? PLoS One 2013; 8(7): e69860.

Pilgaard M. [The Danes’ sport activities 2007]. Copenhagen: Danish Institute for Sports Studies; 2008 [in Danish].

Rennie L, Court-Brown C M, Mok J Y, Beattie T F. The epidemiology of fractures in children. Injury 2007; 38(8): 913-22.

Rothman K J, Greenland, Lash T L. Modern epidemiology, 3rd ed. Philadelphia, PA: Wolthers-Kluwer and Lippincott Williams & Wilkins; 2008.

Sibert J R, Maddocks G B, Brown B M. Childhood accidents: an endemic of epidemic proportion. Arch Dis Child 1981; 56(3): 225-7.

Statistics Denmark. SKAD01 - Skadestuebesøg efter køn, diagnose, område, tid og alder. 2018. https://www.statistikbanken.dk/SKAD01. (date last accessed May 16, 2020) Statistics Denmark. FOLK1A 2008–2020. 2020. http://www.statistikbanken. dk/statbank5a/default.asp?w=1440 (accessed March 13, 2020).

Statistics Denmark. BEF1A 1979-2006. 2019a. http://www.statistikbanken. dk/bef1a (accessed March 13, 2020).

Statistics Denmark. BEF1A07 2005–2009. 2019b. http://www.statistikbanken.dk/statbank5a/default.asp?w=1440 (accessed March 13, 2020).

Tiderius C J, Landin L, Düppe H. Decreasing incidence of fractures in children: an epidemiological analysis of 1,673 fractures in Malmö, Sweden, 1993–1994. Acta Orthop Scand 1999; 70(6): 622-6.

598

Acta Orthopaedica 2020; 91 (5): 598–604

Time trends in pediatric fractures in a Swedish city from 1950 to 2016 Erika BERGMAN, Vasileios LEMPESIS, Jan-Åke NILSSON, Lars JEPHSSON, Björn E ROSENGREN, and Magnus K KARLSSON

Clinical and Molecular Osteoporosis Research Unit, Department of Clinical Sciences and Orthopedics, Lund University, Skåne University Hospital, Malmö, Sweden Correspondence: erika.bergman@med.lu.se Submitted 2020-04-06. Accepted 2020-05-22.

Background and purpose — As previous studies indicate time trends in pediatric fracture incidence, we followed the incidence in a Swedish city between 1950 and 2016.

Patients and methods — Malmö city, Sweden had 322,574 inhabitants in 2015. We used diagnosis registry, charts, and radiographs of the only city hospital to classify fractures in individuals < 16 years in 2014–2016, and compared these with data from 1950–2006. We used joinpoint regression to analyze time trends and present results as mean annual percentage changes (APC). Differences between periods are described as incident rate ratios (IRR). To describe uncertainty, 95% confidence intervals (CI) are used.

Results — During 2014–2016 the pediatric fracture incidence was 1,786 per 105 person-years (boys 2,135 and girls 1,423). From 1950 onwards age-adjusted fracture incidence increased until 1979 in both boys (APC +1.5%, CI 1.2–1.8) and girls (APC +1.6%, CI 0.8–2.5). The incidence remained stable from 1979 to 2016 (APC in boys 0.0%, CI –0.3 to 0.3 and in girls –0.2%, CI –1.1 to 0.7). Age-adjusted incidence 2014–2016 was higher than 2005–2006 in girls (IRR 1.1, CI 1.03–1.3), but not in boys (IRR 1.0, CI 0.9–1.1).

Interpretation — Fracture incidence was in girls higher in 2014–2016 than in 2005–2006. However, only with more than 2 measuring points are meaningful trend analyses possible. When we analyzed the period 1950–2016 with 17 measuring points and joinpoint regression, we found that fracture incidence increased in both sexes until 1979 but has thereafter been stable.

As many as 1/3 of all individuals have sustained a fracture before their 17th birthday (Cooper et al. 2004). However, as there are evident time trends in pediatric fracture incidence (Hedström et al. 2010, Mayranpaa et al. 2010, Jenkins et al.

2018, Koga et al. 2018), an update on occurrence is of interest. Published reports are contradictory. Some reports indicate an increase in pediatric fracture incidence, with higher incidences reported in 2007 than in 1998 in Sweden (Hedström et al. 2010), higher in 1999–2007 than in 1979–1987 in Japan (Koga et al. 2018) and higher in 2015 than in 2005 in Australia (Jenkins et al. 2018). Other authors report a lower incidence in 2005 than in 1983 (Mayranpaa et al. 2010). The discrepancies may be real but may also depend on varying ascertainment methods, time trend differences between regions, or that the years chosen for comparisons were not similar.

In the city of Malmö, Sweden, we have previously found that the unadjusted pediatric fracture incidence was higher in 1979 than in 1950 (Landin 1983), lower in 1993–1994 than in 1975–1979 (Tiderius et al. 1999) and similar in 2005–2006 and 1993–1994 (Lempesis et al. 2017). In contrast, no difference was apparent in the age-adjusted incidence when comparing 1993–1994 with 1976–1979 (Lempesis et al. 2017), indicating that changes in demography between these periods have influenced fracture incidences.

Therefore, we analyzed long-term time trends in pediatric fracture incidence. Secondarily, we compared the present fracture occurrence and fracture etiology with the last previous update a decade ago. Thus, we collected information on fracture occurrence during 2014–2016 in a similar fashion to previous data collection for the period 1950–2006.

© 2020 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 10.1080/17453674.2020.1783484

Acta Orthopaedica 2020; 91 (5): 598–604

Patients and methods Background information Malmö is a city in southern Sweden which in 2014 had a population of 318,107 inhabitants (58,585 < 16 years of age), 2015: 322,574 (60,519 < 16 years of age) and 2016: 328,494 (62,513 < 16 years of age) (Statistics Sweden 2017). The Skåne University Hospital in Malmö is the only hospital that provides trauma care for the population. There are few private clinics in the city, although some have orthopedic surgeons on their staff. However, these clinics evaluate only scheduled patients, and do not have any emergency service capacity. Fractures in the city are thus diagnosed and treated at the Skåne University Hospital. If a patient with a fracture initially attends a primary care center, the family doctor refers the patient to the Department of Radiology at Skåne University Hospital for evaluation. If the radiographic exam reveals a fracture, the patient is directly transferred to the Department of Emergency at the hospital, a visit that renders a fracture classification.

Ascertainment of fracture data All radiographs taken within the general health care system in the region of Skåne, Sweden are, since 2001, when the hospital changed from physical to digital radiographs, filed according to the unique 10-digit national personal identity number of the patient. All radiographs and reports are saved in a digital archive. This digital archive has been used to identify pediatric fractures 2005–2006 (Lempesis et al. 2017). Prior to 2001, all radiographs were organized according to diagnosis, year of injury, and anatomical region in an archive that could be used to identify fractures (Landin 1983, Tiderius et al. 1999). This archive has been used to create a hospital pediatric fracture database that include fractures from 1950, 1955, 1960, 1965, 1970, 1975–1979 (Landin 1983), and 1993–1994 (Tiderius et al. 1999).

In this study we used the same fracture ascertainment method as in 2005–2006 (Lempesis et al. 2017). We searched for pediatric fractures in the digital in- and outpatient diagnosis records at the Departments of Emergency, Orthopedics, Otorhinolaryngology, and Hand Surgery at the hospital. We included records with subsequent criteria: (I) ICD-10 fracture diagnoses S02.3– S02.4, S02.6–S02.9, S12.0–S12.2, S12.7, S22.0–S22.1, S32.0– S32.8, S42.0–S42.9, S52.0–S52.9, S62.0–S62.8, S72.0–S72.9, S82.0–S82.9, and S92.0–S92.9, (II) patient age < 16 years at the time of the fracture event, and (III) city residency in Malmö at the time of the fracture event. During the years 2014–2016, we identified 7,326 visits that fulfilled all 3 criteria. All visits for each patient were then examined in further detail in medical charts, radiographic reports, and referrals to validate fractures.

If the fracture diagnosis was ambiguous, radiographs were rereviewed by the orthopedic surgeon who conducted the review of cases in 2005–2006 (Lempesis et al. 2017), before any verified fracture was included in the study.

599

For verified fractures, we used the same registration protocol as in our 3 previous reports from Malmö (Landin 1983, Tiderius et al. 1999, Lempesis et al. 2017). We categorized bilateral fractures as 2 separate fractures, multiple fractures as independent fractures, with the exception of those in the fingers, toes, metacarpals, and 2 fractures in the same bone, which were categorized as a single fracture. Fractures of skull, sternum, teeth, nose, and ribs were not included. This ascertainment method with chart reviews allowed us to avoid double counting of fractures (due to multiple visits or numerous sequential radiographs).

Through the use of medical records we registered age, sex, date of fracture, fractured region, fractured side, trauma mechanism, trauma activity, and trauma severity (Landin 1983, Tiderius et al. 1999, Lempesis et al. 2017). Trauma severity was, as in previous reports, classified as slight, moderate, or severe. Slight injuries included falls from heights less than 0.5 meters (m) and most sport injuries. Moderate injuries involved falls from heights between 0.5 and 3 m, bicycle injuries, falls from swings and slides, and falls downstairs. Severe injuries comprised falls from heights above 3 m and traffic injuries with a motor vehicle involved. Trauma mechanism was classified into falls, mechanical force (including caught or squeezed, bites, blows, and hit by moving object), non-classifiable, or unknown. We also classified etiology of the fractures during 2014–2016 according to the NOMESCO Classification of External Causes of Injuries (Jørgensen et al. 2007).

Validation To validate the fracture ascertainment method, 1 of the authors (VL) reviewed all digital skeletal radiographs in Malmö on children under the age of 17 during the period from January 1, 2005 to February 28, 2005 and found 103 fractures. During the same 2 months, through the digital hospital in- and outpatient records used in this report, we also found 103 fractures.

Both methods identified 100 fractures, while 3 fractures were identified only by 1 of the methods, indicating a misclassification rate of 3% (Lempesis et al. 2017).

Statistics SPSS (IBM SPSS Statistics 24; IBM Corp, Armonk, NY, USA) and Microsoft Excel 2016 (Microsoft Corp, Redmond, WA, USA) were used for statistical calculations. Data are presented as numbers, proportions (%), and incidences per 105 person-years. We used direct standardization with the average population of the city of Malmö during the examined period (in 1-year classes) as reference, to estimate age-adjusted rates.

Fracture registration was grouped in the following years: 1950, 1955, 1960, 1965, 1970, 1975–1979, 1993–1994, 2005– 2006, and 2014–2016. All evaluated years during a 10-year period were included when calculating the incidences during that specific decade. Difference in fracture incidence between 2 decades are assessed by incident rate ratios (IRRs) with 95% confidence intervals (CI). Trend changes are estimated

600

Acta Orthopaedica 2020; 91 (5): 598–604

Table 1. Number of fractures, fracture incidences (per 105 person-years), and proportions (%) of fractures in all children, boys and girls aged 0–15 years in Malmö, Sweden, during the years 2014–2016

All children Number Incidence Proportion

Boys Number Incidence Proportion

Girls Number Incidence Proportion

All fractures 3,244 1,786 100 1,978 2,135 100 1,266 1,423 100

Axial 29 16 0.9 14 15 0.7 15 17 1.2

Face 12 7 0.4 7 8 0.4 5 6 0.4

Spine 8 4 0.2 3 3 0.2 5 6 0.4

Pelvis 9 5 0.3 4 4 0.2 5 6 0.4 Extremity 3,215 1,770 99 1,964 2,120 99 1,251 1,406 99

Upper extremity 2,520 1,388 78 1,562 1,686 79 958 1,077 76

Scapula 1 1 0.0 0 0 0.0 1 1 0.1 Clavicle 258 142 8.0 169 182 8.5 89 100 7.0 Humerus 358 197 11 183 198 9.3 175 197 14 Proximal 82 45 2.5 40 43 2.0 42 47 3.3

Diaphyseal 13 7 0.4 5 5 0.3 8 9 0.6 Distal 263 145 8.1 138 149 7.0 125 140 9.9 Forearm 1,288 709 40 791 854 40 497 558 39 Proximal 94 52 2.9 51 55 2.6 43 48 3.4 Diaphyseal 203 112 6.3 129 139 6.5 74 83 5.8 Distal 991 546 31 611 660 31 380 427 30 Hand 615 339 19 419 452 21 196 220 15 Carpal and metacarpal 175 96 5.4 145 157 7.3 30 34 2.4 Finger 440 242 14 274 296 14 166 187 13

Lower extremity 695 383 21 402 434 20 293 329 23 Femur 42 23 1.3 25 27 1.3 17 19 1.3

Proximal 4 2 0.1 0 0 0.0 4 4 0.3 Diaphyseal 25 14 0.8 19 21 1.0 6 7 0.5

Distal 13 7 0.4 6 6 0.3 7 8 0.6

Patella 10 6 0.3 6 6 0.3 4 4 0.3 Tibia 278 153 8.6 153 165 7.7 125 140 9.9 Proximal 42 23 1.3 30 32 1.5 12 13 0.9 Diaphyseal 176 97 5.4 92 99 4.7 84 94 6.6 Distal 60 33 1.8 31 33 1.6 29 33 2.3 Fibula 44 24 1.4 24 26 1.2 20 22 1.6

Proximal and diaphyseal 2 1 0.1 2 2 0.1 0 0 0.0 Distal 42 23 1.3 22 24 1.1 20 22 1.6 Foot 321 177 9.9 194 209 9.8 127 143 10

Mid- and hindfoot 11 6 0.3 7 8 0.4 4 4 0.3 Metatarsals 171 94 5.3 103 111 5.2 68 76 5.4 Toe 139 77 4.3 84 91 4.2 55 62 4.3

as mean (CI) annual percentage changes (APC) in fracture incidence, determined by use of joinpoint regression analysis (Kim et al. 2000, Joinpoint Regression Program 2019). We used fracture incidence for each evaluated year (in total 17 points) in the analyses. We considered p < 0.05 to represent a statistically significant difference.

Ethics, funding, and potential conflicts of interest The Regional Ethical Review Board in Lund approved the study (reference number 2016/1080). The study was organized and performed according to the Declaration of Helsinki.

Financial support for the study was provided by ALF, the Herman Järnhardt Foundation, the Greta and Johan Kock Foundation, and Region Skåne FoU. The sources of funding were not involved in the design or in the conduct of the study, in the interpretation of data, or in the writing of the manuscript. None of the authors have any competing interests.

Results Fracture epidemiology 2014–2016 During the years 2014–2016 we identified 3,244 fractures (61% in boys) during 181,617 person-years. This represents a fracture incidence of 1,786 per 105 person-years (2,135 in boys and 1,423 in girls) (Table 1). In this cohort, 2,743 children had sustained 1 fracture and 236 children (8%) had sustained 2 or more fractures. These 236 children (171 boys) with multiple fractures had sustained 501 fractures; 214 had sustained 2 fractures, 16 had sustained 3 fractures, 5 had sustained 4 fractures, and the remaining 5 fractures were sustained by a single child (falling injury).

The age-adjusted boy-to-girl incident rate ratio was 1.6 (CI 1.5–1.7). Before age 10 years, we found no statistically significant sex differences between boys and girls in 1-year age

Acta Orthopaedica 2020; 91 (5): 598–604

601

Incidence per 105 person-years

Incidence per 105 person-years

Incidence per 105 person-years

6,000

5,000

5,000

Boys Girls

1950/1955 2005–2006 2014–2016

5,000

1950/1955 2005–2006 2014–2016

4,000

4,000

3,000

3,000

2,000

2,000

1,000

1,000

4,000

3,000

2,000

1,000

0

0

1

2

3

4

5

6

7

8

9

10 11 12 13 14 15

0

0

1

2

3

4

5

6

Age

Figure 1. Fracture incidence per 105 personyears in boys and girls aged 0–15 years in Malmö, Sweden, during the years 2014–2016.

Data are presented with 95% confidence intervals (CI 95%) per 1-year age class.

7

8

9

10 11 12 13 14 15

0

0

1

2

3

4

5

6

7

8

9

10 11 12 13 14 15

Age

Age

Figure 3. Fracture incidence per 105 person-years in boys (left panel) and girls (right panel) aged 0–15 years in Malmö, Sweden, during the periods 1950/1955 (study start), 2005–2006 (earlier study), and 2014–2016 (present study), representing fracture incidence during different decades.

Table 2. Unadjusted and age-adjusted fracture incidence differences presented as incident rate ratios with 95% confidence intervals between periods of interest (2014–2016 in comparison with 1950/1955, 1960/1965, 1970/1975– 1979, 1993–1994, 2005–2006) in all children aged 0–15 years in Malmö, Sweden Nominator 2014–2016 2014–2016 2014–2016 2014–2016 2014–2016 Denominator 1950/1955 1960/1965 1970/1975–1979 1993–1994 2005–2006 Unadjusted Age-adjusted a Statistically

1.32 (1.24–1.41) a 1.40 (1.31–1.49) a

1.23 (1.16–1.31) a 1.34 (1.26–1.43) a

0.92 (0.88–0.96) a 0.95 (0.89–1.01) 1.03 (0.98–1.07) 0.98 (0.92–1.04)

0.97 (0.92–1.03) 1.05 (0.99–1.11)

significant changes.

classes (except at age 8). Boys from age 10, however, had a higher fracture incidence than girls, with the greatest discrepancy found at age 15, with a boy-to-girl fracture IRR of 4.3 (CI 2.9–6.2). The peak fracture incidence in boys was reached at age 13 (4,396 per 105 person-years) and in girls at age 11 (2,585 per 105 person-years) (Figures 1 and 2, see Supplementary data). The peak fracture incidence in 2014–2016 in boys occurred at similar age to 1950/1955 and 1 year earlier than in 2005–2006 and was at a similar magnitude to 2005–2006.

The peak fracture incidence in girls in 2014–2016 occurred about 1 year earlier than in 1950/1955 and 2005–2006 and the magnitude was higher than in 1950/1955 and 2005–2006 (Figure 3).

78% of all fractures occurred in the upper extremities, 21% in the lower extremities, and 1% in the axial skeleton. Of all fractures 31% were distal forearm fractures, 14% finger fractures, and 8% distal humerus fractures (Table 1). Fractures were 22% more common in the left than the right arm (IRR = 1.2, CI 1.1–1.3), while we found no apparent left to right difference in the legs (IRR = 0.9, CI 0.8–1.05).

We registered the highest incidences in May (214 per 105 person-years), September (209), and June (183). The lowest incidences were found in December (90 per 105 person-years), January (107), and November (109) (Figure 4, see Supplementary data).

Fracture epidemiology 1950–2016 Both the unadjusted and age-adjusted fracture incidences were in all children, as well as in boys and girls separately, higher in 2014–2016 than in 1950/1955 and 1960/1965. The unadjusted and age-adjusted incidence in girls in 2014–2016 was also higher than in 2005–2006. In contrast, the unadjusted fracture incidence in 2014–2016 was in all individuals, as well as boys and girls separately, lower than in 1970/1975–1979 (Table 2, Figure 5).

From 1950 to 1979 age-adjusted fracture incidence increased in boys by +1.5% per year (APC +1.5%, CI 1.2–1.8) and in girls by +1.6% (APC +1.6%, CI 0.8–2.5). From 1979 to 2016 fracture incidence was stable in both boys (APC 0.0%, CI –0.3 to 0.3) and girls (APC –0.2%, CI –1.1 to 0.7) (Figure

602

Acta Orthopaedica 2020; 91 (5): 598–604

Unadjusted incidence per 105 person-years

Age-adjusted incidence per 105 person-years

Age-adjusted rate

3,500

3,500

3,500

1.3 (1.2–1.4) a

1.4 (1.3–1.5) a 1.3 (1.2–1.4) a

1.2 (1.1–1.3) a 3,000

3,000

0.9 (0.9–0.97) a

1.0 (0.9–1.1)

0.9 (0.8–0.97) a

Boys Girls

2,500

3,000

1.0 (0.99–1.1)

0.9 (0.9–1.02)

1.0 (0.9–1.1)

Boys Girls

2,500

2,500

2,000

2,000

2,000

1,500

1,500

1,500

1.1 (1.01–1.2) a

1,000

1.1 (1.03–1.3) a

1,000

1.0 (0.9–1.1) 0.9 (0.9–0.99) a

500

1,000

1.0 (0.9–1.1)

500

1.3 (1.2–1.4) a

500

1.3 (1.2–1.5) a 1.4 (1.3–1.5) a

1.4 (1.2–1.5) a 0

0

0 1950

1960

1970

1980

1990

Boys 1950–1979: APC +1.5 (1.2 to 1.8) a 1979–2016: APC 0.0 (–0.3 to 0.3) Girls 1950–1979: APC +1.6 (0.8 to 2.5) a 1979–2016: APC –0.2 (–1.1 to 0.7)

1.0 (0.9–1.1)

2000

2010

2020

1950

1960

1970

1980

1990

2000

2010

Year

2020

Year

Figure 5. Unadjusted fracture (left panel) and age-adjusted fracture (right panel) incidence per 105 person-years in boys and girls aged 0–15 years in Malmö, Sweden, in 2014–2016 in comparison with 1950/1955, 1960/1965, 1970/1975–1979, 1993–1994, and 2005–2006. These periods are shown with thick lines, with line markers representing the number of years measured in the period.

An arrow indicates the 2 compared periods with incident rate ratios (IRR) in fracture incidence with 95% confidence intervals (95% CI) presented above the arrow. a indicates statistically significant changes.

6). We must, however, acknowledge that different fractures may have different time trends. The proportion of diaphyseal femur fractures for example halved from 1.6% in 1950–1979 to 0.8% in 2014–2016, whereas the proportion of diaphyseal forearm fractures almost doubled, from 3.4% in 1950–1979 to 6.3% in 2014–2016.

Fracture etiology 2014–2016 Fracture etiology is presented according to the Landin classification (Landin 1983) (Table 3, see Supplementary data) and the NOMESCO classification (Table 4, see Supplementary data). Using the NOMESCO classification 28% of all fractures occurred during sports activity and 29% during playing activity. The corresponding proportions according to Landin classification were 24% and 23%, respectively. Of all fractures, 16% occurred during ball games (Table 3, see Supplementary data).

Using NOMESCO the cause of injury can be further specified.

Of the ball game fractures, 78% were sustained during football and 7% during basketball. Among the fractures in the category contact sport (such as wrestling, boxing, and taekwondo) (Table 3, see Supplementary data) 35% occurred during taekwondo.

The most common location for fights was in school (with 29% of all fights). Of all fractures that occurred in school, 45% were sustained during sports activity and 18% during playing activity (Table 4, see Supplementary data).

Fracture etiology 1950–2016 The most common trauma mechanism in children during every time period was a fall on the same plane, and except for 1960/1965 the most common trauma-related activity was sports injuries (Table 3, see Supplementary data).

1950

1960

1970

1980

1990

2000

2010

2020

Year

Figure 6. Age-adjusted fracture incidence per 105 person years in boys and girls aged 0–15 in Malmö, Sweden, during the years 1950–2016, estimated with joinpoint regression. a indicates statistically significant changes.

Discussion Fracture epidemiology 2014–2016 We found a pediatric fracture incidence in 2014–2016 of 1,786 per 105 person-years, slightly lower than the fracture incidence of 2,050 per 105 person-years in Norway in 2010–2011 (Christoffersen et al. 2016). This difference could partly be due to the fact that the studies use different fracture ascertainment methods and that the countries have a differing winter climate and as well as different common leisuretime activities. This view is supported when finding a pediatric fracture incidence of 2,230–2,240 per 105 person-years in northern Sweden in 2005–2006 (Hedström et al. 2010) in comparison with 1,832 per 105 person-years in the south of Sweden during the same period (Lempesis et al. 2017). However, when comparing fracture incidences between different regions it is also important to take into account the proportion of boys and girls, distribution of children between cities and countryside, and proportion of immigration, all factors that may influence the incidence (Moon et al. 2016, Lempesis et al. 2017).

The peak fracture incidence in 2014–2016 occurred at age 13 for boys and age 11 in girls. Most studies infer that peak fracture incidence coincides with puberty (Faulkner et al. 2006, Hedström et al. 2010, Mayranpaa et al. 2010) and current data also infer that peak fracture incidence in 2014–2016 occurs in younger ages compared with previous decades. We speculate that one of the reasons for this could be that puberty seems to start earlier nowadays than historically (Brix et al. 2019). The current data also support previous publications that pediatric

Acta Orthopaedica 2020; 91 (5): 598–604

fractures occur more often in the upper than lower extremity (Tiderius et al. 1999, Lempesis et al. 2017), more often in the left than in the right arm, more often in the non-dominant than in the dominant arm (Tiderius et al. 1999, Anjum et al. 2017, Lempesis et al. 2017) and with a higher incidence during the warmer than during the colder season (Tiderius et al. 1999, Hedström et al. 2010).

Time trends Both boys and girls had higher unadjusted and age-adjusted incidence in 2014–2016 than in 1950/1955 and 1960/1965, indicating that changes in demographics could not entirely explain these higher fracture incidences in the recent period.

During these periods, there may also have been different proportions of children living in rural or urban areas, different sports habits, different patterns regarding seeking medical advice, and differing availability of health care (Landin 1983). The efficacy of traffic safety work in Sweden during the decades examined is supported in this study with lower proportions of traffic injuries during recent years. The large influx of pediatric immigrants in Sweden in recent years (Statistics Sweden 2020), the gradual decrease in physical activity during the most recent decades, and the dramatic increase in sedentary screen time activities since 2010 (Swedish Media Council 2017, Raustorp et al. 2019) may be other important contributors.

A problem with previous published studies (Landin 1983, Tiderius et al. 1999, Hedström et al. 2010, Mayranpaa et al.

2010, Lempesis et al. 2017, Koga et al. 2018) is that these have mainly compared 2 incidences between different years (or periods), not taking the natural variation between years (or periods) into account. For example, if in the study by Lempesis et al. (2017) the year 1970 is chosen to represent unadjusted fracture incidence in this decade, the fracture incidence 2005 was 9% higher, while if instead the year 1979 is used the fracture incidence in 2005 was 23% lower. That is, simply by choosing a different year within the same decade, the natural variation in fracture incidence could lead to different conclusion. A regression using multiple observations may thus be more suitable but may still be inadequate if changes in trends during the period of examination are prevalent. Joinpoint regression takes possible changes in trends into account and with this analysis we were able to examine the entire period from 1950 to 2016. We then found that the annual fracture incidence increased from 1950 to 1979 in both boys and girls, similar to the conclusions drawn by Landin (1983), who compared only 2 periods. However, we also found stable incidences thereafter (from 1979 to 2016) in both boys and girls, which contradicts the conclusions by Tiderius et al.

(1999), who compared fracture incidence in 1975–1979 with 1993–1994 and Lempesis et al. (2017) who compared fracture incidence in 1976–1979 with 2005–2006. This highlights that time-trend inferences should be interpreted with care when based on comparison between 2 defined periods.

603

Strengths and limitations Study strengths include the long study period with the same fracture classification system, the availability of official annual population data, the validation of the ascertainment method (used from 2005), inclusion of only objective verified fractures, and the possibility to preclude double counting of the same fracture. Another strength is inclusion of the more detailed NOMESCO classification as a complement to the Landin classification. The use of joinpoint regression analyses for estimation of time trends, instead of comparing incidences between 2 periods, is also a study strength.

Weaknesses include the use of 2 different ascertainment methods, 1 from 1950 to 1994 and 1 from 2005 to 2016, and that children living in Malmö but treated and followed up at other hospitals, in previous validation studies found to represent 3%–7% misclassification of fractures (Jonsson 1993), would not be registered by our method. The large proportion of unknown trauma (varying from 25% to 54% in different periods) is another weakness that led us to present only descriptive etiology data. A problem with the etiology data is that these are derived from the medical charts, referrals, and reports of radiographs with sometimes limited or conflicting information. It would have been advantageous to improve the classification system regarding the trauma information in the medical charts. It would also have been advantageous to adjust for changes in ethnicity in the population; however, such data were not available.

Conclusion The pediatric fracture incidence in Malmö, Sweden increased from 1950 to 1979, but thereafter was stable until 2016. The higher female incidence in 2014–2016 than in 2005–2006 indicates the necessity to follow fracture incidence continuously during the coming years to reveal any emerging time trends in fracture incidence. This study also highlights the problem when, as in most previous publications, drawing inferences regarding time trends without using multiple observation points.

Supplementary data Tables 3 and 4 and Figures 2 and 4 are available as supplementary data in the online version of this article, http://dx.doi.org/ 10.1080/17453674.2020.1783484

The study design was done by EB, VL, BR, and MK. EB collected the data.

VL re-reviewed some of the radiographs. LL, CJT, and VL undertook the previous collection of the historical data. EB made the calculations with statistical calculation assistance from JÅN and LJ. EB wrote the first draft of the manuscript. EB, VL, BR, MK, JÅN, and LJ completed the manuscript.

The authors would like to thank Lennart Landin and Carl-Johan Tiderius who conducted the studies 1950–1979 and 1993–1994 respectively.

Acta thanks Johannes Mayr and Jaakko Sinikumpu for help with peer review of this study.

604

Anjum R, Sharma V, Jindal R, Singh T P, Rathee N. Epidemiologic pattern of paediatric supracondylar fractures of humerus in a teaching hospital of rural India: a prospective study of 263 cases. Chin J Traumatol 2017; 20(3): 158-60. doi: 10.1016/j.cjtee.2016.10.007.

Brix N, Ernst A, Lauridsen L L B, Parner E, Stovring H, Olsen J, Henriksen T B, Ramlau-Hansen C H. Timing of puberty in boys and girls: a population-based study. Paediatr Perinat Epidemiol 2019; 33(1): 70-8. doi: 10.1111/ppe.12507.

Christoffersen T, Ahmed L A, Winther A, Nilsen O A, Furberg A S, Grimnes G, Dennison E, Center J R, Eisman J A, Emaus N. Fracture incidence rates in Norwegian children, The Tromso Study, Fit Futures. Arch Osteoporos 2016; 11(1): 40. doi: 10.1007/s11657-016-0294-z.

Cooper C, Dennison E M, Leufkens H G, Bishop N, van Staa T P. Epidemiology of childhood fractures in Britain: a study using the general practice research database. J Bone Miner Res 2004; 19(12): 1976-81. doi: 10.1359/ JBMR.040902.

Faulkner R A, Davison K S, Bailey D A, Mirwald R L, Baxter-Jones A D. Size-corrected BMD decreases during peak linear growth: implications for fracture incidence during adolescence. J Bone Miner Res 2006; 21(12): 1864-70. doi: 10.1359/jbmr.060907.

Hedström E M, Svensson O, Bergstrom U, Michno P. Epidemiology of fractures in children and adolescents. Acta Orthop 2010; 81(1): 148-53. doi: 10.3109/17453671003628780.

Jenkins M, Nimphius S, Hart N H, Chivers P, Rantalainen T, Rueter K, Borland M L, McIntyre F, Stannage K, Siafarikas A. Appendicular fracture epidemiology of children and adolescents: a 10-year case review in Western Australia (2005 to 2015). Arch Osteoporos 2018; 13(1): 63. doi: 10.1007/s11657-018-0478-9.

Joinpoint Regression Program, Version 4.7.0.0. Bethesda, MD: Statistical Methodology and Applications Branch, Surveillance Research Program, National Cancer Institute; February 2019.

Jonsson B. Life style and fracture risk. Thesis, University of Lund; Sweden 1993.

Jørgensen K, Nordisk Medicinal-Statistisk K. NOMESCO classification of external causes of injuries. Fourth revised ed.; 2007; cited March 2020.

Available from: https://norden.diva-portal.org/smash/get/diva2:1201255/ FULLTEXT01.pdf.

Kim H J, Fay M P, Feuer E J, Midthune D N. Permutation tests for joinpoint regression with applications to cancer rates. Stat Med 2000;

Acta Orthopaedica 2020; 91 (5): 598–604

19(3): 335-51 (correction: 2001; 20:655). doi: 10.1002/(sici)10970258(20000215)19:3<335::aid-sim336>3.0.co; 2-z.

Koga H, Omori G, Koga Y, Tanifuji O, Mochizuki T, Endo N. Increasing incidence of fracture and its sex difference in school children: 20 year longitudinal study based on school health statistic in Japan. J Orthop Sci 2018; 23(1): 151-5. doi: 10.1016/j.jos.2017.09.005.

Landin L A. Fracture patterns in children: analysis of 8,682 fractures with special reference to incidence, etiology and secular changes in a Swedish urban population 1950–1979. Acta Orthop Scand 1983; (Suppl 202): 1-109.

Lempesis V, Rosengren B E, Nilsson J A, Landin L, Tiderius C J, Karlsson M K. Time trends in pediatric fracture incidence in Sweden during the period 1950–2006. Acta Orthop 2017; 88(4): 440-5. doi: 10.1080/17453674.2017.1334284.

Mayranpaa M K, Makitie O, Kallio P E. Decreasing incidence and changing pattern of childhood fractures: a population-based study. J Bone Miner Res 2010; 25(12): 2752-9. doi: 10.1002/jbmr.155.

Moon R J, Harvey N C, Curtis E M, de Vries F, van Staa T, Cooper C. Ethnic and geographic variations in the epidemiology of childhood fractures in the United Kingdom. Bone 2016; 85: 9-14. doi: 10.1016/j. bone.2016.01.015.

Raustorp A, Froberg A. Comparisons of pedometer-determined weekday physical activity among Swedish school children and adolescents in 2000 and 2017 showed the highest reductions in adolescents. Acta Paediatr 2019; 108(7): 1303-10. doi: 10.1111/apa.14678.

Statistics Sweden. Child population/population in the city of Malmö in 1-year classes, December 31st 2014/2015/2016; 2017; cited March 2020. Available from: http://www.statistikdatabasen.scb.se/sq/83864.

Statistics Sweden. Pediatric immigration in Sweden years 2000–2019 in 1-year classes; 2020; cited March 2020. Available from: http://www.statistikdatabasen.scb.se/sq/84359.

Swedish Media Council. Kids and Media 2017; 2017; cited March 2020.

Available from: https://statensmedierad.se/download/18.7b0391dc15c38ff bccd9a238/1496243409783/Ungar%20och%20medier%202017.pdf.

Tiderius C J, Landin L, Duppe H. Decreasing incidence of fractures in children: an epidemiological analysis of 1,673 fractures in Malmo, Sweden, 1993–1994. Acta Orthop Scand 1999; 70(6): 622-6.

Acta Orthopaedica 2020; 91 (5): 605–610

605

Combined massive allograft and intramedullary vascularized fibula transfer: the Capanna technique for treatment of congenital pseudarthrosis of the tibia Stefanie C M VAN DEN HEUVEL 1, Hay A H WINTERS 1, Klaas H ULTEE 1, Nienke ZIJLSTRA-KOENRADES 2, and Ralph J B SAKKERS 2 1 Amsterdam UMC, location VUmc; 2 University Medical Center Utrecht, The Netherlands Correspondence: scm.vandenheuvel@amsterdamumc.nl Submitted 2019-09-06. Accepted 2020-04-27.

Background and purpose — Congenital pseudarthrosis of the tibia (CPT) is caused by local periosteal disease that can lead to bowing, fracturing, and pseudarthrosis. Current most successful treatment methods are segmental bone transport and vascularized and non-vascularized bone grafting. These methods are commonly hampered by discomfort, reoperations, and long-term complications. We report a combination of a vascularized fibula graft and large bone segment allograft, to improve patient comfort with similar outcomes.

Patients and methods — 7 limbs that were operated on in 6 patients between November 2007 and July 2018 with resection of the CPT and reconstruction with a vascularized fibula graft in combination with a bone allograft were retrospectively studied. The mean follow-up time was 5.4 years (0.9–9.6). Postoperative endpoints: time to discharge, time to unrestricted weight bearing, complications within 30 days, consolidation, number of fractures, and secondary deformities.

Results — The average time to unrestricted weight bearing with removable orthosis was 3.5 months (1.2–7.8).

All proximal anastomoses consolidated within 10 months (2–10). 4 of the 7 grafts fractured at the distal anastomosis between 6 and 14 months postoperatively. After reoperation, consolidation of the distal anastomosis was seen after 2.8 months (2–4). 1 patient required a below-knee amputation.

Interpretation — This case series showed favorable results of the treatment of CPT through a combination of a vascularized fibula graft and large bone segment allograft, avoiding the higher reintervention rate and discomfort with ring frame bone transport, and the prolonged non-weight bearing with vascularized fibula transfer without reinforcement with a massive large bone segment allograft.

Congenital pseudarthrosis of the tibia (CPT) is a rare disease affecting the development of the diaphysis of the tibia with a reported incidence ranging between 1 in 140,000 to 250,000 newborns (Ruggieri and Huson 2001, Horn et al. 2013). CPT is characterized by local periosteal disease, often leading to bowing and fracturing of the tibia and/or fibula, followed by the development of a pseudarthrosis (Stevenson et al. 1999).

The etiology behind CPT remains largely unelucidated.

Many theories regarding the influence of vascular, genetic, and mechanical factors have been proposed over the years, but none provides an entirely satisfactory explanation for the pathological features or its typical location (Hefti et al.

2000, Hermanns-Sachweh et al. 2005). However, there is a clear association with type 1 neurofibromatosis (NF1), as the prevalence of NF1 in CPT patients exceeds 50% (Van Royen et al. 2016).

The challenge in cases of pseudarthrosis in CPT is obtaining solid union of the tibia with minimal limb length discrepancy and angular deformity (Grill et al. 2000). The most used treatments today are resection of the CPT part of the bone and vertical bone transport or the use of a pedicled or free vascularized fibula graft (Kesireddy et al. 2018). A multinational study from Japan, which included both the Ilizarov technique combined with diaphyseal transfer and vascularized fibula grafting, found high rates of union among both treatment groups and concluded that both approaches should be considered (Ohnishi et al. 2005).

For the Ilizarov technique with diaphyseal transfer through proximal metaphyseal corticotomy and distraction at a distance from the dystrophic area, success rates between 50% and 90% have been reported (Paley et al. 1992, Ghanem et al. 1997, Romanus et al. 2000, Choi et al. 2011). Drawbacks of this technique are multiple interventions and prolonged

© 2020 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 10.1080/17453674.2020.1773670

606

Acta Orthopaedica 2020; 91 (5): 605â&#x20AC;&#x201C;610

discomfort of the patient due to many months of wearing a ring frame around the lower leg with pins and K-wires moving through muscle compartments.

Also, pin tract infections and surgery to induce healing at the docking side of the bone transport lead to restrictions in social and psychological functioning (Ramaker et al. 2000, Patterson 2006).

Studies on vascularized fibula grafts report union rates up to over 90% (Weiland et al. 1990, Erni et al. 2010).

Drawbacks of this method include the prolonged period of non-weight bearing due to the lack of primary mechanical strength, resulting in graft fractures requiring reoperations prior to consolidation and hypertrophy of the graft (Weiland et al. 1990, Bos et al. 1993, Romanus et al. 2000, Ohnishi et al.

2005).

In 1993, Capanna et al. reported on his â&#x20AC;&#x153;Capanna techniqueâ&#x20AC;? in the resection of bone tumors in which the use of a vascularized autograft is combined A B C with the use of a solid bone segment allograft in order to achieve instant staFigure 1. A. Postoperative image after Capanna procedure for CPT. B. 8-plates on both distal tibias bility with solid bone during consoli- to correct for ankle valgus at the donor and receptor sites. C. Cross-union between the distal dation and subsequent hypertrophy of anastomosis of graft and tibia and the remnant distal fibula. the vascularized fibula during growth (Capanna et al. 1993).

In order to avoid the drawbacks and discomfort of bone (prior interventions, type 1 neurofibromatosis), affected side, transport or vascularized fibula transfer without adding ini- and donor side. The study was performed in accordance with tial additional stability, we introduced this technique for the the STROBE statement. treatment of pseudarthrosis in CPT. Paley (2019) recently published the outcomes of his cross-union concept, report- Outcome measurements ing union in all 17 treated patients without refracturing with Outcome measurements were: time to discharge, time to unrehis latest technique, with follow-up to 11 years. If these out- stricted weight bearing with a removable orthosis, complicacomes prove to be reproducible, this will probably make the tions within 30 days, radiological consolidation (consolidation cross-union technique the gold standard for treating CPT. We of 3 cortices on the AP and lateral radiograph) of the proximal report a retrospective case series on the Capanna technique and distal anastomosis, number of fractures, time to fracture, in patients with CPT as reference for future strategies in this final alignment, and limb-length discrepancy. disease.

Surgical technique The affected tibia was exposed through an anteromedial incision and wide resection of the affected segment of the tibia Patients and methods with periosteum was performed, including all hamartomatous All 6 patients operated on in the University Medical Center tissue. The fibula was harvested in a standard fashion for free or Utrecht between November 2007 and July 2018 with resec- pedicled transplantation through a lateral approach (At 1999, tion of a congenital pseudarthrosis of the tibia and immediate Stevanovic et al. 1999). The use of bone morphogenic protein reconstruction with a vascularized fibula and a solid segment (BMP) was not included in the present treatment protocol. The bone allograft were retrospectively reviewed. Baseline char- harvested vascularized fibula was placed intramedullary in the acteristics were obtained from hospital charts and included segmental tibial defect. If necessary, reaming of the intramedage, sex, Paley classification of CPT, data on patient history ullary cavity was done. After secure intramedullary placement

Acta Orthopaedica 2020; 91 (5): 605–610

607

Table 1. Results

Table 2. Baseline characteristics

Primary Primary

proximal Refracture distal Refracture Procedures Patients union proximal union distal 7

6

7

0

3

4

of the fibula graft, the microvascular anastomoses were performed when using a free contralateral fibula. The allograft (Bonebank ETB-BISLIFE, Leiden, the Netherlands) was cut to the appropriate length to bridge the gap between proximal and distal tibia. A vertical slot was cut in the allograft allowing it to be placed around the fibula, leaving the vascular pedicle of the fibula in the slot. Fixation of the allograft was done with a large LCP plate for instant stability (Figure 1A). After closure of the surgical incisions, the leg was placed in a lower-leg cast for 6 weeks. After 6 weeks a walking brace was applied at least until consolidation was observed on follow-up imaging.

All patients received antibiotic prophylaxis and patients 3 and 6 additionally received thromboprophylaxis.

In the first 3 unilateral cases, we used the contralateral healthy fibula as a free vascularized graft. After having to use pedicled ipsilateral transplantation in a patient with bilateral CPT and encouraged by the results reported by Tan et al.

(2011), ipsilateral transplantation became first choice in the last 2 patients of this case series with unilateral CPT. Ethics, funding, and potential conflicts of interest The study received a “non-WMO” declaration from the Medical Research Ethics Committee of the University Medical Center Utrecht. No funding was obtained for this study and the authors have no conflicts of interest to declare.

Results (Table 1) Study population (Table 2) 7 limbs were operated in 5 male patients (1 unilateral CPT left tibia, 4 unilateral CPT right tibia), and 1 female patient (bilateral CPT). The patient with bilateral CPT had Paley type 4A of the right lower leg and Paley type 4B of the left lower leg.

The unilateral CPTs were one Paley type 1, one Paley type 3, and three Paley type 4A. Median age at the time of treatment was 5.5 years (3.4–16.5). Type 1 neurofibromatosis was present in 5 of the 6 patients.

3 patients (patients 1, 3, and 6) had been previously operated on unsuccessfully with cancellous bone grafting and placement of intramedullary rods.

The contralateral fibula was harvested and used as a vascularized free fibula in the reconstruction of 3 tibias. The ipsilateral proximal fibula was used as a pedicled fibula in the reconstruction of 4 tibias (3 patients). The dimensions of the allografts

Paley Procedure Age classifi- Previous Donor Patient (years) NF Side cation procedures site 1 1 5.5 2 2 3.4 3 3 a 12.7 4 3 a 16.5 5 4 3.0 6 5 5.1 7 6 14.0

Yes Right Yes Left Yes Right Yes Left Yes Right Yes Right No Right

4A 1 4A 4B 4A 3 4A

4 0 1 0 0 0 1

Contralateral Contralateral Ipsilateral Ipsilateral Contralateral Ipsilateral Ipsilateral

a Bilateral

were length 10 cm (6.6–12) and width (medial–lateral) 1.6 cm (1–2.3) and (anterior–posterior) 2.4 (range 1.4–3.3).

Postoperative outcomes (Table 3) Follow-up was between 0.9 and 9.6 years. Average time to discharge following the combined vascularized fibula and bone allotransplantation was 5 days (3–6). None of the patients suffered a complication within the first 30 days following the procedure. Average time to unrestricted weight bearing with a removable orthosis was 3.5 months (1.2–7.8). Consolidation of the proximal anastomosis was seen after a mean period of 6 months (2–10). 4 of the 7 grafts fractured at the distal anastomosis between 6 and 14 months postoperatively. After reoperation, consolidation of the distal anastomosis was seen after 2.8 months (2–4). One tibia fractured distally to the anastomosis 6 years after reconstruction.

In all 3 patients who had the vascularized fibula from the healthy contralateral side, a valgus deformity of the distal tibia developed due to the hypotrophy of the lateral side of the distal epiphysis. 2 of these 3 patients also developed a valgus in the distal tibia of the acceptor limb due to hypotrophy of the distal epiphysis. A valgus deformity of the proximal tibia occurred in one patient. The limb-length discrepancy at the latest follow-up was 11 mm (0–44). 1 patient (procedures 3 and 4) was followed to skeletal maturity (mean 13.6 years, 8–19).

Additional procedures No additional operations were needed for the union of the proximal anastomosis and only an average of 0.71 (0–1) additional procedures were needed for union of the distal anastomosis. The majority of the other procedures were guidedgrowth procedures in day care.

Patient 1, who had a contralateral vascularized fibula transfer, developed a bilateral valgus deformity in the epiphysis of the distal tibia. Both sides were treated with an 8-plate over the medial side of the epiphyseal plate 4 years after the initial procedure (Figure 1B). 3 years later, the 8-plates were removed and a distal epiphysiodesis was performed on the donor site to minimize the discrepancy in leg length.

608

Acta Orthopaedica 2020; 91 (5): 605–610

Table 3. Postoperative outcomes

Consolidation Months to Procedure Follow-up LOS a proximal consolidation Patient (years) (days) part tibia proximal

Months to Compliconsolidation cation distal < 30 days

Complication 1 > 30 days

Complication 2 > 30 days

1 1 9.6 5 Yes 8 13 – Bilateral valgus Leg length discrepancy

deformity ankle 2 2 9.4 6 Yes 5 13 – Bilateral valgus Leg length discrepancy

deformity ankle 3 3 7 6 Yes 5 4 b – Fracture distal tibia 4 3 3 3 Yes 10 – c – Fracture distal tibia Pseudarthrosis

and fracture plate requiring amputation 5 4 4.1 6 Yes 2 2 b – Persisting distal Valgus deformity

Pseudarthrosis proximal tibia 6 5 3.7 5 Yes 7 2 b – Fracture distal tibia 7 6 0.9 4 Yes 5 3 b – Fracture distal tibia

a Length of stay in hospital. b Consolidation after additional procedure. c Below-knee amputation after refracture.

Patient 2, who had a contralateral vascularized fibula transfer, underwent a second procedure 3 years after the initial procedure during which the LCP plate was removed because the end of the curve of the plate deviated from the bone due to growth of the distal tibia. Additionally, an 8-plate was placed over the distal tibial epiphyseal plate to treat a valgus deformity of the distal epiphysis of the tibia. Progressive migration of the distal fibula with instability of the ankle joint was seen, which was treated by repositioning of the distal fibula with a Taylor spatial frame and subsequent creation of a synostosis of the distal tibia and distal fibula. In addition, the 8-plate on the distal medial side of the left tibia was replaced as the screws had developed maximum divergence and an 8-plate was placed on the distal medial side of the right tibia (donor site) to treat the valgus of the ankle on the donor side.

Patient 3 had bilateral ipsilateral vascularized fibula transfers. The distal anastomosis of the left tibia fractured after 15 months. Fusion by creating a synostosis between the distal tibia and distal remnant fibula segment was not successful at the first attempt and the patient chose to have an amputation and below-knee prosthesis. Her mobility with the below-knee prosthesis increased to such a level that she fractured the earlier operated right leg at the level of the distal screw of the LCP plate. The fracture healed after removing the LCP plate, drilling both distal tibia and fibula, and using a large spongiosa graft from a Reamer Irrigator Aspirator (RIA; DePuy Synthes, Warsaw, IN, USA) procedure from the ipsilateral femur.

Latest follow-up showed a bony crossover between the distal tibia and the distal fibula.

In patient 4 (graft no. 5) a fracture of the distal anastomosis occurred 9 months after surgery. The fracture was treated with removal of the pseudarthrotic tissue, cancellous bone grafting from the iliac crest, and insertion of an intramedullary Rush nail resulting in successful consolidation. 3 years later, an 8-plate over the proximal tibial epiphyseal plate was placed to treat a valgus deformity.

Patient 5 had a fracture of the distal anastomosis of the graft and tibia 8 months after surgery. The fracture was treated with removal of the pseudarthrotic tissue and bone grafting from the iliac crest. Bone healing occurred after 8 weeks with a synostosis between the distal anastomosis of the tibia and the remnant distal fibula (Figure 1C).

Patient 6 (graft no. 7) had a fracture at the distal anastomosis between graft and tibia 6 months after surgery. The fracture was treated with replacement of the LCP plate and an RIA procedure of the femur to harvest spongiosa bone for crossunion between the distal anastomosis of the tibia and the remnant fibula.

Discussion This case series reports on the Capanna technique using a large segment of bone allograft in combination with a vascularized fibula graft after resecting part of the tibia with CPT. The technique was used in order to improve on the current techniques by avoiding the discomfort of external frames for bone transport, or prolonged non-weight bearing when using vascularized fibular grafts without additional bony stabilization.

It has been well documented that external fixation in children and adolescents has a significant physical and physiological impact, with studies reporting pain and consequent sleeping problems in approximately half of the patients (Ramaker et al. 2000). The complications related primarily to the use of an external device and include pin-track infection, loosening, breakage, refracture at the pin insertion site, neurovascular injury, axial deviation, osteopenia, and joint stiffness. Residual limb-length discrepancy and valgus deformity are commonly reported with an overall complication rate of 30–100% (Choi et al. 2011). In addition, a comprehensive review on the physiological effects of external fixation demonstrated depression to be universally evident to varying degrees (Patterson 2006).

Acta Orthopaedica 2020; 91 (5): 605–610

Standardized and validated quality of life measurements were not obtained in this retrospective study. However, we can reasonably assume that the minor discomfort of a lower leg cast does not have the reported outcomes associated with external fixation.

As compared with the technique of vascularized fibula grafts without the support of a massive allograft, the time to full-weight bearing has been reported to be as long as 18–24 months (Kalra and Agarwal 2012). Gilbert (1983) extended this period to up to 3 years, and Weiland et al. (1990) required their patients to wait until skeletal maturity was reached. In this study combining vascularized fibula grafts with a massive allograft, full weight bearing with a removable orthosis was already possible after 1.2 to 7.8 months. The significantly shortened time to weight bearing associated with the present technique underlines the benefits of combining the VFG with the massive allograft, particularly in this young cohort of patients.

One of the undesired side effects that was seen in 3 patients who had the vascularized fibular graft taken from the healthy contralateral side was the development of hypotrophy of the lateral side of the distal epiphysis of the tibia and subsequent valgus of the ankle joint. Valgus deformities at the donor site were also reported in previous studies (Weiland et al. 1990, Ohnishi et al. 2005). The compensatory asymmetric growth on the metaphyseal side of the growth plate, induced by the 8-plate, compensated for the valgus deformity at the epiphyseal side of the growth plate.

The undesirable side effect of hypotrophy of the distal epiphysis with subsequent valgus of the ankle joint in the healthy leg and the publication by Tan et al. (2011) made us switch from the contralateral to the ipsilateral donor leg, and thereby from free vascularized to a pedicled segment of the fibula. Obviously, the amount of available healthy fibula is likely to be smaller in the ipsilateral leg, as the fibula is also affected by the disease in most cases. This must be assessed, and if there is not enough healthy bone in the ipsilateral fibula a free vascularized graft from the contralateral leg is still indicated. In case of a pedicled fibula, the pedicle can be distally or proximally based depending on the position of the defect, the quality of the vascular pedicle, and the position of the healthy segment of the fibula.

The most important complication was the fracturing of the site of the distal anastomosis in 4 out of 7 grafts. Even a vascularized fibula bridging the distal anastomosis and stabilization with massive allograft and large fragment LCP plate does not always create adequate circumstances for primary solid bony fusion at this location. Therefore, we changed the protocol and added the creation of a synostosis between the distal anastomosis and the remnant of the distal fibula with a spongiosa graft from the ipsilateral femur to the initial procedure.

Although slightly different from the crossover technique by Paley (2019), we hope this will primarily strengthen the distal anastomosis and prevent fractures.

609

The fracture that appeared 6 years after the initial surgery was just distal of the most distal screw of the LCP plate. The question therefore arises as to whether removal of the LCP plate after successful consolidation would have prevented this fracture and should become standard protocol.

The use of bone morphogenic protein (BMP) was not included in the present treatment protocol, because the use of BMP in children is off-label in Europe.

All patients in our cohort underwent 1 or more additional procedure(s) during follow-up. No additional operations were needed for the union of the proximal anastomosis and only an average of 0.71 (0–1) additional procedures were needed for union of the distal anastomosis. Most other procedures were guided-growth procedures in day care to correct for a valgus deformity in the upper ankle joint. This is in line with previous studies on the use of vascularized fibula grafts in CPT patients showing residual valgus deformities in up to 80% of patients in the upper ankle joint (Weiland et al. 1990, Ohnishi et al. 2005). Previous studies involving free vascularized fibula grafts reported reintervention rates between 37% and 60% (Bos et al. 1993, Romanus et al. 2000).

The results of the European Paediatric Orthopaedic Society multicenter study by Grill et al. (2000) reported 340 patients who underwent 1,287 procedures (meaning 3.79 interventions per patient) and a fusion rate of 76%.

In summary, the Capanna technique for ipsilateral vascularized fibular transplantation needed only 1.2 interventions to achieve union in 6 out of 7 tibias with a relative short time to full weight bearing and seems therefore to be a more patientfriendly treatment than the conventional methods of bone transport with ring frames or vascularized fibula transfers that do not use the addition of a massive allograft in the reconstruction. We hope that the outcomes will be further improved by adding the creation of a synostosis between the distal anastomosis and the remnant of the distal fibula with a spongiosa graft from the ipsilateral femur to the initial procedure.

SH: conceptualization; data curation; investigation; methodology; visualization; writing—original draft. HW, RS: conceptualization; methodology; supervision; writing—review & editing. KU: data curation; writing—original draft. NZK: data curation; writing—editing.

Acta thanks Søren Kold for help with peer review of this study.

At B. Vascularized bone grafts. In: Green D P, Hotchkiss R N, Pederson W C, editors. Green’s operative hand surgery. Philadelphia, PA: Churchill Livingstone; 1999.

Bos K E, Besselaar P P, van der Eyken J W, Taminiau A H, Verbout A J. Reconstruction of congenital tibial pseudarthrosis by revascularized fibular transplants. Microsurgery 1993; 14(9): 558-62.

Capanna R, Bufalini C, Campanacci C. A new technique for reconstructions of large metadiaphyseal bone defects: a combined graft (allograft shell plus vascularized fibula). Orthop Traumatol 1993; 2: 159–77.

610

Choi I H, Cho T J, Moon H J. Ilizarov treatment of congenital pseudarthrosis of the tibia: a multi-targeted approach using the Ilizarov technique. Clin Orthop Surg 2011; 3(1): 1-8. doi: 10.4055/cios.2011.3.1.1.

Erni D, De Kerviler S, Hertel R, Slongo T. Vascularised fibula grafts for early tibia reconstruction in infants with congenital pseudarthrosis. J Plast Reconstr Aesthet Surg 2010; 63(10): 1699-704. doi: 10.1016/j. bjps.2009.09.016.

Ghanem I, Damsin J P, Carlioz H. Ilizarov technique in the treatment of congenital pseudarthrosis of the tibia. J Pediatr Orthop 1997; 17(5): 685-90.

Gilbert A, Brockman R. Congenital pseudarthrosis of the tibia. Long-term followup of 29 cases treated by microvascular bone transfer. Clin Orthop Relat Res 1995; (314): 37â&#x20AC;&#x201C;44.

Grill F, Bollini G, Dungl P, Fixsen J, Hefti F, Ippolito E, Romanus B, Tudisco C, Wientroub S. Treatment approaches for congenital pseudarthrosis of tibia: results of the EPOS multicenter study. European Paediatric Orthopaedic Society (EPOS). J Pediatr Orthop B 2000; 9(2): 75-89.

Hefti F, Bollini G, Dungl P, Fixsen J, Grill F, Ippolito E, Romanus B, Tudisco C, Wientroub S. Congenital pseudarthrosis of the tibia: history, etiology, classification, and epidemiologic data. J Pediatr Orthop B 2000; 9(1): 11-15.

Hermanns-Sachweh B, Senderek J, Alfer J, Klosterhalfen B, Buttner R, Fuzesi L, Weber M. Vascular changes in the periosteum of congenital pseudarthrosis of the tibia. Pathol Res Pract 2005; 201(4): 305-12. doi: 10.1016/j.prp.2004.09.013.

Horn J, Steen H, Terjesen T. Epidemiology and treatment outcome of congenital pseudarthrosis of the tibia. J Child Orthop 2013; 7(2): 157-66. doi: 10.1007/s11832-012-0477-0.

Kalra G D S, Agarwal A. Experience with free fibula transfer with screw fixation as a primary modality of treatment for congenital pseudarthosis of tibia in children: series of 26 cases. Indian J Plast Surg 2012; 45: 468-77.

Kesireddy N, Kheireldin R K, Lu A, Cooper J, Liu J, Ebraheim N A. Current treatment of congenital pseudarthrosis of the tibia: a systematic review and meta-analysis. J Pediatr Orthop B 2018; 27(6): 541-50. doi: 10.1097/ BPB.0000000000000524.

Ohnishi I, Sato W, Matsuyama J, Yajima H, Haga N, Kamegaya M, Minami A, Sato M, Yoshino S, Oki T, Nakamura K. Treatment of congenital pseud-

Acta Orthopaedica 2020; 91 (5): 605â&#x20AC;&#x201C;610

arthrosis of the tibia: a multicenter study in Japan. J Pediatr Orthop 2005; 25(2): 219-24.

Paley D. Congenital pseudarthrosis of the tibia: biological and biomechanical considerations to achieve union and prevent refracture. J Child Orthop 2019; 13(2): 120-33. doi: 10.1302/1863-2548.13.180147.

Paley D, Catagni M, Argnani F, Prevot J, Bell D, Armstrong P. Treatment of congenital pseudoarthrosis of the tibia using the Ilizarov technique. Clin Orthop Relat Res 1992; (280): 81-93.

Patterson M. Impact of external fixation on adolescents: an integrative research review. Orthop Nurs 2006; 25(5): 300-8; quiz 9-10.

Ramaker R R, Lagro S W, van Roermund P M, Sinnema G. The psychological and social functioning of 14 children and 12 adolescents after Ilizarov leg lengthening. Acta Orthop Scand 2000; 71(1): 55-9. doi: 10.1080/00016470052943900.

Romanus B, Bollini G, Dungl P, Fixsen J, Grill F, Hefti F, Ippolito E, Tudisco C, Wientroub S. Free vascular fibular transfer in congenital pseudoarthrosis of the tibia: results of the EPOS multicenter study. European Paediatric Orthopaedic Society (EPOS). J Pediatr Orthop B 2000; 9(2): 90-3.

Ruggieri M, Huson S M. The clinical and diagnostic implications of mosaicism in the neurofibromatoses. Neurology 2001; 56(11): 1433-43.

Stevanovic M, Gutow A P, Sharpe F. The management of bone defects of the forearm after trauma. Hand Clin 1999; 15(2): 299-318.

Stevenson D A, Birch P H, Friedman J M, Viskochil D H, Balestrazzi P, Boni S, Buske A, Korf B R, Niimura M, Pivnick E K, Schorry E K, Short M P, Tenconi R, Tonsgard J H, Carey J C. Descriptive analysis of tibial pseudarthrosis in patients with neurofibromatosis 1. Am J Med Genet 1999; 84(5): 413-19.

Tan J, Roach J, Wang A. Transfer of ipsilateral fibula on vascular pedicle for treatment of congenital pseudarthrosis of the tibia. J Pediatr Orthop 2011; 31(1): 72-8. doi: 10.1097/BPO.0b013e318202c243.

Van Royen K, Brems H, Legius E, Lammens J, Laumen A. Prevalence of neurofibromatosis type 1 in congenital pseudarthrosis of the tibia. Eur J Pediatr 2016; 175(9): 1193-8. doi: 10.1007/s00431-016-2757-z.

Weiland A J, Weiss A P, Moore J R, Tolo V T. Vascularized fibular grafts in the treatment of congenital pseudarthrosis of the tibia. J Bone Joint Surg Am 1990; 72(5): 654-62.

Acta Orthopaedica 2020; 91 (5): 611–616

611

A new standard radiographic reference for proximal fibular height in children Adrien FROMMER 1, Maike NIEMANN 1, Georg GOSHEGER 2, Gregor TOPOROWSKI 1, Andrea LAUFER 1, Maria EVESLAGE 3, Jan Niklas BRÖKING 1, Robert RÖDL 1, and Bjoern VOGT 1 1 Children’s

Orthopedics, Deformity Reconstruction and Foot Surgery, University Hospital of Muenster; 2 General Orthopedics and Tumor Orthopedics, University Hospital of Muenster; 3 Institute of Biostatistics and Clinical Research, University of Muenster, Germany Correspondence: adrien.frommer@ukmuenster.de Submitted 2019-11-17. Accepted 2020-04-27.

Background and purpose — To date there is a lack of studies defining the anatomical position of the proximal fibula. This is especially relevant when planning surgical interventions affecting the knee joint such as permanent or temporary epiphysiodesis to correct leg length discrepancies or angular deformities in growing patients. The goal of this study is to establish a standardized measurement technique and radiological reference values for the position of the proximal fibula in children.

Patients and methods — 500 measurements were performed in calibrated long standing anteroposterior radiographs of 256 skeletally immature patients (8–16 years; 233 female, 267 male legs). As a radiographic reference in the frontal plane, the distance between the center of the proximal tibial growth plate and a line tangential to the tip of the fibular head and horizontal to the imaging plane was measured (dPTFH).

Results — The average value of dPTFH in the studied population (median age 12 years) was –2.7 mm (SD 3, CI –3.0 to –2.5) and normally distributed (p = 0.1). There were no clinically significant sex or age-dependent differences.

The inter-rater reliability analysis showed excellent ICC values (ICC = 0.88; CI 0.77–0.93).

Interpretation — This study provides a new radiographic reference value to assess the position of the proximal fibula in relation to the proximal tibia in children and adolescents.

This reference can aid preoperative decision-making as to whether additional fibular epiphysiodesis is necessary when performing tibial epiphysiodesis to correct moderate leglength discrepancies.

Detailed knowledge of the physiological limb alignment is of importance for the treatment of limb deformities (Moreland et al. 1987, Chao et al. 1994, Paley 2002). Joint orientation angles of the lower limb and numerical reference values such as the mechanical axis deviation (MAD) help to differentiate physiological from pathological limb alignment (Moreland et al. 1987, Paley 2002) when planning procedures such as corrective osteotomies or epiphysiodesis (ED) for correction of angular deformities or moderate leg-length discrepancies (LLD) (Bowen and Johnson 1984, Canale and Christian 1990, Vogt et al. 2014). While these references aid the assessment of knee joint alignment, to date there is a lack of standardized radiographic references to evaluate the anatomical location of the fibular head in relation to the proximal tibia. The position of the proximal fibula is clinically relevant when planning temporary or permanent tibial ED for moderate LLD. When planning tibial ED the knowledge of remaining growth potential and the presence of pre-existing fibular overgrowth is essential (McCarthy et al. 2003). Some surgeons favor performing a concomitant fibular ED with tibial ED to prevent fibular overgrowth, which might cause discomfort and instability of the knee joint due to laxity of the lateral collateral ligament (LCL) (Canale and Christian 1990, Porat et al. 1991, Metaizeau et al. 1998, McCarthy et al. 2003). Others argue that fibular ED should not be performed due to the risk of peroneal nerve injury and claim the amount of overgrowth is irrelevant (Bowen and Johnson 1984, Gabriel et al. 1994, Siedhoff et al. 2014). Different radiographic approaches have been described previously to measure proximal fibular overgrowth or shortening (Ogilvie and King 1990, McCarthy et al.

2003, Kim et al. 2019) but until today there is no standardized radiographic reference defining the anatomical location of the proximal fibula in children aged 8–16 years.

© 2020 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 10.1080/17453674.2020.1769378

612

Acta Orthopaedica 2020; 91 (5): 611–616

+ dPTFH

dPTFH

– dPTFH

Figure 1. (a) The distance between the center of the proximal tibial growth plate and the tip of the fibular head (dPTFH) is measured in the frontal plane of long standing radiographs of a 13-year-old boy. (b) dPTFH is defined by the distance in millimeters between the center of the tibial growth plate and a line tangential to the tip of the fibular head and horizontal to the imaging plane. (c, d) Negative values indicate that the fibular head is localized more distally than the center of the proximal tibial growth plate and vice versa.

The goal of this study is to define a new radiographic reference for the position of the proximal fibula in skeletally immature patients and to test whether there are age- and sex-dependent differences. We believe that the results can help clinical decision-making, especially in the treatment of LLD by tibial ED in children.

Patients and methods The studied radiographs were all obtained from the archives of our orthopedic clinic from the past 10 years. Most of the radiographs originate from patients who were treated in the outpatient department of our institution. The most common indications for the radiographic examination were: ruling out pathological limb alignment, follow-up of permanent or temporary isolated femoral ED for LLD or angular deformities, and assessment of LLD. Calibrated long-standing anteroposterior radiographs were retrospectively analyzed from a population of 256 skeletally immature patients with a chronological age of 8 to 16 years.

This period of age typically represents the time when growthdependent surgical procedures of the lower limb can be performed. The radiographs were evaluated retrospectively with the following inclusion criteria: chronological patient age at radiologic examination 8–16 years, LLD < 1 cm, MAD < ± 2 cm. Radiographs from patients who underwent operative treatment of the knee joint, who received systemic treatment like chemotherapy or growth hormone application, and who had evidence of maltorsion, congenital disorders, or history of trauma of the leg were excluded from the study. If both legs of one patient met our inclusion criteria bilateral measurements

were conducted. In unilateral congenital disorders or LLD the unaffected contralateral leg was included in the study. This resulted in a radiological assessment of 500 legs, 233 female (f) and 267 male (m) legs.

Long standing radiographs were obtained by digitally stitching 3–4 (depending on the individual’s leg length) sector radiographs together. The images were captured from a defined distance (2.8 m) with a metal calibration sphere (25.4 mm diameter) mounted on an adjustable flexible arm.

As a radiographic reference in the frontal plane, the distance between the center of the proximal tibial growth plate and a line tangential to the tip of the fibular head and horizontal to the imaging plane was measured in a way similar to previous studies (McCarthy et al. 2003) (Figure 1). Negative values indicate that the fibular head is localized more distally than the center of the proximal tibial growth plate and vice versa.

In order to establish a standardized nomenclature, this value will be referred to as the “distance between the proximal tibial physis and the fibular head” (dPTFH). dPTFH was measured in the following age groups (AG): 8–10 years (n = 95: f/m = 44/51), 11–12 years (150: 75/75), 13–14 years (166: 75/91), 15–16 years (89: 39/50) (Figure 2).

3 observers (GT, NB, BV) independently measured dPTFH in 36 randomly chosen radiographs from the study population to assess the inter-rater reliability.

All measurements were performed on calibrated radiographs with the PACS® System (GE Healthcare, Chicago, IL, USA).

Statistics The statistical analyses were performed using IBM SPSS® Statistics 25 for Windows (IBM Corp, Armonk, NY, USA)

Acta Orthopaedica 2020; 91 (5): 611–616

613

Figure 2. Radiological assessment of the center of the proximal tibial growth plate and the tip of the proximal fibula in order to measure dPTFH in different aged patients (a = 10 years, b = 12 years, c = 14 years, d = 16 years). While closer to skeletal maturity the growth plate appears less distinct (d), in general its outlines can still be estimated.

dPTFH (mm) Frequency and R version 3.6.1 (R Foundation for Statistical Computing, Vienna, Austria). Normal distribufemale male tion of the measurements was assessed descrip50 5 tively using a histogram. Data are reported as 40 mean ± standard deviation (SD) and 95% con0 fidence intervals (CI) for the mean. In order to 30 account for intra-individual correlation, the effect of age and sex on the dPTFH values was −5 20 analyzed using a linear mixed model (LMM).

The model included age (centered at the mean), 10 −10 sex, and their interaction as fixed effects and a random intercept for the patient. The model fit 0 –10 –8 –6 –4 –2 0 2 4 8 10 12 14 16 was assessed descriptively using Q–Q plots. An dPTFH (mm) Age (years) additional mixed model was computed using the Figure 4. Scatterplot of age vs. dPTFH Figure 3. The graph demonstrates a age groups as single fixed effect. including the regression lines resulting normal distribution of dPTFH measured from the linear mixed model presented in 500 legs of children and adolescents The intraclass correlation coefficient (ICC) for in Table 2. from the age of 8–16 years. The mean the 3 raters was estimated based on the estimated dPTFH is –2.7 mm with a standard devivariance components of a linear mixed model ation (SD) of 2.8 mm. For clinical practicability mean and SD should be approxiincluding a random effect for each of the patient mated to –3 mm and 3 mm, respectively. and the leg of the patient and a fixed effect for the rater. A 95% confidence interval was computed using a parametric bootstrap (10,000 runs). Sample size Table 1. Results of the linear mixed model for dPTFH (mm) including calculation for 3 independent raters was performed with PASS fixed effects for sex (centered at the mean), sex, and the interaction between age and sex 16.0.4 (NCSS, LLC. Kaysville, UT, USA) by assessing the width of a 2-sided 95% CI for the ICC. Regression No adjustment for multiplicity was applied. All inferential Factor coefficient (95% CI) p-value statistics are intended to be exploratory, not confirmatory. ● ●

Ethics, funding, and potential conflicts of interest The study was approved by the ethical committee of the university of Muenster on November 21, 2017 (registration number: 2017-491-f-S). The authors received no funding for this work and have no conflict of interest.

Results The median age was 12 years (8–16 years, mean 12.4 years) and the dPTFH was normally distributed. The average value of dPTFH was -2.7 mm (SD 2.8, CI -3.0 to -2.5) (Figure 3).

Intercept –2.68 (–3.18 to –2.19) Age (years) –0.0049 (–0.23 to 0.22) Sex (male vs. female) –0.29 (–0.94 to 0.35) Age x sex 0.21 (–0.12 to 0.55)

< 0.001 0.1 0.4 0.2

The LMM analysis showed no statistically significant association with age or sex and no noticeable interaction between these 2 parameters (Figure 4, Tables 1 and 2). The age and sex dependent distribution of dPTFH revealed no clinically relevant difference (Figure 4). The following dPTFH values were measured in the defined AG: 8–10 years: dPTFH = –3.2 mm (CI –3.7 to –2.6), AG 11–12 years: dPTFH = –2.3 mm (CI

614

Acta Orthopaedica 2020; 91 (5): 611–616

Table 2. Results of the linear mixed model for dPTFH (mm) including age group as a fixed effect Regression Factor coefficient (95% CI) Intercept Age group 11–12 vs. 8–10 13–14 vs. 8–10 15–16 vs. 8–10

p–value

–3.18 (–3.96 to –2.39)

< 0.001

0.61 (–0.36 to 1.58) 0.081 (–0.83 to 0.99) 0.81 (–0.38 to 1.99)

0.2 0.9 0.2

The analysis showed no statistically relevant difference between the age groups.

Table 3. Mean dPTFH in the age groups Age group 8–10 11–12 13–14 15–16

Sample (n)

Mean dPTFH (95% CI)(mm)

94 150 166 89

–3.2 (–3.7 to –2.6) –2.3 (–2.8 to –1.8) –3.1 (–3.5 to –2.7) –2.1 (–2.7 to –1.5)

–2.8 to –1.8), AG 13–14 years: dPTFH = –3.1 mm (CI –3.5 to –2.7), AG 15–16 years: dPTFH = –2.1 mm (CI –2.7 to –1.5) (Table 3). No statistically significant differences between the age groups were found in the LMM analysis (Table 2).

For the ease of clinical practicability, the lack of clinical relevance, and taking measurement inaccuracy into consideration the values for the mean and SD of dPTFH can be approximated to –3 mm and 3 mm, respectively.

The estimated ICC value (ICC = 0.88; CI 0.77–0.93) showed excellent reliability for the measurements performed by 3 independent observers.

Discussion Standard radiographic references of joint and limb alignment are of fundamental importance for the treatment of limb deformities and leg-length discrepancies. Various studies have improved the field of deformity reconstruction by providing radiological reference values to distinguish between physiological and pathological limb alignment (Moreland et al.

1987, Chao et al. 1994, Paley 2002). When considering the lower leg, previous studies have mainly assessed the radiological location of the distal fibula in relation to the ankle joint.

Ogden and McCarthy (1983) have shown that during adolescence the distal fibular physis is normally level with the tibial articular surface of the ankle joint. The Shenton line and dime sign are radiographic measurements that have been described in order to analyze the relationship of the distal fibula and distal tibia (Panchbhavi et al. 2018).

These reference values are commonly used in orthopedic and traumatological daily routine and help to analyze malleo-

Table 4. Conditions with preexisting proximal fibular under- and overgrowth Fibular undergrowth

Fibular overgrowth

Idiopathic Idiopathic Posttraumatic or infectious Posttraumatic or infectious (e.g., damage to the fibular (e.g., damage to the tibial growth plate) growth plate) Congenital Congenital (e.g., femoral deficiency, (e.g., achondroplasia, fibular hemimelia, etc.)

Desbuquois dysplasia, etc.)

lar and ankle fractures and to radiologically control the results of surgical reduction (Ogden and McCarthy 1983, Weber and Simpson 1985, Panchbhavi et al. 2018).

However, to date there is a lack of standard radiographic reference values defining the anatomical position of the proximal fibula in relation to the proximal tibia in children and adolescents. Previous radiological examinations evaluated the proximal and distal “tibial–fibular physis distance” in 63 children from the age of 1 to 12 years (Beals and Skyhar 1984). These observations help the assessment of the tibio–fibular relation from an early age onward but do not provide standard radiographic reference values for adolescents in which growthinfluencing surgeries are commonly performed.

Our findings should be seen in the context of different treatment options, especially for moderate LLD in skeletally immature patients. LLD of 2–5 cm can be an impairing condition affecting gait pattern and mobility. While at skeletal maturity lengthening procedures are commonly performed (Schiedel et al. 2014, Reitenbach et al. 2016, Horn et al. 2019), during childhood and adolescence temporary or permanent proximal tibial and/or distal femoral ED of the relatively longer leg are established methods to treat LLD (Canale and Christian 1990, Gabriel et al. 1994, Metaizeau et al. 1998, McCarthy et al.

2003, Siedhoff et al. 2014, Vogt et al. 2014, Boyle et al. 2017).

There is controversy in the literature as to whether an additional ED of the fibular head should be performed together with tibial ED. While some authors argue that fibular ED should be performed to prevent fibular overgrowth in relation to the arrested tibia and consequently laxity of the LCL (Canale and Christian 1990, Draganich et al. 1991, Metaizeau et al. 1998, McCarthy et al. 2003, LaPrade et al. 2010, Arikan and Misir 2019), others argue that the risk of peroneal nerve injury does not justify the intervention and that the possible amount of relative fibular overlength is clinically irrelevant (Bowen and Johnson 1984, Gabriel et al. 1994, Siedhoff et al. 2014).

Part of this controversy results from the lack of standard radiographic reference values to assess the anatomical height of the fibular head in relation to the proximal tibia. When planning the correction of moderate LLD by tibial ED the surgeon must consider if concomitant proximal fibular ED is neces-

Acta Orthopaedica 2020; 91 (5): 611–616

6 mm 2 SD 6 mm 2 SD

615

dPTFH = +3 mm mean dPTFH = –3 mm dPTFH = –9 mm

Figure 5. This study provides dPTFH as a new standard radiographic reference defining the anatomical localization of the fibular head in relation to the center of the proximal tibial growth plate. The mean dPTFH in children and adolescents (8–16 years) is –3 mm with an SD of 3 mm. We propose to consider deviations of dPTFH greater than 2 SD as fibular overlength (dPTFH > +3 mm) or shortening (dPTFH < –9 mm) respectively.

sary or not. Thus, it is essential to evaluate the localization of the proximal fibula in relation to the proximal tibia before the beginning of treatment.

The goal of the latter considerations is to maintain the physiological proximal tibio–fibular relation by prevention of secondary fibular overgrowth. On the other hand, a standard radiographic reference value might be of at least equal importance in conditions with pre-existing proximal fibular underor overgrowth in relation to the tibia (Table 4). Especially in patients with significant LCL instability and subsequent gapping of the medial knee joint due to preexisting proximal fibular overgrowth, as can frequently be seen in achondroplasia, surgical correction of the tibio–fibular disproportion (e.g., fibular ED) can be considered (Lee et al. 2007).

This study shows that the mean dPTFH in children and adolescents (8–16 years) is –3 mm with an SD of 3 mm. We propose to consider deviations of dPTFH greater than 2 SD as fibular overlength (dPTFH = +3 mm) or shortening (dPTFH = –9 mm), respectively (Figure 5).

The inter-rater reliability analysis has shown that dPTFH can be measured accurately and is reproducible by independent observers. These results indicate that dPTFH is a reliable measurement value that can be implemented in routine radiological limb alignment analysis.

Our results should be understood taking into consideration the following limitations. Different techniques of radiographic analysis can lead to variations in the dPTFH depending on the angle of the X-ray beam, therefore this study provides reference values for the height of the proximal fibula only in calibrated long standing, full-weight-bearing anteroposterior radiographs. This study does not supply clinical information regarding the stability of the knee joint or potential discomfort caused by proximal fibular overgrowth or shortening. Further studies will be needed to assess how a dPTFH greater than 2 SD affects the function of the knee joint.

As a new standard radiographic reference dPTFH can aid preoperative decision-making as to whether additional fibular ED is needed when performing tibial ED to correct moderate LLD in children and adolescents by defining the anatomical height of the proximal fibula.

AF: wrote the manuscript, performed and supervised the measurements of dPTFH, supervised the inter-rater reliability analysis, performed statistical analysis, and prepared the figures. MN: performed dPTFH measurements and statistical analysis. GG: provided the radiographs and made substantial changes to the manuscript. GT: performed the measurement for the inter-rater reliability analysis, and critically assessed and corrected the manuscript. AL: assessed and corrected the manuscript, arranged the data, and prepared the tables. ME: revised, performed, and wrote the statistical report. NB: performed the measurement for the inter-rater reliability analysis, and critically assessed and corrected the manuscript. RR: provided the radiographs, analyzed the data, supervised the work, and made substantial changes to the manuscript. BV: designed the study, analyzed the data, supervised the work, performed the measurement for the inter-rater reliability analysis, and critically assessed and corrected the manuscript.

The authors acknowledge support from the Open Access Publication Fund of the University of Münster, Germany.

Acta thanks Joachim Horn and Bjarne Moeller-Madsen for help with peer review of this study.

Arikan Y, Misir A. Clinical and radiologic outcomes following resection of primary proximal fibula tumors: proximal fibula resection outcomes. J Orthop Surg 2019; 27(2): 2309499019837411. doi: 10.1177/2309499019837411.

Beals R K, Skyhar M. Growth and development of the tibia, fibula, and ankle joint. Clin Orthop Relat Res 1984; (182): 289-92.

Bowen J R, Johnson W J. Percutaneous epiphysiodesis. Clin Orthop Relat Res 1984; 190: 170-3.

Canale S T, Christian C A. Techniques for epiphysiodesis about the knee. Clin Orthop Relat Res 1990; (255): 81-5.

Chao E Y, Neluheni E V, Hsu R W, Paley D. Biomechanics of malalignment.

Orthop Clin North Am 1994; 25(3): 379-86.

Draganich L F, Nicholas R W, Shuster J K, Sathy M R, Chang A F, Simon M A. The effects of resection of the proximal part of the fibula on stability of the knee and on gait. J Bone Joint Surg Am 1991; 73(4): 575-83.

Gabriel K R, Crawford A H, Roy D R, True M S, Sauntry S. Percutaneous epiphyseodesis. J Pediatr Orthop 1994; 14(3): 358-62.

Horn J, Hvid I, Huhnstock S, Breen A B, Steen H. Limb lengthening and deformity correction with externally controlled motorized intramedullary nails: evaluation of 50 consecutive lengthenings. Acta Orthop 2019; 90(1): 81-7.

Kim T W, Lee S H, Lee J Y, Lee Y S. Effect of fibular height and lateral tibial condylar geometry on lateral cortical hinge fracture in open wedge high tibial osteotomy. Arthroscopy 2019; 35(6): 1713-20.

LaPrade R F, Spiridonov S I, Coobs B R, Ruckert P R, Griffith C J. Fibular collateral ligament anatomical reconstructions: a prospective outcomes study. Am J Sports Med 2010; 38(10): 2005-11.

Lee S T, Song H R, Mahajan R, Makwana V, Suh S W, Lee S H. Development of genu varum in achondroplasia: relation to fibular overgrowth. J Bone Joint Surg Br 2007; 89(1): 57-61.

McCarthy J, Burke T, McCarthy C. Need for concomitant proximal fibular epiphysiodesis when performing a proximal tibial epiphysiodesis. J Pediatr Orthop 2003; 23(1): 52-4.

616

Metaizeau J P, Wong-Chung J, Bertrand H, Pasquier P. Percutaneous epiphysiodesis using transphyseal screws (PETS). J Pediatr Orthop 1998; 18(3): 363-9.

Moreland J R, Bassett L W, Hanker G J. Radiographic analysis of the axial alignment of the lower extremity. J Bone Joint Surg Am 1987; 69(5): 745-9.

Ogden J A, McCarthy S M. Radiology of postnatal skeletal development, VIII: Distal tibia and fibula. Skeletal Radiol 1983; 10(4): 209-20.

Ogilvie J W, King K. Epiphysiodesis: two-year clinical results using a new technique. J Pediatr Orthop 1990; 10(6): 809-11.

Paley D. Normal lower limb alignment and joint orientation. In: Paley D, editor. Principles of deformity correction. Berlin: Springer; 2002. p. 1-18.

Panchbhavi V K, Gurbani B N, Mason C B, Fischer W. Radiographic assessment of fibular length variance: the case for “fibula minus”. J Foot Ankle Surg 2018; 57(1): 91-4.

Porat S, Peyser A, Robin G C. Equalization of lower limbs by epiphysiodesis: results of treatment. J Pediatr Orthop 1991; 11(4): 442-8.

Acta Orthopaedica 2020; 91 (5): 611–616

Reitenbach E, Rodl R, Gosheger G, Vogt B, Schiedel F. Deformity correction and extremity lengthening in the lower leg: comparison of clinical outcomes with two external surgical procedures. Springerplus 2016; 5(1): 2003.

Schiedel F M, Vogt B, Tretow H L, Schuhknecht B, Gosheger G, Horter M J, Rodl R. How precise is the PRECICE compared to the ISKD in intramedullary limb lengthening? Reliability and safety in 26 procedures. Acta Orthop 2014; 85(3): 293-8.

Siedhoff M, Ridderbusch K, Breyer S, Stucker R, Rupprecht M. Temporary epiphyseodesis for limb-length discrepancy: 8- to 15-year follow-up of 34 children. Acta Orthop 2014; 85(6): 626-32.

Vogt B, Schiedel F, Rodl R. [Guided growth in children and adolescents. Correction of leg length discrepancies and leg axis deformities]. Orthopade 2014; 43(3): 267-84.

Weber B G, Simpson L A. Corrective lengthening osteotomy of the fibula.

Clin Orthop Relat Res 1985; (199): 61-7.

Acta Orthopaedica 2020; 91 (5): 617–619

617

Correspondence

New 3-dimensional implant application as an alternative to allograft in limb salvage surgery: a technical note on 10 cases

Sir,—I would like to congratulate Park et al. (2020a) for this excellent pilot study combining customised 3DiPC with conventional orthopedic prostheses for limb salvage surgery. Its a change in paradigm opening possibilities to adaptation, efficiency and cost minimisation. The concept is adaptable to situations and societies that struggle with allograft banking due to cultural, societal or financial overhead hurdles. Cost reduction will particularly resonate with institutes in low and middle income countries increasing the armamentarium. The method is especially attractive when achieving stable, durable fixation of small peri-articular fragments in children, adolescents and adults (Agarwal et al. 2010). Longevity of allografts, extra corporeal radiated bone and cryotherapy treated bone has unsolved challenges (Gerrand et al. 2003, Puri et al. 2018).

They are prone to acute and delayed failure and do not allow immediate loading. The potential for bone ingrowth within the porous implant and further reduction in risk of aseptic loosening makes it biologically viable. While Park et al. have enumerated the key technical drawbacks and manufacturing challenges, a few technical details in combining them with available off the shelf prosthesis remain unanswered. For instance, the size, length and curvature of the intramedullary stem that would fit the residual native bone is difficult to predict. Would that enforce manufacturing of a wider bore 3DiPC, to combine with a conventional hip or knee implant, as a coupling agent, thus precluding a press fit/uncemented fit of the stem? The key factor, promising longevity, is the integrity of the bone implant interface, and cementation endagers that integrity. The use of lattice structure as a scaffolding in a metastatic case with the hope of bone ingrowth was curious as native biology is certain to lose against the incumbent cancer (Huang et al. 2017, Jenkins et al. 2017). Woud not an intercalary 3DiPC be significantly more robust and withstand post operative irradiation better? How does this compare with Trabecular metal, with outstanding results from the Mayo group in prior irradiated metastatic cases (Jenkins et al. 2017)? Was the size and structure of the lattice a constraint of design or the manufacturing process? Were attempts made to simulate the lattice structure of trabecular metal (Chen et al. 2016)? What concerns do they share about accelerated wear of titanium articulating with conventional chrome cobalt prosthesis, as in their pelvic recon-

structions (Moharrami et al. 2013)? I look forward to a longer outcome analysis of this exciting new concept.

Prakash Nayak Assistant Professor, Bone and soft tissue Unit Department of Surgical Oncology Tata Memorial Hospital Homi Bhabha National Institute, Mumbai, India E-mail: nayakprakash@gmail.com

Sir,—Thank you for the opportunity to respond to the letter from Professor Prakash Nayak, regarding our work entitled “New 3-dimensional implant application as an alternative to allograft in limb salvage surgery: a technical note on 10 cases.” Here, we present a more detailed description on the porous structure (3-dimensional [3D]-printed lattice), design process of the 3D-printed implant and prosthesis composite (3DiPC), and conjugation between 3D-printed and conventional orthopedic implants.

The effectiveness and safety of orthopedic implant fabricated using the 3D-printing technology remains to be proven and requires further accumulation of clinical experiences, but 3D-printed implant is promising especially in orthopedic oncology. Moreover, porosity with lattice structure can be easily implemented using the 3D-printing technology.

The optimal pore size is defined as hundreds of micrometers before 3D printing (Bobyn et al. 1980, Mumith et al. 2017, Park et al. 2020b). For example, the trabecular metal (Zimmer biomet, Warsaw, Indiana, United States) has an open pore with the mean size of 440 um. The electron beam melting (EBM)type 3D printer (ARCAM A1, Arcam AB, Mölndal, Sweden) used to fabricate implants reported in our paper has an accuracy of 200 µm. The pore size varied depending on the surgical location and load for endurance, and the lattice structure with 750-µm pore was often selected, considering the printing accuracy and metal powder removal after printing.

Notably, the lattice unit size is different from the pore size.

© 2020 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 10.1080/17453674.2020.1788698

618

The unit and pore sizes in dode-thin lattice structure.

For example, as regards the dode-thin mesh (Magics 22, Materialise; Leuven, Belgium) we used, the actual pore size is 750 µm if the lattice unit size is 2 mm (Figure).

Bone incorporation is the key of implant longevity, and biocompatibility with the porous structure made of 64 titanium (Ti-6Al-4V) is well known from previous studies (Wu et al.

2013, Guyer et al. 2016, Mumith et al. 2017, McGilvray et al. 2018). For all patients we reported, the junction of the 3D implant to the host bone was designed to achieve a lattice structure depth of > 5 mm. To prevent interference of bone incorporation, PMMA bone cement was carefully applied to prevent it from entering the gap between the host bone and the implant. For some patients undergoing pelvic reconstruction involving the hip joint (patients 2, 3, 5, 6, 7), bone cement was used to attach the THA cup to the 3D-printed pelvic implant and simultaneously prepared outside of the surgical field while performing the surgical approaching procedure (Figure 1D in original paper). For patient 8, injectable bone cement was used to fill the gap between the 3D implant and the intramedullary nail after the 3D implant reduction to the host bone (Figure 1F in original paper). Trabecular metal (Zimmer biomet, Warsaw, Indiana, United States) is a porous tantalum implant with excellent surgical results and relatively long clinical experience. However, fabricating a patient-specific customized implant is more difficult with the trabecular metal than with 3D-printing lattice structure in orthopedic oncology. Remarkably, in revision hip surgery with a massive bone defect using trabecular metal, buttress, shim and restrictor are utilized (Zimmer biomet, Warsaw, Indiana, United States); bone cement is applied between the trabecular metal and revision shell of the hip joint, similarly with the 3DiPC method. Therefore, we believe that using a bone cement mantle with certain thickness is a more proven and realistic method than a press/uncemented fit to fasten 2 types of metal implants.

For metastatic bone tumor cases (patients 4 and 9), the bone biology near the tumor could be impaired by tumor itself or surgical procedures to achieve extended margin after curettage. Therefore, we selected candidates for the 3D implant

Acta Orthopaedica 2020; 91 (5): 617–619

limb salvage surgery with bone metastasis when a wide margin could be justified. Patient 4 already had a NED status preoperatively, and patient 7 had solitary metastasis in the femur diaphysis and underwent limb salvage surgery with wide margins. Therefore, the 3D implant was fixed to the healthy host bone in this case series.

In designing the 3D-printed implant for 3DiPC, special considerations should be made as to where the conventional implant will be combined. One of the disadvantages of the 3D-printed implant surgery is that the surgical plan including the bone margin cannot be adjusted. Therefore, the 3D-printed implant surgery should be carefully indicated for patients with rapidly growing tumors. Under the assumption that the surgery plan remains unchanged, a conventional implant used with 3D-printed implant should be confirmed during the designing process. For example, the diameter of the hip joint socket in a 3D-printed pelvic implant should be larger than that of the THA cup to achieve a 2-mm-thickness cement mantle. For 3DiPC with an intramedullary nail, the nail with the lowest curvature was selected, and the nail length and diameter were determined during the designing process preoperatively. Considering the nail insertion process, approaching direction, and nail curvature, the inner diameter of the 3D implant was increased as necessary so that the planned nail could be easily inserted and the gap between the 3D implant and the nail could be filled with bone cement. Nevertheless, nail insertion through the 3D implant is a difficult process. In patient 9, the mechanical strength of the implant itself was designed to be weak and wrapped with two pieces of the 3D implant around the nail.

A 3D-printed implant in 3DiPC technique is expected to have 2 main advantages over the structural allograft. First, no osteolysis-related problems are observed in a 3D-printed titanium alloy implant with a relatively constant mechanical property over time. However, aside from the absence of osteolysis, further mechanical study on the fatigue profile of the 3D-printed titanium alloy is needed. Second, surgical time can be reduced by omitting allograft trimming and simplification of fixation to the host bone. Pelvic bone tumor surgery with pelvic ring or hip joint reconstruction is large, time-consuming surgery associated with postoperative complications (Mankin et al. 2004). The surgical time cannot be directly compared because pelvic bone tumor surgery encompasses a wide range of surgeries depending on the specific surgical location, extent, and inclusion of the hip joint. In the original paper, the mean surgical time of pelvic reconstruction with hip joint (patient 1–7) was 160 (30–220) min, which is a relatively short time for pelvic reconstruction surgery.

We acknowledge that several aspects of 3D-printed implant surgery remain to be addressed in further studies with longterm follow-up. Currently, we believe that a 3DiPC method for joint reconstruction or strength security using conventional arthroplasty, tumor prosthesis, or intramedullary nail with well proven, stable performance is also a feasible option.

Acta Orthopaedica 2020; 91 (5): 617â&#x20AC;&#x201C;619

Jong Woong Park and Hyun Guy Kang Orthopaedic Oncology Clinic, National Cancer Center, Korea E-mail: jwpark82@ncc.re.kr and ostumor@ncc.re.kr Agarwal M, Puri A, Gulia A, Reddy K. Joint-sparing or physeal-sparing diaphyseal resections: The challenge of holding small fragments. Clin Orthop Rel Res 2010; 468(11): 2924-32.

Bobyn J, Pilliar R, Cameron H, Weatherly G. The optimum pore size for the fixation of porous-surfaced metal implants by the ingrowth of bone. Clin Orthop Rel Res 1980; (150): 263-70.

Chen W-M, Xie Y M, Imbalzano G, Shen J, Xu S, Lee S-J, et al. Lattice Ti structures with low rigidity but compatible mechanical strength: Design of implant materials for trabecular bone. International Journal of Precision Engineering and Manufacturing 2016; 17(6): 793-9.

Gerrand C H, Griffin A M, Davis A M, Gross A E, Bell R S, Wunder J S. Large segment allograft survival is improved with intramedullary cement.

J Surg Oncol 2003; 84(4): 198-208.

Guyer R D, Abitbol J J, Ohnmeiss D D, Yao C. Evaluating osseointegration into a deeply porous titanium scaffold: a biomechanical comparison with PEEK and allograft. Spine 2016; 41(19): E1146-1150.

Huang H-C, Hu Y-C, Lun D-X, Miao J, Wang F, Yang X-G, et al. Outcomes of intercalary prosthetic reconstruction for pathological diaphyseal femoral fractures secondary to metastatic tumors. Orthop Surg 2017; 9(2): 221-8.

Jenkins D R, Odland A N, Sierra R J, Hanssen A D, Lewallen D G. Minimum five-year outcomes with porous tantalum acetabular cup and augment construct in complex revision total hip arthroplasty. J Bone Joint Surg 2017; 99(10): e49.

619

Mankin H J, Hornicek F J, Temple H T, Gebhardt M C. (2004). Malignant tumors of the pelvis: an outcome study. Clin Orthop Rel Res 2004; 425: 212-7.

McGilvray K C, Easley J, Seim H B, Regan D, Berven S H, Hsu W K, Mroz T E, Puttlitz C M. Bony ingrowth potential of 3D-printed porous titanium alloy: a direct comparison of interbody cage materials in an in vivo ovine lumbar fusion model. Spine J 2018; 18(7): 1250-60. doi:10.1016/j. spinee.2018.02.018 Moharrami N, Langton DJ, Sayginer O, Bull S J. Why does titanium alloy wear cobalt chrome alloy despite lower bulk hardness: A nanoindentation study? Thin Solid Films 2013; 549: 79-86.

Mumith A, Coathup M, Chimutengwende-Gordon M, Aston W, Briggs T, Blunn G. Augmenting the osseointegration of endoprostheses using lasersintered porous collars: an in vivo study. Bone Joint J 2017; 99-B(2): 27682. doi:10.1302/0301-620X.99B2.BJJ-2016-0584.R1 Park J W, Kang H G, Kim J H, Kim H-S. New 3-dimensional implant application as an alternative to allograft in limb salvage surgery: A technical note on 10 cases. Acta Orthop 2020a May; doi.org/10.1080/17453674.2020.17 55543 [Ahead of print] Park J W, Song C A, Kang H G, Kim J H, Lim K M, Kim H-S. Integration of a three-dimensional-printed titanium implant in human tissues: case study.

Applied Sciences 2020b; 10(2): 553.

Puri A, Byregowda S, Gulia A, Patil V, Crasto S, Laskar S. Reconstructing diaphyseal tumors using radiated (50â&#x20AC;&#x2020;Gy) autogenous tumor bone graft. J Surg Oncol 2018; 118(1): 138-43.

Wu S H, Li Y, Zhang Y Q, Li X K, Yuan C F, Hao Y L, et al. Porous titanium-6 aluminum-4 vanadium cage has better osseointegration and less micromotion than a poly-ether-ether-ketone cage in sheep vertebral fusion. Artif Organs 2013: 37(12), E191-201. doi:10.1111/aor.12153

5/20 ACTA ORTHOPAEDICA

LONGER IMPLANT SURVIVAL. WITH THE RIGHT BONE CEMENT.

23%

lower revision risk* with PALACOS® R+G compared to other bone cements

* Calculated difference of cumulated revision rates in hip arthroplasty at 13 years of implantation

www.heraeus-medical.com

10034

NJR Data Supplier Feedback (summary reports); Cumulative revision rates (2007â&#x20AC;&#x201C;2020) status February 2020. Current report accessible at http://herae.Us/njr-data We thank the patients and staff of all the hospitals in England, Wales, Northern Ireland and the Isle of Man who have contributed data to the National Joint Registry. We are grateful to the Healthcare Quality Improvement 3DUWQHUVKLS +4,3 WKH 1-5 6WHHULQJ &RPPLWWHH DQG VWDII DW WKH 1-5 &HQWUH IRU IDFLOLWDWLQJ WKLV ZRUN 7KH YLHZV H[SUHVVHG UHSUHVHQW WKRVH RI +HUDHXV 0HGLFDO *PE+ DQG GR QRW QHFHVVDULO\ UHÆ&#x192;HFW WKRVH RI WKH 1DWLRQDO -RLQW Registry Steering Committee or the Health Quality Improvement Partnership (HQIP) who do not vouch for how the information is presented.

Vol. 91, No. 5, 2020 (pp. 501â&#x20AC;&#x201C;619)

The element of success in joint replacement

Volume 91, Number 5, October 2020

2006_19476_AZ_PALACOS_EOS_2.0_Acta_Orthop_Europe_215x280_EN.indd 1 TF-IORT200146.indd 1

14.07.20 13:39

24-10-2020 13:23:59