You are here

Article Text

Article menu
Surgery
Protocol
Prognostic factors associated with failure of total elbow replacement: a protocol for a systematic review
  1. Zaid Hamoodi1,2,
  2. Celina K Gehringer1,3,
  3. Lucy M Bull4,
  4. Tom Hughes5,
  5. Lianne Kearsley-Fleet1,
  6. Jamie C Sergeant1,3,
  7. Adam C Watts2
  1. 1Centre for Epidemiology Versus Arthritis, Centre for Musculoskeletal Research, University of Manchester, Manchester Academic Health Science Centre, Manchester, UK
  2. 2Upper Limb Unit, Wrightington Wigan and Leigh NHS Foundation Trust, Wigan, UK
  3. 3Centre for Biostatistics, School of Health Sciences, Faculty of Biology, Medicine and Health, Manchester Academic Health Science Centre, The University of Manchester, Manchester, UK
  4. 4Technology Department, Health Navigator Ltd, London, UK
  5. 5Department of Health Professions, Manchester Metropolitan University, Manchester, UK
  1. Correspondence to Dr Zaid Hamoodi; zaid.hamoodi-2{at}postgrad.manchester.ac.uk

Abstract

Introduction Total elbow replacement (TER) has higher failure rates requiring revision surgery compared with the replacement of other joints. Understanding the factors associated with failure is essential for informed decision-making between patients and clinicians, and for reducing the failure rate. This review aims to identify, describe and appraise the literature examining prognostic factors for failure of TER.

Methods and analysis This systematic review will be conducted and reported in line with the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols guidelines. Electronic literature searches will be conducted using Medline, EMBASE, PubMed and Cochrane. The search strategy will be broad, including a combination of subject headings (MESH) and free text search. This search will be supplemented with a screening of reference lists of the included studies and relevant reviews. Two independent reviewers will screen all search results in two stages (title and abstract, and full text) based on the Population, Index prognostic factor, Comparator prognostic factor, Outcome, Time and Setting criteria. The types of evidence included will be randomised trials, non-randomised trials, prospective and retrospective cohort studies, registry studies and case–control studies. If the literature lacks enough studies, then case series with 50 or more TERs will be considered for inclusion. Data extraction and risk of bias assessment for included studies will be performed by two independent reviewers using the Checklist for Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies for Prognostic Factors and Quality In Prognostic Studies tools.

Meta-analyses of prognostic estimates for each factor will be undertaken for studies that are deemed to be sufficiently robust and comparable. Several challenges are likely to arise due to heterogeneity between studies, therefore, subgroup and sensitivity analyses will be performed to account for the differences between studies. Heterogeneity will be assessed using Q and I2 statistics. If I2>40% then pooled estimates will not be reported. When quantitative synthesis is not possible, a narrative synthesis will be undertaken. The quality of the evidence for each prognostic factor will be assessed using the Grades of Recommendation Assessment, Development and Evaluation tool.

PROSPERO registration number CRD42023384756.

  • ORTHOPAEDIC & TRAUMA SURGERY
  • Adult orthopaedics
  • Elbow & shoulder
  • Musculoskeletal disorders
https://creativecommons.org/licenses/by/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution 4.0 Unported (CC BY 4.0) license, which permits others to copy, redistribute, remix, transform and build upon this work for any purpose, provided the original work is properly cited, a link to the licence is given, and indication of whether changes were made. See: https://creativecommons.org/licenses/by/4.0/.

https://doi.org/10.1136/bmjopen-2023-071705

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    STRENGTHS AND LIMITATIONS OF THIS STUDY

    • This is the first systematic review to focus on identifying and appraising studies evaluating prognostic factors associated with failure of total elbow replacement.

    • This review will follow the methodological advances in performing systematic reviews and meta-analyses in prognosis research proposed by the PROGnosis REsearch Strategy and The Cochrane Prognosis Methods Group.

    • This review will use the recommended checklists and tools including the PICOTS model (Population, Index prognostic factor, Comparator prognostic factor, Outcome, Time and Setting) for study selection, the CHARMS-PF (Checklist for Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies for Prognostic Factors) for data extraction, and the QUIPS tool (Quality In Prognostic Studies) to assess the risk of bias.

    • This review may be limited by the quality of and the heterogeneity between the included studies. We aim to address this by performing statistical analysis of heterogeneity combined with subgroup and sensitivity analyses.

    Introduction

    Total elbow replacement (TER) is a recognised treatment for painful arthritic elbow and while rheumatoid arthritis remains the main indication, it is increasingly performed for acute trauma and post-traumatic arthritis.1 Despite technological advances, TER has higher failure rates than other total joint replacements, with a reported 5-year failure rate of 6%–10% and 10-year failure rate of 15%–19% compared with a 3% 5-year failure rate and <5% 10-year failure rate in total hip replacement.1–8 In the event of TER failure, surgical revision is usually required but this procedure carries a risk of adverse events such as ongoing pain, disruption of the elbow extensor mechanism, reduction of bone stock, infection, ulnar nerve injury and poor function.9 Understanding the factors associated with failure is essential to inform the development of strategies designed to reduce failure risk and the need for subsequent revision surgery, which is a burden for both patients and society.10

    A prognostic factor is defined as any variable that is associated with a risk of a health outcome among people with a particular health condition, in this context the outcome is failure, and the condition is patients with TER.11 12 Several existing studies have examined possible factors that may be associated with the risk of failure in TER, including primary pathology, prosthesis type, surgeon’s experience, implant fixation and whether surgery is performed in a specialised centre.2 13–19 However, some of the findings have been inconsistent. For example, some studies examining the association between the underlying pathology and TER failure found rheumatoid arthritis to be associated with a worse prognosis while others found post-traumatic arthritis to be associated with a worse prognosis.6 16

    To the best of our knowledge, there are no systematic reviews available that have appraised, pooled and synthesised all existing studies relating to prognostic factors for TER failure, which means that the quality of the current evidence base is unknown. This also means that it is unclear whether potential factors have been derived from single exploratory studies (thus having provisional prognostic value) or replicated and confirmed in multiple studies (thus demonstrating robust prognostic value).

    Therefore, the purpose of this systematic review is to identify, describe, appraise and synthesise all the current literature examining the prognostic factors associated with failure of TER. A better understanding of prognostic factors may help to explain the differences in risk of failure between patients,20 which may facilitate decisions about whether or not to proceed with TER as a definitive management option. An improved understanding of prognostic factors may also help reduce the number of potential TER revisions by improving treatment selection21 and may help to identify areas for novel interventions.11 This review will also identify gaps in the evidence base to aid the planning of future research and could potentially pave the way to developing a prognostic model in TER that could be used to make individualised patient predictions.

    Methods and analysis

    A systematic review will be conducted and reported in line with the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols guidelines.22 It has been registered at the International Prospective Register of Systematic Reviews and will follow the published guidance and recommendation produced by the PROGnosis REsearch Strategy and The Cochrane Prognosis Methods Group on how to undertake a systematic review and meta-analysis in prognostic factors research.11 12 20 23–25

    Patients and public involvement

    The study aim was codesigned with the Wrightington Hospital Patients and Public Involvement group, including members with long-term health conditions and experience of joint replacement surgery. The group also contributed to a list of possible prognostic factors to be included in the review and their ongoing involvement will help ensure that the review is relevant to patients.

    Study eligibility

    All peer-reviewed studies will be included if prognostic factors are investigated in patients who have undergone a primary TER, and if the selection criteria in table 1 are met (based on the Population, Index prognostic factor, Comparator prognostic factor, Outcome, Time and Setting (PICOTS) model).12 Studies must include a human population and provide a prognostic estimate evaluating prognostic factors for failure of TER. This initial list of candidate prognostic factors to be considered is listed in table 2 but, importantly, this is not exhaustive and studies will also be included if other candidate prognostic factors have been investigated. Although the primary interest is in prognostic factors measured before surgery, we will not exclude prognostic factors measured intraoperatively or postoperatively.

    View this table:
    Table 1

    PICOTS used as selection criteria

    View this table:
    Table 2

    Possible prognostic factors investigated in the literature

    The outcome of failure will be defined as revision surgery. The definition of revision surgery varies between joint registries and is likely to vary between other non-registry studies,26 therefore, a broad definition for revision surgery will be used which is any secondary surgery to the prosthesis, this includes addition, removal or alterations to all or part of the construct on the same elbow occurring at any time following the primary TER. This may lead to the inclusion of studies that evaluate non-revision secondary surgeries as part of a composite outcome. If this occurs, these studies will undergo subgroup/sensitivity analyses.

    The types of evidence included will be randomised trials, non-randomised trials, prospective and retrospective cohort studies, registry studies and case–control studies. If the literature is lacking in studies of these types, then a case series with 50 TERs or more will be considered for inclusion. Results from case series will be interpreted with caution due to the higher risk of bias (RoB), especially selection bias. Review articles, surveys, case reports and conference abstracts will be excluded.

    If the literature is published in a non-English language, an English version will be identified, or the author contacted to check if an English version is available. If an English version is unavailable, the study will be excluded. A list of excluded non-English language studies will be summarised in online supplemental material.

    Studies that report only laboratory/biomechanical work, methodology for identifying new prognostic factors, or results from animal studies will be excluded. Furthermore, if multiple papers are published using the same or overlapping datasets, then only one of these papers will be included, which will be based on the largest number of patients, the most detailed results and/or the longest follow-up time.

    Search strategy

    A broad search will be conducted to ensure all the relevant studies are captured in the search (high sensitivity). As the reporting of prognostic factor studies is generally poor,27 the search will aim to capture all published studies for TER/arthroplasty (low specificity).

    Electronic searches will be conducted using OVID Medline, Embase, PubMed and Cochrane Library databases. The search strategy will include a combination of subject headings (MESH) and free text searches (online supplemental file 1). The proposed search terms were developed with guidance from an experienced information scientist at the University of Manchester. The electronic search will not be limited by the date of publication or the language, or to the human population. This search will be supplemented with a screening of reference lists of the included studies and relevant review studies to identify any further studies that could have been missed in the electronic search. All duplicates will be removed using EndNote V.20 (Philadelphia, USA).

    Study screening

    The study screening process will undergo two phases: (1) screening of titles and abstracts and (2) full-text screening. The screening process will be blinded and all studies in each phase will be screened independently by two reviewers against the specified inclusion and exclusion criteria (table 1). ZH will screen 100% of the studies and the second independent review will be performed by CKG and LMB (50% each) (figure 1). The screening will be conservative, and studies will be included in the next round of screening unless there is an agreement to remove them. Copies of the relevant and unclear studies will be obtained and read thoroughly by the two independent reviewers. Any disagreement will be resolved by discussion or involvement of a third reviewer (JCS). The reason to exclude any study in the full-text screening phase will be provided in online supplemental file. Rayyan software (Cambridge, USA) will be used for the management of the search results.

    Figure 1
    Figure 1

    Summary of the systematic review process.

    Pilot testing of the selection process will be undertaken using 30 articles selected at random and screened by the reviewers. A meeting will be held to review the screening process to determine whether any changes need to be made.

    Data extraction

    The data will be extracted using a standardised tool that will be based on the Checklist for Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies for Prognostic Factors (CHARMS-PF).12 CHARMS-PF covers nine domains: (1) source of data, (2) participants, (3) outcomes to be predicted, (4) prognostic factors, (5) sample size, (6) missing data, (7) analysis, (8) results and (9) interpretation and discussion.

    The data from all included studies will be extracted by two independent reviewers, ZH will complete 100% of data extraction and the second independent data extraction will be performed by CKG and LMB (50% each), and any disagreement will be resolved by discussion. Where disagreement remains, a third reviewer (JCS) will be asked to assess the data and the majority opinion will be accepted. The data extraction tool will initially undergo pilot testing using 10% of the included studies chosen at random and a meeting will be held to review and determine whether any changes need to be made.

    Risk of bias

    The methodological quality or RoB in the included studies will be assessed using the Quality In Prognostic Studies tool which has been recommended and developed specifically for prognostic factor studies28 29. The tool has six domains: (1) selection of study participation, (2) study attrition, (3) prognostic factor measurement, (4) outcome measurement, (5) study covariates and (6) statistical analysis and reporting.

    Two independent reviewers, (ZH) and (TH), will assess each domain and mark with responses ‘yes’, ‘no’ or ‘unclear’. The RoB in each domain will then be classified into ‘low’, ‘moderate’ or ‘high’. Any disagreement between the two reviewers will be resolved by discussion. Where there is disagreement, a third reviewer (JCS) will be asked to assess the study and the majority opinion will be accepted.

    Data synthesis and meta-analysis

    For the quantitative synthesis, prognostic estimates and their precisions will be extracted for each prognostic factor. This will include the risk ratio (RR), OR or HR to measure the prognostic effect, and SEs and CIs to measure the precision. Mean difference will also be accepted to measure associations for continuous variables. Both adjusted and unadjusted prognostic effects will be considered; however, adjusted effect measures will be favoured in the interpretations.12

    Meta-analyses for each prognostic factor will be undertaken for studies that are deemed to be sufficiently robust and comparable. Several challenges are likely to arise due to heterogeneity between studies; therefore, subgroup analysis will be undertaken of groups of studies with similar characteristics. If the number of studies allows, then separate meta-analyses will be undertaken for groups with similar prognostic effect measures, for example:

    • HRs, ORs and RRs separately.

    • Unadjusted and adjusted associations separately.

    • Prognostic factor effects at distinct cut points (or groups of similar cut points) separately.

    • Prognostic factor effects corresponding to a linear trend (association) separately.

    • Prognostic factor effects corresponding to non-linear trends separately.

    • Prognostic factors at different times of measurement (preoperatively, intraoperatively or postoperatively) separately.

    • Each method of measurement (for factors and outcomes) separately.

    Sensitivity analysis will also be performed when there is heterogeneity between studies caused by differences in:

    • The outcome (revision surgery) definition.

    • The participations baseline characteristics (casemix).

    • The RoB.

    • The study designs.

    • The type of TER included in the study.

    If heterogeneity persists despite subgroup analyses, then a random effect approach will be used to allow for unexplained heterogeneity across included studies. This will be performed if five or more studies are included in a meta-analysis, otherwise, a fixed effect approach will be used.30

    Statistical analyses will be performed using R V.4.2.2 or later and RStudio software (Vienna, Austria). If quantitative synthesis is not possible because of issues such as a paucity of evidence, overall evidence quality and methodological heterogeneity, a narrative synthesis of the evidence will be undertaken.

    Assessment of heterogeneity

    Pooled data will be used in the review and heterogeneity will be assessed between each group using forest plots with 95% CIs for the pooled measures. Heterogeneity will also be assessed by calculating the Q statistic and I2. If there is heterogeneity across the included studies (I2>40%),31 then the overall pooled data estimate will not be reported. The heterogeneity results will still be reported.

    Reporting bias

    Reporting deficiencies will be assessed using a funnel plot which addresses ‘small-study effects’. The funnel plot will be used when 10 or more studies are examining the same prognostic factor and included in the same meta-analysis.32 The Peters’ and Debray’s tests will be used to test for asymmetry in the funnel plot.32 33 The asymmetry in the plot cannot differentiate between differences caused by bias (publication or selective reporting biases) or heterogeneity; therefore, the results will be interpreted with caution.

    Reporting

    The quality of evidence relating to each prognostic factor will be evaluated using the adapted Grades of Recommendation Assessment, Development and Evaluation (GRADE) framework that is specific to systematic reviews of prognostic factor research.34 The GRADE contains five domains:

    • RoB.

    • Inconsistency.

    • Imprecision.

    • Indirectness.

    • And publication bias.

    The GRADE system will be used to summarise whether the evidence for each prognostic value is of high, moderate, low or very low quality. The online The GRADEpro Guideline Development Tool will be used.

    Ethics statements

    Patient consent for publication

    References

    1. ↵
      1. Macken AA,
      2. Prkic A,
      3. Kodde IF, et al
      . Global trends in indications for total elbow arthroplasty: a systematic review of national registries. EFORT Open Rev 2020;5:215–20. doi:10.1302/2058-5241.5.190036
      OpenUrlPubMed
    2. ↵
      1. Chou T-FA,
      2. Ma H-H,
      3. Wang J-H, et al
      . Total elbow arthroplasty in patients with rheumatoid arthritis. Bone Joint J 2020;102-B:967–80. doi:10.1302/0301-620X.102B8.BJJ-2019-1465.R1
      OpenUrl
    3. ↵
      1. Evans JP,
      2. Evans JT,
      3. Mohammad HR, et al
      . How long does an elbow replacement last? A systematic review and meta-analysis of case-series and national registry reports with more than 10 years of follow-up. Acta Orthop 2022;93:495–502. doi:10.2340/17453674.2022.2753
      OpenUrl
    4. ↵
      1. Krenek L,
      2. Farng E,
      3. Zingmond D, et al
      . Complication and revision rates following total elbow arthroplasty. J Hand Surg Am 2011;36:68–73. doi:10.1016/j.jhsa.2010.09.036
      OpenUrlCrossRef
    5. ↵
      1. Little CP,
      2. Graham AJ,
      3. Carr AJ
      . Total elbow Arthroplasty: a systematic review of the literature in the English language until the end of 2003. J Bone Joint Surg Br 2005;87:437–44. doi:10.1302/0301-620X.87B4.15692
      OpenUrlCrossRefPubMed
    6. ↵
      1. Prkic A,
      2. Welsink C,
      3. The B, et al
      . Why does total elbow arthroplasty fail today? A systematic review of recent literature. Arch Orthop Trauma Surg 2017;137:761–9. doi:10.1007/s00402-017-2687-x
      OpenUrl
    7. ↵
      1. Voloshin I,
      2. Schippert DW,
      3. Kakar S, et al
      . Complications of total elbow replacement: a systematic review. J Shoulder Elbow Surg 2011;20:158–68. doi:10.1016/j.jse.2010.08.026
      OpenUrlCrossRefPubMedWeb of Science
    8. ↵
      1. Welsink CL,
      2. Lambers KTA,
      3. van Deurzen DFP, et al
      . Total elbow Arthroplasty: a systematic review. JBJS Rev 2017;5:e4. doi:10.2106/JBJS.RVW.16.00089
    9. ↵
      1. Chin K,
      2. Lambert S
      . Revision total elbow replacement. J Clin Orthop Trauma 2021;20:101495. doi:10.1016/j.jcot.2021.101495
    10. ↵
      1. Wagner ER,
      2. Ransom JE,
      3. Kremers HM, et al
      . Comparison of the hospital costs for two-stage reimplantation for deep infection, single-stage revision and primary total elbow arthroplasty. Shoulder Elbow 2017;9:279–84. doi:10.1177/1758573217706364
      OpenUrl
    11. ↵
      1. Steyerberg EW,
      2. Moons KGM,
      3. van der Windt DA, et al
      . Prognosis research strategy (PROGRESS) 2: prognostic factor research. PLoS Med 2013;10:e1001381. doi:10.1371/journal.pmed.1001381
    12. ↵
      1. Riley RD,
      2. Moons KGM,
      3. Snell KIE, et al
      . A guide to systematic review and meta-analysis of prognostic factor studies. BMJ 2019;364:k4597. doi:10.1136/bmj.k4597
    13. ↵
      1. Fevang B-TS,
      2. Lie SA,
      3. Havelin LI, et al
      . Results after 562 total elbow replacements: a report from the Norwegian Arthroplasty register. J Shoulder Elbow Surg 2009;18:449–56. doi:10.1016/j.jse.2009.02.020
      OpenUrlCrossRefPubMedWeb of Science
    14. ↵
      1. Jenkins PJ,
      2. Watts AC,
      3. Norwood T, et al
      . Total elbow replacement: outcome of 1,146 arthroplasties from the Scottish arthroplasty project. Acta Orthop 2013;84:119–23. doi:10.3109/17453674.2013.784658
      OpenUrlCrossRefPubMedWeb of Science
    15. ↵
      1. Krukhaug Y,
      2. Hallan G,
      3. Dybvik E, et al
      . A survivorship study of 838 total elbow replacements: a report from the Norwegian arthroplasty register 1994-2016. J Shoulder Elbow Surg 2018;27:260–9. doi:10.1016/j.jse.2017.10.018
      OpenUrl
    16. ↵
      1. Plaschke HC,
      2. Thillemann TM,
      3. Brorson S, et al
      . Implant survival after total elbow arthroplasty: a retrospective study of 324 procedures performed from 1980 to 2008. J Shoulder Elbow Surg 2014;23:829–36. doi:10.1016/j.jse.2014.02.001
      OpenUrlCrossRefPubMed
    17. ↵
      1. Skyttä ET,
      2. Eskelinen A,
      3. Paavolainen P, et al
      . Total elbow arthroplasty in rheumatoid arthritis: a population-based study from the finnish arthroplasty register. Acta Orthop 2009;80:472–7. doi:10.3109/17453670903110642
      OpenUrlCrossRefPubMedWeb of Science
    18. ↵
      1. Viveen J,
      2. van den Bekerom MPJ,
      3. Doornberg JN, et al
      . Use and outcome of 1,220 primary total elbow arthroplasties from the australian orthopaedic association national joint arthroplasty replacement registry 2008-2018. Acta Orthop 2019;90:511–6. doi:10.1080/17453674.2019.1657342
      OpenUrl
    19. ↵
      1. Baghdadi YMK,
      2. Veillette CJH,
      3. Malone AA, et al
      . Total elbow Arthroplasty in obese patients. J Bone Joint Surg Am 2014;96:e70. doi:10.2106/JBJS.M.00364
    20. ↵
      1. Riley RD,
      2. Moons KG,
      3. Hayden JA, et al
      . Prognostic factor research in Healthcare. In: van der Windt DA, Hemingway H, Croft P, et al., eds. Prognosis Research in Healthcare: Concepts, Methods, and Impact. Oxford University Press, 2019.
    21. ↵
      1. Damen J,
      2. Hooft L
      . The increasing need for systematic reviews of prognosis studies: strategies to facilitate review production and improve quality of primary research. Diagn Progn Res 2019;3:2. doi:10.1186/s41512-019-0049-6
    22. ↵
      1. Moher D,
      2. Shamseer L,
      3. Clarke M, et al
      . Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Syst Rev 2015;4:1. doi:10.1186/2046-4053-4-1
    23. ↵
      1. Hemingway H,
      2. Croft P,
      3. Perel P, et al
      . Prognosis research strategy (PROGRESS) 1: a framework for researching clinical outcomes. BMJ 2013;346(feb05 1):e5595. doi:10.1136/bmj.e5595
    24. ↵
      1. Hingorani AD,
      2. Windt DA van der,
      3. Riley RD, et al
      . Prognosis research strategy (PROGRESS) 4: stratified medicine research. BMJ 2013;346:e5793. doi:10.1136/bmj.e5793
    25. ↵
      1. Steyerberg EW,
      2. Moons KGM,
      3. van der Windt DA, et al
      . Prognosis research strategy (PROGRESS) 3: prognostic model research. PLOS Med 2013;10:e1001381. doi:10.1371/journal.pmed.1001381
    26. ↵
      1. Liebs TR,
      2. Splietker F,
      3. Hassenpflug J
      . An analysis of national arthroplasty registries' definitions of revision. Clin Orthop Relat Res 2015;473:3421–30. doi:10.1007/s11999-015-4255-4
      OpenUrl
    27. ↵
      1. Geersing G-J,
      2. Bouwmeester W,
      3. Zuithoff P, et al
      . Search filters for finding prognostic and diagnostic prediction studies in MEDLINE to enhance systematic reviews. PLoS One 2012;7:e32844. doi:10.1371/journal.pone.0032844
    28. ↵
      1. Hayden JA,
      2. Côté P,
      3. Bombardier C
      . Evaluation of the quality of prognosis studies in systematic reviews. Ann Intern Med 2006;144:427–37. doi:10.7326/0003-4819-144-6-200603210-00010
      OpenUrlCrossRefPubMedWeb of Science
    29. ↵
      1. Hayden JA,
      2. van der Windt DA,
      3. Cartwright JL, et al
      . Assessing bias in studies of prognostic factors. Ann Intern Med 2013;158:280–6. doi:10.7326/0003-4819-158-4-201302190-00009
      OpenUrlCrossRefPubMedWeb of Science
    30. ↵
      1. IntHout J,
      2. Ioannidis JPA,
      3. Borm GF
      . The Hartung-Knapp-Sidik-Jonkman method for random effects meta-analysis is straightforward and considerably outperforms the standard Dersimonian-Laird method. BMC Med Res Methodol 2014;14:25. doi:10.1186/1471-2288-14-25
    31. ↵
      1. Higgins JPT,
      2. Thomas J,
      3. Chandler J, et al
      . Cochrane Handbook for systematic reviews of interventions. In: Cochrane Handbook for Systematic Reviews of Interventions. 2nd Edition. Chichester (UK): John Wiley & Sons, 23 September 2019. doi:10.1002/9781119536604
    32. ↵
      1. Debray TPA,
      2. Moons KGM,
      3. Riley RD
      . Detecting small-study effects and funnel plot asymmetry in meta-analysis of survival data: a comparison of new and existing tests. Res Synth Methods 2018;9:41–50. doi:10.1002/jrsm.1266
      OpenUrl
    33. ↵
      1. Peters JL,
      2. Sutton AJ,
      3. Jones DR, et al
      . Comparison of two methods to detect publication bias in meta-analysis. JAMA 2006;295:676–80. doi:10.1001/jama.295.6.676
      OpenUrlCrossRefPubMedWeb of Science
    34. ↵
      1. Huguet A,
      2. Hayden JA,
      3. Stinson J, et al
      . Judging the quality of evidence in reviews of prognostic factor research: adapting the GRADE framework. Syst Rev 2013;2:71. doi:10.1186/2046-4053-2-71

    Supplementary materials

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Footnotes

    • LK-F, JCS and ACW are joint senior authors.

    • Twitter @zaidahamoodi, @GehringerCelina, @Lucyy31, @dt_hughes, @KearsleyFleet, @jamiecsergeant, @myelbowdoc

    • LK-F, JCS and ACW contributed equally.

    • Contributors ZH, LK-F, ACW and JCS provided the idea of the topic. ZH designed and wrote the protocol. CKG and LMB contributed to designing the search strategy, screening process and data extraction. TH contributed to the methodology of the risk of bias assessment and group for sensitivity analysis. JCS contributed to the design of the statistical methods. ACW contributed to a list of a priori prognostic factors. ZH coordinated the whole process. All authors read, provided feedback, input into the methodology and approved the final manuscript. LK-F, ACW and JCS contributed to this work equally.

    • Funding This work was supported by funding from: The joint National Joint Registry (NJR) and Royal College of Surgeons of England (RCS England) Research Fellowship (grant number: NA). The John Charnley Trust (grant number: NA). Versus Arthritis (grant number: 21755). ZH research fellowship is funded by the NJR/RCSEng and his MD degree is funnded by the John Charnley Trust.

    • Competing interests None declared.

    • Patient and public involvement Patients and/or the public were involved in the design, or conduct, or reporting, or dissemination plans of this research. Refer to the Methods section for further details.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.