Surgical interventions for post-traumatic anterior shoulder instability in adults

  • Protocol
  • Intervention


  • François Desmeules,

    Corresponding author
    1. Maisonneuve-Rosemont Hospital Research Center, University of Montreal Affiliated Research Center, Orthopaedic Clinical Research Unit, Montreal, Quebec, Canada
    2. University of Montreal, School of Rehabilitation, Faculty of Medicine, Montreal, Quebec, Canada
    Search for more papers by this author
  • Janie Barry,

    1. Maisonneuve-Rosemont Hospital Research Center, University of Montreal Affiliated Research Center, Orthopaedic Clinical Research Unit, Montreal, Quebec, Canada
    Search for more papers by this author
  • Jean-Sébastien Roy,

    1. Laval University, Department of Rehabilitation, Faculty of Medicine, Quebec City, Canada
    2. Centre for Interdisciplinary Research in Rehabilitation and Social Integration, Quebec City, Canada
    Search for more papers by this author
  • Pascal-André Vendittoli,

    1. University of Montreal, Department of Surgery, Montreal, Québec, Canada
    2. University of Montreal Affiliated Research Center, Orthopaedic Clinical Research Unit, Maisonneuve-Rosemont Hospital Research Center, Montreal, Canada
    Search for more papers by this author
  • Dominique M Rouleau

    1. University of Montreal, Department of Surgery, Montreal, Québec, Canada
    Search for more papers by this author


This is the protocol for a review and there is no abstract. The objectives are as follows:

To assess the effects (benefits and harms) of different surgical interventions for post-traumatic anterior shoulder instability in adults.

Our main comparisons will be:

  • anatomic (anatomical) versus non-anatomic (non-anatomical) methods

  • arthroscopic versus open surgical approaches

  • different suture techniques or fixation materials


Description of the condition

The glenohumeral or shoulder joint is a ball-and-socket joint that consists of the head of the humerus (upper-arm bone) and the glenoid fossa, a shallow dish-shaped part of the scapula (shoulder blade). The shoulder joint has the greatest mobility of all joints in the body; this reflects the relative positioning, dimensions and mobility of its components, where mainly soft tissues (e.g. the rotator cuff tendons and the muscles linking the scapula to the thoracic cage) rather than bone act as restraints. In consequence, it is the most commonly dislocated joint in the body (Veeger 2011).

Shoulder or glenohumeral instability may be defined as the inability, during active shoulder movements, to maintain the humeral head within the shoulder joint or, more precisely, centred with the glenoid fossa of the scapula. Shoulder instability is a debilitating condition most often encountered in active young men; it can cause severe pain and disability, often leading to inability to work or participate in sports (Cadet 2010). The most common form of glenohumeral instability is said to be anterior (the humeral head is pushed out of the joint in a forward direction) or anteroinferior (forward and downwards), as up to 98% of instability cases have been reported to occur in these directions (Liu 1996). This results, in almost all cases, from a traumatic event such as a fall with the arm outstretched or a direct blow to the shoulder. These types of traumatic events cause partial dislocation (subluxation) or dislocation of the glenohumeral joint, leading to injuries to the shoulder joint structures. The overall incidence of traumatic shoulder dislocation has been reported to be 1.7% in the general population (Romeo 2001), and it accounts for 50% of all dislocations encountered in the emergency department (Blake 1999).

The most commonly injured structures, which are often stretched or torn, are the labrum (a fibrocartilaginous rim attached around the margin of the glenoid fossa of the shoulder blade), deep shoulder tendons of the rotator cuff and shoulder joint ligaments. The humeral head or the glenoid may also be fractured (Grumet 2010). Common pathoanatomy findings following traumatic anterior dislocation include a Bankart lesion and a torn labrum at the anteroinferior part of the glenoid (Grumet 2010). In more severe cases, a bony Bankart lesion may be present when the glenoid bone is broken off with the anterior labrum. A Hill-Sachs lesion may often be present at the posterior aspect of the humeral head. The Hill-Sachs lesion is an osseous defect caused by the impaction of the posterior humeral head on the anteroinferior labrum when the shoulder dislocates (Provencher 2012). Once these structures have been injured, the stability of the glenohumeral joint is compromised, in particular with dynamic movements of the arm in overhead positions. Up to 50% of people may experience recurrent episodes of subluxation or other episodes of dislocation subsequent to their initial traumatic dislocation (Robinson 2006).

There is still much debate regarding the best treatment for people suffering from anterior shoulder instability (Grumet 2010). Conservative (non-surgical) treatment typically comprises shoulder immobilisation in a sling followed by rehabilitation involving exercises to regain dynamic control of the humeral head and to strengthen the shoulder rotator cuff muscles. A Cochrane review has concluded that the effectiveness of different conservative management options for this population is uncertain (Handoll 2006). Emerging evidence suggests that, especially in young active patients, early surgical intervention to repair shoulder joint structures followed by immobilisation and rehabilitation may be more efficacious (Grumet 2010). The Cochrane review comparing surgical versus non-surgical management found that highly active young people treated surgically after an acute anterior shoulder dislocation were less likely to have an unstable shoulder (Handoll 2004).

Description of the intervention

There are two basic types of surgical approach for shoulders with post-traumatic anterior instability: ‘anatomic’ (anatomical) and ‘non-anatomic’ (non-anatomical) repairs.

Anatomic repairs, which aim to restore normal anatomy, involve the repair of damaged structures, such as a Bankart lesion, and repair or reconstruction of damaged capsule and torn ligaments. Thus the aim of a typical Bankart repair procedure is to reattach the torn labrum to the anterior glenoid fossa with transosseous sutures to re-establish normal shoulder anatomy and function (Randelli 2012). In some cases when a large and engaging Hill-Sachs lesion is present, the remplissage technique may be added, i.e. a transfer of the infraspinatus tendon and the posterior capsule is done to fill the Hill-Sachs defect (Zhu 2011).

The goal of non-anatomic repairs, such as the Bristow-Laterjet procedure and the less commonly performed Magnusson-Stack and Putti-Platt procedures (Ahmad 2005), is to restore stability to the glenohumeral joint by compensating for capsulolabral and osseous injuries with an osseous or soft tissue block to passively limit glenohumeral translation (Van Tongel 2011). The Bristow-Latarjet procedure is performed mainly in people who have bone loss from the front of the glenoid. This procedure involves an osteotomy of the coracoid process with pectoral tendon conservation. The coracoid tip is transferred with the conjoint tendon and is attached with one or two screws to the anterior glenoid (Matthes 2007). The goal of this surgery is to compensate for bone loss, and the transferred muscle acts as an additional muscular strut to prevent further dislocations (Omidi-Kashani 2008). The Magnusson-Stack procedure aims to tighten the subscapularis by altering its insertion site from the lesser tuberosity to a groove created lateral to the bicipital groove. The Putti-Platt procedure involves shortening the subscapularis tendon.

Each of these methods of surgery has undergone several modifications since its first appearance in orthopaedic practice; changes involve not only the surgical approach (arthroscopic or open) but also anchor or suture types and their placement. Following any of these surgical interventions, immobilisation and a rehabilitation exercise regimen are generally warranted to allow proper healing of repaired structures and to regain normal range of motion and normal dynamic control of the glenohumeral joint (Hayes 2002).

How the intervention might work

Common surgical procedures for post-traumatic anterior shoulder instability in adults are aimed at restoring the integrity of damaged structures and ultimately enabling return to normal activities without functional disability (Brophy 2009). Several risk factors for recurrent instability after surgical repair have, however, been identified: bony Bankart lesions, engaging Hill–Sachs lesions, poor glenohumeral ligament mechanical properties, specific glenoid anatomy described as an inverted pear, participation in contact sports and younger age. It is important to note that these factors may also influence the type of surgery that will be performed (Van Tongel 2011).

Nowadays, surgeons generally use surgical interventions that aim to restore normal anatomy; such as the Bankart procedure, which involves repair of the Bankart lesion. The reported complications of such anatomic procedures are generally low. Open anatomical repairs can provide good results, but arthroscopically performed anatomical repairs, now often used, seem to be of equal efficacy while being better tolerated by patients and decreasing morbidity (Van Tongel 2011). Non-anatomic methods, such as the Putti-Platt procedure, where tendon or muscles are shortened/transferred in an attempt to stabilise and strengthen the shoulder joint, are often preferred by surgeons when more extensive damage to the soft tissues and bony structures of the shoulder is present. While these methods are considered effective in reducing the recurrence of instability, complications are more frequent, including loss of range of motion and arthropathy (Murray 2013). Non-anatomic repairs are performed most often with an open approach. Arthroscopically performed non-anatomic repair is also an option but is more technically challenging (Van Tongel 2011).

Why it is important to do this review

Overall, it is uncertain what is the best surgical procedure for treating traumatic anterior shoulder instability (Grumet 2010). This uncertainty includes also the type of surgical approach (e.g. arthroscopic or open surgery) and the type of sutures or anchors used to stabilise and repair shoulder structures (Brophy 2009). The previous Cochrane review (Pulavarti 2009), which focused on a similar topic to this review, included three trials comparing arthroscopic versus open surgery. This review found "insufficient evidence" to inform practice for this comparison. Our review will update the search for evidence and will expand the scope to include all comparisons of surgical interventions, including different types of fixation materials and techniques, used for treating adults with post-traumatic anterior shoulder instability.


To assess the effects (benefits and harms) of different surgical interventions for post-traumatic anterior shoulder instability in adults.

Our main comparisons will be:

  • anatomic (anatomical) versus non-anatomic (non-anatomical) methods

  • arthroscopic versus open surgical approaches

  • different suture techniques or fixation materials


Criteria for considering studies for this review

Types of studies

Randomised or quasi-randomised controlled trials (in which the method of allocating participants to a treatment is not strictly random; e.g. by date of birth, hospital record number, alternation) evaluating surgical interventions for post-traumatic anterior shoulder instability in adults.

Types of participants

Individuals, 16 years of age or older, being treated for post-traumatic shoulder instability. We will include adults with either first-time (acute) traumatic anterior shoulder instability or multiple events following post-traumatic anterior shoulder instability, including subluxation or dislocation. Diagnosis can be based on participant history, clinical examination and investigations such as examination under anaesthesia, imaging (x-ray, computed tomography, magnetic resonance imaging) or arthroscopy. Participants with concomitant rotator cuff tears will not be excluded. Although trials comparing different surgical interventions for treating rotator cuff tears will be excluded, trials evaluating surgical interventions for shoulder instability in which rotator cuff tears are repaired will be included, provided the policy applies to all trial participants.

Types of interventions

We will include various surgical interventions such as the Bankart procedure (with or without remplissage), the Bristow-Laterjet or other types of interventions aimed at restoring the integrity of damaged structures in people suffering from traumatic anterior shoulder instability. Procedures may be performed arthroscopically or by an open method. Studies comparing different types of sutures or anchor types for specific procedures will also be included in the review. The newer technique under study, which is often less invasive, will be considered the experimental treatment, and the control intervention will be considered the usual approach - it is often the more invasive approach.

Types of outcome measures

Primary outcomes
  • Validated participant-reported functional assessment and shoulder rating tools such as Western Ontario Shoulder Instability Index (WOSI); Constant-Murley Score; Shoulder Instability Questionnaire (SIQ); American Shoulder and Elbow Surgeons Evaluation Form (ASES); University of California–Los Angeles (UCLA) Shoulder Score; Rowe Score; Disabilities of the Arm, Shoulder, and Hand scale (DASH) and the Simple Shoulder Test (SST) (Fayad 2005; Roy 2009)

  • Recurrence: participants with episodes of instability following surgical intervention (although 'instability' can be diagnosed by history and clinical examination, with further dislocation confirmed by x-ray, preference will be given to the episodes diagnosed by a medical practitioner or that required a medical intervention)

  • Adverse effects

    • Short term (blood loss, nerve injury, infections, other adverse effects associated with anaesthesia, rotator cuff tear, loss of range of motion)

    • Long term (nerve injury, rotator cuff tear, loss of range of motion)

    • Revision surgery

    • Degenerative arthritis

Secondary outcomes
  • Pain (measured by visual analogue scale (VAS) and validated questionnaires)

  • Self-reported health-related quality of life (measured by validated questionnaires such as Short Form (SF)-36, EuroQol-5D (standardised measure of health outcome), Sickness Impact Profile

  • Participant satisfaction (any type of measure)

  • Recovery defined as return to pre-injury level of activity (sports and/or work)

  • Objective instability during physical clinical tests

  • Range of motion

  • Shoulder muscle strength (isometric or isotonic, measured with a dynamometer)

Timing of outcome measurement

Timing of primary outcome measurement will also be assessed (short, medium and long term) and considered in analyses for all primary outcomes. We propose to present the results in the following time frames: short term (up to three months); medium term (> three to 12 months); and long term (> 12 months). Preference will be given to long-term outcomes.

Search methods for identification of studies

Electronic searches

We will search the Cochrane Bone, Joint and Muscle Trauma Group Specialised Register (to present), the Cochrane Central Register of Controlled Trials (CENTRAL)(current issue), MEDLINE (1946 to present), EMBASE (1980 to present) and CINAHL (Cumulative Index to Nursing and Allied Health Literature) (1981 to present). No language restrictions will be applied.

In MEDLINE, a subject-specific strategy will be combined with the sensitivity-maximising version of the Cochrane highly sensitive search strategy for identifying randomised trials (Lefebvre 2011) (see Appendix 1). Search strategies for CENTRAL and EMBASE are also shown in Appendix 1. Other search strategies will be developed at a later stage.

Searching other resources

We will search the World Health Organization (WHO) International Clinical Trials Registry Platform for ongoing and recently completed trials. Conference proceedings published in Orthopaedic Proceedings, a supplement of The Bone & Joint Journal (BJJ: formally Journal of Bone and Joint Surgery (JBJS) (British)), will also be searched. This database contains abstracts of papers presented at recent scientific meetings organised by orthopaedic associations and specialist societies. The table of contents of the following journals will be searched (2010 onwards): BJJ, JBJS (American), American Journal of Sports Medicine, British Journal of Sports Medicine and Journal of Elbow and Shoulder Surgery. We will handsearch references of the included randomised controlled trials (RCTs) and previous reviews on the subject.

Data collection and analysis

Selection of studies

Two review authors (FD and JB) will independently screen the search results and, after obtaining full texts of potentially eligible trials when necessary, will independently select trials that fulfil our inclusion criteria. Disagreements will be resolved through discussion and, when necessary, by consultation with a third review author (DR or PV). We will attempt to contact trial authors for clarification of study methods and characteristics if necessary to establish trial eligibility.

Data extraction and management

Two review authors (FD and JB) will independently extract information on study characteristics and results using a piloted data extraction form. Collected data will include: study type, authors, year of publication and objective; inclusion and exclusion criteria; randomisation procedure, allocation and blinding and interventions and care programmes as well as co-interventions; types of outcome measures and follow-up time points; baseline characteristics of participants (age, sex, hyperlaxity, leisure or competition sport participation, contact or non-contact sport participation, first-time acute traumatic instability event versus recurrent, presence of osseous deficit, bony Bankart lesion and/or a rotator cuff tear); attrition and results; and trial author conclusions. Disagreements will be resolved through consensus. We will attempt to contact trial authors when details of study methods or results are incomplete.

Assessment of risk of bias in included studies

We will assess the risk of bias of the included studies using The Cochrane Collaboration's tool for assessing risk of bias (Chapter 8; Higgins 2011). This tool includes seven domains: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective reporting and other sources of bias. Examples of other sources of bias that we will conisder are premature stopping due to some data-dependent process, extreme baseline imbalance and different experience or expertise of surgeons performing the operations in different intervention groups. We will assess risk of bias from blinding of outcome assessment separately for subjective and objective outcome measures. After independent evaluation of each study by pairs of raters, two review authors (FD and DR) will compare ratings and will resolve differences. An independent third evaluator (JSR) will make a judgement when consensus is not achieved. For each of these seven domains, we will assign a judgement of 'high risk' of bias, 'low risk' of bias or 'unclear risk' of bias based on the criteria in Table 8.5.d of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011).

Measures of treatment effect

We will calculate risk ratios (RRs) and 95% confidence intervals (95% CIs) for dichotomous outcomes; mean differences (MDs) with 95% CIs for continuous outcomes and standardised mean differences (SMDs) with 95% CIs when data from continuous outcomes are pooled using different scales.

Unit of analysis issues

We anticipate that the unit of randomisation in the included trials will be the individual participant. However, when trials included adults with bilateral instability, results may be presented for shoulders rather than for individual participants. When such unit of analysis issues arise and appropriate corrections have not been made, we will present the data for such trials only when the disparity between the units of analysis and randomisation is small. We will avoid unit of analysis issues related to repeated observations of the same outcome, such as by presenting separate data for different periods of follow-up, as described above (Section 9.3.1; Higgins 2011).

Dealing with missing data

When possible, we will perform intention-to-treat analysis. If there is a discrepancy in the number randomised and the numbers analysed in each treatment group, we will calculate the percentage loss to follow-up in each group and will report this information. If dropouts exceed 10% for any trial, we will assign the worst outcome to those lost for the related outcomes and will assess the impact of this in sensitivity analyses with the results of completers. We will calculate missing standard deviations from other available data such as standard errors when possible (Section; Higgins 2011). Values will be extracted from graphs when numerical data are not reported. We will contact trial investigators to request missing information and data, in particular for primary outcomes.

Assessment of heterogeneity

We will assess clinical heterogeneity in terms of participants, interventions and outcomes of the clinical trials. We will assess statistical heterogeneity through visual examination of the forest plot, primarily to check for overlapping confidence intervals, and by a Chi² test, where a P value < 0.l will be taken as indicating significant heterogeneity. We will also use the I² statistic to describe the percentage of the variability in effect estimates that is due to heterogeneity rather than to random error. We will base our interpretation of the I² results as suggested in Higgins 2011: 0% to 40% might not be important; 30% to 60% may represent moderate heterogeneity; 50% to 90% may represent substantial heterogeneity; and 75% to 100% may represent very substantial heterogeneity.

Assessment of reporting biases

To reduce reporting bias, we plan to search for published and unpublished studies without language restrictions. We will look for multiple publications from single studies. When possible, we will try to obtain the protocol or the clinical registration documents for all trials and will contact the authors of unpublished trials to ask for unpublished results. All included studies will be assessed for adequacy of reporting data and outcomes. Reporting biases will be appraised with the risk of bias assessments tool for each trial. To investigate the likelihood of publication bias in our review, funnel plots will be generated, provided that at least 10 studies assessing any one outcome are identified.

Data synthesis

When considered appropriate, we will pool results of comparable groups of trials using both fixed-effect and random-effects models. The model we choose to report will be guided by a careful consideration of the extent of heterogeneity and whether it can be explained, in addition to other factors such as the number and size of studies that are included. Ninety-five per cent confidence intervals will be used throughout. We will consider not pooling data when considerable heterogeneity is noted (I² > 75%) that cannot be explained by the diversity of methodological or clinical features among trials. When it is inappropriate to pool data, we will still present trial data in the analyses or tables for illustrative purposes and will report these in the text.

When outcome measurements are determined using the same scale, the mean difference will be combined as a summary statistic. The SMD will be used when studies assess the same outcome but measure it on different scales. If the scales used in the trials differ in the direction of scoring, then the mean values from one set of scales will be multiplied by -1 to ensure that the direction of scores across trials is comparable.

Subgroup analysis and investigation of heterogeneity

When possible, the following subgroup analyses will be performed for each comparison.

  • Adults with first-time acute traumatic anterior shoulder instability versus adults with multiple event post-traumatic anterior shoulder instability

  • Surgical interventions with and without remplissage

  • People up to 40 years of age versus people over 40 years of age (age categories of younger than 30, 30 to 40, 40 to 50 and over 50 years will also be considered)

  • Athletes versus non-athletes

We will investigate whether the results of subgroups are significantly different by inspecting the overlap of confidence intervals and performing the test for subgroup differences available in Review Manager software.

Sensitivity analysis

We will conduct sensitivity analyses to investigate the robustness of the results for the primary outcomes by excluding trials at high risk of bias, primarily related to failure of allocation concealment. We will also perform sensitivity analyses with trials reporting dropout rates of 10% or greater, to ascertain differences in outcomes of intention-to-treat (ITT) analysis (all dropouts will be assigned to the worst outcome) and analysis with completers only (i.e. those available at follow-up). If significant heterogeneity is detected, and if there are some outlying studies with results that conflict with the other studies, sensitivity analyses will be performed; we will consider potential confounding clinical characteristics of participants, such as age, hyperlaxity, leisure or competition sport participation, contact or non-contact sport participation and first-time acute traumatic instability event versus recurrent event, as well as the presence of an osseous deficit, a bony Bankart lesion or a rotator cuff tear. Methodological characteristics of the outlying studies will also be assessed and sensitivity analyses will be performed to examine the effect of their removal.

Additional sensitivity analyses will test the selection of statistical model (fixed-effect versus random-effects) for pooling, and the effects of including trials in which some participants have bilateral involvement.

Assessing the quality of the body of evidence

We shall use the GRADE approach to assess the quality of evidence related to the primary outcomes listed in Types of outcome measures and four secondary outcomes (pain, self-reported health-related quality of life, participant satisfaction and recovery defined as return to pre-injury level of activity in sports and/or work) (see Chapter 12.2, Higgins 2011). We will prepare 'Summary of findings' tables for the main comparisons if there is sufficient evidence.


We would like to thank Mario Lenza, Muhammad-Adeel Ahktar and Helen Handoll for providing helpful comments about this protocol. We are also grateful to Joanne Elliott for her assistance in developing the search strategies and Laura MacDonald for editorial processing.


Appendix 1. Search strategies

The Cochrane Library (Wiley Online Library)

#1    MeSH descriptor: [Shoulder Dislocation] this term only  
#2    [mh ^shoulder] or [mh ^"Shoulder Joint"]           
#3    MeSH descriptor: [Shoulder Dislocation] this term only
#4    MeSH descriptor: [Dislocations] this term only and with qualifiers: [Surgery - SU]
#5    dislocate* or sublux* or instability or unstable or stabili?ation:ti,ab,kw  (Word variations have been searched)       
#6    #3 or #4 or #5   
#7    #2 and #6        
#8    (shoulder* or glenohumeral or gleno-humeral) near/3 (dislocat* or sublux* or instability or unstable or stabili?ation):ti,ab,kw  (Word variations have been searched)   
#9    #1 or #7 or #8   
#10  MeSH descriptor: [Orthopedic Procedures] this term only           
#11  MeSH descriptor: [Surgical Procedures, Operative] explode all trees       
#12  MeSH descriptor: [Orthopedic Fixation Devices] explode all trees           
#13  MeSH descriptor: [Arthroscopy] this term only   
#14  MeSH descriptor: [Suture Techniques] this term only     
#15  surg* or operat* or arthroscop* or repair*:ti,ab,kw  (Word variations have been searched)  
#16  Any MeSH descriptor with qualifier(s): [Surgery - SU]     
#17  {or #10-#16}     
#18  #9 and #17       

MEDLINE (Ovid Online)

1   Shoulder Dislocation/
2   Shoulder/ or Shoulder Joint/
3   Joint Instability/
4   Dislocations/su
5   (dislocate* or sublux* or instability or unstable or stabili?ation).tw.
6   or/3-5
7   2 and 6
8   ((shoulder* or glenohumeral or gleno-humeral) adj3 (dislocat* or sublux* or instability or  unstable or stabili?ation)).tw. (4501)
9   1 or 7 or 8
10  Orthopedic Procedures/
11  exp Surgical Procedures, Operative/
12  exp Orthopedic Fixation Devices/
13  Arthroscopy/
14  Suture Techniques/
15  (surg* or operat* or arthroscop* or repair*).tw.
16  su.fs.
17  or/10-16
18  9 and 17
19  Randomized controlled
20  Controlled clinical
21  randomized.ab.
22  placebo.ab.
23  Drug Therapy.fs.
24  randomly.ab.
25  trial.ab.
26  groups.ab.
27  or/19-26
28  exp Animals/ not Humans/
29  27 not 28
30 18 and 29

EMBASE (Ovid Online)

1    Shoulder dislocation/
2    Shoulder/ or Shoulder injury/
3    Joint instability/
4    Dislocation/su [Surgery]
5    (dislocate* or sublux* or instability or unstable or stabili?ation).tw.
6    or/3-5
7    2 and 6
8    ((shoulder* or glenohumeral or gleno-humeral) adj3 (dislocat* or sublux* or instability or unstable or stabili?ation)).tw.
9    1 or 7 or 8
10  exp Orthopedic surgery/
11  exp Arthroscopy/
12  Suturing method/
13  (surg* or operat* or arthroscop* or repair*).tw.
14  su.fs.
15  or/10-14
16  9 and 15
17  Randomized controlled trial/
18  Clinical trial/
19  Controlled clinical trial/
20  Randomization/
21  Single blind procedure/
22  Double blind procedure/
23  Crossover procedure/
24  Placebo/
25  Prospective study/
26  ((clinical or controlled or comparative or placebo or prospective* or randomi#ed) adj3 (trial or study)).tw.
27  (random* adj7 (allocat* or allot* or assign* or basis* or divid* or order*)).tw.
28  ((singl* or doubl* or trebl* or tripl*) adj7 (blind* or mask*)).tw.
29  (cross?over* or (cross adj1 over*)).tw.
30  ((allocat* or allot* or assign* or divid*) adj3 (condition* or experiment* or intervention* or treatment* or therap* or control* or group*)).tw.
32  or/17-31
33  Case Study/ or Abstract Report/ or Letter/
34  32 not 33
35  16 and 34

Contributions of authors

All authors participated in the design and draft of the protocol. All authors approved the present version. François Desmeules is the guarantor of the protocol.

Declarations of interest

François Desmeules: none known
Janie Barry: none known
Jean-Sébastien Roy: none known
Pascal-André Vendittoli: I have no potential conflict of interest regarding the submitted work. All potential conflicts of interest are related to hip and knee surgery
Dominique M Rouleau: In the past received research and teaching funding paid to my institution from: Zimmer, Stryker, Synthes, Tornier, Arthrex, Smith and Nephew,Conmed, Linvatec and Depuy.