Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Effectiveness of early versus delayed rehabilitation following rotator cuff repair: Systematic review and meta-analyses

  • Bruno Mazuquin ,

    Roles Conceptualization, Formal analysis, Methodology, Writing – original draft

    b.mazuquin@mmu.ac.uk

    Affiliation Department of Health Professions, Faculty of Health, Psychology and Social Care, Manchester Metropolitan University, Manchester, United Kingdom

  • Maria Moffatt,

    Roles Methodology, Validation, Writing – review & editing

    Affiliation Department of Health Professions, Faculty of Health, Psychology and Social Care, Manchester Metropolitan University, Manchester, United Kingdom

  • Peter Gill,

    Roles Validation, Writing – review & editing

    Affiliations Department of Health Professions, Faculty of Health, Psychology and Social Care, Manchester Metropolitan University, Manchester, United Kingdom, Northern Care Alliance NHS Group, Manchester, United Kingdom

  • James Selfe,

    Roles Methodology, Writing – review & editing

    Affiliation Department of Health Professions, Faculty of Health, Psychology and Social Care, Manchester Metropolitan University, Manchester, United Kingdom

  • Jonathan Rees,

    Roles Writing – review & editing

    Affiliation Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Science, University of Oxford and NIHR Oxford Biomedical Research Centre, Oxford, United Kingdom

  • Steve Drew,

    Roles Writing – review & editing

    Affiliation University Hospitals Coventry and Warwickshire, Coventry, United Kingdom

  • Chris Littlewood

    Roles Conceptualization, Methodology, Writing – original draft, Writing – review & editing

    Affiliation Department of Health Professions, Faculty of Health, Psychology and Social Care, Manchester Metropolitan University, Manchester, United Kingdom

Effectiveness of early versus delayed rehabilitation following rotator cuff repair: Systematic review and meta-analyses

  • Bruno Mazuquin, 
  • Maria Moffatt, 
  • Peter Gill, 
  • James Selfe, 
  • Jonathan Rees, 
  • Steve Drew, 
  • Chris Littlewood
PLOS
x

Abstract

Objective

To investigate the effectiveness of early rehabilitation compared with delayed/standard rehabilitation after rotator cuff repair for pain, function, range of movement, strength, and repair integrity.

Design

Systematic review and meta-analyses.

Methods

We searched databases and included randomised controlled trials (RCTs) comparing early with delayed/standard rehabilitation for patients undergoing rotator cuff repair surgery. We assessed risk of bias of the RCTs using the Cochrane RoB 2 tool.

Results

Twenty RCTs, with 1841 patients, were included. The majority of the RCTs were of high or unclear risk of overall bias. We found substantial variations in the rehabilitation programmes, time in the sling and timing of exercise progression. We found no statistically significant differences for pain and function at any follow-up except for the outcome measure Single Assessment Numeric Evaluation at six months (MD: 6.54; 95%CI: 2.24–10.84) in favour of early rehabilitation. We found statistically significant differences in favour of early rehabilitation for shoulder flexion at six weeks (MD: 7.36; 95%CI: 2.66–12.06), three (MD: 8.45; 95%CI: 3.43–13.47) and six months (MD: 3.57; 95%CI: 0.81–6.32) and one year (MD: 1.42; 95%CI: 0.21–2.64) and similar differences for other planes of movement. In terms of repair integrity, early mobilisation does not seem to increase the risk of re-tears (OR:1.05; 95%CI: 0.64–1.75).

Discussion

Current approaches to early mobilisation, based largely on early introduction of passive movement, did not demonstrate significant differences in most clinical outcomes, although we found statistically significant differences in favour of early rehabilitation for range of movement. Importantly, there were no differences in repair integrity between early and delayed/standard rehabilitation. Most rehabilitation programmes did not consider early active movement as soon as the patient feels able. With ongoing variation in rehabilitation protocols there remains a need for large high quality RCT to inform the optimal approach to rehabilitation after rotator cuff repair surgery.

Introduction

Shoulder pain is experienced by one in four people at any one time [1] and is one of the most common musculoskeletal pain presentations [2]. Tears of the rotator cuff, the muscles and tendons that closely surround the shoulder joint, are a common cause of shoulder pain [1]. Many people with symptomatic rotator cuff tears can be successfully managed non-surgically; where this treatment proves insufficient, surgery to repair the torn rotator cuff might be offered [3]. In 2018/2019, almost 9,000 surgical repairs were undertaken in the UK National Health Service [4]. However, as the number of operations to repair the rotator cuff increases and surgical techniques advance, there are ongoing uncertainties about the optimal approach to postoperative rehabilitation, a key component of the recovery process [5].

A recent international survey of practice reported that postoperative rehabilitation has not evolved over the last two decades [6]. A generally cautious approach to standard postoperative rehabilitation still remains, and includes sling immobilisation for several weeks [7]. In 2018 a systematic review [8] was published that reported conflicting evidence in relation to early versus delayed/standard rehabilitation following rotator cuff repair. The meta-analyses suggested that early mobilisation did not provide additional clinical benefit in terms of pain or function, although recovery of range of movement of movement was more rapid. It also reported no statistically significant difference in repair integrity, i.e. the number of re-tear events was similar between groups, which is one of the historical concerns for this surgery and the justification for cautious approaches to postoperative rehabilitation. However, most of the randomised controlled trials (RCTs) included in the review were rated as presenting a high risk of bias and data for large tears were lacking.

A number of new RCTs (eight) have been published recently providing the opportunity for up to date evidence. Our aim was to investigate the effectiveness of early compared to standard or delayed rehabilitation following rotator cuff repair for pain, function, range of movement, strength and repair integrity.

Methods

We reported the systematic review according to the Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) [9]. The protocol was registered in the International prospective register of systematic reviews database (PROSPERO)—PROSPERO 2020 CRD42020209330.

Eligibility criteria

We included studies that met the following criteria:

  • Design: RCT
  • Participants: patients aged 18 years or older who had undergone surgical repair of the rotator cuff
  • Intervention and comparison: early rehabilitation compared with delayed rehabilitation (as per study definitions).
  • Outcomes: pain, function, range of movement, strength and repair integrity.

Search

We searched MEDLINE, CINAHL, Scopus, SPORTDiscus, SciELO and the Cochrane Library for relevant papers from inception to December 2020. We decided not to limit the searches by date of publication to identify other RCTs that could be relevant but were not included in the previous version. The electronic search strategies were supplemented by hand searching the reference lists and citations of the included RCTs. There was no restriction to date or language of publication. For the search strategies, we combined MeSH terms and keywords such as: rotator cuff, shoulder, shoulder joint, rehabilitation, physiotherapy, immobilisation and RCT. The detailed search strategy is available in the S1 File.

Screening

Searches results were imported to Rayyan (https://rayyan.qcri.org) where removal of duplicates and screening for eligibility was undertaken by two independent authors (BM and MM). Studies were first screened by their titles and abstracts; the full text was retrieved if further information was needed for a decision.

Data extraction

Data was extracted by one author (BM) and reviewed by a second author (PG) using a pre-established Excel template. Data extraction included author names, year of publication, country, participants characteristics, characteristics of the intervention and comparator, tools used for outcomes assessment, results for the variables of interest and public and patient involvement and engagement activities. In case of missing data, we contacted the authors via email to request additional available data.

Risk of bias and grading of evidence

The risk of bias of the included RCTs was assessed by one author (BM) and reviewed by a second author (MM) using the Cochrane risk of bias tool (RoB2) [10]. The Grades of Recommendation, Assessment, Development and Evaluation (GRADE) framework and GRADEpro GDT [11] software were used to rate the certainty of the effect. Outcomes were rated and downgraded according to the presence or absence of factors (risk of bias, inconsistency, indirectness, imprecision) affecting the quality of the body of RCTs included in each outcome.

Measurement of treatments effect

We generated meta-analyses and forest plots using RevMan 5 (The Nordic Cochrane Centre, Copenhagen, Denmark) [12]. The meta-analyses were presented according to the different outcome measures used across RCTs, for example American Shoulder and Elbow Score (ASES), and follow-up timing. Continuous data was expressed as mean difference and 95% confidence interval. For dichotomous outcomes, the odds ratio (OR) was used with 95% confidence interval. As re-tear or failure to heal, indicative of repair integrity, is regarded as an unfavourable outcome, we also calculated the Absolute Risk Increase and Number Needed to Harm (NNH) for the repair integrity data. Statistical heterogeneity was tested using the I2 test. In addition, we observed variation in the rehabilitation protocols for exercise dosage and time in the sling; therefore, we used the random effects methods for the meta-analyses. Funnel plots were not created to check for heterogeneity and bias as this is not recommended where meta-analyses have fewer than 10 RCTs, none of our meta-analyses had more than 10 RCTs [13, 14].

Results

Study selection

Initially 2142 records were found, after removing duplicates 1805 studies were screened. Of these, 22 were selected for full text review. Another two studies were excluded due to wrong interventions: Baumgarten, Osborn [15] investigated the use of pulley exercises at six weeks and Michael, König [16] explored the effect of continuous passive movement and physiotherapy versus physiotherapy alone. Twenty RCTs were eligible for inclusion; in total, 13 RCTs were included in the meta-analyses (Fig 1).

Study characteristics

Participants mean age ranged from 54 to 65.4 years. Sample sizes varied from 18 to 206 participants; only nine out of the 20 RCTs (45%) reported a sample size calculation and the majority, 16/20 (80%), recruited patients from a single hospital. The total number of patients from the 20 RCTs was 1841. Only eight of 20 RCTs included patients with large tears [3, 1723]. There was variation in the use of single or double row repairs as well as additional surgical procedures such as subacromial decompression, acromioclavicular joint excision and biceps tenodesis. Further details are available in the S2 File.

Risk of bias

Fig 2 summarises the results of the risk of bias assessment. S3 File reports the risk of bias assessment of each RCT. Most RCTs were rated as of unclear or high risk of bias for randomisation process and overall bias.

Rehabilitation protocols

The method of immobilisation after rotator cuff repair was variable. The majority of the RCTs [3, 20, 21, 2431](11/20, 55%) reported the use of a standard sling, six RCTs [19, 22, 23, 3234] used a sling with an abduction component and one RCT [35] used a standard sling for the early group and a sling with an abduction component for the delayed group. The time in the sling was also variable; eight RCTs (40%) [19, 22, 2427, 30, 33] used a six-week period for both groups. Three RCTs prescribed a sling for the early group to be used for comfort only and discarded as the patient felt appropriate [3, 20, 21]. One RCT did not prescribe a sling for those in the early group [31].

Exercise progression was similar across RCTs, starting with passive exercises, moving to active-assisted, active and then resisted exercises. However, there was variation in the timing for the exercise progression and information about frequency and intensity of exercises. All studies had a time driven protocol at all stages except Littlewood, Bateman [3], Sheps, Bouliane [20], Sheps, Silveira [21]. Littlewood, Bateman [3] used a patient-led approach using acceptable symptom response to prescribe and progress exercises, regardless of the postoperative time. Sheps, Bouliane [20] used a similar protocol to Littlewood, Bateman [3]; however, there is limited information about the protocol after six weeks postoperative. Sheps, Silveira [21] also used a patient-led approach; only pain-free active movements were permitted but exercise progression was based on the number of weeks postoperative. More details about the rehabilitation programmes are available in the S4 File.

Outcomes

Pain at rest.

Meta-analyses for pain at rest, measured with the Visual Analogue Scale, were possible at six weeks, three and six months, and one- and two-years follow-up. No statistically significant differences were found for any comparisons (Table 1). Figs 3 and 4 shows the forest plots at six weeks and three months, respectively. All forest plots for meta-analyses in Tables 14 that are not shown in the main text are available in the S5 File. The GRADE summary of findings with reasons for downgrading the certainty of effect is available in the S6 File.

thumbnail
Fig 3. Meta-analysis of pain at rest by Visual Analogue Scale at six weeks (a negative effect estimate favours early rehabilitation).

https://doi.org/10.1371/journal.pone.0252137.g003

thumbnail
Fig 4. Meta-analysis of pain at rest by Visual Analogue Scale at three months (a negative effect estimate favours early rehabilitation).

https://doi.org/10.1371/journal.pone.0252137.g004

thumbnail
Table 1. Meta-analyses of pain at rest by Visual Analogue Scale (a negative effect estimate favours early rehabilitation).

https://doi.org/10.1371/journal.pone.0252137.t001

thumbnail
Table 2. Meta-analyses of function by outcome measures (A positive effect estimate favours early rehabilitation).

https://doi.org/10.1371/journal.pone.0252137.t002

thumbnail
Table 3. Meta-analyses of range of movement by movement (A positive effect estimate favours early rehabilitation).

https://doi.org/10.1371/journal.pone.0252137.t003

Function.

There was variation in the use of composite measures of shoulder pain and function and outcome measures for shoulder function. Given the wide range of data available, we combined studies according to the outcome measure used. Meta-analyses were possible for the ASES score (three and six months, and one and two years), Constant-Murley score (CM) (three and six months, and one and two years), Single Assessment Numeric Evaluation (SANE) (three and six months), Simple Shoulder Test (three and six months, and one year) and Western Ontario Rotator Cuff Index (three and six months, and one year). There were no statistically significant differences for any outcome measures at any follow-up except for the SANE at six months in favour of early rehabilitation (Table 2). Overall, the mean differences were small in the short- and long-term follow-ups as illustrated in Figs 5 and 6 for the ASES. Other outcome measures used by RCTs included the Disabilities of the Arm, Shoulder and Hand, Oxford Shoulder Score, Shoulder Pain and Disability Index and University of California Los Angeles shoulder rating scale. The certainty of effects ranged from very low to moderate (Table 2).

thumbnail
Fig 5. Meta-analysis of function at 3 months by American Shoulder and Elbow Surgery score (A positive effect estimate favours early rehabilitation).

https://doi.org/10.1371/journal.pone.0252137.g005

thumbnail
Fig 6. Meta-analysis of function at 1 year by American Shoulder and Elbow Surgery score (A positive effect estimate favours early rehabilitation).

https://doi.org/10.1371/journal.pone.0252137.g006

Range of movement.

Meta-analyses were possible for shoulder flexion, abduction, external rotation at 90° of abduction and internal rotation at 90° of abduction at six weeks, three and six months, and one and two years. All measurements were made using a goniometer. Statistically significant differences were found for flexion at six weeks, three and six months and one year, abduction at six weeks, external rotation at three and six months and internal rotation at six weeks, three and six months (Table 3). The certainty of effects ranged from very low to high.

Strength.

Nine RCTs reported muscle strength [1922, 24, 2730]. Due to heterogeneity of testing position and data reporting, we did not pool the data into meta-analyses. Six RCTs used a hand-held dynamometer [19, 21, 22, 27, 28, 30]; other instruments included isokinetic dynamometer, tensiometer and the strength composite of the CM. The follow-up assessment ranged from six weeks to two years. Muscle strength was tested for flexion (five RCTs) [19, 21, 22, 27, 28], abduction (four RCTs) [2022, 27], internal (four RCTs) [19, 22, 28, 29] and external rotation (six RCTS) [19, 22, 2730]. All RCTs reported that both groups improved in strength; another consistent finding across all RCTs was that no statistically significant differences between group were found for any strength test at any follow-up assessed.

Repair integrity.

Meta-analyses were possible for three and six months and one-year follow-up. Only Arndt, Clavert [24] used a CT arthrography to assess the repair integrity, all other RCTs used ultrasound or MRI scan. There were no statistically significant differences between groups at any follow-up. At one year, the number needed to harm (NNH) was 42.5. We carried out a sensitivity analysis for the one-year meta-analysis by including only RCTs with overall low risk of bias. The sensitivity analysis showed a reduction in the odds ratio from 1.26 (95% CI: 0.82–1.93) (Fig 7) to 1.05 (95% CI: 0.64–1.75) (Fig 8); the NNH increased to 651. The certainty of effects ranged from low to moderate.

thumbnail
Fig 7. Meta-analysis of repair integrity (re-tear events) at 1 year (An OR lower than 1 favours early rehabilitation).

https://doi.org/10.1371/journal.pone.0252137.g007

thumbnail
Fig 8. Sensitivity meta-analysis of repair integrity (re-tear events) at 1 year (An OR lower than 1 favours early rehabilitation).

https://doi.org/10.1371/journal.pone.0252137.g008

Complications.

Overall, the number of complications were low. Nine RCTs [3, 20, 21, 23, 24, 27, 28, 31, 35] reported post-operative complications. The most common complication, reported in five RCTs [20, 21, 23, 24, 35], was related to limited shoulder range of movement. Jenssen, Lundgreen [35] reported two cases of capsulitis in the early group and none in the delayed group. Sheps, Silveira [21] found one case of frozen shoulder in the early group and two cases in the delayed group. Koh, Lim [23] found no difference in the proportion of patients with stiffness (defined as having any one of the following three: forward elevation of <120°, internal rotation lower than L3, and/or external rotation with the arm at the side of <20°) at three months postoperatively (early: 53% vs delayed: 50%). However, at 2 years follow-up the proportion of patients with shoulder stiffness in the delayed group (38%, n = 18/48) was greater than in the early group (18%, n = 7/40). Arndt, Clavert [24] and Sheps, Bouliane [20] did not report the number of patients with complications by group. Other complications reported included deep and superficial infection, loose anchors, suture pull-out, persistent shoulder pain, biceps subluxation, detached biceps tendon, deep vein thrombosis and pulmonary embolism. No differences between groups were observed for these other complications.

Patient and public involvement.

The only RCT to describe the participation of patients in the study development was Littlewood, Bateman [3]. Their patient and public involvement and engagement activities involved three meetings facilitated by the lead researcher. The group supported the co-production of patient-facing materials and development of study processes such as recruitment and informed consent. They also had a patient representative as part of their trial management group [36].

Discussion

We aimed to summarise the effectiveness of early compared to delayed/standard rehabilitation following rotator cuff repair on clinical outcomes and repair integrity. We found no statistically significant differences for pain. For function, the only statistically significant difference was for the SANE at six months in favour of early rehabilitation (MD:6.54, 95%CI:2.24, 10.84, p = 0.003). The mean difference found for outcome measures for function were small and did not reach the minimal clinically important differences (MCID) [3739]. The MCID for the SANE for patients undergoing rotator cuff repair is reported to be 16.9 [38]. The meta-analyses for range of movement showed statistically significant differences for flexion, abduction, internal and external rotation, mainly in the short-term in favour of early rehabilitation. However, similar to function, the mean differences were small and unlikely to be clinically significant [40]. In terms of repair integrity, we also found no statistically significant differences between groups at any follow-up. Compared to the 2018 systematic review [8], we included another eight RCTs published since 2016 [3, 21, 26, 28, 30, 31, 34, 35], two RCTs [20, 29] that were excluded in the 2018 review due to the inclusion of patients with traumatic tears and the full text from Mazzocca, Arciero [33], which in the 2018 review was included only as an unpublished abstract. The addition of these other RCTs did not change the overall results and conclusions. The optimal approach to postoperative rehabilitation following surgical repair of the rotator cuff remain unclear. The majority of RCTs were rated at high risk or unclear for overall bias, this was mainly due to issues with the randomisation process, but also partially related to the blinding of personnel and patients, which is not often possible in RCTs delivering exercises. Risk of bias was one of the main reasons for downgrading the certainty of the evidence.

Despite rotator cuff repair being such a common operation, we observed substantial variations in the rehabilitation protocols in relation to when patients were permitted to begin moving their shoulder and specific timelines for exercise progression. Almost all RCTs only allowed passive shoulder exercises in the first few weeks post-surgery for their early rehabilitation groups. Only Littlewood, Bateman [3], Sheps, Bouliane [20] used an individualised patient-directed approach, which facilitated a controlled and progressive introduction of active movements according to an acceptable symptoms response [41]. Sheps, Silveira [21] also used a patient-directed approach; however only pain-free active movements were allowed and exercise progression was still based on the number of weeks post-surgery. Restricting to only passive exercises in early postoperative stages may not provide sufficient load to stimulate and assist tendon healing [42] and, therefore, may not provide an optimal stimulus for tissue repair and remodelling to enable patients return to their usual activities, including leisure and work. We also found variations in the frequency of face-to-face appointments and duration of the programme.

Only four RCTs had clear distinctions between their groups regarding sling use; three RCTs prescribed the sling for comfort only and one RCT recommended patients to not use a sling at all. Although sling immobilisation has been traditionally viewed by many as important in protecting the tendon to facilitate healing, this is open to question. Stephens, Littlewood [43] interviewed patients who were part of the RaCeR RCT [3], where patients were supported to remove their sling after rotator cuff repair surgery as soon as they felt able. They reported that some patients who were in the delayed or standard rehabilitation group, and had to use a sling for four weeks, found that the sling contributed to their pain instead of relieving it. The restrictions imposed by the sling also impacted patients self-efficacy and in some cases even their self-identity. In contrast, some patients in the early rehabilitation group felt more confident and had the perception that moving their shoulder was contributing to their recovery. Another qualitative study [44] exploring patients perceptions of rehabilitation after shoulder arthroplasty reported that the sling, especially when using an abduction component, was impairing their sleep. Patients reported that they were unable to sleep because of the position, itching and temperature changes.

One of the main justifications for sling immobilisation and delayed rehabilitation following rotator cuff repair is the risk to the integrity of the repair, i.e. re-tear or failure to heal. However, as observed in our meta-analysis of repair integrity at one-year follow-up with RCTs at low risk of bias, the absolute risk reduction is very small (0.2%) and the NNH is 651. Thus, the chances of a re-tear that is caused by starting an early controlled and progressive mobilisation of the shoulder is low and no higher than delayed/standard rehabilitation; this needs to be considered in the context of the negative effects of immobilisation. We also undertook a sensitivity meta-analysis as four out of the eight RCTs included in the primary meta-analysis of repair integrity were rated at high risk of overall bias. Therefore, the sensitivity meta-analysis with only RCTs of low risk of bias provided a more reliable and robust result. There is debate that for small and medium tears early rehabilitation may be appropriate; however, for large tears the risk is less acceptable. Subgroups analyses by tear size was not possible and further recommendations for specific groups cannot be made. Only eight RCTs included patients with large tears in their sample, but data reported by tear size was not available. The RCT from Sheps, Silveira [21] (n = 206; low risk of bias), the largest RCT included in our systematic review, included patients with large tears. They found that despite patients with a large tear having a higher risk of a re-tear, this risk was not affected by the postoperative rehabilitation protocol. In contrast, using a sling for weeks may possibly cause problems such restricted range of movement and may contribute to further deconditioning of an already weakened rotator cuff muscle. As observed by Koh, Lim [23], delaying mobilisation and restricting movement may increase the risk of patients having range of movement limitations in the long-term.

Strengths and limitations

We followed strict methods for this systematic review. However, the certainty of effects were affected by the methodological quality of the body of evidence. The majority of the RCTs were considered of high or unclear overall risk of bias, had small sample sizes and their definition of early and delayed rehabilitation were not consistent. Further subgroup analyses were not possible due to the lack of data reported by tear size. Therefore, our results should be interpreted with caution [45].

Implications for clinical practice

Current approaches to early rehabilitation following rotator cuff lead to earlier restoration of range of movement and, importantly, based on current data, risk of re-tear does not seem to be increased.

We found substantial variation in the time that patients used a sling, how exercises were progressed, and a lack of information about exercise dosage. This limits the ability to make specific clinical recommendations in relation to an optimal rehabilitation programme.

Implications for research

Further large, high-quality RCTs are needed to investigate the effectiveness of early rehabilitation, particularly in relation to individual, patient-directed rehabilitation. Future RCTs must ensure that their sample size is of adequate power to allow to enable clinical recommendations with confidence.

Conclusion

Although rotator cuff repair is a common surgery, postoperative rehabilitation has not evolved for over twenty years. The addition of new RCTs in this systematic review and meta-analysis has not changed the overall conclusion from a systematic review and meta-analysis reported in 2018. Rehabilitation protocols remain variable and cautious with regards to sling use and exercise progression. Our systematic review suggests that patients treated with early rehabilitation may regain range of movement earlier and are not at a higher risk of compromising the repair integrity, which has been a concern for clinicians. A large, high-quality multi-centre RCT, including all rotator cuff tear sizes and with a more progressive and individualised approach to early rehabilitation (gradual introduction of active use of the arm as soon as able and within acceptable limits of the individual patients’ pain) is needed to advance knowledge and for conclusive recommendations on the optimal rehabilitation programme following rotator cuff repair.

References

  1. 1. Littlewood C, May S, Walters S. Epidemiology of rotator cuff tendinopathy: a systematic review. Shoulder & Elbow. 2013;5(4):256–65.
  2. 2. Jordan KP, Kadam UT, Hayward R, Porcheret M, Young C, Croft P, et al. Annual consultation prevalence of regional musculoskeletal problems in primary care: an observational study. BMC Musculoskeletal Disorders. 2010;11(1):144. pmid:20598124
  3. 3. Littlewood C, Bateman M, Butler-Walley S, Bathers S, Bromley K, Lewis M, et al. Rehabilitation following rotator cuff repair: A multi-centre pilot & feasibility randomised controlled trial (RaCeR). Clinical Rehabilitation. 2020:0269215520978859.
  4. 4. HES. Hospital Episode Statistics, Hospital Admitted Patient Care Activity 2018–19 [Internet] https://digital.nhs.uk/data-and-information/publications/statistical/hospital-admitted-patient-care-activity/2018-19#
  5. 5. Funk L. Arthroscopic shoulder surgery has progressed, has the rehabilitation? International Musculoskeletal Medicine. 2012;34(4):141–5.
  6. 6. Littlewood C, Mazuquin B, Moffatt M, Bateman M. Rehabilitation following rotator cuff repair: A survey of current practice (2020). Musculoskeletal Care. 2020(epub (ahead of print)).
  7. 7. Kane LT, Lazarus MD, Namdari S, Seitz AL, Abboud JA. Comparing expert opinion within the care team regarding postoperative rehabilitation protocol following rotator cuff repair. Journal of Shoulder and Elbow Surgery. 2020;29(9):e330–e7. pmid:32386779
  8. 8. Mazuquin BF, Wright AC, Russell S, Monga P, Selfe J, Richards J. Effectiveness of early compared with conservative rehabilitation for patients having rotator cuff repair surgery: an overview of systematic reviews. British Journal of Sports Medicine. 2018;52(2):111–21. pmid:28039127
  9. 9. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement. Journal of Clinical Epidemiology. 2009;62(10):1006–12. pmid:19631508
  10. 10. Sterne JAC, Savović J, Page MJ, Elbers RG, Blencowe NS, Boutron I, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ. 2019;366:l4898. pmid:31462531
  11. 11. GRADEpro GDT: GRADEpro Guideline Development Tool [Software]. McMaster University, 2020 (developed by Evidence Prime, Inc.). gradepro.org.
  12. 12. Review Manager (RevMan) [Computer program]. Version 5.3. Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2014.
  13. 13. Sterne JAC, Sutton AJ, Ioannidis JPA, Terrin N, Jones DR, Lau J, et al. Recommendations for examining and interpreting funnel plot asymmetry in meta-analyses of randomised controlled trials. BMJ (Clinical research ed). 2011;343(1756–1833 (Electronic)):d4002. pmid:21784880
  14. 14. Ioannidis JP, Trikalinos TA. The appropriateness of asymmetry tests for publication bias in meta-analyses: a large survey. Canadian Medical Association Journal. 2007;176(8):1091–6. pmid:17420491
  15. 15. Baumgarten KM, Osborn R, Schweinle WE Jr., Zens MJ, Helsper EA. Are Pulley Exercises Initiated 6 Weeks After Rotator Cuff Repair a Safe and Effective Rehabilitative Treatment? A Randomized Controlled Trial. Am J Sports Med. 2016;44(7):1844–51. pmid:27159310
  16. 16. Michael JW, König D, Imhoff A, A I, Martinek S, Braun S, et al. [Efficiency of a postoperative treatment after rotator cuff repair with a continuous passive motion device (CPM)]. Z Orthop Ihre Grenzgeb. 2005;143(4):438–45. pmid:16118760
  17. 17. Duzgun I, Baltaci G, Atay OA. Comparison of slow and accelerated rehabilitation protocol after arthroscopic rotator cuff repair: pain and functional activity. Acta Orthopaedica Et Traumatologica Turcica. 2011;45(1):23–33. pmid:21478659
  18. 18. Duzgun I, Baltaci G, Turgut E, Atay OA. Effects of slow and accelerated rehabilitation protocols on range of motion after arthroscopic rotator cuff repair. Acta Orthopaedica et Traumatologica Turcica. 2014;48(6):642–8. pmid:25637728
  19. 19. Lee BG, Cho NS, Rhee YG. Effect of Two Rehabilitation Protocols on Range of Motion and Healing Rates After Arthroscopic Rotator Cuff Repair: Aggressive Versus Limited Early Passive Exercises. Arthroscopy-the Journal of Arthroscopic and Related Surgery. 2012;28(1):34–42. pmid:22014477
  20. 20. Sheps DM, Bouliane M, Styles-Tripp F, Beaupre LA, Saraswat MK, Luciak-Corea C, et al. Early mobilisation following mini-open rotator cuff repair: a randomised control trial. The bone \& joint journal. 2015;97-B(9):1257–63. pmid:26330594
  21. 21. Sheps DM, Silveira A, Beaupre L, Styles-Tripp F, Balyk R, Lalani A, et al. Early Active Motion Versus Sling Immobilization After Arthroscopic Rotator Cuff Repair: A Randomized Controlled Trial. Arthroscopy: The Journal of Arthroscopic \& Related Surgery. 2019;35(3):749—60.e2. pmid:30827428
  22. 22. De Roo P, Muermans S, Maroy M, Linden P, den Daelen L. Passive mobilization after arthroscopic rotator cuff repair is not detrimental in the early postoperative period. Acta orthopaedica Belgica. 2015;81(3):485–92. pmid:26435245
  23. 23. Koh KH, Lim TK, Shon MS, Park YE, Lee SW, Yoo JC. Effect of immobilization without passive exercise after rotator cuff repair: randomized clinical trial comparing four and eight weeks of immobilization. The Journal of bone and joint surgeryAmerican volume. 2014;96(6):e44. pmid:24647511
  24. 24. Arndt J, Clavert P, Mielcarek P, Bouchaib J, Meyer N, Kempf JFa. Immediate passive motion versus immobilization after endoscopic supraspinatus tendon repair: A prospective randomized study. Orthopaedics \& Traumatology-Surgery \& Research. 2012;98(6 Suppl):S131—S8. pmid:22944392
  25. 25. Cuff DJ, Pupello DR. Prospective randomized study of arthroscopic rotator cuff repair using an early versus delayed postoperative physical therapy protocol. Journal of Shoulder and Elbow Surgery. 2012;21(11):1450–5. pmid:22554876
  26. 26. Fawzy S, Rizk Mohamed A, Sameer A, Foad A. Difference between early versus delayed postoperative physical rehabilitation protocol following arthroscopic rotator cuff repair. Egyptian Rheumatology and Rehabilitation. 2016;43(3):137–42.
  27. 27. Keener JD, Galatz LM, Stobbs G, Patton R, Yamaguchi K, Stobbs-Cucchi G, et al. Rehabilitation Following Arthroscopic Rotator Cuff Repair A Prospective Randomized Trial of Immobilization Compared with Early Motion. Journal of Bone and Joint Surgery-American Volume. 2014;96(1):11–9.
  28. 28. Kjær BH. Physiotherapy for patients with traumatic rotator cuff tear and associations within patient characteristics.: University of Southern Denmark; 2019.
  29. 29. Klintberg IH, Gunnarsson A-CC, Svantesson U, Styf J, Karlsson J. Early loading in physiotherapy treatment after full-thickness rotator cuff repair: a prospective randomized pilot-study with a two-year follow-up. Clinical Rehabilitation. 2009;23(7):622–38. pmid:19482895
  30. 30. Raschhofer R, Poulios N, Schimetta W, Kisling R, Mittermaier C. Early active rehabilitation after arthroscopic rotator cuff repair: a prospective randomized pilot study. Clinical rehabilitation. 2017;31(10):1332–9. pmid:28933605
  31. 31. Tirefort J, Schwitzguebel AJ, Collin P, Nowak A, Plomb-Holmes C, Lädermann A. Postoperative Mobilization After Superior Rotator Cuff Repair: Sling Versus No Sling: A Randomized Prospective Study. JBJS. 2019;101(6).
  32. 32. Kim YSS, Chung SW, Kim JY, Ok JHH, Park I, Oh JH. Is Early Passive Motion Exercise Necessary After Arthroscopic Rotator Cuff Repair? American Journal of Sports Medicine. 2012;40(4):815–21. pmid:22287641
  33. 33. Mazzocca AD, Arciero RA, Shea KP, Apostolakos JM, Solovyova O, Gomlinski G, et al. The Effect of Early Range of Motion on Quality of Life, Clinical Outcome, and Repair Integrity After Arthroscopic Rotator Cuff Repair. Arthroscopy: The Journal of Arthroscopic \& Related Surgery. 2017;33(6):1138–48.
  34. 34. Oyarzún DR, Poblete LA, Nieto FF, Huerta CO, Lange CC, Cifuentes AT. Efectividad de ejercicios submaximales versus protocolo convencional en el rango de movimiento glenohumeral tras reparación artroscópica del supraespinoso. Revista Chilena de Ortopedia y Traumatología. 2017;58(01):013–20.
  35. 35. Jenssen KK, Lundgreen K, Madsen JE, Kvakestad R, Pripp AH, Dimmen S. No Functional Difference Between Three and Six Weeks of Immobilization After Arthroscopic Rotator Cuff Repair: A Prospective Randomized Controlled Non-Inferiority Trial. Arthroscopy. 2018;34(10):2765–74. pmid:30195953
  36. 36. Littlewood C, Bateman M, Cooke K, Hennnings S, Cookson T, Bromley K, et al. Protocol for a multi-centre pilot and feasibility randomised controlled trial with a nested qualitative study: rehabilitation following rotator cuff repair (the RaCeR study). Trials. 2019;20(1):328. pmid:31171031
  37. 37. Buchbinder R, Ramiro S, Huang H, Gagnier JJ, Jia Y, Whittle SL. Measures of Adult Shoulder Function. Arthritis Care & Research. 2020;72(S10):250–93. pmid:33091271
  38. 38. Cvetanovich GL, Gowd AK, Liu JN, Nwachukwu BU, Cabarcas BC, Cole BJ, et al. Establishing clinically significant outcome after arthroscopic rotator cuff repair. Journal of Shoulder and Elbow Surgery. 2019;28(5):939–48. pmid:30685283
  39. 39. Kim DM, Kim TH, Kholinne E, Park JH, Shin MJ, Kim H, et al. Minimal Clinically Important Difference, Substantial Clinical Benefit, and Patient Acceptable Symptomatic State After Arthroscopic Rotator Cuff Repair. The American Journal of Sports Medicine. 2020;48(11):2650–9. pmid:32813985
  40. 40. Muir SW, Corea CL, Beaupre L. Evaluating change in clinical status: reliability and measures of agreement for the assessment of glenohumeral range of motion. North American journal of sports physical therapy 2010;5(3):98–110. pmid:21589666
  41. 41. Rees S, Mazuquin B, Richmond H, Williamson E, Bruce J. Experiences of women undergoing breast cancer surgery and physiotherapists participating in the UK PROSPER trial. Physiotherapy. 2020;107:e50–e1.
  42. 42. Sharma P, Maffulli N. Tendon injury and tendinopathy: healing and repair. The Journal of bone and joint surgery American volume. 2005;87(1):187–202. pmid:15634833
  43. 43. Stephens G, Littlewood C, Foster NE, Dikomitis L. Rehabilitation following rotator cuff repair: A nested qualitative study exploring the perceptions and experiences of participants in a randomised controlled trial. Clinical Rehabilitation. 2020:0269215520984025.
  44. 44. Horrall Stith E, Borrero L, Santurri L, Breeden L, Shemanski S. Patient Perceptions of Acute Phases of Rehabilitation Following Shoulder Arthroplasty. The Open Journal of Occupational Therapy. 2020;8(2):14.
  45. 45. Carr AJ, Cooper CD, Campbell MK, Rees JL, Moser J, Beard DJ, et al. Clinical effectiveness and cost-effectiveness of open and arthroscopic rotator cuff repair the UK Rotator Cuff Surgery (UKUFF) randomised trial. Health technology assessment (Winchester, England). 2015;19(80):1–218. pmid:26463717