Up to a quarter of adults have experienced shoulder pain over the past year, and it represents the third most common musculoskeletal problem (Bot, 2005; Luime, 2004). About half of those affected will recover completely within six months (Bruls, 2015). Pain beyond three months is associated with poorer recovery, disability, and reduced ability to work. Subacromial pain is the most common form (up to 70%) of shoulder pain, and it can impair the ability to work or do household tasks (Mitchell, 2005; Harkness, 2003; van der Windt, 2000). Most patients presenting with subacromial pain, without a history of trauma, receive a diagnosis of subacromial pain syndrome (SAPS), shoulder impingement, or rotator cuff disease. Each of these labels describe similar clinical presentations, but there is inconsistency about how they are defined and overlap between these diagnoses. Here, we use the term SAPS (see box 1 for details of its presentation). This recommendation addresses the role of surgery for adults with symptoms lasting more than three months, who approach health professionals for treatment.

Current Practice

First line treatment options for SAPS include simple analgesia such as paracetamol, non-steroidal anti-inflammatory drugs (NSAIDs), glucocorticoid injections, and exercise therapy (Whittle, 2015). Subacromial acromial decompression surgery is a second line treatment option for patients with more longstanding symptoms. Current guidelines provide inconsistent recommendations. Such surgery includes removal of the subacromial bursa (bursectomy and removal of bone from the under surface of the acromion (acromioplasty). Surgeons initially performed subacromial decompression surgery as an open procedure. It evolved to less invasive keyhole surgery: arthroscopy. Despite trials dating back to 1993 (Brox, 1993) and systematic reviews failing to demonstrate benefit from surgery, (Coghlan, 2008) the number of arthroscopies performed has risen dramatically, although there is substantial geographical variation (Vitale, 2010; Judge, 2014). There were 21 000 procedures performed in NHS hospitals in 2010, which cost approximately £50 million (Judge, 2014).

Disclaimer

The information in this guideline module is adapted from Rapid Recommendation published in the BMJ (Vandvik, 2019). For more information see:

Vandvik PO, Lähdeoja T, Ardern C, et al. Subacromial decompression surgery for adults with shoulder pain: a clinical practice guideline. BMJ 2019;364:l294

Or visit the website of MAGICapp: https://app.magicapp.org/public/guideline/nBMa0L

PICO 1

PICO 2

Description of studies and results

What are the benefits and harms of subacromial decompression surgery? The linked systematic review and meta-analysis pooled data from seven randomised controlled trials with 1014 participants diagnosed with SAPS (Lahdoaja, 2019). In general, the patients included in the trials are representative of patients with SAPS presenting to primary care centres and outpatient clinics. Participants were around 49 years (median) and had had symptoms for around two years (median). Planned evaluation of trials at lower risk of bias The panel planned to focus on evidence at lower risk of bias. Two trials included placebo surgery and were at low risk of bias (Paavola, 2018; Beard, 2018). At one year after treatment, they showed that surgery did not have meaningful benefit over placebo surgery:

• High certainty evidence for little or no effect on
  • Pain (mean difference −0.26 (95% confidence interval −0.84 to 0.33), MID 1.5)
  • Function (mean difference 2.8 (−1.4 to 6.9), MID 8.3)
  • Health related quality of life (mean difference −0.03 points (−0.11 to 0.06), MID 0.07)
• Moderate certainty evidence for little or no global perceived effect (risk ratio 1.10 (0.94 to 1.30))
• Low certainty evidence for little or no effect on return to work (risk ratio 1.05 (0.89 to 1.23)).

Similar results were seen at six months, two years, and at five year follow-up, with the latter supported by low certainty evidence due to imprecise estimates from unblinded trials (Lahdoaja, 2019)

Planned evaluation of surgery compared with exercise therapy

This analysis compared subacromial decompression surgery (including postoperative exercise therapy) with exercise therapy alone. Six trials reported such comparisons, and all were at high risk of bias due to lack of blinding. Some had imprecise estimates of effect. Compared with exercise therapy, there was no important benefit of surgery on pain, function, quality of life, global perceived effect, and return to work (Lahdoaja, 2019). About a third (32%) of all participants included in the trials continued to have more than minor symptoms (such as mild to moderate pain) at one year, irrespective of treatment. The average pain scores in the trials at two years were 1.6 to 3.0 units (0-10 scale), reflecting mild to moderate pain.

Harms

Potential harms from surgery were incompletely reported in the trials. The trials were also underpowered to detect rare events. There were around 12 more frozen shoulders per 1000 patients undergoing subacromial decompression surgery, based on the two placebo controlled trials (low certainty evidence). Because harms data from randomised trials were anticipated to be so limited, the guideline panel requested the systematic review to include observational studies designed to evaluate harms after subacromial decompression surgery.(Lahdoaja, 2019). The systematic review assessed 140 publications in full text, of which four reported results from a large prospective cohort study from the United States considered to represent best current evidence on serious harms (Gill, 2014; Neer, 1972; Paavola, 2018; Beard, 2018; Hao, 2019; Lahdeoja, 2019; Karjalainen, 2019; Diercks, 2014; Brox, 1993; Coghlan, 2008; Vitale, 2010; Judge, 2014; Shields, 2015). This registry study investigated 30-day complications resulting in readmission to hospitals after mixed arthroscopic procedures including subacromial decompression surgery from 2006 to 2013 (Brox, 2003).

The risk of serious harms after mixed shoulder arthroscopic procedures was 0.5% (95% confidence interval 0.4% to 0.7%) during years 2006-11 and 0.6% (0.5% to 0.7%) during 2011-13. Reported harms included events such as major bleeding, deep infections, serious anaesthetic complications, venous thromboembolism, and peripheral nerve injury. The indirectness caused by inclusion of mixed arthroscopic shoulder procedures in the registry study results in moderate certainty evidence for estimated harms.

Level of evidence of the literature

Use this link to see all conclusions and GRADE levels: https://app.magicapp.org/goto/guideline/nBMa0L/rec/42100

The article of Vandvik (2019) published in the BMJ is used to answer this clinical questions. is published in the BMJ (Vandvik, 2019).

Two PICOs were investigated:

PICO 1:

P: Adults with Subacromial Pain Syndrome (SAPS)

I: Subacromial decompression surgery

C: Placebo surgery

PICO 2:

P: Adults with Subacromial Pain Syndrome (SAPS)

I: Subacromial decompression surgery (+postoperatieve exercise therapy)

C: Exercise therapy

Outcome measures and minimally important differences

The international panel included patients with lived experience of shoulder pain and surgery, orthopaedic surgeons, physiotherapists, rheumatologists, general internists, general practitioners, epidemiologists, and methodologists. The panel initially decided on the scope of the recommendation and the outcomes that are most important to patients. The panel identified the following important outcomes: pain, patient global perceived effect, physical function, participation in work and recreation activities, health related quality of life, development of full-thickness rotator cuff tears, and potential harms from surgery (such as adhesive capsulitis [iatrogenic frozen shoulder], death, infection, venous thromboembolism, and anaesthesia related events). This selection was also informed by the Outcome Measures in Rheumatology (OMERACT) preliminary shoulder trial core domain outcome set (Buchbinder, 2017). To inform the recommendation the panel members requested two systematic reviews addressing the following questions:

1. What is the smallest change in pain, function and quality of life that patients with shoulder conditions such as SAPS consider important—the minimally important difference—to make surgery worthwhile? Such patient-reported outcomes measures (PROMs) were measured with a variety of instruments in the trials and are challenging to interpret.
2. What are the benefits and harms of subacromial decompression surgery in patients with SAPS, as compared to placebo and nonoperative management strategies?

Parallel teams conducted these systematic reviews (Hao, 2019; Lahdeoja, 2019) Another team updated a Cochrane systematic review synchronised with this BMJ Rapid Recommendation (Karjalainen, 2019). The panel asked the review team to explore potential subgroup effects for risk of bias in trials and different types of comparisons to surgery, such as exercise therapy. The panel used this evidence and followed BMJ Rapid Recommendations procedures for creating a trustworthy recommendation. This includes the GRADE approach.

What is the minimum difference in symptoms and function important to patients? The systematic review of minimally important differences (MIDs) identified 22 original studies of 5562 patients. They reported results for 74 MID estimates judged to be of variable and mostly low credibility (Hao, 2019). The most credible MID estimates were used to help interpret the results of the systematic review, as shown in the infographic. The panel were, due to credible estimates, confident that patients valued

• A difference in pain of at least 1.5 units as important (visual analogue scale 0-10)
• A difference in function of at least 8.3 units as important (constant score 0-100)

The panel were less confident in the difference in health related quality of life reported by patients to be important (EQ 5-D, MID 0.07 units, low credibility median estimate). The panel met by videoconference to discuss the evidence and formulate a recommendation. The panel considered the balance of benefits, harms, and burdens of surgery versus placebo surgery and nonoperative treatments, the certainty of the evidence for each outcome, typical and expected variations in patient values and preferences, as well as feasibility and acceptability (practical issues) (Alonso-Coello, 2016). Recommendations using GRADE can be strong or weak, for or against a course of action (Guyatt, 2008). The panel made the recommendation from an individual patient’s perspective assuming that all options were available and affordable to the patient. It does not take a public health, societal, or health payer perspective. Healthcare systems can adapt these recommendations by including costs and other key issues of relevance, contextualised to national and local circumstances.

Search and select (Methods)

The evidence is based on two linked systematic reviews; one on minimally important differences in pain, shoulder and quality of life in adults with shoulder pain (Hao, 2019) and another review of 7 trials on benefits and harms of subacromial decompression surgery in adults with SAPS (Lahdeoja, 2019), which also included 54 observational studies investigating harms after such surgery

Results

Two systematic reviews were included in the analysis of the literature. Important study characteristics and results are summarized in the evidence tables.