Skip to content
View commentaries on this research

This is a plain English summary of an original research article. The views expressed are those of the author(s) and reviewer(s) at the time of publication.

An increasingly common surgical procedure for shoulder pain, subacromial decompression, was only slightly more effective than no treatment. In the first trial of this sort, improvements in pain and function following decompression or arthroscopy (a placebo surgery) did not reach a threshold of clinical importance compared with people allocated to no treatment at all.

This NIHR-funded study involved 313 people with shoulder pain that had not responded to physiotherapy and a steroid injection. The main intervention, decompression, involved removing any bony outgrowths that could have been causing the tendons around the shoulder to be “pinched”. This was compared with an arthroscopy used for investigation only; this did not include removing any bone or tissue. Those in the third “no treatment” arm did not receive any physiotherapy or advice.

The minimal differences between surgery and no treatment call into question the value of decompression for such a wide group of patients. It may be best reserved for people with a clear mechanical obstruction such as those with definite bony spurs.

Why was this study needed?

Shoulder pain affects 14 in every hundred adults in the UK. A common cause is inflammation and pressure on some of the tendons in the shoulder as they pass through a small arch termed the subacromial space. The arch is formed by the bone above the shoulder joint and a ligament attaching to the shoulder blade.  When the swollen tendon rubs on the acromion bone leading to pain on raising the arm. It can also cause the bone to produce spurs which can further inflame the tendon in a vicious cycle.

Arthroscopic subacromial decompression opens up this space by removing bone and damaged soft tissue through keyhole surgery. In the UK in 2015/16 there were 30,669 admissions for this type of operation funded by the NHS. Rates of subacromial decompression have risen almost 8-fold from 5.2 per 100,000 population to 40.2 per 100,000 between 2001 and 2010 in England.

As it is not clear how well the procedure works, this study aimed to assess the effectiveness of the operation compared to a placebo procedure or nothing.

What did this study do?

This randomised controlled trial took place at 32 hospitals in the UK (51 surgeons). It included 313 adults with subacromial shoulder pain for at least three months that had not responded to exercise or at least one steroid injection. People with a rotator cuff tear were excluded from the study. They were randomised to either arthroscopic subacromial decompression, arthroscopy (an investigation with no bone or soft tissue removal) or no treatment. Both surgical groups received postoperative exercise therapy.

Interpretation of the results needs to take into account that a high proportion did not receive the intended intervention (23% decompression, 42% arthroscopy and 12% no treatment). The additional exercise therapy may have had a separate effect but would have been expected to increase the difference between the surgical groups and no treatment. Long and any differences between surgical waiting times may also have contributed to unfair between-group comparisons.

The main outcome measure for the study was the Oxford Shoulder Score assessed at six months, a validated patient-reported questionnaire it runs from 0 (worst) to 48 (best). A change of 4.5 points is the minimum change considered clinically important.

What did it find?

By six months:

  • There was no difference between the two surgical groups regarding pain and function (Oxford Shoulder Score). The decompression group average score improved from 25.2 to 32.7 while the arthroscopy group improved from 26.7 to 34.2 (mean difference [MD] ‑1.3, 95% confidence interval ‑3.9 to 1.3).
  • The no treatment group score improved slightly from 25.5 to 29.4 by six months. Though this improvement was less than for the decompression group (MD 2.8, 95% CI 0.5 to 5.2) and arthroscopy group (MD 4.2, 95% CI 1.8 to 6.6),
  • The differences were not considered to be clinically meaningful as they were less than 4.5 points.

By 12 months there had been further improvement in all groups. The decompression group scored 38.2, arthroscopy scored 38.4, and the score for those after no treatment had increased to 34.3. None of the differences between groups reached the threshold of clinical importance.

What does current guidance say on this issue?

The Royal College of Surgeons, Chartered Society of Physiotherapists and British Orthopaedic Association 2014 commissioning guideline recommends rest, simple pain killers and then physiotherapy for subacromial shoulder pain. A single injection of steroid can be tried, with a second injection after six weeks if there is a benefit. If there is still pain and disability, they recommend referral for surgical decompression.

The surgical recommendation is now being reconsidered by the British Orthopaedic Association and the British Elbow and Shoulder Society following the publication of this study.

What are the implications?

This study calls into question the benefits of surgical decompression. The small benefit of arthroscopy could have been a “placebo effect” or perhaps due to the post-operative physiotherapy this group received.

It may be that this, or a modified procedure, is effective for people with specific features such as bony spurs or there may be other subgroups of patients who could be further researched. This uncertainty could be discussed with patients alongside their non-surgical options.

These results might discourage surgeons from offering this procedure and dissuade patients from undergoing it. It is likely that the orthopaedic surgeons will consider this evidence and collectively review criteria for patient selection and any speciality guidance offered.

Citation and Funding

Beard DJ, Rees JL, Cook JL, et al. Arthroscopic subacromial decompression for subacromial

shoulder pain (CSAW): a multicentre, pragmatic, parallel group, placebo-controlled, three-group, randomised surgical trial. Lancet. 2017. [Epub ahead of print].

This project was funded by the National Institute for Health Research Biomedical Research Centre, Arthritis Research UK (Clinical Studies Grant 19707) and the Royal College of Surgeons (England).


BOA/BESS. Statement in response to recent studies regarding subacromial decompression. British Orthopaedic Association and British Elbow and Shoulder Society: London; 2017.

NHS Digital. Hospital Admitted Patient Care Activity 2015/16: Procedures and Interventions. 2016.

Judge A, Murphy RJ, Maxwell R, et al. Temporal trends and geographical variation in the use of subacromial decompression and rotator cuff repair of the shoulder in England. Bone Joint J. 2014;96-B(1):70-4.

Produced by the University of Southampton and Bazian on behalf of NIHR through the NIHR Dissemination Centre

  • Share via:
  • Print article
Back to top