Progressive exercise compared with best-practice advice, with or without corticosteroid injection, for rotator cuff disorders: the GRASP factorial RCT

Health Technol Assess. 2021 Aug;25(48):1-158. doi: 10.3310/hta25480.

Abstract

Background: Rotator cuff-related shoulder pain is very common, but there is uncertainty regarding which modes of exercise delivery are optimal and the long-term benefits of corticosteroid injections.

Objectives: To assess the clinical effectiveness and cost-effectiveness of progressive exercise compared with best-practice physiotherapy advice, with or without corticosteroid injection, in adults with a rotator cuff disorder.

Design: This was a pragmatic multicentre superiority randomised controlled trial (with a 2 × 2 factorial design).

Setting: Twenty NHS primary care-based musculoskeletal and related physiotherapy services.

Participants: Adults aged ≥ 18 years with a new episode of rotator cuff-related shoulder pain in the previous 6 months.

Interventions: A total of 708 participants were randomised (March 2017-May 2019) by a centralised computer-generated 1 : 1 : 1 : 1 allocation ratio to one of four interventions: (1) progressive exercise (n = 174) (six or fewer physiotherapy sessions), (2) best-practice advice (n = 174) (one physiotherapy session), (3) corticosteroid injection then progressive exercise (n = 182) (six or fewer physiotherapy sessions) or (4) corticosteroid injection then best-practice advice (n = 178) (one physiotherapy session).

Main outcome measures: The primary outcome was Shoulder Pain and Disability Index (SPADI) score over 12 months. Secondary outcomes included SPADI subdomains, the EuroQol 5 Dimensions, five-level version, sleep disturbance, fear avoidance, pain self-efficacy, return to activity, Global Impression of Treatment and health resource use. Outcomes were collected by postal questionnaires at 8 weeks and at 6 and 12 months. A within-trial economic evaluation was also conducted. The primary analysis was intention to treat.

Results: Participants had a mean age of 55.5 (standard deviation 13.1) years and 49.3% were female. The mean baseline SPADI score was 54.1 (standard deviation 18.5). Follow-up rates were 91% at 8 weeks and 87% at 6 and 12 months. There was an overall improvement in SPADI score from baseline in each group over time. Over 12 months, there was no evidence of a difference in the SPADI scores between the progressive exercise intervention and the best-practice advice intervention in shoulder pain and function (adjusted mean difference between groups over 12 months -0.66, 99% confidence interval -4.52 to 3.20). There was also no difference in SPADI scores between the progressive exercise intervention and best-practice advice intervention when analysed at the 8-week and 6- and 12-month time points. Injection resulted in improvement in shoulder pain and function at 8 weeks compared with no injection (adjusted mean difference -5.64, 99% confidence interval -9.93 to -1.35), but not when analysed over 12 months (adjusted mean difference -1.11, 99% confidence interval -4.47 to 2.26), or at 6 and 12 months. There were no serious adverse events. In the base-case analysis, adding injection to best-practice advice gained 0.021 quality-adjusted life-years (p = 0.184) and increased the cost by £10 per participant (p = 0.747). Progressive exercise alone was £52 (p = 0.247) more expensive per participant than best-practice advice, and gained 0.019 QALYs (p = 0.220). At a ceiling ratio of £20,000 per quality-adjusted life-year, injection plus best-practice advice had a 54.93% probability of being the most cost-effective treatment.

Limitations: Participants and physiotherapists were not blinded to group allocation. Twelve-month follow-up may be insufficient for identifying all safety concerns.

Conclusions: Progressive exercise was not superior to a best-practice advice session with a physiotherapist. Subacromial corticosteroid injection improved shoulder pain and function, but provided only modest short-term benefit. Best-practice advice in combination with corticosteroid injection was expected to be most cost-effective, although there was substantial uncertainty.

Future work: Longer-term follow-up, including any serious adverse effects of corticosteroid injection.

Trial registration: Current Controlled Trials ISRCTN16539266 and EudraCT 2016-002991-28.

Funding: This project was funded by the National Institute for Health Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 25, No. 48. See the NIHR Journals Library website for further project information.

Keywords: ADULT; COST–BENEFIT ANALYSIS; EXERCISE THERAPY; HUMANS; INJECTIONS; PHYSICAL THERAPISTS; ROTATOR CUFF; SHOULDER PAIN.

Plain Language Summary

The rotator cuff is a group of muscles and tendons that stabilise the shoulder and allow it to move. Problems with the rotator cuff are very common. Symptoms include pain, which can affect a person’s ability to work, sleep well or perform daily tasks. It is not known which treatments work best for shoulder pain, how exactly they should be delivered and whether or not people do better if they are given a steroid injection. The GRASP (Getting it Right: Addressing Shoulder Pain) trial tested whether or not people with a rotator cuff disorder would do better after a progressive exercise programme (supervised by a physiotherapist over six appointments spread out over 16 weeks) compared with a one-off best-practice advice session with a physiotherapist. The trial also tested whether or not giving a corticosteroid injection in the shoulder before starting either regime would help people recover more. We assessed the cost of delivering these treatments to the NHS. We recruited 708 people from 20 NHS-based musculoskeletal centres in the UK. People were allocated to one of four treatment groups at random: (1) progressive exercise (six or fewer physiotherapy sessions), (2) best-practice advice (one physiotherapy session), (3) corticosteroid injection then progressive exercise (six or fewer physiotherapy sessions) or (4) corticosteroid injection then best-practice advice (one physiotherapy session). Trial participants were asked to complete a questionnaire that asked about their level of shoulder pain and their ability to perform basic daily tasks before treatment, and then again at 8 weeks and at 6 and 12 months. Participants’ shoulder pain and function improved over time in each of the four treatment groups. The GRASP trial showed that there was no difference between the best-practice advice session with a physiotherapist and the more comprehensive exercise programme. Corticosteroid injection improved people’s shoulder pain and function, but only by a small amount and in the short term. No serious side effects were observed during the 12-month follow-up period. Best-practice advice in combination with corticosteroid injection is likely to be most cost-effective to the NHS.

Associated data

  • ISRCTN/ISRCTN16539266