SARAH

In Review

Lamb SE, et al. "Exercises to improve function of the rheumatoid hand (SARAH): a randomized control trial". The Lancet. 2015. 385(9966):421-9.
PubMed • Full text • PDF

Clinical Question

In patients with rheumatoid arthritis of the hand, are tailored hand exercises in addition to usual care effective in improving overall hand function and quality of life while being cost-effective compared to usual care alone?

Bottom Line

Tailored hand exercise programs are a worthwhile, low-cost intervention as an adjunct to a range of drug regimens for patients with rheumatoid hand arthritis. A hand exercise regimen is effective in restoring and retaining hand function in rheumatoid arthritis, with associated positive effects on activities of daily living, work, and physical and emotional roles.

Major Points

Rheumatoid arthritis (RA) is a complex systemic inflammatory condition which causes symmetrically swollen and tender joints in the hands and feet. It affects millions and can lead to joint destruction, disability, and even premature death. Middle-aged females is the main demographic which is affected by RA. Smoking tobacco, diet, and disease are other risk factors which can start the onset of RA. Biological and nonbiological DMARDs are the main medications used to treat the disease while corticosteroids and NSAIDs may be used as add-on therapies. Hand exercises may also be an effective, low-cost intervention in addition to the drug regimens in order to increase hand mobility and functionality.

In the SARAH trial, patients were randomly assigned to usual care or to an exercise program. The exercises were completed on a daily basis for a minimum of 12 weeks and included 6 sessions with a physiotherapist or occupational therapist. The sessions utilized seven mobility exercises and 4 strength/endurance exercises with resistance from bands, balls, or putty. The participants tailored their exercise prescription using a modified Borg Scale and were provided with exercise booklets to demonstrate the exercise techniques. Therapists detailed the sessions in standardized log books and clinical records. The exercise group’s improvement was more than double of the usual care group.

Based on the outcomes, tailored hand exercises are shown to be effective, low-cost interventions to help patients with rheumatoid arthritis.However, there are a few criticisms within this trial. For example, clinician masking was impossible in this study and there was low ethnic diversity. Selection bias may also be a factor since there is very little data pertaining to the people who refused to participate. Finally, two of seven of the American College of Rheumatologists criteria related to global rating of disease, due to not being applicable to the intervention.

Guidelines

As of July 2016, no guidelines have been published in the U.S. that reflect the results of this trial. Furthermore, no applicable guidelines were published elsewhere.

Design

• Multicenter, investigator blinded, parallel group randomized control trial
• N = 490 participants
  • Tailored hand exercises (experimental)= 246 participants
  • Standard care = 244 participants
• Setting: Completed at 17 National Health Service Hospital trusts in England
• Enrollment: October 2009 - May 2011
• Follow-up: 12 months
• Analysis: Intention-to-treat
• Primary outcome: Overall hand function subscale of the Michigan Hand Outcome Questionnaire (MHQ)

Population

Inclusion Criteria

• Adults with rheumatoid arthritis meeting the American College of Rheumatology clinical and immunological criteria
• Participant reported active pain and hand dysfunction
• Not on a DMARD regimen or on a stable DMARD regimen (including biologicals) for 3 months or more

Exclusion Criteria

• Upper limb surgery/fracture in previous 6 months
• Pregnancy
• Waiting for upper limb surgery

Baseline Characteristics

Baseline characteristics similar between groups

• Mean age in early 60's
• Majority female (76% in both control and experimental groups)
• Majority Caucasian (98% in the control group, 97% in the experimental group)
• Majority right-hand dominant (90% in the control group, 92% in the experimental group)

Interventions

Participants were randomly assigned to exercise program or usual care

• Exercise Program: DMARD or biological drug regimen plus tailored hand exercise program
  • Exercise programs included 7 mobility exercises and 4 endurance exercises against resistance from balls, bands, or therapeutic putty performed daily at home for a minimum of 12 weeks
• Usual care: DMARD or biological drug regimen

Outcomes

Comparisons are exercise vs. usual care

Primary Outcomes

Measure of overall hand function subscale of the Michigan Hand Outcome Questionnaire (MHQ) at 12 months

Mean change from baseline: Exercise 7.93 (5.98-9.88) vs. Usual Care 3.56 (1.45-5.68) (p = 0.0028)

Secondary Outcomes

Self-reported improvements

81% vs. 63% at 12 months (p<0.0001, NNT=6)

Subgroup Analysis

• No significant difference in subgroup effects based on: time since diagnosis, baseline medication regimen
• No significant interaction between treatment and medication

Adverse Events

• 103 reported serious adverse events total, with 1 of these events appearing to be related to the provided treatments
• 2 deaths in the control group
• 3 life-threatening conditions (1 in the control group, 2 in the experimental group)
• 10 hospital admissions (3 in the control group, 7 in the experimental group)
  • 2 requiring medical intervention (1 from each group)
• 86 reported disabilities (38 in the control group, 48 in the experimental group), all of which were accounted for by rheumatoid arthritis flares
• 2 reports of transient exacerbation of arm pain (both in the experimental group)

Criticisms

• Due to methodological limitations clinician masking is impossible
• Little ethnic diversity
• Due to Data Protection Law, very little data is available for people who declined to participate, therefore selection bias cannot be ruled out
• Two of seven of the American College of Rheumatologists criteria related to global rating of disease were excluded, due to not being applicable to the intervention
• Older population of participants

Funding

Funding for this study was provided by the UK National Institute of Health Research Technology Assessment Programme (project number 07/32/05). The funder had no role in data analysis, data interpretation, or the decision to submit this article for publication.