To review the efficacy of nonmedicinal, noninvasive therapies in knee and hip osteoarthritis.

MEDLINE was searched from 1966 to 1993 by cross-referencing the medical therapy subheadings of 'osteoarthritis' (including 'drug therapy', 'radiotherapy', 'surgery' and 'rehabilitation') with 'clinical trials', 'comparative study' or 'trial(s)'. Bibliographies of articles meeting the inclusion criteria were examined. English language articles were retrieved.

Study designs of evaluations included in the review

Controlled trials in which patients were assigned to either (1) a treatment group consisting of nonmedicinal, noninvasive therapy, or (2) a concurrent control group that received either no therapy, an alternative therapy or a placebo intervention.

Specific interventions included in the review

The interventions studied were of nonmedicinal, noninvasive therapies including: diathermy (shortwave and ultrasound), exercise (quadriceps strengthening and aerobic), topical (capsaicin), laser, acupuncture, transcutaneous nerve stimulation (TENS) and pulsed electromagnetic fields (PEMF).

Participants included in the review

Patients with osteoarthritis of hip or knee.

Outcomes assessed in the review

The outcomes assessed were: pain, which was usually defined as self-reported pain measured on an analogue scale as the change in pain from baseline; and a measure of lower limb function, most commonly defined as a change from baseline in a timed walk.

How were decisions on the relevance of primary studies made?

[A: Consensus by 2 authors]

The quality of the primary studies was assessed using the following criteria: whether assignment was randomised; clinically-relevant outcomes were reported; patients had symptoms typical of the disease studied; clinical and statistical significance was considered; the therapeutic manoeuvre was subject to contamination or cointervention; whether follow-up was completed; whether the assessor or the patient were blinded to treatment group; and whether the sample size was adequate. [A: Consensus by 2 authors]

[A: Summarised by each of two authors, compared and consensus reached]

How were the studies combined?

The studies were combined in a narrative review.

How were differences between studies investigated?

There was a discussion on the potential causes of bias in the studies.

Fourteen randomised controlled trials (RCTs) with 760 patients were included:

1 RCT (42 patients) examined shortwave pre-exercise diathermy,

1 RCT (74 patients) examined ultrasound pre-exercise diathermy,

2 RCTs (97 patients) examined quadriceps strengthening exercise,

1 RCT (222 patients) examined stretching, strengthening andaerobic exercise,

1 RCT (70 patients) examined capsaicin,

2 RCTs (58 patients) examined laser therapy,

2 RCTs (72 patients) examined acupuncture,

3 RCTs (98 patients) examined TENS, and

1 RCT (27 patients) examined PEMF.

Of the seven therapies exercise had the strongest evidence for a beneficial effect. There were no data on the efficacy of exercise specifically for hip osteoarthritis.

Available evidence suggests that diathermy provides no benefit in terms of pain reduction or improvement in function when added to an exercise programme.

Data were insufficient to evaluate the efficacy of the other five therapies. A single trial of topical treatment with capsaicin suggests it may be useful in reducing pain associated with knee osteoarthritis. Laser treatment may also be useful in reducing the pain and disability associated with knee osteoarthritis, but additional evaluation is required in view of an unusually strong intervention effect and weak placebo effect. The trials of acupuncture gave inconsistent results. TENS studies all reported superior pain control with active treatment but all exhibited a strong placebo effect. PEMF therapy was reported to reduce pain, but the study was small and further evaluation is required.

More data are required to determine the optimal exercise regime for treating knee osteoarthritis, and to evaluate the role of topical capsaicin, laser therapy, acupuncture, TENS and PEMF. No data address the role of any of these therapies in hip osteoarthritis.

Limiting the literature search to only one database and English language articles may have resulted in a failure to retrieve other relevant studies. The primary studies were subjected to a rigorous evaluation, which revealed several potential sources of bias including lack of standardisation of the exercise regime, lack of power, variable length of follow-up, unknown compliance with prescribed exercise regimes, lack of blinding of assessors, lack of an objective measure of function, and absence of comparable placebo therapy. There is an excellent discussion on these sources of bias by the authors, and the conclusions of the review are made in view of these factors.

Food and Drug Administration, grant number FD-U-000073; Agency for Health Policy and Research, grant number 1 R01AG10566-01AZ.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.