BACKGROUND Osteoarthritis (OA) is a disease that affects synovial joints, which mainly causes degeneration and destruction of hyaline cartilage. To date, no curative treatment for OA exists. The primary goals for OA therapy are to relieve pain, maintain or improve functional status, and minimize deformity. Transcutaneous electrical nerve stimulation (TENS) is a noninvasive modality in physiotherapy that is commonly used to control both acute and chronic pain arising from several conditions. A number of trials evaluating the efficacy of TENS in OA have been published. OBJECTIVE To assess the effectiveness of TENS in the treatment of knee OA. The primary outcomes of interest were those described by the Outcome Measures in Rheumatology Clinical Trials (OMERACT) 3, which included pain relief, functional status, patient global assessment, and change in joint imaging for studies of one year or longer. The secondary objective was to determine the most effective mode of TENS application in pain control. METHODS We searched MEDLINE, EMBASE, CINAHL, HEALTHSTAR, PEDro, Current Contents and the Cochrane Controlled Trial Register using the Cochrane Musculoskeletal Group search strategy for trials up to and including December 1999. We also hand-searched reference lists and consulted content experts. METHODS Two independent reviewers selected the trials that met predetermined inclusion criteria. METHODS Two independent reviewers extracted the data using standardized forms and assessed the quality of randomization, blinding and dropouts. A third reviewer was consulted to resolve any differences. For dichotomous outcomes, relative risks (RR) were calculated. For continuous data, weighted mean differences (WMD) or standardized mean difference (SMD) of the change from baseline were calculated. A fixed effects model was used unless heterogeneity of the populations existed. In this case, a random effects model was used. RESULTS Seven trials were eligible to be included in this review. Six used TENS as the active treatment while one study used acupuncture-like TENS (AL-TENS). A number of 148 and 146 patients were involved in the active TENS treatment and placebo, respectively. Three studies were cross-over studies and the others were parallel group, randomized controlled trials (RCTs). Median methodological quality of these studies was two. Pain relief from active TENS and AL-TENS treatment was significantly better than placebo treatment. Knee stiffness also improved significantly in active treatment group compared to placebo. Different modes of TENS setting (High Rate and Strong Burst Mode TENS) demonstrated a significant benefit in pain relief of the knee OA over placebo. Subgroup analyses showed a heterogeneity in the studies with methodological quality of three or more and those with repeated TENS applications. CONCLUSIONS TENS and AL-TENS are shown to be effective in pain control over placebo in this review. Heterogeneity of the included studies was observed, which might be due to the different study designs and outcomes used. More well designed studies with a standardized protocol and adequate number of participants are needed to conclude the effectiveness of TENS in the treatment of OA of the knee.