Objective : To assess the effectiveness of conservative therapy in carpal tunnel syndrome. Data sources : A computer-aided search of MEDLINE and the Cochrane Collaboration was conducted for randomized controlled trials (RCTs) from January 1985 to May 2006. Review methods : RCTs were included if: (1) the patients, with clinically and electrophysiologically confirmed carpal tunnel syndrome, had not previously undergone surgical release, (2) the efficacy of one or more conservative treatment options was evaluated, (3) the study was designed as a randomized controlled trial. Two reviewers independently selected the studies and performed data extraction using a standardized form. In order to assess the methodological quality, the criteria list of the Cochrane Back Review Group for systematic reviews was applied. The different treatment methods were grouped (local injections, oral therapies, physical therapies, therapeutic exercises and splints). Results : Thirty-three RCTs were included in the review. The studies were analysed to determine the strength of the available evidence for the efficacy of the treatment. Our review shows that: (1) locally injected steroids produce a significant but temporary improvement, (2) vitamin B6 is ineffective, (3) steroids are better than non-steroidal anti-inflammatory drugs (NSAIDs) and diuretics, but they can produce side-effects, (4) ultrasound is effective while laser therapy shows variable results, (5) exercise therapy is not effective, (6) splints are effective, especially if used full-time. Conclusion : There is: (1) strong evidence (level 1) on efficacy of local and oral steroids; (2) moderate evidence (level 2) that vitamin B6 is ineffective and splints are effective and (3) limited or conflicting evidence (level 3) that NSAIDs, diuretics, yoga, laser and ultrasound are effective whereas exercise therapy and botulinum toxin B injection are ineffective.