Objective: The aim of this study was to analyze the effect of radon therapy on pain in rheumatic diseases.

Methods: MEDLINE and MedKur databases were searched for the terms radon plus therapy, rheum, arthritis, and osteo. Radon therapy centers and experts in the field were contacted, proceedings hand-searched, and bibliographies checked for references of potential importance. Included were all prospective randomized controlled clinical trials that compared clinical effects of radon therapy with other interventions in patients with rheumatic diseases and studied pain intensity. Information concerning patients, interventions, results, and methodology were extracted in a standardized manner by all authors independently and summarized descriptively. Reports on pain reduction were pooled for meta-analysis.

Results: Five clinical trials with a total of 338 patients and comparing the effect on pain of radon baths (three trials) or radon speleotherapy (two trials) with control intervention in degenerative spinal disease (two trials), rheumatoid arthritis (one trial) and ankylosing spondylitis (two trials) met the inclusion criteria. In meta-analysis, the pooled data showed no difference immediately after treatment (P=0.13) but significantly better pain reduction in the radon group than the control group at 3 months (P=0.02) and 6 months (P=0.002) after treatment.

Conclusions: The existing trials suggest a positive effect of radon therapy on pain in rheumatic diseases. With respect to the potential clinical effect and given the increasing public interest in radon therapy, there is an urgent need for further randomized controlled clinical investigations with long-term follow-up.

Objective: To quantify the efficacy of a series of baths containing natural radon and carbon dioxide (1.3 kBq/l, 1.6 g carbon dioxide/l on average) versus artificial carbon dioxide baths alone in patients with rheumatoid arthritis.

Subjects: Sixty patients participating in an in-patient rehabilitation programme including a series of 15 baths were randomly assigned to two groups.

Design: Pain intensity (100 mm visual analogue scale) and functional restrictions [Keitel functional test, Arthritis Impact Measurement Scales (AIMS questionnaire)] were measured at baseline, after completion of treatment and 3 and 6 months thereafter. To investigate whether the overall value of the outcomes was the same in both groups, the overall mean was analysed by Student's t-test for independent samples.

Results: The two groups showed a similar baseline situation. After completion of treatment, relevant clinical improvements were observed in both groups, with no notable group differences. However, the follow-up revealed sustained effects in the radon arm, and a return to baseline levels in the sham arm. After 6 months, marked between-group differences were found for both end-points (pain intensity: -16.9%, 95% confidence interval -27.6 to -6.2%; AIMS score: 0.57, 95% confidence interval 0.16 to 0.98). The between-group differences were statistically significant for both overall means (pain intensity, P: = 0.04; AIMS, P: = 0.01).

Conclusion: Marked short-term improvements in both groups at the end of treatment may have masked potential specific therapeutic effects of radon baths. However, after 6 months of follow-up the effects were lasting only in patients of the radon arm. This suggests that this component of the rehabilitative intervention can induce beneficial long-term effects.