This meta-analysis was performed to determine the effect of balneotherapy on relieving pain and stiffness and improving physical function, compared to controls, among patients with knee osteoarthritis. We searched electronic databases for eligible studies published from 2004 to December 31, 2016, with language restrictions of English or Japanese. We screened publications in Medline, Embase, Cochrane library, and the Japan Medical Abstracts Society Database using two approaches, MeSH terms and free words. Studies that examined the effect of balneotherapy for treating knee osteoarthritis of a ≥2-week duration were included. Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) scores were used as the outcome measure. A total of 102 publications were assessed according to the exclusion criteria of the study; eight clinical trial studies, which comprised a total of 359 cases and 375 controls, were included in this meta-analysis. The meta-analysis analyzed improvement in WOMAC score at the final follow-up visit, which varied from 2 to 12 months post-intervention. Our meta-analysis indicates that balneotherapy was clinically effective in relieving pain and stiffness, and improving function, as assessed by WOMAC score, compared to controls. However, there was high heterogeneity (88 to 93%). It is possible that balneotherapy may reduce pain and stiffness, and improve function, in individuals with knee osteoarthritis, although the quality of current publications contributes to the heterogeneity observed in this meta-analysis.