Rheumatoid arthritis (RA) is a chronic systemic autoimmune disease characterized by persistent inflammation of synovial joints with pain, often leading to joint destruction and disability, and despite intensive research, the cause of RA remains unknown. Balneotherapy—also called mineral baths or spa therapy—uses different types of mineral water compositions like sulphur, radon, carbon dioxin, etc. The role of balneotherapy is on debate; Sukenik wrote that the sulphur mineral water has special proprieties to rheumatologic diseases, including in the course of active inflammatory phases in RA. The aim of this review is to summarize the available evidence on the effects of balneotherapy on patients with rheumatoid arthritis. We have made a systematic search of the articles published from 1980 to 2014 on this topic in PubMed, Scopus, CRD, PEDro, Web of Science and Embase databases. We have followed the method set by the Preferred Reporting Items for Systematic reviews and Meta-Analysis (PRISMA). These that have compared balneotherapy with other therapeutic modalities or with no intervention were considered. The inclusion criteria of these papers were randomized control trial (RCT); languages: English, French, Spanish, Italian and Portuguese; evaluation of efficacy (analysis of outcomes); use of natural mineral water baths; and participants with RA. A total of eight articles documenting RCTs were found and included for full review and critical appraisal involving a total of 496 patients. The studies selected highlighted an important improvement and statistically significant in several clinical parameters, in spite of their heterogeneity between the various studies. One study emphasized an important improvement on functional capacity up to 6 months of follow-up (FU). Some of the studies (std.) reveal an improvement on morning stiffness (5 std.), number of active joints (3 std.), Ritchie index (2 std.) and activities of daily living (2 std.) up to 3 months of FU. Three studies reveal the improvement on handgrip strength up to 1 month of FU. About pain (VAS), the three studies which evaluated this parameter were inconclusive about real significant improvement. Our tables summarize the published papers about this topic. Different authors emphasize the same problems: methodologies differing from study to study, treatment modalities, outcomes and their analysis. On the one hand, it is particularly difficult to have homogeneity on this population in all the parameters (patient’s clinical heterogeneity, diverse clinical course of the disease, variety of the drugs), and on the other hand, natural mineral water composition is always unique with potential specific biological effects. This comprehensive review has revealed that there are very few published studies about the use of natural mineral water in RA. International multicentre studies, using the same methodologies, could be achieved by carrying the scientific arguments to support our clinical practice.