Objective: To perform a systematic review with meta-analyses that evaluates the effectiveness of the Pilates method on the pain and functionality outcome in adults with non-specific chronic low back pain. Data sources: The search was performed in the following databases: Medline, Embase, AMED, Cinahl, Lilacs, Scielo, SportDiscus, ProQuest, Web of Science, PEDro, Academic Search Premier and the Cochrane Central Register of Controlled Trials from 1950 to 2011; the following keywords were used: 'Pilates', 'Pilates-based', 'back exercises', 'exercise therapy', 'low back pain', 'back pain' and 'backache'. Review methods: The inclusion criteria were studies that assessed the effects of the Pilates method on patients with chronic low back pain. Results: Five studies met the inclusion criteria. The total number of patients was 71 in the Pilates group and 68 in the control group. Pilates exercise did not improve functionality (standardized mean difference (SMD = -1.34; 95% confidence interval (CI) -2.80, 0.11; P = 0.07) or pain between Pilates and control groups (SMD = -1.99; 95% CI -4.35, 0.37; P = 0.10). Pilates and lumbar stabilization exercises presented no significant difference in functionality (mean difference (MD) = -0.31; 95% CI -1.02, 0.40; P = 0.39) or pain (MD = -0.31; 95% CI -1.02, 0.40; P = 0.39). Conclusion: The Pilates method did not improve functionality and pain in patients who have low back pain when compared with control and lumbar stabilization exercise groups.