Background: Persistent post-surgical pain affects 10-80% of individuals after common operations, and is more common among patients with psychological factors such as depression, anxiety, or catastrophising. Methods: We conducted a systematic review and meta-analysis of randomised, controlled trials to evaluate the efficacy of perioperative psychotherapy for persistent post-surgical pain and physical impairment. Paired independent reviewers identified studies, extracted data, and assessed risk of bias. The Grading of Recommendations, Assessment, Development and Evaluation system was used to assess the quality of evidence. Results: Our search of five electronic databases, up to September 1, 2016, found 15 trials (2220 patients) that were eligible for review. For both persistent post-surgical pain and physical impairment, perioperative education was ineffective, while active psychotherapy suggested a benefit (test of interaction P=0.01 for both outcomes). Moderate quality evidence showed that active perioperative psychotherapy (cognitive-behaviour therapy, relaxation therapy, or both) significantly reduced persistent post-surgical pain [weighted mean difference (WMD) -1.06 cm on a 10 cm visual analogue scale for pain, 95% confidence interval (CI) -1.56 to -0.55 cm; risk difference (RD) for achieving no more than mild pain (≤3 cm) 14%, 95% CI 8-21%] and physical impairment [WMD -9.87% on the 0-100% Oswestry Disability Index, 95% CI -13.42 to -6.32%, RD for achieving no more than mild disability (≤20%) 21%, 95% CI 13-29%]. Conclusions: Perioperative cognitive behavioural therapy and relaxation therapy are effective for reducing persistent pain and physical impairment after surgery. Future studies should explore targeted psychotherapy for surgical patients at higher risk for poor outcome.