Objective: To systematically review the effects of Tai Chi exercise as a nonpharmacological therapeutic strategy on the physical and mental health in individuals with knee osteoarthritis. Data sources: A systematic literature retrieval has been conducted in PubMed, Web of Science, EMbase, CENTRAL, OVID, CINAHL, Physiotherapy Evidence Database, Chinese Biomedical Database and China National Knowledge Infrastructure up to June 2020 to identify the relevant randomised controlled trials (RCTs). Methods: Two authors assessed independently the risk of bias using the Cochrane Collaboration's tool. Standardised mean difference (SMD) and 95% CI were calculated and data were combined using the fixed or random-effect model. The strength of evidence was rated with Grading of Recommendations Assessment, Development and Evaluation (GRADE) system. Results: A total of 16 RCTs involving 986 patients with knee osteoarthritis met the established inclusion criteria. The strength of evidence for the main outcomes was low or moderate. The systematic review illustrated the efficacy of Tai Chi exercise in treating and managing knee osteoarthritis. Patients' outcomes practising Tai Chi exercise were improved significantly, including pain (SMD = ‒0.69, 95%CI: ‒0.95 to ‒0.44, P < 0.001), stiffness (SMD = ‒0.59, 95%CI: ‒0.91 to ‒0.27, P < 0.001), physical function (SMD = ‒0.92, 95%CI: -1.16 to ‒0.69, P < 0.001), dynamic balance (SMD = 0.69, 95%CI: 0.38 to 0.99, P < 0.001), physiological and psychological health (SF-36 PCS: SMD = 0.48, 95%CI: 0.28 to 0.68, P < 0.001; SF-36 MCS: SMD = 0.26, 95%CI: 0.06 to 0.45, P = 0.01). No adverse events associated with Tai Chi exercise were reported. Conclusion: Tai Chi exercise was beneficial for ameliorating physical and mental health of patients with knee osteoarthritis and should be available as an alternative non-pharmacological therapy in rehabilitation programmes.