||The aim of this study was to update previous reviews and examine recent evidence from randomised clinical trials (RCTs) of the use of moxibustion for osteoarthritis (OA). Twelve databases were searched from inception through to September 2016 with no language limits applied. Data extraction and risk-of-bias assessments were performed by two independent reviewers. A total of 19 RCTs met all inclusion criteria and were evaluated. Three RCTs compared the effects of moxibustion with those of sham moxibustion in patients with knee OA (KOA) and found favourable effects of moxibustion on pain reduction (n=305; SMD, -0.46; 95% CI: -0.86 to -0.06, P=0.02, I2=65%), including at follow-up (n=305; SMD, -0.36; 95% CI: -0.70 to -0.01, P=0.04, I2=54%). Eleven RCTs compared the effects of moxibustion with those of conventional oral drug therapies. Eight RCTs reported a total symptom score and the meta-analysis showed superior effects of moxibustion compared with drug therapies for this measure (n=691; SMD, -0.24; 95% CI: -0.78 to 0.29; P=0.37, I2=91%) and response rate (n=758 knees; RR, 1.10; 95% CI: 1.05-1.16, P <0.0001, I2=0%). Three RCTs found superior or equivalent effects of moxibustion on symptom score compared with intra-articular injection or topical drug therapy. The existing trial evidence is sufficiently convincing to suggest that moxibustion, compared with sham moxibustion and oral drugs, is effective for pain reduction and symptom management in KOA. The level of evidence is moderate, given the high risk of bias and small sample size.