Objective To evaluate the efficacy and safety of diacerein in patients with knee osteoarthritis (OA).Methods A total of 223 patients satisfying the American College of Rheumatology criteria for knee OA were chosen for this 17-week, randomized, double-dummy, diclofenac sodium-controlled trial, with diacerein dosage of 100mg/d and diclofenac sodium of 75mg/d. Efficacy and safety of both drugs were evaluated.Results Totally 106 patients in the diacerein group and 107 patients in the diclofenac group were considered qualified for the evaluation. After 12 weeks of treatment, the total effective rates of patients/physicians' overall assessment in diacerein and diclofenac groups were 65.4%/61.6% and 61.2%/61.2%, respectively (P＞0.05). The primary efficacy parameter [visual analog scale (VAS) assessment of pain on 20 metres walking] and the secondary efficacy parameters [tenderness on palpation, Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC), and 36-item Short-Form (SF-36) Health Survey] significantly improved compared with baseline in both groups ( P＜0.05). In the follow-up period, there were no obvious changes in above parameters in diacerein group. However, in diclofenac group,pain on 20 metres walking, tenderness on palpation, and WOMAC became aggravated after withdrawing the drug for 4weeks (P＜0.05). Moreover, the consumption of paracetamol was significantly lower in diacerein group than in diclofenac group during follow-up (P＜0.001). The incidences of related adverse events were 35.7% in diacerein and 45.1%in diclofenac group, respectively. Mild-to-moderate gastrointestinal disorders were the most frequent adverse events.Conclusions Diacerein is as effective as diclofenac sodium in treating patients with knee OA. Furthermore, it has better extended effect and a good safety profile. It is generally well tolerated and has no severe adverse effect.