Relative efficacy and tolerability of etoricoxib, celecoxib, and naproxen in the treatment of osteoarthritis A Bayesian network meta-analysis of randomized controlled trials based on patient withdrawal

This study aimed to assess the relative efficacy and tolerability of etoricoxib, celecoxib, and naproxen at recommended dosages in patients with osteoarthritis (OA). Randomized controlled trials (RCTs) examining the efficacy and tolerability of etoricoxib 30-60 mg, celecoxib 200-400 mg, and naproxen 1000 mg, based on the number of patient withdrawals among those with OA, were included in this network meta-analysis. We performed a Bayesian random-effects network meta-analysis to combine direct and indirect evidence from the RCTs. Eight RCTs, including 5,942 patients, met the inclusion criteria. The proportion of patient withdrawals due to lack of efficacy was significantly lower in the etoricoxib 30-60 mg (OR 0.21, 95 % CrI 0.12-0.38), celecoxib 200-400 mg (OR 0.29, 95 % CrI 0.18-0.47), and naproxen 1000 mg (OR 0.31, 95 % CrI 0.18-0.51) groups than in the placebo group. The number of patient withdrawals due to lack of efficacy tended to be lower in the etoricoxib 30-60 mg group than in the naproxen 1000 mg and celecoxib 200-400 mg groups, although they did not reach statistical significance (OR 0.68, 95 % CrI 0.36-1.33 and OR 0.70, 95 % CrI 0.38-1.37, respectively). Ranking probabilities based on the surface under the cumulative ranking curve (SUCRA) indicated that etoricoxib 30-60 mg had the highest probability of being the best treatment based on the number of withdrawals due to lack of efficacy (SUCRA = 0.9168) followed by celecoxib 200-400 mg (SUCRA = 0.5659), naproxen 1000 mg (SUCRA = 0.5171), and placebo (SUCRA = 0.000189). With respect to tolerability, the number of withdrawals due to adverse events was not significantly different among etoricoxib, celecoxib, naproxen, and placebo, although it tended to be lower with etoricoxib and placebo. Etoricoxib 30-60 mg, celecoxib 200-400 mg, and naproxen 1000 mg were more efficacious than placebo. However, there was no significant difference in efficacy and tolerability between the medications.