Objective. To estimate the cardiovascular (CV) and gastrointestinal (GI) risks of etoricoxib in the treatment of osteoarthritis (OA) compared to a placebo and other non-steroidal anti-inflammatory drugs (NSAIDs). Methods. A systematic review of randomized, controlled trials (RCTs) of etoricoxib were performed. Bayesian network meta-analysis was used over a duration of 12 weeks. The incidence of CV and GI events for a duration ≥26 weeks were also tabulated and presented using descriptive statistics. Results. From this search, 10 studies were identified. Of these, 6 and 5 RCTs that measured the CV and GI events at 12 weeks were included in meta-analysis. They showed that etoricoxib did not increase the CV events compared to the placebo or NSAIDs during the 12 week period (odds ratio [OR]=0.59 compared to celecoxib, OR=0.89 with ibuprofen, OR=0.70 with placebo, and OR=2.16 with naproxen). The risk of GI events was comparable to that of most comparators, with the exception of naproxen, which had a significantly lower risk of GI events (OR=0.18) during the 12 week period. For a duration ≥26 weeks, the incidence of CV and GI events with etoricoxib increased with increasing duration. Conclusion. Etoricoxib is an alternative short-term treatment option for OA, showing comparable CV and GI complications to other NSAIDs. Nevertheless, further studies will be needed to elucidate the long-term safety of etoricoxib in the treatment of OA. (J Rheum Dis 2017;24:293-302)