Loading...
Multiple studies have demonstrated the association between nonsteroidal anti-inflammatory drug (NSAID) use and gastrointestinal (GI) complications. Results of these studies might be difficult to apply to clinical practice because of uncontrolled confounding in observational studies or a rigid protocol in randomized trials. Now, to apply the rigor of a randomized, controlled design and reflect the real-life variability of clinical practice, researchers conducted an industry-sponsored, prospective, randomized, open-label, blinded endpoint (PROBE) trial to compare the incidence of GI complications with use of celecoxib versus a nonselective nonsteroidal anti-inflammatory drug (nsNSAID) in 8067 patients with osteoarthritis from 783 clinics in the U.S.
Patients were stratified by Helicobacter pylori infection status and randomized to receive either celecoxib or an nsNSAID of the treating physician's choice for 6 months. Patients taking aspirin were excluded. Adjustments in drug doses and the use of gastroprotective agents were allowed. The primary endpoint of clinically significant upper or lower GI complications was determined by a blinded adjudication panel.
More GI complications occurred in the nsNSAID group than in the celecoxib group (2.4% vs. 1.3%; odds ratio, 1.82; 95% confidence interval, 1.31–2.55). The vast majority of complications were occult GI bleeding (44 of 54 in the celecoxib group and 75 of 98 in the nsNSAID group). Upper GI bleeding occurred in only 2 patients — both in the nsNSAID group. Fewer moderate-to-severe abdominal symptoms were reported in the celecoxib versus nsNSAID group (2.3% vs. 3.4%; P=0.004). Frequencies of other complications were similar, including cardiovascular events. The dropout rate was approximately 35% in both groups.
Cryer B et al. GI-REASONS: A novel 6-month, prospective, randomized, open-label, blinded endpoint (PROBE) trial. Am J Gastroenterol 2013 Mar; 108:392. (http://dx.doi.org/10.1038/ajg.2012.467)
Comment
This study has two distinct messages. First, celecoxib results in fewer gastrointestinal complications than nonselective NSAIDs in clinical practice. Second, the PROBE study design provides valid information in the setting of variable treatment inherent in clinical practice. Of interest, the clinical conclusion is based on lower rather than upper GI complications, which have been more common in other studies. Possible explanations for this difference include (but are not limited to) aspects of the PROBE study design: the adjudication panel that might have reported more lower GI complications than previous trials; the high dropout rate; or the allowed use of gastroprotective agents that reduced the risk for upper GI complications in both groups.