Risk of Recurrent Gastrointestinal Bleeding in Patients Taking Celecoxib
While treatment with selective cyclooxygenase 2 inhibitors (COX-2) has been associated with fewer clinical upper gastrointestinal adverse events than conventional NSAIDs, in patients with average risk, few data exist on patients at high risk for recurrent ulcer complications. To address this issue, Chan et. al. (N Engl J Med 347:2104, 2002) did a comparison study to assess whether treatment with celecoxib (Celebrex®) would be similar to diclofenac plus omeprazole (Prilosec®) in reducing recurrent gastrointestinal bleeding in patients with prior ulcer bleeding.
Methods: 287 arthritis patients with prior ulcer bleeds, in whom ulcer healing was confirmed endoscopically, were randomly assigned to receive either 200 mg of celecoxib twice daily plus daily placebo (n=144) or 75 mg of diclofenac twice daily plus 20 mg of omeprazole dialy (n=143) for six months. Patients were permitted to take antacids, acetaminophen, non-NSAID analgesics, DMARDs and low-dose aspirn (< 325mg daily). The primary endpoint was recurrent bleeding.
Results: The mean age of patients in each group was 60-65, and one-half of the patients had a concurrent medical condition. The disease activity scores and pain scales were similar in both groups at baseline and throughout the six-month treatment period. The probability of recurrent ulcer bleeding was 4.9% for patients treated with celecoxib and 6.4% for patients treated with the combination of diclofenac and omeprazole. In patients not taking low-dose aspirin (n=260), the probability of recurrent ulcer bleeding was 4.5% and 5.6%, respectively. 24% of patients receiving celecoxib, and 31% of patients in the diclofenac group, had renal adverse events, which included hypertension, peripheral edema, and renal failure. Among patients with renal impairment at baseline, these numbers were significantly increased, 51% in patients receiving celecoxib and 41% in patients receiving diclofenac.
Conclusion: In patients who are at high risk for ulcer complications, treatment with celecoxib is as effective as treatment with diclofenac plus omeprazole at preventing recurrent gastrointestinal bleeding. Renal adverse events are common with both treatment regimens.
Editorial Comments: Selective COX-2 inhibitors are thought to cause fewer GI ulcers and bleeding than conventional NSAIDs. However, most studies to date have targeted relatively healthy arthritis patients. This study differs in targeting older, sicker arthritis patients who have already had bleeding gastric or duodenal ulcers. In these patients, the good news is that the rate of recurrent bleeding ulcers was no higher in the celecoxib group compared to a group receiving a conventional NSAID (diclofenac) in combination with a proton pump inhibitor (omeprazole). The bad news is that the rate of recurrent bleeding ulcers in both groups was alarmingly high (5% in 6 months, or 10% in one year). Furthermore, renal complications were far more prevalent in this older, sicker cohort regardless of treatment assignment. In patients at high risk for recurrent GI bleeds, alternate treatments to coxibs and NSAIDs should be sought where possible. If not possible, patients should be watched carefully for signs and symptoms of GI bleeding, and renal insufficiency.