Current issues of ACP Journal Club are published in Annals of Internal Medicine


Celecoxib combined with esomeprazole prevented recurrent ulcer bleeding in patients with previous NSAID-induced ulcer bleeding


ACP J Club. 2007 Sep-Oct;147:38. doi:10.7326/ACPJC-2007-147-2-038

Clinical Impact Ratings

GIM/FP/GP: 6 stars

Hospitalists: 5 stars

Gastroenterology: 7 stars

Source Citation

Chan FK, Wong VW, Suen BY, et al. Combination of a cyclo-oxygenase-2 inhibitor and a proton-pump inhibitor for prevention of recurrent ulcer bleeding in patients at very high risk: a double-blind, randomised trial. Lancet. 2007;369:1621-6. [PubMed ID: 17499604]



In patients with previous nonsteroidal antiinflammatory drug (NSAID)–induced ulcer bleeding, is combination treatment with celecoxib and esomeprazole more effective than celecoxib alone for preventing recurrent ulcer bleeding?


Design: Randomized controlled trial.

Allocation: Concealed.*

Blinding: Blinded (clinicians, patients, outcome assessors, {data collectors, data analysts, and safety and monitoring committee}†).*

Follow-up period: Median 13 months (range 0.4 to 13 mo).

Setting: Prince of Wales Hospital, Hong Kong, China.

Patients: 273 patients (mean age 71 y, 52% women) who were taking nonselective NSAIDs for arthritis and presented to the hospital with upper gastrointestinal (GI) bleeding. All patients discontinued NSAIDs and took proton-pump inhibitors (PPIs) for 8 weeks. Patients were enrolled after their ulcers had healed by endoscopy if they tested negative for Helicobacter pylori infection and required regular use of NSAIDs during the trial. Exclusion criteria included concomitant use of low-dose aspirin (ASA), anticoagulants, or corticosteroids before index bleeding; previous gastric or duodenal surgery except patch repair; allergy to celecoxib; or erosive esophagitis, gastric outlet obstruction, terminal illness, cancer, or renal failure (serum creatinine > 200 µmol/L). After the cardiovascular hazards of cyclooxygenase-2 inhibitors became known, patients were allowed to take prophylactic low-dose ASA during the study period.

Intervention: Celecoxib, 200 mg, plus esomeprazole, 20 mg (n = 137), or celecoxib alone (n = 136), twice daily for 12 months.

Outcomes: Recurrent ulcer bleeding (hematemesis or melena with ulcers or bleeding erosions confirmed by endoscopy, or decrease in hemoglobin ≥ 20 g/L with ulcers or erosions), global assessment of disease activity, arthritis pain, and adverse events.

Patient follow-up: 84% (intention-to-treat analysis).

Main results

At a median 13 months, the combination-treatment group had a lower cumulative incidence of recurrent ulcer bleeding than did the celecoxib-alone group (Table). Groups did not differ for global assessment of disease activity, arthritis pain, or incidence of adverse events.


Combination treatment with celecoxib and esomeprazole was more effective than celecoxib alone for preventing recurrent ulcer bleeding in patients with previous nonsteroidal antiinflammatory drug–induced ulcer bleeding.

*See Glossary.

†Information provided by author.

Source of funding: Research Grant Council of Hong Kong.

For correspondence: Professor F.K. Chan, Chinese University of Hong Kong, Hong Kong, China. E-mail

Table. Celecoxib plus esomeprazole vs celecoxib alone in patients with previous NSAID-induced ulcer bleeding at median 13 months‡

Outcome Celecoxib + esomeprazole Celecoxib alone RRR (95% CI) NNT (CI)
Recurrent ulcer bleeding 0% 8.8% 100% (69 to 100) 12 (7 to 20)

‡NSAID = nonsteroidal antiinflammatory drug; other abbreviations defined in Glossary. RRR, NNT, and CI calculated from data in article.


The highest incidence of GI complications associated with NSAID use occurs in patients who have had prior GI bleeding or prior complicated or uncomplicated ulcer; are > 75 years of age; and are taking concomitant clopidogrel, warfarin, steroids, or other NSAIDs. Combining ASA with a PPI reduces bleeding compared with placebo (1) or clopidogrel alone (0.7% vs 8.6%, P = 0.001) (2). However, neither combining diclofenac with omeprazole nor using celecoxib alone is sufficiently effective for preventing recurrent bleeding from NSAIDs (4.9% vs 6.4%, P = 0.6) (3).

The study by Chan and colleagues reinforces that switching to coxib monotherapy does not adequately prevent GI injury in high-risk patients, including those on concomitant ASA. However, the absence of recurrent bleeding in patients randomized to receive coxibs plus PPIs (even among the 15% of patients on ASA) makes this the optimal approach to prevent recurrent bleeding in patients who bleed while taking nonselective NSAIDs. Results from the subgroup taking ASA must be interpreted with caution because of small numbers.

PPI dosage is problematic. Taking PPIs once daily reduces ASA bleeds and uncomplicated NSAID ulcers but may not be sufficient to prevent bleeding. Taking PPIs twice daily affects compliance and cost. In balancing cardiovascular and GI risks, data support taking PPIs plus celecoxib twice daily to prevent bleeding in persons at high risk, substituting celecoxib with naproxen if cardiovascular risk is unacceptable, and using PPIs once daily for patients taking ASA alone. Older patients at risk who need ASA or NSAIDs plus clopidogrel or warfarin should take PPIs twice daily.

Philip O. Katz, MD
Albert Einstein Medical Center
Philadelphia, Pennsylvania, USA


1. Lai KC, Lam SK, Chu KM, et al. Lansoprazole for the prevention of recurrences of ulcer complications from long-term low-dose aspirin use. N Engl J Med. 2002;346:2033-8. [PubMed ID: 12087138]

2. Chan FK, Ching JY, Hung LC, et al. Clopidogrel versus aspirin and esomeprazole to prevent recurrent ulcer bleeding. N Engl J Med. 2005;352:238-44. [PubMed ID: 15659723]

3. Chan FK, Hung LC, Suen BY, et al. Celecoxib versus diclofenac and omeprazole in reducing the risk of recurrent ulcer bleeding in patients with arthritis. N Engl J Med.2002;347:2104-10. [PubMed ID: 12501222]