From the AGA Journals

Drug combinations found to increase upper gastrointestinal bleeding risk

View on the News

'Real world' results

Gastrointestinal toxicity is the major issue limiting nonsteroidal anti-inflammatory use. The excess annual risk of upper gastrointestinal bleeding per 1,000 patients is about 1 with low-dose aspirin, about 2 with coxibs, and about 4-6 with traditional NSAIDs (ibuprofen, naproxen). However, the risk of upper gastrointestinal bleeding increases markedly with several factors, including the use of concomitant medications.

Ideally, large randomized trials comparing NSAIDs with and without a concomitant medication would inform our assessment of risk. However, few such trials are available, so we commonly rely on observational database studies, such as that of Masclee et al. These studies have the important benefit of large sample size and "real world" results, but also have potential limitations, including reliability of data (for example, accuracy of diagnostic coding) and potential bias because of unequal distribution of confounding factors between cases and controls.

Masclee et al. report significant synergy (more than additive risk) of traditional NSAIDs with corticosteroids, SSRIs, aldosterone antagonists, and antithrombotic agents other than low-dose aspirin (although risk was increased with traditional NSAIDs plus low-dose aspirin). Low-dose aspirin was synergistic with antithrombotic agents and corticosteroids, while coxibs were synergistic with low-dose aspirin and SSRIs.

The results of Masclee et al. support current North American guidelines, which suggest use of proton pump inhibitors or misoprostol for traditional NSAID users taking concomitant medications such as antithrombotics, corticosteroids, or SSRIs, and use of PPIs for low-dose-aspirin users taking antithrombotics or taking corticosteroids if greater than or equal to 60 years old. Their results also suggest further evaluation of aldosterone antagonists is warranted as another possible risk factor.

Dr. Loren Laine is professor of medicine, department of internal medicine, Yale University, New Haven, Conn. He is on the Data Safety Monitoring Boards of Eisai, BMS, and Bayer; and is a consultant for AstraZeneca.




Combining nonsteroidal anti-inflammatory drugs with selective serotonin reuptake inhibitors increased the risk of upper gastrointestinal bleeding by up to 190% beyond the baseline risk found for NSAID monotherapy, researchers reported in the October issue of Gastroenterology.

Patients also faced excess risks of upper GI bleeding when they took corticosteroids, aldosterone antagonists, or anticoagulants together with low-dose aspirin or nonselective NSAIDs, although the effect was not seen for COX-2 inhibitors, said Dr. Gwen Masclee at Erasmus Medical Center in Rotterdam, the Netherlands and her associates.

Source: American Gastroenterological Association

The findings should help clinicians tailor treatments to minimize chances of upper gastrointestinal bleeding, particularly for elderly patients who often take multiple drugs, the investigators said (Gastroenterology 2014 [doi:10.1053/j.gastro.2014.06.007]).

The researchers analyzed 114,835 cases of upper gastrointestinal bleeding, including all gastroduodenal ulcers and hemorrhages extracted from seven electronic health record databases from the Netherlands, Italy, and Denmark. Three databases included primary care data, and four were administrative claims data, the investigators said. Cases served as their own controls, they noted.

Monotherapy with prescription nonselective NSAIDs increased the chances of an upper gastrointestinal bleed by 4.3 times, compared with not using any of the drugs studied (95% confidence interval, 4.1-4.4), the researchers said. Notably, bleeding risk from taking either nonselective NSAIDs or corticosteroids was the same, they said, adding that previous studies have yielded inconsistent findings on the topic. The incidence ratios for monotherapy with low-dose aspirin and COX-2 inhibitors were slightly lower at 3.1 (95% CI, 2.9-3.2) and 2.9 (95% CI, 2.7-3.2), respectively, they added.


Next Article: