https://gastroenterology.acponline.org/archives/2022/11/18/9.htm

In adults using oral anticoagulants, PPI use may be associated with lower risk for upper GI bleeding

An ACP Journal Club commentary on the meta-analysis stressed the importance of assessing patients' underlying risk for GI bleeding and using overall clinical risk to guide shared decision making in those at higher risk or with lower baseline bleeding tolerance.


A recent meta-analysis of six observational studies and one randomized controlled trial (RCT) found that proton-pump inhibitors (PPIs) reduce upper GI bleeding in patients prescribed oral anticoagulants, with the most benefit seen in those at higher bleeding risk.

The meta-analysis was published online June 6 by the American Journal of Medicine. The following commentary by Daniel J. Stein, MD, MPH, and Joseph D. Feuerstein, MD, was published in the ACP Journal Club section of the November Annals of Internal Medicine.

Upper GI bleeding is a frequent complication of oral anticoagulant and antiplatelet use. GI bleeding is the leading reason for GI-related hospitalization (>500,000 cases per year), with most due to upper GI causes. Antisecretory therapy is one of the few preventive measures available. Kurlander and colleagues' meta-analysis of mostly observational data suggests that combining anticoagulants with a PPI may reduce risk for GI bleeding.

Classification of outcomes varied and included any gastric or duodenal bleeding during ≥3 days of hospitalization and any PPI-preventable upper GI bleeding. Despite this, the beneficial effect of PPIs was consistent across the observational data, probably because most illness in these studies was caused by peptic-related conditions.

The pooled PPI effect was dominated by the large observational data sets, with most data from a study of >1.6 million participants, which allowed for more extensive confounding adjustment and subset analysis. The single RCT (>17,500 patients) found a small absolute reduction with a PPI for bleeding from gastroduodenal lesions but not for composite upper GI bleeding events. However, the overall bleeding rate in the RCT was lower than expected (0.14% vs. >1%/y in most observational studies). This highlights the importance of underlying risk, which was reduced in the RCT by design (excluding patients with previous PPI indications). The authors' conclusion that benefits of PPI cotherapy seemed “most clearcut and substantial” in patients with increased upper GI bleeding risk is based on the smaller numbers needed to treat in this group compared with low-risk groups.

In clinical practice, initiation of PPI therapy as prophylaxis against GI bleeding is supported when treating patients with anticoagulants. As the meta-analysis was based on mostly observational data, it is important to assess a patient's underlying risk for GI bleeding. In the absence of further RCT data for higher-risk patients, overall clinical risk for GI bleeding should be used to guide shared decision making and PPI prophylaxis for those with higher baseline risk or lower bleeding risk tolerance.