LAUSR

The ischemic and bleeding event risk trade‐off must be managed after coronary stenting. Prophylactic or continued proton pump inhibitor (PPI) therapy is often prescribed in addition to antiplatelet therapy. Almost half of the patients undergoing stenting are either taking or prescribed a PPI. The factorial design COMPASS trial comparing rivaroxaban + low‐dose aspirin versus low‐dose aspirin alone in patients with chronic atherosclerotic disease with high‐risk features also provides a randomized evaluation of the utility and safety of long‐term pantoprazole (n = 8791) versus placebo (n = 8807). In these stable patients, cardiovascular events were similar with pantoprazole versus placebo (hazard ratio [HR] = 1.04 [95% confidence interval: 0.93–2.15], p = 0.51). There was also no meaningful difference in clinically significant upper gastrointestinal (GI) events (0.39% PPI vs. 0.44% placebo annual rates, p = 0.35). By randomizing PPI assignment, COMPASS further established the safety of long‐term PPI therapy convincingly countering previous observational reports of a wide variety of putative long‐term adverse effects (gastric atrophy, hip fractures, dementia, etc.). In the 100mg daily aspirin monotherapy arm, 0.8% of the patients on PPI had a clinically significant upper GI event versus 1.4% on placebo over 3 years (HR = 0.58 [0.35–0.95], p interaction = 0.124 technically suggesting no appreciable difference). The COGENT trial randomized (n = 3873) patients receiving DAPT (clopidogrel/aspirin) to omeprazole versus placebo. There was no appreciable difference in cardiovascular events (omeprazole [4.9%] vs. placebo [5.4%], p = 0.96), but GI events were reduced (omeprazole 1.1% vs. placebo 2.9% at 180 days) (HR = 0.34 [0.18–0.63], p < 0.001). In the all‐comers, GLOBAL LEADERS trial n = 15,839 patients undergoing coronary stenting were randomized to either dual antiplatelet therapy for 1 year (ticagrelor + aspirin if acute coronary syndrome/clopidogrel + aspirin if chronic coronary syndrome) followed by aspirin monotherapy in the second year versus ticagrelor + aspirin for 1 month followed by 23 months of ticagrelor monotherapy. Ischemic and bleeding outcomes were equivalent at 2 years. Almost 50% of the patients were discharged on PPI following routine practice. Adjusting for baseline differences, there was no significant difference in ischemic or bleeding events during the first year of therapy with or without PPI therapy. In the second year, however, patients assigned to aspirin monotherapy had a significantly higher adjusted risk of ischemic events associated with taking a PPI of 6.9% versus no PPI of 4.5% (adjusted HR [aHR] = 1.7 [1.30–2.22], p < 0.001). Among the patients assigned ticagrelor monotherapy in the second year, there was no difference in ischemic events associated with taking a PPI of 5.1% versus no PPI of 5.3% (aHR = 0.89 [0.68–1.16], p = 0.370). While not a randomized PPI versus placebo comparison, the observed association between aspirin monotherapy + PPI treatment and increased ischemic events in the