Front Cardiovasc Med. 2026 Jun 16;13:1816099. doi: 10.3389/fcvm.2026.1816099. eCollection 2026.
ABSTRACT
BACKGROUND: Non-vitamin K antagonist oral anticoagulants (NOACs) are widely prescribed for stroke prevention in atrial fibrillation (AF). Although they reduce intracranial hemorrhage compared with warfarin, gastrointestinal bleeding (GIB), particularly upper GIB (UGIB), remains a clinically significant complication. Proton pump inhibitors (PPIs) are frequently co-prescribed for gastroprotection, yet the available evidence remains largely observational and subject to methodological heterogeneity and confounding by indication.
EVIDENCE BASE AND REVIEW METHODOLOGY: We conducted a structured narrative review of nationwide cohort studies, multi-database analyses, randomized trials, and meta-analyses evaluating PPI co-therapy in NOAC-treated populations, with emphasis on clinically adjudicated UGIB outcomes and methodological considerations including confounding by indication and endpoint variability.
CLINICAL EVIDENCE AND IMPLICATIONS FOR CARDIAC RHYTHMOLOGY PRACTICE: A large Korean nationwide cohort demonstrated that PPI co-therapy in NOAC-treated AF patients was associated with reduced risks of UGIB hospitalization (weighted HR 0.825) and transfusion-requiring UGIB (weighted HR 0.798), particularly in high-risk subgroups. A meta-analysis of approximately 1.97 million oral anticoagulant users reported lower odds of total and major GIB with PPI use (OR ∼0.67-0.68). Randomized evidence suggests biological plausibility for reduction of gastroduodenal bleeding but highlights endpoint specificity. However, heterogeneity across analytic designs underscores persistent residual confounding.
CONCLUSIONS: PPI co-therapy may represent a clinically pragmatic strategy pending dedicated randomized evidence in high-risk NOAC-treated patients. A pragmatic, risk-stratified approach appears most appropriate while awaiting confirmation from dedicated randomized trials.
PMID:42382444 | PMC:PMC13314432 | DOI:10.3389/fcvm.2026.1816099