Proton pump inhibitors in myocardial infarction: safety paradox in Type 1 vs. Type 2---a multicenter retrospective cohort study

Background Proton pump inhibitors (PPIs) are commonly prescribed for gastrointestinal protection in patients with myocardial infarction (MI). However, their safety profile remains contentious, particularly across MI subtypes. While recent studies have highlighted an independent association between long-term PPI use and increased cardiovascular risk, the implications of this risk remain poorly defined in the context of MI. Type 1 MI (T1MI), characterized by coronary artery obstruction, and Type 2 MI (T2MI), characterized by ischemia without significant coronary artery disease, may exhibit distinct pathophysiological mechanisms and responses to PPI therapy. This clinical equipoise underscores the need for large-scale investigations to clarify whether PPI use differentially affects adverse outcomes in T1MI versus T2MI patients, ultimately guiding evidence-based prescribing in this high-risk population. Methods This multicenter retrospective study sourced from 82 hospitals in China from January 2010 to March 2024, involving 2,247 T2MI and 84,505 T1MI patients. Participants were categorized based on PPI use, and propensity score matching (PSM) was applied within each subtype to balance baseline covariates. The primary endpoint was net adverse clinical events (NACE), a composite of all-cause death, recurrent MI, and bleeding events. Secondary endpoints included all-cause death, cardiac death, and bleeding events. Cox proportional hazards models were used to estimate adjusted hazard ratios (aHR) and 95% confidence intervals (CI). Results In the T2MI cohort (n = 2,247; age 76.1 ± 10.1 years, 41.6% female), a total of 314 pairs were well matched after 1:1 PSM, no significant differences were observed between PPI users and non-users for NACE (aHR 0.91, 95%CI 0.77-1.07, P = 0.251), all-cause death (aHR 0.92, 95%CI 0.78-1.08, P = 0.281), cardiac death (aHR 0.95, 95%CI 0.80-1.14, P = 0.586), or bleeding events (aHR 0.30, 95%CI 0.04-2.50, P = 0.265). Conversely, in the T1MI cohort (n = 84,505; age 63.8±13.9 years, 29.0% female), after 1:1 PSM (11,214 pairs), PPI use was not associated with reduced NACE (aHR 0.97, 95%CI 0.92-1.01, P = 0.154), all-cause death (aHR 0.95, 95%CI 0.89-1.01, P = 0.123), cardiac death (aHR 0.97, 95%CI 0.90-1.05, P = 0.476) but significantly increased bleeding risk (aHR 1.12, 95%CI 1.03-1.23, P = 0.012) (Table 1). Conclusions This study provides evidence that PPI use neither improves nor worsens clinical outcomes in T2MI patients, suggesting routine use may be safe in this population. Conversely, PPI use was associated with a modestly increased bleeding risk in T1MI patients, emphasizing the need for risk-benefit assessments. Further research should be conducted to confirm these findings.