Intra-aortic balloon pump use and in-hospital mortality in acute myocardial infarction with Killip class III or IV: a nationwide cohort study

BACKGROUND: The intra-aortic balloon pump (IABP) remains widely used in acute myocardial infarction (AMI) complicated by severe heart failure or shock, particularly where access to advanced mechanical circulatory support is limited. Contemporary nationwide evidence in AMI presenting with Killip class III or IV is scarce. We therefore evaluated IABP use during hospitalization and in-hospital outcomes in this setting. METHODS: We did a nationwide cohort study using the China Chest Pain Center Database (4089 hospitals; 2019 to 2023). Adults aged ≥ 18 years with AMI and Killip class III or IV who underwent percutaneous coronary intervention (PCI) were included. The exposure was IABP use during hospitalization versus no recorded mechanical circulatory support; the primary outcome was in-hospital all-cause mortality. Associations were estimated by stabilized inverse probability of treatment weighting based on Bayesian additive regression trees (sIPTW-BART) and weighted Cox models, supported by prespecified sensitivity analyses. RESULTS: Among 37 477 patients, 6051 received IABP. In the sIPTW-BART-weighted cohort, IABP use was associated with higher in-hospital all-cause mortality (28.1% vs. 19.1%; hazard ratio [HR] 1.48, 95% CI 1.37-1.61) and cardiovascular mortality (27.5% vs. 18.8%; HR 1.48, 1.36-1.61), and recorded in-hospital complications differed between groups. Findings were consistent across prespecified sensitivity analyses. The E-value for the lower confidence limit was 2.08, indicating the minimum strength of unmeasured confounding required to explain away the primary association. CONCLUSIONS: Among patients undergoing PCI for AMI with Killip class III or IV presentation, IABP use during hospitalization was associated with higher in-hospital all-cause and cardiovascular mortality. These findings suggest potential risk associated with routine IABP use in this population and support further prospective studies to identify patients most likely to benefit or be harmed. Residual confounding by indication cannot be excluded; these findings should therefore be interpreted as associations rather than as causal effects of IABP.