Initial intra-aortic balloon pump versus Impella support strategies and short-term outcomes in acute myocardial infarction complicated by cardiogenic shock: a MIMIC-IV cohort study

Comparative evidence to guide the initial choice of temporary mechanical circulatory support in acute myocardial infarction complicated by cardiogenic shock remains limited, particularly in real-world intensive care settings. We compared short-term outcomes associated with an initial intra-aortic balloon pump strategy versus an initial Impella strategy using the MIMIC-IV database. We conducted a retrospective cohort study using adult first intensive care unit stays in MIMIC-IV. Acute myocardial infarction and cardiogenic shock were identified using ICD-based definitions, and the complete code list is provided in Supplementary Table S1. Adults with acute myocardial infarction complicated by cardiogenic shock who received mechanical circulatory support within 24 h after first intensive care unit admission were screened. The exposure of interest was the earliest documented device within that 24-hour window. Patients whose initial device was an intra-aortic balloon pump or Impella were included in the primary analysis, whereas those with initial extracorporeal membrane oxygenation were excluded from the main comparison. The primary outcome was in-hospital mortality. The primary analysis used propensity score overlap weighting with an expanded model incorporating calendar era, pre-device clinical severity variables, PCI status, and time from intensive care unit admission to device initiation. Prespecified sensitivity analyses included doubly robust logistic regression, a weighted log-link model, a single-device cohort sensitivity analysis, exclusion of patients with pre-intensive care unit device documentation, multiple imputation, and complete-case analysis. Among 85,242 adult first intensive care unit stays, 6,678 met the acute myocardial infarction definition and 1,007 met the acute myocardial infarction plus cardiogenic shock definition. Of these, 443 patients had any mechanical circulatory support documented during the hospitalization, and 395 had mechanical circulatory support within 24 h after first intensive care unit admission. After excluding 10 patients with initial extracorporeal membrane oxygenation, the final primary cohort included 385 patients, of whom 313 received an initial intra-aortic balloon pump and 72 received an initial Impella. The median time from intensive care unit admission to device initiation was 0.80 h in the intra-aortic balloon pump group and 0.97 h in the Impella group. Crude in-hospital mortality was 29.4% in the intra-aortic balloon pump group and 58.3% in the Impella group. Initial Impella remained associated with higher in-hospital mortality in the overlap-weighted model (OR 2.89, 95% CI 1.58–5.30; P = 0.001), doubly robust model (OR 4.15, 95% CI 2.05–8.39; P < 0.001), weighted log-link model (RR 1.80, 95% CI 1.31–2.48; P < 0.001), single-device cohort sensitivity analysis (OR 3.41, 95% CI 1.74–6.68; P < 0.001), and exclusion of patients with pre-intensive care unit device documentation (OR 2.59, 95% CI 1.40–4.78; P = 0.003). Multiple-imputation analyses were directionally consistent. In this single-center critical care cohort, an initial Impella strategy was associated with higher in-hospital mortality than an initial intra-aortic balloon pump strategy among adults with acute myocardial infarction complicated by cardiogenic shock who received mechanical circulatory support within 24 h after first intensive care unit admission. These findings should be interpreted cautiously because residual confounding and treatment-selection bias cannot be excluded in this observational study.