BACKGROUND: Left ventricular end-diastolic pressure (LVEDP) is frequently measured during primary percutaneous coronary intervention (PCI). However, little is known of this measurement's utility in predicting outcomes or informing treatment decisions. We sought to determine the prognostic value of LVEDP measured during primary PCI for ST-segment elevation myocardial infarction (STEMI). METHODS: We studied 1,909 (33.2%) of 5,745 STEMI patients in whom LVEDP was measured during primary PCI in the APEX-AMI trial. Cox regression analysis was used to evaluate whether LVEDP was an independent predictor of mortality and the composite of death, cardiogenic shock, or congestive heart failure (CHF) at 90 days. RESULTS: The median (25th, 75th percentiles) LVEDP level was 22 mm Hg (16, 29); compared with patients with LVEDP ≤ 22 mm Hg, those with LVEDP > 22 mm Hg had higher rates of CHF (7.3% vs 3.1%, P < .001), cardiogenic shock (4.6% vs 1.7%, P < .001), and death (4.1% vs 2.2%, P = .014) at 90 days. After multivariable adjustment, LVEDP was associated with increased risk of mortality through 90 days (adjusted hazard ratio 1.22, 95% CI 1.02-1.46, per 5-mmHg increase, P = .044) and the composite of death, cardiogenic shock, or CHF within the first 2 days (adjusted hazard ratio 1.40, 95% CI 1.23-1.59, per 5-mm Hg increase, P < .001), but not from day 3 to 90 (P = .25). CONCLUSIONS: Left ventricular end-diastolic pressure measured during primary PCI for STEMI is an independent predictor of inhospital and longer term cardiovascular outcomes. Measuring LVEDP may be useful to stratify patient risk and guide postinfarct treatment.