Background: Data is lacking in the literature regarding the prognostic impact of left ventricular-end diastolic pressure (LVEDP) across acute coronary syndromes (ACS). Objective: To assess LVEDP and its prognostic implications in ACS patients. Methods: Prospective, longitudinal and continuous study of 1,329 ACS patients from a single center between 2004 and 2006. Diastolic function was determined by LVEDP. Population was divided in two groups: A - LVEDP < 26.5 mmHg (n = 449); group B - LVEDP >= 26.5 mmHg (n = 226). Results: There were no significant differences between groups with respect to risk factors for cardiovascular disease, medical history and medical therapy during admission. In group A, patients with non-ST elevation ACS were more frequent, as well as normal coronary angiograms. In-hospital mortality was similar between groups, but one-year survival was higher in group A patients (96.9 vs 91.2%, log rank p = 0.002). On a multivariate Cox regression model, a LVEDP >= 26.5 mmHg (HR 2.45, 95% CI 1.05 - 5.74) remained an independent predictor for one-year mortality, when adjusted for age, LV systolic ejection fraction, ST elevation ACS, peak troponin, admission glycemia, and diuretics at 24 hours. Also, a LVEDP >= 26.5 mmHg was an independent predictor for a future readmission due to congestive HF (HR 6.65 95% CI 1.74 - 25.5). Conclusion: In our selected population, LVEDP had a significant prognostic influence. (Arq Bras Cardiol 2011; 97(2) : 100-110)