OBJECTIVES We assessed the contribution of difference in duration of pulmonary venous and mitral flow at atrial contraction (ARd-Ad) for prognostic stratification of patients with left ventricular (LV) systolic dysfunction. BACKGROUND Although pulmonary venous flow (PVF) variables may supplement mitral flow patterns in evaluating left ventricular (LV) diastolic function, their value to the prognostic stratification of patients has not been investigated. METHODS Pulsed wave Doppler mitral and PVF velocity curves were recorded in 145 patients (mean age: 70 > years) with LV systolic dysfunction secondary to ischemic or nonischemic cardiomyopathy who were followed for 15 +/- 8 months. In 38% of patients, PVF signal was enhanced by the intravenous (IV) administration of a galactose-based echo-contrast agent. Based on E-wave deceleration time less than or equal to or >130 ms and ARd-Ad, patients were grouped into restrictive (group 1, n = 40), nonrestrictive with ARd-Ad greater than or equal to 30 ms (group 2, n = 55) and nonrestrictive with ARd-Ad <30 ms (group 3, n = 50). RESULTS During follow-up, 29 patients died from cardiac causes acid 28 were hospitalized for worsening heart failure (HF). On multivariate Cox model, ARd-Ad greater than or equal to 30 ms provided important prognostic information with regard to cardiac mortality and emerged as the single best predictor of cardiac events (cardiac mortality, hospitalization). The 24-month cardiac event-free survival was best (86.3%) for group 3; it was intermediate (37.9%) for group 2; and it was worst (22.9%) for group 1 (p < 0.0002 group 1 vs. 3; p < 0.0005 group 2 vs. 3; p < 0.0003 group 1 vs. group 2). CONCLUSIONS Assessment of ARd-Bd exhibited an independent value in the prognostic evaluation of patients with LV systolic dystfunction. Moreover, it contributed to identify patients at low, intermediate and high risk of cardiac events. (J Am Coll Cardiol 2000;36:1295-302) (C) 2000 by the American College of Cardiology.