BACKGROUND Dilated cardiomyopathy, a heart muscle disease of unknown cause, is characterized by high mortality and is a major cause of cardiac transplantation. It has become, therefore, increasingly important to identify patients at higher risk. The aim of this study was to assess which of the data obtained at the time of diagnosis are the best predictors of survival. RESULTS One hundred forty-four patients with dilated cardiomyopathy (118 men; mean age, 39 years) were assessed clinically, noninvasively, and hemodynamically. The effect of variables derived from the evaluation on outcome (death or heart transplantation) was examined. During a mean follow-up time of 4.1 years, 68 patients (47%) died and 9 (6%) underwent heart transplantation. The 1-, 2-, and 5-year transplant-free survival rate was 79, 69, and 44%, respectively. Cox multivariate regression analysis identified three variables as independent predictors of outcome: (1) pulmonary artery systolic pressure, P = .0001; (2) left ventricular ejection fraction, P = .0013; and (3) left ventricular end-diastolic dimension, P = .007. The prognostic index was constructed from regression coefficients and parameters significant in the Cox model. The minimal prognostic index in the study group was 1.4 and the maximal was 6.0 with a corresponding 1-year survival of 98 and 18%, respectively. The validity of the prognostic index was tested in the consecutive group of 81 patients, who were followed for a mean 2.3 years. The prognostic index of the poor outcome group differed significantly from that in survivors (3.7 vs 2.9, respectively, P < .01). The sensitivity and specificity of model predictions were 68 and 52%, respectively. CONCLUSIONS The severity of pulmonary hypertension and left ventricular dysfunction provides an independent insight into the prognosis of patients with dilated cardiomyopathy. The prognostic index is useful when assessing prognosis and may be helpful in the timing of heart transplantation.