BACKGROUND: Evidence for the role of right ventricular (RV) function is emerging in patients with heart failure of different etiologies. Studies conducted in dilated cardiomyopathy (IDC) showed a high prevalence of RV dysfunction unrelated to the severity of pulmonary hypertension. The aim of the study was to investigate the role of RV dysfunction in ischemic versus nonischemic patients. METHODS: A series of 153 patients with left ventricular (LV) dysfunction (defined as a LV ejection fraction <45%) of either ischemic (n = 61, coronary artery disease [CAD] group) or nonischemic (n = 92, IDC group) origin were studied invasively. Besides routine catheterization data, RV volumes and ejection fractions were obtained angiographically. Reference data were collected in a control group of healthy subjects. RV dysfunction was defined as a RV ejection fraction <35% and ventricular concordance as a <10% difference between RV and LV ejection fraction. The LV/RV end-diastolic volume ratio was calculated to assess the relative dilatation of the ventricular chambers. Hemodynamic and angiographic data were compared in the 2 groups by univariate and multivariate logistic regression analysis. RESULTS: Patients with IDC and CAD had comparable LV ejection fractions (29% +/- 3% vs 31% +/- 8%, P not significant) and mean pulmonary pressures (27 +/- 12 mm Hg vs 26 +/- 11 mm Hg, P not significant); the LV/RV end-diastolic volume ratio was identical in the 2 groups (1.26 +/- 0.4 vs 1.24 +/- 0.4, P not significant). RV ejection fraction was significantly lower in IDC compared with CAD (33% +/- 10 % vs 46% +/- 11%, P <.0001), with a prevalence of RV dysfunction in the IDC group of 65% compared with 16% in the CAD group (P <.0001); similarly, the prevalence of ejection fraction concordance was 74% versus 33%, respectively (P <.0001). At multivariate analysis, a low RV ejection fraction was a powerful independent predictor of IDC compared with CAD (odds ratio 0.91, 95% confidence interval 0.87-0.94, P <.0001). RV dysfunction had a positive predictive value of 75% and a negative predictive value of 78% for the diagnosis of IDC; for ventricular concordance, these values were 81% and 69%, respectively. The correlation between mean pulmonary artery pressure and RV ejection fraction was weaker in the IDC group compared with the CAD group (R(2) = 0.032, P =.047 and R(2) = 0.172,P <.0001, respectively). CONCLUSION: In the presence of LV dysfunction, a reduced RV ejection fraction is a powerful marker for IDC compared with CAD, independent of age, pulmonary hypertension, LV function, and ventricular dimensions. These findings support the concept that IDC is frequently characterized by a biventricular involvement and that the presence of RV dysfunction represents a distinguishing feature of this disease.