Three indexes developed originally to assess left ventricular contractile performance were applied instead to the right ventricle (RV) in 11 conscious dogs: the relation between stroke work and end-diastolic volume (EDV), termed the preload recruitable stroke work (PRSW) relation; the end-systolic pressure-volume (ESPV) relation; and the maximum dP/dt (dP/dtmax)-EDV relation. The reproducibility, inotropic sensitivity, chronotropic sensitivity, and afterload sensitivity of these RV relations were compared. RV volume was determined with an ellipsoidal shell subtraction model from orthogonal dimensions measured by sonomicrometry. RV transmural pressure was measured with micromanometers. After autonomic blockade, preload was varied by repeated, transient vena caval occlusions before and during partial occlusion of the main pulmonary artery, after release of the pulmonary arterial occlusion, after calcium infusion, and over a range of heart rates induced by atrial pacing. The slope and volume-axis intercept of the PRSW relation were more reproducible (SD/mean, 7.8±33%and 6.2±4.1%, respectively) than the slope and volume-axis intercept of the ESPV relation (10.1±6.7% and 23.0±313%, bothp<0.05) or the slope and volume-axis intercept of the dP/dtmax-EDV relation (43.4±70.4% and 153.8±184.6%, bothp<0.05). The slope of the PRSW relation increased 32±17% (p<0.05) after calcium infusion, but the volume-axis intercept did not change significantly. In contrast, the slopes of the ESPV and dP/dtmax-EDV relations did not change significantly after calcium infusion, but the volume-axis intercepts decreased significantly (bothp<0.05). Despite a 71±26% increase in mean RV ejection pressure during partial occlusion of the main pulmonary artery, the slopes and volume-axis intercepts of both the PRSW and dP/dtmax-EDV relations did not change significantly, but the slope of the ESPV relation increased 45±22% (p<0.05) without significant change in the volume-axis intercept. None of the relations demonstrated significant chronotropic sensitivity. The PRSW relation is the preferred index of RV contractile performance because 1) it is the most reproducible, 2) its slope alone sensitively detects changes in contractile state, and 3) unlike the ESPV relation, it is relatively insensitive to afterload.