The independence of the slope of the left ventricular (LV) end-systolic stress/end systolic diameter (ESS/ESD) relation of preload (PL) after normalization for end-diastolic diameter (EDD), and the reliability of both the original and the normalized slope as contractility and prognostic indices, were assessed. We studied echocardiographically, hemodynamically, and with carotid pulse tracing, 25 normal subjects, 18 patients with coronary artery disease having a normal ejection fraction (EF), and 30 patients with idiopathic dilated cardiomyopathy (DCM). The arterial pressure was changed with intravenous infusion of phenylephrine. To investigate the effect of PL on the ESS/ESD slope, in 12 normals the PL was decreased with intravenous infusion of 40 mg of furosemide and increased with 1 l isotonic NaCl. The studied population was followed up for 18-61 months. The mean values of the slope after furosemide and after NaCl differed significantly but when the above values were divided by the EDD the differences were nullified. The mean values of the slope differed highly significantly between the three groups. The normalized slope increased the sensitivity in the discrimination. During the follow-up period 10 patients with DCM died. The original and the normalized slopes separated the deceased and survivors significantly (P < 0.002 for both indices) as compared to EF (P < 0.05). The ESS/ESD slope is a very sensitive contractility index and is also superior to EF as a prognostic parameter. The normalization of the slope eliminates its dependence on PL and improves the sensitivity in assessing decreased contractility, although it slightly decreases its prognostic value.