Preload affects left ventricular pump function through the Frank-Starling curve by changing sarcomere length. However, common indices of preload, such as end-diastolic volume, or pressure, or stress, do not necessarily reflect sarcomere length. Ultimately, this depends on the elastic stiffness constant (k) and end-diastolic stress (sigma), which are not in simple relation with the above mentioned indices. An index of preload is proposed, (k sigma)(1/k). This index has been evaluated in 148 patients with different degrees of hemodynamic overload, and in 24 normal subjects. The preload index was found to be 1.448 +/- 0.034 in normal subjects. However, in the other patients evaluated preload index increased in mitral insufficiency (1.490 +/- 0.035), in decompensated aortic insufficiency (1.490 +/- 0.89) and in dilated cardiomyopathy (1.52 +/- 0.125), and markedly decreased in aortic stenosis (1.367 +/- 0.039) and in hypertrophic cardiomyopathy (1.41 +/- 0.034). It was always positively related to the afterload, measured as peak systolic stress. No positive relationship was found with end-diastolic volume nor pressure. Therefore, preload as a compensatory mechanism is differently recruited in response to various degrees of hemodynamic overload and parallels the afterload, in agreement with the concept of preload-afterload mismatch.