Transmitral color Doppler early diastolic flow propagation velocity (Vp) has been correlated with the left ventricular (LV) relaxation time constant τ in dilated cardiomyopathy and ischemic heart disease. The aim of this study was to investigate the independent influence of LV systolic function and geometry, and of LV relaxation, on Vp in an unselected outpatient population. We studied 30 normal subjects and 130 patients (hypertensive LV hypertrophy, aortic valve stenosis or prosthesis, hypertrophic cardiomyopathy, coronary artery disease, dilated cardiomyopathy, aortic or mitral valve regurgitation). In all, we noninvasively measured LV geometry, mass, systolic function, wall motion dyssynergy, and diastolic function (abnormal relaxation or restrictive LV Doppler filling patterns). The Vp was similar in normal subjects and in patients (51 ± 14 vs 53 ± 25 cm/s). In normal subjects, the determinants of Vp at multiple regression analysis were isovolumic relaxation time, 2-dimensional cardiac index, and mitral E-wave velocity-time integral. In all, the main determinants were LV ejection fraction, percent of segmental wall dyssynergy, and isovolumic relaxation time and age. The Vp was highest in hypertrophic (75 ± 25 cm/s, p <0.05 vs normal subjects) and lowest in dilated (35 ± 13 cm/s, p = NS) cardiomyopathy. During multivariate analysis of variance, percent of wall dyssynergy (but not diffuse LV hypokinesia) independently reduced Vp (p = 0.02). The latter was not influenced by the LV filling pattern. Thus, in an unselected clinical population, prolonged relaxation per se does not influence Vp if LV systolic dyfunction and/or wall dyssynergy is absent—the latter factors are important independent determinants of Vp, which is determined by multiple factors.