The possible alteration in diastolic function as a result of physiological left ventricular hypertrophy was evaluated in 28 male marathon runners (mean age 38.6 +/- 5.7 years) using Doppler echocardiography. The subjects were divided into two groups according to the intensity of training and chronometered running times: Group I (n = 14), 80-100 km weekly and marathon times under 3 hours; and Group II (n = 14), 30-50 km weekly and marathon times between 3 hours 15 min. and 3 hours 45 min. A control group (n = 14) of similarly aged sedentary individuals free of cardiovascular disease was used for comparisons. The following parameters were analyzed: (a) left ventricular mass and left ventricular mass index by means of M mode two-dimensional guided echocardiography; (b) peak velocities of E and A waves, E/A peak velocities ratio, E/A velocity-time integrals ratio, E wave/total diastolic flow velocity-time integrals and A wave/total diastolic flow velocity-time integrals (last one also called atrial filling fraction) that, together with E wave deceleration time, were obtained from pulsed Doppler recordings of transmitral flow; (c) isovolumic relaxation time measured from a continuous Doppler recording of aortic ejection flow. Left ventricular mass index was significantly greater in both groups of marathon runners with respect to controls (group I: 121.1 +/- 15.5 vs 81.8 +/- 11.1 g/m2, p < 0.01; group II: 106.5 +/- 21.1 vs 81.8 +/- 11.1 g/m2, p < 0.01), but there were no significant differences in any of the diastolic function parameters analyzed (Anova and Scheffe tests).(ABSTRACT TRUNCATED AT 250 WORDS)