This study examined whether left ventricular function was reduced during 3 h of semi-recumbent ergometer cycling at 70% of maximal oxygen uptake while preload to the heart was maintained via saline infusion. Indices of left ventricular systolic function (end-systolic blood pressure-volume relationship, SBP/ESV) and diastolic filling (ratio of early to late peak filling velocities into the left ventricle, E:A) were calculated during recovery and compared with baseline resting data. During exercise in seven healthy, trained male subjects, an arterial catheter allowed continuous assessment of arterial pressure, stroke volume (SV), cardiac output () and an index of contractility (dP/dt max). A venous catheter assessed that central venous pressure (CVP) was maintained throughout rest, exercise and 10 min into recovery. Both systolic blood pressure and heart rate (HR) increased with the onset of exercise (from 132 ± 5 to 185 ± 19 mmHg and from 66 ± 9 to 135 ± 23 beats min-1; increases from rest to the end of the first 5 min of exercise in SBP and HR, respectively) but systolic blood pressure did not change from 30 to 180 min of exercise (∼150 mmHg), while heart rate only increased by 8 ± 9 beats min-1 (means ± s.d.; P > 0.05). The attenuated increase in HR compared with other studies suggests that the maintained CVP (∼5 mmHg) helped to prevent cardiovascular drift in this protocol. Stroke volume, and dP/dtmax were all increased with the onset of exercise (from 85 ± 8 to 120 ± 18 ml, from 5.4 ± 1.3 to 16.5 ± 3.3 l min-1 and from 14.4 ± 4 to 28 ± 8 mmHg s-1; values from rest to the end of the first 5 min of exercise for SV, and dP/dtmax, respectively) and were maintained during exercise. There was no difference in the SBP/ESV ratio from pre- to postexercise. Conversely, E:A was reduced from 2.0 ± 0.4 to 1.6 ± 0.5 postexercise (P < 0.05), returning to normal values at 24 h postexercise. This change in diastolic filling could not be fully explained (r2 = 0.39) by an increased heart rate and, with CVP unchanged, it is likely to represent some depression of intrinsic relaxation properties of left ventricular myocytes. Three hours of semi-supine cycling resulted in no evidence of a depression in left ventricular systolic function, while left ventricular diastolic function declined postexercise.