Previous studies using echocardiography in healthy subjects have reported conflicting data regarding the percentage of the stroke volume (SV) of the left ventricle (LV) resulting from longitudinal and radial function, respectively. Therefore, the aim was to quantify the percentage of SV explained by longitudinal atrioventricular plane displacement (AVPD) in controls, athletes, and patients with decreased LV function due to dilated cardiomyopathy (DCM). Twelve healthy subjects, 12 elite triathletes, and 12 patients with DCM and ejection fraction below 30% were examined by cine magnetic resonance imaging. AVPD and SV were measured in long- and short-axis images, respectively. The percentage of the SV explained by longitudinal function (SV(AVPD%)) was calculated as the mean epicardial area of the largest short-axis slices in end diastole multiplied by the AVPD and divided by the SV. SV was higher in athletes [140 +/- 4 ml (mean +/- SE), P = 0.009] and lower in patients (72 +/- 7 ml, P < 0.001) when compared with controls (116 +/- 6 ml). AVPD was similar in athletes (17 +/- 1 mm, P = 0.45) and lower in patients (7 +/- 1 mm, P < 0.001) when compared with controls (16 +/- 0 mm). SV(AVPD%) was similar both in athletes (57 +/- 2%, P = 0.51) and in patients (67 +/- 4%, P = 0.24) when compared with controls (60 +/- 2%). In conclusion, longitudinal AVPD is the primary contributor to LV pumping and accounts for approximately 60% of the SV. Although AVPD is less than half in patients with DCM when compared with controls and athletes, the contribution of AVPD to LV function is maintained, which can be explained by the larger short-axis area in DCM.