Mechanoelectrical feedback effects of altering preload, afterload, and ventricular shortening

Electrophysiological consequences of altering ventricular load (mechanoelectrical feedback) were characterized in an isolated canine heart preparation. A computerized servo pump system controlled left ventricular volume and allowed ventricular ejection against a simulated arterial load (3-element Windkessel model). In 12 ventricles, end-diastolic volume (Ved) was held constant (end-diastolic pressure 6-12 mmHg) as arterial resistance (R) was varied (0.5-12 mmHg.s.ml-1), but afterload-dependent changes in the monophasic action potential (MAP) were not observed despite a large stroke volume effect. In contrast, when R was held constant in eight ventricles while Ved was increased from 20 to 40 ml, the plateau phase of the MAP was abbreviated, the terminal portion of phase 3 repolarization was delayed, and MAP duration measured at 20, 70, and 90% repolarization decreased (P < 0.05). In six ventricles, immediate transitions from isovolumic to ejecting mode at constant Ved did not alter MAP duration, but the magnitude of early afterdepolarizations (EADs), observed during isovolumic beats at high Ved, was reduced with resumption of ventricular ejection. As stroke volume of the initial ejecting contraction was increased by stepwise reductions of R, the magnitude of the EADs decreased progressively. Thus altering ventricular afterload does not modulate action potential duration in ventricles subjected to elevated, physiological, or even greatly reduced levels of afterload, whereas diastolic filling to high Ved does. Under conditions that lead to reduced stroke volume and high end-systolic volume, EADs are produced that are virtually abolished when ventricular ejection fraction is normalized.