Stretch-induced voltage changes in the isolated beating heart: importance of the timing of stretch and implications for stretch-activated ion channels

OBJECTIVES: It is now well recognized that myocardial stretch can cause arrhythmias due to stretch-induced depolarizations. The effects of transient stretch applied during the various phases of the cardiac action potential have not been investigated. This study (1) examined the effects of short stretch pulses and sustained stretch on the monophasic action potential (MPA) repolarization time course and diastolic potential, (2) examined the arrhythmic response to differently timed stretch pulses, and (3) tested by comparison with computer simulations whether these effects are compatible with stretch-activated channel characteristics known from patch-clamp studies. METHODS: We studied the MAP changes elicited by short transient stretch pulses applied at different times during the cardiac cycle to 8 isolated Langendorff-perfused rabbit hearts. The left ventricle (LV) was instrumented with a fluid-filled balloon, the volume of which was altered rapidly and precisely by means of a computer-controlled linear motor-driven piston. MAPs were recorded simultaneously from one right ventricular (RV) and two LV sites while short volume pulses of increasing amplitude were applied to the LV at variable delays after the last of 8 regular electrical pacing stimuli. The effect of pulsatile volume pulses applied at different phases of electrical systole and diastole was compared to the effect of sustained stretch pulses (60 s duration) of the same amplitude. The experimental results were compared with computer simulations of stretch-induced effects on the action potential to further validate the experimentally measured effects with theoretical predictions based on the Oxford Heart model with added stretch channel terms. RESULTS: Stretch pulses applied during early systole caused a brief transient repolarization during the LV MAP plateau phase, with a maximal amplitude of 24 +/- 10% of the total MAP amplitude. Stretch pulses at the end of the MAP caused a transient depolarization, with a maximal amplitude of 13 +/- 5%. These oppositely polarized stretch effects crossed over during a transitional range of repolarization (mean 65 +/- 9% of repolarization) when stretch produced neither transient repolarizations nor depolarizations. Only stretch pulses applied at a mean repolarization level of 77 +/- 5% or later led to arrhythmias, preceded by transient depolarizations. No corresponding de- or repolarizations were seen in MAPs recorded simultaneously from the unstretched RV. The effects of long pulses on the MAP waveform were nearly identical to an overlay plot of the effects of many differently timed short transient pulses. When the stretch-induced voltage changes in the MAP were plotted against the repolarization level at which they were produced, a linear relationship was found (mean correlation coefficient r = 0.97; P < 0.0001) with a reversal at approximately half the total MAP amplitude. The computer simulations of the influence of stretch-activated channels reproduced both the effects of short and sustained stretch seen in the MAP recordings. CONCLUSIONS: We demonstrated in the isolated beating heart that the electrophysiologic effects of sudden myocardial stretch depend on the timing of the stretch relative to electrical systole or diastole. These findings are in agreement with patch clamp studies on stretch-activated ion channels which showed a linear current/voltage relation with a reversal potential between -20 and -30 mV. Only stretch pulses applied at the end of the action potential or during diastole elicit ectopic beats as a result of transient depolarizations, while stretch pulses applied during phase 2 and 3 cause transient repolarizations or no effect, respectively.