Hypoxia and hypothermia enhance spatial heterogeneities of repolarization in guinea pig hearts: analysis of spatial autocorrelation of optically recorded action potential durations.

INTRODUCTION: Regional dispersions of repolarization (DOR) are arrhythmogenic perturbations that are closely associated with reentry. However, the characteristics of DOR have not been well defined or adequately analyzed because previous algorithms did not take into account spatial heterogeneities of action potential durations (APDs). Earlier simulations proposed that pathologic conditions enhance DOR by decreasing electrical coupling between cells, thereby unmasking differences in cellular repolarization between neighboring cells. Optical mapping indicated that gradients of APD and DOR are associated with fiber structure and are largely independent of activation. We developed an approach to quantitatively characterize APD gradients and DOR to determine how they are influenced by tissue anisotropy and cell coupling during diverse arrhythmogenic insults such as hypoxia and hypothermia. METHODS AND
RESULTS: Voltage-sensitive dyes were used to map APs from 124 sites on the epicardium of Langendorff-perfused guinea pig hearts during (1) cycles of hypoxia and reoxygenation and (2) after 30 minutes of hypothermia (32 degrees to 25 degrees C). We introduce an approach to quantitate DOR by analyzing two-dimensional spatial autocorrelation of APDs along directions perpendicular and parallel to the longitudinal axis of epicardial fibers. A spatial correlation length L was derived as a statistical measure of DOR. It corresponds to the distance over which APDs had comparable values, where L is inversely related to DOR. Hypoxia (30 min) caused a negligible decrease in longitudinal thetaL (from 0.530 +/- 0.138 to 0.478 +/- 0.052 m/sec) and transverse thetaT (from 0.225 +/- 0.034 to 0.204 +/- 0.021 m/sec) conduction velocities and did not alter thetaL/thetaT or activation patterns. In paced hearts (cycle length [CL] = 300 msec), hypoxia decreased APDs (123 +/- 18.2 to 46 +/- 0.6 msec; P < 0.001) within 10 to 15 minutes and enhanced DOR, as indicated by reductions of L from 1.8 +/- 0.9 to 1.1 +/- 0.5 mm (P < 0.005). Hypothermia caused marked reductions of thetaL (0.53 +/- 0.138 to 0.298 +/- 0.104 m/sec) and thetaT (0.225 +/- 0.034 to 0.138 +/- 0.027 m/sec), increased APDs (128 +/- 4.4 to 148 +/- 14.5 msec), and reduced L from 2.0 +/- 0.3 to 1.3 +/- 0.6 mm (P < 0.05). L decreased with increased time of hypoxia and recovered upon reoxygenation. Hypoxia and hypothermia reduced L measured along the longitudinal (L(L)) and transverse (L(T)) axes of cardiac fibers while the ratio of L(L)/L(T) remained constant.
CONCLUSION: Conventional indexes of DOR (i.e., APD "range" or "standard deviation," evaluated with extracellular electrodes) did not convey the spatial inhomogeneities of repolarization revealed by L. Spatial autocorrelation analysis provides a statistically significant measurement of DOR, which can take into account intrinsic heterogeneities of APDs and fiber orientation. The data show that hypoxia and hypothermia produce reductions of L, even though they have different effects on mean APD and conduction velocity. The preservation of a constant L(L)/L(T) ratio during hypoxia and hypothermia, despite large reductions in L, is consistent with a mechanism in which reduced cell-to-cell coupling unmasks intrinsic dispersions of APD and reduces L(L) and L(T) by the same factor. Thus, the spatial autocorrelation of APDs provides a sensitive index of DOR under normal and arrhythmogenic conditions. It incorporates the anisotropic nature of the myocardium and therefore is preferable to conventional indexes of DOR.