Acute epicardial ECG parameters as quantitative predictors of infarct size at 1 week in the baboon.

We describe an experimental baboon model that allows quantitative prediction of myocardial necrosis measured at 1 week from acute epicardial ECG parameters recorded from a high-resolution matrix of fixed epicardial electrodes. The electrode grid overlies a circumscribed area of ultimate necrosis, produced by the occlusion of a selected diagonal branch of the left anterior descending coronary artery (LAD). This grid allowed examination of the pattern of changes in ST segment elevation (ST increases) throughout their return to control levels, and profiled changes in the distribution of electrodes recording TQ-ST segment deflections. Those points more centrally located within the area of ST increases consistently showed greater absolute values of ST increases and remained elevated longer than the more peripheral electrodes. Areas of the electrode matrix corresponding to those electrode points showing significant ST increases (2 mV above control) at each recording interval through 8 hr were fitted to the area of necrosis underlying this electrode grid. While the maximum area of ST increases (maxAst) uniformly overestimated infarct size between animals on the order of 25%, regression analysis allowed prediction of the extent of infarct from maxAst with an error of only 5%. Correlation of maxAst with the epicardial extent of infarct, total weight, and volume yielded coefficients of 0.95, 0.85, and 0.91 respectively, while mean ST increases (ST increases) showed a poorer correlation with respective coefficients of 0.49, 0.55, and 0.39. MaxAst proved to be the single best predictor of infarct size assessed at 1 week.