Purpose of our investigation was to ascertain whether the electrocardiographic mapping of the anterior thoracic wall can provide more precise information on the extent of an anterior myocardial infarction (MI) than the 12 conventional leads do. Thirty-seven patients were studied 1 to 72 months after an acute MI of the anterior wall. All patients underwent left heart catheterization which included selective coronary arteriography and left ventriculography, to evaluate the indication for surgery. Electromaps were obtained by means of 71 Ag-AgCl electrodes located at regular intervals on the thoracic wall (between the right midclavicular line and the left posterior axillary line). The following parameters were considered: total number of Q waves, R waves and ST elevations greater than or equal to 0.05 mV (NQ, NR, NST); the sum of Q, R and ST voltages (sigma Q, sigma R, sigma ST); the sum of Q-wave and R-wave areas (sigma aQ, sigma aR). The electrocardiographic data were correlated with the percentage of left ventricular dyssynergy (corresponding to the ratio between the length of the akinetic and/or dyskinetic portion of the left ventricular silhouette and the total enddiastolic perimeter) and with the ejection fraction obtained from the left ventricular angiograms in right anterior oblique projection. A significant but weak correlation was found only between sigma R, sigma aR and percentage of dyssynergy and between NST, sigma ST and ejection fraction. Thus the amplitude and duration values of positive activation potentials (sigma aR, sigma R) were better predictors of dyssynergy extent than the classical direct signs of necrosis (NQ, sigma Q). The poor correlation observed in our patients between ECG and angiographic data can mainly be due: a) to a lack of concordance between the dyssynergic area and the truly infarcted region; b) to the well-known limits of surface electrocardiography in defining the cardiac generator characteristics. In particular, as far as the adequacy of various ECG recording systems in determining infarct size is concerned, our study suggests that exploring a large thoracic area is not definitely more advantageous than using 12 - lead ECG, when only traditional analysis of electrocardiographic tracings is performed.