The 10-20% incidence of cardiac rupture in acute fatal myocardial infarction has not changed in the past century, and little is known about its pathophysiology. To determine whether expansion - acute dilatation and thinning of the area of infarction not explained by resorption of necrotic tissue - may be a variable predictive for rupture, we studied 110 consecutive autopsied patients who died of acute myocardial infarction. The presence and severity of expansion was determined qualitatively on a scale of 0-4+ from review of the postmortem radiographs, and quantitatively by a ratio of wall thickness adjacent to the rupture site to thickness of noninfarcted wall. By radiographs, 54 (49%) had expansion, of whom 23 (43%) had rupture in the zone of expansion; of the 56 cases without expansion only one (2%) had ruptured (p<0.005). The severity of expansion related to rupture: 50% with 3-4+ expansion ruptured, compared with only 23% with 1-2+ expansion. Wall thickness ratios for patients without expansion were 0.93 ± 0.10; for those with expansion but no rupture, 0.62 ± 0.16; and 0.49 ± 0.13 for those with expansion and rupture (p<0.01). Wall thinning in these three groups was noted at average postinfarction intervals of 8, 7 and 5.5 days, respectively - too early to be caused by resorption. Thus, expansion of acute myocardial infarction is a previously unappreciated variable that appears to be predictive of rupture, suggesting that regional dilatation and thinning of newly infarcted myocardium may be of pathogenetic importance in the development of rupture. Since expansion can be identified with two-dimensional echocardiography, and occurs over days, these findings suggest that patients at high risk for rupture might be recognized early and interventions to prevent further expansion and rupture evaluated.