PATHOPHYSIOLOGY AND NATURAL HISTORY MYOCARDIAL INFARCTION Correlation between echocardiographic endocardial surface mapping of abnormal wall motion and pathologic infarct size in autopsied hearts

We previously developed a cross-sectional echocardiographic technique for quantitatively mapping the endocardial surface of the left ventricle and on which regions of abnormal wall motion can be superimposed in their correct spatial distribution. This endocardial mapping technique (EMT) provides a measure of the left ventricular endocardial surface area (ESA in cm2), the area of abnormal wall motion (AWM in cm2), and the overall percent dysfunction (%AWM) as a measure of the functional "infarct size." To test this approach, we compared the EMT measurements with the actual endocardial surface area (in cm2) and pathologic infarct size (both percent infarct by volume and percent endocardial surface overlying infarct) measured at later autopsy in 20 adults (14 men, six women) ranging in age from 47 to 76 years (mean 64 + 9.6 years). The median interval from echocardiographic study to death was 19 days (range 1 to 269 days). Patients were divided into two groups based on the age of their infarcts at the time of death: (1) recent (infarct age < 14 days; mean age 5.3 + 4.6 days) and (2) old (infarct age > 6 months; mean age 3.6 + 3 years). When the left ventricular endocardial surface area at autopsy was compared with the EMT-derived ESA, a close correlation was found (EMT area = 1.17 x autopsy area + 20.4; r = .94, p = .0001), with the systematic difference in the measurements accounted for by systolic arrest, loss of distending pressure, and specimen shrinkage. The echocardiographic measure of infarct size (%AWM) correlated well with the autopsy percent infarction by volume (%AWM 1.1 x infarct volume + 5.5; r .82, p = .0001). Similarly, a good correlation was found for the percent abnormal wall motion and the autopsy percent endocardial surface area overlying infarction (%AWM = 0.89 x infarct area 0.9; r .89, p = .0001). When the data were examined in relation to the age of the myocardial infarct, the echocardiographic %AWM appeared to overestimate the autopsy infarct size (by percent infarct volume) in the recent infarct group (n = 6), and underestimate the extent in the old infarct group (n = 13). The findings suggest that the EMT will provide a useful quantitative measure of left ventricular endocardial surface area and the extent of ischemic/infarct-related dysfunction. Circulation 77, No. 5, 978-987, 1988. WITH INCREASING INTEREST in myocardial salvage after presumed acute coronary occlusion, an accurate, repeatable measure of the amount of left ventricle rendered dysfunctional by the ischemic process is needed.1 Ideally, such a method should be quantitatively accurate, noninvasive, and easily repeatable so From The Cardiac Unit of the Massachusetts General Hospital and Harvard Medical School, Boston. Address for correspondence: Arthur E. Weyman, M.D., Cardiac Ultrasound Laboratory, Phillips House 8, Massachusetts General Hospital, Fruit Street, Boston, MA 02114. Received Sept. 21, 1987; revision accepted Feb. 4, 1988. Dr. Wilkins was supported in part by the New Zealand National Heart Foundation and the Royal Australasian College of Physicians, Wellington, New Zealand. Dr. Choong was supported by the National Heart Foundation of Australia, Canberra, A.C.T., Australia. Dr. Thomas was supported in part by grant HL07535 from the National Heart, Lung, and Blood Institute, Bethesda. 978 that comparisons through time and between subjects can be made. Any acceptable functional measure of "infarct size" should also demonstrate reasonable agreement with the extent of pathologic change. We previously developed a quantitative echocardiographic method for mapping the left ventricular endocardial surface on which regions of abnormal wall motion can be superimposed and measured.2. 3Although this method has been experimentally validated in a canine preparation of infarction, the relationship of the mapped area of dysfunction to the pathologic infarct size in humans has not been examined.