Assessment of location and size of myocardial infarction with contrast-enhanced echocardiography. II. Application of digital imaging techniques

Abstract

Contrast echocardiography can be used to identify nonperfused regions of myocardium and localize and quantify infarcted myocardium. Analysis is usually undertaken by visual inspection of an analog two-dimensional echocardiographic image. The purpose of our study was to apply digital imaging techniques to contrast-enhanced echocardiograms for the determination of myocardial infarct size. Myocardial contrast was produced by an injection into the aortic root of a mixture of hydrogen peroxide and blood. Sixteen open chest dogs were studied 4 hours after coronary artery occlusion. Echocardiograms were evaluated by two independent observers. The results were compared with infarct location and size determined with nitro-blue tetrazolium staining of the corresponding slice of the left ventricle. Both the routine analog echocardiographic image and the digital subtraction image were analyzed. For the latter, three precontrast and three postcontrast echocardiographic enddiastolic fields were digitized in a 256 X 256 X 6 bit matrix and then averaged. Average pre- and postcontrast images were mathematically subtracted to form the digital subtraction image. There was excellent correlation between the percent of infarct determined with digital subtraction contrast echocardiography and results of nitro-blue tetrazolium staining (r = 0.97, SEE = 0.04, p less than 0.001). Using linear regression, the relation between infarct size by the two studies was best described by the equation DSI = 0.92 NBT + 0.03, where DSI = digital subtraction image and NBT = infarct size by nitro-blue tetrazolium. Inter- and intraobserver variability were also excellent (r = 0.93 and 0.96, respectively).(ABSTRACT TRUNCATED AT 250 WORDS)