This study was designed to test the accuracy of echocardiographic radial shortening measurements during variable degrees of regional dysfunction produced by 14 transient (≤10 min) coronary occlusions (8 left anterior descending coronary artery, 6 left circumflex coronary artery) Followed by up to 24 h of reperfusion in chronically instrumented conscious dogs. Independent measurement of regional myocardial thickening was obtained using epicardial pulsed Doppler probes and served as a standard for comparison. Radial shortening fraction was derived from two-dimensional short-axis views along 12 equidistant radii. Six reference systems from the epicardial and endocardial centers of geometry (centroids) in a fixed or a floating position were explored. In the ischemic zone, percent thickening fraction averaged 22 ± 5% at baseline, decreased to -4 ± 4% during occlusion and gradually returned to baseline values after reperfusion. Percent change in radial shortening correlated significant by with percent change in thickening fraction in the ischemic zone. The worst correlation was seen with the Heating endocardial centroid (τ = 0.68), and the best was observed with the epicardial Heating reference (r = 0.91). Moreover, the epicardial floating reference provided narrower 95% confidence limits of radial shortening and less heterogeneity among radii than did fixed reference systems. Thus, compared with an independent standard, echocardiographic measurement of radial shortening from the short axis provided recognition of discrete grades of regional dysfunction induced by transient reversible ischemia. This technique may be amenable for serial assessment of regional function after interventions on the ischemic myocardium.