A quantitative nuclear magnetic resonance (NMR) imaging method of evaluating regional left ventricular function was compared with histochemical evidence of infarction in dogs and functional measurements in patients. Short-axis images of the heart were obtained at end-diastole and at 100 ms intervals thereafter. Regional diastolic left ventricular wall thickness and maximal percent systolic wall thickening were measured at the level of the papillary muscles in each of six segments. In six normal dogs, the mean end-diastolic wall thickness was 9 +/- 1.6 mm, and the mean maximal percent thickening was 61 +/- 11%. In eight dogs with a 4 day old infarct, maximal percent thickening was 5 +/- 8% (p less than 0.001) in the infarcted segments. In 10 normal human volunteers, the mean end-diastolic wall thickness was 10.1 +/- 1 mm, and the mean maximal percent systolic wall thickening was 60 +/- 18%. Reduced maximal percent systolic wall thickening was defined as a value greater than or equal to 2 SD below the mean value obtained in normal volunteers. Seven patients with regional wall motion abnormalities were independently assessed by NMR imaging and biplane ventriculography. With a sensitivity of 94% and a specificity of 80%, NMR imaging demonstrated reduced maximal percent systolic wall thickening in the same segments identified as akinetic or dyskinetic by biplane ventriculography. Thus, abnormalities of regional systolic wall thickening are accurately identified with this quantitative imaging technique.