Circumferential myocardial shortening in the normal human left ventricle. Assessment by magnetic resonance imaging using spatial modulation of magnetization.

BACKGROUND: Conventional cardiac imaging methods do not depict true segmental myocardial shortening, since they cannot determine segment length between fixed points in the myocardium. METHODS AND RESULTS: We used electrocardiographically gated magnetic resonance imaging with spatial modulation of magnetization to noninvasively "tag" the myocardium with dark stripes at uniform 7-mm intervals center to center at end diastole. We then determined end-systolic stripe separation and thereby calculated circumferential shortening. When end systole was not reached in the first image series, a second temporally overlapped series starting in late systole was used to determine late-systolic shortening. Septal, anterior, lateral, and inferior segments were assessed at endocardium, midwall, and epicardium on five midventricular short-axis sections each in 10 normal volunteers. A transmural gradient in circumferential shortening was observed, with the percentage of endocardial segment shortening consistently greater than epicardial segment shortening (epicardial, 22 +/- 5%; midwall, 30 +/- 6%; and endocardial, 44 +/- 6%; p less than 0.0001 by analysis of variance). Circumferential shortening varied from apex to base with slices closer to the base of the left ventricle showing less shortening at the midwall (28 +/- 9%) and endocardium (39 +/- 6%) than more apical slices at the midwall (34 +/- 13%) and endocardium (49 +/- 9%) (p less than 0.05 and p less than 0.01, respectively, by analysis of variance). CONCLUSIONS: Transmural and longitudinal heterogeneity of circumferential shortening is present in the normal human left ventricle. Magnetic resonance imaging with spatial modulation of magnetization is a powerful new tool for assessment of circumferential shortening and provides information unobtainable with conventional imaging methods.