BackgroundIn hypertensive left ventricular hypertrophy (LVH), intrinsic myocardial systolic function may be normal or depressed. Magnetic resonance tagging can depict intramural myocardial shortening in vivo.Methods and ResultsTagged left ventricular magnetic resonance images were obtained in 30 hypertensive subjects with LVH (mean LV mass index, 142±41 g/m) and normal ejection fraction (mean, 64±9%) using spatial modulation of magnetization. In 26 subjects, circumferential myocardial shortening (%S) was compared with results obtained in 10 normal subjects at endocardium, midwall, and epicardium on up to 4 short-axis slices each. Similarly, in 10 subjects, midwall longaxis shortening at basal, midventricular, and apical sites was compared with results obtained in 12 normal volunteers. Circumferential %S was reduced in hypertensive subjects. Mean shortening was 29±6% at the endocardium in hypertensive subjects versus 44±6% in normal subjects (P= .0001); 20±6% at the midwall versus 30±6% (P= .0001); and 13±5% at the epicardium versus 21±5% (P= .0002). However, the transmural gradient in percent shortening from endocardium to epicardium in hypertensive subjects paralleled that in normal subjects. The normal base-to-apex gradient in circumferential %S was absent in LVH. In contrast to normal subjects, circumferential %S showed regional heterogeneity in hypertensive subjects, being maximal in the lateral wall and least in the inferior wall. Longitudinal shortening was also uniformly depressed in hypertensive subjects: 10±9% at the base versus 21±6% in normal subjects (P= .0001); 14±8% at the midventricle versus 18±3% (P= .03); and 14±8% at the apex versus 18±4% (P= .04).ConclusionsIn hypertensive LVH with normal pump function, intramural circumferential and longitudinal myocardial shortening are depressed.