In vivo measurement of water diffusion in the human heart

Abstract

Existing magnetic resonance methods for diffusion imaging, including echo planar, are ineffective in the beating heart due to motion-induced signal attenuation. To overcome this problem, we used a diffusion-weighted stimulated echo-echo planar magnetic resonance imaging sequence. The two lobes of the diffusion-sensitizing gradient were synchronized to the same point in successive cardiac cycles in order to fix the cardiac position and avoid bulk motion effects. The apparent diffusion coefficients (ADCs) of the interventricular septum in 12 healthy subjects for diffusion gradients along the x-, y-, and z-directions were 1.40 ± 0.27, 1.48 ± 0.35, and 1.78 ± 0.27 × 10⁻³ mm²/s. The ADCs of the interventricular septum in a second group of 15 healthy subjects for diffusion gradients along the short axis, horizontal and vertical long axes were 0.92 ± 0.15. 1.50 ± 0.15, and 1.10 ± 0.24 × mm²/s. Because the ADCs were less than the measured values for skeletal muscle and their standard deviations were low, it seems unlikely that bulk motion effects made the dominant contribution to the measured myocardial ADC for the interventricular septum, although motion and/or susceptibility artifacts frequently degraded measurements in the free wall of the left ventricle. Additional evidence that ADC was not predominantly determined by wall motion was obtained in a third group of patients with various cardiac abnormalities, in whom there was only a weak correlation between ADC and ejection fraction. Although further study is needed to better understand the factors contributing to the myocardial ADC, we hypothesize that the measured diffusional anisotropy in the septum might be explained largely on the basis of myofiber orientation.