Demonstration of penetrating intramyocardial coronary arteries with high-frequency transthoracic echocardiography and Doppler in human subjects

Characterization of intramyocardial coronary artery flow may offer insight into the spectrum of coronary physiology. The purposes of this study were to test the feasibility of detection and measurement of intramyocardial coronary artery flow by using high-frequency transthoracic ultrasound and to evaluate the hemodynamic and morphologic differences in intramyocardial coronary arteries between patients with echocardiographically normal myocardium and patients with diseased myocardium. In 116 subjects (age 58 ± 19 years; male:female 67:49; 58 normal [control subjects], 40 with left ventricular hypertrophy [LVH], 18 with systolic left ventricular dysfunction [cardiomyopathy, CM]), we examined the myocardium just beneath the apical impulse window at a depth of 3 to 5 cm by using a 6- or 7-MHz centerline frequency transducer. For color Doppler examination, a special preset coronary program with a low Nyquist limit (12 to 20 cm) was used. After obtaining linear color signals, the width and length, peak and mean diastolic pulsed Doppler flow velocities, diastolic velocity time integrals, and percent duration of diastolic Doppler flow were measured. The number of linear color flow signals per square centimeter was counted in 520 different cardiac cycles, and the angles formed by their inner curvature was measured with a graduated protractor. We identified color flow Doppler signals within the myocardium having a mean width of 1.1 ± 0.4 mm and flow direction from epicardium to endocardium in 104 (89.7%) subjects and spectral Doppler signals in 74 (63.8%) subjects. In 33 (45.8%) subjects, only diastolic flow was detected and in 39 (54.2%) subjects, diastolic flow was predominant with systolic reversal. Peak and mean diastolic flow velocities and velocity time integrals of spectral Doppler signal in control subjects were 26.2 ± 8.6 cm/s, 19.0 ± 6.3 cm/s, and 9.5 ± 2.7 cm, respectively. There were no significant differences in width and density of linear color flow signals among the 3 groups. The color flow signals in the LVH and CM groups had a narrower angle of inner curvature (P < .005 for LVH, P < .05 for CM, respectively), and their spectral Doppler signals showed significantly higher diastolic velocities and shorter diastolic flow duration (P < .005 for LVH, P < .05 for CM, respectively) than those of the control subjects. Detection and measurement of flow signals consistent with penetrating intramyocardial coronary arteries are feasible in a high percentage of subjects by use of high-frequency transthoracic ultrasound. The findings in patients with LVH and CM suggest that there are distinct hemodynamic and morphologic departures from those with normal left ventricles that may be a consequence of disordered myocardial perfusion in diseased myocardium.