Quantification of left ventricular mass using cardiac magnetic resonance imaging compared with echocardiography in domestic cats.

Abstract

The hypotheses were that cardiac magnetic resonance imaging (cMRI) would accurately determine LV mass in domestic cats and would do so more accurately than echocardiography (ECHO). ECHO was performed on seven sedated cats. LV mass was calculated using the truncated ellipse formula from a right parasternal long-axis view. T1 weighted gradient echo cMRI was acquired from anesthetized cats during multiple phases of the cardiac cycle. Short-axis images were obtained by acquiring 3 mm thick contiguous slices perpendicular to the cardiac long axis. LV mass was determined using Simpson's rule. Endocardial and epicardial borders were traced on each slice at end-systole, end-diastole, and mid-cycle and the difference in areas was myocardial area. Myocardial area was multiplied by slice thickness to calculate myocardial volume. Total (summated) myocardial volume was multiplied by myocardial density (1.05) to obtain LV mass at three measured phases of the cardiac cycle. Cats were euthanized and the LV was dissected and weighed to determine true mass. CMRI at end-systole most accurately quantified LV mass and was more accurate than echocardiography (P = 0.0078). Actual LV mass ranged from 6.5 to 10.5 g (mean = 8.5 g, SD = 1.6 g) compared with MRI LV mass at end-systole, which ranged from 6.7 to 11.1 g (mean = 8.7 g, SD = 1.7 g) and echocardiographic LV mass at enddiastole, which ranged from 5.2 to 9.1 g (mean= 7.1 g, SD = 1.8 g). Inter- and intraobserver variability for cMRI was 2%. CMRI obtained at end-systole accurately and reliably quantifies LV mass in domestic cats. It is more accurate than the echocardiographic method used in this study.