Determination of left ventricular mass in dogs with rapid-acquisition cardiac computed tomographic scanning.

Abstract

The development of left ventricular hypertrophy in patients with heart disease often has far-reaching clinical implications with respect to overall morbidity and mortality. Approaches used to assess left ventricular mass include electrocardiography, echocardiography, contrast ventriculography, single photon-emission tomography, and conventional computed tomography. However, all of these modalities suffer from some major draw back that precludes widespread application to all patients. In this study we assessed the accuracy of determinations of left ventricular mass in 22 dogs by rapid-acquisition (50 msec) computed axial tomography (RACAT), an ultrafast computed tomographic (CT) instrument. Electrocardiographically triggered, end-diastolic, short-axis cardiac scans were obtained from apex to base during administration of intravenous iodinated contrast. Myocardial edges were determined for each tomographic scan by two methods: the regional half-contour method (the CT density half way between that of the left ventricular myocardium and adjacent ventricular cavities or lung) and "interactive plateau thresholding" of the cardiac borders. Left ventricular mass by RACAT was calculated as the sum of the mass of each individual scan from apex to base (modified Simpson's rule). Postmortem left ventricular mass ranged from 58 to 160 g. The correlation between true left ventricular mass and tomographically determined mass was excellent (r = .99), with the slope and y intercept not statistically different from 1 and 0, respectively. The standard error of the estimate was 4.1 g. Interobserver and intraobserver variability for determining left ventricular mass demonstrated excellent agreement (r = .99 and r = .99, respectively). We conclude that quantitative assessment of left ventricular mass can be accurately and reproducibly performed in dogs by rapid acquisition CT scanning. It is likely that this technique will be readily transferable to the clinical settings and prove to be an important method for quantifying left ventricular mass in patients.