Background: Cardiac computed tomography (CCT) has the potential to assess both coronary anatomy and ventricular function in a single study. We examined the agreement between CCT and myocardial perfusion scintigraphy (MPS) for the assessment of global and regional ventricular function.
Methods and results: Research CCT was performed in 52 patients with a low to intermediate likelihood of coronary artery disease referred for MPS. Left ventricular end-diastolic volume, left ventricular end-systolic volume, left ventricular ejection fraction (LVEF), and myocardial wall motion and thickening were compared between techniques. In addition, myocardial contrast attenuation on CCT was compared with radiotracer uptake on MPS. LVEF values agreed well (mean difference, 4.1%; SD, 15.13%), but CCT left ventricular end-diastolic volume was greater compared with MPS (mean difference, 46.0 mL; SD, 33.34 mL) (P < .001). There was moderate agreement for segmental myocardial motion and thickening, with kappa values of 0.57 (95% confidence interval, 0.51-0.63) and 0.47 (95% confidence interval, 0.41-0.53), respectively. Seventeen patients had hypoattenuation in at least 1 myocardial segment on CCT. Three of four patients with concomitant abnormalities of wall motion and thickening on CCT had infarction in the same territory on MPS.
Conclusions: There was good agreement for LVEF between CCT and MPS but myocardial volumes differed, and these modalities cannot be used interchangeably. Mild abnormalities of regional function are detected more commonly by CCT than by MPS. Myocardial hypoattenuation on CCT is highly specific for myocardial infarction when associated with reduction of systolic wall thickening and regional wall motion abnormality.