Accuracy of 16-detector multislice spiral computed tomography in the initial evaluation of dilated cardiomyopathy

Background Multislice computed tomography (MSCT) recently proved its accuracy in the detection of coronary artery disease (CAD). It can also give information about left ventricular function and venous network anatomy. We here sought to validate a MSCT-based strategy in the initial evaluation of patients with dilated cardiomyopathy (DCM). Methods Thirty-six patients with DCM underwent cardiac MSCT before conventional coronary angiography with ventriculography. We analysed arterial calcium score [Agatston score equivalent (ASE)], coronary stenosis, left ventricular parameters, and venous network. Results The sensitivity of an MSCT-based strategy in detecting significant CAD was 100%, and the specificity, 80%. The positive and negative predictive values were, respectively, 67% and 100%. For ASE <1.000 (75% of patients), MSCT detected all non-CAD patients (motion artifacts), enabling conventional angiography to be avoided in 21 (77.7%) of 27 patients. For ASE ≥1000, MSCT enabled conventional angiography to be avoided in only 2 (22.2%) of 9 patients. The ventricle was assessable in 83.4% (30 patients) on MSCT. Correlation coefficient Rs with ventriculography were 0.78 (P<.0001), 0.77 (P<.0001) and 0.82 (P<.0001) respectively for end-diastolic volume, end-systolic volume, and ejection fraction. The venous network was assessable in all patients both on MSCT and angiography. Conclusion In patients undergoing an initial evaluation of DCM, MSCT appears to be an effective alternative to conventional angiography. The following attitude may be proposed: when ASE >1.000, conventional coronary angiography is mandatory due to MSCTʹs poor interest in such cases; when ASE <1.000, a contrast-enhanced MSCT may, when normal, replace coronary angiography.