Abstract 16320: Reliability of 64-Slice Multidetector Computed Tomography in Predicting Coronary Flow During Percutaneous Coronary Intervention

Abstract

Background: Multidetector computed tomography (MDCT) is a noninvasive method for assessment of atherosclerotic coronary plaque morphology. However, association between coronary plaque morphology on MDCT and coronary flow during percutaneous coronary intervention (PCI) has not been fully investigated.

Method: We investigated consecutive 77 lesions with 77 patients with coronary artery disease (CAD) who underwent MDCT within 30 days before PCI. Plaque morphologies and CT density at the culprit lesion were assessed by MDCT analysis. Plaque volume on MDCT was calculated by adding plaque area every 1mm. Coronary flow after PCI was investigated according to the Thrombolysis in Myocardial Infarction flow grade (TMG). All lesions were classified into 2 types of lesions, lesions with occurrence of slow-flow during PCI (TMG 0-2) and those without.

Result: Slow-flow phenomenon was detected in 16 lesions (20.8%) during PCI. CT density was lower (35.6±17.7 HU vs. 68.9±30.5 HU, p<0.0003) and plaque volume was greater (123.0±86.4mm³ vs. 66.9±55.9mm³ p=0.02) in lesions with slow-flow phenomenon than without. Signet ring-like appearance (12.5% vs. 1.65%, p=0.04) and positive remodeling (37.5% vs. 9.83%, p=0.006) were significantly more frequent in lesions with slow-flow than without. Multiple logistic regression analysis revealed that CT density (OR: 0.92, 95% CI: 0.869-0.956, p=0.0003) and plaque volume (OR: 1.02, 95% CI: 1.004-1.027, p=0.007) were independently related with slow-flow during PCI. The cut-off values for CT density and plaque volume for slow-flow during PCI were 39.0 HU (sensitivity 75.0%, specificity 88.2%, area under the receiver-operating characteristic curve (AUC) 0.85 ) and 73.4mm³ (sensitivity 87.5%, specificity 66.6%, AUC 0.75), respectively. For predicting slow-flow during PCI, the diagnostic power of combination of lower CT density <39.0 HU and greater plaque volume >73.4 mm³ showed 55.6% of sensitivity, 96.6% of specificity, 83.3% of positive predictive value, 87.7% of negative predictive value, and 87.0% of diagnostic accuracy.

Conclusion: Our study suggests that non-invasive assessment of plaque morphologies by MDCT may help to predict poor outcome after PCI in patients with CAD.