Significance of systolic-phase imaging on full-phase ECG-gated CT angiography to detect intimal tears in aortic dissection.

ABSTRACT

PURPOSE: For patients with aortic dissection (AD) and intramural hematoma (IMH), the optimal cardiac phase to detect intimal tears (IT) and ulcer-like projections (ULP) on retrospective electrocardiogram (ECG)-gated computed tomography angiography (CTA) remains unclear. The purpose of this study was to compare the accuracy of retrospective ECG-gated CTA for detecting IT in AD and ULP in IMH between each cardiac phase. MATERIALS AND METHODS: A total of 75 consecutive patients with AD and IMH of the thoracic aorta were enrolled in this single-center retrospective study. The diagnostic performance to detect IT and ULP in the thoracic aortic regions (including the ascending aorta, aortic arch, and proximal and distal descending aorta) was compared in each cardiac phase on retrospective ECG-gated CTA. RESULTS: In the systolic phase (20%), the accuracy, sensitivity, and specificity to detect IT in AD was 64% (95% confidence interval [CI] 56-72%), 69% (95%CI 60-78%), and 25% (95%CI 3.3-45%), respectively. In the diastolic phase (70%), the accuracy, sensitivity, and specificity to detect IT in AD was 52% (95%CI 43-60%), 52% (95%CI 42-61%), and 50% (95%CI 25-75%), respectively. The accuracy to detect IT in AD on ECG-gated CTA was significantly higher in the systolic phase than that in the diastolic phase (P = 0.025). However, there were no differences in the accuracy (83%; 95%CI 78-89%), sensitivity (71%; 95%CI 62-80%), or specificity (100%; 95%CI 100%) to detect ULP in IMH between the cardiac cycle phases. CONCLUSION: Although it is currently recommended for routine diagnosis of AD and IMH, single-diastolic-phase ECG-gated CTA has risk to miss some IT in AD that are detectable in the systolic phase on full-phase ECG-gated CTA. This information is critical for determining the optimal treatment strategy for AD.