Thoracic Aortic Strain is Irrelevant Regarding Endograft Sizing in Most Young Patients

Abstract

The introduction of electrocardiogram gating in computed tomography (CT) angiography imaging of aortic disease raised questions whether diameters measured on diastolic images can lead to the undersizing of aortic endografts. As previous studies suggested that young patients may have the highest risk of an unintended undersize, the aim of our study was to analyze the strain of the thoracic aorta in a young patient cohort. We assessed the descending aorta of 52 patients (35 men, mean age 41.1±7.3years) who underwent coronary CT angiography (CCTA) with suspected coronary artery disease. To reduce radiation dose, native calcium score scans triggered on late systole were compared with diastolic phase CCTA images. Cross-section areas were measured, and effective diameters were derived at 3 levels of the visible segment of the descending aorta (P1, P2, and P3) in systole and diastole. Aortic pulsatility (mm, dsystolic-ddiastolic) and strain (%, [dsystolic-ddiastolic]/ddiastolic) were calculated at each level. All measurements were performed 3 times by 2 independent readers to evaluate interreader and intrareader reproducibility. A total of 936 measurements were performed. Significant differences were found between systolic and diastolic diameters at each location (all P<0.001). Average aortic pulsatility was 1.5±0.6mm at P1, 1.6±0.7mm at P2, and 1.7±0.7mm at P3, with a corresponding aortic strain of 6.7 ±3.1% at P1, 7.4 ±3.5% at P2, and 8.1 ±3.6% at P3. The differences between the strain of the measurement levels were not significant (P=0.344). Aortic strain and pulsatility did not show significant correlation with pulse pressure (P=0.693), patient age (P=0.649), or other anamnestic data. Intraclass correlation coefficient was in the range of 0.95-0.96 for interobserver and in the range of 0.95-0.97 for intraobserver analysis. This study shows that descending aortic strain can be measured precisely and reliably on images of routine CCTA examinations with native scans acquired during systole. We demonstrated that young adults have an aortic strain of 6.7-8.1%. As the average thoracic aortic strain was still lower than the recommended prosthesis oversize of 10%, routine use of systolic phase imaging cannot be recommended: it has no clinical benefit for the vast majority of the patients but increases the risk of motion artefacts. We also demonstrated that large interindividual differences are present in the scale of thoracic aortic strain, a phenomenon that needs further investigations to be fully understood.