Comparison of Computed Tomography Angiography and Transesophageal Echocardiography for Evaluating Aortic Arch Disease

Abstract

Aortic arch (AA) atheroma is a common source of artery-to-artery embolism. Identification of AA atherosclerotic disease is an important component of the embolic stroke workup. Transesophageal echocardiography (TEE) is the gold standard for AA evaluation, but it has associated risks and is not always readily available. Computed tomography angiography (CTA) is a rapid and noninvasive alternative. This study was conducted to compare the sensitivity and specificity of CTA and TEE for detecting AA disease. We performed a retrospective review of 250 consecutive patients at a tertiary stroke center who underwent both TEE and CTA within a 90-day period. We compared the presence and characteristics of AA plaques using a predetermined grading system for plaques in the ascending, transverse, and descending arch for both modalities (grades 1-4). Out of 750 AA segments (ascending, transverse, and descending AA in 250 patients), 494 were adequately imaged by CTA and TEE. The sensitivity of CTA in detecting grade 1-4 AA atheromas was 53%, and the specificity was 89%. For only high-grade atheromas, the specificity improved to 99%, but the sensitivity decreased to 23%. The negative predictive value of CTA for detection of AA atheromas was 60% (range 54%-65%) for all grades and 95% (range 92%-96%) for high-grade atheromas. CTA has a high negative predictive value for AA atheromas, especially for higher-grade atheromas, and thus may be a useful screening tool to exclude high-grade plaques, indicating a possible complementary role for CTA in detecting AA atheromas.