Association between large aortic arch atheromas and high-intensity transient signals in elderly stroke patients.

BACKGROUND AND PURPOSE Aortic arch atheromas (AAs) have been shown to be a risk factor for ischemic stroke (IS) in the elderly because of their potential for cerebral embolization. However, the association between AAs and the presence of cerebral microemboli has not been clearly established. The aim of this study was to determine whether large AAs are associated with an increased frequency of high-intensity transient signals (HITS) in elderly patients with IS. METHODS We performed bitemporal simultaneous HITS monitoring of both middle cerebral arteries in 62 consecutive elderly patients with acute IS (mean age 72.5+/-8.8 years, 65% men). In 16 patients, one or both temporal windows were inadequate; therefore, the analysis of HITS was performed in the remaining 46 patients. All patients underwent omniplane transesophageal echocardiography (TEE), and they had no significant extracranial or intracranial artery disease and no cardiac prosthetic valves. Large AA was defined as > or = 4 mm in thickness. Complex AA was defined as ulcerated or mobile, regardless of plaque thickness. HITS monitoring was performed within 24 hours of TEE and analyzed by an experienced neurologist-sonographer blinded to TEE findings. A 9-dB threshold was chosen to discriminate HITS from background Doppler signal. The HITS counts in the left and in the right middle cerebral arteries were added and reported as a total number of HITS in 30 minutes. RESULTS HITS were detected in 14 (78%) of 18 patients with large AAs versus 8 (29%) of 28 patients with no or small AAs (odds ratio [OR] 8.8, 95% CI 2.2 to 34.8; P=0. 001). The association was also present in 27 patients with no other cardiac embolic sources, such as atrial fibrillation, patent foramen ovale, spontaneous echo contrast, and thrombus (7 of 10 patients with large AAs versus 3 of 17 patients with small or no AA; OR 10.9, 95% CI 1.7 to 68.5; P=0.013). Complex AAs were associated with a higher frequency of HITS than were noncomplex AAs (6 of 6 patients with complex AAs versus 15 of 39 patients with noncomplex AAs; OR 2. 6, 95% CI 1.7 to 3.9; P=0.005). CONCLUSIONS HITS are significantly associated with large AAs in elderly stroke patients. This observation may support the causal role of large AAs in IS.