Transient Ischemic Attack (TIA) is defined as transient neurological deficit without evidence of magnetic resonance imaging restricted diffusion. Work up for TIA is undertaken to identify risk factors that could be modified to prevent stroke. Data is sparse on Trans Esophageal Echo (TEE) for TIA work up. We aimed to determine the frequency, predictors, and yield of TEEs in this group.

Retrospective data from a prospectively collected dataset was analyzed. Two groups were formed: TIA with no TEE and TIA with TEE. We identified the difference among these groups and compared the baseline characteristics. Abnormal TEE findings that would have changed management are summarized. 

360 TIA patients were included in this study. Of 360 TIAs, 15 patients (4%) underwent TEE. Significant differences in these two groups were for weakness as one of TIA presenting symptom (n = 7; 47%; p=0.013), intracranial vessel occlusion on CT angiogram (n= 3; 30%; p=0.015) and old cortical infarct on CT head (n=5; 33%; p=<0.001). 50% had left ventricular hypertrophy (n = 6) on TTE. On TEE; 40% (n = 6) patients were found to have left atrial enlargement, and 30% (n=3) had spontaneous echo contrast. No patients were found to have left atrial thrombus. 

Almost one out of three patients with TIA who had TEE had spontaneous echo contrast warranting a change in treatment (anticoagulation). TEE was selected for TIA patients with clinical (unilateral weakness) and imaging findings (intracranial vessel occlusion or old cortical infarct).  Our findings indicate our selection bias for TEE and utility in our selected patients with TIA.  A large study of unselected TIA patients is needed to determine utility and predictors of a finding, which would warrant anticoagulation. Perhaps, classification of TIA etiology per the TOAST criteria could aid in identifying this subgroup as well.