Abstract 1832: A Noninvasive Strategy of Carotid Intimal-Media Thickness Assessment and Transthoracic Echocardiography Identifies Unexplained Stroke Patients with a Low Yield of Transesophageal Echocardiography.
Background: Transesophageal echocardiography (TEE) to assess for cardiovascular source of embolus (CSE) is standard in the evaluation of patients (pts) with unexplained stroke or TIA. Carotid intimal-media thickness (CIMT) predicts aortic atherosclerosis and is associated with cardioembolic conditions. Harmonic transthoracic echocardiography (TTE) with bubble study is highly sensitive for detection of PFO. We hypothesized that a noninvasive strategy of CIMT assessment and TTE with bubble study would identify unexplained stroke or TIA pts in whom TEE provides little incremental diagnostic yield.
Methods: We prospectively performed CIMT measurement/carotid plaque screen and harmonic TTE with bubble study on 60 pts referred for TEE after stroke or TIA of undetermined origin. Pts with high risk embolic conditions (Afib/flutter, LVEF<30%, or carotid stenosis >50% by carotid Doppler or MRA) were ineligible. Carotid plaque screen was performed 1.5 cm above and below the carotid bulb bilaterally. Far wall mean common carotid IMT measurement was performed in the right and left common carotid artery using automated border detection software. A preset CIMT was selected for strategy testing based on prior pilot data.
Results: TEE findings included patent foramen ovale (20%), LA/LV thrombus (2%), spontaneous echo contrast (3%), valve vegetation/ mass/ tumor (7%), complex aortic atheroma [>4mm, or with ulceration or mobile features] (18%), and the composite of any potential cardiac source of embolus (CSE) (38%). TTE bubble study had a sensitivity of 92% and a specificity of 96% for detection of PFO on TEE. Among the 37% of pts with a CIMT < 0.78 mm without carotid plaque and with a negative transthoracic bubble study, the prevalence of any CSE on TEE was only 9%. Among the 63% of pts with CIMT ≥ 0.78 mm, with carotid plaque, or with a positive transthoracic bubble study, the prevalence of CSE on TEE was 55%. The sensitivity and NPV of this strategy for the detection of any CSE were 91%.
Conclusion: In pts with stroke or TIA of undetermined origin, a non-invasive strategy of CIMT assessment and TTE with bubble study can identify pts in whom further invasive evaluation with TEE will be of low diagnostic yield. This finding may be useful in the selection unexplained stroke or TIA for TEE.