Abstract WMP39: Incorporation of Automated Robotic Transcranial Doppler to Screen for Patent Foramen Ovale and Quantify Right to Left Shunt Severity in the Evaluation of Ischemic Stroke Patients for Stroke Etiology at a Comprehensive Stroke Center

Abstract

Background: Patent foramen ovale (PFO) as a source of right-left shunting (RLS) is a recognized risk factor for recurrent stroke in young patients. Current standard care for RLS diagnosis includes transthoracic echocardiography (TTE), transesophageal echocardiography (TEE) and manual transcranial Doppler (TCD) which have disadvantages. The purpose of this retrospective review is to evaluate incorporation of an automated robotic TCD (rTCD) into the standard evaluation for RLS as cause of stroke, as prior studies have indicated higher sensitivity and quantification of RLS size to direct cardiology referral for potential PFO closure. Methods: Between Dec 2021-Oct 2022, 200 stroke patients age <65 were included in this retrospective review. Average age was 55.7 ± 11.1 yrs. All patients had standard TTE and an rTCD (NovaSignal) with saline bubble injection at rest and Valsalva strain for assessment of RLS performed by echo and ultrasound technicians. Neurologists read results, including RLS shunt severity using the Spencer Shunt Grading scale (0-5), where Grade 3-5 is considered large. Results: The rate of positive RLS shunt using rTCD was 55.5% (n=111), of which 23.5% (n=47) were Spencer grade 3+ or higher. No bone window rate was 8.5% (n=17), which is lower than reported manual TCD rates. There were no rTCD RLS shunt negative patients who had a positive TTE or TEE. Of patients who had positive rTCD (n=14) and TTE with bubble, only 5 TTE’s were positive (35.7% sensitivity). Of rTCD positive patients (n=23) who also had TEE with bubble, only 11 were positive (48% sensitivity). And of 15 rTCD Grade 3+ RLS, only 11 TEE with bubble were positive (73% sensitivity). Conclusion: We successfully replaced rTCD for TTE with bubble to screen for RLS on patients who were admitted with a stroke or TIA, with the following advantages. Non-TCD technicians could be trained to perform the tests with accuracy. rTCD was more sensitive in detecting RLS than TTE with bubble. Quantification of RLS severity identified stroke patients with potentially higher risk of recurrence, prioritized to more aggressive medical management and cardiology referral for TEE and potential PFO closure.