Comparative Accuracy of Transcranial Doppler, Transthoracic Echocardiography, Transesophageal Echocardiography, and Cardiac CT in Detecting Right-to-Left Shunt in Embolic Stroke of Undetermined Source
Abstract Body: Introduction: Embolic Stroke of Undetermined Source (ESUS) represents about 25% of ischemic strokes, with patent foramen ovale (PFO) involved in roughly 25% of those cases. Recent studies (REDUCE, CLOSE, RESPECT) indicate that PFO closure may benefit select patients. Accurate PFO diagnosis is crucial for treatment decisions but current methods, such as transcranial Doppler (TCD) with bubble study, transthoracic echocardiography with bubble study (TTE), transesophageal echocardiography (TEE), and cardiac computed tomography (CT), vary in accuracy. This study aims to assess the diagnostic performance of TCD compared to TTE, TEE, and cardiac CT, to determine the most reliable modality for detecting PFO in patients with ESUS.
Methods: We conducted a retrospective analysis of ESUS patients aged 18 and older referred to our vascular lab between July 2023 and September 2024. Each patient underwent at least two PFO detection methods: TCD with bubble study, TTE with bubble study, TEE, and cardiac CT. PFO presence was confirmed if identified by at least two detection methods. Patients with extracardiac shunts were excluded. Sensitivity, specificity, and overall test performance were evaluated using the Area Under the Curve (AUC) of a Receiver Operating Characteristic (ROC) curve, with higher AUC values indicating better performance. Differences in AUCs were assessed using DeLong’s test. Statistical analysis was performed using Analyse-it for Microsoft Excel (version 2.30).
Results: Of 187 enrolled patients, 63 (33.7%) had evidence of right-to-left shunt. The mean age of the cohort was 63 ± 18 years, with 51% being male. PFO was confirmed in 36 patients. Sensitivity was highest for TCD (93.1%), followed by TTE (87.1%), TEE (86.7%), and cardiac CT (42.9%). Specificity was highest for cardiac CT (87.5%), followed by TTE (84%), TCD (59.1%), and TEE (47.1%). The AUC was highest for TTE (AUC: 0.85, 95% CI: 0.76–0.95), followed by TCD (AUC: 0.75, 95% CI: 0.64–0.87), TEE (AUC: 0.66, 95% CI: 0.50–0.81), and cardiac CT (AUC: 0.65, 95% CI: 0.42–0.88). No statistically significant differences in diagnostic performance across pairwise comparisons were found (p > 0.05).
Conclusion: TCD with bubble study is highly sensitive for screening right-to-left shunt, while cardiac CT is highly specific for detecting PFO in ESUS patients. Further studies with larger sample sizes are needed to confirm these findings and refine diagnostic strategies.
Ravi, Vikas
( University of California, San Diego
, San Diego
, California
, United States
)
Meyer, Dawn
( University of California, San Diego
, San Diego
, California
, United States
)
Hailey, Lovella
( University of California, San Diego
, San Diego
, California
, United States
)
Mortin, Melissa
( University of California, San Diego
, San Diego
, California
, United States
)
Bavarsad Shahripour, Reza
( University of California, San Diego
, San Diego
, California
, United States
)
Pirahanchi, Yasaman
( University of California, San Diego
, San Diego
, California
, United States
)
Jazayeri, Seyed Behnam
( Mayo Clinic
, Rochester
, Minnesota
, United States
)
Carrion Penagos, Julian
( University of California, San Diego
, San Diego
, California
, United States
)
Bowers, Jeffrey
( University of California, San Diego
, San Diego
, California
, United States
)
Meyer, Brett
( University of California, San Diego
, San Diego
, California
, United States
)
Agrawal, Kunal
( University of California, San Diego
, San Diego
, California
, United States
)
Modir, Royya
( University of California, San Diego
, San Diego
, California
, United States
)
Hemmen, Thomas
( University of California, San Diego
, San Diego
, California
, United States
)
Author Disclosures:
Vikas Ravi:DO NOT have relevant financial relationships
| Dawn Meyer:DO have relevant financial relationships
;
Speaker:Chiesi:Active (exists now)
; Speaker:Astra Zeneca:Active (exists now)
| Lovella Hailey:No Answer
| Melissa Mortin:DO NOT have relevant financial relationships
| Reza Bavarsad Shahripour:DO NOT have relevant financial relationships
| Yasaman Pirahanchi:DO NOT have relevant financial relationships
| Seyed Behnam Jazayeri:DO NOT have relevant financial relationships
| Julian Carrion Penagos:DO NOT have relevant financial relationships
| Jeffrey Bowers:DO NOT have relevant financial relationships
| Brett Meyer:DO have relevant financial relationships
;
Advisor:Sevaro Health:Past (completed)
| Kunal Agrawal:DO have relevant financial relationships
;
Consultant:Community Health Accreditation Partners (CHAP):Active (exists now)
| Royya Modir:DO have relevant financial relationships
;
Consultant:Medtronic:Active (exists now)
| Thomas Hemmen:DO NOT have relevant financial relationships