Abstract TP260: Paste-able Soft Ultrasound Probe as a Reliable Alternative for Right-to-Left Shunt Detection in Patent Foramen Ovale-Related Stroke

Background and Objective: Patent foramen ovale (PFO) is a cause of cryptogenic stroke, and right-to-left shunt (RLS) detection is key for diagnosis and management. Transcranial color flow imaging (TC-CFI) is limited by inadequate temporal bone windows in up to 30% of patients. We developed a novel paste-able soft ultrasound probe (PSUP) applied to the neck to detect microbubbles via the carotid artery, offering a noninvasive, operator-independent alternative (Figure). This study prospectively evaluated PSUP reliability compared with transthoracic echocardiography (TTE) and TC-CFI. Methods: Consecutive patients undergoing RLS evaluation with TTE and TC-CFI for cerebral infarction workup or post–PFO closure follow-up were included. During TC-CFI, the middle cerebral artery was observed; if not visualized, the vertebral artery was assessed. Simultaneously, a PSUP was taped to the ipsilateral common carotid artery and assessed using the same protocol. Contrast (air, saline, blood) was injected intravenously, and microbubbles were counted at rest and during the Valsalva maneuver. RLS grades were: negative (0 bubbles), low (1–19), moderate (≥20 with discernible particles), and high (≥20 with indistinguishable particles). For unilateral TC-CFI and PSUP, thresholds were halved. Agreement was assessed using Fleiss’ κ, and reliability using Cronbach’s α (0–1 scale; higher values indicate better agreement/consistency). α was calculated for all three modalities and recalculated after excluding one at a time; a greater drop indicated greater contribution to reliability. Results: A total of 528 tests from 314 patients (175 males; mean age 63 years) were analyzed. Fleiss’ κ was 0.67 at rest and during Valsalva (both p < 0.01), indicating good agreement. Cronbach’s α was 0.95 at rest and 0.96 during Valsalva. When excluding each modality, α at rest was: TTE 0.95, TC-CFI 0.94, PSUP 0.91; during Valsalva: TTE 0.96, TC-CFI 0.94, PSUP 0.91 (Table). Excluding PSUP caused the largest α decrease, showing its greatest contribution to reducing variability and improving consistency. Conclusion: PSUP demonstrated reliability comparable to or exceeding conventional methods. Its noninvasive, neck-applied design provides a practical alternative, especially for patients unsuitable for transcranial imaging, potentially broadening screening and follow-up options in PFO-related stroke care.