ALGORITHMIC APPROACH FOR DIFFERENTIATING CARDIOEMBOLIC AND LARGE ARTERY ATHEROSCLEROSIS ETIOLOGY OF ACUTE LARGE VESSEL OCCLUSIONS FOR MECHANICAL THROMBECTOMY

Document Type

Conference Proceeding

Publication Date

7-21-2024

Publication Title

J Neurointerv Surg

Abstract

Introduction Mechanical thrombectomy (MT) devices were fundamentally designed for the treatment of cardioembolic (CE) strokes despite 10-20% of LVOs being caused by large artery atherosclerosis (LAA) which may not respond comparably in outcome measures. Therefore, the differentiation between CE and LAA is critical for selecting the ideal MT technique and device. Here, we hypothesized that we can differentiate between the two using clinical and radiographic features to construct an algorithm for use in the emergent setting. Methods We developed a scoring system with multiple variables based on well-established risk factors for CE and LAA strokes with each allocated point weighed in association to the degree of contribution to its respective etiology (tables 1 and 2). We then tested the system using data from a single center retrospective chart review of acute MCA occlusions treated with MT with confirmed etiology as either CE or LAA. We utilized these variables to generate an area under the ROC curve (AUC; range 0.5 to 1) using logistic regression to assess how well a score distinguished stroke etiology and report the sensitivity and specificity computed for each score level (table 3). Results Final analysis included 27 LAA patients and 43 CE patients. Based on clinical criteria a score >3 reflects a high probability of CE etiology while a score <3 reflects a high probability of LAA etiology (70.4% sensitivity, 86.0% specificity). Meanwhile, radiographic imaging criteria scores >3 reflects a high probability of CE while scores <3 reflects a high probability of LAA (100% sensitivity, 86.0% specificity). Conclusion Our scoring system reliably differentiates between CE and LAA LVO. This data requires validation with a larger dataset of patients including those with LVO in other territories and may be of great utility for decisions regarding MT technique and devices if found reliable in a prospective study. (Table Presented).

Volume

16

First Page

A80

Last Page

A81

Share

COinS