Abstract DP381: CHIRP ³ M _-1 Is Superior to Prior Stroke Risk Prediction Scores in Discriminating Cardioembolic Stroke From Large Artery Stroke

Introduction: There is no existing prediction system to distinguish risk of cardioembolic stroke (CES) from large artery stroke (LAS). CHIRP 3 M -1 score is a novel surrogate for reduced left atrial appendage (LAA) emptying velocity. It assigns points for coronary artery disease (1 point), heart failure (1 point), left atrial volume index ≥ 42 mL/m 2 (1 point), rhythm atrial fibrillation (AF, 1 point), history of paroxysmal AF (2 points), persistent AF (3 points), longstanding/permanent AF (4 points), and subtracts 1 point for more than moderate mitral regurgitation. We evaluated the performance of the CHIRP 3 M -1 score as a marker of CES risk. Methods: We conducted a retrospective case-control study using The University of Kansas Medical Center stroke registry to identify patients with neurologist-classified CES or LAS admitted in calendar year 2021. Data collection was supplemented through medial chart review. Cases with CES were compared to controls with LAS. Differences between the CHIRP 3 M -1 , CHIRP 3 , CHA 2 DS 2 -VASc, CHADS 2 , R 2 CHADS 2 and ATRIA scores were examined using Mann-Whitney U tests. Discriminative performance of these scores was compared using area under the receiver operating characteristic curves (AUCs) curves. Results: Among the 341 patients (age 70.2 ± 12.3 years, 153 [44.9%] females) included for analysis, 189 (55.4%) had CES. The mean clinical stroke risk scores for CES vs. LAS as shown in Table 1 and Figure 1 were as follows: CHIRP 3 M -1 (3.4 ± 3.6 vs. 0.9 ± 1.4, p < 0.0001), CHA 2 DS 2 -VASc (3.6 ± 1.5 vs. 3.1 ± 1.6, p = 0.0089) and R 2 CHADS 2 (2.6 ± 1.4 vs. 2.3 ± 1.5, p = 0.0667). The CHIRP 3 M -1 score demonstrated superior discrimination (AUC 0.725) compared to the CHA 2 DS 2 -VASc score (AUC 0.581) in differentiating CES from LAS ( Figure 2 ). A CHIRP 3 M -1 score of ≥ 2 points had a sensitivity of 55.6% and specificity of 78.5% in distinguishing CES from LAS. Conclusions: The CHIRP 3 M -1 score is superior to existing clinical stroke risk models like CHA 2 DS 2 -VASc score in discriminating CES from LAS. Therefore, CHIRP 3 M -1 may be better suited to identify patients for CES mitigation using oral anticoagulation or LAA occlusion.