The Rapid Occluded MCA Vessel Etiology (ROME) Score - Identifying the Etiology of Large Vessel Occlusions of the Middle Cerebral Artery.

BackgroundDifferentiating between intra-cerebral atherosclerotic disease (ICAD) and non-ICAD large vessel occlusion (LVO) is crucial for selecting the appropriate mechanical thrombectomy (MT) technique and device. We developed an algorithm to predict LVO etiology using clinical and radiographic features in the emergent setting. MethodsWe conducted a retrospective chart review of middle cerebral artery (MCA) occlusions treated with MT and confirmed as ICAD or non-ICAD. We recorded common risk factors and radiographic features from CT angiography to identify significant differences between groups. These factors were used in a multivariable logistic regression to create the algorithm. The ROME score was then tested against the ABC2D algorithm for predicting ICAD LVO in a prospective cohort. ResultsThe analysis included 33 ICAD and 327 non-ICAD LVO strokes. ICAD LVO patients were less likely to have atrial fibrillation (9.1% vs 53.8%; [points: 4]) or systolic heart failure with EF[≤]35% (9.1% vs 27.8%; [points: 1) and more likely to present with progressive or fluctuating symptoms (21.2% vs 4.6%; [points: 1). ICAD patients had a higher incidence of multi-vessel atherosclerotic disease (84.8% vs 37%; [points: 1]), tapered appearance of occlusion (60.6% vs 0.9%; [points: 6]), and extra-cranial ICA atherosclerotic plaque with high-risk features (plaques with lengths [≥]1cm or thickness >3mm perpendicular to the long axis of the artery with associated ulceration or with soft plaque component (87.9% vs 37.6%; [points: 4]). AUC for the ROME score was 0.9666 with the highest sensitivity (97%) and specificity (88%) at a cut-off of 9. In the prospective cohort of 201 patients, the ROME score showed 81.3% sensitivity and 98.8% specificity, while the ABC2D score showed 90.6% sensitivity and 50.3% specificity. ConclusionOur scoring system effectively differentiates between ICAD and non-ICAD LVO, with greater specificity than the ABC2D score. Future steps will include validation in external databases and clinical trials.