Clinical Risk Factors and In-Hospital Scoring Systems for Detection of Large Vessel Occlusion in Anterior Circulation Acute Ischemic Stroke: A Comparative Study

Background: Anterior circulation large vessel occlusion (LVO) accounts for a substantial proportion of morbidity in acute ischemic stroke and requires rapid identification to enable timely endovascular therapy. Bedside clinical scales may aid early detection, particularly where immediate vascular imaging is limited. Hence, we aimed to evaluate clinical risk factors for anterior circulation LVO and compare the diagnostic performance of the National Institutes of Health Stroke Scale (NIHSS) and the GAI₂AA (Gaze deviation, aphasia, inattention, arm weakness, atrial fibrillation) score in an in-hospital acute ischemic stroke cohort. Methodology: This observational study included 168 consecutive adults with acute ischemic stroke admitted to a tertiary care centre between January 2025 and October 2025. All patients underwent baseline NIHSS assessment and vascular imaging with CT or MR angiography. GAI₂AA scores were derived from documented clinical findings. Diagnostic accuracy metrics and receiver operating characteristic (ROC) analyses were performed. Results: LVO was identified in 55 patients (33%). LVO was associated with older age as well as a history of smoking and prior cerebrovascular events. At optimal cutoffs, GAI₂AA ≥3 demonstrated sensitivity of 78% and specificity of 81%, outperforming NIHSS ≥8, which showed higher sensitivity (87%) but lower specificity (58%). GAI₂AA had superior discriminative ability (AUC 0.92 vs. 0.78). Gaze deviation was the single best discriminant (Odds ratio12.7 [95% Confidence intervals 5.8-27.8]). Conclusions: The GAI2AA score showed higher overall accuracy than NIHSS for predicting anterior circulation LVO and may serve as a practical, rapid bedside tool to guide triage and imaging decisions in resource-constrained environments, though prospective multicenter validation is needed before widespread adoption.