Predicting clinical outcome in posterior circulation large-vessel occlusion patients with endovascular recanalisation: the GNC score.

Abstract

Background and purpose: Acute ischaemic strokes caused by posterior circulation large-vessel occlusions (pc-LVOs) are associated with particularly poor prognoses, including significant disability and mortality rates. This study sought to develop and validate a novel scoring system for predicting functional outcomes in pc-LVO cases following successful endovascular recanalisation.

Methods: We derived a predictive model from the DETECT-China cohort and externally validated it using the DETECT2-China dataset. Poor outcome was defined as a modified Rankin Scale score of 4-6 at 90 days. Cerebral circulation time (CCT), measured via digital subtraction angiography (DSA), served as a key predictor. Multivariable logistic regression was employed to construct the scoring system.

Results: The training cohort comprised 92 patients, of whom 52 (56.5%) experienced poor outcomes. Multivariate analysis identified prolonged CCT (adjusted OR (aOR) 1.365; 95% CI 1.105 to 1.686; p=0.004), elevated admission National Institutes of Health Stroke Scale (NIHSS) (aOR 1.235; 95% CI 1.120 to 1.363; p<0.001) and higher blood glucose levels (aOR 1.345; 95% CI 1.023 to 1.769; p=0.034) as independent predictors. These variables were integrated into the GNC score (Glucose-NIHSS-CCT). The GNC score demonstrated excellent predictive performance for clinical outcome, good discrimination and calibration in this cohort, as well as the bootstrap validation. Importantly, the excellent performance of this score was further validated in DETECT2-China.

Conclusions: This is the first report that CCT based on DSA is an independent prognostic marker in pc-LVO patients with successful recanalisation post-endovascular treatment. The GNC score, incorporating readily available clinical and angiographic parameters, offers a reliable tool for outcome prediction in this high-risk population.