Relationship between postinterventional cerebral hyperdensities and malignant brain edema in patients with acute ischemic stroke after mechanical thrombectomy

This study aimed to evaluate the predictive value of postinterventional cerebral hyperdensities (PCHDs) for malignant brain edema (MBE) in acute ischemic stroke (AIS) patients after mechanical thrombectomy (MT). We sought to establish and validate a nomogram for predicting MBE in this population. This study included patients with acute anterior circulation large vessel occlusion stroke treated with MT at our hospital from May 2017 to July 2024. PCHDs were classified based on their distribution characteristics and extent. A multivariate logistic regression analysis was used to assess the predictive value of PCHDs subtypes for MBE. Receiver operating characteristic (ROC) curve analysis identified optimal predictive thresholds. Least absolute shrinkage and selection operator regression was applied to select significant predictors of MBE to construct a nomogram. The nomogram's performance was evaluated using the C-index, calibration plots, and decision curve analysis. Among 516 enrolled patients, 126 (24.4%) developed MBE. The cortex sign [adjusted odds ratio (OR) = 6.290, 95% confidence interval (CI): 2.581-15.329], basal ganglia sign (adjusted OR = 4.081, 95% CI: 1.831-9.096), and combined sign (adjusted OR = 8.295, 95% CI: 3.942-17.454) were independent risk factors for MBE. Higher PCHDs scores were inversely associated with MBE risk (adjusted OR = 0.620, 95% CI: 0.534-0.719). The nomogram incorporating age, atrial fibrillation, baseline National Institutes of Health Stroke Scale (NIHSS) score, occlusion site, white blood cell (WBC) count, total cholesterol level, and PCHDs score demonstrated good discrimination (C-index: 0.904) and calibration (Hosmer-Lemeshow test, p = 0.851) to predict MBE. PCHDs show a strong association with MBE in AIS patients. Our nomogram provides individualized prediction of post-MT MBE risk; however, multicenter validation is warranted.