Abstract
Malignant brain edema (MBE) is a life-threatening complication of large hemispheric infarction (LHI). While early decompressive hemicraniectomy has been shown to improve outcomes in selected patients, identifying those at risk for MBE remains a clinical challenge. This study aimed to develop a simple and practical scoring system based on non-contrast CT imaging obtained approximately 24 h after stroke onset to support early prediction and timely intervention. We conducted a retrospective cohort study of LHI patients who underwent non-contrast CT imaging within 24 ± 8 h of symptom onset. Radiographic parameters assessed included infarct volume (ABC/2 method), ASPECTS, anterior or posterior cerebral artery territory involvement, temporal lobe involvement, midline shift, lateral ventricle compression, intercaudate distance, Sylvian fissure effacement and sulcal effacement. Multivariate logistic regression identified independent predictors of MBE. A weighted scoring system was then developed based on the regression model and internally validated using ROC analysis. Among 62 patients included, 38 (61.3%) developed MBE. Infarct volume and midline shift were independently associated with MBE. A scoring system (SIriraj Malignant Brain Edema Prediction by Lesion Volume and Edema Shift – SIMPLE) was derived using infarct volume ≥ 106 mL (two points) and midline shift ≥ 1.6 mm (one point). The score demonstrated excellent discriminative performance (AUC 0.951, 95% CI: 0.899–1.002). A cutoff score ≥ 2 yielded 89.5% sensitivity and 91.7% specificity. A dual-threshold strategy, using score 0 to rule out and score three to rule in, achieved 97.4% sensitivity and 100% specificity, providing clear clinical guidance for observation, medical management, or early neurosurgical intervention. The SIMPLE score is a rapid and reliable tool using routine CT imaging to predict MBE in LHI patients at an optimal 24-hour timepoint. Its simplicity and strong predictive accuracy may assist clinicians in early risk stratification and timely treatment decisions. Further prospective validation is recommended.