Pediatric ASPECTS predicts outcomes following acute symptomatic neonatal arterial stroke

Objective: To test the hypothesis that the Alberta Stroke Program Early Computed Tomography Score (ASPECTS) is useful in determining outcomes after neonatal arterial ischemic stroke (NAIS), we assessed accuracy of the modified pediatric ASPECTS (pedASPECTS) to predict cerebral palsy (CP), neurologic impairment, and epilepsy.

Methods: Cross-sectional study included newborns with acute NAIS whose outcomes were assessed at ≥18 months after stroke. PedASPECTS accuracy to predict outcomes was determined by sensitivity, specificity, and receiver operator characteristic (ROC) curves, and correlation between pedASPECTS and infarct volume was determined by the Spearman correlation coefficient.

Results: Ninety-six children met the inclusion criteria. Median percentage infarct to supratentorial brain volume was 6.8% (interquartile range [IQR] 3.0%-14.3%). Median pedASPECTS was 7 (IQR 4-10). At a median age of 2.1 years, 35% developed CP, 43% had neurologic impairment, and 7% had epilepsy. Median pedASPECTS predicted outcomes of interest: CP (10, IQR 8-12) vs no CP (5, IQR 4-8) (p < 0.0001), poor (9, IQR 7-12) vs good (6, IQR 4-8) neurologic outcomes (p < 0.0001), and epilepsy (10, IQR 8-12) vs no epilepsy (7, IQR 4-10) (p = 0.033). PedASPECTS accuracy was good for CP (ROC 0.811) and fair for neurologic impairment (ROC 0.760) and epilepsy (ROC 0.761). A pedASPECTS ≥8 had ≥69% sensitivity and ≥54% specificity for clinical outcomes. PedASPECTS correlated with infarct volume (Spearman rank 0.701, p < 0.0001).

Conclusions: This study provides Class II evidence that pedASPECTS has fair to good accuracy for predicting CP, neurologic impairment, and epilepsy after NAIS and correlates with infarct volume. PedASPECTS may assist with early identification of babies requiring close developmental surveillance.