Electroencephalography (EEG) and amplitude-integrated EEG (aEEG) have been utilized to predict neurodevelopmental impairment (NDI) in preterm infants.

MEDLINE, SCOPUS, Embase, Web of Science, and Cochrane Library were searched from inception until November 1, 2023. Observational studies evaluating EEG and aEEG and neurodevelopment assessment performed at least after six months corrected age (CA) were included. A diagnostic test accuracy (DTA) meta-analysis using a Bayesian random-effects bivariate model was performed. QUADAS-2 was used to assess the risk of bias, and GRADE approach was used to ascertain the certainty of evidence.

Thirty-eight studies (n = 4667) were included in the systematic review of which 26 studies (n = 3363) were synthesized in a DTA meta-analysis. EEG and aEEG had a sensitivity of 63.9% (95% credible intervals [CrI]: 52.1%, 76.3%) and specificity of 87.6% (80.9%, 94.1%) for the outcome of any NDI (<1 S.D. from the mean value for the CA). For the outcome of severe NDI (<2 S.D.s from the mean value for the CA) (10 studies, n = 1075), the sensitivity was 59.7% (42.3%, 75.9%) and specificity was 85.2% (72.7%, 92.4%). Evidence certainty was very low to low. Subgroup analysis of studies that had utilized Bayley Scales of Infant and Toddler Development between 18 and 30 months CA (12 studies, n = 1202) for neurodevelopmental assessment revealed a sensitivity of 59% (42.3%, 74.7%) and specificity of 83.7% (75.8%, 90.1%) for any NDI.

EEG and aEEG possibly have suboptimal sensitivity and an acceptable-to-high specificity in predicting any NDI and severe NDI in preterm infants.