Background EEG features and prediction of cognitive outcomes in very preterm infants: A systematic review

Objectives: Very preterm infants are at risk of cognitive impairment, but current capacity to predict at-risk infants is sub-optimal. Electroencephalography (EEG) has been used to assess brain function in development. This review investigates the relationship between EEG and cognitive outcomes in very preterm infants. Methods: Two reviewers independently conducted a literature search in April 2018 using PubMed, CINAHL, PsycINFO, Cochrane Library, Embase and Web of Science. Studies included very preterm infants (born <= 34 weeks gestational age, GA) who were assessed with EEG at <= 43 weeks postmenstrual age (PMA) and had cognitive outcomes assessed >= 3 months of age. Data on the subjects, EEG, cognitive assessment, and main findings were extracted. Meta-analysis was undertaken to calculate pooled sensitivity and specificity. Results: 31 studies (n = 4712 very preterm infants) met the inclusion criteria. The age of EEG, length of EEG recording, EEG features analysed, age at follow-up, and follow-up assessments were diverse. The included studies were then divided into categories based on their analysed EEG feature(s) for meta-analysis. Only one category had an adequate number of studies for meta-analysis: four papers (n = 255 very preterm infants) reporting dysmature/disorganised EEG patterns were meta-analysed and the pooled sensitivity and specificity for predicting cognitive outcomes were 0.63 (95% CI: 0.53-0.72) and 0.83 (95% CI: 0.74-0.89) respectively. Conclusions: There is preliminary evidence that background EEG features can predict cognitive outcomes in very preterm infants. Reported findings were however too heterogeneous to determine which EEG features are best at predicting cognitive outcome.