To identify perinatal factors in children born extremely preterm (EP) that were associated with motor impairment (MI) at 2 and 10years of age and develop a predictive algorithm to estimate the risk of MI during childhood. Participants of the Extremely Low Gestational Age Newborns Study (ELGANS) were classified as: no MI, MI only at 2years, MI only at 10years, and MI at both 2 and 10years, based on a standardized neurological examination at 2 and the Gross Motor Function Classification System (GMFCS) at 10years of age. Least Absolute Shrinkage and Selection Operator (LASSO) regression was used to develop the final predictive model. Of the 849 study participants, 64 (7.5%) had a diagnosis of MI at both 2 and 10years and 63 (7.4%) had a diagnosis of MI at 1 visit but not the other. Of 22 total risk factors queried, 4 variables most reliably and accurately predicted MI: gestational age, weight z-score growth trajectory during neonatal intensive care unit (NICU) stay, ventriculomegaly, and cerebral echolucency on head ultrasound. By selecting probability thresholds of 3.5% and 7.0% at ages 2 and 10, respectively, likelihood of developing MI can be predicted with a sensitivity and specificity of 71.2%/72.1% at age 2 and 70.7%/70.7% at age10. In our cohort, the diagnosis of MI at 2years did not always predict a diagnosis of MI at 10years. Specific risk factors are predictive of MI and can estimate an individual infant's risk at NICU discharge of MI at age 10years.
Read full abstract