Neurosonograms in very preterm neonates: Findings, risk factors, and association with neurodevelopmental outcome: A retrospective study

Abstract

Neurosonogram (NSG) is a widely used, noninvasive bedside investigation in neonatal intensive care. Yet, actual data on the proportions of abnormal findings, specific conclusions, and its ability to predict outcomes are sparse. We aimed to study these aspects in very preterm infants who are a high-risk group. This retrospective descriptive study aimed at analyzing reports of postnatal age-specific NSG reports done in very preterm neonates, and their associations with risk factors. Follow-up information prospectively recorded by the child development services was analyzed for the association of NSG abnormalities with neurodevelopmental impairment (NDI). Clinical and radiological details of 323 very preterm neonates were analyzed. Abnormal NSG (defined as ≥Grade 3 intraventricular hemorrhage/≥Grade 2 periventricular leukomalacia/ventriculomegaly) was reported in 42 infants (13%) at or less than term postmenstrual age; 8% of babies at ≤2 weeks postnatal age; and 12% at 36–40 weeks. Choroid plexus cyst was the most common “other” findings. Gestational age <28 weeks, chorioamnionitis, extensive resuscitation, ventilation >24 h, and culture-proven sepsis were independent risk factors for abnormal NSG in the first 2 weeks of life. Of the babies followed up, 19.25% babies had NDI. The sensitivity of any abnormal NSG was only 16% and the specificity was 91% to predict NDI. The negative predictive value of abnormal NSG for NDI was 79.5%. In very preterm, any abnormal NSG (till term postmenstrual age) was reported in 13% of very preterm. The ability to predict NDI using abnormal NSG reports has a test accuracy of 73.8%.