Neurosonographic abnormalities associated with maternal history of cocaine use in neonates of appropriate size for their gestational age.

PURPOSE To determine whether increased incidence of neurosonographic abnormalities (predominantly of the basal ganglia and thalamus) in cocaine-exposed neonates who are small for their gestational age is attributable to the cocaine or to neonatal size.

METHODS Neonates whose sizes were appropriate for their gestational age with no evidence of hypoxia or respiratory distress were identified prospectively by a maternal history of cocaine use. Scans were performed within 72 hours of birth using a 7.5-MHz transducer following a standard protocol. The images were analyzed without access to patient information. Forty study neonates were compared with 34 control subjects who were appropriate in size for their gestational age, scanned using the same protocol. Comparisons were made using Fisher Exact Test, t test, and logistic regression.

RESULTS No control infant had neurosonographic abnormalities. In the study group, gestational age ranged from 27 to 41 weeks. Of the 40 study neonates, 14 (35%) had one neurosonographic abnormality; two had two abnormalities. The predominant lesion was focal echolucencies, mainly in the area of the basal ganglia (10 of 40, 25%). Other findings were caudate echogenicity (3 of 40, 7.5%), ventricular dilation (2 of 40, 5%) and one "moth-eaten" appearance of the thalamus. Lesions were more likely approaching term and were not related to prematurity or alcohol use.

CONCLUSION Apparently normal neonates with a maternal history of cocaine use are likely to have degenerative changes or focal infarctions in their basal ganglia attributable to cocaine. Neurosonography should be used to evaluate these neonates. The long-term significance of these lesions needs further evaluation.