I thank Dr Willoughby for his comments and careful reading of our article.1 We did our best to control for multiple confounding variables. However, we concur with Dr Willoughby that there may have been other variables, such as precipitous delivery, that could not be controlled for and thus may have accounted for some of the variability in cystic periventricular leukomalacia (PVL). We agree that longer-term follow-up, looking for outcomes such as cerebral palsy, would be ideal. However, we disagree with Dr Willoughby's sentiment that cystic echolucencies are clinically unimportant. Cystic PVL has been associated with a substantial risk of later CP or motor delay.2- 4 Thus, although some infants with cystic PVL may ultimately be free of later motor disability, cystic PVL remains a clinically significant marker of early brain injury. We would like to reemphasize that our data show an association between antenatal antibiotics and reduced risk of later cystic PVL, and not proof of protection. We agree with Dr Willoughby that our data do not provide for an expanded indication for antimicrobial use in pregnancy and that randomized trials are needed prior to any initiation of antenatal antibiotic treatment strategies to decrease the occurrence of neonatal brain injury.
Coleman MM, Leef KH, Tuttle D, Stefano JL, Paul D. Beware Observational Studies in Neonatal Practice—Reply. Arch Pediatr Adolesc Med. 2003;157(9):935. doi:10.1001/archpedi.157.9.934-a