RT Journal A1 Willoughby, Jr RE T1 BEware observational studies in neonatal practice JF Archives of Pediatrics & Adolescent Medicine JO Archives of Pediatrics & Adolescent Medicine YR 2003 FD September 1 VO 157 IS 9 SP 934 OP 935 DO 10.1001/archpedi.157.9.934-a UL http://dx.doi.org/10.1001/archpedi.157.9.934-a AB Paul et al1 showed, in an observational study, an apparent protective effect of antepartum antibiotics on the outcome of cystic lesions detected by cranial ultrasound. I wish to underscore the concluding recommendation of the authors that their results should not alter clinical practice. Cystic echolucencies are imperfect predictors of cerebral palsy (CP), the outcome of ultimate clinical interest. Any discussion must also consider the association of maternal antibiotic use with CP. There is no conclusive evidence that antimicrobial therapy, outside its use for treating proven infections, alters that risk. Early-onset sepsis is rarely the cause of CP. There is no consensus that chorioamnionitis is associated with later CP in preterm infants when other risks are taken into account.2