The burden of inflammatory stress and antibiotic treatment in preterm neonates and neuro-developmental outcome

Abstract

Objective: To evaluate the possible association between antibiotic treatment in preterm infants and the occurrence of adverse radiologic and neurologic outcomes. We hypothesize that antibiotic treatment may represent inflammation and thus be associated with adverse outcomes. Methods: Preterm infants (≤ 36 weeks gestational age), hospitalized in the Neonatal Intensive Care Unit (NICU) were eligible for inclusion. Follow-up performed by a pediatric neurologist included assessment by the Griffiths Mental Developmental Scales (GMDS–ER). NICU admission history was retrospectively reviewed. Extended antibiotic treatment was defined as treatment for a duration greater than 72 hours. Infants were further categorized by the number of antibiotic treatment courses given. Adverse neurologic outcome was defined as DQ < 55. Adverse radiological outcome included any of the following: intraventricular hemorrhage grade 2 or above, periventricular leukomalacia or post-hemorrhagic hydrocephalus. Statistical analysis included univariate and multivariate analyses. Results: Included were 203 patients. Adverse radiologic and neurologic outcomes were present in 46 of 203 (22.7%) and 21 of 203 (10.3 %) preterm infants, respectively. Maternal infection or antibiotic treatment were associated significantly with adverse radiological outcome, as were extended antibiotic therapy or multiple antibiotic treatment courses (p<0.001). Unadjusted analysis showed that extended antibiotic therapy (OR 5.03, 95% CI, 2.23, 11.33) and repeated antibiotic courses (two courses OR 2.81, 95% CI, 1.1, 7.8, three courses or more OR 9.44, 95% CI, 3.50, 25.49) were associated with adverse radiological outcome, A non-significant trend remained after adjustment to other known risk factors. Among 46 neonates with adverse radiologic outcome, 16 (34.8%) had an adverse neurological outcome. The most significant risk factors were NICU length of stay, duration of mechanical ventilation and culture-proven sepsis. Conclusion: Our sample despite its small size, suggests an association between prolonged antibiotic treatment and neonatal brain injury. These findings need further investigation using larger cohorts.