Antibiotics were administered to newborns at low risk for early-onset sepsis as frequently as to newborns with EOS risk factors, based on data from approximately 7,500 infants.
EOS remains a significant cause of morbidity and mortality, and predicting which newborns are at risk remains a challenge for neonatal care that often drives high rates of antibiotic use, Dustin D. Flannery, DO, of Children’s Hospital of Philadelphia and colleagues wrote.
Antibiotic exposures are associated with short- and long-term adverse effects in both preterm and term infants, which highlights the need for improved risk assessment in this population, the researchers said.
“A robust estimate of EOS risk in relation to delivery characteristics among infants of all gestational ages at birth could significantly contribute to newborn clinical management by identifying newborns unlikely to benefit from empirical antibiotic therapy,” they emphasized.
In a study published in Pediatrics, the researchers identified 7,540 infants born between Jan. 1, 2009, and Dec. 31, 2014, at two high-risk perinatal units in Philadelphia. Gestational age ranged from 22 to 43 weeks. Criteria for low risk of EOS were determined via an algorithm that included cesarean delivery (with or without labor or membrane rupture), and no antepartum concerns for intra-amniotic infection or nonreassuring fetal status.
A total of 6,428 infants did not meet the low-risk criteria; another 1,121 infants met the low-risk criteria. The primary outcome of EOS was defined as growth of a pathogen in at least 1 blood and/or cerebrospinal fluid culture obtained at 72 hours or less after birth. Overall, 41 infants who did not meet the low-risk criteria developed EOS; none of the infants who met the low-risk criteria developed EOS. Secondary outcomes included initiation of empirical antibiotics at 72 hours or less after birth and the duration of antibiotic use.
Although fewer low-risk infants received antibiotics, compared with infants with EOS (80.4% vs. 91.0%, P < .001), the duration of antibiotic use was not significantly different between the groups, with an adjusted difference of 0.6 hours.
Among infants who did not meet low-risk criteria, 157 were started on antibiotics for each case of EOS, the researchers noted in their discussion of the findings. “Because no cases of EOS were identified in the low-risk group, this proportion could not be calculated but suggests that antibiotic exposure in this group was disproportionately higher for incidence of EOS.”
The study findings were limited by several factors including the possible lack of generalizability to other centers and the use of data from a period before more refined EOS strategies, the researchers noted. Other limitations include the inability to assess the effect of lab results on antibiotic use, a lack of data on the exact indication for delivery, and potential misclassification bias.
Risk assessment tools should not be used alone, but should be used to inform clinical decision-making, the researchers emphasized. However, the results were strengthened by the inclusion of moderately preterm infants, who are rarely studied, and the clinical utility of the risk algorithm used in the study. “The implications of our study include potential adjustments to sepsis risk assessment in term infants, and confirmation and enhancement of previous studies that identify a subset of lower-risk preterm infants,” who may be spared empirical or prolonged antibiotic exposure, they concluded.