According to a retrospective cohort study, antibiotics were given to most neonates with blood cultures, even those at low risk of early sepsis (EOS).
Among more than 7,500 neonates whose cultures were performed at two Philadelphia perinatal units, 80.4% of those at low risk of developing EOS received antibiotics, compared to 91% of neonates not considered at risk. low risk (P
In addition, the duration of antibiotic administration was not significantly different between those born with and those born without low-risk birth characteristics (adjusted difference 0.6 hours, 95% CI -3, 8 to 5.1), they noted in Pediatrics.
“Delivery characteristics can help providers predict which infants are at the lowest risk of EOS and determine if empiric antibiotic therapy is indicated,” they wrote.
Antibiotic use in premature and very low birth weight infants is associated with necrotizing enterocolitis, a disease in which bacteria invade the intestinal tissue. On the other hand, EOS – although rare – can be fatal, especially for premature babies, with a study noting a mortality rate of nearly 30%.
“Sepsis remains one of the most feared events in any neonatal setting,” wrote Ivana Culic, MD, of Beth Israel Deaconess Medical Center in Boston, and Amy O’Connell, MD, PhD, of Boston Children’s Hospital, in a accompanying editorial. “Practitioners who have seen a newborn fall ill with sepsis have good reason to be cautious; neonatal sepsis can progress rapidly and can be fatal even after initiation of appropriate antibiotics.”
A previous study by Flannery and colleagues found that antibiotic use in very low birth weight and extremely low birth weight infants who were considered to be at low risk of early infection could be reduced without adverse effects.
But methods for assessing the risk of infection in newborns need to be refined, the study authors noted. Currently, there are algorithms to predict the risk of EOS in term infants. However, studies of preterm infants examining criteria for low-risk delivery generally do not include moderately preterm or late preterm infants.
Of 53,575 births at two Philadelphia perinatal units from 2009 to 2014, this study evaluated the 7,549 infants (14.4%) of all gestational ages who underwent blood cultures – of which 1,091 were considered at low risk for ‘EOS. The median gestational age was 37 weeks, the median birth weight was 2859 grams, and 43.9% were girls.
Forty-one infants developed EOS. In this group, the median gestational age was 35 weeks, the median birth weight was 2,415 grams, and 68% were girls. None were considered low-risk deliveries.
Flannery and her colleagues called a birth low risk if it met all of the following criteria:
- Cesarean birth
- Rupture of amniotic membranes during childbirth
- Absence of labor or attempts to induce labor
- Absence of suspected or confirmed maternal intra-amniotic infection
- Absence of non-reassuring unexplained acute fetal status
Two infants with EOS met most criteria except for the absence of unexplained and non-reassuring acute fetal status. Sixteen cases were due to Escherichia coliand 16 were due to group B Streptococcus.
However, a single set of criteria should not be applied to infants of all gestational ages, Culic and O’Connell cautioned.
“As clinicians, we all recognize that the risks and outcomes of sepsis are different for the population of extremely preterm versus term infants,” they noted.
Although application of the EOS risk calculator may reduce antibiotic use in term infants, “application of the same risk assessment tool in preterm infants would undoubtedly increase antibiotic use. as their initial clinical signs and symptoms (temperature instability, need for respiratory support, and need for inotropic support) are likely to be present,” they wrote.
Flannery and colleagues acknowledged that more refined risk assessment strategies for EOS have been implemented since their study.
Flannery reported research funding from the Agency for Health Care Research and Quality under two contracts with the CDC and Children’s Hospital of Philadelphia. The co-authors reported research funding from the NIH and the Eunice Kennedy Shriver National Institute of Child Health and Human Development.
Culic and O’Connell reported no financial information.