March 16, 2017
2 min read

Algorithm reduces lab testing, antibiotic treatment for sepsis among newborns

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact

The use of clinical care algorithms estimating the risk of early-onset sepsis for individual infants reduced the percentage of newborns who had to undergo laboratory testing and receive empirical antibiotic treatment without noticeable adverse effects.

“[The current] guidelines are based on epidemiologic data obtained before the widespread obstetric use of intrapartum antibiotic prophylaxis — when early-onset sepsis incidence was 5- to 10-fold higher than currently observed. These guidelines result in a large percentage — 15% to 20% — of term and later preterm infants being evaluated for sepsis, with 5% to 8% receiving empirical antibiotics,” Michael W. Kuzniewicz, MD, MPH, and colleagues wrote. “Persistent high rates of evaluation and treatment contrast with the decreasing incidence of early-onset sepsis — 0.3 to 0.8 cases per 1000 births.”

Michael W. Kuzniewicz

The researchers aimed to observe the outcome of neonatal early-onset sepsis risk prediction models on infants who receive sepsis evaluations and antibiotic use, as well as to assess the models’ safety within a large, integrated health care system. The study included 204,485 infants born at Kaiser Permanente Northern California Hospital between Jan. 1, 2010, and Dec. 31, 2015. All participants were born at 35 weeks’ gestation or later.

Kuzniewicz and colleagues compared three periods of time. The baseline period (Jan. 1, 2010-Nov. 31, 2012) was based on national recommended guidelines for sepsis management. The learning period (Dec. 1, 2012-June 30, 2014) was based on sepsis management of multivariable estimates of sepsis risk at birth, and the early-onset sepsis calculator period (July 1, 2014-Dec. 31, 2015) was based on multivariable risk estimate combined with the infant’s clinical condition in the first 24 hours after birth.

The baseline, learning and early-onset sepsis calculator periods included 95,343, 52,881 and 56,261 infants, respectively. When the baseline period was compared with the sepsis calculator period, the use of blood cultures decreased from 14.5% to 4.9%. The administration of empirical antibiotics within 24 hours after birth decreased from 5.0% to 2.6%, and antibiotic use within 24 and 72 hours after birth also decreased from 0.5% to 0.4%. In all periods, the occurrence of culture-confirmed sepsis was comparable (0.3% during baseline, 0.3% during learning and 0.2% during early-onset sepsis calculator). The chance of readmissions regarding sepsis within 7 days of birth was infrequent. Out of 100,000 births, 5.2, 1.9 and 5.3 readmissions were seen within the baseline, learning and early-onset sepsis calculator periods, respectively. Undesirable clinical outcomes, including the need for inotropes, mechanical ventilation, meningitis and death, saw no change in chance of occurrence after the early-onset sepsis calculator was implemented.  

“Although the use of predictive analytics is garnering increased attention in the scientific literature, use of patient-specific, multivariable sepsis risk estimates to guide the care of newborns represents a significant shift from the current recommended practice in neonatology,” Kuzniewicz and colleagues wrote. “Our work provides prospective validation of the efficacy and safety of this approach.” – by Katherine Bortz

Disclosure: The researchers report no relevant financial disclosures.