Risk Factors for Neonatal Sepsis and Perinatal Death Among Infants Enrolled in the Prevention of Perinatal Sepsis Trial, Soweto, South Africa

Background: Factors associated with neonatal sepsis, an important cause of child mortality, are poorly described in Africa. We characterized factors associated with early-onset (days 0-2 of life) and late-onset (days 3-28) -sepsis and perinatal death among infants enrolled in the Prevention of Perinatal Sepsis Trial (NCT00136370 at ClinicalTrials.gov), Soweto, South Africa.

Methods: Secondary analysis of 8011 enrolled mothers and their neonates. Prenatal and labor records were abstracted and neonatal wards were monitored for hospitalized Prevention of Perinatal Sepsis-enrolled neonates. Endpoint definitions required clinical and laboratory signs. All univariate factors associated with endpoints at P < 0.15 were evaluated using multivariable logistic regression.

Results: About 10.5% (837/8011) of women received intrapartum antibiotic prophylaxis; 3.8% of enrolled versus 15% of hospital births were preterm. Among 8129 infants, 289 had early-onset sepsis, 34 had late-onset sepsis, 49 had culture-confirmed neonatal sepsis and 71 died in the perinatal period. Factors associated with early-onset sepsis included preterm delivery [adjusted relative risk (aRR) = 2.6; 95% confidence interval (CI): 1.4-4.8]; low birth weight (<1500 g: aRR = 6.5, 95% CI: 2.4-17.3); meconium-stained amniotic fluid (MSAF) (aRR = 2.8, 95% CI: 2.2-3.7) and first birth (aRR = 1.8; 95% CI: 1.4-2.3). Preterm, low birth weight, MSAF and first birth were similarly associated with perinatal death and culture-confirmed sepsis. MSAF (aRR = 2.4, 95% CI: 1.1-5.0) was associated with late-onset sepsis.

Conclusions: Preterm and low birth weight were important sepsis risk factors. MSAF and first birth were also associated with sepsis and death, warranting further exploration. Intrapartum antibiotic prophylaxis did not protect against all-cause sepsis or death, underscoring the need for alternate prevention strategies.