Incidence, risk factors and outcomes for neonatal sepsis in The Gambia: Descriptive cohort analysis embedded in a randomised clinical trial

BackgroundNeonatal sepsis is a major contributor to adverse neonatal outcomes in West Africa. High quality data on risk factors for sepsis in this region is limited, yet important for surveillance, prevention and management of newborns at greatest risk. This study aimed to describe the clinical epidemiology of sepsis for health-facility born neonates in an urban West African setting. MethodsThis study comprised secondary data analyses from the Gambian cohort of the PregnAnZI-2 randomised clinical trial. Relatively healthy pregnant women and their offspring who delivered at two urban health facilities were included, with a combination of active and passive surveillance until 28 days. Neonatal sepsis was defined as suspected if clinical and laboratory (abnormal FBC or CRP) criteria were met, and confirmed if blood-culture was positive. A novel conceptual framework informed logistic regression models to identify 1) factors associated with neonatal sepsis and 2) contribution of sepsis towards neonatal mortality. ResultsA total of 6515 neonates were included. The health-facility based incidence of neonatal sepsis was 20.2 cases/1000 live births (N=131 cases), predominantly early-onset (<3 days)(15.7 cases/1000 livebirths). Confirmed sepsis accounted for 22% (29/131) of all cases, with Burkholderia cepacia and Staphylococcus aureus the most prevalent bacteria in 24% (7/29) of confirmed sepsis each. Risk factors for sepsis included low 1-minute Apgar score (aOR 13.2, 95% CI 8.40-20.73), pre-labour maternal fever (aOR 5.0, 95% CI 1.15-21.69), easily recognisable congenital malformation (aOR 3.39, 95% CI 1.55-7.38) and low-birth weight (aOR 2.85, 95% CI 1.75-4.65). 40.7% of all neonatal deaths in the cohort occurred in neonates with sepsis, with 40-fold increase in mortality compared to neonates without sepsis (OR: 39.98, 95% CI 22.5 - 71.1). ConclusionSepsis, especially early onset, is a major morbidity for health facility born neonates delivered following relatively healthy pregnancy in urban Gambia, with high associated mortality. We identify neonatal phenotypes (low birth weight, newborns with low 1-minute apgar scores, or those with easily recognisable congenital malformations) who may benefit from enhanced postnatal surveillance or antibiotics to prevent or treat neonatal sepsis and reduce neonatal mortality. Trial RegistrationNCT03199547: Clinicaltrials.gov. Registered on 23rd June 2017