Analysis from the NeoOBS Global Neonatal Sepsis Prospective Observational Cohort Study Across 19 Hospitals in 11 Countries; Clinical Presentation, Treatment, Mortality Outcomes and Development of the NeoSEP Sepsis Severity Score

Background: Neonatal sepsis is associated with high mortality, but there is limited clinical data on outcomes in hospital facilities in low and middle income countries (LMIC). The NeoOBS study aimed to determine the clinical features, pathogens and clinical outcomes of infants presenting with neonatal sepsis in hospitals globally. Methods: Infants <60 days of age, fulfilling at least 2 sepsis criteria were enrolled across 19 sites in 11 countries, with prospective data collection to 28 days. A multivariable Cox regression model was built to predict mortality and develop a sepsis severity score in 85% of infants (randomly selected) and validated in 15%. Findings: 3204 infants were enrolled between 2018-2020; median (IQR) postnatal age was 5 (2-15) days, birth weight 2500g (1400-3000g), with 4 (2-5) clinical and 1 (0-2) laboratory sepsis criteria, and 69% hospitalised since birth. Overall mortality was 11.3% (95%CI 10.2-12.5%; n=350), but higher in infants with culture positive blood stream infection at baseline (17.7%, 95%CI 14.7-21.1%, n=564) than culture negative (n=2631, 9.9%, 95%CI 8.8-11.2%). Among 309 infection-related deaths, 59% (n=183) occurred in healthcare-associated infections (HAI). Klebsiella pneumoniae was the most common pathogen identified (n=132, 23.4% of baseline cultures). A wide variety of antibiotic regimens were prescribed, with frequent use of carbapenems. A sepsis severity score was developed using infants’ characteristics, supportive care, and clinical signs. In the validation sample, mortality (95%CI) was 1% (0-5%), 6% (4-10%), and 28% (19-39%) in low (score 0-3), medium (4-6) and high (7-14) risk groups, respectively. Interpretation: High rates of HAI, mortality and carbapenem use were observed in neonatal units globally. Novel strategic treatment and prevention trials are urgently required, where the NeoSEP score may have utility. Funding Information: The study was sponsored by the Global Antibiotic Research and Development Partnership (GARDP) with additional funding from: Bill and Melinda Gates Foundation, German Federal Ministry of Education and Research (BMBF), South African Medical Research Council, UK Department of Health and Social Care, Monaco and in-kind contribution from Indian Council for Medical Research. Declaration of Interests: None to declare. Ethics Approval Statement: Ethical approval was obtained both from St. George’s, University of London Research Ethics Committee and sites’ local, central or national ethics committees and other relevant local bodies, where required.