Abstract
This study aims to estimate the incidence, prevalence of morbidity and mortality rate (MR), associated risks, and the applicability of neonatal sequential organ failure assessment (nSOFA), of neonatal sepsis from a database of all hospitalized infants in a representative sub-provincial region in China. An observational study was retrospectively conducted based on an established perinatal cohort database from 59,056 livebirths linking all the hospitalized neonates (n = 7,960), from January to December, 2015, in Huai'an, Jiangsu Province. Their clinical conditions and outcomes were followed up within postnatal 28 days for term, and postmenstrual 44 weeks for preterm, infants. There were 1,225 cases of neonatal sepsis diagnosed (15.4% of hospitalized, incidence of 20.7 per 1,000 livebirths), with 95 all-cause-related in-hospital deaths (MR, 7.8% for the cases, 1.6 per 1,000 livebirths). For 755 (61.6%) hospitalized at Huai'an Women and Children's Hospital (HWCH), 29.9% developed severe sepsis and 6.4% septic shock, with corresponding MR 16.4% and 64.6%, and the median [interquartile range] of nSOFA scoring as 3 [2, 5] and 9 [5,10], respectively. The area under receiver operating characteristic curves of nSFOA predicting sepsis mortality in those as total, clinical and culture-proven sepsis were 0.80-0.85, with preterm and term ones being 0.79 and 0.88. By multiple Poisson regression models in patients in HWCH, birthweight <1500 g, severe hypoxemic respiratory failure and congenital anomalies were associated with increased, whereas inborn and multiple use of antibiotics decreased, death risks. Increment of nSOFA scores was associated with increasing MR (relative risk 1.21, 95% confidence interval, 1.15, 1.27). This study depicted the incidence, prevalence and MR of neonatal sepsis of Huai'an, and highlighted the eligibility of nSOFA for outcome prediction, in a representative emerging region with expanding advanced neonatal intensive care capabilities.