Neonatal Sepsis: Prevalence and Contributing Factors Among Neonates Admitted to the Special Care Unit at Kawempe National Referral Hospital, Kampala, Uganda.

Abstract

Background: Despite recent declines in under-5 mortality, neonatal sepsis rates remain high, especially in low- and middle-income countries. This study examined the prevalence and risk factors of neonatal sepsis at Kawempe National Referral Hospital (KNRH), Uganda's primary maternal and neonatal care centre.

Methods: A cross-sectional survey was conducted in KNRH's special care unit from October 2021 to December 2021. Sociodemographic and clinical data were collected from mothers and patient case records of consecutively sampled neonates using an interviewer-administered electronic questionnaire. Neonatal sepsis was defined based on diagnosis in the patient case record, regardless of bacteriological confirmation, and classified as early-onset (diagnosed within 72 hours) or late-onset (after 72 hours). Logistic regression was used to identify factors associated with neonatal sepsis.

Results: Out of 265 neonates enrolled, 56.8% were boys with a median age of 4 days (interquartile range = 2-7). Half (51.1%) were born pre-term, and most (71.3%) were delivered vaginally. Hospital deliveries were predominant (70.9%), while 25.7% occurred in peripheral clinics and 3.4% at home. One-third of the mothers reported experiencing fever (pyrexia) around the time of delivery. The prevalence of neonatal sepsis was 35.8%, with 62.1% of the cases being late-onset. Significant factors associated with neonatal sepsis included delivery from a peripheral clinic (aOR = 2.2, 95% CI = 1.2-4.1, P = .010), maternal perinatal pyrexia (aOR = 3.4, 95% CI = 1.8, 6.3, P < .001), meconium stained liquor (aOR = 2.4, 95% CI = 1.3, 4.6, P = .005), poor cord care (aOR = 7.4, 95% CI = 1.3, 42.3, P = .025), and prelacteal feeding (aOR = 3, 95% CI = 1.4, 6.7, P = .007) while neonates delivered by caesarean section (aOR = 0.4, 95% CI = 0.2, 09, P = .02) and those who initiated breastfeeding in the first hour of life (aOR = 0.1, 95% CI = 0.1, 0.2, P < .001) were associated with lower chances of getting neonatal sepsis.

Conclusions: The prevalence of neonatal sepsis at KNRH is high, with the majority of cases being late-onset. Mothers should be educated on treating fevers early during pregnancy, and antenatal care should focus on hygiene, sanitation, and nutrition. Strengthening infection prevention and control practices in health facility settings could reduce late-onset neonatal sepsis, especially in peripheral clinics.