A comparison of early neonatal deaths among preterm infants with term neonatal deaths at the University Teaching Hospital, Lusaka, Zambia

Abstract

Background: Prematurity is a common complication that contributes  significantly to high neonatal mortality. In spite of many efforts by the government and other partners, non-significant decline has been achieved in the recent past. Globally, 15 million babies are born preterm (<37 weeks gestation) each year, and more than 1 million of those do not survive their first month of life. Preterm birth accounts for 75% of all perinatal mortality in some series thus identifying the determinants of preterm deaths is very crucial for policy improvement. This study was aimed at establishing factors associated with preterm deaths at UTH compared to those of term neonatal deaths.

Methods: A case-control study was conducted among 208 neonates that were early neonatal deaths i.e. within 7 days in neonatal intensive care unit (NICU) at UTH in 2015. Antenatal and intrapartum details (parity, multiple pregnancy, birth weight, antenatal steroid exposure, antibiotic exposure, and the indication of admission to NICU) were obtained from 104 neonates that were preterm (between 24- 36 completed weeks  gestation) and had died and of a further 104 term neonates (>37 weeks gestation) that died around the same time. The data was collected by int e rvi ewe r- administered structured questionnaire and analyzed by SPSS v21. Bivariate analysis was used to identify variables for multivariate logistic regression model to identify obstetric determinants amongst deaths in neonates that were preterm compared to those born at term.

Results: There were few differences between the two groups. The sex of the neonate significantly influenced the odds of dying. We confirmed that male neonates had a 57.1% higher risk than females (42.9%) of dying during the early neonatal period. More term neonates that died were male (P=0.0031) and had a very poor Apgar score (1-3) (P=0.0048). Both the indications for admission to NICU and cause of death were different in the two groups with preterms (P<0.0001) and terms P=0.0309. On
multivariate regression analysis, poor Apgar score was associated with six-fold odds of RDS. More preterm neonates had died despite receiving steroids. None of the other factors reached statistical significance (adjOR 6.0, 95% CI 3.03-11.92, p<0.0001). Poor Apgar score was also the only factor associated with sepsis, though it was a neonate with a good Apgar score that had higher odds of dying due to sepsis. Primiparity was associated with a 2.6-fold odds (95% CI 1.03 to 6.68, p=0.04) of hypoxic ischaemic encephalopathy. On logistic regression, a preterm neonate dying only had a higher odds of being a LBW (<2500g) than any other factor [adjusted OR 132.72 (95% CI 39.49 to 387.66) P<00001]. Considering the main causes of death, hypoxic ischemic encephalopathy in preterm neonates was only associated with poor Apgar score (i.e. <7) [adjusted OR 2.03 (95% CI 1.12 to 3.67) P =0.02]. Sepsis in term neonates OR 0.2 (95% CI 0.15 to 0.54) P<00001]. Respiratory distress syndrome in preterm neonates dying was only associated with poor Apgar score [adjusted OR 6.01 (3.03 to 11.92) P<00001].

Conclusions: Hypoxic ischemic encephalopathy as a cause of early neonatal death is commoner in term neonates but also common in preterm. Sepsis is commoner in preterm neonates as a cause of early neonatal death. Comparing different causes of death, poor Apgar score featured in all cases calling for improved resuscitation.