Background and objective Neonatal sepsis is a severe disease in preterm infants that can
result in significant morbidity in this vulnerable patient group. Our aim was to identify and
assess the risk factors and incidence of neonatal sepsis in two different groups of preterm
infants (very preterm-and moderate to late preterm infants) in a level 2 neonatal ward.
Methods We conducted a retrospective cohort study in preterm infants with a gestational age
(GA) of <36weeks transferred to or born in our centre from 2009 to 2012. Sepsis was defined
as a positive blood culture and clinical symptoms, with early -and late onset sepsis defined as
onset of symptoms before 72 hours and after 72 hours of life, respectively. Univariate and
multivariate logistic analyses were performed.
Results We included 163 very preterm infants (GA<32weeks) and 434 moderate-to late
preterm infants (GA 32-36weeks). Neonatal sepsis was found in 6% (36/597) of preterm
infants. There was a significant association between the type of group a neonate belonged to
(according to GA) and whether or not sepsis was diagnosed (p < 0.002). The odds of sepsis
for neonates were 2.7 times higher if they were born with a GA <32 weeks (group A) than if
they were born with a GA 32- 36 weeks (group B). Early-onset sepsis (EOS) with group B
streptococcus occurred in one late preterm infant (0.7%) of 146 suspected episodes. Lateonset
sepsis (LOS) occurred in 35 preterm infants of a total of 90 suspected episodes (39%).
Coagulase-negative staphylococci was the most frequently pathogen isolated in LOS. Total
parenteral nutrition administration, vaginal spontaneous birth and gestational age were
significantly associated with LOS.
Conclusion This study shows a very low prevalence of EOS and LOS in both very preterm
and moderate preterm infants. Vaginal spontaneous birth was found to be a significant
protective factor for LOS. TPN administration (by central venous catheter, peripheral inserted
catheter), vaginal spontaneous birth and lower GA were found to be significant risk factors
for development of LOS.
|Please use this identifier to cite or
link to this item:
© 2003-2007 RUG : The University of Groningen holds the intellectual property rights of this repository. All rights reserved.Powered by WildFire