The use of molecular techniques for bacterial detection in the analysis of gastric aspirates collected from infants on the first day of life.

Abstract Prospective service evaluation of the utility of molecular methods to analyse neonatal gastric aspirate specimens in a single neonatal unit and associated maternity unit. 43 newborn infants investigated for sepsis with median gestational age of 39 weeks (range 31–41 weeks) and median birth weight 3050 grams (range 1250–4220 g). Gastric aspirates routinely collected within 12 h of birth were analysed using conventional and molecular methods for bacterial detection, bacterial DNA load and sequencing to identified bacterial species. Results Bacterial DNA loads varied from 0.03 to 1736 pg/µl of DNA extract (1 µl of DNA extract equivalent to 4 µl gastric aspirate). Bacteria were identified in 30/43 (70%) of samples by molecular methods and 10/43 (23.3%) of samples by culture. Cultures were only positive when the bacterial DNA exceeded 4.5 pg/µl of extract. Infants with prolonged rupture of membranes (> 24 h prior to delivery) had a DNA load on average 23 times higher than those without (95%CI 3.7 to 141; p  = 0.001). Additional bacteria detected by molecular methods included many species that are fastidious and potentially pathogenic including Leptotrichia spp. , Serratia spp ., Ureaplasma spp . , Veillonella spp ., Haemophilus influenzae and Group B Streptococcus . Due to a low rate of adverse outcomes it was not possible to correlate bacterial identifications or DNA load with infant outcome. Conclusions Molecular methods can identify bacteria from a greater proportion of gastric aspirate specimens that conventional culture. Further work is required to establish whether this information can be used to improve infant outcomes.