Uncovering the novel Enterobacter cloacae complex species responsible for septic shock deaths in newborns: a cohort...

Enterobacter cloacae complex contains nosocomial pathogens responsible for infection outbreaks. Identification at the species level within the E cloacae complex remains difficult. Using whole genome sequencing, our aim was to decipher the transmission routes that led to the death of six of seven neonates who had bacteraemia caused by E cloacae complex isolates in a neonatal intensive care unit (NICU) over a 13 month period.

In this cohort study, E cloacae complex isolates were taken from 186 newborns in an NICU: 14 were clinical samples (eg, blood cultures), 728 rectal swab samples, and 38 environmental samples (20 from siphons, 18 from incubators, and one from a mattress). The samples were analysed to decipher the possible role of manual cross transmission or environmental source in an outbreak of fatal septic shocks related to E cloacae complex bacteraemia. After the replacement of the incubators suspected to be the reservoir of some outbreak-related isolates on Feb 1, 2018, E cloacae complex strains were screened again for 10 months (503 rectal swab samples from 163 newborns). The 71 E cloacae complex isolates recovered from screening, clinical, and environmental samples across both study periods were compared by whole genome sequencing. The pathogenicity of E cloacae complex isolates responsible for fatal septic shocks was assessed using a Galleria mellonella in-vivo model.

From Dec 9, 2016, to Jan 31, 2018, 249 (34%) of 728 rectal swab samples were positive for E cloacae complex, with 66 (35%) of 186 newborns colonised. E cloacae complex were also recovered from four (20%) of 20 siphons and 11 (61%) of 18 incubators. During this period, whole genome sequencing identified that the isolates responsible for the six fatal septic shocks were all Enterobacter bugandensis. A G mellonella infection model showed a higher virulence of E bugandensis. Genomic analysis highlighted the role of incubators as a long-term reservoir and source of cross contamination, leading to their replacement on Feb 1, 2018. Following incubator replacement, a 10-month follow-up investigation identified a physiological gut colonisation with polyclonal E cloacae complex in 52 (34%) of 163 neonates within a median of 9 days (5-14), but no E cloacae complex-related septic shocks.

Despite around 30% of neonates being physiologically colonised with E cloacae complex, fatal sepsis was systematically linked with bacteraemia caused by E bugandensis. Our findings highlight the need for accurate identification methods to detect the hypervirulent species within the E cloacae complex recovered in neonates.

Assistance Publique-Hôpitaux de Paris.