Infections bactériennes néonatales précoces en maternité : peut-on limiter les prélèvements bactériologiques périphériques en salle de naissance ?

Early-onset neonatal sepsis in the Paris area: a population-based surveillance study from 2019 to 2021

BACKGROUND

Early-onset neonatal sepsis (EOS) is a rare condition but an important cause of severe morbidity and mortality in neonates.

METHODS

This is a prospective observational study in neonates born at ≥34 weeks of gestation (WG). The primary endpoint was EOS, defined by isolation of pathogenic species from blood culture and/or cerebrospinal fluid culture within 72 hours after birth. Data on EOS were collected exhaustively from all maternity wards in Paris area (April 2019-March 2021).

RESULTS

108 EOS were recorded (annual incidence, 0.32 per 1000 live births; 95% CI 0.26 to 0.38). In term infants, the most frequent pathogens were group B Streptococcus (GBS) (n=47) and Escherichia coli (n=20); in late preterm infants, the most frequent pathogens were E. coli (n=15) and GBS (n=7). Fifteen meningitis cases were diagnosed. Five E. coli strains (14%) were resistant to both amoxicillin and gentamicin, which is an empiric treatment for EOS. Of the 54 infants with GBS infections, 35 were born from mothers with negative GBS prepartum screening test and 8 from mothers with no screening. Two deaths were reported, both in term infants (Proteus mirabilis and E. coli).

CONCLUSION

In neonates ≥34 WG born in the Paris area, GBS was twice as frequent as E. coli in term infants. EOS was six times more frequent in late preterm than in term infants and was due to E. coli in 60% of cases. Prevention of GBS EOS and empiric antibiotic treatment of EOS could be improved.

[Value of gastric fluid to start antibiotics in premature babies suspected of having early neonatal bacterial infection]

The objective of this study was to evaluate the value of direct examination and culture of gastric fluid in the treatment of early neonatal bacterial infections (INBP) in pre-term infants.

MATERIALS AND METHODS

Observational study conducted over 6 months in a Type III center. All hospitalized premature babies who had routine gastric fluid sampling at birth during the period of the study were included. They were classified into two groups: premature infants with probable or suspected infection and treated as such (Group 1) and premature infants with no infection or only having colonization (Group 2).

RESULTS AND DISCUSSION

In total, 255 pre-term infants were included in the study. Group 1 consisted of 127 newborns and group 2 consisted of 128 newborns. The direct gastric fluid examination was positive in 51 newborns in Group 1 and in 46 newborns in group 2. The culture was positive in 25 newborns in group 1 and eight newborns in group 2. Direct examination of gastric fluid of the 255 children studied had low sensitivity (40.1%) and low specificity (64%) of INBP, with 52.6% positive predictive value (PPV) and 51.8% negative predictive value (NPV). The gastric fluid culture was specific (93.7%) of the INBP, sensitivity was low (19.6%), with PPV at 75.7% and NPV at 54%.

CONCLUSION

These results undermine the relevance of the direct examination of gastric fluid in the delicate diagnosis of INBP. This direct examination has a low PPV and NPV. It is advisable not to start or stop antibiotic therapy solely on this argument; however, it can guide the choice of antibiotic therapy and remains useful for this reason. The culture of gastric fluid has very good specificity (93.7%).

[Diagnostic value of suspicion criteria for early-onset neonatal bacterial infection: report ten years after the Anaes recommendations]

BACKGROUND

Because clinical symptoms and biological markers are neither sensitive nor specific, newborns are frequently suspected of having an infection. In France, 30-50% of newborns are suspected of having early-onset sepsis (EOS) and many of them undergo laboratory tests and empirical antibiotic treatments while awaiting results. The aim of this study was to evaluate the diagnostic value of various suspicion criteria for EOS as recommended by the Anaes since 2002, and the value of umbilical cord blood procalcitonin (PCT), currently assayed in our maternity ward.

MATERIAL AND METHODS

This 4-year retrospective study in the CHU of Nantes included hospitalized newborns with suspected early neonatal infection. Infection status was established according to the Anaes definitions and clinical evolution.

RESULTS

The study included 2151 newborns. Among anamnestic criteria, only prematurity significantly increased the risk of EOS (relative risk of 3.1; 95% CI 1.4-7.0). The relative risk of infection for a symptomatic newborn was 12.2 (95% CI 4.9-30.2; P<0.0001). Laboratory test results were the most predictive criteria. The relative risk to be infected was 291.6 (95% CI 70.7-1,214.0; P<0.0001) with a blood cord PCT value>0.6 ng/L. The positive post-test probability was 28% (95% CI: 23-33) and the negative post-test probability was close to 0 (95% CI: 0-0).

CONCLUSION

Clinical criteria of postnatal life adaptation are more predictive of early-onset neonatal infection than anamnestic criteria are. The blood cord PCT value could be a helpful marker in the identification of infected newborns. PCT measured in umbilical cord blood could be included in a general algorithm in order to identify as soon as possible newborns with a high risk of EOS.