Neonatal bacterial infection remains a serious pathology in industrialized countries despite the use of prophylaxis measures for group B streptococcus (GBS) (peri-partum antibiotic in women with GBS colonization), which was implemented in the United States in 1996 and in France in 2001 and has led to a dramatic decrease in the incidence of neonatal bacterial infections. However, early onset neonatal infection (EONI), which is defined as an infection occurring during the first 6 days after birth (as opposed to late onset neonatal infections (LONI) occurring between days 7-89), is still one of the leading causes of neonatal morbidity and mortality. Physicians consider EONI a significant diagnostic and therapeutic emergency due to the potential for sudden onset and rapid evolution of sepsis in newborns with immature immune systems. Currently, in France, detection of EONI is based on national consensus guidelines published in 2002 (ANAES recommendations). There are broad indications to provide empirical antibiotic treatment pending diagnostic confirmation through different complementary exams. To ensure that every infected newborn is diagnosed, biological assessments are often repeated and result in the use of invasive and painful procedures, anemia and financial concerns. Moreover, in cases of abnormal biological results, many newborns are subjected to intravenous (IV) antibiotic treatments requiring hospitalization and separation from their mother. However recent studies have shown that antibiotics can have a potentially deleterious effect on the neonatal digestive microbiota and result in the appearance of antibiotic-resistant bacteria, with possible long-term consequences on the health of the child.

Procalcitonin (PCT) is a calcitonin prohormone secreted from the parenchymal tissues. This marker of inflammation has been shown to be a valuable diagnostic marker for bacterial infection in adults and in children. It also seems to be a reliable marker for neonatal bacterial infection, which would make it useful in the detection of EONI. Because physiological levels of PCT vary during the first days of life, possibly due to postnatal intestinal bacterial colonization, levels of this marker are difficult to interpret in the early neonatal period. However, in a study of 2151 newborns with suspected EONI, Nicolas Joram et al. found that PCT obtained from the umbilical blood cord, prior to newborn intestinal colonization, bypasses this postnatal physiological peak of PCT and effectively constitutes a discriminant marker to distinguish between infected and healthy infants using a cutoff value of 0.6 ng/ml.

Subsequent to this pilot study, several studies on PCT in umbilical blood cord confirmed its good diagnostic performance for EONI, particularly when included in a diagnostic algorithm. This marker could contribute to a better estimation of EONI risk in order to limit the use of unnecessary complementary exams and prescription of antibiotics and their associated short- and long-term side effects in healthy newborns.

Therefore, in this study, the investigators propose to test the diagnostic value of a PCT-based algorithm in newborns suspected of having EONI. The investigators hypothesize that this algorithm is as efficient as those currently used (ANAES), but will limit coinciding biological exams and exposure to antibiotics during the neonatal period.

Inclusion Criteria:

• All children born at > 36 weeks gestation in one of the 15 participating maternity or neonatology units and suspected having EONI according to the ANAES recommendations (clinical suspicion of chorioamnionitis, intrapartum maternal fever > 38°C, infected twin, spontaneous premature delivery at < 37 gestational weeks, prolonged rupture of membrane for > 12 hours, maternal group B Streptococcus colonization without full prophylactic antibiotic treatment, or signs of fetal asphyxia) will be included in the study.
• Oral Consent (Non Opposition).

Exclusion Criteria:

• Newborns will be considered ineligible if:
• Parental non opposition is not obtained, if the parents do not speak French, present severe dementia and/or cannot be reached on day 6.
• Nosocomial neonatal infection, severe congenital malformation or obstetrically explained neonatal asphyxia are diagnosed.
• Secondary parental opposition.

Children included in the DIACORD study will be born at > 36 weeks gestation in one of the 15 participating maternity or neonatology units and suspected of EOSI according to the ANAES recommendations (clinical suspicion of chorioamnionitis; intrapartum maternal fever > 38°C, infected twin, spontaneous premature delivery at < 37 gestational weeks, prolonged rupture of membrane for > 12 hours, maternal colonization with group B Streptococcus without full prophylactic antibiotic treatment, or signs of fetal asphyxia).