Intérêt des prélèvements périnataux dans la prise en charge des nouveau-nés suspects d'infections bactériennes précoces

Early-onset neonatal diseases (EOND) require optimal diagnostic and therapeutic strategy to prevent severe (including neurological…) and life threatening complications notably in preterm neonates. The early diagnosis of EOND is challenging because of non-specific clinical signs by the newborn. So the decision to treat can only be taken on a range of biological, clinical and anamnestic arguments. Therefore, many newborns were treated by an empiric antibiotic ; this is not without consequences for their digestive colonization, the immune response of the gastrointestinal tract and thereby the duration of their hospitalization. Since the last guideline of ANAES in 2002, the type of bacteriological samples to conduct in the newborn at birth remains controversial and varies greatly from hospital. Blood culture is currently the gold standard for the diagnosis of sepsis. However, in addition to the fact that culture reports are available only after 48-72 hours, blood cultures are frequently false negative due to the small amount of blood that can be drawn from neonates. Non central samples (ear, gastric fluid…) have limited value to diagnose EOND and the contribution of multiple peripheral sampling remains to be demonstrated. However non central sampling may help to isolate the pathogen and to determine its antibiotics susceptibilities. Molecular diagnoses based on acid nucleic detection are still under assessment. Biochemical marker such as umbilical blood cord procalcitonin has a promising bacterial infection diagnostic value in neonates with a better diagnosis performance than CRP, limiting complementary exams and exposure to antibiotics. New guidelines are desirable for 2015 to optimize and standardize the diagnosis of EOND in France.