Laurence Generenaz (France)
Joint Unit Hospices Civils de Lyon/Biomérieux Hospices Civils de LyonAuthor Of 1 Presentation
A SINGLE HOST BIOMARKER COULD HELP IN REDUCING UNJUSTIFIED ANTIBIOTIC TREATMENTS IN HOSPITALIZED NEONATES
Abstract
Backgrounds:
Late-onset neonatal sepsis (LOS) is frequently observed in Neonatal Intensive Care Units (NICUs) and potentially severe. Despite its high prevalence, diagnostic remains difficult. Biological markers (CRP and PCT) had low performance at LOS onset. Blood culture results are too late and lack sensitivity. These difficulties lead to antibiotic overuse in hospitalized newborns, resulting in an increased antibiotic resistance, microbiota modification, neonatal complications. Our aim was to identify a biomarker combination to early exclude the diagnosis of LOS in newborn with suggestive clinical signs.
Methods
We conducted a prospective, multicenter cohort study (EMERAUDE) (NCT03299751). Hospitalized preterm neonates of at least 7 days with signs of suspected LOS were enrolled. Serum samples were collected at the time of the venipuncture prescribed for standard care. We assessed the performance of 11 protein markers using Simple PlexTM technology (Protein simple©, CA, USA). An adjudication committee, assigned each patient to one category among “confirmed infection”, “infirmed infection” or “undetermined infection”.
Results:
234 patients were enrolled and 230 had analyzable samples. The patients were mainly preterm (80%) with a median gestational age of 27 weeks, a median birth weight of 940 grams. 22% had a confirmed infection and all of them received antibiotics. Among the infirmed infection group, 27% received antibiotics. Among all analysed biomarkers IL10 had the best performance, with an area under the curve [IC95%] of 0.845[0.777-0.914]. Using optimal threshold value, we highlighted that more than a half of unjustified antibiotics in non-infected neonates could be avoided.
Conclusions/Learning Points:
At the onset of clinical suspicion of LOS, additional biomarkers could help the clinician in its choice of prescribing or not antibiotics. Validation studies will be performed to assess the interest of IL10 biomarker in a clinical decision rule.