Procalcitonin at 12-36 hours of fever for prediction of invasive bacterial infections in hospitalized febrile neonates

Anne-Sophie Romain; Romain Guedj; Anais Chosidow; Nicolas Mediamolle; Aurélie Schnuriger; Sophie Vimont; Charlène Ferrandiz; Nicolas Robin; Marie-Hélène Odièvre; Emmanuel Grimprel; Mathie Lorrot

doi:10.3389/fped.2022.968207

Procalcitonin at 12-36 hours of fever for prediction of invasive bacterial infections in hospitalized febrile neonates

Front Pediatr. 2022 Sep 29:10:968207. doi: 10.3389/fped.2022.968207. eCollection 2022.

Authors

Affiliations

¹ Department of General Pediatrics, Trousseau Hospital, Assistance Publique-Hôpitaux de Paris (AP-HP), Sorbonne Université, Paris, France.
² Department of Pediatrics Emergency, Trousseau Hospital, Assistance Publique-Hôpitaux de Paris (AP-HP), Sorbonne Université, Paris, France.
³ Perinatal and Pediatric Epidemiology Research Team (EPOPé), Centre de Recherche Épidémiologie et Statistique Sorbonne Paris Cité (CRESS) INSERM UMR1153, Paris, France.
⁴ Department of Virology, Trousseau Hospital, Assistance Publique-Hôpitaux de Paris (AP-HP), Sorbonne Université, Paris, France.
⁵ Sorbonne Université, INSERM UMR_S 938, Centre de Recherche Saint Antoine (CRSA), Paris, France.
⁶ Department of Bacteriology, Saint Antoine Hospital, Assistance Publique-Hôpitaux de Paris (AP-HP), Sorbonne Université, Paris, France.
⁷ INSERM UMR_S 1155, Hôpital Tenon, Paris, France.
⁸ Biologie intégrée du globule rouge, UMR_S1134, INSERM, Université de Paris, Paris, France.

Abstract

Aim: We aimed to investigate the performance of procalcitonin (PCT) assay between 12 and 36 h after onset of fever (PCT H12-H36) to predict invasive bacterial infection (IBI) (ie, meningitis and/or bacteremia) in febrile neonates.

Methods: We retrospectively included all febrile neonates hospitalized in the general pediatric department in a teaching hospital from January 2013 to December 2019. PCT assay ≤ 0.6 ng/ml was defined as negative. The primary outcome was to study the performance of PCT H12-H36 to predict IBI.

Results: Out of 385 included neonates, IBI was ascertainable for 357 neonates (92.7%). We found 16 IBI: 3 meningitis and 13 bacteremia. Sensitivity and specificity of PCT H12-H36 in the identification of IBI were, respectively, 100% [95% CI 82.9-100%] and 71.8% [95% CI 66.8-76.6%], with positive and negative predictive values of 14.3% [95% CI 8.4-22.2%] and 100% [95% CI 98.8-100%] respectively. Of the 259 neonates who had a PCT assay within the first 12 h of fever (< H12) and a PCT assay after H12-H36, 8 had IBI. Two of these 8 neonates had a negative < H12 PCT but a positive H12-H36 PCT.

Conclusions: PCT H12-H36 did not miss any IBI whereas < H12 PCT could missed IBI diagnoses. PCT H12-H36 might be included in clinical decision rule to help physicians to stop early antibiotics in febrile neonates.

Keywords: antibiotics; fever; invasive bacterial infection; neonates; procalcitonin.