Editor’s Note: Dr. Ella Perrin (she/her/hers) is a resident physician in pediatrics at Naval Medical Center San Diego. Her interests include disordered eating, obesity, decreasing weight stigma and bias, and the field of hospital medicine. The views expressed in this blog are those of the author and do not necessarily reflect the official policy or position of the Department of the Navy, Department of Defense, or the US Government. Dr. Perrin declares no conflicts of interest. -Rachel Y. Moon, MD, Associate Editor, Digital Media, Pediatrics
Fever in a well-appearing neonate is a common chief complaint in the pediatric emergency department (ED). The majority of these patients are febrile due to viral infection, but it is critical that invasive bacterial infection (IBI), including bacteremia and meningitis, is not missed, as the outcomes from untreated bacterial infection can be devastating. However, management with hospitalization and intravenous antibiotics is not without risk, including exposure to other infectious organisms and increased stress on new families.
There are several evidence-based clinical decision tools that use patient age and lab values to estimate the risk of IBI. These tools (for example, PECARN, AAP, and Step-by-Step guidelines) allow medical providers to better identify low-risk neonates who are appropriate for outpatient management. They hinge on the reliability of inflammatory markers, including:
- C-reactive protein (CRP)
- Procalcitonin
- White blood cell count (WBC)
- Absolute neutrophil count (ANC)
Over time, the epidemiology of the well-appearing febrile neonate has changed. Patients are presenting to the ED earlier in the course of the disease. Because biomarkers take several hours to increase, this earlier presentation makes the biomarkers less reliable.
In an article being early released in Pediatrics this week, entitled “Performance of Febrile Infant Algorithms by Duration of Fever,” Roberto Velasco, MD, from Hospital Universitari Parc Tauli in Spain and pediatric emergency medicine colleagues in Spain analyzed the performance of the above inflammatory markers, depending on duration of fever, in identifying those well-appearing febrile infants younger than 90 days of age who had IBI (10.1542/peds.2023-064342).
The data in this study came from a prospective registry of 2,565 infants younger than 90 days old with fever who were seen in one pediatric ED in Spain between 2008 and 2021.
There were several remarkable findings:
- From 2008 to 2021, the rate of infants who were evaluated within the first 2 hours of fever almost doubled.
- The performance of WBC, ANC, and CRP in predicting IBI decreased significantly in those who were evaluated within 2 hours of fever.
- Procalcitonin maintained equivalent sensitivity (detection ability) for IBI at <2 hours of fever.
- With fever <2 hours, procalcitonin of ≥0.5 ng/ml (the traditional cutoff point) had a sensitivity of 37.5% for IBI. Sensitivity increased to 100% at ≥0.14 ng/ml.
- Procalcitonin of ≥0.14 ng/ml alone was a better predictor of IBI than the PECARN, AAP, or Step-by-Step guidelines in this group.
This study highlights an important flaw in current pathways for the management of well-appearing febrile neonates, as available clinical tools may have a decreased sensitivity when duration of fever is <2 hours. The presence of normal inflammatory markers may be less reassuring in this group of infants.
The study also brings up a potential solution—a procalcitonin level of ≥0.14 ng/ml may be a more appropriate risk-stratification tool in infants with short duration of fever.
Neonatal IBI is a diagnosis no clinician wants to miss. As such, clinicians should be willing to accept little risk in managing them. I urge you to read this article (and watch the accompanying video abstract) and incorporate its findings into your decision-making process for the management of neonates with short duration of fever.
