Written by Joshua Belfer
Spoon Feed
In a cohort of febrile infants, the performance of WBC, ANC, and CRP decreased in patients with less than 2 hours of fever; PCT remained similar.
How long does it take for inflammatory markers to rise?
Using a single center cohort of febrile infants ≤90 days old, a secondary analysis on 2,565 infants analyzed the performance of biomarkers and of the PECARN, AAP, and Step-by-Step clinical rules based on the duration of fever (<2 hours, 2-12, >12). Overall, 24.7% of febrile infants were seen within 2 hours of the fever, accounting for 24 of the 76 total cases of invasive bacterial infection (IBI, bacteremia and meningitis).
When comparing the area under the curve (AUC) for biomarkers in the three subgroups, the worst performance was seen for WBC, ANC, and CRP in the fever <2 hour group. Procalcitonin (PCT) had a similar AUC for all three groups analyzed. Each of the three clinical rules had the lowest sensitivity in patients with <2 hours of fever. Interestingly, the highest sensitivity was seen for the Step-by-Step rule in the 2-12 hour group, and for the PECARN and AAP rules in the ≥12 hour group.
As with any single center study, generalization of results must be cautioned. Additionally, the degree of vigilance of parents of the included infants may be different than other hospitals or regions of the country; any discrepancy between the duration of fever reported by parents and the “true” duration would confound the results of this study.
How will this change my practice?
In febrile infants, the serious consequences of missing bacteremia or meningitis stresses the importance of high sensitivity biomarkers. Next time I see a well appearing, febrile infant within a few hours of fever onset, I will hesitate if the inflammatory markers return reassuring. I’m not sure I have enough information from this single center study to consider repeating labs in this situation, but I’d consider even closer pediatrician follow up and stricter return precautions than I would already typically advise. With mind to the risks of keeping these children in the hospital longer, one could also consider a longer period of observation (whether that’s in the ER or inpatient).
Source
Performance of Febrile Infant Algorithms by Duration of Fever. Pediatrics. 2024 May 1;153(5):e2023064342. doi: 10.1542/peds.2023-064342. PMID: 38563061.
