Procalcitonin and Other Inflammatory Markers in Febrile Infants Aged 60 Days or Younger

Febrile infants aged 60 days or younger are at risk of bacteremia and bacterial meningitis, collectively termed invasive bacterial infections (IBIs).1 Clinical practice guidelines from the American Academy of Pediatrics (AAP) recommend lumbar puncture (LP), empirical antibiotics, and hospitalization for all febrile infants aged 8 to 21 days. Infants aged 22 to 60 days should be risk stratified using inflammatory markers (IMs),1 of which procalcitonin (PCT) offers the best diagnostic accuracy for IBIs.2 The AAP guideline initially recommended that “PCT should be obtained along with absolute neutrophil count (ANC) or C-reactive protein (CRP).” However, this was subsequently changed to “PCT should be obtained along with ANC.”1 This latter PCT+ANC recommendation closely resembles the Pediatric Emergency Care Applied Research Network (PECARN) prediction rule, which also includes urinalysis (UA) results but was not derived for IBIs specifically.3 No study has compared the diagnostic accuracy of PCT+CRP vs PCT+ANC for identification of IBIs among febrile young infants.We conducted a secondary analysis of prospective quality improvement data for all febrile infants aged 60 days or younger at a tertiary pediatric emergency department (ED) between June 2020 and September 2024.4 Standardized clinical, laboratory, and telephone follow-up data were collected for all infants with a rectal temperature of 38.0 °C or higher measured at home or ED triage. Nearly all infants had IMs, UA, and blood and urine cultures; LP was at the discretion of the treating physician. The primary outcome of IBI (bacteremia or bacterial meningitis) was ascertained for all infants by review of microbiologic culture results and telephone follow-up more than 14 days postdischarge. IBIs and urinary tract infections (UTIs) were defined according to current definitions.1,3,4This was a complete case analysis of previously healthy, term-born infants (≥37 weeks’ gestation) aged 8 to 60 days meeting AAP inclusion criteria. Infants were excluded for critical illness (eg, highest triage acuity level; emergent interventions, such as endotracheal intubation or vasopressors; intensive care admission; abnormal Pediatric Assessment Triangle).3,5 The primary outcome was accuracy for detecting IBIs using AAP-recommended IM thresholds (PCT ≤0.5 ng/mL, CRP ≤20.0 mg/L, ANC ≤4000/mm3).1 We report test characteristics (negative predictive value NPV, sensitivity, specificity) using 3 risk-stratification strategies: (1) PCT and CRP, (2) PCT and ANC, and (3) PCT, ANC, and UA.3 This study received approval from the institutional Research Ethics Board and followed the 2015 Standards for the Reporting of Diagnostic Accuracy guidelines.A total of 2032 febrile infants were included; 1703 (83.8%) were more 21 days old, and 29 (1.4%) had IBIs, including 23 (1.1%) with bacteremia and 6 (0.3%) with bacterial meningitis (Table 1). Table 2 summarizes the diagnostic accuracy of each risk-stratification strategy. All strategies performed with 100% sensitivity and NPV with no missed IBIs among infants aged 22 to 60 days and overall. The combination of PCT+CRP correctly classified the most infants as low risk, yielding the highest specificity in all age groups (85.6%; 95% CI, 83.8–87.3 for infants aged >21 days). Specificity using PCT+ANC among infants aged older than 21 days was 73.7% (95% CI, 71.5–75.8) and 62.0% (95% CI, 59.6–64.4) with the addition of a negative UA.This comparison of PCT in combination with CRP or ANC for detection of IBIs among well-appearing febrile infants aged 60 days or younger informs our understanding of AAP recommendations in 3 important ways. First, all PCT-based strategies performed with perfect sensitivity and NPV, adding to literature supporting the safety of AAP recommendations.4,6,7 However, the combination of PCT+CRP, as initially recommended by the AAP, significantly improved specificity compared with PCT+ANC. Second, the addition of UA testing, while necessary to rule out UTIs, lowered specificity for IBIs. Findings here support the AAP recommendation for infants aged older than 21 days to assess risk of UTI by UA but that decisions regarding antibiotics and LP for UA-positive infants be made separately using IMs.1,8,9 Third, the NPV of all PCT-based strategies was excellent for ruling out IBIs, including infants aged 8 to 21 days, highlighting the need for larger studies comparing outcomes of PCT-based risk stratification vs systematic LPs for these infants, as currently recommended. When using PCT, infant age may not be a relevant variable, as seen with the methodologically derived PECARN rule.3This study has limitations. This was a single-center study, and results may not be generalizable to other settings. Only 29 infants had IBIs; however, the study cohort was similarly large compared with large multicenter cohorts.3,6 There were too few infants/IBIs to meaningfully estimate diagnostic accuracy among infants aged 22 to 28 days. Duration of fever was not considered and could impact the diagnostic accuracies reported here.10 However, if externally validated, PCT+CRP should be considered as an alternative and potentially preferable combination of IMs that could further reduce unnecessary LPs, antibiotic exposure, and hospitalizations among febrile young infants.The authors thank Amelie Hilditch for clerical assistance.