CRP, procalcitonin, Step-by-Step approach predict infections in febrile infants

Biomarkers such as C-reactive protein (CRP) and procalcitonin are robust predictors of serious bacterial infections (SBIs) and invasive bacterial infections (IBIs), respectively, in febrile infants, results of a Singapore study have shown.

In addition, the Step-by-Step approach delivers the highest sensitivity and negative predictive value (NPV), while the Lab-score shows the highest specificity and area under receiver operating characteristic curve (AUC) in predicting both SBIs and IBIs.

This prospective cohort study involved 258 infants aged 0‒90 days who presented to an emergency department (ED) from July 2020 to August 2021. The researchers assessed the performances of Lab-score, Step-by-Step (original and modified), and Pediatric Emergency Care Applied Research Network (PECARN) rule in predicting SBIs and IBIs.

SBIs consisted of bacterial meningitis, bacteraemia, and/or urinary tract infections, while IBIs included bacteraemia and/or bacterial meningitis.

Of the infants, 86 (33.3 percent) presented with SBIs and nine (3.5 percent) with IBIs. [Ann Acad Med Singap 2022;51:595-604]

In predicting SBIs, absolute neutrophil count achieved the highest sensitivity and NPV, while procalcitonin ≥1.7 ng/mL had the highest specificity and positive predictive value (PPV). CRP ≥20 mg/L showed the highest AUC of 0.741 (95 percent confidence interval [CI], 0.672‒0.810) among biomarkers.

Among diagnostic approaches, the Step-by-Step (original) approach achieved the highest sensitivity (97.7 percent), while Lab-score showed the highest AUC of 0.695 (95 percent CI, 0.621‒0.768) compared to the PECARN rule at 0.625 (95 percent CI, 0.556‒0.694) and Step-by-Step (original) at 0.573 (95 percent CI, 0.502‒0.644).

In predicting IBIs, procalcitonin ≥1.7 ng/mL delivered the highest sensitivity, specificity, PPV, and NPV, while the Step-by-Step (original and modified) approach showed the highest sensitivity of 100 percent. Lab-score achieved the highest AUC of 0.854 (95 percent CI, 0.731‒0.977) relative to the PECARN rule at 0.589 (95 percent CI, 0.420‒0.758) and Step-by-Step at 0.562 (95 percent CI, 0.392‒0.732).

“We found that CRP and procalcitonin as single biomarkers were strong predictors of SBIs and IBIs,” the researchers said. “In line with the findings of previous studies, CRP ≥20 mg/L performed best in predicting SBIs with an AUC of 0.741.” [Ann Emerg Med 2012;60:591-600; Pediatr Infect Dis J 2014;33:e273-279; BMJ Paediatrics Open 2021;5:e000861]

Peak levels

A commonly available biomarker, CRP peaks later in the course of an illness, normally 4‒6 hours. On the other hand, procalcitonin levels immediately rise in response to bacterial infection, normally within 2‒4 hours, but may take 6‒12 hours to peak. [J Intensive Care 2017;5:1-7]

“CRP performed better than PCT in predicting for SBIs in our population,” the researchers said. “We postulate that this could be because majority of these infants receive their workup after hospitalization rather than on presentation to the ED, therefore providing an adequate window for CRP to rise.”

Due to the seriousness of missed diagnoses among febrile infants, physicians must choose a diagnostic approach with high sensitivity and NPV. In this regard, the original Step-by-Step algorithm by Gomez and colleagues showed a high sensitivity of 92 percent and NPV of 99.3 percent for IBIs in this population. [Pediatrics 2016;138:e20154381]

The present study revealed a similarly high sensitivity and NPV in predicting SBIs (97.7 percent and 93.5 percent, respectively) and IBIs (100 percent and 100 percent, respectively). Of note, the original Step-by-Step approach achieved the highest sensitivity and NPV in predicting both SBIs and IBIs, but due to low specificity, the overall AUC was not optimal.

“The team is working to refine the algorithm by recruiting a larger study population to drive locally derived thresholds rather than depend on the existing published thresholds,” the researchers said.

“Going forward, we are implementing a modified Step-by-Step algorithm to examine outcomes and missed cases, towards validating it as a safe tool for use in Singapore,” they added.