Inflammatory Markers and Invasive Bacterial Infection in Febrile Infants With Positive Urinalyses

The American Academy of Pediatrics (AAP) clinical practice guideline (CPG) for febrile infants provides varying management recommendations for infants with positive urinalyses (UAs) depending on age and the presence of normal vs abnormal inflammatory markers (IMs).1 In a large prospective study conducted at children’s hospital emergency departments (EDs), infants with positive UAs and normal procalcitonin (PCT) and absolute neutrophil count (ANC) had a low prevalence of bacteremia and bacterial meningitis.2 However, there are limited studies in non–children’s hospital EDs or when using the AAP recommended combination of IMs when PCT is not available: maximum temperature (Tmax), ANC, and C-reactive protein (CRP).1 We aimed to describe the age-based prevalence of bacteremia and bacterial meningitis in febrile infants aged 8 to 60 days with positive UAs and normal vs abnormal IMs in a large, diverse sample of EDs.We conducted a secondary analysis of a retrospective cohort of well-appearing, previously healthy infants aged 8 to 60 days with temperatures greater than or equal to 38.0 °C between November 1, 2020, and October 31, 2022, from the Reducing Excessive Variability in Infant Sepsis Evaluation II (REVISE II) data set.3 The study included 33 freestanding children’s hospitals, 37 nonfreestanding children’s hospitals, and 35 general hospitals located in the United States and Canada.Our study was limited to infants with positive UAs, defined as positive leukocyte esterase, presence of nitrites, or greater than 5 white blood cells per high-powered field on microscopy.1,4 We defined bacteremia and bacterial meningitis as the presence of an identified pathogen in blood or cerebrospinal fluid (CSF) cultures, respectively, in infants who received a course of treatment. For IMs, the data collection form listed the normal and abnormal values per the AAP CPG. IMs were considered normal at the following thresholds: PCT less than or equal to 0.5 ng/mL, ANC less than or equal to 4000/mm3, CRP less than 20 mg/L (2 mg/dL), and Tmax less than 38.6 °C.1 We calculated the prevalence of bacteremia and bacterial meningitis among infants with positive UAs, comparing those with normal vs abnormal IMs, overall and stratified by age group, using the AAP CPG recommended combinations: PCT+ANC and Tmax+ANC+CRP. Infants were required to have all IMs in the combination. We used Stata version 18.0 (StataCorp, Inc) for the analyses. The study was deemed exempt by the AAP Institutional Review Board.Of 2144 infants with positive UAs, 135 (6.3%) had bacteremia without meningitis and 12 (0.6%) had meningitis. The prevalence of bacteremia was low in infants with positive UAs and normal IMs (0.3% for both PCT+ANC and Tmax+ANC+CRP) (Table 1). Only 1 infant with normal IMs had bacterial meningitis, across age groups and IM combinations. This 34-day-old infant had a normal PCT and ANC but an abnormal CRP and no bacteremia, and was admitted at the index visit (Table 2). Conversely, the prevalence of bacteremia was high in infants with positive UAs and abnormal IMs, highest in infants aged 8 to 28 days (Table 3), and varied by which IM was abnormal (Supplemental Table 1). The prevalence of bacterial meningitis in infants with abnormal IMs varied by age group, lowest in infants aged 29 to 60 days (0.3%–0.5%) (Table 3).In a multicenter cohort study, we found a low prevalence of bacteremia and bacterial meningitis in febrile infants aged 8 to 60 days with positive UAs and normal IMs. Findings support the AAP CPG recommendations that CSF testing is not needed in most febrile infants with positive UAs and normal IMs and that infants aged 29 to 60 days can be managed as outpatients.1 Additionally, the low prevalence of bacteremia and meningitis in infants aged 8 to 21 days with normal IMs may support shared decision-making about CSF testing in this youngest age group. Although one 34-day-old infant with normal PCT+ANC had meningitis, the infant was admitted at the index visit, raising the question of whether the infant was ill-appearing.Infants with positive UAs and abnormal IMs had a high prevalence of bacteremia, particularly infants aged 8 to 28 days, with some variation by which IM was abnormal. The prevalence of meningitis in infants with abnormal IMs was low though nonzero in infants aged 29 to 60 days. Aligning with the AAP CPG, clinicians can consider shared decision-making about CSF testing for these older infants with positive UAs and abnormal IMs.1Our study builds on prior work by including a large sample of infants evaluated at both children’s and non–children’s hospitals and by assessing the AAP CPG recommended IM combinations. Limitations include the retrospective design, which may have resulted in misclassifying the clinical appearance of febrile infants. Additionally, we used the AAP CPG IM cutoff values, but not those derived more recently.5The authors acknowledge all the sites and site team members that collected data for this study as part of the AAP REVISE II QI Collaborative.