Background: Fever is an exceedingly common complaint in the pediatric emergency department (ED), accounting for nearly 20% of pediatric ED visits. The vast majority of children evaluated with fever do not have a serious bacterial infection (e.g., bacteremia) and many of them have a viral infection without any evidence of bacterial infection. However, in the neonate a missed serious bacterial infection (SBI) can have devastating consequences. The goal of this analysis was to determine the fever characteristics (height, duration, location) associated with risk of serious bacterial infection (SBI) in a secondary analysis of a large cohort of febrile infants.
Methods: This is a secondary analysis of the Pediatric Emergency Care Applied Research Network study on febrile infants. Patients were prospectively infant enrolled at 26 enrolling emergency departments between 2008 and 2013 with a documented fever. We analyzed association of height of fever, location of where temperature was taken (enrolling ED vs. non-healthcare location) and duration of fever (>24 hours vs. <12 hours) using odds ratio (OR) and reporting 95% confidence intervals (95% CI).
Results: We included 4,821 patients who had at least a blood culture completed. We also did a subgroup analysis of 3,663 patients who had a blood, cerebrospinal fluid (CSF), and urine culture. Height of fever was significantly associated with risk of SBI, with an odds ratio of 1.5 (95% CI 1.2-1.8). This was true for positive blood culture (OR 1.4, 95% CI 1.2-1.7), urine culture (OR 1.4, 95% CI 1.3-1.5), and CSF culture (OR 1.5, 95% CI 1.1-2.0). This association held when results from indeterminate cultures were excluded from the analysis (OR for SBI 1.5, 95% 1.3-1.6). Duration of fever was not associated with risk of SBI, and a fever taken in the enrolling ED versus at a non-healthcare facility was minimally associated with risk of SBI (OR 1.3, 95% CI 1.0-1.5), specifically urine culture (odds ratio 1.3, 95% CI 1.0-1.5). This relationship was stronger when indeterminate results were presumed positive (OR for SBI 1.4, 95% CI 1.1-1.8)
Conclusion: In all analyses, height of fever was associated with all 3 major types of SBI in febrile infants. Duration and location of fever were less reliably associated with risk of SBI, but there was a small association of risk of SBI and a fever taken at the enrolling ED vs. at a non-healthcare location.