Management of the Febrile Young Infant: Update for the 21st Century

Abstract

Infants aged 90 days or younger with fever are frequently evaluated in the pediatric emergency department. Physical examination findings and individual laboratory investigations are not reliable to differentiate benign viral infections from serious bacterial infections in febrile infants. Clinical prediction models were developed more than 25 years ago and have high sensitivity but relatively low specificity to identify bacterial infections in febrile infants. Newer laboratory investigations such as C-reactive protein and procalcitonin have favorable test characteristics compared with traditional laboratory studies such as a white blood cell count. These novel biomarkers have not gained widespread acceptance because of lack of robust prospectively collected data, varying thresholds to define positivity, and differing inclusion criteria across studies. However, C-reactive protein and procalcitonin, when combined with other patient characteristics in the step-by-step approach, have a high sensitivity for detection of serious bacterial infection. The RNA biosignatures are a novel biomarker under investigation for detection of bacterial infection in febrile infants.

Figure 1

Micro-array showing the difference in transcriptional profiles of host leukocytes in response to different pathogens. Heat maps (middle diagram) show over-expressed genes labeled in red and under-expressed genes labeled in blue. Alternatively, modular analysis (right diagram) shows genes with similar function being grouped together and are labeled red for over-expressed or blue for under-expressed. From Mahajan et al, Pediatr Emerg Care. 2015.