To the Editor We commend Milcent et al1 on their prospective study on the use of procalcitonin in the diagnosis of serious bacterial infections in infants. We believe there are 3 issues that limit applicability of the study as presented.
First, the Milcent et al study1 included patients classified with the clinical appearance determined as “minimally, moderately, or very ill.” There is general agreement to perform a full workup on all febrile infants who do not appear well. Additionally, for those ill-appearing infants who undergo full evaluation, urinalyses and cerebrospinal fluid studies have high sensitivity for urinary tract infection and meningitis, respectively, in febrile infants. The management of well-appearing febrile infants is more controversial, with generally good outcomes demonstrated even with highly variable evaluation strategies.2,3 Therefore, the crucial clinical struggle is to more quickly and accurately detect systemic bacterial infections, and particularly bacteremia, in well-appearing febrile infants. In terms of clinical decision making, the study could provide crucial information by providing the test characteristics for procalcitonin in well-appearing infants or those with low Yale Observation Scale scores. We did note that in this study, the Yale Observation Scale was not helpful in detecting serious or invasive bacterial infection. Again, exploring Yale Observation Scale scores and the general impression of illness when applying the results could be helpful.
Antoon JW, Steiner MJ. Use of Procalcitonin Assays to Predict Serious Bacterial Infection in Young Febrile Infants. JAMA Pediatr. 2016;170(6):623. doi:10.1001/jamapediatrics.2016.0382