RESUMO
INTRODUCTION: The use of procalcitonin (PCT) in the evaluation of the febrile infant in the emergency care unit has been widespread. The aim of this study is to assess whether the introduction of PCT has changed the management of hospitalised febrile infants and the cost/effectiveness of this marker. MATERIALS AND METHODS: A retrospective study was performed comparing 2 periods: January-December 2009 (without PCT) and January-December 2011 (routine use of PCT). Infants aged 7 to 90 days with fever who were admitted to a university hospital and had a blood test performed were included in the study. Bacterial infection rate, antibiotic use, hospitalisation days, and analytical costs were compared. Evaluations were made using PCT, C-reactive protein (CRP), white cell count, Rochester score, and the lab-score proposed by Galetto-Lacour for the diagnosis of bacterial infection. RESULTS: A total of 109 patients were included in period 1, and 111 in period 2 (87 of which had a PCT value). The prevalence of bacterial infection, use of antibiotics, number of blood tests, and days of hospital admission was similar in both periods. The blood test cost was significantly higher in the second period. Sensitivity, specificity, positive predictive value and negative predictive value were 70.6, 58.1, 52.6 and 75%, respectively for the CRP (cut-off 1mg/dL) and 41.7; 78.4; 57.7, and 65.6% for the PCT (cut-off value 0.5ng/ml). CONCLUSIONS: The use of PCT does not seem to have a significant impact on the management of the hospitalised febrile infant.