Key points are not available for this paper at this time.
INTRODUCTION: Identification of bacterial infections is crucial if treatment is to be initiated early and antibiotics used rationally. The primary objective of this study was to test the efficiency of procalcitonin (PCT) in identifying bacterial/parasitic episodes among febrile adult patients presenting to an emergency department. Secondary objectives were to identify clinical or biological variables associated with either bacterial/parasitic infection or critical illness. METHODS: This was a prospective, single centre, non-interventional study, conducted in the adult emergency department of an academic tertiary care hospital. We included patients with body temperature of 38.5 degrees C or greater. A serum sample for measurement of PCT was collected in the emergency room. Patients were followed up until day 30. After reviewing the medical files, two independent experts, who were blind to the PCT results, classified each of the patients as having a bacterial/parasitic infection, viral infection, or another diagnosis. RESULTS: Among 243 patients included in the study, 167 had bacterial/parasitic infections, 35 had viral infections and 41 had other diagnoses. The PCT assay, with a 0.2 microg/l cutoff value, had a sensitivity of 0.77 and a specificity of 0.59 in diagnosing bacterial/parasitic infection. Of the patients with PCT 5 microg/l or greater, 51% had critical illness (death or intensive care unit admission) as compared with 13% of patients with lower PCT values. CONCLUSION: Bearing in mind the limitations of an observational study design, the judgements of the emergency department physicians were reasonably accurate in determining the pretest probability of bacterial/parasitic infection. PCT may provide additional, valuable information on the aetiology and prognosis of infection in the emergency department.
Hausfater et al. (Wed,) studied this question.