RESUMEN
OBJECTIVES: Overuse of broad-spectrum antibiotics in emergency departments (EDs) results in antibiotic resistance. We determined whether procalcitonin (PCT) -guided therapy can be used to reduce antibiotic regimens in EDs by investigating efficacy, safety and accuracy. METHODS: This was a non-inferiority multicentre randomized clinical trial, performed in two Dutch hospitals. Adult patients with fever ≥38.2°C (100.8°F) in triage were randomized between standard diagnostic workup (control group) and PCT-guided therapy, defined as standard workup with the addition of one single PCT measurement. The treatment algorithm encouraged withholding antibiotic regimens with PCT <0.5 µg/L, and starting antibiotic regimens at PCT ≥0.5 µg/L. Exclusion criteria were immunocompromised conditions, pregnancy, moribund patients, patients <72 h after surgery or requiring primary surgical intervention. Primary outcomes were efficacy, defined as number of prescribed antibiotic regimens; safety, defined as combined safety end point consisting of 30 days mortality, intensive-care unit admission, ED return visit within 2 weeks; accuracy, defined as sensitivity, specificity and area-under-the-curve (AUC) of PCT for bacterial infections. Non-inferiority margin for safety outcome was 7.5%. RESULTS: Between August 2014 and January 2017, 551 individuals were included. In the PCT-guided group (n = 275) 200 (73%) patients were prescribed antibiotic regimens, in the control group (n = 276) 212 (77%) patients were prescribed antibiotics (p 0.28). There was no significant difference in combined safety end point between the PCT-guided group, 29 (11%), and control group, 46 (16%) (p 0.16), with a non-inferiority margin of 0.46% (n = 526). AUC for confirmed bacterial infections for PCT was 0.681 (95% CI 0.633-0.730), and for CRP was 0.619 (95% CI 0.569-0.669). CONCLUSIONS: PCT-guided therapy was non-inferior in terms of safety, but did not reduce prescription of antibiotic regimens in an ED population with fever. In this heterogeneous population, the accuracy of PCT in diagnosing bacterial infections was poor. TRIAL REGISTRATION IN NETHERLANDS TRIAL REGISTER: http://www.trialregister.nl/trialreg/admin/rctview.asp?TC=4949.