Procalcitonin-guided Antibiotic Treatment in Patients With Positive Blood Cultures: A Patient-level Meta-analysis of Randomized Trials

Abstract

Whether procalcitonin (PCT)-guided antibiotic management in patients with positive blood cultures is safe remains understudied. We performed a patient level meta-analysis to investigate effects of PCT-guided antibiotic management in patients with bacteremia overall and stratified based on type of pathogen. We extracted and analysed individual data of 523 patients with positive blood cultures included in 13 trials (7 studies evaluating patients with acute respiratory illnesses, 6 studies patients with sepsis), in which patients were randomly assigned to receive antibiotics based on PCT levels (PCT group) or a control group. The main efficacy endpoint was duration of antibiotic treatment. The main safety endpoint was mortality within 30 days. Mean duration of antibiotic therapy for 253 patients who received PCT guided treatment was significantly shorter compared to 270 control patients (-2.86 days (95%CI -4.88 to -0.84, p=0.006). Mortality was similar in both arms (16.6% vs 20.0%, p=0.263). In subgroup analyses by type of pathogen, we noted a trend of shorter mean antibiotic durations in the PCT arm for patients infected with gram positive organisms or Escherichia coli and significantly shorter treatment for subjects with pneumococcal bacteremia (-4.52 days [95%CI -7.40 to -1.64, p=0.002). In analysis by site of infection, antibiotics exposure was shortened in PCT group subjects with Streptococcus pneumoniae respiratory infection and those with Escherichia coli urogenital infections. This meta-analysis of patients with bacteremia receiving PCT-guided antibiotic management demonstrates lower antibiotic exposure without an apparent increase in mortality. Few differences were demonstrated in subgroup analysis stratified by type or site of infection but notable for decreased exposure in patients with pneumococcal pneumonia and E Coli urogenital infections.