Use of procalcitonin point-of-care testing to guide de-escalation of antibiotic therapy in adults with suspected bacterial sepsis in a tertiary hospital in Bangladesh: a randomised controlled open-label trial

Résumé fourni par la source

Background Important drivers of antibiotic resistance in low- and middle-income countries (LMICs) include antibiotic overuse and scarcity of microbiological diagnostics. Procalcitonin point-of-care testing is a potential tool to guide antibiotic de-escalation in patients admitted with suspected bacterial sepsis. Methods In an open-label randomised controlled clinical trial in Chittagong Medical College Hospital in Bangladesh, adult patients with suspected bacterial sepsis were randomised to procalcitonin (PCT)-guided de-escalation of antibiotic therapy or standard-of-care. In the intervention group, serum procalcitonin was assessed daily and clinicians received a non-binding advice to stop antibiotics after 72 h if procalcitonin dropped below pre-defined threshold concentrations (0.5 ng/mL or 80% from the peak value if baseline value > 2 ng/mL). The primary endpoint was length of antibiotic therapy analysed by intention-to-treat. The study is registered with ClinicalTrials.gov (NCT05955612) and was completed in July 2024. Findings Between 26 July 2023 and 20 June 2024, 532 of the 1002 patients screened for eligibility were randomly assigned to procalcitonin-guided antibiotic therapy (n = 266) or standard-of-care (n = 266). Follow-up was completed on 30 July 2024. Median (interquartile range [IQR]) length of antibiotic therapy was 4.7 (2.8–8.0) days with procalcitonin-guided therapy compared to 9.0 (5.8–12.0) days in the control arm (p < 0.0001). Mortality was similar between the two groups. In the procalcitonin-guided arm, 23 of 265 participants died (8.7%; 95% confidence interval [CI], 5.6%–12.7%), compared with 21 of 265 participants (7.9%; 95% CI, 5.0%–11.9%) in the control arm. The absolute risk difference was 0.8 percentage points (95% CI, −3.9 to 5.5 percentage points) (p = 0.875). Following review, the Data Safety Monitoring Board did not identify any deaths as attributable to antibiotic discontinuation. The incidence of recurrent infections was 8/265 (3.0%) in procalcitonin-guided antibiotic therapy and 10/265 (3.8%) in the control arm (p = 0.811). Interpretation In this single-centre trial in Bangladesh, procalcitonin-guided de-escalation reduced the duration of antibiotic therapy in adults with suspected bacterial sepsis. Mortality and recurrent infection were similar between the groups, although the trial was not powered to assess these outcomes formally. This promising approach of point-of-care procalcitonin-guided antibiotic discontinuation should be evaluated further in comparable resource-limited settings. Funding Wellcome Trust of Great Britain.