Eficácia e segurança de um protocolo utilizando a proteína C reativa para orientação da terapia antibiótica aplicado por meio de ferramenta digital de suporte à decisão clínica: ensaio clínico randomizado

Abstract

Introduction: The use of circulating biomarkers, such as procalcitonin (PCT) and C-reactive protein (CRP), has been studied to help reduce antibiotic consumption and control the emergence of multidrug-resistant microorganisms. Although the evidence for the use of CRP in this context is less robust, it offers advantages such as lower cost and greater availability. Objectives: To determine whether an algorithm based on CRP levels and strategic clinical variables, implemented through a Digital Clinical Decision Support System accessible via mobile devices, can safely reduce antimicrobial exposure in hospitalized adult patients. Methods and Analysis: A single-center, prospective, open-label, randomized clinical trial involving patients with clinical suspicion or microbiological confirmation of bacterial infection, who had started antibiotic therapy within the last 72 hours. The intervention consisted of applying an algorithm via a mobile application, in which serial CRP values and clinical variables were used to guide decisions on antibiotic discontinuation. The control group received treatment duration recommendations based on the best available evidence. The primary outcome was the duration of antibiotic treatment for the infectious episode that led to the patient's inclusion in the study, in each group. Results: A total of 110 patients were included, with a mean age of 67 years and a predominance of males. The duration of antibiotic treatment was significantly shorter in the intervention group compared to the control group (median of 6.0 [5.0; 7.0] vs. 7.0 [5.0; 9.0] days; p = 0.015). No differences were found in total antimicrobial exposure during hospitalization, antibiotic-free days, or hospital length of stay. All-cause mortality, treatment failure, reinfection rate, subsequent infections by resistant microorganisms, and Clostridioides difficile infection rates were similar between the groups. Conclusion: A digital application-based algorithm incorporating CRP values and clinical variables safely reduced antibiotic use for the initial infectious episode in hospitalized adults but did not decrease total antimicrobial exposure. Further studies are needed to evaluate this strategy's performance in other settings, as well as its safety and impact on clinical practice.