Procalcitonin to Reduce Long-Term Infection-associated Adverse Events in Sepsis. A Randomized Trial

<b>Rationale:</b> Although early antimicrobial discontinuation guided by procalcitonin (PCT) has shown decreased antibiotic consumption in lower respiratory tract infections, the outcomes in long-term sepsis sequelae remain unclear.<b>Objectives:</b> To investigate if PCT guidance may reduce the incidence of long-term infection-associated adverse events in sepsis.<b>Methods:</b> In this multicenter trial, 266 patients with sepsis (by Sepsis-3 definitions) with lower respiratory tract infections, acute pyelonephritis, or primary bloodstream infection were randomized (1:1) to receive either PCT-guided discontinuation of antimicrobials or standard of care. The discontinuation criterion was ≥80% reduction in PCT levels or any PCT ≤0.5 μg/L at Day 5 or later. The primary outcome was the rate of infection-associated adverse events at Day 180, a composite of the incidence of any new infection by <i>Clostridioides difficile</i> or multidrug-resistant organisms, or any death attributed to baseline <i>C. difficile</i> or multidrug-resistant organism infection. Secondary outcomes included 28-day mortality, length of antibiotic therapy, and cost of hospitalization.<b>Measurements and Main Results:</b> The rate of infection-associated adverse events was 7.2% (95% confidence interval [CI], 3.8-13.1%; 9/125) versus 15.3% (95% CI, 10.1-22.4%; 20/131) (hazard ratio, 0.45; 95% CI, 0.20-0.98; <i>P</i> = 0.045); 28-day mortality 15.2% (95% CI, 10-22.5%; 19/125) versus 28.2% (95% CI, 21.2-36.5%; 37/131) (hazard ratio, 0.51; 95% CI, 0.29-0.89; <i>P</i> = 0.02); and median length of antibiotic therapy 5 (range, 5-7) versus 10 (range, 7-15) days (<i>P</i> < 0.001) in the PCT and standard-of-care arms, respectively. The cost of hospitalization was also reduced in the PCT arm.<b>Conclusions:</b> In sepsis, PCT guidance was effective in reducing infection-associated adverse events, 28-day mortality, and cost of hospitalization.Clinical trial registered with www.clinicaltrials.gov (NCT03333304).