Abstract
Background: Reducing critically ill patients' exposure to antibiotics is mandatory. In randomized controlled trials, procalcitonin (PCT)-guided algorithms (i.e., antibiotic therapy [ABT] should be stopped whenever PCT is less than 0.5 µg/L or is below 80% of the peak value) reduced the duration of (ABT) more than compliance with the current guidelines. However, the interest of such stopping rules in daily practice remains debated. Thus, we carried out a real-life study addressing this issue. Results: During the study period, 112 patients with sepsis upon intensive care unit admittance were included. The median age was 66 years (56-79). Half of the patients presented with acute respiratory failure. Pneumonia was diagnosed in 78% of them, and 41% met septic shock criteria. The initial ABT was empirical in most cases, and appropriateness rate to the isolated bacteria reached 71%. A median number of four PCT measurements was achieved in both groups. The compliance rate with the PCT algorithm was 54%. The median duration of ABT was 5 (4-7) days if the PCT algorithm was followed, as compared to 7 (5-10) days otherwise (p < 0.001). This ABT stopping rule allowed a 2-day reduction in the treatment duration as compared with those recommended by the guidelines (p < 0.001). The only independent factor associated with shorter treatment duration was compliance with the PCT algorithm (OR = 0.74, 95% CI [0.62; 0.88]; p < 0.001). Regarding safety, no difference in outcome was found between the two groups. Conclusions: Complying with one PCT-based stopping rule is associated with a significant reduction in the duration of ABT in septic critically ill patients, without apparent impact on patient outcomes.