Effect of Broad-Spectrum Antibiotic De-escalation on Critically Ill Patient Outcomes: A Retrospective Cohort Study

Abstract

Purpose: Antibiotic de-escalation (ADE) in critically ill patients is controversial. Previous studies mainly focused on mortality; however, data are lacking about superinfection. Therefore, we aimed to identify the impact of ADE versus continuation of therapy on superinfections rate and other outcomes in critically ill patients.

Methods: This was a two-center retrospective cohort study of adults initiated on broad-spectrum antibiotics in the intensive care unit (ICU) for ≥ 48 h. The primary outcome was the superinfection rate. Secondary outcomes included 30-day infection recurrence, ICU and hospital length of stay, and mortality.

Results: 250 patients were included, 125 in each group (ADE group and continuation group). Broad spectrum antibiotic discontinuation occurred at a mean of 7.2 ± 5.2 days in the ADE arm vs. 10.3 ± 7.7 in the continuation arm (P value = 0.001). Superinfection was numerically lower in the ADE group (6.4% vs. 10.4%; P = 0.254), but the difference was not significant. Additionally, the ADE group had shorter days to infection recurrence (P = 0.045) but a longer hospital stay (26 (14-46) vs. 21 (10-36) days; P = 0.016) and a longer ICU stay (14 (6-23) vs. 8 (4-16) days; P = 0.002).

Conclusion: No significant differences were found in superinfection rates among ICU patients whose broad-spectrum antibiotics were de-escalated versus patients whose antibiotics were continued. Future research into the association between rapid diagnostics with antibiotic de-escalation in the setting of high resistance is warranted.

Keywords: Antibiotics; Broad spectrum; De-escalation; Stewardship; Superinfection.