Antibiotic de-escalation in severe intra-abdominal infections: a single-center retrospective cohort study on safety and prognostic factors.

BACKGROUND

Antibiotic de-escalation is a key antimicrobial stewardship strategy aimed at reducing the ecological and adverse effects of broad-spectrum therapy without compromising efficacy. However, its real-world safety and impact on clinical outcomes in patients with severe intra-abdominal infections (IAIs) remain underexplored.

METHODS

A single-center, retrospective cohort study was conducted from June 2022 to June 2025. Adults with severe IAIs who received initial broad-spectrum antibiotics and achieved source control within 48 h were included. Patients were stratified into de-escalation and non-de-escalation groups on the basis of the antibiotic management strategy. Propensity score matching (1:1) was used to balance baseline characteristics. The primary outcome was clinical cure at 30 days. Secondary outcomes included 28-day mortality, lengths of ICU and hospital stay, antibiotic duration, and the incidence of secondary infections and Clostridioides difficile infection (CDI).

RESULTS

Among 405 eligible patients, 212 were matched (106 per group). The de-escalation group had a significantly higher clinical cure rate (79.8% vs. 63.5%; OR 2.24, 95% CI 1.29-3.89; P = 0.004) and lower 28-day mortality (9.6% vs. 21.2%; OR 0.40, 95% CI 0.18-0.88; P = 0.02). De-escalation was also associated with shorter ICU and hospital stays, reduced antibiotic duration, and fewer secondary infections. Multivariable analysis confirmed de-escalation (aOR 2.71, 95% CI 1.48-4.97) and adequate source control (aOR 3.52, 95% CI 1.81-6.85) as independent predictors of clinical cure.

CONCLUSION

Antibiotic de-escalation following adequate source control is associated with improved clinical outcomes and reduced mortality in severe IAIs without increasing the risk of complications. These findings support its integration into antimicrobial stewardship programs for this high-risk population.

CLINICAL TRIAL NUMBER: Not applicable.