Real-world use of cefiderocol as monotherapy or combination therapy for the treatment of Gram-negative bacterial infections: the multicentre retrospective CEFI-BAC study

Objectives This study aimed to characterise the real-world use of cefiderocol in treating Gram-negative bacterial infections (GNBIs) across Italian hospitals. Methods We conducted a multicentre retrospective study enrolling patients with GNBI treated with cefiderocol from January 2021 to February 2023. Statistical analyses included Kaplan–Meier survival estimates and multivariable Cox regression. A propensity score analysis with inverse probability of treatment weighting (IPTW) was also performed to compare the treatment effect of combination therapy versus monotherapy adjusting for imbalances between treatment groups. Results A total of 239 patients were included. Bloodstream infections were the most common (49.8%), followed by ventilator-associated and hospital-acquired pneumonia. Acinetobacter baumannii was the most common isolate (64.8%), followed by Klebsiella spp. (23%), Pseudomonas aeruginosa (17.6%), and Stenotrophomonas maltophilia (8.8%). Overall 30-day survival was 71% (95% CI: 65-76), with no significant differences between monotherapy and combination therapy. Independent predictors of higher 30-day mortality were: having received 2 or 3 previous lines of antibiotic therapy (aHR: 4.26, 95% CI: 1.00-18.20; aHR: 7.33, 95% CI: 1.53-35.05), SARS-CoV-2 coinfection (aHR: 4.19, 95% CI: 2.04-8.59), and isolation of NDM-producing Klebsiella spp. (aHR: 6.22, 95% CI: 2.09-18.50). Conclusions Real-world experience supports the role of cefiderocol as a valuable option for GNBIs, with no clinical advantage of combination therapy over monotherapy. Notably, NDM-producing infections and use of cefiderocol as salvage therapy are associated with poor outcomes, highlighting the need for optimised treatment strategies.