Abstract Background New Delhi Metallo-beta-lactamase (NDM)-producing Enterobacterales infections pose a significant health threat given limited therapeutic options. A rise in cases within our health system prompted us to compare clinical outcomes of patients with NDM-producing bacteremia treated with ceftazidime/avibactam plus aztreonam (CZA-ATM) and/or cefiderocol (FDC). Methods Patients identified to have blood cultures with an NDM-producing organism (detection of blaNDM by PCR) from 2022 to 2024 were analyzed retrospectively. Only patients who completed therapy with an active agent (CZA-ATM or FDC) were included in the analysis. Baseline characteristics were compared using Fisher’s Exact test and Wilcoxon rank-sum test, as appropriate. Logistic regression, adjusting for baseline variables with p-values of 0.10, was used to determine the adjusted odds ratio of mortality at 30 days. Results A total of 97 infections were identified. The most common organisms were Klebsiella pneumoniae (78%, n=76) and E. coli (12%, n=12). Eight percent (3/38) of isolates were non- susceptible to CZA-ATM and 17% (9/54) were non-susceptible to FDC. Seventeen (17%) patients died within 48 hours from culture. Of the 60 patients who received targeted therapy without a regimen change, 48 (80%) patients received CZA-ATM and 12 (20%) received FDC. Median (IQR) age of patients was 70 (60-78.5) years, 23 (34%) were admitted to an ICU within 24 hours of culture collection, and 8 (13%) were immunocompromised. Baseline characteristics in the two groups were similar, except for male sex and Pitt bacteremia score. A total of 22 (37%) patients died by 30 days (17 (35%) in CZA-ATM group and 5 (42%) in FDC group) but this was not significantly different (OR=1.30; 95% CI: 0.36-4.74; p=0.688). Days to bacteremia clearance was significantly higher in the FDC group (p=0.02). Recurrent infection at 90 days occurred in 13 (27%) and 3 (25%) in the CZA-ATM and FDC groups, respectively (p=1.000). Conclusion Infections with NDM-producing organisms are associated with high mortality. FDC resistance may be a concern for its empiric use. Although a larger cohort is required, our data suggest that mortality and likelihood of recurrent infection are similar whether CZA-ATM or FDC is used for treatment of NDM bacteremias. Disclosures All Authors: No reported disclosures
Dahal et al. (Thu,) studied this question.