Ceftazidime-avibactam (CAZ/AVI) is a broad-spectrum combination that effectively targets carbapenem-resistant Enterobacterales (CRE), including Klebsiella pneumoniae and Escherichia coli. Pharmacokinetically, CAZ/AVI follows a time-dependent killing mechanism, where the effectiveness of ceftazidime is optimized when its concentration stays above the minimum inhibitory concentration (MIC) for at least 50% of the dosing interval. For avibactam, the target is to maintain a concentration of at least 1.0 mg/L for 50% of the dosing interval to ensure optimal activity. Dimelow et al. demonstrated that both ceftazidime and avibactam achieve adequate epithelial lining fluid (ELF) concentrations, with ELF: plasma ratios of 52% for ceftazidime and 42% for avibactam, exceeding plasma PK/PD targets and supporting their efficacy in treating infections, including potential gut decolonization8. This study provides valuable insights into the efficacy of CAZ/AVI for achieving PAS negativity in patients colonized with CRE, particularly CRKP and CREC. Our results indicate that CAZ/AVI is more effective in achieving short-term PAS negativity compared to other antimicrobial regimens, including colistin, polymyxin B, combinations with meropenem, and fosfomycin. This aligns with findings from Perrotta et al., where CAZ/AVI plus aztreonam therapy successfully treated NDM producing Klebsiella pneumoniae and contributed to rectal decolonization in a critically ill patient9. However, Lima et al. identified that the use of CAZ/AVI was associated with persistent CRE gut colonization, emphasizing the complex dynamics of decolonization in patients undergoing long-term antimicrobial treatment10.
Our findings align with Bar-Yoseph et al., who emphasized that while CRE colonization rates decrease over time, persistent colonization remains a challenge, highlighting the need for effective decolonization therapies11. In our study, CAZ/AVI demonstrated a significantly higher rate of PAS negativity at the two-week mark compared to non-CAZ/AVI regimens. This aligns with recent studies suggesting CAZ/AVI’s potential as a targeted decolonization strategy in high-risk populations, although its role in decolonization remains underexplored5. Furthermore, Herrera et al. identified several risk factors, such as male sex, clinical sample, and multiple readmissions, that are associated with prolonged carriage of multidrug-resistant Gram-negative bacilli, underscoring the complexity of persistent colonization and the potential benefit of targeted therapeutic strategies like CAZ/AVI for managing CRE carriage in high-risk patients12. Loukili et al.‘s study on intestinal clearance of carbapenemase-producing Enterobacterales (CPE-IC) found that decolonization can be a prolonged process, with a median time of 698 days to achieve CPE-IC in their cohort13. While their study did not focus on CAZ/AVI, it highlighted the challenges of achieving intestinal clearance in patients with CPE, which aligns with our findings that decolonization, even with effective initial therapies like CAZ/AVI, can be complicated by factors such as relapse. This emphasizes the need for ongoing surveillance and potentially more nuanced decolonization strategies that consider individual pathogen characteristics. Despite the initial success of CAZ/AVI, relapse rates between CAZ/AVI and non-CAZ/AVI groups were similar, highlighting a critical challenge in long-term management of CRE colonization. Tacconelli et al. have previously noted that current decolonization strategies often provide only transient benefits, underscoring the need for continuous surveillance and tailored management to prevent recurrence5.
Lubbert et al. reported that selective digestive decontamination (SDD) using oral colistin and gentamicin in ICU patients with KPC-2-producing Klebsiella pneumoniae showed no significant difference in decolonization rates between the treatment and control groups14. On the other hand, Oren et al.’s study observed a significant eradication rate of 56% (22/50) with oral nonabsorbable antibiotics compared to 7% (7/102) in the control group15. This result supports the effectiveness of targeted decolonization therapies using nonabsorbable antibiotics. In addition, Saidel-Odes et al. reported a significant reduction in carriage rates 7 days post-treatment with oral gentamicin and colistin (38.8% versus 83.9%; OR 0.13; 95% CI 0.02–0.74; p < 0.0016), though this effect diminished by 28 days, leading to a non-significant difference (41.5% versus 66.7%)16. Machuca et al. found that decolonization rates at 180 days post-treatment were significantly higher with gentamicin compared to controls, though streptomycin and neomycin showed no significant association17. These findings suggest that while initial decolonization is achievable, long-term outcomes and resistance development require careful monitoring. Although SDD may aid in CRE management, concerns remain regarding resistance emergence and suboptimal decolonization rates. Nevertheless, recent evidence from systematic reviews suggests that SDD may still offer short-term benefits in selected high-risk patients, supporting its continued evaluation. In line with these findings, Zhang et al. conducted a recent meta-analysis evaluating the effectiveness of SDD for decolonization of ESBL-producing and carbapenem-resistant Enterobacterales. SDD demonstrated short-term efficacy, with decolonization rates ranging from 53.8 to 68.0% in single-arm trials, and significantly higher eradication compared to control arms in both randomized and non-randomized studies18. Similarly, Feria et al. reported that oral and/or intravenous antibiotic therapy significantly increased CRE eradication rates (OR: 2.66; 95% CI: 1.55–4.55), although data on infection rates and mortality were insufficient19. These reviews underscore the growing body of evidence supporting the use of targeted antimicrobial-based strategies for CRE decolonization, while also highlighting the need for further high-quality trials to evaluate long-term efficacy and safety.
This study has several limitations. First, it is a single-center study with a relatively small sample size, which may limit the generalizability of the findings. Second, the retrospective nature of the study introduces potential biases and limits the ability to control for confounding factors. Third, molecular analysis of the resistance mechanisms in CRE isolates was not conducted, which restricts our understanding of the genetic dynamics of resistance transfer among colonized patients. Ji et al. highlighted the importance of identifying carbapenemase gene transfer between different species within the host, emphasizing how in-host plasmid transmission can significantly impact CRE colonization and dissemination20. The absence of such molecular data in our study prevents a deeper exploration of the potential for interspecies resistance transmission, which could have provided additional insights into the effectiveness and limitations of CAZ/AVI. Fourth, we did not analyze carbapenemase types. However, metallo-beta-lactamases are rare in our hospital, with OXA-48 and KPC enzymes being predominant. Notably, all isolates in our cohort were susceptible to CAZ/AVI. Fifth, the use of Chromid® Carba Agar alone for CRE surveillance without confirmatory phenotypic or molecular tests may have introduced misclassification bias. While chromogenic media are practical for routine screening, they may exhibit limited sensitivity and specificity, potentially leading to false-negative or false-positive results. In particular, colonization may have been underdetected due to the suboptimal sensitivity of the method, which is a recognized limitation of this screening approach. Finally, the study relied solely on surveillance data without including clinical evaluations or mortality outcomes, potentially overlooking critical aspects of patient health and the broader impact of treatment efficacy.
In spite of these limitations, we consider that our results might be important for infection control in ICU. The spectrum, antibacterial mechanism, PK and PD paremeters are important for antibiotics. We believe that decolonization effect of antibiotics could be added to these paremeters.
