CRKP infections are associated with significant morbidity and mortality. Polymyxins are the last resort of the antibiotics used in the management of these infections. The safety and efficacy of polymyxin drugs are compromised due to their poor pharmacokinetic properties and toxicity [3]. Ceftazidime-avibactam is a newer beta-lactam-beta-lactamase inhibitor which has efficacy against OXA-48 and KPC carbapenemases. Multiple studies in the past have proven the superior safety and efficacy of ceftazidime-avibactam compared to polymyxins in CRKP infections mediated by KPC and OXA-48 [11–18] (Table 3).
We reviewed eight observational studies on CRKP infections (KPC or OXA-48) treated by ceftazidime-avibactam and polymyxins. The mortality with ceftazidime-avibactam ranged from 8-37% whereas it was 30-70% for patients treated with polymyxin. All studies showed better mortality outcomes with ceftazidime-avibactam. Despite the low quality of evidence, these are important findings that will help standardize the treatment of CRKP infections. We found only one study that focused on the efficacy of ceftazidime-avibactam when combined with aztreonam in patients with NDM infections [17]. This is particularly important for countries like India, where NDM carbapenemase production is the major mechanism of carbapenem resistance. To the best of our knowledge, this is the second study with real-world data comparing the efficacy of ceftazidime-avibactam with aztreonam combination against polymyxins.
Table 3: Mortality associated with ceftazidime-avibactam and polymyxins in previous studies on CRKP infections
S.n
|
Author
|
Study design
|
Sample size
|
Predominant carbapenemases
|
Outcome
|
Ceftazidime-avibactam
|
Polymyxins
|
1
|
Van Duin et al [11]
|
Prospective observational study
|
137
|
KPC
|
30-day all-cause mortality
|
8%
|
33%
|
2
|
Almangour et al [12]
|
Retrospective observational cohort study
|
230
|
OXA-48
|
In-hospital mortality
|
35%
|
44%
|
3
|
Hakeam et al [13]
|
Retrospective cohort study
|
61
|
OXA-48
|
30-day mortality
|
37.5%
|
41.4%
|
4
|
Fang et al [14]
|
Retrospective observational study
|
115
|
-
|
28-day mortality
|
8.1%
|
29.5%
|
5
|
Castón et al [15]
|
Retrospective observational study
|
339
|
OXA-48 and KPC
|
|
21.9%
|
46.9%
|
7
|
Satlin et al [16]
|
Retrospective observational study
|
137
|
KPC
|
30-day mortality
|
10%
|
31%
|
8.
|
Marco falcone et al [17]
|
Prospective observational study
|
102 patients
|
Metallo beta lactamases
|
30-day mortality
|
19.2%
|
44%
|
9.
|
Current study
|
Retrospective observational study
|
106
|
OXA-48 and NDM
|
30-day mortality
|
29%
|
57%
|
In our study, there was significantly lower overall 30-day mortality and time to mortality in the ceftazidime-avibactam arm compared to polymyxins, and the results remained significant after adjusting for covariates. The clinical success was more in the ceftazidime-avibactam arm compared to the polymyxin arm, however, it did not reach statistical significance. These findings are similar to the experience obtained with studies on KPC and OXA. The overall mortality was, however, very high in our cohort. This was mostly because we included critically ill patients with high baseline severity. The mortality in the ceftazidime-avibactam arm remained low even after adjusting for severity. Our results are different from a recently published study from South India that has shown no difference in mortality after introducing ceftazidime-avibactam ± aztreonam for the treatment of CRKP infections. However, it must be noted here that the study was conducted in two different timelines where improvement in supportive care could influence the results despite the adjustments [19].
There was no difference between the incidence of the new requirement of dialysis, new onset thrombocytopenia, or liver dysfunction between the ceftazidime-avibactam and the polymyxin arms in this study. We were not able to assess the neurological side effects in our study due to its retrospective design. Further high usage of adjunctive antibiotics in both groups makes it difficult to attribute a side effect to one particular antibiotic
It is important to note here that the recommended method for polymyxin susceptibility testing is micro broth dilution. The susceptibility testing for polymyxin in this study was done by the VITEK2 method which is not reliable [20, 21]. It is possible that CRKP isolates in the polymyxin arm were resistant in the control arm and this would explain the poor outcome. In an ideal scenario, the susceptibility of polymyxins should be available to guide treatment decisions. In resource-limited settings, in the real-world, polymyxin susceptibility by micro broth dilution is rarely available, even in the best centres. Similarly, Xpert Carba-R or synergy testing or ceftazidime-avibactam susceptibility is not available in all centres. The intention of this study is to evaluate the impact of these drugs on the outcome.
Since both drug strategies have a poor evidence base, it is common practice to combine the antibiotics in question with adjunctive antibiotics. Combination therapy with adjunctive antibiotics was used in 71/106 (66.9%) in our study cohort. The most used antibiotic in combination with polymyxin and ceftazidime-avibactam was tigecycline followed by carbapenems. To avoid polymyxin monotherapy, adjunctive antibiotics were more commonly added to the polymyxin arm. Despite this, the mortality in the polymyxin arm remained higher. The role of adding an additional antibiotic with proven in vitro susceptibility needs to be studied further.
Our study has a few limitations. Being a retrospective study, the unknown confounding factors or confounding by indication could have influenced our study results. At baseline more patients in the polymyxin arm compared to ceftazidime-avibactam + aztreonam were on mechanical ventilation, requiring vasopressors and dialysis, however, the difference was not statistically significant. Our results did not change after adjusting to these covariates.
Despite these limitations, our study results suggest better efficacy of ceftazidime-avibactam in combination with aztreonam compared to polymyxins in the treatment of CRKP isolates in settings with a high prevalence of NDM carbapenemases. Further randomized controlled studies are required to confirm our results.