Introduction
The high prevalence of antimicrobial resistance (AMR) to broad-spectrum antibiotics in gram-negative bacteria (GNR), such as carbapenem-resistant Enterobacterales (CRE), Pseudomonas aeruginosa, and Acinetobacter baumannii, has been a global therapeutic challenge for the last two decades.1–4 The World Health Organization has categorized these bacteria as ‘Priority 1 (Critical)’ pathogens in 2024.5
Ceftazidime-avibactam (CAZ-AVI), a combination of a third-generation cephalosporin and a beta-lactamase inhibitor, is one of the new antimicrobials developed to address concerns associated with AMR, such as treatment failure and increasing morbidity, mortality, and treatment costs. Avibactam is a novel beta-lactamase inhibitor that enhances the activity of ceftazidime, a third-generation cephalosporin, by hydrolyzing most beta-lactamases.6–8
The Infectious Diseases Society of America and published research have supported the use of CAZ-AVI for complicated CRE infections, especially in severe cases of pneumonia, complicated urinary tract infections (UTIs), and intra-abdominal wound infections.8,9 It exhibits potent bactericidal activity against GNR harboring extended-spectrum beta-lactamases, AmpC beta-lactamases, and serine carbapenemases. However, it has no activity against metallo-beta-lactamase (MBL)-producing organisms.7,8,10
Antibiotic options for carbapenem-resistant GNR are limited in our country due to the unavailability of other options such as plazomicin, eravacycline, CAZ-AVI and aztreonam combination, and cefiderocol. Although tigecycline and colistin are available, they are associated with serious side effects, including blood dyscrasias and nephrotoxicity, respectively. There is limited published data on the in-vitro efficacy of CAZ-AVI from our region. Hence, we aimed to evaluate the in-vitro susceptibility of CAZ-AVI in carbapenem-resistant GNR in this cross-sectional study from a tertiary care hospital laboratory.
Materials and methods
Study design and setting
This observational, cross-sectional study was conducted at the Microbiology Laboratory of Indus Hospital & Health Network in Karachi, Pakistan, from January 2023 to October 2024. All patient samples from the outpatient (OPD), emergency (ER), and inpatient (IPD) departments, including blood, urine, tissue, and respiratory specimens that were positive for carbapenem-resistant GNR, were included in the study using consecutive, non-probability sampling. Clinical samples included in the study were obtained from the hospital's biobank and fully de-identified prior to analysis to ensure patient confidentiality. Bacterial identification and antimicrobial susceptibility testing were performed after the clinical samples were processed for bacterial culture in the laboratory. After the bacterial strains were characterized as carbapenem-resistant GNR, CAZ-AVI susceptibilities were determined.
Identification and susceptibility testing
Samples were inoculated onto Sheep Blood Agar, Chocolate Agar, and MacConkey Agar, except for urine, which was inoculated onto Cystine Lactose Electrolyte Deficient Medium agar and incubated at 35 ± 2 °C for 24–48 h. Preliminary bacterial identification was performed by gram staining and biochemical testing, including H2S production, oxidase, triple sugar iron, citrate, indole, motility, and urease tests. Identification was confirmed using API® ID strips and APIWEBTM database (bioMérieux). The Kirby-Bauer disc diffusion method was used to perform antibiotic susceptibility testing for carbapenems and CAZ-AVI on Mueller-Hinton Agar. Zone of inhibition diameters were interpreted to assess antimicrobial susceptibility as sensitive, intermediate, or resistant in accordance with the clinical breakpoints outlined in the Clinical and Laboratory Standards Institute M100 guidelines.9
Data collection and statistical analysis
Demographic, clinical, and laboratory data were retrieved from the hospital’s electronic records and maintained on a standardized proforma. Data were entered into Microsoft Excel (Microsoft Excel 2013 32-bit) and MedCalc Statistical Software version 20.027 (MedCalc Softwarebv, Ostend, Belgium) for statistical analysis. Data were double-checked for accuracy during transfer from hard copies to the statistical software. To ensure confidentiality, data were de-identified before analysis. The Chi-squared test was used to identify significant associations between demographic characteristics and clinical diagnoses with the frequency of CAZ-AVI susceptibility. A p-value of ≤0.05 was considered statistically significant.
Results
Study isolates
During the study period, a total of 158 carbapenem-resistant GNR were reported, of which 92 (58%) were Enterobacterales and 66 (42%) were P. aeruginosa. Among the Enterobacterales, 65% (n = 60) of the isolates were E. coli, and 35% (n = 32) were Klebsiella spp. (CAZ-AVI. 1). The majority of patients were males (n = 85; 54%) compared to females (n = 73; 46%). Regarding age distribution, 55% (n = 87) of the patients were between 0–25 years, 15% (n = 23) were between 26–50 years, and 30% (n = 48) were between 51–80 years of age. The majority of patients (n = 76; 48%) were admitted to IPD, followed by ER (n = 47; 30%) and OPD (n = 35; 22%). The clinical diagnoses were UTI in 46% (n = 72), bacteremia in 19% (n = 30), skin and soft tissue infection in 27% (n = 43), and lower respiratory tract infection in 8% (n = 13) of the cases (Fig. 1).
CAZ-AVI susceptibility
Of the total carbapenem-resistant isolates (n = 158), CAZ-AVI susceptibility was observed in 11% (n = 17) of the isolates (Table 1). Among the organisms, 13% (n = 4) of Klebsiella spp., 7% (n = 4) of E. coli, and 14% (n = 9) of P. aeruginosa strains showed CAZ-AVI susceptibility (Fig. 2). CAZ-AVI-susceptible strains were most predominant among patients aged 26–50 years (n = 6; 35%), the majority of whom were female (n = 10; 59%). In terms of location, most of the CAZ-AVI-susceptible isolates were from IPD patients (n = 8; 47%), followed by ER (n = 5; 29%) and OPD (n = 4; 24%). Clinical samples from patients with UTI grew the most CAZ-AVI-susceptible strains (n = 9; 53%), followed by patients with skin and soft tissue infection (n = 4; 23%), lower respiratory tract infection (n = 3; 18%), and bacteremia (n = 1; 6%) (Fig. 1). The association of CAZ-AVI susceptibility with the age group 26–50 years was found to be statistically significant (p = 0.02) (Table 1).
Table 1Association of patient demographics and clinical diagnosis with CAZ-AVI susceptibility among different carbapenem-resistant gram-negative organisms
| Characteristic | CAZ-AVI Susceptibility n (%) p-value
|
|---|
| E. coli (n = 60) | Klebsiella spp. (n = 32) | P. eruginosa (n = 66) | Total (n = 158) |
|---|
| Overall positivity |
| 4/60 (7%), p = 0.2 | 4/32 (13%), p = 0.7 | 9/66 (14%), p = 0.3 | 17/158 (11%), p = 1 |
|
| Age |
| 0–25 | 2/35 (6%), p = 0.7 | 2/20 (10%), p = 0.5 | 2/32 (6%), p = 0.9 | 6/87 (7%), p = 0.09 |
| 26–50 | 1/5 (20%), p = 0.7 | 0/1 (0%), p = 0.9 | 5/17 (29%), p = 0.6 | 6/23 (26%), p = 0.02* |
| 51–80 | 1/20 (5%), p = 0.3 | 2/11 (18%), p = 0.4 | 2/17(12%), p = 0.8 | 5/48 (10%), p = 0.9 |
|
| Gender |
| Male | 1/27 (4%), p = 0.3 | 1/13 (8%), p = 0.9 | 5/45 (11%), p = 0.3 | 7/85 (8%), p = 0.3 |
| Female | 3/33 (9%), p = 0.3 | 3/19 (16%), p = 0.8 | 4/21 (19%), p = 0.4 | 10/73 (14%), p = 0.3 |
|
| Department |
| IPD | 1/26 (4%), p = 0.2 | 1/18 (6%), p = 0.4 | 6/32 (19%), p = 0.07* | 8/76 (11%), p = 0.9 |
| OPD | 1/13 (8%), p = 0.6 | 2/5 (40%), p = 0.06* | 1/17 (6%), p = 0.3 | 4/35 (11%), p = 0.9 |
| ER | 2/21 (10%), p = 0.8 | 1/9 (11%), p = 1 | 2/17 (12%), p = 0.9 | 5/47 (11%), p = 1 |
|
| Diagnosis |
| Bacteremia | 0/8 (0%), p = 0.9 | 1/14 (7%), p = 0.4 | 0/8 (0%), p = 0.9 | 1/30 (3%), p = 0.2 |
| UTI | 3/32 (9%), p = 0.5 | 2/10 (20%), p = 0.4 | 4/30 (13%), p = 0.5 | 9/72 (13%), p = 0.5 |
| LRTI | 0/2 (0%), p = 0.7 | 0/3 (0%), p = 0.5 | 3/8 (38%), p = 0.2 | 3/13 (23%), p = 0.2 |
| SSTI | 1/18 (6%), p = 0.5 | 1/5 (20%), p = 0.4 | 2/20 (10%), p = 0.9 | 4/43 (9%), p = 0.7 |
Discussion
Our study findings revealed a CAZ-AVI susceptibility rate of 11% among carbapenem-resistant GNR tested from various clinical samples. CAZ-AVI is a novel antibiotic that combines the cephalosporin ceftazidime with a β-lactamase inhibitor (avibactam), which is effective against serine carbapenemases produced by Enterobacterales and P. aeruginosa.11,12 It has shown activity against highly resistant pathogens globally, with improved clinical outcomes observed, especially in intra-abdominal infections and UTIs.2,13,14 Wang Y. et al. demonstrated its effectiveness in treating KPC- and OXA-48 CRE bacteremia, leading to better clinical outcomes.15 However, CAZ-AVI is ineffective against MBL-positive strains and KPC enzymes with substitutions such as the D179Y substitution.4,10,16,17
Commonly reported carbapenemases in Enterobacterales and P. aeruginosa strains from Pakistan include New Delhi MBL (NDM)-1, NDM-7, VIM, IMP, OXA-48, and KPC-2, showing high prevalence rates of MBL producers in our region.10,18,19 Although the use of CAZ-AVI is relatively recent in Pakistan, our findings demonstrate a very low overall susceptibility rate (11%), consistent with studies conducted in other countries revealing rising resistance to CAZ-AVI.8,15 One potential explanation is the prevalence of MBLs in the region, highlighted by a local study attributing this resistance to NDM production.10 While CAZ-AVI susceptibility is not evident in MBL-producing strains, the synergistic combination of CAZ-AVI and aztreonam has shown promising results in several in-vitro studies.20–22 Clinical data on the efficacy of such combinations is limited and requires further investigation.
A recent study from Pakistan showed findings consistent with ours. Of the total isolates tested, 77% of Enterobacterales and 80.1% of P. aeruginosa strains showed resistance to CAZ-AVI, predominantly due to MBL production.10 Another surveillance study from India demonstrated 51% susceptibility in carbapenem-resistant K. pneumoniae and 24% in carbapenem-resistant E. coli against CAZ-AVI.8 Molecular characterization of the isolates showed the presence of carbapenemases, with OXA-48-like enzymes found in 52% of K. pneumoniae isolates, and 27% co-producing NDM and OXA-48-like enzymes. In E. coli isolates, NDM was detected in 68% of strains, followed by OXA-48-like enzymes in 24%.8 Furthermore, a global surveillance program report from the Asia-Pacific region also supports our findings.23
The optimal treatment for MBL-producing gram-negative bacterial infections remains unclear. However, recent guidelines by the Infectious Diseases Society of America recommend the combination of CAZ-AVI and aztreonam as the preferred treatment option, as aztreonam is stable against hydrolysis by MBLs, and recent studies have shown good clinical outcomes with this synergistic effect.24,25 In addition to colistin and aminoglycosides, other recommended drugs include newer agents like cefiderocol and ervacycline, which can be used in combination with other active agents when available treatment options are limited.26,27
Resistance to CAZ-AVI among carbapenem-resistant GNR in South Asia is a significant concern. Studies have shown that pre-existing resistance mechanisms, such as the presence of certain β-lactamases due to amino acid substitutions and efflux pumps, contribute to the high resistance rates observed in this region.28,29 For instance, the emergence of blaKPC genes in P. aeruginosa in China highlights the role of specific resistance mechanisms in limiting the efficacy of CAZ-AVI.30 Furthermore, the high prevalence of OXA-48-like enzymes in some parts of Asia also impacts CAZ-AVI susceptibility.31 Regional disparities in CAZ-AVI resistance, such as those observed in Asia, underscore the need for tailored strategies based on local resistance profiles.
The reliance on conventional antibiotics such as CAZ-AVI, which are ineffective against MBL producers, poses significant limitations in the era of escalating AMR. A crucial consideration for mitigating AMR is the use of alternative strategies, such as phage therapy, in which bacteriophages specifically target and lyse bacteria. There is supportive evidence that phage therapy can enhance antibiotic efficacy and restore sensitivity in resistant strains.32,33 Additionally, Clustered Regularly Interspaced Short Palindromic Repeats-based antimicrobials and antimicrobial peptides represent innovative approaches that could provide targeted and effective treatments. Novel efflux pump inhibitors also hold promise by preventing bacteria from expelling antibiotics, thereby increasing their effectiveness.34 These emerging options not only offer new therapeutic pathways but also underscore the need for a multifaceted approach to address the complex challenges of AMR. While these alternatives require further research to establish their efficacy and safety, they provide a forward-looking perspective on combating AMR.
There are several limitations in our study. Firstly, molecular characterization to identify the exact underlying resistance mechanisms was not performed due to the technical and financial constraints of conducting such analyses, which require specialized molecular techniques for detecting specific resistance genes or sequencing to identify mutations in β-lactamases. Additionally, our study focused primarily on the prevalence of CAZ-AVI resistance, and we aimed to perform future studies prioritizing molecular characterization to provide a comprehensive understanding of resistance mechanisms against CAZ-AVI. Secondly, we acknowledge the limitations related to potential biases introduced by the sampling method. The convenience sampling approach may have resulted in an overrepresentation or underrepresentation of certain groups, which could have limited the generalizability of our results to the broader population. To address this, we plan to implement strategies for mitigating these biases in future studies through more diverse sampling methods and statistical adjustments. Furthermore, statistical significance for the association of CAZ-AVI susceptibility among bacterial isolates could not be demonstrated adequately, possibly due to the smaller sample size. This necessitates a multi-center study incorporating a larger number of clinical samples and bacterial isolates from the wider population of Pakistan. Lastly, there is a need to evaluate the synergistic combination of CAZ-AVI and aztreonam, which has demonstrated in-vitro and clinical efficacy in several past studies.20–22,25
Future directions
The findings of this study highlight the challenges in addressing the global rise of AMR. Future research efforts should focus on elucidating the molecular mechanisms underlying resistance to inform the development of targeted therapeutic strategies. Moreover, there is an urgent need for enhanced surveillance systems to monitor AMR trends and for comprehensive educational programs aimed at promoting judicious antibiotic stewardship among healthcare providers. Exploring combination therapies, such as pairing CAZ-AVI with other antibiotics like aztreonam, may offer a promising approach to mitigate resistance development and improve clinical outcomes. Additionally, investments in novel antimicrobial agents and alternative treatments, including bacteriophage therapy, could provide critical solutions in the face of evolving AMR. Ultimately, a multifaceted strategy encompassing public health policy, clinical practice, and research will be essential for effectively managing and treating infections caused by carbapenem-resistant GNR.
Conclusions
Our study demonstrated low CAZ-AVI susceptibility in carbapenem-resistant gram-negative bacteria, posing a significant challenge due to the limited availability of therapeutic options. We recommend large-scale genotypic studies to better understand the mechanisms of resistance to this novel antibiotic, which has not been widely prescribed.
Declarations
Ethical statement
This study was carried out after receiving approval from the Institutional Review Board of the hospital (IHHN_IRB_2023_10_028). The requirement for informed consent was waived as all data were de-identified before analysis.
Data sharing statement
All data generated during the course of this study are included in this article.
Funding
The authors did not receive any funding for this work.
Conflict of interest
The authors declare no competing interests.
Authors’ contributions
Conceptualization (FA, NK), data curation (MA), formal analysis (MA, MAK), investigation (MA), methodology (FA, NK), project administration (FA), resources (NK), software (MAK), supervision (FA), validation (MA, FA), visualization (MAK), writing – original draft (MA), and writing – review & editing (FA, MAK, NK).