Background
Increasing antibacterial resistance and associated infections have become a global health threat [
1]. Antibacterial resistance can lead to unresponsiveness to treatment, resulting in persistent illness and increased risk of death [
2]. Multidrug-resistant organisms harm not only patients who are infected but also those not infected with multidrug-resistant bacteria. All patients are affected by the lack of appropriate antibiotic regimens due to less efficient antimicrobial agents [
3].
Among the many antibiotic-resistant bacteria infections, carbapenem-resistant Gram-negative bacteria (CRGN), which mainly include carbapenem-resistant Enterobacteriaceae (CRE),
Acinetobacter baumannii (CRAB) and
Pseudomonas aeruginosa (CRPA), pose a significant threat to humanity in terms of mortality, medical burden, and trends in antimicrobial resistance [
4]. The emergence and spread of CRGN are considered to be mainly related to the hydrolysis of carbapenemase, changes in membrane permeability [
5], overuse of antimicrobials, and healthcare-associated infections [
6]. In a report, the European Union and European Economic Area estimated that the burden of carbapenem-resistant
Klebsiella pneumonia (CRKP) increased the most (by 6.16 times) among the bacterial species studied in terms of the number of infections and deaths during 2007–2015, followed by carbapenem-resistant
Escherichia coli (CREC) [
7]. In the face of rapidly increasing antimicrobial resistance, the World Health Organization (WHO) ranked CRGN as the most urgent bacteria in 2017 [
8].
According to data reported by the China Antibacterial Surveillance Network, the usage rate and defined daily doses (DDDs)/1000 patient-days in inpatients have shown a downward trend from 2011 to 2017 since the rectification of the clinical application of antimicrobials in China in 2011 [
9]. However, consumption of carbapenems has increased, with a rise of carbapenem resistance rate in Gram-negative bacteria. Therefore, it was proposed to strengthen the management of carbapenems and tigecycline to reduce carbapenem resistance in China in February 2017. It was shown that reduced antibacterial use is positively associated with improved bacterial resistance without negatively affecting medical quality indexes [
10]. Since use of antibiotics can create selective pressure favoring the spread of resistance [
11], assessing the association of antibiotic usage with antimicrobial resistance could help local physicians and decision-makers make better use of antibacterials and distribute healthcare funds more effectively, while improving infection control strategies [
12]. Therefore, the aim of this study was to describe the changing trend of antibacterial usage and the prevalence of CRGN from 2012 to 2019 in a tertiary hospital and to evaluate the correlation between them.
Methods
Study design and setting
The data in this study were collected from local monitoring of carbapenem resistance in Gram-negative bacteria and antibacterial consumption at the Sixth Affiliated Hospital of Guangxi Medical University, which is a comprehensive, tertiary care, university-affiliated, and teaching hospital in the south of China. It is also the largest hospital in Southeast Guangxi, with currently 2400 available beds.
Bacterial isolates and susceptibility testing
Antibiotic resistance data, which were provided by the hospital’s clinical microbiology laboratory, were extracted from the hospital information system. A total of 37,716 cases with data on Gram-negative bacterial resistance were recorded, including all positive clinical specimens, and 34,489 cases with data on six isolated species, including A. baumannii, Pseudomonas aeruginosa (P. aeruginosa), Klebsiella pneumonia (K. pneumonia), E. coli, E. cloacae, and B. cepacia, which were target bacteria to be monitored in this hospital. The resistance rate was reported as the percentage of resistant isolates among all tested isolates. Antimicrobial susceptibility was tested based on the latest performance standards set by the Clinical and Laboratory Standards Institute (CLSI). Tested isolates that had intermediately resistance were eliminated from this analysis. We retained the first isolate of duplicate test results from the same patient during the same in-patient stay. The isolation rate was calculated as the percentage of total gram-negative bacterial isolates.
Antibacterial consumption
Antibacterial consumption data from 2012 to 2019 were obtained from the hospital information system. According to the Guidelines for Anatomical Therapeutic Chemical (ATC) classification and DDD assignment 23rd edition (2020, [
13]), which was developed by the WHO, data on antibacterial consumption were expressed as DDDs/1000 patient-days. Antibacterials were classified according to the ATC classification system.
Statistical analysis
The changing trends of antimicrobial use and resistance were analyzed by linear regression. Pearson correlation analysis was used for evaluating the relationship between antibiotic consumption and bacterial resistance rates. SPSS 18.0 (SPSS, USA) was used for statistical analysis. A p-value < 0.05 was considered statistically significant.
Discussion
The present study revealed significant associations of consumption of antibiotics with the rates of carbapenem resistance in Gram-negative bacteria.
As shown above, cephalosporins (except for C/BLI combinations) had the largest consumption variety, despite showing a downward trend. In addition, consumption of BL/BLI combinations showed an increasing trend, and piperacillin-tazobactam was the most frequently used antibiotic combination, corroborating previous findings [
14]. It is known that a history of using 3rd generation cephalosporins independently predicts extended-spectrum β-lactamase (ESBL)-producing
K. pneumoniae or
E. coli bacteremia [
15]. The replacement of 3rd generation cephalosporins by BL/BLI combinations is suitable for reducing the incidence of ESBL-producing Enterobacteriaceae [
16‐
18]. Additionally, our results revealed significantly increased consumption of carbapenems, corroborating global trends [
19,
20]. Furthermore, the present findings confirmed that elevated carbapenem usage was associated with increased number of infections resistant to carbapenems. Yet, it is undeniable that overuse of carbapenems presents an important problem, and carbapenems should be used more rationally.
In this study, the isolation rates of
E. coli,
E. cloacae, and
B. cepacia showed a downward trend. As expected,
E. coli was the most common isolate in 2012–2019, but still highly sensitive to carbapenems. Carbapenem resistance rates in
K. pneumoniae were significantly lower than the national levels and relatively stable, ranging from 2% in 2012 to 3% in 2019. Previous studies in China revealed CRKP incidence increased from 8.9% in 2005 to 26.3% in 2018 [
21,
22]. Surprisingly, the detection rate of CRKP declined for the first time in 2019 [
21].
CRAB had the highest rate among all assessed species, rapidly rising between 2012 and 2019, as reported for the whole country [
21‐
23]. Since
A. baumannii has a resistance rate above 50% to most antibacterial drugs, susceptibility testing should always be performed before its application [
24]. Our results showed significant positive associations of resistance to carbapenems in
A. baumannii and
E. coli with carbapenem consumption, corroborating a large multicenter study in China [
25]. Previous studies have indicated that CRAB rate increase is related to carbapenem exposure [
26,
27].
Different from findings by Tan et al [
28], which showed no significant association between usage of BL/BLI combinations and CRAB prevalence,
A. baumannii resistance to carbapenems was positively correlated with cefoperazone-sulbactam usage in the present study. We also observed a positive correlation between tetracycline consumption and
A. baumannii resistance to carbapenems. The cefoperazone/sulbactam-based combination regimen, which is usually combined with tigecycline, minocycline, carbapenems or aminoglycosides, is the commonest treatment option for carbapenem-resistant and extensively drug-resistant
A. baumannii infections [
29]. Therefore, it is understandable that increased consumption of cefoperazone-sulbactam and tetracyclines is associated with carbapenem resistance in
A. baumannii.
Unexpectedly, increasing use of carbapenems and P/BLI combinations was significantly correlated with reduced resistance of
E. cloacae to carbapenems. Nevertheless
E. cloacae has genomic heterogeneity, and consumption of carbapenems is an independent risk factor for infection of imipenem-heteroresistant
E. cloacae [
30], which may be selected for highly resistant and pathogenic strains. Accordingly, this finding did not provide a strong basis for the selection of antimicrobials to
E. cloacae infections. Given the lack of related research on the correlation between antibacterial consumption and carbapenem resistance in
E. cloacae, further research is needed to verify this association. It is difficult to treat infections caused by
B. cepacia, because of the high level of intrinsic and acquired resistance to many antimicrobial agents. Carbapenems are one of the most reliable antibacterial drugs [
31]. Yet, we found that use of quinolones was positively correlated with
B. cepacia resistance to carbapenems, which prompts more attention to be paid to the use of quinolones to slow down carbapenem resistance in
B. cepacia.
Because of the urgent need to improve the rational use of antibiotics, a multidisciplinary antibiotic team, comprising infectious disease specialists, microbiologists, clinical pharmacists, and clinicians, was established in 2017 in our hospital. This team ensures appropriate administration of antibiotic therapy for inpatients, pointing out medication errors and offering corrective suggestions. Coincidentally, carbapenem resistance rates in K. pneumoniae, P. aeruginosa, E. cloacae, and B. cepacia showed a downward trend from 2017 to 2019. Consequently, reducing the unreasonable use of antibacterial drugs may be an effective measure for reducing the spread of CRGN; however, further studies are needed to confirm these observations.
There were some limitations in this study. Firstly, it was conducted only at one tertiary hospital, while carbapenem resistance rates and antibacterial consumption may vary widely across hospitals; therefore, a multicenter study with more relevant data is needed to explore the correlation between antibacterial consumption and CRGN. In addition, due to the retrospective nature of this study, safety could not be examined, and appropriate and inappropriate uses of antibiotics could not be clearly distinguished. Furthermore, the development of carbapenem resistance in Gram-negative infection is associated with various factors, and the selective pressure of antibacterial drugs is only one of them. Multi-factor analysis needs to be further studied.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.