Emergence of KPC-2-Producing ST11-K64 Hypervirulent Klebsiella Pneumoniae for the Elderly in Intensive Care Unit: Antimicrobial Resistance, Molecular and Clinical Characteristics

Background: In recent years, carbapenem-resistant hypervirulent Klebsiella pneumoniae (CR-hvKP) has been increasingly reported and poses severe therapeutic challenge to global public health, but has not yet been systematically studied in elderly patients. This study aimed to investigate the clinical and molecular characteristics and risk factors of CR-hvKP infections in elderly patients. Methods: We retrospectively investigated elderly patients with carbapenem-resistant Klebsiella pneumoniae (CRKP) infections in intensive care unit (ICU),between January 2020 and December 2020, the clinical data being collected from medical records, and microbiological data including antimicrobial susceptibility testing, phenotype experiment, carbapenemases production, string test, virulence gene, capsular serotype-specic (cps) genes and multilocus sequence typing, of the CR-hvKP group dened by the presence of some combination of rmpA, rmpA2, iucA, iroB, and peg-344 was compared with those of carbapenem-resistant non-hypervirulent Klebsiella pneumoniae (CR-non-hvKP) strains. Results: Of 80 CRKP strains, 51(63.8%) met the denition of CR-hvKP and 7 of them had all ve of the virulence genes tested. The main mechanism of resistance to carbapenems found in CR-hvKP strains was the presence of the bla KPC-2 gene. There was no statistical signicance in the resistance rates of antimicrobial agents between the CR-hvKP and CR-non-hvKP subgroups (p ≥ 0.05). Compared with the minimum inhibitory concentration (MIC) 50 values of CR-non-hvKP group, those of antimicrobials, including ceftazidime, ceftazidime/avibactam, imipenem/avibactam, tigecycline, levooxacin, cefoperazone-sulbactam, exhibited higher levels in the CR-hvKP group. Avibactam (4 µg/mL) signicantly decreased the MIC90 values more than sixteen-fold than that of ceftazidime and imipenem alone against KPC-2-producing Klebsiella pneumoniae. K64 and ST11 were highly prevalent and strongly associated with CR-hvKP (P<0.05). Cardiovascular disease (odds ratio [OR] = 11.956) and ST11-K64 (OR= 8.385) appeared to be independent variables associated with CR-hvKP infection by multivariate analysis. Conclusion: The CR-hvKP strains showed higher MIC50 values than CR-non-hvKP strains. KPC-2-producing ST11-K64 CR-hvKP is emerging, which might become new signicant “superbugs” and a threat to elderly patients. carbapenem-resistant hypervirulent Klebsiella pneumoniae; CR-non-hvKP:carbapenem-resistant non-hypervirulent Klebsiella pneumoniae; PBA:Phenyloronic acid; EDTA:ethylenediamine tetra-acetic acid; E. coli: Escherichia coli. EUCAST: European Committee on Antimicrobial Susceptibility Testing; FDA:Food and Drug Administration; hvKP: hypervirulent Klebsiella pneumoniae; K. pneumoniae: Klebsiella pneumoniae; ICU: intensive care unit ;KPC: Klebsiella pneumoniae carbapenemase; MBL:metal β-lactamase; MIC: Minimal Inhibitory Concentration; MLST: Multilocus sequence type; NCBI: National Center for Biotechnology Information; OR: odds ratio: PCR: Polymerase Chain Reaction; ST:


Introduction
Klebsiella pneumoniae (K. pneumoniae), an important community-acquired and nosocomial pathogen, especially in the intensive care unit (ICU), causes widespread infections including pneumonia, bacteremia, urinary tract infection, endophthalmitis, liver abscesses and sometimes even life-threatening septic shock. Two clinical pathotypes: hypervirulent Klebsiella pneumoniae (hvKP) and carbapenem resistant Klebsiella pneumoniae (CRKP) often cause life-threatening infections [1] . In the conventional sense, hvKP is seldom resistant to antibiotics, however, because of virulence genes and antibiotic resistance genes transmitted by mobile genetic elements, increasing researches show that carbapenem-resistance hypervirulent klebsilla pnemoniae (CR-hvKP) strains are widely found [2][3][4] . In China, the prevalence of CR-hvKP infections is 0% 2 5.8%, with Henan and Shandong being the most serious [5] .The emergence of CR-hvKP could lead to the next clinical crisis, has attracted worldwide attention [2] .
During the last decade, the global spread, of CRKP posed a severe challenge to public health, particularly in China. According to the China Antimicrobial Resistance Surveillance System report [6] , most CRKP strains were isolated from ICU, and were common in the elderly. The main resistant mechanism for CRKP in China is the production of KPC-2 carbapenemase, and ST11 is the dominant sequence type (ST) [7] . A recent genomic analysis showed that ST11 classic CRKP strains can acquire the pLVPK-like plasmid and become CR-hvKP strains [2] .
In the past, the de nition of hvKP relied on clinical manifestations and phenotypes, which proven to have poor speci city and sensitivity [8] .The rmpA, rmpA2,iucA, iroN and peg-344 genes on virulence plasmids can be used for the identi cation of hvKP and classic KP(cKP) strains with pinpoint accuracy [8] . The emergence of CR-hvKP will be a great challenge for clinician and result in adverse outcome. So far, researches on CR-hvKP in elderly patients remain limited [5] . Therefore, a cognitive of the risk factors of infections, microbiological and molecular characteristics in CR-hvKP will provide important insights into the control and treatment for CR-hvKP infections among the elderly patients.
In response to this emerging problem, a retrospective study in elderly patients in a Chinese hospital was conducted, the aim was to advance our cognitive of the clinical characteristics, risk factors, hypervirulence and epidemiology of CR-hvKP infections based on newly recognized biomarkers.

Bacterial strains and identi cation
We collected clinical CRKP strains from patients over sixty-ve years old between January 2020 and December 2020 from ICU in Yueyang Hospital of Integrated Traditional Chinese and Western Medicine, Shanghai University of Traditional Chinese Medicine, and the capacity of the hospital is more than 1200 beds. All the strains were collected consecutively and non-duplicated, only the rst strain was collected from the same patient. All the K. pneumoniaeidenti cation and the drug resistance of imipenem and ertapenem were detected by Vitek 2 system (bioMérieux, France) and further species identi cation was con rmed with MALDI-TOF MS system (Bruker Daltonics, USA), the drug resistance of meropenem was tested by using a Kirby-Bauer disk diffusion susceptibility test. CRKP was de ned as that resistant to either imipenem, meropenem or ertapenem [9] in accordance with the breakpoints of 2020 Clinical and Laboratory Standards Institute (CLSI) guidelines [10] . We collected all clinical information for CRKP positive patients from electronic medical records, including basic demographics, underlying diseases, clinical presentations, exposure to hospital care, treatment process, and outcomes.

Antimicrobial Susceptibility Testing
The Antimicrobial Susceptibility Testing (AST) for the CRKP strains were further performed by the broth microdilution assay (BIO-NONT, China). The antibiotics included ceftazidime, ceftazidime/avibactam, cefepime, aztreonam, piperacillin/tazobactam, imipenem,imipenem/ avibactam, meropenem, amikacin, polymyxin B, tigecycline, levo oxacin, cefoperazone/sulbactam. The AST breakpoints of 2020 CLSI guideline were adopted [10] with exception of tigecycline and polymyxin B. The breakpoint of tigecycline was based on the US Food and Drug Administration(FDA) standard [11] ,and the breakpoint of polymyxin B was based on the European Committee on Antimictobial Susceptibiity Testing(EUCAST) standard [12] . The breakpoints of imipenem/avibactam and cefoperazone/sulbactam were referred to imipenem and cefoperazone in CLSI, respectively [13][14] .The standard strains E. coliATCC 25922 and K. pneumoniae ATCC700603 were used as quality control strains in each experiment.

DNA extraction
The strains were inoculated into the blood agar plate and incubated overnight at 35℃. The colonies were suspended in 500ml sterile distilled water. The suspension was boiled at 100℃ for 10 minutes, centrifuged at 10000rmp for 13 minutes. The supernatant was used as a template for DNA reaction. The DNA was stored at -35℃.

Phenotype experiment and detection of Carbapenemases
We used phenylboronic acid (PBA)and ethylenediamine tetra-acetic acid (EDTA) [15] companying with imipenem to detect the phenotypes of KPC β-lactamase and metal β-lactamase (MBLs), respectively. In both tests, compared with imipenem alone, the diameter of the inhibition zone has increased > or =5 mm by PBA or EDTA, which is considered to be a positive combined-disc test result that shows the presence of KPC β-lactamase or MBL enzyme, respectively. The presence of carbapenemase genes (bla KPC , bla NDM , bla IMP and bla OXA-48) were detected by PCR as previously described [16][17] . The primers are listed in Supplementary table 1. The PCR ampli ed products were sequenced (Sangon Biotech, Shanghai) and the results were compared with the BLAST searches in the NCBI Genbank.
Detection of virulence-associated features and capsular serotype-speci c (cps)genotyping Hypermucoviscosity as a virulence factor was identi ed by the string test. If a viscous string longer than 5mm is generated by stretching a bacterial colony on blood agar plate with a bacteriology inoculation loop, the string test is positive [18] . Five virulence genes including rmpA, rmpA2, iucA, iroB, peg-344 [8,[19][20] and eight capsular serotype-speci c (cps) genes (K1, K2, K5, K20, K47,K54, K57 and K64) [20][21][22][23] were identi ed by polymerase chain reaction(PCR) as previous described. The primers are listed in Supplementary table 1. The CRKP strain which detected with any one of the ve virulence genes was de ned as CR-hvKP.

Statistical analysis
IBM SPSS Statistics version 19.0 was used for statistical analysis. The χ2 or Fisher's exact test were used for categorical variables. P<0.05 was considered statistically signi cant. To further identify variables associated with CR-hvKP infections, Logistic regression was used. All variables with P<0.05 were included in the multivariate analysis.

CRKP strains and Virulence-associated features
A total of 80 CRKP strains isolated from 80 patients in ICU were included in the study. The percentages of 80 CRKP strains isolated from respiratory tract, urine, blood, pus and puncture uid were 72.5%, 15.0%, 6.3%, 5.0% and 1.3%, respectively. CRKP strains were divided as CR-hvKP and carbapenem-resistant non-hypervirulent Klebsiella pneumoniae (CR-non-hvKP) based on whether they contained virulence genes. The detection of the presence of virulence genes showed that 51 (63.8%) of 80 CRKP strains were positive, suggesting carrying virulence plasmids, which were designated as CR-hvKP group, while the remaining 29 strains were designated as CR-non-hvKP group.

Susceptibility results and Carbapenemase resistance Genes
The antimicrobial susceptibility pro les of the CR-hvKP and CR-non-hvKP strains were listed in Table2.
According to phenotypic experiment results and detection of carbapenemase genes, all CR-hvKP and 96.6% (28/29) CR-non-hvKP strains showed only produced bla KPC-2 carbapenemase gene and the presence of KPClactamas. Only one CR-non-hvKP strains showed the presence of MBL enzyme, harbored bla IMP-4 .

Clinical factors for CR-hvKP infection of elderly patients
Demographic and clinical factors of patients with CR-hvKP and CR-non-hvKP infections are summarized in Table 1. The age, gender, antibiotic exposure, hospitalization within the last 90 days, or length of stay from CRKP isolation to outcome (In-hospital mortality or discharge) didn't have signi cant difference between CR-hvKP and CR-non-hvKP (P>0.05). In addition, there was no statistical difference among most underlying conditions such as pulmonary disease, diabetes mellitus, hypertension, cerebrovascular diseases, malignancy, hypoproteinemia and liver abscess (P>0.05), except cardiovascular disease (60.8% vs.17.2%, P<0.001). Gastric tube was more prevalent in CR-non-hvKP group than in CR-hvKP group (96.6% vs.80.4%, P=0.045), whereas the other invasive procedures and devices had no signi cant difference. Most of the outcomes between the two groups (P>0.05) had no statistical difference, the frequency of change of initial antibiotics due to clinical worsening was higher in CR-hvKP group than CR-non-hvKP group (58.8% vs.34.5%, P=0.036) ( Table 1).

Discussion
More than 30 years ago, a report rst described hvKP in China [24] , in recent years, reports on CR-hvKP have increased gradually, which poses a greater challenge to clinical treatment due to its high drug resistance and high virulence [5] . So far a limited number of data based on the microbiological and molecular characteristics in the elderly can be searched. According to the results of China's census in 2021, elderly patients over 65 years old account for 13.5% of the total population [25] , China has entered an aging society. As is known to all, ICU is a relatively popular setting for CRKP infections in hospitals and due to low immune function and cumulative basic diseases, the elderly patients are more prone to nosocomial infections. In this study, we focused on elderly patients hospitalized in ICU, reported antimicrobial resistance, molecular and clinical characteristics of CR-hvKP, and found the main CR-hvKP clone KPC-2-producing ST11-K64 among elderly patients.
It is worth noting that the main cause of liver abscess is hvKP infection [26][27] ,however, we found that the main specimen type of both CR-hvKP strains and CR-non-hvKP strains isolated from was sputum, and only 7.8% (4/51) patients with CR-hvKP infections had symptoms of liver abscess, suggesting that CR-hvKP can also invade other parts of the body [28] .The production of carbapenemases is the most common mechanism of drug resistant in CRKP strains [29] ,KPC-2 was the most frequently identi ed among CR-hvKP strains [2] . As reported in previous studies [30] , the CRKP strains were resistant to most antibiotics with exception of ceftazidimeavibactam, imipenem/ avibactam, polymyxin B and tigecycline, only one CR-non-hvKP strain which harbored blaIMP-4 displayed resistance to ceftazidime/avibactam, this is consistent with the fact that avibactam is inactive against MBL producing strains [31] . Avibactam, as the most effective β-lactamase inhibitor [32] , signi cantly increased the activity of ceftazidime and imipenem against KPC-producing K. pneumoniae in vitro, which decreased the MIC90 values of ceftazidime and imipenem more than sixteen-fold. We also identi ed one KPC-2-producing strain exhibited resistant to ceftazidime/ avibactam, with MIC value 16 ug/mL, it might due to OmpK35/36 defects, blaKPC-2 point mutation and high KPC expression [33][34] . We found a phenomenon that deserves special attention, namely, compared with CR-non-hvKP strains, CR-hvKP strains exhibited high level MIC50 values, especially ceftazidime/avibatam, imipenem/avibatam, tigecycline, which are our last line in treating CRKP-infected patients.
Previously, a positive string test was used as screening test for hvKP, the de nition of hvKp is controversial. so we employed ve virulence genes, including two regulator genes which have been associated with hypermucoviscous phenotype (rmpA and rmpA2), siderophore aerobactin (iucA) and salmochelin (iroB) [5] and the putative metabolite transporter peg344 [35] attribute to the hypervirulent phenotype of K. pneumoniae. We found 51 strains which de ned as CR-hvKP harbored a part of combination of these markers, rmpA2 and icuA were the most common virulence genes in our study [36] . In term of the thick polysaccharide capsules encoded by the cps [1] , unlike carbapenem-sensitive hypervirulent K. pneumoniae which belong to serotypes K1, K2, or K57 [37] , the most popular CR-hvKP clone was K64 in our study [38] .The most predominant MLST type of the 80 strains was ST11 [39] . Notably 86.3% (44/51)CR-hvKP strains belonged to ST11-K64 clone, which was an independent predictor factor for CR-hvKP infections, being more resistant and exhibited elevated virulence, leading to a high mortality and transmissibility in infected patients [5,40−41] . Since all the strains were collected from the same ward ICU, we speculated clonal dissemination of KPC-2-producing ST11-K64 K. pneumoniae exited, and during the process of transmission, KPC-2-producing ST11-K64 K. pneumoniae acquire of a PLVKPlike virulence plasmid, leading to KPC-2-producing ST11-K64 CR-hvKP. ZHANG et al [42] also provided evidence that ST11-K64 K.pneuminiae may be a competent host strain for a hypervirulent plasmid. The emergence of KPC-2-producing ST11-K64 CR-hvKP might pose a serious challenge to the management of elderly patients infected with these strains in the future, as these strains contain both drug resistance and virulence genes and show higher resistance than CR-non-hvKP strains [32] .
In order to take appropriate intervention measures correctly, a better understanding of risk factors is essential.
Univariate analysis showed that change of initial antibiotics due to clinical worsening was strongly associated with CR-hvKP. This reminds the individualized therapy must be used to treat CR-hvKP infections based on in vitro antimicrobial susceptibility pro les, molecular type, and the patient's health status [43] . Cardiovascular disease was independent predictor to CR-hvKP infections. Patients with underlying diseases are more likely to develop hvKP infections [44] , such patients typically receive inappropriate empiric therapy, more frequent hospital exposure associated with severe comorbidities may result in exposure to CR-hvKP and possible infections [45] . Meanwhile, older age and serious complications could reduce immunity and thus increase the risk of infection/settlement and even death [46] . Clinicians should pay close attention to these risk factors in clinical practice to reduce emergence of CR-hvKP strains.
There were some limitations in our study. First, this is a one-year retrospective study conducted in a single center, not a multi-center, longitudinal molecular epidemiological study on CR-hvKP. Second, Pulsed eld gel electrophoresis was not done in this study, we speculated nosocomial transmission based on MLST typing, capsule serotype and drug resistance genotype of the clinically strains.

Conclusions
CR-hvKP is emerging as a common pathogen in the elderly in China. Cardiovascular disease and ST11-K64 appeared to be independent variables associated with CR-hvKP infections. We rstly found the CR-hvKP strains showed higher MIC50 values than CR-non-hvKP strains. KPC-2-producing ST11-K64 CR-hvKP might become new signi cant "superbugs" and a threat to elderly patients.

Consent for publication
Not applicable.

Availability of data and materials
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.