High frequency of multidrug-resistant (MDR) Klebsiella pneumoniae harboring several β-lactamase and integron genes collected from several hospitals in the north of Iran

Background Klebsiella pneumoniae is one of the leading causes of hospital outbreaks worldwide. Also, antibiotic-resistant K. pneumoniae is progressively being involved in invasive infections with high morbidity and mortality. The aim of the current study was to determine antimicrobial susceptibility patterns and the incidence of resistance genes (integron types and β-lactamase-encoded genes) among clinical isolates of K. pneumoniae. Methods In this cross-sectional study, a total of 100 clinical samples were obtained from hospitalized patients in three teaching hospitals in the north of Iran, from November 2018 and October 2019. Antimicrobial susceptibility testing was performed using disk agar diffusion test in line with CLSI recommendations. For colistin, minimum inhibitory concentration (MIC) was determined using broth microdilution. Based on antibiogram, multi-drug resistant (MDR) and extensive-drug resistant (XDR) strains were detected. Finally, integron types and β-lactamase resistance genes were identified using polymerase chain reaction technique. Results The most and least clinical samples were related to the urine and bronchoalveolar lavage, respectively. Based on the antibiogram results, amikacin and gentamicin exhibited good activity against K. pneumoniae strains in vitro. The high resistance rate (93%) to ampicillin/sulbactam predicts the limited efficacy of this antibiotic, in the hospitals studied. Among all the 100 isolates, the frequency of MDR and XDR phenotypes were 58% and 13%, respectively, while no pan-drug resistant (PDR) strains were found. In the MDR K. pneumoniae strains, the prevalence of blaSHV, blaTEM, blaCTX-M-15, blaKPC, blaOXA-48, blaNDM β-lactamase genes were 91.4%, 82.7%, 79.3%, 29.3%, 36.2% and 6.9%, respectively, however 91.4% of the isolates were carrying intI gene. Class II and III integrons were not detected in any isolates. Conclusion The MDR K. pneumoniae is becoming a serious problem in hospitals, with many strains developing resistance to most available antimicrobials. Our results indicate co-presence of a series of β-lactamase and integron types on the MDR strains recovered from hospitalized patients. The increasing rate of these isolates emphasizes the importance of choosing an appropriate antimicrobial regimen based on antibiotic susceptibility pattern. Supplementary Information The online version contains supplementary material available at 10.1186/s12941-021-00476-1.


Background
Among the Klebsiella spp, Klebsiella pneumoniae (K. pneumoniae or KP) a gram-negative encapsulated bacterium, is responsible for up to 10% of nosocomial infections [1]. This organism causes a wide range of infections, such as pneumonia, burn and urinary tract infections (UTIs), septicemia, and meningitis [2]. Management of the infections caused by antibiotic-resistant K. pneumoniae is problematic due to the bacterium's intrinsic and acquired resistance to a broad spectrum of the drugs, such as β-lactams. MDR strains can be fairly challenging to treat, especially for elderly, immunosuppressed individuals, or infants with immature immunity [3]. The β-lactamase-producing K. pneumoniae can destroy a varied range of β-lactams such as penicillins, carbapenems, and cephalosporins [4]. The key mechanisms of resistance K. pneumoniae uses against these antimicrobials are hyperexpression of chromosomal cephalosporinases and production of plasmid-encoded Ambler class A [Extended spectrum β-lactamases (ESBLs)], B (Metalloβ-lactamases) and D (oxacillinases) β-lactamases [5]. ESBLs are plasmid-borne enzymes that hydrolyze the oxyimino β-lactam ring found in 3rd generation cephalosporins and aztreonam. The extensive use of numerous β-lactam agents in recent decades has led to the appearance of ESBLs, which are frequently derivatives of TEM-1 and SHV-1 enzymes [6]. Carbapenems are the β-lactams of choice for the treatment of infections caused by ESBL-producing K. pneumoniae. Ambler class B enzymes which play a critical role in drug resistance against carbapenems, are zinc dependent and inhibited by EDTA [7]. Resistance genes have a high ability to spread, because the genes have commonly found on the transferable elements such as integrons, insertion sequences (IS) and transposons [8,9]. Integrons, a segment of double-strand DNA sequence, are immobilized, but contain an integrase (intI)-encoding gene that allows the insertion of the resistance gene cassettes between highly conserved nucleotide sequences. Although several types of integrons have been identified, class I, II and III integrons are the most common types in the clinical settings [10,11]. In recent years, MDR K. pneumoniae strains producing ESBLs-, MBLs, and KPC resistance genes have been progressively found in many regions of Iran [12][13][14]. Despite the high importance of this issue, only a restricted number of reports have originated from the north of Iran addressing the frequency and coexistence of resistance genes among the clinical isolates of K. pneumoniae. Therefore, this study was performed to determine the antibiotic resistance profiles, incidence of MDR, XDR and PDR phenotypes and also prevalence of β-lactamase and integron resistance genes among K. pneumoniae strains isolated from hospitalized patients in the north of Iran.

Study design and sampling
In this descriptive cross-sectional study, based on the previous study [15], using the following formula n = z 2 P (1 − P)/d 2 , on a total of 100 non-duplicated samples, which were obtained from the one-year from November 2018 to October 2019. The clinical samples were collected from hospitalized patients of three educational hospitals in Mazandaran province. 1: Razi Teaching Hospital located in Qaem Shahr (Qaem Shahr is a city in Mazandaran province, in northern Iran) 2: Imam Khomeini Teaching Hospital located in Sari (Sari is the largest and the capital city of Mazandaran province, in northern Iran) and Zare Teaching Hospital located in Sari. These hospitals currently operate under the Mazandaran University of Medical Sciences (MAZUMS). The geographic region of the studied hospitals is shown in Fig. 1. One hundred K. pneumoniae isolates were recovered from different clinical sources including blood, sputum, bronchoalveolar lavage (BAL), wound exudates, urine, cerebrospinal fluid (CSF) and synovial fluid. In this study, all clinical specimens from any human source which contain K. pneumoniae, including both gender, and from all age groups including infants to elderly were included. Hence, K. pneumoniae strains isolated from out-patients, as well as other species of Klebsiella and mixed and/or contaminated plates were excluded.

Microbiological method
The samples were cultured on MacConkey agar (Merck, Germany), then incubated at 37 °C for 24 h. The recovered colonies were initially identified using conventional biochemical and microbiological tests, including motility (−), urease (+), oxidase (−), lysine decarboxylase (+), citrate utilization (+), Triple sugar iron agar (acid/ acid), Hydrogen sulfide (-), Methyl Red-Voges Proskauer (−/+), and Indole (−) [16]. Further confirmation of K. pneumoniae strains at the species level was accomplished by the analytical profile index (API) 20E kit (bioMérieux, La-Balme-les-Grottes, France) according to the manufacturer's instructions [17]. All isolates were preserved in the Luria-Bertani (LB) broth (Merck, Co., Germany) containing 20% glycerol at −80 °C for further use. Center for Disease Control and Prevention (ECDC) [19]. E. coli ATCC 25922 was used as a quality control (QC) organism. Also, colistin susceptibility assay was done for carbapenems-resistance isolates by broth microdilution method according to the European Committee on Antimicrobial Susceptibility Testing (EUCAST) breakpoints [20]. The E. coli NCTC 13846 (a colistin-resistant strain) was used as QC in colistin minimum inhibitory concentration (MIC) determination. The primary screening of extended-spectrum β-lactamases (ESBL)-producing pathogens was assess by the profile of AST. Strains that display reduced susceptibilities to CAZ and/or CTX were temporarily regarded as ESBL-producer, and finally confirmed as instructed by CLSI.

Statistical analysis
After collection of the data, statistical analysis was performed with the IBM SPSS Statistics 20 (SPSS Inc., Chicago, IL, USA) and a p-value less than 0.05 was considered as statistically significant.

Determination of antibiotic resistance pattern
Based on the acquired antibiotic resistance pattern, the highest and lowest resistance rate was related to SAM (93%) and AK (8%), respectively (Table 1). Also, 58% (58/100) of the isolates were resistant to three or more antimicrobials (MDR), and 13% (13/100) isolates were XDR. No PDR isolates were found. No non-MDR strains were resistant to AK. Altogether, the frequency of resistance genes among MDR strains was significantly higher than in non-MDR strains (p < 0.05). Overall, among 45 carbapenem-resistant K. pneumoniae (CRKP) strains, 28 cases were resistant to colistin antibiotic with a MIC > 2. Beside, ESBLs were phenotypically identified in 64 (64%) of the all tested isolates, molecular analysis exhibited that all strains had at least one ESBL gene.

Molecular detection of resistance genes
Molecular analysis results showed that class I integrons were the predominant resistance transferable elements in our isolates. However, 91.4% and 11.9% of the MDR and non-MDR isolates were carrying intI gene, respectively. Class II and III integrons were not detected in any isolates. As shown in Table 2, the frequency of resistance genes in MDR isolates was higher than non-MDR.

Discussion
The increasing rate of K. pneumoniae resistant strains against multiple antimicrobials is a major challenge in medical centers [30]. In a systematic review and  [32]. Another study also showed disparity with our data, where 89.5% of the K. pneumoniae strains were MDR [33], thus far higher than reported in the current work. Geographic distance, level of hygiene, type of specimens, date of study, sample size, and restriction on antibiotic usage may be the reasons for these inconsistencies. In our study, colistin resistance was found in 62.22% CRKP strains, which we consider a surprisingly high level. Possibly the over-use of this antimicrobial during recent years in the treatment of infections caused by organisms resistant to less toxic antibiotics could be the cause of this phenomenon. In an earlier study from Iran, Farivar et al. reported that 16.9% of K. pneumoniae isolated from various clinical sources were resistant to colistin [34]. In another recent work performed in an Iranian teaching hospital involving 100 CRKP strains from hospitalized patients, 50% of them were resistant to colistin [35]. Previous reports from India, Kuwait, Turkey and United Arab Emirates revealed that the prevalence of colistin resistance among K. pneumoniae was 4%, 8%,   bla TEM genes were 83% (n = 25), 70% (n = 21) and 57% (n = 17), respectively [40]. In another study, directed by Ghafourian [42]. Also, the discrepancy could be due to at least one extendedspectrum cephalosporin and another mechanism such as an ESBL or AmpC-type enzyme with porin loss [43,44]. An exciting point in our study is the low prevalence of bla NDM gene. According to the findings of Fallah et al., the close distance of India and Pakistan to Iran, the large number of journeys between the countries, and the ease of resistance transfer among microorganisms have led us to think that it may be likely for antibiotic-resistant bacteria to have the same gene [45].
PCR test results showed that intI gene was detected in 58% of the isolates (53 MDR and 5 non-MDR). Concordance with our study, Derakhshan et al. showed that 25.8% (8/31) of their K. pneumoniae isolates were carried intI. In addition, they did not find class II and III integrons [46]. In contrast, Haddadi et al. (2019), found that the class I integron gene frequency among 54 MDR K. pneumoniae clinical strains was 37.6% [47]. Firoozeh et al. showed that 100% (150/150) and 36.7% (55/150) of their MDR K. pneumoniae carried intI and intII genes, respectively [48]. But this finding is contradictory to our reports. We did not find any class II and III integrons among all samples. This discrepancy could be due to differences in the source of samples, microbial genetic diversity and level of hygiene.

Conclusion
This study has the potential to add to the body of literature regarding to MDR and/or XDR organisms in Iran. MDR K. pneumoniae is becoming a severe problem in hospitals, as many strains are developing resistance to most available antimicrobials. The increasing rate of these isolates emphasizes the importance of choosing an appropriate antimicrobial regimen based on antibiotic susceptibility patterns. The finding of the present study exposed a high prevalence of class I integron among MDR K. pneumoniae isolates and resistance genes especially bla SHV and bla TEM from Sari (north of Iran), which led to more attention to MDR strains.