Materials and Methods
A prospective analytical study was carried out over a period of 8 months (November 2014 to June 2015) in our tertiary care hospital in Pondicherry after obtaining the Institutional Ethical Committee (IEC) clearance (Registration number: Faculty/2014/19). All non-duplicate isolates of K. pneumoniae recovered from various clinical samples received in microbiology department for bacterial culture during the study period were included in this study. In case of recovery of more than one isolates of K. pneumoniae from the same patients having identical antibiogram, the first isolate was included and further isolates were considered as duplicate and were excluded from our study. Samples were collected and processed as per standard operative procedures. These isolates were subjected to antibiotic susceptibility test against a Gram negative antibiotic panel comprising of cotrimoxazole (1.25/23.75 μg), cefoxitin disk (30 μg), cefotaxime (30 μg), cefepime-tazobactam (30/10 μg), ciprofloxacin (5 μg), gentamycin (10 μg), amikacin (30 μg), imipenem (10 μg), meropenem (10 μg), piperacillin-tazobactam (100/10 μg), cefoperazone-sulbactam (75/30 μg). Antibiotic susceptibility was determined by Kirby-Bauer disc diffusion method. E. coli ATCC 25922 was used for quality control as per CLSI 2015 guidelines [8].
ESBL Detection by Phenotypic Method
K. pneumoniae strains showing cefotaxime zone ≤27 mm were considered as ESBL producers and were confirmed by combined disc test using ceftazidime (30 μg) and ceftazidime with clavulanic acid (30/10 μg) discs [8].
Lawn culture of the test organism with a bacterial suspension of 0.5 McFarland was made on a Mueller Hinton agar plate. The Ceftazidime (CAZ) disc and ceftazidime with clavulanic acid (CAC) discs were placed 25 mm apart on the Mueller Hinton agar plate. The plates were aerobically incubated at 37°C. Isolates showing an increase ≥5 mm in a zone diameter in CAC versus CAZ alone were considered to be ESBL producers. K. pneumoniae ATCC 700603 (positive control) and E. coli ATCC 25922 (negative control) were used for quality control [Table/Fig-1].
ESBL combined disc test using ceftazidime and ceftazidime with clavulanic acid.
ESBL production was further compared phenotypically by HiCrome ESBL Agar (Himedia, Mumbai, India). ESBL producing K. pneumoniae strains had luxuriant growth with bluish green colonies [Table/Fig-2].
HiCrome ESBL agar showing bluish green colonies.
AmpC Detection by Phenotypic Method
An isolate was considered as AmpC beta lactamase producer if it was cefoxitin resistant (zone ≤14 mm) and showed positive AmpC disc test [9,10].
For AmpC disc test, filter paper discs containing Tris-EDTA (20 μL of a 1:1 mixture of saline and 100× Tris-EDTA) were prepared and stored. Lawn of E. coli ATCC 25922 on Mueller-Hinton agar plate was made. A cefoxitin disk (30 μg) was placed on Mueller-Hinton agar. An AmpC disk is moistened with 20 μL of saline, smeared with test isolate colonies and kept close to the cefoxitin disk on Mueller-Hinton agar. We also included AmpC discs with known AmpC positive and negative laboratory isolates as Positive Control (PC) and Negative Control (NC) respectively. An indented inhibition zone after overnight incubation is indicative of AmpC beta lactamase production [Table/Fig-3].
AmpC disc test showing indented zone of inhibition of cefoxitin disc (CX) near the Test isolate (T) and Positive Control (PC) discs. Inhibition zone near Negative Control (NC) disc shows no indentation.
Phenotypic Tests for Carbapenemase Detection
Isolates showing resistance to one or more carbapenems i.e., imipenem and/or meropenem were subjected to Modified Hodge test (MHT), meropenem-EDTA combined disc test and culture on HiCrome KPC Agar (Himedia, Mumbai, India) to differentiate various carbapenemase producers phenotypically. Isolates that grew were considered as carbapenemase producers. The KPC producing K. pneumoniae had luxuriant growth with bluish green colonies [Table/Fig-4].
HiCrome KPC agar showing bluish green colonies.
Modified Hodge Test
K. pneumoniae strains considered to produce carbapenemase if they were positive in MHT. The test was carried out as per CLSI 2015 guidelines [8]. A lawn culture of imipenem sensitive E. coli ATCC 25922 was made on a Mueller Hinton agar plate. An imipenem disk (10 μg) was placed in the centre. K. pneumoniae isolates were streaked as straight lines from the centre to the periphery of the plate. After overnight incubation, a clover leaf-like distorted zone of inhibition of the imipenem disc produced by a test isolate was interpreted as a positive result [Table/Fig-5].
Meropenem-EDTA Combined Disc Test
A lawn culture of the test organism with a bacterial suspension of 0.5 McFarland was made on a Mueller Hinton agar plate [11,12]. A meropenem disc (10 μg) and another meropenem disc with 10 μL of 0.1 M EDTA were placed 25 mm apart on the Mueller Hinton agar plate. The plates were aerobically incubated at 37°C. Isolates showing an increase ≥5 mm in a zone diameter in meropenem-EDTA versus meropenem alone were considered to be MBL producers [Table/Fig-6].
Combined disc test using meropenem and meropenem-EDTA.
Molecular Methods
All cefotaxime resistant 117 K. pneumoniae isolates were screened for blaCTX-M, blaSHV-1, blaTEM genes by multiplex PCR with specific primers as described by Kiratisin P et al., and Abujnah AA et al., [13,14]. Carbapenem resistant 33 strains were tested for blaNDM-1, blaIMP-1, blaVIM-2, blaSIM, and blaKPC genes using specific primers as described by Kazi M et al, and Saranathan R et al., [15,16]. The forward and reverse primers used for PCR are given in [Table/Fig-7]. PCR condition followed were 95°C for 5 minutes, 30 cycles with 95°C for 30 seconds, 55°C for 30 seconds, 72°C for 90 seconds and final extension at 72°C for 7 minutes. Amplicons were identified by electrophoresis of PCR products on 1% agarose gel.
Primers used for PCR [13-16].
| Beta lactamase genes | Forward and reverse primers | Sequence (5’ to 3’) | Base pair (bp) size | Reference |
|---|
| blaCTX-M | F | TCTTCCAGAATAAGGAATCCC | 909 bp | Kiratisin P et al., [13] |
| R | CCGTTTCCGCTATTACAAAC |
| blaSHV-1 | F | TGGTTATGCGTTATATTCGCC | 868 bp | Kiratisin P et al., [13] |
| R | GGTTAGCGTTGCCAGTGCT |
| blaTEM | F | CAGCGGTAAGATCCTTGAGA | 643 bp | Abujnah AA et al., [14] |
| R | ACTCCCCGTCGTGTAGATAA |
| blaNDM-1 | F | GTAGTGCTCAGTGTCGGCA | 475 bp | Kazi M et al., [15] |
| R | GGGCAGTCGCTTCCAACGGT |
| blaIMP-1 | F | CTACCGCAGCAGAGTCTTTGC | 640 bp | Saranathan R et al., [16] |
| R | GAACAACCAGTTTTGCCTTACC |
| blaVIM-2 | F | ATGTTCAAACTTTTTGAGTAGTAAG | 801 bp | Saranathan R et al., [16] |
| R | CTACTAACGACTGAGCG |
| blaSIM-1 | F | TTGCGGAAGAAGCCCAGC | 681 bp | Saranathan R et al., [16] |
| R | GCGGCGGTTTTGATTTGC |
| blaKPC | F | SACCRCSTCGCRGSACCSRT | 275 bp | Kazi M et al., [15] |
| R | SCCRSCASGCCSGRTRTCS |
Results
In our study, the prevalence of MDR K. pneumoniae was 20.8%. Out of a total of 562 non-duplicate K. pneumoniae strains isolated during study period, 117 isolates were found to be MDR having resistance to three or more classes of antibiotics and were included for further characterisation. Most of these MDR isolates were recovered from respiratory (n=45) and pus/exudate samples (n=40), followed by urine (n=19) and blood (n=13) samples. Gender-wise distribution of our samples showed, the majority of our MDR isolates were from male patients with age group of 26 to 45 (76%) followed by females.
We performed antibiotic susceptibility test against a panel of Gram negative antibiotics by disc diffusion test. Among the non-beta-lactam agents, highest resistance was found against co-trimoxazole (82%, n=97) and ciprofloxacin (75%, n=88), followed by gentamycin (56%, n=66) and amikacin (46%, n=54) [Table/Fig-8].
Antibiotic susceptibility pattern of MDR K. pneumoniae.
| Antibiotics | Sensitive | Resistant | Intermediate |
|---|
| Cotrimoxazole | 9% (10) | 82% (97) | 9% (10) |
| Gentamycin | 23% (27) | 56% (66) | 21% (24) |
| Amikacin | 32% (37) | 46% (54) | 22% (26) |
| Ciprofloxacin | 21% (24) | 75% (88) | 4% (5) |
| Imipenem | 72% (84) | 28% (33) | 0% (0) |
| Meropenem | 77% (90) | 23% (27) | 0% |
| Cefoxitin | 31% (36) | 63% (74) | 6% (7) |
| Cefotaxime | 0% | 100% (117) | 0% |
| Cefoperazone-sulbactam | 69% (81) | 25% (29) | 6% (7) |
| Cefepime-tazobactam | 78% (91) | 20% (23) | 2% (3) |
| Piperacillin-tazobactam | 35% (41) | 45% (53) | 20% (23) |
All 117 MDR K. pneumoniae isolates were cefotaxime resistant. Out of these isolates, 91 isolates were ESBL positive by ceftazidime-clavulanic acid combined disc method and 95 isolates by HiCrome ESBL agar. Amp C screening was positive in 74 isolates which showed cefoxitin resistance. Of them, 27 isolates (23%) were identified as AmpC beta lactamase producer Amp C by disc test.
A total of 33 isolates resistant to one or more carbapenems were tested for MBL and KPC phenotypically and genotypically. Out of the 33 isolates, 16 were positive in Meropenem-EDTA combined disc test, 17 were Modified Hodge test positive and 18 isolates were identified as carbapenemase producer by HiCrome KPC agar [Table/Fig-9].
Resistance detected by phenotypic methods.
| Resistance mechanism | Phenotypic methods | Number of Isolates (%) |
|---|
| ESBL | ESBL combined disc test | 91 (77.7%) |
| HiCrome ESBL agar | 95 (81%) |
| AmpC beta-lactamase | AmpC disc test | 27 (23%) |
| Carbapenemase | MBL combined disc test | 16 (48%) |
| Modified Hodge test | 17 (51.5%) |
| HiCrome KPC agar | 18 (54.5%) |
The results of molecular tests are outlined in [Table/Fig-10]. Two ESBL genes i.e., blaSHV-1 and blaCTX-M were identified in 81 and 71 isolates respectively. Whereas, blaNDM-1 (n=11) and blaIMP-1 (n=3) were the carbapenemase genes found among our 33 carbapenem resistant isolates.
Resistance genes detected by PCR.
| Resistance mechanism | Genes detected | Number of Isolates (%) |
|---|
| ESBL | blaCTX-M | 71 (60.6%) |
| blaSHV-1 | 81 (69%) |
| blaTEM | 0 (0%) |
| Carbapenemase | blaNDM-1 | 11 (33%) |
| blaIMP-1 | 3 (9%) |
| blaVIM-2 | 0 (0%) |
| blaSIM-1 | 0 (0%) |
| blaKPC | 0 (0%) |
Discussion
K. pneumoniae is notorious for its drug resistance. While MDR is a more common scenario, extensively drug-resistant and pan-drug-resistant clones of K. pneumoniae have also been reported [17,18]. Isolates displaying resistance to three or more categories of antibiotics are considered as MDR [19]. In this study, all 117 MDR K. pneumoniae isolates were cefotaxime resistant. Among these isolates, positive results obtained in HiCrome ESBL agar (81%) was higher than that of ceftazidime-clavulanic acid combined disc test (77.7%) [Table/Fig-9]. This higher positivity of ESBL in chromogenic media in comparison to combined disc test was also noted in previous studies. Kałużna E et al., found 92.9% and 95.2% positive results using CHROM agar ESBL (GRASO) and ChromID ESBL (bioMérieux) respectively which was higher than double-disc synergy test (47.6%), combined disc test (40.5%), E-test ESBL (26.2%) and VITEK 2 (35.7%) [20]. Overdevest ITMA et al., also reported lower specificity of chromogenic ESBL detection media indicating the chance of obtaining false-positive results [21].
In the present study, AmpC production was seen in 27 (23%) K. pneumoniae by AmpC disc test [Table/Fig-9]. The prevalence of AmpC enzyme has been variable in studies from different parts of India. While Singhal S et al., from Delhi and Hemalatha V et al., from Chennai reported 36.06% and 47.3% AmpC prevalence in gram-negative bacilli respectively [22,23], it is less in a study from Karnataka [24]. The mechanism of cefoxitin resistance in gram negative bacteria could be either AmpC beta-lactamase or porin mutations [10]. Since 47 of our isolates were cefoxitin resistant and AmpC disc test negative, these may indicate presence of other mechanism of cefoxitin resistance such as porin mutations. However, this finding was not substantiated by molecular methods.
Carbapenem resistance in gram negative bacteria is an emerging problem and a serious threat to infection control in hospital. Expression of carbapenem hydrolysing enzymes like MBLs, KPC, OXA and overproduction of ESBLs or AmpC along with porin loss can confer resistance to this class of antibiotics [25,26]. In this study, 33 carbapenem resistant isolates were phenotypically tested by HiCrome KPC agar, Modified Hodge test and combined disc test. Several chromogenic media, namely CHROM agar KPC, chromID CARBA, Brilliance CRE and HiCrome KPC agar are available commercially and have been recommended for screening gram negative isolates with reduced carbapenem susceptibility.
Modified Hodge test has been effectively used for detection of carbapenemase production in Enterobacteriaceae. A positive result in MHT may indicate carbapenemase of diverse class viz. MBLs including New Delhi metallo betalactamase (NDM), KPC and SME-1 [27]. In our study, HiCrome KPC agar was used and 18 out of 33 carbapenems resistant isolates were found positive for carbapenemase. Among this 18 HiCrome KPC agar positive isolates, 17 were MHT positive and 16 were MBL producer by meropenem-EDTA combined disc test [Table/Fig-9]. This may suggest presence of other carbapenemases in at least one isolate. The remaining 15 isolates were non-carbapenemase producing and are assumed to have other mechanisms of carbapenem resistance.
We have done molecular identification of ESBL and carbapenemase genes [Table/Fig-11,12,13 and 14]. Among the ESBL genes, blaSHV-1 was most common (n=81, 69%), followed by blaCTX-M (n=71, 60.6%) in 117 MDR K. pneumoniae isolates [Table/Fig-10]. No isolates were positive for blaTEM gene. This is in accordance with other studies. High prevalence of blaCTX-M, and blaSHV have been reported in K. pneumoniae in Indian subcontinent [28,29]. Goyal A et al., from India found blaCTX-M in 85.4% ESBL producing isolates, followed by blaTEM (54.9%) and blaSHV (32.9%) [28]. A study from Bangladesh identified blaCTX-M gene in 51.4% K. pneumoniae isolates, followed by blaSHV (27%) [29]. blaCTX-M have been the most clinically significant ESBL gene due to its worldwide dissemination. It has replaced blaTEM in bacterial population in several countries including India [3]. Although blaTEM is not a rarity in Enterobacteriaceae, it was not detected in our isolates.
PCR amplification of blaCTX-M among the phenotypic ESBL positive Klebsiella pneumoniae isolates. Lane M-10 Kb Gene O’ Ruler (Thermo Scientific), Lane 1,3,4,5,7,8,9,12,13 and 14-Positive for blaCTX-M, Lane 2,6,10 and 11-Negative for blaCTX-M.
PCR amplification of blaSHV among the phenotypic ESBL positive Klebsiella pneumoniae isolates. Lane M-10 Kb Gene O’ Ruler (Thermo Scientific), Lane 1,4,5,6 and 9-Positive for blaSHV, Lane 2,3,7,8 and 10-Negative for blaSHV.
PCR amplification of blaNDM-1 among the phenotypic MBL positive Klebsiella pneumoniae isolates. Lane M-10 Kb Gene O’ Ruler (Thermo Scientific), Lane 8 and11 -Positive for blaNDM-1, Lane 1-7,9,10,12-14 - Negative for blaNDM-1.
PCR amplification of blaIMP among the phenotypic MBL positive Klebsiella pneumoniae isolates. Lane M-10 Kb Gene O’ Ruler (Thermo Scientific), Lane 12 -Positive for blaIMP, Lane 1-11, 13 and 14 -Negative for blaIMP.
The molecular analysis of our 33 carbapenem resistant isolates showed preponderance of blaNDM-1 gene (n=11, 33%) followed by blaIMP-1 in 3 (9%) isolates [Table/Fig-10]. blaVIM-2, blaSIM-1, blaKPC genes were not detected in our isolates. Similar findings were noted in other studies. Various authors have reported prevalence of various genes for carbapenem resistance. blaNDM, blaKPC, blaIMP, blaVIM and blaOXA-48 were commonly implicated [26,30,31]. In a study from Taiwan, Tseng IL et al., found 18.4% carbapenem non-susceptible K. pneumoniae had carbapenemase genes and blaKPC (15.8%), blaIMP-8 (1.6%), blaVIM-1 (0.9%) and blaNDM-1 (0.1%) were the prevalent genotypes [31]. However, in Saudi Arabia blaOXA-48 was commonest (81.5%) followed by blaNDM-1 and blaVIM [32]. NDM-1 is a novel MBL enzyme, first reported in New Delhi in a Swedish patient colonised with K. pneumoniae. The initial outbreaks were from India and Pakistan rapidly followed by its worldwide spread [2,30].
The co-production of two or more of these enzymes has become frequently now-a-days. This not only confers wider spectrum of antibiotic resistance, but also leads to greater chance of survival and dissemination of the resistant bacterial strains. Furthermore, presence of high-level of AmpC enzymes may preclude the detection of the ESBLs [33]. In the present study, 32% isolates were co-producers [Table/Fig-15]. Several of these isolates were found to be positive for multiple ESBL and MBL genes. Out of the three blaIMP-1 positive isolates, two were co-expresser of blaCTX-M, blaSHV-1, blaIMP-1, and blaNDM-1 genes and one isolate co-expressed blaIMP-1 and blaSHV-1. Likewise, out of 11 blaNDM-1 positive isolates, blaCTX-M and blaSHV-1 genes were co-expressed along with blaNDM-1 in 3 and 5 isolates respectively. This is in accordance to a study from North India, which found co-production of ESBL and AmpC (13.5%), ESBL and MBL (10%), AmpC and MBL (1%) and ESBL, AmpC and MBL (2.5%) among E. coli isolates [9]. A higher prevalence of these co-producer strains were reported in other studies [33,34].
K. pneumoniae isolates showing multiple resistance mechanisms.
| Mechanism | Number of Co-producer isolates detected by phenotypic method (%) |
|---|
| ESBL+AmpC | 9 (8%) |
| Carbapenemase+AmpC | 5 (4%) |
| ESBL+Carbapenemase | 14 (12%) |
| ESBL+Carbapenemase+AmpC | 9 (8%) |
Limitation
The limitations of our study were inability to confirm OXA, AmpC and porin mediated resistance mechanisms by molecular methods due to cost constrains. This is especially important for OXA enzymes as there are no phenotypic tests for its detection due to absence of specific inhibitor. Recently high level Temocillin resistance on MIC or disc diffusion test has been described to be an effective indicator of OXA-48 [26]. However, it needs further confirmation.
Conclusion
MDR strains of K. pneumoniae is prevalent in our hospital. While, blaSHV and blaCTX-M were the major ESBL genes, blaNDM-1 and blaIMP-1 were prevalent MBL genes. Co-production of ESBL, AmpC and MBL enzymes was found in our K. pneumoniae isolates. It indicates the need for close surveillance of antimicrobial use and resistance profile of gram negative bacteria for effective infection control.