Keywords
enterobacteriaceae - multidrug resistant - beta-lactamases - co-occurrence
Introduction
Infectious disease burden and antimicrobial resistance (AMR) are serious global problems
related to health not only in humans but also in animals, particularly in developing
countries as in the case of India. India is known for its largest antibiotic use globally,
popularly known as “AMR capital of the world.” Infections caused by gram-negative
bacteria are considerably more worrisome than those caused by gram-positive bacterial
infections as they are more commonly multidrug-resistant (MDR).[1] MDR organisms are those organisms that show resistance to one agent in any three
or more antibiotics classes. Infections due to MDR organisms are consistently increasing
and hence pose a challenge toward effective therapeutic options. As per the data of
the World Health Organization (WHO), the mortality rate due to MDR organisms in patients
is significantly much greater than that of non-MDR organisms.[2] The national pharmaceutical sales data 2000–2010 stated that more than 10 units
of antibiotics consumption per person in India were highlighted in 2010 alone.[3] MDR Enterobacteriaceae is emerging globally as one of the most serious health problems, leading to treatment
failure of both community-acquired as well as nosocomial infections.[4] One of the major causes of bacterial resistance is the inappropriate and unnecessary
use of β-lactam drugs, leading to the selection of a variety of mutated forms of β-lactamases.
ESBLs, AmpC, and MBL have presently emerged as the most worrisome resistance mechanisms,
leading to an uncontrollable impact on antimicrobial chemotherapy. The plasmid helps
in carrying these genes, facilitating the spread between microorganisms of the same
family, and is often co-expressed in the same isolate.[5]
ESBLs are β-lactamases showing resistance to penicillins, cephalosporins, and aztreonam
(but not to cephamycins or carbapenems) by hydrolyzing these antibiotics but inhibited
by β-lactamase inhibitors such as clavulanic acid. Despite being resistant to β-lactam
drugs, ESBL-producing organisms are also frequently found to show resistance to other
classes of drugs such as aminoglycosides, cotrimoxazole, tetracycline, and fluoroquinolones.[6] AmpC β-lactamases are cephalosporins that have the ability to hydrolyze and inactivate
cephalosporins, cephamycins, aminopenicillins, and monobactams but are less inhibited
by clavulanic acid.[7] Carbapenems were known to be the only treatment for ESBL and AmpC producing infections
until the emergence of carbapenem-resistant isolates. Hence, the future of antibiotics
has fallen into the darkness due to the emergence of MBL producers. Adding up to this
global health security threat, carbapenem-resistance Enterobacteriaceae (CRE) in time and again is found to be co-associated with ESBL or AmpC β-lactamase
or sometimes both and can co-transferred with the plasmids.[8]
Such co-occurrence of different types of β-lactamases in a single organism may lead
to diagnostic and treatment failure in crucial times, mostly in severe cases. Hence,
for effective treatment of infections, it is necessary to identify the co-occurrence
as antibiotic susceptibility testing alone cannot detect these resistant organisms.
So, further confirmation is required by various phenotypic tests in laboratory settings.
There are insufficient data regarding ESBL, AmpC, and MBL detection; also, to the
best of our knowledge, no studies were found on the co-occurrence of ESBL, AmpC, and
MBL β-lactamases among the members of Enterobacteriacae strains causing infections in Gangtok, East Sikkim, India. Sikkim is one of the northeastern
states in India mainly of hilly regions having a total population of over 6 lakhs
with more rural areas and fewer healthcare facilities and hospitals.[9] Detecting and analyzing these β-lactamases and their co-existence may be of great
awareness in the prevention and control from further spread of such infections as
well as in the treatment of severe cases. Further, MDR infections are increasing rapidly
in hospital settings due to the direct use of expanded spectrum cephalosporins avoiding
effective control measures. With this background, the present study has been undertaken
to highlight the ESBL, AmpC, and MBL production by various phenotypic methods and
their co-occurrence among the multidrug-resistant (MDR) Enterobacteriaceae isolates in a tertiary care hospital in Sikkim.
Materials and Methods
The present study was performed from June 2018 to May 2019 in the department of Microbiology,
Sikkim Manipal Institute of Medical Sciences (SMIMS), Gangtok, Sikkim. A total of
400 non-repetitive clinical isolates of Enterobacteriaceae were collected from the clinical specimens (urine, sputum, pus, blood, endotracheal
tip [ET], catheter tip [CT], and body fluid) sent to the microbiology laboratory of
the Central Referral Hospital affiliated to SMIMS. All the isolates were stored at
−80°C. The sample size was calculated using the formula, n = z
2
p(1-p)/d
[2],. where n is the sample size, z is the statistic corresponding to the level of confidence, p is the expected prevalence from studies, and d is precision (corresponding to effect size).[10] In the present study, the estimation of sample size was done using the prevalence
value p = 50% (0.5) based on previous studies, correspondingly, the z value of 1.96 and precision of 5% (0.05) were considered.[2] Based on this calculation, the n value was estimated and obtained to be 400 in the present study.
Inclusion Criteria
In this study, only members of Enterobacteriaceae isolated from different clinical specimens that is, urine, sputum, pus, blood, ET,
CT, and body fluids were included.
Exclusion Criteria
All clinical isolates other than Enterobacteriaceae were excluded.
Identification of the clinical isolates
Microscopy was done for each specimen by Gram staining and was inoculated into MacConkey
agar (MA) and blood agar (BA) plates (HiMedia Laboratories Pvt. Ltd, Mumbai, India)
and incubated for 18 to 24 hours at 37°C aerobically. All isolates were then identified
up to the species level for the members of Enterobacteriaceae by studying morphology, Gram staining, and by standard biochemical tests.[11]
Antimicrobial Susceptibility Testing
All the identified members of Enterobacteriaceae were subjected to antibiotic susceptibility testing using the Kirby–Bauer disk diffusion
method,[12] Mueller–Hinton agar (MHA) following the CLSI guidelines.[13] The antibiotics used are ampicillin (10 μg), amoxicillin clavulunic acid (20/10 μg),
piperacillin–tazobactam (100/10 μg), cefuroxime (30 µg), cefuroxime axetil (30/20
µg), ceftazidime (30 µg), cefoperazone–sulbactam (75/25 μg), cefepime (30 µg), ertapenem
(10 µg), imipenem (10 µg), meropenem (10 µg), amikacin (30 μg), gentamicin (10 μg),
nalidixic acid (30 μg), ciprofloxacin (5 μg), nitrofurantoin (300 µg) and trimethoprim-sulfamethoxazole
(1.25/23.75 μg). Escherichia coli ATCC 25922 and K. pneumoniae ATCC 700603 were used as controls in each set of susceptibility tests.
Beta-lactamases Detection by Phenotypic Methods
Three hundred four Enterobacteriaceae isolates were found to be MDR as they showed resistance to at least one antibiotic
in the three or more antimicrobial categories. These MDR isolates were further screened
using various phenotypic methods for the detection of ESBL, AmpC, and MBL production.
E. coli ATCC 25922 (ESBL negative) and K. pneumoniae ATCC 700603 (ESBL positive) were used as control strains.
ESBL Detection Tests
Double-disc synergy test: The isolates were screened for ESBL production following
the method reported by Kolhapura et al.[14] Phenotypic confirmatory disk-diffusion test: ESBL production was confirmed using
the method and interpretation from the previous study by Shukla et al.[15]
AmpC Detection Tests
AmpC E-test: Double-sided AmpC E-test strips (AB Biomerieux, Sweden) containing cefotetan
in one end and cefotetan–cloxacillin was used following as per the previous publication.[16] AmpC boronic acid disc diffusion test: Screening for AmpC was done using the phenotypic
test following the previous study.[5]
MBL Phenotypic Detection Test
Detection using imipenem/imipenem EDTA disc: Phenotypic detection of MBL production
was performed using a disc of imipenem (10 μg) and another combination disc of imipenem.
EDTA disc (10/750 μg) was performed as per the method described by Chanu et al.[7]
Carba-NP test: The test was performed and interpreted as per the study done by Nordmann
et al.[17]
Data Analysis
The statistical data analyses were performed using the computer software program Statistical
Package for Social Sciences (SPSS) version 20. Chi-square test for ESBL-AmpC and ESBL-MBL
was done using an online calculator open-epi version-3.0 2 × 2 considering p-value less than 0.05 as significant. Multidrug-resistance Index (MDR Index) was calculated
using the reported study by Krumperman,[18] formulated as a/b where “a” is the number of antibiotics showing resistance by the
isolate and “b” is the number of antibiotics used. In our study, the value of “a”
is taken as the MDR isolates showing resistance to three or more antimicrobial categories
and “b” is the number of antibiotics (19 in total) used. The MDR index was calculated
only for the maximum isolated pathogens that are E.coli and K. pneumoniae as other organisms were significantly low. MDR index of less than 0.2 and greater
than 0.2 was taken as an indicator to differentiate between low- and high-risk drug-resistant
pathogens.
Results
Bacterial Isolates
Out of 400 non-duplicate members of Enterobacteriaceae isolates obtained from various clinical samples. E. coli (283) 70.8% was the mostly pathogen isolated, followed by K. pneumoniae (88) 22.0%, Enterobacter cloaceae (9) 2.25%, Morganella morganii (8) 2%), 1% (4) isolates each of Serratia marcescens and Salmonella enteric serovar Typhi, and 0.25% (1) isolate each of Proteus vulgaris, Providencia rettgeri, Citrobacter fruendii, and Shigella sonnei. The majority of the isolates were from clinical specimen of urine (271) 67.8% and
others from sputum (49) 12.3%, pus (35) 8.8%, blood (30) 7.5%, ET (10) 2.5%, CT (4)
1.0%, body fluid (1) 0.25% isolated from various in-patient (IP) and out-patient departments
(OPDs).
Antimicrobial Susceptibility Test
All 400 Enterobacteriaceae isolates were tested for antimicrobial susceptibility following the CLSI guidelines.[13] Out of which, 304 (76%) isolates were found to be MDR, showing resistance to at
least one of the agents in three or more antimicrobial categories. E.coli (74.91%) and K. pneumoniae (73.86%) isolates showed the maximum MDR. The single isolated pathogen of P. vulgaris, P. rettgeri, C. fruendii, and S. sonnei also showed MDR ([Table 1]). The MDR isolates exhibited maximum resistance to antimicrobial categories of penicillins
(32–98%), cephalosporins (25–94%), quinolones classes (79–86%), followed by aminoglycosides
(18–36%), nitrofurantoin (41%) and sulphonamides (40%). Also, Salmonella enteric serovar Typhi showed higher resistance to aminoglycosides (75%) than cephalosporins (50%) and fluoroquinolones
(25%). These 304 MDR Enterobacteriaceae were isolated from IP wards (77.6%) and OP wards (22.4%). Out of these, specifically,
10.9% were from ICUs and 9.5% were from pediatric patients.
Table 1
Multidrug-resistant (MDR) pattern in different Enterobacteriaceae isolates
Enterobacteriaceae isolates (n = 400)
|
MDR
|
None MDR
|
Escherichia coli (n = 283) 70.8%
|
212 (74.91%)
|
71 (25.08%)
|
Klebsiella pneumoniae (n = 88) 22%
|
65 (73.86%)
|
23 (26.13%)
|
Enterobacter cloaceae (n = 9) 2.25%
|
8
|
1
|
Morganela morganii (n = 8) 2%
|
8
|
0
|
Serratia marcescens (n = 4) 1%
|
4
|
0
|
Salmonella enteric serovar Typhi (n = 4) 1%
|
3
|
1
|
Proeus vulgaris (n = 1) 0.25%
|
1
|
0
|
Providencia rettgeri (n = 1)
|
1
|
0
|
Citrobacter fruendii (n = 1)
|
1
|
0
|
Shigella sonnei (n = 1)
|
1
|
0
|
Total: 400
|
304 (76%)
|
96 (24%)
|
Beta-lactamases Detection by Phenotypic Methods
All 304 MDR Enterobacteriaceae isolates were detected for ESBLs, AmpCs, and MBLs production by various phenotypic
methods as shown in [Table 2].
Table 2
Beta-lactamase production in different Enterobacteriaceae isolates
MDR Enterobacteriaceae isolates (n = 304)
|
ESBLs
|
AmpCs
|
MBLs
|
DDST
|
PCDDT
|
E-test
|
BDD test
|
I/I-EDTA
|
CarbaNP
|
Escherichia coli (212) 69.7%
|
111 (52.3%)
|
96 (45.2%)
|
25 (11.8%)
|
25 (11.7%)
|
14 (6.6%)
|
17 (8%)
|
Klebsiella pneumoniae (64) 21%
|
49 (76.5%)
|
43 (67.1%)
|
7 (10.9%)
|
11 (17%)
|
20 (31%)
|
23 (35.9%)
|
Enterobacter cloacae (9) 2.9%
|
5 (55.5%)
|
4 (44.4%)
|
2 (22.2%)
|
2 (22.2%)
|
2 (22.2%)
|
1 (11.1%)
|
Morganella morganii (8) 2.6%
|
5 (62.5%)
|
5 (62.5%)
|
|
2 (25%)
|
1 (12.5%)
|
0
|
Serratia marcescens (4) 1.3%
|
2 (50%)
|
3 (75%)
|
|
0
|
1 (25%)
|
2 (50%)
|
Salmonella enteric serovar Typhi (3) 0.9%
|
0
|
0
|
0
|
0
|
0
|
0
|
Proeus vulgaris (1) 0.3%
|
0
|
0
|
0
|
0
|
0
|
0
|
Providencia rettgeri (1) 0.3%
|
1
|
1
|
1
|
0
|
1
|
1
|
Citrobacter fruendii (1) 0.3%
|
1
|
0
|
1
|
0
|
0
|
0
|
Shigella sonnei (1) 0.3%
|
1
|
1
|
0
|
0
|
0
|
0
|
Total
|
175 (58%)
|
153 (50.4%)
|
36 (11.8%)
|
40 (13.1%)
|
39 (12.8%)
|
44 (14.8%)
|
Around 56 (18.4%) isolates of the overall MDR isolates showed co-occurrence either
with any two or all three β lactamases as represented in [Table 3]. The calculated value was found to be P* = 0.01073 and P** = 0.00561 for the co-production of ESBL-AmpC and ESBL-MBL, respectively.
Table 3
Co-occurrence of β-lactamases in different Enterobacteriaceae isolates
Isolates
|
ESBL + AmpC
|
ESBL + MBL
|
AmpC + MBL
|
ESBL + AmpC + MBL
|
Escherichia coli
|
8
|
11
|
2
|
0
|
Klebsiella pneumoniae
|
5
|
19
|
2
|
1
|
Enterbacter cloacae
|
2
|
1
|
0
|
0
|
Serratia marcescens
|
0
|
2
|
0
|
0
|
Morganella morganii
|
1
|
1
|
0
|
0
|
Providentia rettgiri
|
0
|
1
|
0
|
0
|
Total
|
16 (5.2%)
|
35 (11.5%)
|
4 (1.3%)
|
1 (0.3%)
|
Discussion
In our study, the majority (304 [76%]) of the total 400 Enterobacteriaceae isolates were found to be MDR, which is the main cause of worry as this could hamper
the current therapeutic scenario. One possible reason could be the rise in the pharmaceutical
sector in Sikkim, which could have contributed to a greater rate of antibiotic resistance
due to the amount of waste reaching the various waterways that may indirectly act
as a continuous source of AMR.[19] Other associated reasons could be the increasing rate of diseases, inadequate hospitals,
or healthcare centers, lack of appropriate diagnostic methods, poor infection control
practices, and the affinity of clinicians with the empirical treatment practices may
have further supported the global crisis of AMR.[3] The increase in healthcare costs could be another main reason in developing countries
such as India. Considering the male-female ratio (111:193) among the MDR isolates,
females (63.48%) were much higher than males (36.51%). One of the possible reasons
could be due to high-risk factors for urinary tract infections in females than males.[20]
Out of the 304 MDR isolates, the majority (77.6%) were isolated from IP compared with
OP 22.4%. This could be due to prolonged hospitalization, overuse of third-generation
cephalosporins in hospital settings, and the presence of invasive devices.[21] Yet, another cause of worry is that from the 68 isolates from OPDs, ESBL production
was seen in 20 isolates, 8 AmpC, and 4 MBL producers, mostly observed was E. coli followed by K. pneumoniae and E. cloacae. The co-production was also seen in four of the isolates. This could be a risk factor
favoring the community spread as members of Enterobacteriaceae are known to cause community as well as hospital-acquired infections.[21]
Enterobacteriaceae family, especially E. coli and K. pneumoniae are also known to cause UTIs that may further become critical if not treated.[22] In the present study, cephalosporins and aminoglycosides resistance were seen slightly
higher in K. pneumoniae (45–73%) and (26–30%) than in E. coli that showed (25–70%) and (9–25%), respectively. The majority of the isolates showed
carbapenem resistance in K. pneumoniae (28–32%), followed by E. coli (7–8%). This could be possibly due to the expression of carbapenemases, plasmid AmpC,
and permeability changes predominantly in K. pneumoniae than E. coli.[23] On the contrary, fluoroquinolones resistance was more in E. coli (65–76%) than K. pneumoniae (56–60%). A possible reason could be that fluoroquinolones are the drug of choice
for bacterial infections such as urinary tract infections (UTIs), which are known
to be mainly caused by E. coli.[20] Though lesser numbers of isolates of S. enteric serover Typhi 4 (1%) was isolated in our study, three (75%) isolates showed resistance to fluoroquinolones
and one isolate (25%) was found to be ciprofloxacin-resistant. This thoroughly co-relates
with the analysis done by Britto et al in India.[24] Ciprofloxacin is recommended as the antibiotic most appropriate for enteric fever
as first-line cephalosporins are restricted to avoid ciprofloxacin-resistant S. enteric (ICMR AMRS 2016–2018). MDR index was calculated for the major isolated pathogens
of E.coli and K. pneumoniae that showed a value of less than 0.2 in 74.91% of E.coli and 73.86% of K. pneumoniae isolates. Yet again, a high rise in the MDR index among commonly found hospital isolates
indicates that there is a higher risk of infection by such MDR pathogens to humans.
It also highlights a prompt investigation to provide a better risk assessment to patients
infected by such MDR pathogens.
Based on the phenotypic test, the present study showed ESBL detection of 50.4% which
is lesser than the study reported by Mirza et al,[8] but much higher than the other studies reported by Kolhapura et al,[14] Khanna et al,[25] Shivanna and Rao.[5] Higher rates of ESBL production were seen in K. pneumoniae (67.1%) than E. coli (45.2%) isolates, in which a similar rate of ESBL among K. pneumoniae (42%) than E. coli (33%) has also been reported from a multicentric study done in India earlier.[21] The ESBL detection rate in major hospitals of India highlights a range from 19%
to 60%.[26] AmpC production was detected in 13.1% isolates, which is slightly lower than the
study done by Shivanna and Rao,[5] but similar to Mirza et al.[8] Many studies too indicate boronic acid disc-diffusion test as a better method than
other phenotypic methods to identify the producers of AmpC although no specific confirmatory
phenotypic tests have been announced for the detection of AmpC enzymes by CLSI so
far.[27] Our result for MBL production was found to be much higher (14.8%) than the studies
done by Mirza et al,[8] but lower than that reported by Chanu et al.[7] The difference in the prevalence rate in our study could also be endorsed due to
various factors such as our hospital antibiotic guidelines and practices, ethnic differences
in various populations, different phenotypic methods and procedures performed in other
studies.[25]
[28]
The present study showed the co-occurrence of ESBL and AmpC in 5.2% of isolates, which
is similar to the study done by Chanu et al (5.7%)[7] and Khanna et al (5.6%),[25] but much lower than that reported by Shivanna and Rao (19%).[5] The present study reported a co-occurrence pattern of ESBL and MBL in 11.5% of isolates,
which is higher than that reported in other studies.[7]
[8]
[25] The AmpC and the MBL co-production in our study was found in only four (1.3%) isolates
as compared with the study done by Kolhapura et al (6.2%),[14] but similar as reported by Mirza et al (1.7%).[8] The present study also found co-occurrence of the three β-lactamases, that is, ESBL,
AmpC, and MBL together in one isolate, whereas none of these studies[5]
[7]
[8]
[25] had shown it, except the study reported by Kolhapura et al (5.1%),[14] which reported a much higher co-occurrence than our study.
Co-production of these β lactamases in this study gives the idea of horizontal transfer
of multiple resistance enzyme genes in the same isolate. This re-emphasizes the utmost
need for continuous supervision, especially MDR Enterobacteriaceae in the hospital as well as community settings, for timely and suitable therapy.[8] The co-production of β-lactamases in ESBL-AmpC and ESBL-MBL was statistically proven
using the chi-square test that showed that such co-production in β-lactamases is statistically
significant. Hence, whenever such MDR organisms are isolated, they should be screened
and dealt with proper antibiotics to avoid therapeutic failure. The infections caused
by various β-lactamase pathogens, especially Enterobacteriaceae is life-threatening as there are no specific guidelines provided to detect such β-lactamases
production. This may lead to inappropriate antibiotic therapy, further worsening the
present situation of antimicrobial resistance.[25] The high MDR rate detected in such a small populated and remote region highlights
a peak of danger in bigger populated cities of India. With the present scenario of
the pandemic crisis of COVID-19, people may consume antibiotics by themselves because
of fear or ignorance; this may show a more dangerous elevated graph of antibiotic
resistance pattern in India. Molecular methods are more specific and reliable but
costly to be affordable by a common setting in developing countries such as India.
However, these phenotypic tests can detect various β-lactamases in simple laboratory
settings, are faster and easy to access on a routine basis, and are more valid and
cost-effective. Such phenotypic methods can be implemented in every simple laboratory
setting with a lower cost to screen, report, and record data for the presence of these
β-lactamases in different rural regions of India.
Conclusion
The members of Enterobacteriaceae in this geographical region showed high multidrug resistance. A high prevalence of
β-lactamases and their co-production were also found among the Enterobacteriaceae family, mainly in K. pneumoniae and E. coli isolates. The present study highlights the necessity to identify the MDR β-lactamases
stains for effective therapy in severe as well as mild bacterial infections, thereby
enabling to reduce the risk of MDR in hospital and community settings. Further, similar
studies in specific geographical regions may be encouraged to have a brief idea of
organism-based antibiotic susceptibility patterns and β-lactamase production for effective
management and treatment regime.