|Home | About | Journals | Submit | Contact Us | Français|
Extended-spectrum beta-lactamases (ESBLs) mediated resistance is more prevalent worldwide, especially among Gram-negative bacterial isolates, conferring resistance to the expanded spectrum cephalosporins. As limited data were available on the prevalence of ESBLs in this area, the current study was undertaken to determine the prevalence, antibacterial resistance patterns, and molecular detection and characterization of ESBL encoding resistance genes among ocular Gram-negative bacterial isolates from ocular infections.
A prospective study was done on 252 ocular Gram-negative bacterial isolates recovered from ocular infections during a study period from February 2011 to January 2014. All isolates were subjected to detection of ESBLs by cephalosporin/clavulanate combination disc test and their antibacterial resistance pattern was studied. Molecular detection and characterization of ESBL encoding blaTEM-, blaSHV, blaOXA-, and blaCTX-M (phylogenetic groups 1, 2, 9, and 8/25) resistance genes by multiplex polymerase chain reaction and DNA sequence analysis.
Of all Gram-negative bacteria, Pseudomonas aeruginosa (44%) was the most common strain, followed by Enterobacter agglomerans and Klebsiella pneumoniae each (10%). Among the 252, 42 (17%) were ESBL producers. The major source of ESBL producers were corneal scraping specimens, highest ESBL production was observed in P. aeruginosa 16 (38%) and Escherichia coli 7 (16.6%). Among ESBL-producing genes, the prevalence of blaTEM-gene was the highest (83%) followed by blaOXA-gene (35%), blaSHV-gene (18.5%), and blaCTX-M-1-gene (18.5%) alone or together.
The higher rate of prevalence of ESBLs-encoding genes among ocular Gram-negative bacteria is of great concern, as it causes limitation to therapeutic options. This regional knowledge will help in guiding appropriate antibiotic use which is highly warranted.
The eye is a unique organ that is almost impermeable to all external infectious agents, though the eye surface invariably is exposed to a wide variety of microorganism. Bacterial infection of the eye is mainly due to external sources or through the intraocular invasion of microorganism carried by the blood stream. Alteration of the normal flora contributes to cause various internal and external ocular infections.[1,4,5,6] The most common Gram-negative bacteria causing ocular infections include Pseudomonas aeruginosa, Escherichia coli, and Enterobacter spp.[1,7] Despite the protection by the elements of a tear along with the blinking action of the eyelids, the resident bacteria of the conjunctival sac or environmental bacteria can establish infection, resulting in the need for antibiotic intervention to treat these infections. Understanding of the antibacterial resistance, especially to commonly used antibiotics, is very important in choosing the appropriate antibiotic to prevent sight-threatening complications. The factors contributing to the development of drug resistance among ocular isolates includes overuse of antibiotics for systemic infections as well as overuse of topical antibiotics in the eye. As a consequence, routine antibacterial susceptibility testing accompanied by molecular biological techniques for the detection of the drug resistance among ocular isolates of Gram-negative bacteria is highly warranted to understand the prevalence of drug resistance to extended-spectrum of drugs, and also to confirm the availability of broad-spectrum antibacterial agents. The development of resistance to beta-lactam group antibiotics is an emerging problem and production of beta-lactamases is the most common drug resistance mechanism among Gram-negative bacteria, especially Enterobacteriaceae family. The extended-spectrum beta-lactamases (ESBLs) are typically mutant, plasmid-mediated, clavulanate-susceptible enzymes which are derived from the older, broad-spectrum beta-lactamases and that can hydrolyze penicillins, expanded-spectrum cephalosporins, and monobactams but are inactive against cephamycins and carbapenems. ESBLs are commonly found in Gram-negative bacteria isolates mainly in Enterobacteriaceae (Klebsiella pneumoniae and E. coli leap out as the significant ESBL producers) and have been reported worldwide. In addition, ESBL-producing organisms frequently show the cross-resistance to other classes of non-β-lactam antibiotics including trimethoprim-sulfamethoxazole, aminoglycosides, and fluoroquinolones thus treatment of these infections is often a therapeutic challenge. In the 1990s, blaTEM and blaSHV type ESBLs were more prevalent and recently rapid, and immense spread of blaCTX-M-type ESBLs has been described. Several reports and surveys have documented the prevalence and incidence of ESBL-mediated resistance among the Gram-negative bacterial isolates from community and hospital-acquired infections and very few reports from ocular infections. In India, SENTRY surveillance study reported, the prevalence of ESBLs ranges increased from 62% to 100% in E. coli and Klebsiella spp. isolated from respiratory infections, blood stream infections, skin and soft tissue infections, and ocular infections caused by K. pneumoniae 58% and E. coli 42% have also been reported. However, the prevalence studies on ESBLs producing ocular isolates from India are very few. Studies reported by Jayahar Bharathi et al. and Sowmiya et al. recorded the incidence of ESBL among ocular population to be 7% and 77%, respectively, and blaCTX-M-gene (14%) was the common ESBL gene found among ocular Enterobacteriaceae family. As there is scanty data on the prevalence of ESBLs genes among ocular pathogens in India, the current study was undertaken to determine the prevalence of ESBL producing Gram-negative bacteria isolated from ocular infections and to detect and characterize the blaTEM-, blaSHV-, blaOXA-, and blaCTX-M genes by multiplex polymerase chain reaction (mPCR) and DNA sequencing.
A prospective analysis was done on 252 Gram-negative bacteria isolated from patients clinically diagnosed with bacterial ocular infections, such as blepharitis, conjunctivitis, internal and external hordeolum, scleritis, canaliculitis, keratitis, dacryocystitis, and internal infections such as endophthalmitis and panophthalmitis during the 3-year study period from February 2011 to January 2014, from a total of 4917 ocular specimens which were received at a tertiary eye care center in South India. Using standard techniques, culture and smears obtained from each ocular specimen were evaluated for significant microbiological features.[13,14] Bacterial isolates were identified up to species level using their cultural characteristics and standard biochemical tests. In vitro susceptibility testing was performed by Kirby-Bauer disc diffusion method and interpreted using Clinical and Laboratory Standards Institute (CLSI) serum standards. Pure, nonduplicate, single species of Gram-negative bacteria were isolated and subcultured onto MacConkey agar (M008; Hi-Media Laboratories Pvt. Ltd., India) for further analysis. The study was approved by the Institutional Ethical and Research Committee of the Aravind Eye Care system.
In vitro antibacterial susceptibility testing for each pure isolate was performed by Kirby-Bauer disc diffusion method against the third generation cephalosporins, (cefpodoxime, ceftazidime, cefotaxime, and ceftriaxone), monobactams (aztreonam), aminoglycosides (amikacin, tobramycin, and gentamicin), fluoroquinolones (norfloxacin, ciprofloxacin, levofloxacin, ofloxacin, and gatifloxacin), and phenolics (chloramphenicol) using Bio-Rad susceptibility discs and interpreted according to the CLSI standards guidelines.
According to the CLSI guidelines, laboratories using disc diffusion method for antibiotic susceptibility testing can screen for ESBL producing isolates showing inhibition zone size of ≤17 mm with 10 μg cefpodoxime, ≤22 mm with 30 μg ceftazidime, ≤25 mm with 30 μg ceftriaxone, ≤27 mm with 30 μg cefotaxime, and ≤27 mm with 30 μg aztreonam discs which can be identified as potential ESBL producers and shortlisted for the phenotypic confirmation of ESBL production.
For the confirmation of ESBL-production, as per the guidelines of CLSI, the disc test was done with confluent growth of the test isolates (0.5 McFarland standard) on Mueller-Hinton agar plates, with a distance of 25 mm of disc containing cefotaxime disc (30 μg/disc) and ceftazidime disc (30 μg/disc) with the combination of cefotaxime with clavulanate (30 μg + 10/μg/disc) and ceftazidime with clavulanate (30 μg + 10/μg/disc). Plates were incubated overnight at 37°C, and after incubation a zone of inhibition that showed ≥5 mm increased in the cephalosporin discs and their respective cephalosporin/clavulanate discs was taken to be a phenotypic confirmation of ESBL production.
The antibacterial agents used were tested for their efficacy against the standard American Type Culture Collection (ATCC) non-ESBL-producing organism (E. coli ATCC 25922) and an ESBL-producing organism (K. pneumoniae ATCC 700603) for reference and quality.
A single colony of each nonduplicate Gram-negative isolate was inoculated from MacConkey agar into 5 ml of Luria-Bertani broth (M1245; Hi-Media Laboratories Pvt. Ltd., Mumbai, India) and incubated for 20 h at 37°C. Cells from 1.5 ml of the overnight culture were harvested by centrifugation at 12,000 rpm for 5 min. The DNA was extracted from each bacterial isolate by alkaline lysis method using QIAamp DNA Mini Kit (Code No. 51304) according to the manufacturer's instruction (Qiagen Gmbh, Hilden, Germany). Along with the test isolates, DNA of blaCTX-M-gene carrying E. coli 4712 (positive control for blaCTX-M-15; GenBank accession number KC200080), DNA of blaSHV-gene carrying K. pneumoniae (ATCC 700603) (positive control for blaSHV-gene), DNA of blaTEM-gene carrying E. coli (ATCC 35218) (positive control for blaTEM-gene), DNA of blaOXA-gene laboratory isolate of K. pneumoniae 7888 (positive control for blaOXA-gene GenBank accession number JN565742), and DNA of non-ESBL-producing organism, E. coli ATCC 25922 were also extracted for performing positive and negative control during each mPCR. All DNA extraction procedures were carried out in a Class II biological safety cabinet (Clean Air, India) placed in a room (room B) physically separated from that used to prepare nucleic acid amplification mixes (room A) and that used for post-PCR analysis (room C).
Total DNA (2 μL) was subjected to each mPCR in a 50 μL reaction mixture containing 1X PCR buffer (10 mM Tris-HCl, pH 8.3, and 1.5 mM MgCl2) 200 μM each deoxynucleotide triphosphate, 1U of Taq DNA polymerase (Bangalore Genei, Bengaluru, India), and variable concentration of gene-specific primers [Table 1]. Amplification was carried out as follows: 94°C for 10 min, 30 cycles of 94°C for 40s, 60°C for 40s, and 72°C for 1 min and final extension step at 72°C for 8 min for the amplification of ESBL genes blaTEM, blaSHV, blaOXA, and blaCTX-M (phylogenetic groups 1, 2, 9, and 8/25). Amplicons were visualized under ultraviolet transilluminator and documented by gel document system (Vilber Lourmat, France) after running at 120V for 45 min on a 2% agarose gel containing ethidium bromide. A 100 bp DNA ladder (Bangalore Genei, Bengaluru, India) was used as a marker.
In order to study the molecular characterization of the gene (s) that were identified in the mPCR which code for enzymes, blaTEM, blaSHV, blaOXA, and blaCTX-M a total of six mPCR amplicons were performed. After gel electrophoresis, targeted gene of PCR product was eluted from the gel and purified with the EZ-10 Spin Column DNA Gel Extraction Kit (Bio Basic Inc., Canada). Purified products were sequenced at MWG AG Biotech, Bengaluru, by using an ABI 3730 XI automated sequencer (PE Applied Biosystems, Foster City, CA, USA). The sequences were analyzed and identified using the MegaBLAST search program of GenBank databases. The percentage similarity was determined using the sequence alignment of the test isolate with the GenBank sequence of the type strain by BLAST program.
During the 3 years study period, a total of 252 Gram-negative bacterial isolates were recovered from 221 ocular specimens. From ocular specimens of 199 patients, single species were isolated and while from samples of 22 patients; two or three species were isolated. Among the 252 isolates, 162 (64%) showed non-Enterobacteriaceae family, of which 56% were from male and 44% from female patients. The average age of male patients was 48.3 years (4 months to 90 years) and that of female patients 47.9 years (1 month to 85 years). Whereas, 90 (36%) showed Enterobacteriaceae family, of which 50% were from male and 50% from female patients. The average age of male patients was 36.5 years (2 months to 71 years) and that of female patients 46.3 years (2 months to 95 years). The predominant Gram-negative bacterial species isolated were P. aeruginosa (112 of 252; 44.4%) followed by Enterobacter agglomerans (25 of 252; 10%), K. pneumoniae (25 of 252; 10%), Alcaligenes denitrificans (20 of 252; 8%), and E. coli (13 of 252; 5%). P. aeruginosa was more prevalent in keratitis cases (61 of 112; 54.5%), while E. agglomerans dominated in contact lens-associated keratitis (7 of 25; 28%), K. pneumoniae in dacryocystitis (7 of 25; 28%), A. denitrificans in contact lens-associated keratitis (9 of 20; 45%), and E. coli in conjunctivitis (4 of 13; 31%). Distribution of Gram-negative bacterial isolates recovered from various ocular infections is presented in Table 2.
In the present study, of the total of 252 isolates of ocular Gram-negative bacteria tested for the production of ESBLs by the cephalosporin combination disc test, 42 (17%) were detected as ESBL producers, and 210 (83%) were detected as non-ESBL producers. Among the 42, the highest percentage of ESBL production was seen in P. aeruginosa 16 (38%), followed by E. coli 7 (17%) and A. denitrificans 6 (14%). Ocular Gram-negative bacterial isolates of non-Enterobacteriaceae family showed 69% positivity, whereas in Enterobacteriaceae family showed 31% positivity for ESBL production. The highest rate of isolation of ESBL producing non-Enterobacteriaceae was from contact lens-associated keratitis cases 27.5% (8 of 29) followed by bacterial keratitis and postoperative endophthalmitis 24% (7 of 29), whereas among Enterobacteriaceae ESBL producing isolates were obtained from dacryocystitis case 31% (4 of 13) and contact lens-associated conjunctivitis case 23% (3 of 13). The phenotypic test results of ESBL producing Gram-negative bacteria isolated from various ocular infections are presented in Table 3 and Fig. 1.
The susceptibility of the ESBL producing ocular Enterobacteriaceae and non-Enterobacteriaceae family isolates to commonly used antibiotics is depicted in Table 4. Out of Enterobacteriaceae (36% [90 of 252]) isolates 14% (13 of 90) were ESBL producers, and the remaining 86% (77 of 90) were non-ESBL producers. Among the 13 ESBL producing Enterobacteriaceae family isolates, the highest rate of susceptibility observed to chloramphenicol 12 (92%), followed by amikacin 11 (85%), gatifloxacin 9 (69%), and tobramycin 7 (54%) and the highest percentage of resistance was observed in the following order: Cefazolin 11 (85%), cefotaxime 10 (77%), ceftazidime 10 (77%), ciprofloxacin 10 (77%), norfloxacin 9 (69%), and cefpodoxime 9 (69%). Among the 77 non-ESBL producing Enterobacteriaceae family isolates the rate of susceptibility to various antibacterials was as follows: Tobramycin 76 (99%), amikacin 73 (95%), levofloxacin 72 (94%), and the highest rate of resistance was observed with respect to cefazolin 47 (61%) and cefpodoxime 23 (30%).
Among the ocular non-Enterobacteriaceae 64% (162 of 252) isolates tested, 18% (29 of 162) were ESBL producers and the remaining 82% (133 of 162) were non-ESBL producers. The 29 ESBL producing non-Enterobacteriaceae family isolates showed the highest rate of susceptibility to gatifloxacin 19 (66%), ofloxacin and gentamicin each 16 (55%), and the highest percentage of resistance was observed to cefazolin 7 (87.5%) (34 isolates tested) and ceftazidime 8 (80%) (38 isolates tested). Among the 133 non-ESBL producing non-Enterobacteriaceae family isolates showed the highest rate of susceptibility was observed to tobramycin 125 (94%), gatifloxacin 122 (92%), levofloxacin 121 (91%), and ciprofloxacin 118 (89%) and the highest rate of resistance was observed with cefazolin 20 (77%) (34 isolates tested) and cefpodoxime 15 (54%) (38 isolates tested).
Among the 252 isolates, 156 (62%) showed positive amplification for ESBL encoding resistant genes, of which 83 (53%) were non-Enterobacteriaceae isolates and 73 (47%) were Enterobacteriaceae isolates. Among the 83 non-Enterobacteriaceae isolates, blaTEM-gene was more prevalent in 94% (78 of 83), followed by blaOXA in 29% (24 of 83) and blaSHV in 4% (3 of 83). Among the 73 Enterobacteriaceae isolates, blaTEM-gene was more prevalent in 70% (51 of 73), followed by blaOXA-gene in 42.5% (31 of 73), blaSHV-gene in 36% (26 of 73), and blaCTX-M-1 16% (12 of 73). All the 252 Gram-negative bacterial isolates were subjected to genotypic detection of ESBL encoding resistant genes by mPCR analysis and the results are presented in Table 5 and Figs. Figs.2,2, ,33.
For further confirmation and characterization, six amplified products (two amplified with blaTEM-gene-specific primer, two with blaSHV-gene-specific primer, one with blaOXA-gene-specific primer, and one with blaCTX-M-1-gene-specific primer) from multidrug-resistant bacterial isolates were subjected to DNA sequencing. A DNA database comparison of partial sequencing of blaTEM, blaSHV, blaOXA-, and blaCTX-M-1-genes demonstrated six EBSL enzymes – TEM-193 (JN588556), TEM-195 (JN588557), SHV-142 (JN596240), SHV-143 (JN565741), OXA-162 (JN565742), and CTX-M-15 (KC200081) have been reported in NCBI.
ESBL producing Gram-negative bacterial isolates mainly Enterobacteriaceae are now an increasing problem worldwide. The emergence and continuous spread of these bacteria seem to be caused mainly by extensive use of broad-spectrum antibacterial agents to treat the infections, and the presence of mobile genes on plasmid encoding ESBLs among the pathogens have been on an increase every year complicating the treatment strategies in patients. ESBL producing organisms are usually reported as hospital-acquired, especially in the Intensive Care Units. Other hospital units that are at risk includes surgical wards, pediatrics and neonatology, rehabilitation units, and oncology wards. Countries with a high rate of prevalence include Turkey (60%), Brazil (45.4%), Western Pacific (24.6%), the Netherlands (22.6%), and Iran (44-74%).[22,23] However, from the current study, the prevalence of ESBL producing ocular pathogens of the Enterobacteriaceae and non-Enterobacteriaceae family was found to be 47% and 53%, respectively. The previous study reported by Sowmiya et al. presented the prevalence of ESBL producing ocular Enterobacteriaceae as 77% (57 of 73) at the molecular level. Few surveillance studies that have documented the resistance pattern of ocular pathogens include, Ocular TRUST and The Surveillance Network with the exclusion of ESBL.[24,25] Proper identification of genes involved in ESBL-mediated resistance is a necessity for the surveillance and epidemiological analysis and on this basis the present study was envisaged to determine the prevalence of ESBL encoding blaTEM, blaSHV, blaOXA-, and blaCTX-M – (variant 1, 2, 9, 8/25) resistant genes among ocular Gram-negative bacterial isolates. In the present study, the rate of isolation of ocular Enterobacteriaceae and non-Enterobacteriaceae is similar to the rates in our earlier study reported by Jayahar Bharathi et al., which showed 40% and 60%, respectively.
In India, the incidence of ESBL mediated resistance was observed among 60-68% of clinical pathogens that were isolated from major hospitals, but there are very few reports for ocular pathogens from the Enterobacteriaceae family which shows the incidence to be ranging from 7% to 77%.[7,12] In our current analysis, the rate of ESBL production among ocular pathogens from Enterobacteriaceae family was 14% (13 of 90) and non-Enterobacteriaceae family 18% (29 of 162). ESBLs have been reported most frequently in E. coli and K. pneumoniae from nosocomial and community-acquired infections. Similarly, the present study shows the prevalence of ESBL-production was 54% in ocular Enterobacteriaceae family isolates. E. coli was isolated from patients with conjunctivitis and dacryocystitis case indicated the existence of community-associated strains. Isolation of ESBL producing non-Enterobacteriaceae family isolate P. aeruginosa (55%) in the case of postoperative endophthalmitis also indicated that the strains were associated with hospital-acquired predominantly during surgery notwithstanding all standard aseptic preventive measures, includes the use of povidone-iodine, proper instilling of preoperative antibiotics, dedicated sterilized instruments, and intraocular lenses. Similarly, our current study reports well correlated with our earlier study reported by Jayahar Bharathi et al.
ESBL producing ocular Enterobacteriaceae family isolates showed increased susceptibility to chloramphenicol (92%) and amikacin (85%). A higher degree of resistance was seen to cefazolin (85%), cefotaxime, ceftazidime, and ciprofloxacin (each 77%). In a study conducted at our hospital by Bharathi et al., in 2010, ocular Enterobacteriaceae family isolates from community-acquired ocular infections showed the highest percentage of susceptibility to amikacin 92% and gatifloxacin 93%. The switch in the resistance rates of cefotaxime and ceftazidime from 40% and 45% to 77% demonstrate the widening spread of drug resistance among ocular pathogens. ESBL positive ocular non-Enterobacteriaceae family isolates showed increased susceptibility to gatifloxacin (66%), followed by ofloxacin and gentamicin (each 55%) and a higher degree of resistance to cefazolin, cefotaxime, and ceftazidime 87.5%, 80%, and 52%, respectively. This also confirms the development of resistance rates of cefotaxime and ceftazidime from 35% and 18% to 80% and 52%, respectively.
The prevalence of ESBLs types greatly vary with geographical regions, and in most cases,[11,19] blaTEM-gene has been reported with a higher frequency, while in some places,[28,29] blaSHV-gene are more prevalent than other types. The present study reports the most prevalent gene detected to be blaTEM-gene (83%) followed by blaOXA-gene (35%) and blaSHV-gene (18.5%) alone or together. Our present study shows the greater difference when compared with previous findings. It was higher than the previous study conducted by Sowmiya et al. showed that blaTEM-gene (77%) was more prevalent followed by blaSHV-gene (35%) and blaOXA-gene (33%). In addition to TEM, SHV, and OXA-types, another class of ESBL, CTX-M type-β-lactamases has been reported to be more active against cefotaxime and ceftriaxone than ceftazidime, but a possibility of point mutation can always increase the activity against ceftazidime. The prevalence of blaCTX-M-gene also stands reported indicating 73% in Indian hospitals among clinical isolates, especially among E. coli and K. pneumoniae. However, from the current study shows a prevalence of blaCTX-M-1 gene (8%) majorly among E. coli (67%); more specifically CTX-M-15 type. It was lower than the previous study reported by Sowmiya et al. showed that blaCTX-M-gene (56%). CTX-M enzymes like CTX-M-9 have been reported from Spain, CTX-M-2 from most South American countries, Japan, and Israel, whereas CTX-M-15 exist worldwide.
Our current study demonstrates that the high prevalence of common beta-lactamase genes, blaTEM (16%) were detected more from patients with corneal ulcer and contact lens-related conjunctivitis cases, followed by blaOXA-genes (25%) detected more from contact lens-related conjunctivitis cases. Current findings undoubtedly could be ascribed to microbial contamination of contact lens storage, especially by Gram-negative bacteria that lead to the development of contact lens-related corneal and conjunctival infections. The contaminating source of contact lens during storage appear to be mainly of external origin, especially water contaminated with Gram-negative bacteria leading to a widespread incidence of EBSL producing strains. Topical broad-spectrum aminoglycoside and fluoroquinolones were used to treat these infections. In nosocomial settings, the prevalence of blaTEM gene (9%) were detected more among P. aeruginosa (17%) isolated from vitreous specimens collected from postoperative endophthalmitis cases referred from outside hospital. The major source of the ESBL producing P. aeruginosa strains is contaminated irrigating fluids, contaminated instruments, and poor operating room environments. Topical or intraocular piperacillin/tazobactam has been used for the treatment of such cases.[34,35]
The present study shows the low prevalence of ESBL encoding resistant genes among ocular pathogens isolated in our hospital setup. The prevalence of ESBL producing isolates is of crucial importance among ocular Enterobacteriaceae and P. aeruginosa. Molecular detection and characterization of ESBL producing ocular pathogens confirm the presence of common ESBL genotypes majority from community-acquired infection and nosocomial settings. It is a real challenge to an ophthalmologist and one should limit the therapeutic problem. Our data suggest that the prevalence of ESBL producing Gram-negative ocular pathogens in nosocomial and community settings in developing countries is probably under-reported. Our single center study report could not reflect a broader regional context among ocular isolates and study data suggest more study has to be conducted in different parts of South India.
Indian Council of Medical Research, New Delhi, India.
There are no conflicts of interest.
This work was supported by the Indian Council for Medical Research, India.