- Email: [email protected]
Shifting Trends in Bacterial Keratitis in Toronto
Shifting Trends in Bacterial Keratitis in Toronto An 11-Year Review Alejandro Lichtinger, MD, Sonia N. Yeung, MD, PhD, Peter Kim, MBBS(Hons), Maoz D. Amiran, MD, Alfonso Iovieno, MD, Uri Elbaz, MD, Judy Y.F. Ku, MBChB, Rachel Wolff, MD, David S. Rootman, MD, Allan R. Slomovic, MA, MD Objective: To review the distribution, current trends, and resistance patterns of bacterial keratitis isolates in Toronto over the last 11 years. Design: Retrospective, observational, case series. Participants: Microbiology records of suspected bacterial keratitis cases that underwent a diagnostic corneal scraping and cultures from January 1, 2000, through December 31, 2010, were reviewed. Methods: Culture results and antibiotic sensitivity profiles were reviewed and analyzed. Main Outcome Measures: Distribution of the main isolated pathogens as well as in vitro laboratory minimum inhibitory concentration testing results to identify resistance patterns. Results: A total of 1701 consecutive corneal scrapings were taken during the 11 years of the study. A pathogen was recovered in 977 samples (57.4%), with bacterial keratitis accounting for 897 of the positive cultures (91.8%). The total number of Gram-positive and Gram-negative isolates was 684 and 213, respectively. We identified a decreasing trend in Gram-positive isolates (P ⫽ 0.016). The most common isolate overall was coagulase-negative Staphylococcus (CNS) and the most common Gram-negative bacteria isolated was Pseudomonas aeruginosa. Methicillin-resistant Staphylococcus aureus (MRSA) was present in 1.3% of the S aureus isolates, whereas methicillin-resistant CNS (MRCNS) was present in 43.1% of the CNS isolates. There was a trend toward increasing laboratory resistance to methicillin from 28% during the first 4 years of the study to 38.8% for the last 3 years (P ⫽ 0.133). When analyzing the sensitivities of MRSA and MRCNS isolates to other antibiotics, there was resistance to cefazolin and sensitivity to vancomycin in all isolates, whereas resistance to other antibiotics was variable. Conclusions: There was a significant decrease in the percentage of Gram-positive microorganisms over time. The sensitivity of Gram-negative isolates to tested antimicrobials was ⬎97% response for all the reported antibiotics; this was not the case for Gram-positive isolates, in which resistance to the antibiotics was more common. Methicillin-resistant organisms accounted for 29.1% of all Gram-positive cultures in our series, suggesting that the empiric use of vancomycin in the setting of severe suspected bacterial keratitis may be justified. Financial Disclosure(s): Proprietary or commercial disclosure may be found after the references. Ophthalmology 2012;119:1785–1790 © 2012 by the American Academy of Ophthalmology.
Microbial keratitis is a significant cause of ocular morbidity that can result in severe visual loss and represents one of the most common causes of corneal blindness.1 Bacterial keratitis accounts for 54% to 94.2% of all corneal infections.2– 6 Early determination of the causative infective organisms and knowledge of their sensitivities to available antibiotics is essential to effective treatment. Broad-spectrum, intensive therapy with ⬎1 antimicrobial is usually instituted until causative organisms can be identified or clinical response monitored.7 With widespread use of broad-spectrum antibiotics, a corresponding change in the microbial spectrum and susceptibility to antibiotics may occur. Regional differences exist in the microbial isolates encountered and their sensitivities and emerging resistance to available antimicrobials. Therefore, local epidemiologic studies are necessary to pro© 2012 by the American Academy of Ophthalmology Published by Elsevier Inc.
vide evidence-based guidance for the successful management of bacterial keratitis. Currently, many community-based ophthalmologists in the Toronto area use monotherapy with a fourthgeneration fluoroquinolone as initial empirical therapy with or without corneal scrapings and cultures for the treatment of suspected bacterial keratitis. At our institution, we routinely scrape and culture these cases and start frequent topical therapy with a combination of cefazolin or vancomycin (according to the preference of the treating physician) and tobramycin. This study reviews the incidence and distribution of bacterial keratitis in the greater Toronto area, and examines any changing trends in corneal isolates and their susceptibility to common antimicrobials during the last 11 years. ISSN 0161-6420/12/$–see front matter http://dx.doi.org/10.1016/j.ophtha.2012.03.031
1785
Ophthalmology Volume 119, Number 9, September 2012
Patients and Methods This study conformed to the provisions of the Declaration of Helsinki and was approved by the Institutional Research Ethics Board of the University Health Network, University of Toronto. All patients with suspected bacterial keratitis that underwent a diagnostic corneal scraping and cultures from January 1, 2000, through December 31, 2010, were included in the study. The study was divided into 3 periods for analysis: 2000 to 2003, 2004 to 2007, and 2008 to 2010. Corneal scrapings were obtained with a Kimura spatula or a surgical blade, inoculated on chocolate agar, sheep blood agar, Sabouraud’s agar, and in thioglycolate broth. Gram- and Giemsastained smears were routinely prepared. Selective media and stains for aerobic and anaerobic bacteria, Mycobacteria, and Acanthamoeba were used in patients with suggestive clinical pictures. The specimens were sent to the University Health Network/Mount Sinai Hospital Department of Microbiology for analysis. Resistance to the appropriate antibiotics was routinely tested as indicated by the bacterial group. In vitro susceptibility was determined by the Vitek system (bioMérieux, Marcy l’Etoile, France) and Kirby-Bauer disk diffusion method with minimal inhibitory concentrations determined using the Clinical and Laboratory Standards Institute guidelines. Statistical analysis was performed using SPSS software, version 11 (SPSS Inc, Chicago, IL). Descriptive statistics, and mean and standard deviations were used for continuous variables; rates and percentages for categorical variables were used to describe the sample. The Cochran-Armitage test and Spearman rank correlation coefficient were used to test for trend. P⬍0.05 was considered significant.
Results During the study’s 11-year period from 2000 through 2010, a total of 1701 corneal scrapings were undertaken in 1413 patients. Female patients represented 53.7% of the cases. The mean age was 36.1⫾22.6 years. The mean number of corneal scrapings per year was 154.6⫾32.4 (Fig 1). A significant decrease in the number of scrapings per year was noted during the 11 years of the study (P ⫽ 0.003).
All samples were cultured and a pathogen was recovered in 977 samples (57.4%). The percentage of positive corneal cultures per year ranged from 52.3% to 62.3%. The percentage of positive corneal cultures per period was 59.9%, 55%, and 56.5%, for 2000 to 2003, 2004 to 2007, and 2008 to 2010, respectively. Bacterial keratitis accounted for 897 of the positive growths (91.8% of all isolates). Fungal isolates accounted for 6% and Acanthamoeba species accounted for the remaining 2.2% of all cultured microorganisms. The percentage of bacterial isolates per year ranged from 85.43% to 96.1% of all positive cultures (Fig 1). The total number of Gram-positive and -negative isolates over the 11-year period of the study was 684 (76.2%) and 213 (23.8%), respectively; the number and percentage per period can be seen in Table 1. During the 11 years of the study there was an increase in the percentage of Gram-negative isolates (rs ⫽ 0.7; P ⫽ 0.0165). Coagulase-negative Staphylococcus (CNS) was the most commonly cultured bacterial organism overall (328 isolates; 36.5% of all bacterial growths) and hence the most common Grampositive bacteria, accounting for 48% of Gram-positive isolates. The most common Gram-negative bacteria in the study was Pseudomonas aeruginosa (91 isolates; 10.1% of all bacterial isolates), accounting for 43% of the Gram-negative growths. Table 1 shows the most common isolated bacteria overall and by Gram group per period. We detected a decrease in both CNS and Staphylococcus aureus (P ⫽ 0.025 and 0.048, respectively). Trends for other isolated bacteria were not significant.
Sensitivity to Antibiotics of Gram-Negative Microorganisms From all the Gram-negative isolates tested against ciprofloxacin, 97.4% (range, 93.3%–100% per year) of them were sensitive to the drug. Similarly, sensitivity to gentamicin was 96.8% (range, 87.5%–100% per year). Ceftazidime and tobramycin had an equal sensitivity of 98.2% (range, 88.9%–100% and 91.7%–100% per year, respectively), whereas piperacillin/tazobactam was effective in 100% of the cases. None of the antibiotics tested showed any significant trend in resistance patterns (Fig 2).
250
# of cases
200 150 100 50 0 2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
Year Corneal Scrapings
Positive v cultures
Figure 1. Number of corneal scrapings, positive cultures, and bacterial isolates per year.
1786
Bacterial Isolates
2010
Lichtinger et al 䡠 Trends in Bacterial Keratitis in Toronto Table 1. Most Commonly Isolated Bacteria by Gram Group Per Period and Total 2000–2003
2004–2007
2008–2010
Bacteria
N
%
N
%
N
%
Staphylococcus aureus Coagulase-negative Staphylococcus Streptococcus species Other Total Gram-positive Pseudomonas aeruginosa Moraxella species Serratia marcescens Other Total Gram-negative Total bacterial isolates
81 163 66 19 329 29 12 13 21 75 404
25 50 20 6 81 39 16 17 28 19
43 106 63 20 232 41 16 12 14 83 315
18 46 27 9 74 49 19 14 17 26
30 59 27 7 123 21 15 3 16 55 178
24 48 22 6 69 38 27 5 29 31
Sensitivity to Antibiotics of Gram-Positive Microorganisms The sensitivity of Gram-positive microorganisms to erythromycin was 62.7% (range, 46.7%– 80.9% per year) with a significant increase in resistance over time (P ⫽ 0.0357). Sensitivity to trimethoprim/sulfamethoxazole was 85.3% (range, 74.3%–93.8% per year), whereas 70.4% (range, 53.8%– 86.7% per year) of the tested microorganisms were sensitive to cefazolin. Oxacillin was effective in 70.9% (range, 53.9%– 86.7% per year) of the isolates tested. One hundred percent of the Gram-positive isolates tested against vancomycin were sensitive to this antibiotic (Fig 3). When analyzing microorganism-specific resistance patterns in our study, we found that all the common Gram-negative microorganisms (Moraxella species, Serratia marcescens, and Proteus species) were sensitive in all cases to the antibiotics screened in the study, whereas P aeruginosa, the most common Gram-negative microorganism overall, was sensitive in every case to ceftazidime, tobramycin, and piperacillin/tazobactam. Pseudomonas aeruginosa was also sensitive to ciprofloxacin and gentamicin in 97.8% and 98.9% of the cases, respectively.
Total from Gram Group (%)
Total from Bacterial Isolates (%)
22 48 23 7 100 (684) 43 20 13 24 100 (213) 897
17 37 17 5 76 10 5 3 6 24 100
For the Gram-positive microorganisms, we found that 81.7% of S aureus specimens were sensitive to erythromycin. Ninety-eight percent, 98.1%, and 98.7% of S aureus were sensitive to trimethoprim/sulfamethoxazole, cefazolin, and oxacillin, respectively, and 100% of the isolates were sensitive to vancomycin. Streptococcus species were sensitive to erythromycin in 88.7% of the cases, whereas 100% sensitivity was present for vancomycin. For CNS, resistance to erythromycin and cefazolin was found in 51% and 43.3% of the isolates, respectively; oxacillin was effective only in 56.9% of the cases. A better sensitivity profile was found with trimethoprim/sulfamethoxazole, demonstrating only 20.8% resistance, and with vancomycin, for which the sensitivity was 100%. Methicillin/oxacillin-resistant S aureus (MRSA) was present in only 1.3% of the isolates whereas methicillin/oxacillin-resistant CNS (MRCNS) was present in 43.1% of the tested isolates, with no significant increasing trend over time (P ⫽ 0.1336). When analyzing the sensitivities of MRSA and MRCNS isolates to other antibiotics, we found 100% resistance to cefazolin and 100% sensitivity to vancomycin; resistance to other antibiotics was variable (Fig 4).
100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 2000-2003
2004-2007
2008-2010
Ciprofloxacin Ciprofl f oxacin (P = 0.122)
Gentamicin (P = 0.410)
Tobramycin (P = 0.298)
Piperacillin/tazobactam
Total
Ceftazidime Ceft f azidime (P = 0.29 0.298)
Figure 2. Sensitivity of Gram-negative microorganisms to the tested antibiotics per period (trend P value).
1787
Ophthalmology Volume 119, Number 9, September 2012
100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 2000-2003
2004-2007
2008-2010
Total
Erythromycin (P = 0.035)
Cefazolin (P = 0.136)
Trimethoprim/sulfa (P = 0.476)
Oxacillin (P = 0.133)
Vancomycin Figure 3. Sensitivity of Gram-positive microorganisms to the tested antibiotics per period (trend P value).
Discussion Bacterial keratitis is a devastating ophthalmologic emergency and is the most common reason leading to corneal blindness in developing countries.8 Rapid diagnosis and institution of appropriate therapy are important for eradication of infection and successful visual recovery. Regional variations in microbial constituents and their responses to available antibiotics are pivotal in the decision-making process for the successful management of bacterial keratitis. To our knowledge, this is the first Canadian series reporting the incidence of bacterial keratitis, trends in the composition of microbial isolates, and resistance patterns over an 11-year period. In our 11-year study, 1701 corneal scrapings were performed, with a mean of 154.6 cases per year. There was a
significant decrease in the number of cases per year over this period. This trend may reflect the fact that more patients are being treated successfully in the community with broad spectrum antibiotics, such as fourth-generation fluoroquinolones, without diagnostic corneal scrapings or referral to tertiary care centers, rather than an actual overall reduction in the incidence of bacterial keratitis. Our positive culture rate was 57.4%, similar to previous reports that found positive cultures in 35% to 86% of the specimens,5,6,9,10 although pathogen recovery rates can be as low as 14%.8 As expected, in our series, bacterial isolates accounted for the majority of positive cultures (91.8%). We found a significant increase in the percentage of Gram-negative microorganisms, with P aeruginosa being the most common pathogen in this group and responsible for 10% of positive cultures overall (43% of Gram-negative
100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Cefazolin zolin
Erythromycin Ery r thromycin
Clindamycin
SENSITIVE
RESISTANT
Figure 4. Resistance patterns to different antibiotics of methicillin-resistant isolates.
1788
Trimethoprim/sulfa f
Vancomycin Vanco
Lichtinger et al 䡠 Trends in Bacterial Keratitis in Toronto isolates). The increase in Gram-negative isolates might be related to the rise in contact lens wear in Canada.11 The association between Gram-negative bacterial infection and contact lens wear is well recognized.12 Contact lens wear has been identified as the major risk factor for infectious keratitis in various studies.10,13–17 The increase in Gramnegative isolates has been reported in previous series including a recent study by Shalchi et al,18 in which 61.1% of all bacterial isolates were Gram-negative, thereby representing the highest documented level of gram-negative keratitis in the literature. With regard to common pathogens, recent studies of bacterial keratitis from Hyderabad, India,19 Oxford, England,5 and Waikato, New Zealand,9 have found CNS to be the most common pathogen, present in 42.3%, 24.8%, and 40.8%, of their isolates respectively. This was found to be the most common isolated microorganism in our study, where it accounted for 36.5% of all bacterial cultures. Others have found P aeruginosa to be the most common pathogen overall.10,18 There is extensive variability with regard to the distribution and frequency of the different pathogens, as well as their antibiotic resistance profiles among series from different locations, making local monitoring data invaluable for the clinician. In recent years, MRSA and MRCNS have become increasingly important multi– drug-resistant ocular pathogens.20,21 The prevalence of community-acquired MRSA is increasing.21–23 Although the rate of nosocomial-acquired MRSA in hospitalized patients in Europe ranges between 1% and 20%,24 it can be as high as 57% for MRSA, and 25% for MRCNS in elderly hospitalized patients in the United States.25 We found a trend toward an increase in the percentage of methicillin/oxacillin-resistant isolates during the study from 28% during the first 4 years to 38.8% during the last period of the study, which was not significant (P ⫽ 0.133). Other institutions have reported a significant increase in the identification of these pathogens in corneal scrapings.26 –29 In a previous report by Elsahn et al,30 as found in our study, all MRSA and MRCNS isolates where sensitive to vancomycin and resistant to cefazolin. In the same study, the authors found that 90% and 45% of MRSA and MRCNS isolates, respectively, were resistant to fourth-generation fluoroquinolones, a cause of concern for the use of empirical monotherapy with these agents. These findings are consistent with a previous study by Freidlin et al20 on the spectrum of eye diseases caused by MRSA. Methicillin-resistant microorganisms were found in 29.1% of all Gram-positive isolates tested in our study. These organisms were only sensitive to vancomycin, as determined by laboratory testing. Based on these findings, some authors27 have recommended the use of vancomycin as the drug of choice for suspected bacterial keratitis in combination with tobramycin/gentamicin. In contrast, others have proposed the use of vancomycin only for confirmed methicillin-resistant cases for fear of creating resistance, which is clearly becoming an important issue.31 The limitations of our study include its retrospective nature and the possibility that in vitro resistance data may underestimate the true efficacy of treatment in clinical prac-
tice. One of the limitations of using in vitro antibiotic sensitivity for in vivo effectiveness is that the antibiotic sensitivity does not always mirror the clinical response to antimicrobials. Reasons for this discrepancy include pharmacodynamic and pharmacokinetic properties of the drugs, as well as host immunologic factors.9,32,33 We also cannot exclude the possibility that bacteria identified in some cases may represent nonpathogenic commensal species from the ocular surface. Finally, our results are specific to the setting of our catchment population, which is a tertiary care, referral university center in Toronto, Canada; therefore, our findings may not be extrapolated to other regions or populations. Differences in microbial isolates according to geographic location, referral patterns, extent of empiric antimicrobial therapy before cultures, or other factors may account for the differences seen between our study and other investigators.6 Based on the findings of our study, we have now moved to using a combination of vancomycin and tobramycin as our initial empirical therapy for all cases of severe suspected bacterial keratitis and for those cases pretreated at the community with fourth-generation fluoroquinolones that have not responded to therapy. Bacterial keratitis continues to be the most common cause of infectious keratitis at our hospital. We documented a decreasing trend in the percentage of Gram-positive microorganisms recovered over the years with an increase in the percentage of Gram-negative microorganisms. Coagulase-negative Staphylococcus was the most common bacteria overall throughout the study. P aeruginosa continues to be the most common Gram-negative microorganism isolated. The sensitivity of Gram-negative isolates to tested antimicrobials was excellent, with ⬎97% response for all the reported antibiotics. This was not the case for Grampositive isolates, in which resistance to tested antibiotics was more common. Methicillin-resistant organisms accounted for 29.1% of all Gram-positive cultures in our series, suggesting that the empiric use of vancomycin in the setting of severe suspected bacterial keratitis may be justified.
References 1. Daniell M. Overview: initial antimicrobial therapy for microbial keratitis. Br J Ophthalmol 2003;87:1172– 4. 2. Asbell P, Stenson S. Ulcerative keratitis: survey of 30 years’ laboratory experience. Arch Ophthalmol 1982;100:77– 80. 3. Jones DB. Polymicrobial keratitis. Trans Am Ophthalmol Soc 1981;79:153– 67. 4. Liesegang TJ, Forster RK. Spectrum of microbial keratitis in South Florida. Am J Ophthalmol 1980;90:38 – 47. 5. Orlans HO, Hornby SJ, Bowler IC. In vitro antibiotic susceptibility patterns of bacterial keratitis isolates in Oxford, UK: a 10-year review. Eye (Lond) 2011;25:489 –93. 6. Alexandrakis G, Alfonso EC, Miller D. Shifting trends in bacterial keratitis in south Florida and emerging resistance to fluoroquinolones. Ophthalmology 2000;107:1497–502. 7. Baum JL. Initial therapy of suspected microbial corneal ulcers. I. Broad antibiotic therapy based on prevalence of organisms. Surv Ophthalmol 1979;24:97–105.
1789
Ophthalmology Volume 119, Number 9, September 2012 8. Zhang C, Liang Y, Deng S, et al. Distribution of bacterial keratitis and emerging resistance to antibiotics in China from 2001 to 2004. Clin Ophthalmol 2008;2:575–9. 9. Pandita A, Murphy C. Microbial keratitis in Waikato, New Zealand. Clin Experiment Ophthalmol 2011;39:393–7. 10. Green M, Apel A, Stapleton F. Risk factors and causative organisms in microbial keratitis. Cornea 2008;27:22–7. 11. Woods CA, Jones DA, Jones LW, Morgan PB. A seven year survey of the contact lens prescribing habits of Canadian optometrists. Optom Vis Sci 2007;84:505–10. 12. Keay L, Radford C, Dart JK, et al. Perspective on 15 years of research: reduced risk of microbial keratitis with frequentreplacement contact lenses. Eye Contact Lens 2007;33:167– 8. 13. Saeed A, D’Arcy F, Stack J, et al. Risk factors, microbiological findings, and clinical outcomes in cases of microbial keratitis admitted to a tertiary referral center in Ireland. Cornea 2009;28:285–92. 14. Dart JK. Predisposing factors in microbial keratitis: the significance of contact lens wear. Br J Ophthalmol 1988;72:926 –30. 15. Dart JK, Stapleton F, Minassian D. Contact lenses and other risk factors in microbial keratitis. Lancet 1991;338:650 –3. 16. Fong CF, Tseng CH, Hu FR, et al. Clinical characteristics of microbial keratitis in a university hospital in Taiwan. Am J Ophthalmol 2004;137:329 –36. 17. Bourcier T, Thomas F, Borderie V, et al. Bacterial keratitis: predisposing factors, clinical and microbiological review of 300 cases. Br J Ophthalmol 2003;87:834 – 8. 18. Shalchi Z, Gurbaxani A, Baker M, Nash J. Antibiotic resistance in microbial keratitis: ten-year experience of corneal scrapes in the United Kingdom. Ophthalmology 2011;118:2161–5. 19. Gopinathan U, Sharma S, Garg P, Rao GN. Review of epidemiological features, microbiological diagnosis and treatment outcome of microbial keratitis: experience of over a decade. Indian J Ophthalmol 2009;57:273–9. 20. Freidlin J, Acharya N, Lietman TM, et al. Spectrum of eye disease caused by methicillin-resistant Staphylococcus aureus. Am J Ophthalmol 2007;144:313–5. 21. Said-Salim B, Mathema B, Kreiswirth BN. Communityacquired methicillin-resistant Staphylococcus aureus: an emerging pathogen. Infect Control Hosp Epidemiol 2003; 24:451–5.
22. Iwao Y, Yabe S, Takano T, et al. Isolation and molecular characterization of methicillin-resistant Staphylococcus aureus from public transport. Microbiol Immunol 2011;56:76 – 82. 23. Yamamoto T, Nishiyama A, Takano T, et al. Communityacquired methicillin-resistant Staphylococcus aureus: community transmission, pathogenesis, and drug resistance. J Infect Chemother 2010;16:225–54. 24. Dulon M, Haamann F, Peters C, et al. MRSA prevalence in European healthcare settings: a review. BMC Infect Dis [serial online] 2011;11:138. Available at: http://www.biomedcentral. com/1471-2334/11/138. Accessed March 11, 2012. 25. Fukuda M, Ohashi H, Matsumoto C, et al. Methicillinresistant Staphylococcus aureus and methicillin-resistant coagulase-negative Staphylococcus ocular surface infection efficacy of chloramphenicol eye drops. Cornea 2002; 21(suppl):S86 –9. 26. Solomon R, Donnenfeld ED, Holland EJ, et al. Microbial keratitis trends following refractive surgery: results of the ASCRS infectious keratitis survey and comparisons with prior ASCRS surveys of infectious keratitis following keratorefractive procedures. J Cataract Refract Surg 2011;37:1343–50. 27. Eiferman RA, O’Neill KP, Morrison NA. Methicillin-resistant Staphylococcus aureus corneal ulcers. Ann Ophthalmol 1991;23: 414–5. 28. Lee KM, Lee HS, Kim MS. Two cases of corneal ulcer due to methicillin-resistant Staphylococcus aureus in high risk groups. Korean J Ophthalmol 2010;24:240 – 4. 29. Appelbaum PC. MRSA–the tip of the iceberg. Clin Microbiol Infect. 2006;12(suppl):3–10. 30. Elsahn AF, Yildiz EH, Jungkind DL, et al. In vitro susceptibility patterns of methicillin-resistant Staphylococcus aureus and coagulase-negative Staphylococcus corneal isolates to antibiotics. Cornea 2010;29:1131–5. 31. Falcone M, Russo A, Pompeo ME, et al. Retrospective casecontrol analysis of patients with staphylococcal infections receiving daptomycin or glycopeptide therapy. Int J Antimicrob Agents 2012;39:64 – 8. 32. Oguz H, Oguz E, Karadede S, Aslan G. The antibacterial effect of topical anesthetic proparacaine on conjunctival flora. Int Ophthalmol 1999;23:117–20. 33. Goldstein MH, Kowalski RP, Gordon YJ. Emerging fluoroquinolone resistance in bacterial keratitis: a 5-year review. Ophthalmology 1999;106:1313– 8.
Footnotes and Financial Disclosures Originally received: December 16, 2011. Final revision: March 15, 2012. Accepted: March 16, 2012. Available online: May 23, 2012.
David S. Rootman: Consultant—AMO. Allan R. Slomovic: Consultant—Bausch & Lomb, Alcon, and Allergan. Manuscript no. 2011-1799.
From the Toronto Western Hospital, University of Toronto, Toronto, Ontario, Canada. Financial Disclosure(s): The authors have made the following disclosures:
1790
Alejandro Lichtinger: Consultant—Bausch & Lomb. Correspondence: Alejandro Lichtinger, 399 Bathurst St, East Wing 6-401, Toronto, Ontario M5T 2S8. E-mail: [email protected].
Microbial keratitis is a significant cause of ocular morbidity that can result in severe visual loss and represents one of the most common causes of corneal blindness.1 Bacterial keratitis accounts for 54% to 94.2% of all corneal infections.2– 6 Early determination of the causative infective organisms and knowledge of their sensitivities to available antibiotics is essential to effective treatment. Broad-spectrum, intensive therapy with ⬎1 antimicrobial is usually instituted until causative organisms can be identified or clinical response monitored.7 With widespread use of broad-spectrum antibiotics, a corresponding change in the microbial spectrum and susceptibility to antibiotics may occur. Regional differences exist in the microbial isolates encountered and their sensitivities and emerging resistance to available antimicrobials. Therefore, local epidemiologic studies are necessary to pro© 2012 by the American Academy of Ophthalmology Published by Elsevier Inc.
vide evidence-based guidance for the successful management of bacterial keratitis. Currently, many community-based ophthalmologists in the Toronto area use monotherapy with a fourthgeneration fluoroquinolone as initial empirical therapy with or without corneal scrapings and cultures for the treatment of suspected bacterial keratitis. At our institution, we routinely scrape and culture these cases and start frequent topical therapy with a combination of cefazolin or vancomycin (according to the preference of the treating physician) and tobramycin. This study reviews the incidence and distribution of bacterial keratitis in the greater Toronto area, and examines any changing trends in corneal isolates and their susceptibility to common antimicrobials during the last 11 years. ISSN 0161-6420/12/$–see front matter http://dx.doi.org/10.1016/j.ophtha.2012.03.031
1785
Ophthalmology Volume 119, Number 9, September 2012
Patients and Methods This study conformed to the provisions of the Declaration of Helsinki and was approved by the Institutional Research Ethics Board of the University Health Network, University of Toronto. All patients with suspected bacterial keratitis that underwent a diagnostic corneal scraping and cultures from January 1, 2000, through December 31, 2010, were included in the study. The study was divided into 3 periods for analysis: 2000 to 2003, 2004 to 2007, and 2008 to 2010. Corneal scrapings were obtained with a Kimura spatula or a surgical blade, inoculated on chocolate agar, sheep blood agar, Sabouraud’s agar, and in thioglycolate broth. Gram- and Giemsastained smears were routinely prepared. Selective media and stains for aerobic and anaerobic bacteria, Mycobacteria, and Acanthamoeba were used in patients with suggestive clinical pictures. The specimens were sent to the University Health Network/Mount Sinai Hospital Department of Microbiology for analysis. Resistance to the appropriate antibiotics was routinely tested as indicated by the bacterial group. In vitro susceptibility was determined by the Vitek system (bioMérieux, Marcy l’Etoile, France) and Kirby-Bauer disk diffusion method with minimal inhibitory concentrations determined using the Clinical and Laboratory Standards Institute guidelines. Statistical analysis was performed using SPSS software, version 11 (SPSS Inc, Chicago, IL). Descriptive statistics, and mean and standard deviations were used for continuous variables; rates and percentages for categorical variables were used to describe the sample. The Cochran-Armitage test and Spearman rank correlation coefficient were used to test for trend. P⬍0.05 was considered significant.
Results During the study’s 11-year period from 2000 through 2010, a total of 1701 corneal scrapings were undertaken in 1413 patients. Female patients represented 53.7% of the cases. The mean age was 36.1⫾22.6 years. The mean number of corneal scrapings per year was 154.6⫾32.4 (Fig 1). A significant decrease in the number of scrapings per year was noted during the 11 years of the study (P ⫽ 0.003).
All samples were cultured and a pathogen was recovered in 977 samples (57.4%). The percentage of positive corneal cultures per year ranged from 52.3% to 62.3%. The percentage of positive corneal cultures per period was 59.9%, 55%, and 56.5%, for 2000 to 2003, 2004 to 2007, and 2008 to 2010, respectively. Bacterial keratitis accounted for 897 of the positive growths (91.8% of all isolates). Fungal isolates accounted for 6% and Acanthamoeba species accounted for the remaining 2.2% of all cultured microorganisms. The percentage of bacterial isolates per year ranged from 85.43% to 96.1% of all positive cultures (Fig 1). The total number of Gram-positive and -negative isolates over the 11-year period of the study was 684 (76.2%) and 213 (23.8%), respectively; the number and percentage per period can be seen in Table 1. During the 11 years of the study there was an increase in the percentage of Gram-negative isolates (rs ⫽ 0.7; P ⫽ 0.0165). Coagulase-negative Staphylococcus (CNS) was the most commonly cultured bacterial organism overall (328 isolates; 36.5% of all bacterial growths) and hence the most common Grampositive bacteria, accounting for 48% of Gram-positive isolates. The most common Gram-negative bacteria in the study was Pseudomonas aeruginosa (91 isolates; 10.1% of all bacterial isolates), accounting for 43% of the Gram-negative growths. Table 1 shows the most common isolated bacteria overall and by Gram group per period. We detected a decrease in both CNS and Staphylococcus aureus (P ⫽ 0.025 and 0.048, respectively). Trends for other isolated bacteria were not significant.
Sensitivity to Antibiotics of Gram-Negative Microorganisms From all the Gram-negative isolates tested against ciprofloxacin, 97.4% (range, 93.3%–100% per year) of them were sensitive to the drug. Similarly, sensitivity to gentamicin was 96.8% (range, 87.5%–100% per year). Ceftazidime and tobramycin had an equal sensitivity of 98.2% (range, 88.9%–100% and 91.7%–100% per year, respectively), whereas piperacillin/tazobactam was effective in 100% of the cases. None of the antibiotics tested showed any significant trend in resistance patterns (Fig 2).
250
# of cases
200 150 100 50 0 2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
Year Corneal Scrapings
Positive v cultures
Figure 1. Number of corneal scrapings, positive cultures, and bacterial isolates per year.
1786
Bacterial Isolates
2010
Lichtinger et al 䡠 Trends in Bacterial Keratitis in Toronto Table 1. Most Commonly Isolated Bacteria by Gram Group Per Period and Total 2000–2003
2004–2007
2008–2010
Bacteria
N
%
N
%
N
%
Staphylococcus aureus Coagulase-negative Staphylococcus Streptococcus species Other Total Gram-positive Pseudomonas aeruginosa Moraxella species Serratia marcescens Other Total Gram-negative Total bacterial isolates
81 163 66 19 329 29 12 13 21 75 404
25 50 20 6 81 39 16 17 28 19
43 106 63 20 232 41 16 12 14 83 315
18 46 27 9 74 49 19 14 17 26
30 59 27 7 123 21 15 3 16 55 178
24 48 22 6 69 38 27 5 29 31
Sensitivity to Antibiotics of Gram-Positive Microorganisms The sensitivity of Gram-positive microorganisms to erythromycin was 62.7% (range, 46.7%– 80.9% per year) with a significant increase in resistance over time (P ⫽ 0.0357). Sensitivity to trimethoprim/sulfamethoxazole was 85.3% (range, 74.3%–93.8% per year), whereas 70.4% (range, 53.8%– 86.7% per year) of the tested microorganisms were sensitive to cefazolin. Oxacillin was effective in 70.9% (range, 53.9%– 86.7% per year) of the isolates tested. One hundred percent of the Gram-positive isolates tested against vancomycin were sensitive to this antibiotic (Fig 3). When analyzing microorganism-specific resistance patterns in our study, we found that all the common Gram-negative microorganisms (Moraxella species, Serratia marcescens, and Proteus species) were sensitive in all cases to the antibiotics screened in the study, whereas P aeruginosa, the most common Gram-negative microorganism overall, was sensitive in every case to ceftazidime, tobramycin, and piperacillin/tazobactam. Pseudomonas aeruginosa was also sensitive to ciprofloxacin and gentamicin in 97.8% and 98.9% of the cases, respectively.
Total from Gram Group (%)
Total from Bacterial Isolates (%)
22 48 23 7 100 (684) 43 20 13 24 100 (213) 897
17 37 17 5 76 10 5 3 6 24 100
For the Gram-positive microorganisms, we found that 81.7% of S aureus specimens were sensitive to erythromycin. Ninety-eight percent, 98.1%, and 98.7% of S aureus were sensitive to trimethoprim/sulfamethoxazole, cefazolin, and oxacillin, respectively, and 100% of the isolates were sensitive to vancomycin. Streptococcus species were sensitive to erythromycin in 88.7% of the cases, whereas 100% sensitivity was present for vancomycin. For CNS, resistance to erythromycin and cefazolin was found in 51% and 43.3% of the isolates, respectively; oxacillin was effective only in 56.9% of the cases. A better sensitivity profile was found with trimethoprim/sulfamethoxazole, demonstrating only 20.8% resistance, and with vancomycin, for which the sensitivity was 100%. Methicillin/oxacillin-resistant S aureus (MRSA) was present in only 1.3% of the isolates whereas methicillin/oxacillin-resistant CNS (MRCNS) was present in 43.1% of the tested isolates, with no significant increasing trend over time (P ⫽ 0.1336). When analyzing the sensitivities of MRSA and MRCNS isolates to other antibiotics, we found 100% resistance to cefazolin and 100% sensitivity to vancomycin; resistance to other antibiotics was variable (Fig 4).
100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 2000-2003
2004-2007
2008-2010
Ciprofloxacin Ciprofl f oxacin (P = 0.122)
Gentamicin (P = 0.410)
Tobramycin (P = 0.298)
Piperacillin/tazobactam
Total
Ceftazidime Ceft f azidime (P = 0.29 0.298)
Figure 2. Sensitivity of Gram-negative microorganisms to the tested antibiotics per period (trend P value).
1787
Ophthalmology Volume 119, Number 9, September 2012
100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 2000-2003
2004-2007
2008-2010
Total
Erythromycin (P = 0.035)
Cefazolin (P = 0.136)
Trimethoprim/sulfa (P = 0.476)
Oxacillin (P = 0.133)
Vancomycin Figure 3. Sensitivity of Gram-positive microorganisms to the tested antibiotics per period (trend P value).
Discussion Bacterial keratitis is a devastating ophthalmologic emergency and is the most common reason leading to corneal blindness in developing countries.8 Rapid diagnosis and institution of appropriate therapy are important for eradication of infection and successful visual recovery. Regional variations in microbial constituents and their responses to available antibiotics are pivotal in the decision-making process for the successful management of bacterial keratitis. To our knowledge, this is the first Canadian series reporting the incidence of bacterial keratitis, trends in the composition of microbial isolates, and resistance patterns over an 11-year period. In our 11-year study, 1701 corneal scrapings were performed, with a mean of 154.6 cases per year. There was a
significant decrease in the number of cases per year over this period. This trend may reflect the fact that more patients are being treated successfully in the community with broad spectrum antibiotics, such as fourth-generation fluoroquinolones, without diagnostic corneal scrapings or referral to tertiary care centers, rather than an actual overall reduction in the incidence of bacterial keratitis. Our positive culture rate was 57.4%, similar to previous reports that found positive cultures in 35% to 86% of the specimens,5,6,9,10 although pathogen recovery rates can be as low as 14%.8 As expected, in our series, bacterial isolates accounted for the majority of positive cultures (91.8%). We found a significant increase in the percentage of Gram-negative microorganisms, with P aeruginosa being the most common pathogen in this group and responsible for 10% of positive cultures overall (43% of Gram-negative
100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Cefazolin zolin
Erythromycin Ery r thromycin
Clindamycin
SENSITIVE
RESISTANT
Figure 4. Resistance patterns to different antibiotics of methicillin-resistant isolates.
1788
Trimethoprim/sulfa f
Vancomycin Vanco
Lichtinger et al 䡠 Trends in Bacterial Keratitis in Toronto isolates). The increase in Gram-negative isolates might be related to the rise in contact lens wear in Canada.11 The association between Gram-negative bacterial infection and contact lens wear is well recognized.12 Contact lens wear has been identified as the major risk factor for infectious keratitis in various studies.10,13–17 The increase in Gramnegative isolates has been reported in previous series including a recent study by Shalchi et al,18 in which 61.1% of all bacterial isolates were Gram-negative, thereby representing the highest documented level of gram-negative keratitis in the literature. With regard to common pathogens, recent studies of bacterial keratitis from Hyderabad, India,19 Oxford, England,5 and Waikato, New Zealand,9 have found CNS to be the most common pathogen, present in 42.3%, 24.8%, and 40.8%, of their isolates respectively. This was found to be the most common isolated microorganism in our study, where it accounted for 36.5% of all bacterial cultures. Others have found P aeruginosa to be the most common pathogen overall.10,18 There is extensive variability with regard to the distribution and frequency of the different pathogens, as well as their antibiotic resistance profiles among series from different locations, making local monitoring data invaluable for the clinician. In recent years, MRSA and MRCNS have become increasingly important multi– drug-resistant ocular pathogens.20,21 The prevalence of community-acquired MRSA is increasing.21–23 Although the rate of nosocomial-acquired MRSA in hospitalized patients in Europe ranges between 1% and 20%,24 it can be as high as 57% for MRSA, and 25% for MRCNS in elderly hospitalized patients in the United States.25 We found a trend toward an increase in the percentage of methicillin/oxacillin-resistant isolates during the study from 28% during the first 4 years to 38.8% during the last period of the study, which was not significant (P ⫽ 0.133). Other institutions have reported a significant increase in the identification of these pathogens in corneal scrapings.26 –29 In a previous report by Elsahn et al,30 as found in our study, all MRSA and MRCNS isolates where sensitive to vancomycin and resistant to cefazolin. In the same study, the authors found that 90% and 45% of MRSA and MRCNS isolates, respectively, were resistant to fourth-generation fluoroquinolones, a cause of concern for the use of empirical monotherapy with these agents. These findings are consistent with a previous study by Freidlin et al20 on the spectrum of eye diseases caused by MRSA. Methicillin-resistant microorganisms were found in 29.1% of all Gram-positive isolates tested in our study. These organisms were only sensitive to vancomycin, as determined by laboratory testing. Based on these findings, some authors27 have recommended the use of vancomycin as the drug of choice for suspected bacterial keratitis in combination with tobramycin/gentamicin. In contrast, others have proposed the use of vancomycin only for confirmed methicillin-resistant cases for fear of creating resistance, which is clearly becoming an important issue.31 The limitations of our study include its retrospective nature and the possibility that in vitro resistance data may underestimate the true efficacy of treatment in clinical prac-
tice. One of the limitations of using in vitro antibiotic sensitivity for in vivo effectiveness is that the antibiotic sensitivity does not always mirror the clinical response to antimicrobials. Reasons for this discrepancy include pharmacodynamic and pharmacokinetic properties of the drugs, as well as host immunologic factors.9,32,33 We also cannot exclude the possibility that bacteria identified in some cases may represent nonpathogenic commensal species from the ocular surface. Finally, our results are specific to the setting of our catchment population, which is a tertiary care, referral university center in Toronto, Canada; therefore, our findings may not be extrapolated to other regions or populations. Differences in microbial isolates according to geographic location, referral patterns, extent of empiric antimicrobial therapy before cultures, or other factors may account for the differences seen between our study and other investigators.6 Based on the findings of our study, we have now moved to using a combination of vancomycin and tobramycin as our initial empirical therapy for all cases of severe suspected bacterial keratitis and for those cases pretreated at the community with fourth-generation fluoroquinolones that have not responded to therapy. Bacterial keratitis continues to be the most common cause of infectious keratitis at our hospital. We documented a decreasing trend in the percentage of Gram-positive microorganisms recovered over the years with an increase in the percentage of Gram-negative microorganisms. Coagulase-negative Staphylococcus was the most common bacteria overall throughout the study. P aeruginosa continues to be the most common Gram-negative microorganism isolated. The sensitivity of Gram-negative isolates to tested antimicrobials was excellent, with ⬎97% response for all the reported antibiotics. This was not the case for Grampositive isolates, in which resistance to tested antibiotics was more common. Methicillin-resistant organisms accounted for 29.1% of all Gram-positive cultures in our series, suggesting that the empiric use of vancomycin in the setting of severe suspected bacterial keratitis may be justified.
References 1. Daniell M. Overview: initial antimicrobial therapy for microbial keratitis. Br J Ophthalmol 2003;87:1172– 4. 2. Asbell P, Stenson S. Ulcerative keratitis: survey of 30 years’ laboratory experience. Arch Ophthalmol 1982;100:77– 80. 3. Jones DB. Polymicrobial keratitis. Trans Am Ophthalmol Soc 1981;79:153– 67. 4. Liesegang TJ, Forster RK. Spectrum of microbial keratitis in South Florida. Am J Ophthalmol 1980;90:38 – 47. 5. Orlans HO, Hornby SJ, Bowler IC. In vitro antibiotic susceptibility patterns of bacterial keratitis isolates in Oxford, UK: a 10-year review. Eye (Lond) 2011;25:489 –93. 6. Alexandrakis G, Alfonso EC, Miller D. Shifting trends in bacterial keratitis in south Florida and emerging resistance to fluoroquinolones. Ophthalmology 2000;107:1497–502. 7. Baum JL. Initial therapy of suspected microbial corneal ulcers. I. Broad antibiotic therapy based on prevalence of organisms. Surv Ophthalmol 1979;24:97–105.
1789
Ophthalmology Volume 119, Number 9, September 2012 8. Zhang C, Liang Y, Deng S, et al. Distribution of bacterial keratitis and emerging resistance to antibiotics in China from 2001 to 2004. Clin Ophthalmol 2008;2:575–9. 9. Pandita A, Murphy C. Microbial keratitis in Waikato, New Zealand. Clin Experiment Ophthalmol 2011;39:393–7. 10. Green M, Apel A, Stapleton F. Risk factors and causative organisms in microbial keratitis. Cornea 2008;27:22–7. 11. Woods CA, Jones DA, Jones LW, Morgan PB. A seven year survey of the contact lens prescribing habits of Canadian optometrists. Optom Vis Sci 2007;84:505–10. 12. Keay L, Radford C, Dart JK, et al. Perspective on 15 years of research: reduced risk of microbial keratitis with frequentreplacement contact lenses. Eye Contact Lens 2007;33:167– 8. 13. Saeed A, D’Arcy F, Stack J, et al. Risk factors, microbiological findings, and clinical outcomes in cases of microbial keratitis admitted to a tertiary referral center in Ireland. Cornea 2009;28:285–92. 14. Dart JK. Predisposing factors in microbial keratitis: the significance of contact lens wear. Br J Ophthalmol 1988;72:926 –30. 15. Dart JK, Stapleton F, Minassian D. Contact lenses and other risk factors in microbial keratitis. Lancet 1991;338:650 –3. 16. Fong CF, Tseng CH, Hu FR, et al. Clinical characteristics of microbial keratitis in a university hospital in Taiwan. Am J Ophthalmol 2004;137:329 –36. 17. Bourcier T, Thomas F, Borderie V, et al. Bacterial keratitis: predisposing factors, clinical and microbiological review of 300 cases. Br J Ophthalmol 2003;87:834 – 8. 18. Shalchi Z, Gurbaxani A, Baker M, Nash J. Antibiotic resistance in microbial keratitis: ten-year experience of corneal scrapes in the United Kingdom. Ophthalmology 2011;118:2161–5. 19. Gopinathan U, Sharma S, Garg P, Rao GN. Review of epidemiological features, microbiological diagnosis and treatment outcome of microbial keratitis: experience of over a decade. Indian J Ophthalmol 2009;57:273–9. 20. Freidlin J, Acharya N, Lietman TM, et al. Spectrum of eye disease caused by methicillin-resistant Staphylococcus aureus. Am J Ophthalmol 2007;144:313–5. 21. Said-Salim B, Mathema B, Kreiswirth BN. Communityacquired methicillin-resistant Staphylococcus aureus: an emerging pathogen. Infect Control Hosp Epidemiol 2003; 24:451–5.
22. Iwao Y, Yabe S, Takano T, et al. Isolation and molecular characterization of methicillin-resistant Staphylococcus aureus from public transport. Microbiol Immunol 2011;56:76 – 82. 23. Yamamoto T, Nishiyama A, Takano T, et al. Communityacquired methicillin-resistant Staphylococcus aureus: community transmission, pathogenesis, and drug resistance. J Infect Chemother 2010;16:225–54. 24. Dulon M, Haamann F, Peters C, et al. MRSA prevalence in European healthcare settings: a review. BMC Infect Dis [serial online] 2011;11:138. Available at: http://www.biomedcentral. com/1471-2334/11/138. Accessed March 11, 2012. 25. Fukuda M, Ohashi H, Matsumoto C, et al. Methicillinresistant Staphylococcus aureus and methicillin-resistant coagulase-negative Staphylococcus ocular surface infection efficacy of chloramphenicol eye drops. Cornea 2002; 21(suppl):S86 –9. 26. Solomon R, Donnenfeld ED, Holland EJ, et al. Microbial keratitis trends following refractive surgery: results of the ASCRS infectious keratitis survey and comparisons with prior ASCRS surveys of infectious keratitis following keratorefractive procedures. J Cataract Refract Surg 2011;37:1343–50. 27. Eiferman RA, O’Neill KP, Morrison NA. Methicillin-resistant Staphylococcus aureus corneal ulcers. Ann Ophthalmol 1991;23: 414–5. 28. Lee KM, Lee HS, Kim MS. Two cases of corneal ulcer due to methicillin-resistant Staphylococcus aureus in high risk groups. Korean J Ophthalmol 2010;24:240 – 4. 29. Appelbaum PC. MRSA–the tip of the iceberg. Clin Microbiol Infect. 2006;12(suppl):3–10. 30. Elsahn AF, Yildiz EH, Jungkind DL, et al. In vitro susceptibility patterns of methicillin-resistant Staphylococcus aureus and coagulase-negative Staphylococcus corneal isolates to antibiotics. Cornea 2010;29:1131–5. 31. Falcone M, Russo A, Pompeo ME, et al. Retrospective casecontrol analysis of patients with staphylococcal infections receiving daptomycin or glycopeptide therapy. Int J Antimicrob Agents 2012;39:64 – 8. 32. Oguz H, Oguz E, Karadede S, Aslan G. The antibacterial effect of topical anesthetic proparacaine on conjunctival flora. Int Ophthalmol 1999;23:117–20. 33. Goldstein MH, Kowalski RP, Gordon YJ. Emerging fluoroquinolone resistance in bacterial keratitis: a 5-year review. Ophthalmology 1999;106:1313– 8.
Footnotes and Financial Disclosures Originally received: December 16, 2011. Final revision: March 15, 2012. Accepted: March 16, 2012. Available online: May 23, 2012.
David S. Rootman: Consultant—AMO. Allan R. Slomovic: Consultant—Bausch & Lomb, Alcon, and Allergan. Manuscript no. 2011-1799.
From the Toronto Western Hospital, University of Toronto, Toronto, Ontario, Canada. Financial Disclosure(s): The authors have made the following disclosures:
1790
Alejandro Lichtinger: Consultant—Bausch & Lomb. Correspondence: Alejandro Lichtinger, 399 Bathurst St, East Wing 6-401, Toronto, Ontario M5T 2S8. E-mail: [email protected].