Doi:10.1016/s0002-9394(03)00905-x

Susceptibilities for Common Ocular Isolates MARIA REGINA CHALITA, MD, ANA LUISA HO
¨ FLING-LIMA, MD, MBA,
AUGUSTO PARANHOS, JR., MD, PAULO SCHOR, MD, AND
RUBENS BELFORT, JR., MD, PHD, MBA
● PURPOSE: To assess the in vitro susceptibility of the
in the treatment of ocular infections, with high suscep-
most common ocular bacterial isolates to several antibi-
tibility to all pathogens tested. Chloramphenicol also
otics and verify changing trends in the antibiotic suscep-
revealed an increase in its susceptibility to all bacteria
tibility in a 15-year period.
evaluated.
Ophthalmol
2004;137:43–51.
● DESIGN: Experimental study.
2004 by Elsevier Inc. All rights reserved.)
● METHODS: All cultures positive for Staphylococcus au-
reus, coagulase-negative Staphylococcus (CNS), Streptococ-
cus sp, and Pseudomonas sp in conjunctival (n ؍ 4,585)
BACTERIAL OCULAR INFECTIONS ARE POTENTIALLY sight threatening if appropriate antibiotic therapy is and corneal (n ؍ 3,779) samples from patients seen at the
Federal University of Sa˜o Paulo from 1985 to 2000 were
of bacterial eye infections, the treatment must be initiated evaluated. Cultures were performed in liquid and solid
usually before pathogen identification and antibiotic sus- media, and susceptibility tests were done to amikacin,
ceptibility test results are available. Therefore, the antibi- gentamicin, neomycin, tobramycin, ciprofloxacin, norfloxa-
otic chosen must have demonstrated efficacy against a cin, ofloxacin, cephalothin, and chloramphenicol.
broad spectrum of possible ocular pathogens, providing ● RESULTS: Amikacin and neomycin showed an im-
good coverage against most gram-positive and gram-nega- provement of their sensitivity during the study period
(88%–95% and 50%– 85%, respectively) for corneal
The frequent and sometimes indiscriminate use of anti- and conjunctival samples. Gentamicin and tobramycin
biotics has led to the development of bacterial strains revealed a decrease of sensitivity in time, from 95% to
resistant to many commonly used Hence, peri- less than 80% in corneal and conjunctival samples.
odic susceptibility testing should be performed to ensure Ciprofloxacin, norfloxacin, and ofloxacin had good sen-
that the currently available antibiotics are providing good sitivity to all evaluated bacteria, better in conjunctiva
coverage against recent clinical isolates of pathogenic (95%) than in cornea (90%). Sensitivity of S. aureus to
bacteria. As suggested by Jensen and this type cephalothin decreased during the study but was still 98%
of testing should be done every 2 to 3 years to detect for CNS. Chloramphenicol had good sensitivity to S.
resistance trends with currently used antibiotics. Such aureus (85% in corneal and 92%in conjunctival sam-
studies are of paramount value to health care providers ples), CNS (87% and 88.5%), and Streptococcus sp
who often have to select a first-line antibiotic treatment (95% and 96%).
without the benefit of having the microbiological testing ● CONCLUSIONS: Gentamicin, tobramycin, and cephalo-
thin decreased their in vitro susceptibility to all tested
The purpose of this study is to assess the in vitro suscepti- pathogens. The fluoroquinolones remained a good choice
bility of the most common ocular bacterial isolates to severalantibiotics and verify changing trends in the antibioticsusceptibility of these bacteria during 15 consecutive years.
Additional material for this article can be found on ajo.com.
Accepted for publication July 29, 2003.
From the Department of Ophthalmology Federal University of Sa˜o Inquiries to Maria Regina Chalita, MD, Cleveland Clinic Foundation, ALL CONSECUTIVE CASES OF BACTERIAL KERATITIS AND 9500 Euclid Ave, i31, Cleveland, OH 44195; fax: (216) 445-8475; e-mail:[email protected] bacterial conjunctivitis that underwent a diagnostic corneal 2004 BY ELSEVIER INC. ALL RIGHTS RESERVED.
TABLE 1. Total Number of Tested Bacteria for Each Antibiotic, Divided by Sample Site, From 1985 to 2000
or conjunctival culture from January 1985 through May 2000 no statistically difference between years. The total number at the Federal University of Sao Paulo, Brazil were reviewed.
of each bacterial group isolates in this period was as Cultures were performed using both liquid (brain and heart follows: Pseudomonas sp, 170 samples (12 conjunctival; 158 infusion) and solid media (5% sheep blood agar, chocolate corneal); S epidermidis, 1,004 samples ( 831 conjunctival; agar) and all isolates were identified. The four most frequent 173 corneal); S. aureus, 2,260 samples (1,861 conjunc- were analyzed and had their in vitro susceptibility test tival; 399 corneal); and Streptococcus sp, 314 samples (128 performed (coagulase negative Staphylococcus (CNS), Staph- conjunctival; 186 corneal). The number of bacteria iso- ylococcus aureus, Streptococcus pneumoniae, and Pseudomonas lated from corneal and conjunctival cultures that were aeruginosa). In vitro susceptibility testing was determined by tested for each antibiotic susceptibility is shown in the Kirby-Bauer disk diffusion method and interpreted using The total number and statistical data for each type of the National Committee for Clinical Laboratory Standards bacteria analyzed from corneal and conjunctival samples, (NCCLS) serum Isolates of intermediate sensitiv- ity were categorized with the resistant organisms, because the The antibiotic-susceptibility profiles for the four differ- number of intermediates was insignificant compared with the ent bacteria tested are represented in all of the Tables.
whole sample. Susceptibility testing of the bacterial isolates For the aminoglycosides, shows a decrease of was performed to the fluoroquinolones (ciprofloxacin, nor- Streptococcus sp susceptibility to amikacin in corneal and floxacin, and ofloxacin), aminoglycosides (amikacin, neomy- conjunctival samples (P ϭ .0002 and P ϭ .0004, respec- tively) and an increase in CNS (P Յ .0001) and Staphy- (cephalothin), and chloramphenicol. Samples from conjunc- lococcus aureus (P ϭ .0001) in conjunctival samples over tiva and from cornea were analyzed separately. All laboratory time. The analysis of gentamicin in vitro activity over time tests were performed using the same methodology over theanalyzed period and by the same microbiologist.
revealed a significant decrease of Streptococcus and S. For statistical analysis, logistic regression was used to fit aureus susceptibilities in corneal samples (P ϭ .004 and P odds of susceptibility vs month. Plots were shown as Յ .0001, respectively) and conjunctival samples (P ϭ .006 kernel-smoothed estimates of probability of susceptibility and P ϭ .002, respectively; demonstrated vs time. A gaussian (normal) kernel with a 4-year band- an increase of S. aureus and CNS susceptibilities to neomycin in corneal samples (P Յ .0001 and P ϭ .001,respectively) and conjunctival samples (P Յ .0001 inboth) and an increase of Pseudomonas susceptibility in corneal samples (P Յ .0001). The analysis of tobramycinin vitro activity showed a decrease of Streptococcus and S. FROM 1985 TO 2000, 3,779 CORNEAL CULTURES AND 4,585 aureus susceptibilities in corneal samples (P Յ .0001 and P conjunctival cultures were obtained. The number of cor- ϭ .0003, respectively) and conjunctival samples (P Յ neal and conjunctival cultures obtained and the percent- .0001 in both) but an increase of CNS susceptibility in age of positive cultures remained relatively constant with TABLE 2. Total Number of Samples Analyzed From 1985 to 2000
CI ϭ confidence interval; CNS ϭ coagulase-negative staphylococcus.
Odds ratio (OR) representing the change in odds every year, P values of all bacteria analyzed, divided For the fluoroquinolones, shows an increase shows an increase in Streptococcus, S. aureus, of Streptococcus, S. aureus, and Pseudomonas susceptibil- and Pseudomonas susceptibilities to ofloxacin in corneal ities to ciprofloxacin in corneal samples (P Յ .0001, P ϭ samples (P Յ .0001, P ϭ .04, and P ϭ .004, respec- .02, P ϭ .02, respectively) and an increase of Strepto- tively) and also an increase in S. aureus and CNS coccus and CNS susceptibilities to ciprofloxacin in susceptibilities to ofloxacin in conjunctival samples (P conjunctival samples (P ϭ .04 and P Յ .0001, respec- ϭ .04 and P ϭ .0001, respectively).
tively). The analysis of norfloxacin in vitro activity For the cephalosporins, shows a decrease of S. demonstrated an increase of Streptococcus and Pseudo- aureus and Pseudomonas susceptibilities to cephalothin in monas susceptibilities in corneal samples (P ϭ .0004 and corneal samples (P ϭ .005 and P Յ .0001, respectively) P ϭ .005, respectively) and an increase of CNS suscep- and a decrease of S. aureus susceptibility in conjunctival tibility in conjunctival samples (P Յ .0001; SHIFTING TRENDS IN IN VITRO ANTIBIOTIC SUSCEPTIBILITIES TABLE 3. Total Number of Samples Analyzed From 1985 to 2000
CI ϭ confidence interval; CNS ϭ coagulase-negative staphylococcus.
Odds ratio (OR) representing the change in odds every year, P values of all bacteria analyzed, divided by antibiotic, in conjunctival samples.
The analysis of chloramphenicol in vitro activity revealed antibiotic susceptibility testing can be Com- an increase of all tested bacteria susceptibilities in corneal monly, a broad-spectrum antibiotic is chosen as a first-line samples and also an increase in Streptococcus, S. aureus and treatment. Although the results of standardized in vitro tests CNS susceptibilities in conjunctival samples are based on drug concentrations that are safely achievable inplasma and do not reflect the concentrations achievable withtopical application or the effect of local factors on drug activity, they provide a useful estimate of the comparativesusceptibility of organisms to different BACTERIAL EXTERNAL OCULAR INFECTIONS, ESPECIALLY The emergence of antibiotic-resistant ocular isolates has bacterial keratitis, can be sight-threatening, leading to a always been a The resistance pattern varies in prompt treatment even before pathogen identification and different countries and different parts of the globe owing to TABLE 4. Number of Bacteria Susceptible to Amikacin, Total Number of Bacteria Tested, and Susceptibility Percentage of the
Different Bacterial Groups in Cornea and Conjunctiva Samples, Grouped in 4-Year Analysis CNS ϭ coagulase-negative Staphylococcus; N ϭ number of bacteria; %S ϭ percentage of susceptible organisms; S ϭ number of TABLE 5. Number of Bacteria Susceptible to Gentamicin, Total Number of Bacteria Tested, and Susceptibility Percentage of the
Different Bacterial Groups in Cornea and Conjunctiva Samples, Grouped in 4-Year Analysis CNS ϭ coagulase-negative Staphylococcus; N ϭ number of bacteria; %S ϭ percentage of susceptible organisms; S ϭ number of seasonal and climatic differences as well as cultural differ- tobramycin, two antibiotics widely used, showed a decrease The bacteria analyzed were the most frequent of their sensitivity over time. This finding correlates with causes of bacterial conjunctivitis and keratitis described in some published reports of emerging resistance to amino- The analysis of aminoglycosides susceptibilities to each The fluoroquinolones started to be tested in our labora- pathogen tested varied from each antibiotic tested. Ami- tory in 1990. All three different fluoroquinolones tested kacin, an aminoglycoside that is not frequently and rou- presented the same sensitivity pattern during the study tinely used in our community, showed an improvement in period. There was a slight decrease in all pathogen suscep- its sensitivity during the study period. Neomycin was used tibilities to quinolones in 1993 and 1994, with a subse- more in the past, and its increase of susceptibility may be quent increase. Interestingly, this decrease in sensitivity due to its lesser use nowadays. Conversely, gentamicin and happened at the same time that several cases of emerging SHIFTING TRENDS IN IN VITRO ANTIBIOTIC SUSCEPTIBILITIES TABLE 6. Number of Bacteria Susceptible to Neomycin, Total Number of Bacteria Tested, and Susceptibility Percentage of the
Different Bacterial Groups in Cornea and Conjunctiva Samples, Grouped in 4-Year Analysis CNS ϭ coagulase-negative Staphylococcus; N ϭ number of bacteria; %S ϭ percentage of susceptible organisms; S ϭ number of TABLE 7. Number of Bacteria Susceptible to Tobramycin, Total Number of Bacteria Tested, and Susceptibility Percentage of the
Different Bacterial Groups in Cornea and Conjunctiva Samples, Grouped in 4-Year Analysis CNS ϭ coagulase-negative Staphylococcus; N ϭ number of bacteria; %S ϭ percentage of susceptible organisms; S ϭ number of fluoroquinolone resistance were published in the litera- The analysis of the first-generation cephalosporin, ceph- alothin, revealed a decrease of S. aureus and Pseudomonas decrease in sensitivity was exacerbated in corneal samples susceptibility with time. The Pseudomonas sensitivity to compared with conjunctival samples. One of the possible cephalotin was expected to be low, as the first-generation explanations for this is that bacteria that cause corneal cephalosporin has its main activity against gram-positive infections tend to be more pathogenic than the ones microorganisms. Cephalothin is frequently used in our causing conjunctival infections. Besides that, all fluoro- community as a fortified antibiotic in treating bacterial quinolones tested in this study showed high efficacy when keratitis. This frequent use could have contributed to the analyzing the four different bacteria since 1998, proving decrease of S. aureus sensitivity during the study period. It that in our community quinolones are still a good choice is important to point out that the first-generation cepha- losporin used in ophthalmology in the United States is TABLE 8. Number of Bacteria Susceptible to Ciprofloxacin, Total Number of Bacteria Tested, and Susceptibility Percentage of
the Different Bacterial Groups in Cornea and Conjunctiva Samples, Grouped in 4-Year Analysis CNS ϭ coagulase-negative Staphylococcus; N ϭ number of bacteria; %S ϭ percentage of susceptible organisms; S ϭ number of TABLE 9. Number of Bacteria Susceptible to Norfloxacin, Total Number of Bacteria Tested, and Susceptibility Percentage of the
Different Bacterial Groups in Cornea and Conjunctiva Samples, Grouped in 4-Year Analysis CNS ϭ coagulase-negative Staphylococcus; N ϭ number of bacteria; %S ϭ percentage of susceptible organisms; S ϭ number of cefazolin, and in Brazil it is cephalothin, but cefazolin is for this is that after so many years of this drug being considered comparable to cephalothin.
discredited and not being used, the bacterial susceptibility Chloramphenicol, a broad-spectrum antibiotic, fell into to it improved again, especially for gram-positive patho- discredit since the mid 1980s owing to its high resistance indexes against most ocular pathogens. In our community, There are some points in this study that should be chloramphenicol was the most commonly used antibiotic addressed. Resistance based on in vitro testing may not in the 1960s and 1970s, but most of the ophthalmologists reflect true clinical resistance, because topical intensive stopped using chloramphenicol in the 1990s because bac- use of antibiotics can produce corneal concentrations that teria susceptibility was very low. The analysis of chloram- exceed mean inhibitory concentration Antibiotic phenicol susceptibility to the four pathogens tested showed sensitivities may not correspond to clinical response to an an important increase with time. One possible explanation antibiotic because of host factors and penetration of the SHIFTING TRENDS IN IN VITRO ANTIBIOTIC SUSCEPTIBILITIES TABLE 10. Number of Bacteria Susceptible to Ofloxacin, Total Number of Bacteria Tested, and Susceptibility Percentage of the
Different Bacterial Groups in Cornea and Conjunctiva Samples, Grouped in 4-Year Analysis CNS ϭ coagulase-negative Staphylococcus; N ϭ number of bacteria; %S ϭ percentage of susceptible organisms; S ϭ number of TABLE 11. Number of Bacteria Susceptible to Cephalothin, Total Number of Bacteria Tested, and Susceptibility Percentage of
the Different Bacterial Groups in Cornea and Conjunctiva Samples, Grouped in 4-Year Analysis CNS ϭ coagulase-negative Staphylococcus; N ϭ number of bacteria; %S ϭ percentage of susceptible organisms; S ϭ number of drug. However, in vitro susceptibility testing is the only the pathogens tested over these 15 years, along with the established method of determining antibiotic resistance cephalosporin tested, cephalothin. The gram-negative and was performed in a standardized method at our bacteria had a good response to the aminoglycosydes amikacin, gentamicin, and tobramycin, and we would microorganisms isolated in our population, and the anti- suggest using aminoglycosides to treat these infections.
biotics analyzed were the most prescribed and used in our All the fluoroquinolones tested in this study, conversely, community. There are several different microorganisms represented a great choice for treating ocular infections, and antibiotics that could be studied, and our plan is to with high susceptibility to all pathogens tested. Chlor- analyze different bacteria and drugs in the near future.
amphenicol also revealed an increase in its susceptibility In conclusion, this study showed that, in South to all pathogens, showing that after a long period America, the most used aminoglycosides, gentamicin without using this drug, the sensitivity tends to get and tobramycin, decreased their in vitro susceptibility to TABLE 12. Number of Bacteria Susceptible to Chloramphenicol, Total Number of Bacteria Tested, and Susceptibility Percentage
of the Different Bacterial Groups in Cornea and Conjunctiva Samples, Grouped in 4 Years Analysis CNS ϭ coagulase-negative Staphylococcus; N ϭ number of bacteria; %S ϭ percentage of susceptible organisms; S ϭ number of 14. Liesegang TJ, Forster RK. Spectrum of microbial keratitis in South Florida. Am J Ophthalmol 1980;90:38 –47.
1. Jones DB. Decision-making in the management of microbial 15. Upadhyay MP, Karmacharya PCD, Koirala S, Tuladhar NR, keratitis. Ophthalmology 1981;88:814 –820.
Bryan LE, Smolin G. Epidemiologic characteristics, predis- 2. Tuft SJ, Matherson M. In vitro antibiotic resistance in posing factors and etiologic diagnosis of corneal ulceration in bacterial keratitis in London. Br J Ophthalmol 2000;84:687– Nepal. Am J Ophthalmol 1991;111:92–99.
16. Everett SL, Kowalski RP, Karenchak LM, Landsittel D, Day 3. Jensen HG, Felix C. In vitro antibiotic susceptibilities of R, Gordon YJ. An in vitro comparison of the susceptibilities ocular isolates in North and South America. Cornea 1998; of bacterial isolates from patients with conjunctivitis and blepharitis to newer and established topical antibiotics.
4. Kunimoto DY, Sharma S, Garg P, Rao GN. In vitro susceptibility of bacterial keratitis pathogens to ciprofloxa- 17. Osato MS, Jensen HG, Trousdale MD, et al. The compara- cin. Ophthalmology 1999;106:80 –85.
tive in vitro activity of ofloxacin and selected ophthalmic 5. Goldstein MH, Kowalski RP, Gordon YJ. Emerging fluoro- antimicrobial agents against ocular bacterial isolates. Am J quinolone resistance in bacterial keratitis. Ophthalmology 18. Montan P. Endophthalmitis. Curr Opin Ophthalmol 2001; 6. Garg P, Sharma S, Rao GN. Ciprofloxacin-resistant Pseudo- monas keratitis. Ophthalmology 1999;106:1319 –1323.
19. Ooishi M, Miyao M. Antibiotic sensitivity of recent clinical 7. Molinari H, Belfort R Jr, Almada AT, Silva VLP. Laborato- isolates from patients with ocular infections. Ophthalmo- rial analysis of 247 patients with conjuctivitis, keratitis and endophthalmitis [translation]. Rev Bras Oftalmol 1978;37: 20. Liesegang TJ. Use of antimicrobials to prevent postoperative infection in patients with cataract. Curr Opin Ophthalmol 8. National Committee for Clinical Laboratory Standards. Per- formance standards for antimicrobial disk susceptibility test.
21. Bovelacci A, Montini G, Ramacciotti PG. Changes in The National Committee for Clinical Laboratory Standards.
resistance to aminoglycoside antibiotics of different bacterial strains isolated during the period 1978 to 1983. Chemio- 9. Gritz DC, Hwang DG, Whitcher JP Jr. New horizons in antibacterial therapy. Int Ophthalmol Clin 1996;36:153– 22. Gwon A. Topical ofloxacin compared with gentamicin in the treatment of external ocular infection. Br J Ophthalmol 10. Knauf HP, Silvany R, Southern PM, Risser RC, Wilson SE.
Susceptibility of corneal and conjunctival pathogens to 23. Alexandrakis G, Alfonso EC, Miller D. Shifting trends in ciprofloxacin. Cornea 1996;15:66 –71.
bacterial keratitis in South Florida and emerging resistance 11. Serdarevic ON. Role of the fluoroquinolones in ophthalmol- to fluoroquinolones. Ophthalmology 2000;107:1497–1502.
ogy. Int Ophthalmol Clin 1993;33:163–178.
24. von Rosenstiel N, Adam D. Quinolone antibacterials: an 12. Carmichael TR, Wolpert M, Koornhof HJ. Corneal ulcer- update of their pharmacology and therapeutic use. Drugs ation at an urban African hospital. Br J Ophthalmol 1985; 25. Chaudhry NA, Flynn HW, Murray TG, Tabandeh H, Mello 13. Wahl JC, Katz HR, Abrams DA. Infectious keratitis in MO, Miller D. Emerging ciprofloxacin-resistant Pseudomonas Baltimore. Ann Ophthalmol 1991;23:234 –237.
aeruginosa. Am J Ophthalmol 1999;128:509 –510.
SHIFTING TRENDS IN IN VITRO ANTIBIOTIC SUSCEPTIBILITIES

Source: http://www.doutorachalita.com.br/pdf/DoutoradoAJO.pdf