REVIEW / Vis. Pan-Am.

2014;13(3):70-72

Management of acute bacterial keratitis:

Fortified antibiotics or fluoroquinolones?
Ana Luisa Höfling-Lima MD, PhD1, Francisco Bandeira e Silva MD2 Corresponding author:
1. Professor of Ophthalmology and Chair, Federal University of Sao Paulo, Brazil. Ana Luisa Höfling-Lima
Av. Ibijau, 331 - 17o. andar Moema - SP - Brasil 04524.020
2. Post-graduate cornea fellow, Department of Ophthalmology, Federal University of Sao Paulo, Brazil. Email: analhofling@gmail.com
Funding: None

Proprietary/financial interest: None

Abstract or tobramycin) and a cephalosporin (cefazolin) there is a lack of consensus regarding which
Bacterial keratitis (BK) is one of the in order to cover both Gram-positive and fluoroquinolone generation is most suitable
most frequent causes for emergency hospital Gram-negative bacteria, with special concern for empirical therapy. For instance, in some
admissions.1 Identifying the causative to Staphylococcus spp., Streptococcus countries, such as the UK, the preferred initial
microorganism promptly and properly is spp. and Pseudomonas aeruginosa.8 In the therapy is still ciprofloxacin, which is partly
mandatory to achieve acceptable outcomes. 1990’s, quinolones have been made widely justified due to better in vitro activity against
Nevertheless, appropriate initial management of accessible for topical use in ophthalmology.9 Pseudomonas aeruginosa.13 However, there
these cases requires laboratory-based diagnosis Structural changes over the years led to have been some case reports and retrospective
and even a modest laboratory set may not always the production of different molecules of studies showing the development of bacterial
be available at some clinical settings. fluoroquinolones presenting with distinct resistance to these new fluoroquinolones,
Key words: bacterial keratitis; pharmacokinetic and pharmacodynamic probably due to over-the-counter availability
diagnosis; treatment. properties. The characteristics of 4th in some places and widespread topical and
generation fluoroquinolones , which includes systemic use for both prophylaxis and infection.
Relevant evidence-based gatifloxacin and moxifloxacin, made them
information an excellent choice for initial therapy of Results
Guidelines support an empiric approach for bacterial keratitis with the perks of a very broad Choy et al14 found that 11% of 65
initial treatment of bacterial keratitis2-5 but, there spectrum, higher activity against Gram-positive Pseudomonas aeruginosa isolates in both
is still no consensus over which antibiotics pathogens, readily accessibility, improved contact lens- and non-contact lens-related
should be a clinician start the treatment.6,7 aqueous humor and corneal penetration10 and keratitis were non-susceptible in vitro to both
For about four decades, a combination of no special conditions for conservation.11,12 ofloxacin and moxifloxacin, an increasing
fortified antibiotics was the preferred therapy, Fourth generation fluoroquinolones have been trend compared to previous years. Betanzos-
usually with an amynoglicoside (gentamicin used worldwide for the last decade; however, cabreras et al15 found a resistance rate of
14.2% to gatifloxacin and moxifloxacin
resistance after analyzing S. epidermidis
isolates recovered from corneal ulcers.
National surveillance studies in the United
States have reported a high frequency of
resistant to the fluoroquinolones among
ocular staphylococci isolates, particularly
Table 1. Resistance rates of 539 cultured bacterial isolates from keratitis, São Paulo, Brazil for methicillin-resistant strains.16 Along with
MRSA (methicillin-resistant Staphylococcus aureus); MRCNS (methicillin-resistant coagulase-negative these alarming findings, the controversy about
Staphylococcus); MSCNS (methicillin-susceptible coagulase-negative Staphylococcus). the best choice for empirical therapy has been
† Total number of samples cultured for each group of bacteria. brought back into the light again.
* Total number of samples submitted to antimicrobial susceptibility testing. We conducted a retrospective analysis of
a) Only third generation cephalosporins presented anti-pseudomonas activity. samples from 893 cases of BK with positive
b,c,d) Reference used was based on disk diffusion breakpoints provided by CLSI.
culture and studied separately 75 Pseudomonas
e) Breakpoints used were based on MICs and disk diffusion test provided by CLSI.
f) Reference used for Pseudomonas spp was adapted from disk diffusion breakpoints provided by CLSI for urinary
spp; 21 methicillin-resistant S. aureus
tract isolates. Other references were based on provided MICs and disk diffusion breakpoints in the CLSI. (MRSA), 129 methicillin-resistant (MRCNS)
g) The only breakpoint available for moxifloxacin in the CLSI referred to Staphylococci spp (both S. aureus and 314 susceptible S. coagulase-negative
and Coagulase-negative Staphylococci; susceptible or resistant to methicillin). Breakpoints for disk diffusion staphylococci (MSCNS). Resistance rates are
were used as reference. displayed individually for cephalothin, amikacin,

70 PAN-AMERICA

tobramycin, gentamicin, moxifloxacin and
gatifloxacin (Table 1).
Our findings display an overall high rate of
resistance to 4th generation fluoroquinolones
comparable to the resistance rates found for
cephalothin and amynoglicosides. MRSA
and MRCNS seem to be the most concerning
bacterial agents, since resistance levels for
both fortified combinations and 4th generation
fluoroquinolones are the highest found among
our isolates. Newer reviews show that main
aspects one has to take into account when
choosing an empirical antibiotic therapy
are the following: treatment success, time
to treat, serious complications of infection
and adverse effects of the medication.7
Studies had compared and reviewed
thoroughly the sensitivity profile, advantages
and drawbacks related to commercial
accessibility, bioavailability presentation,
toxicity, preservation, and effectiveness of
each drug.10,17 Treatment outcomes with either
group vary greatly according to geographical
differences, patient profile (severe or non-
severe keratitis), clinical setting (hospital
and outpatient clinics). Absolute data seems
solid at first in most of these papers; however,
when their design is unraveled, results often
lose power. The main reason for that is related
to the criteria used to recruit patients and
protocol used to determine which antibiotics
and posology would be applied, which often,
do not match with other studies. A metanalysis
evaluation of this data is not very feasible.
Besides, when it comes to bacterial resistance
to antibiotics, analysis becomes even more
difficult since culture and antimicrobial
susceptibility methods vary widely among
the studies and in several publications the
amount of positive samples for each identified
microorganism is small for a reliable statistical
evaluation. Another hiccup for microbiology
studies in BK is that methods for antimicrobial
susceptibility testing for ophthalmology
purposes are bound to major groups, such as
Clinical and Laboratory Standards Institute and
EUCAST, and all of their data is based on the
levels of antibiotics achieved in tissues after
systemic use, which may not work the same
way for topical application of drugs, since
the relationship between drug bioavailability
and ocular penetration depends on other
factors related to the specific anatomy of
ocular structures. Furthermore, some of the
drugs used in ophthalmology are not available
Höfling-Lima AL, Bandeira e Silva F. Management of acute bacterial keratitis.
for systemic use, hence, they are not listed
in these guidelines. For instance, 2014
EUCAST provided a guidance document on
breakpoints for topical use of antimicrobial
agents. However, topical breakpoints for
4th generation fluoroquinolones or other
amynoglicosides other than gentamicin are
not listed in this document. Alternatively to the
status-quo dilemma, new treatments are under
evaluation such as the effect of crosslinking
on antimicrobial activity18 and effectiveness
of besifloxacin. Besifloxacin, for topical use,
is a new fluoroquinolone that has an adhesive
component in its composition, allowing it
to remain for longer periods in the ocular
surface, thus providing a greater concentration
over time when compared to antibiotics that
are quickly removed by blinking and/or tear
drainage. Nevertheless, there are no clinical
trials evaluating the role of besifloxacin in
BK. In one of the few papers available on the
matter, Chung et al compared the penetration
of four fluoroquinolones and demonstrated
that besifloxacin does not reach high levels
within corneal tissues.19 Furthermore, when
there is already epithelial disruption, such
as in BK, it would probably penetrate better20
and, since its retention time is the longest of
all topical antibiotics, it may be taken into
account as a therapeutic option. McDonald and
colleagues conducted a thorough systematic
review and metanalysis that resulted in
16 high quality clinical trials concerning
topical antibiotics for management of BK and
concluded that none of the trials took into
consideration cost analysis and time without
antibiotics until treatment. They support that
there are significant differences between
fluoroquinolones and fortified antibiotics
when it comes to ready accessibility, and
suggest that whereas fluoroquinolones
are easily dispensed from hospitals or
community pharmacies and can be kept in
room temperature, fortified aminoglycoside-
cephalosporin combination must be prepared
in compound pharmacies and must remain
refrigerated for approximately 4 days. In
their review, they also denote that the time
requirement for fortified antibiotics formulation
might be hazardous for patients with severe
corneal infections and melting that require
immediate treatment, such as in P. aeruginosa
keratitis. Although we are in agreement with
both statements, we believe that the rise in
fluoroquinolone resistance is a major factor
PAN-AMERICA 71
 REVIEW / Vis. Pan-Am. 2014;13(3):70-72
to take into account when choosing an initial
therapy. Antibiotic resistance due to genetic
mutation is the most dreaded phenomena
to infectious diseases experts, while there
have been some published papers on the
subject21-24 but we also agree that lack of
fluoroquinolones susceptibility for some
bacteria may be due to intrinsic antibiotic,
rather than acquired resistance through
mutation.25 Nevertheless, some bacteria
that show resistance to ciprofloxacin have a
high probability to be resistant to treatment
with 4th generation quinolones, disregarding
the mechanism of resistance. This
hypothesis is supported based on several
essays demonstrating that ciprofloxacin
is the most active fluorquinolone against
P. aeruginosa.26 However, further research
has not been able to show strains that are
resistant to ciprofloxacin and susceptible to
other generation of fluorquinolones, such
as 4th generation.27,28 On that account, in a
community setting with a high prevalence
of P. aeruginosa, ciprofloxacin could be
preferred initial treatment over fourth
generation fluoroquinolones for BK. In most
South American ophthalmology hospitals
and outpatient wards, the preferred initial
treatment for BK starts with 4th generation
fluoroquinolones. We suggest that for
both severe and non-severe BK, treatment
should be initiated with 4th generation
fluoroquinolones and adjusted according
to clinical course and culture/antimicrobial
susceptibility results. In the setting of severe
and/or fast evolving infections, the time
taken in order to get results from cultures and
antimicrobial infections is golden; because
a delay in the decision taking of exchanging
4th generation fluoroquinolones for fortified
antibiotics might result in poorer outcomes.
Sometimes, the sole identification of
pathologic bacteria may help in choosing
whether and when to start fortified antibiotics,
specially if there is previous epidemiologic
data available for BK.
Conclusion and recommendation
In conclusion, gold standard initial therapy
for BK is far from a clinical and academic
consensus. Bacterial resistance to the
antimicrobials used in ophthalmology varies
greatly according to geographic location, and
initial treatment choice should be guided by local
antimicrobial resistance profile data, clinical
course and laboratory evidence of antibiotic
72 PAN-AMERICA
susceptibility. In severe cases that need prompt treatment, 4th generation
fluoroquinolones seem to be a more suitable option, once they are readily
available virtually anywhere and they do not need special conditions for
storage. Clinical course will dictate further change in therapy to fortified
antibiotics, when there is no objective or subjective evidence of clinical
improvement, and laboratory results are unavailable.
Acknowledgements: Maria Cecília Zorat Yu, Thomas Chagas-
Neto and Paulo José Martins Bispo.
REFERENCES
1. Suwan-Apichon O, Reyes JM, Herretes S, et al. Topical corticosteroids as adjunctive therapy for
bacterial keratitis. Cochrane Database Syst Rev 2007(4):CD005430.
2. Blanton CL, Rapuano CJ, Cohen EJ, Laibson PR. Initial treatment of microbial keratitis. CLAO J
1996;22(2):136-40.
3. Bennett HG, Hay J, Kirkness CM, et al. Antimicrobial management of presumed microbial kerati-
tis: guidelines for treatment of central and peripheral ulcers. Br J Ophthalmol 1998;82(2):137-
45.
4. American Academy of Ophthalmology Cornea/External Disease Panel. Preferred Practice Pattern
Guidelines: Bacterial Keratitis-Limited Revision. San Francisco, CA: American Academy of
Ophthalmology, 2011.
5. International Clinical Guidelines: Bacterial Keratitis. International Council of Ophthalmology.
International Council of Ophtalmology, 2010.
6. Shah VM, Tandon R, Satpathy G, et al. Randomized clinical study for comparative evaluation of
fourth-generation fluoroquinolones with the combination of fortified antibiotics in the treatment
of bacterial corneal ulcers. Cornea 2010;29(7):751-7.
7. McDonald EM, Ram FS, Patel DV, McGhee CN. Topical antibiotics for the management of
bacterial keratitis: an evidence-based review of high quality randomised controlled trials. Br J
Ophthalmol 2014.
8. Baum JL. Initial therapy of suspected microbial corneal ulcers. I. Broad antibiotic therapy based
on prevalence of organisms. Surv Ophthalmol 1979;24(2):97-105.
9. Wong RL, Gangwani RA, Yu LW, Lai JS. New treatments for bacterial keratitis. J Ophthalmol
2012;2012:831502.
10. Robertson SM, Curtis MA, Schlech BA, et al. Ocular pharmacokinetics of moxifloxacin after
topical treatment of animals and humans. Surv Ophthalmol 2005;50 Suppl 1:S32-45.
11. Fisher LM, Gould KA, Pan XS, et al. Analysis of dual active fluoroquinolones in Streptococcus
pneumoniae. J Antimicrob Chemother 2003;52(2):312-3; author reply 3-4.
12. Fisher LM, Heaton VJ. Dual activity of fluoroquinolones against Streptococcus pneumoniae. J
Antimicrob Chemother 2003;51(2):463-4; author reply 4-5.
13. 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(11):2161-5.
14. Choy MH, Stapleton F, Willcox MD, Zhu H. Comparison of virulence factors in Pseudomonas
aeruginosa strains isolated from contact lens- and non-contact lens-related keratitis. J Med
Microbiol 2008;57(Pt 12):1539-46.
15. Betanzos-Cabrera G, Juarez-Verdayes MA, Gonzalez-Gonzalez G, et al. Gatifloxacin, moxifloxa-
cin, and balofloxacin resistance due to mutations in the gyrA and parC genes of Staphylococcus
epidermidis strains isolated from patients with endophthalmitis, corneal ulcers and conjunctivitis.
Ophthalmic Res 2009;42(1):43-8.
16. Haas W, Pillar CM, Torres M, et al. Monitoring antibiotic resistance in ocular microorganisms:
results from the Antibiotic Resistance Monitoring in Ocular micRorganisms (ARMOR) 2009
surveillance study. Am J Ophthalmol 2011;152(4):567-74 e3.
17. Wilhelmus KR, Gilbert ML, Osato MS. Tobramycin in ophthalmology. Surv Ophthalmol
1987;32(2):111-22.
18. Iseli HP, Thiel MA, Hafezi F, et al. Ultraviolet A/riboflavin corneal cross-linking for infectious
keratitis associated with corneal melts. Cornea 2008;27(5):590-4.
19. Chung JL, Lim EH, Song SW, et al. Comparative intraocular penetration of 4 fluoroquinolones
after topical instillation. Cornea 2013;32(7):1046-51.
20. Fukuda M, Inoue A, Sasaki K, Takahashi N. The effect of the corneal epithelium on the intraocu-
lar penetration of fluoroquinolone ophthalmic solution. Jpn J Ophthalmol 2004;48(2):93-6.
21. Agnello M, Wong-Beringer A. Differentiation in quinolone resistance by virulence genotype in
Pseudomonas aeruginosa. PLoS One 2012;7(8):e42973.
22. Mouneimne H, Robert J, Jarlier V, Cambau E. Type II topoisomerase mutations in ciprofloxacin-
resistant strains of Pseudomonas aeruginosa. Antimicrob Agents Chemother 1999;43(1):62-6.
23. Nakano M, Deguchi T, Kawamura T, et al. Mutations in the gyrA and parC genes in fluoroqui-
nolone-resistant clinical isolates of Pseudomonas aeruginosa. Antimicrob Agents Chemother
1997;41(10):2289-91.
24. Higgins PG, Fluit AC, Milatovic D, et al. Mutations in GyrA, ParC, MexR and NfxB in clinical
isolates of Pseudomonas aeruginosa. Int J Antimicrob Agents 2003;21(5):409-13.
25. Agarwal T, Jhanji V, Satpathy G, et al. Moxifloxacin resistance: intrinsic to antibiotic or related to
mutation? Optom Vis Sci 2012;89(12):1721-4.
26. Swiatlo E, Moore E, Watt J, McDaniel LS. In vitro activity of four fluoroquinolones against
clinical isolates of Pseudomonas aeruginosa determined by the E test. Int J Antimicrob Agents
2000;15(1):73-6.
27. Duggirala A, Joseph J, Sharma S, et al. Activity of newer fluoroquinolones against gram-positive
and gram-negative bacteria isolated from ocular infections: an in vitro comparison. Indian J
Ophthalmol 2007;55(1):15-9.
28. Moshirfar M, Mirzaian G, Feiz V, Kang PC. Fourth-generation fluoroquinolone-resistant bacterial
keratitis after refractive surgery. J Cataract Refract Surg 2006;32(3):515-8.