Introduction

Ulcerative keratitis is one of the most common canine ocular diseases (Ollivier 2003; Gilger 2007). A corneal ulcer is present when there is a break in the corneal epithelium that exposes the underlying corneal stroma. Uncomplicated superficial ulcers usually heal rapidly but it can become an eye threatening disease if infection occurs (Gilger 2007). Most commensal bacteria constitute the conjunctival sac microbiota, having a role in maintenance of normal ocular health by preventing overgrowth of potentially pathogenic agents (Gaskin 1980; Gerding & Kakoma 1990; Dupont et al. 1994, Andrade et al. 2002). However, microorganisms of the normal flora can become potential pathogens if tissue damage to the cornea occurs, or if the organism's resistance to infection has decreased (Winn et al. 2006). Few canine conjunctival microflora studies are available and predominating microbiota includes S. epidermidis, S. aureus, Streptococcus spp., Neisseria spp., Enterobacteriaceae, and Pseudomonas spp. (Gerding & Kakoma 1990; Prado et al 2005; Andrade et al. 2002). Fungi are rarer as described by Samuelson et al. (1984), relating Cladosporium oxysporum and Curvularia lunata. In an investigation in Beijing, China, from 29 eyes with ulcerative keratitis, Staphylococcus spp. was the most frequently isolated bacteria. Streptococcus spp. and Pseudomonas spp. were the second and the third, respectively (Wang et al. 2008). Gram-positive bacteria predominated over gram-negative in the conjunctival sac of clinically normal dogs and dogs with ulcerative keratitis in Ceará, Brazil, and Staphylococcus spp. was the most frequently isolated genus (Prado et al. 2005). Culture and sensitivity testing provide useful information for the diagnosis and determination of appropriate antimicrobial therapy in many corneal diseases (Ollivier 2003). The purpose of this study was to analyze the potential pathogens found in corneal ulcers in dogs and compare the antimicrobial sensitivity tests results with commercial available eye drop for empiric therapy and its importance for clinical outcome.

Materials and Methods

This investigation included 86 samples taken from corneal ulcers of 80 dogs of different breeds referred to our private clinic located in Valinhos, São Paulo, Brazil, in the period of 2004 to 2008. Most of the dogs were already in pretreatment. The samples were collected using sterile swabs moistened by sterile saline that were passed directly on the corneal ulcer of each affected eye. The swabs were removed taking care to avoid contact with the eyelashes or skin of the eyelids. The swabs were transported in Cary Blair medium. Each swab was inoculated onto a blood agar, MacConkey agar and Sabouraud agar plates. The blood agar plate was incubated at 37°C in a 5% CO₂ environment for 5 days. The MacConkey agar plate was incubated at 37°C in an aerobic environment for 5 days. The Sabouraud agar plate was incubated at 35°C in aerobic environment for 7 days. Those plates without bacterial growth after 5 days or 7 days for fungal growth were treated as negative results (Wilhelmus 1994; Murray et al. 2007). Significant growth and isolated organisms were identified by automated identification in Vitek BioMérieux® and antibiotic sensitivity tests by manual methodology as standardized by CLSI 2008 and almost all drugs tested were of antibiotics available for topical use.

Results

Population: The sample age ranged from 6 months to 14 years, with greater concentration between 1 and 6 years and represented mostly by Shih-Tzu (31.25%), followed by Poodle (15%) and Lhasa Apso (8.75%).

Microbiology: From 86 samples cultured, 22 were negative (25.58%) and 64 (74.42%) positive, 59 (92.18%) for one or more bacteria and 5 (7.82%) for yeasts. The gram-negative rods were the most frequent group (48.43%) and the most common isolated bacteria was Pseudomonas aeruginosa, alone in 21 cases and associated with enterobacteria in other two cases (tab. 1), followed by the gram-positive cocci (Staphylococcus sp + Streptococcus sp) [39.06%]. The susceptibility tests revealed gram-negative bacteria strains susceptible for the most common antibiotics of clinical use: quinolones (ciprofloxacin, levofloxacin, gatifloxacin and moxifloxacin), aminoglycosides (amikacin, tobramycin and gentamicin). Only one P. aeruginosa strain resistant to all antibiotics except polymixin B and two E. coli resistant to quinolones were found. One of them was associated with an unusual profile of ESBL (Extended Spectrum Beta-lactamase) resistance. For chloramphenicol and tetracycline, 52.17% of the P. aeruginosa strains were resistant. The gram-positive cocci were also susceptible to all antibiotics tested except the Staphylococcus sp for tetracycline with almost 50% resistant strains. None of the Staphylococcus sp was oxacillin resistant and only one Staphylococcus aureus and one Streptococcus equii were ciprofloxacin resistant but susceptible to other quinolones.

Antibiotic therapy: Among the negative cultures, 22.09% (19) of the animals were pretreated with antibiotics. Nevertheless, in 52.32% of positive cultures the dogs also were already treated with antibiotics. The most commonly used antibiotics alone or in association in initial empirical therapy in the latter group were chloramphenicol (30.23%), ciprofloxacin (16.27%), tobramycin (13.95%) among others such as tetracycline, gentamicin, neomycin and polymyxin B, gatifloxacin and ofloxacin. As the animal was attended at our clinic, a new antibiotic drug was introduced in almost all of the cases, once the great majority of the pretreatments have been done improperly. The therapy instituted after the animal care was modified after the result of the drug susceptibility test (antibiogram) in 20.93% of cases while in the reminder 79.07% of events the initial therapy was maintained in agreement with the antibiogram result. In some cases surgical procedures had to be performed, combined with drug therapy, being 16.28% conjunctival pedicle grafts and 12.79% third eyelid flaps. In only 9.3% of cases was not possible to determine the evolution of the ulcer since the animals did not return for the following assessments and 98.7% of the followed cases showed a good development, with ulcer healing and visual function preservation. The worst results were observed in ulcers infected by fungi and resistant bacteria.

Discussion and Conclusions

Although there are many predisponent factors for the development of corneal ulcers, such as KCS, entropion, corneal exposure (Gilger 2007), the right choice of the antimicrobial drug is fundamental, since infection is one of the most important causes of complication. The empirical initial therapy must be directed against the most common bacterial agents, e.g., Pseudomonas, Staphylococcus and Streptococcus, using broad spectrum antibiotics, among other drugs, like lubrificants and anticolagenolitics (Slatter 2005). Our findings related to the microorganisms most frequently isolated from ulcers of dogs are in agreement with literature. Some variations in frequency are considered as several factors appear to influence the prevalence of individual microorganisms; these include geography and climate, season, species, sampling and culturing techniques, and the immediate environment (Galle & Moore 2007). As expected a specialized service almost always receives complicated cases of corneal ulcers and most of the time the eyes are being treated improperly, with inadequate choice of drugs or frequency of medication. Long-term inappropriate antibiotic treatment may result in overgrowth of pathogenic bacteria, yeasts, or fungi (Galle & Moore 2007). Cultures and antibiograms are, therefore, fundamental for the success of the treatment. From our results, quinolones and aminoglycosides are the first choice antibiotics to be used immediately, before the results of the cultures and susceptibility tests. Anytime, however, it is mandatory to change the drug if there is resistance or if the clinical valuation aspect of the ulcer is getting worse. The frequency of the antibiotics application is also very important and intensive topical antibiotic therapy (every 1-2 hours) is recommended (Gilger 2007). Our high success rate (98.7%) reinforces the importance of collecting samples for cultures and sensitivity test. Although fungal keratitis are rare in dogs, they can be very aggressive and directing the therapy against the agent is fundamental to preserve the eye. The results of the culture and drug susceptibility test were essential to properly management of 18.60% of cases of our inadequate empiric therapy, showing the appropriate antimicrobial option and responsible for a better outcome. Resistance for quinolones and other antimicrobial drugs are still not frequent, but as result of human close contact this problem should emerge as signaled by two cases of two multiresistant strains, one E. coli ESBL producer, an important agent of human hospital infections (Moreno et al. 2008; Caratolli 2007) and frequently cited as animal colonizer and potential pathogen and one P. aeruginosa susceptible only to polymixin, probably as result as chronic and improperly use of antimicrobial drugs. The challenge of etiologic diagnosis of ulcerative keratitis depends on the appropriated specimen collection, transport and especially an adequate Microbiological Laboratory support (Wilhelmus, 1994). The integration of the clinician and the Laboratory is essential for correct diagnosis and optimizing drug therapy.

Table 1. Microorganisms isolated from 86 canine ocular samples associated with ulcerative keratitis.

References

1.  Ollivier FJ. Bacterial corneal diseases in dogs and cats. Clinical Techniques in Small Animal Practice, 2003; 18 (3): 193-198.

2.  Gilger BC. 2007. Diseases and surgery of the canine cornea, p.690-752. In: Gellat K.N. (Ed.), Veterinary Ophthalmology. 4th ed. Blackwell Publishing, Oxford.

3.  Gaskin JM. Microbiology of the canine and feline eye. Veterinary Clinics of North America: Small Animal Practice 1980; 10: 303-316.

4.  Gerding PA, Kakoma I. Microbiology of the canine and feline eye. Veterinary Clinics of North America: Small Animal Practice 1990; 20: 615-625.

5.  Dupont C, Carrier M, Higgins R. Bacterial and fungal flora in healthy eyes of birds of prey. Canadian Veterinary Journal 1994; 35: 699-671.

6.  Andrade AL, Stringhini G, Bonello FL, Marinho M, Perri SHV. Microbiota conjunctival de cães sadios da cidade de Araçatuba (SP). Arquivos Brasileiros de Oftalmologia 2002, 65: 323-326.

7.  Winn WJ, Allen SD, Janda WM, Koneman EW, Procop G., Schrenkenberger PC. 2006. Koneman's Color Atlas and Textbook of Diagnostic Microbiology. 6 th ed., Lippincott Willians and Willkins, Baltimore.

8.  Prado MR, Rocha MFG, Brito MDG, Monteiro AJ, Teixeira MFS, Sidrim JJC. Survey of bacterial microorganisms in the conjunctival sac of clinically normal dogs and dogs with ulcerative keratitis in Fortaleza, Ceará, Brazil. Veterinary Ophthalmology, 2005; 8:33-37.

9.  Samuelson DA, Andresen TL, Gwin RM. Conjunctival fungal flora in horses, cattle, dogs and cats. Journal of the American Animal Hospital Association, 1984; 184: 1240-1242.

10. Wang L, Qingshan P, Libo Z, Xue Q, Jun C, Qi C. Investigation of bacterial microorganisms in the conjunctival sac of clinically normal dogs and dogs with ulcerative keratitis in Beijing, China. Veterinary Ophthalmology, 2008; 11: 145-149.

11. Wilhelmus KR, Liesegang TJ, Osato MS, Jones DB. 1994. Cumitech 13A. Laboratory diagnosis of ocular infections. Coord ed. SC Specter. American Society for Microbiology, Washington DC.

12. Murray PR, Baron EJ, Jorgensen JH, Pfaler MA, Yolken RH. 2007. Manual of Clinical Microbiology. 8th ed. Ed.ASM Press, Washington, DC.

13. CLSI. 2008. Performance Standards for Antimicrobial Susceptibility Testing; Eighteenth Informational Supplement. CLSI document M100 -S18. Clinical and Laboratory Standards Institute, Wayne.

14. Slatter D. 2005. Córnea e sclera, p.283-338. In: Slatter D. Fundamentos de Oftalmologia Veterinária. 3^TM ed. Ed. Roca, São Paulo.

15. Galle LE, Moore MP. 2007. Clinical microbiology, p.236-270. In: Gellat K.N. (Ed.), Veterinary Ophthalmology. 4th ed. Blackwell Publishing, Oxford.

16. Moreno A, Bello H, Guggiana D, Domínguez M, González G. Extended-spectrum ,lactamases belonging to CTX-M group produced by Escherichia coli strains isolated from companion animals treated with enrofloxacin. Veterinary Microbiology, 2008; 129:203-208.

17. Carattoli A. Animal reservoirs for extended spectrum ,-lactamase producers. Clinical Microbiology and Infection, 2007, 14(1):117-23