Abstract
Which treatment approach?
Thorough examination to identify a primary cause for the corneal ulcer is essential, as treatment of any underlying condition is important to effect clinical success. For example, lack of correction of an entropion will result in failure of ulcer healing or early recurrence. The aetiology of corneal ulcers was reviewed by the author in a recent article in this journal. 1
The question of whether medical or surgical treatment is most appropriate depends on a number of factors, as outlined below. In practice, depending on the depth of the ulcer, many cases can be managed with topical antibiotics and/or antivirals and systemic analgesia alone. The importance of supportive care, with bathing of ocular (and nasal) discharges, assisted feeding and rehydration if necessary, and adequate pain relief should not be underestimated.
SURGICAL TREATMENT OF CORNEAL ULCERS
A sister article addressing the indications for surgical treatment appears on pages 398–405 of this issue of J Feline Med Surg, and at:
Medical or surgical therapy?
Factors to consider
Depth of the ulcer (eg, superficial versus stromal defect or descemetocoele)
Progression of the ulcer (eg, malacia, infection)
Concurrent ophthalmic disease processes (eg, tear film abnormalities, FHV-1 infection, sequestrum, entropion)
Patient's general health (risks of anaesthesia, side effects of some drugs in, eg, renal disease, etc)
Patient's demeanour (eg, feasibility of medicating topically or orally)
Owner compliance and financial circumstances
Placement of a therapeutic soft contact lens can be considered for analgesia, as described on the right.
Bacterial keratitis
The choice of antimicrobial drug in cases of bacterial keratitis will depend on the expected scope of sensitivity of the suspected micro-organism. Preparation of a cytological smear is highly recommended to guide this choice. Adequate levels of the therapeutic agent then need to be established and maintained at the site of infection, with the lowest toxicity.
Gram-positive coccoid bacteria (eg, Staphylococcus and Streptococcus species) are the most common isolates from the feline con-junctiva.3,4 Similarly, the most common bacterial isolates in feline (and canine) patients with external eye disease were Staphylococcus species, followed by Streptococcus species and then Pseudomonas species. 5 The presence of Gram-negative bacteria, while less common, might indicate a different topical antibiotic as being the most appropriate choice.
Mycoplasma species have been reported in association with ulcerative keratitis in cats. 6 On cytology inclusion bodies may be identified within the cytoplasm of epithelial cells, 6 but are not specific for Mycoplasma; nor is culture particularly sensitive. The development of a polymerase chain reaction (PCR)-based test for Mycoplasma has, however, improved diagnostic rates. 6 The authors of the above-mentioned report concluded that the Mycoplasma species were unlikely to be the primary aetiological agents but were nonetheless clinically relevant. In an earlier study, experimental inoculation of the conjunctival sacs of healthy cats with Mycoplasma species resulted in conjunctivitis that was self-limiting in the absence of ulcerative keratitis. 7
Treatment of ulceration associated with Mycoplasma species has been reported with topical tetracyclines, ofloxacin, chloramphenicol, erythromycin and oral azithromycin.6,8 Resistance of Mycoplasma species to neomycin has been described. 9
Anaerobic and aerobic bacterial culture and sensitivity testing is advisable before commencing treatment, particularly where secondary infection is suspected or an ulcer is rapidly progressing. Anaerobic bacteria were isolated from 7.9% of feline ulcerative keratitis bacterial swab samples in one study; 10 Clostridium, Peptostreptococcus, Actinomyces, Fusobacterium and Bacteroides species were the most common of these isolates. Cytology, however, has the advantage that it is both fast and low cost, and can be employed while awaiting culture and sensitivity test results.
Adjunctive therapies
Therapeutic contact lens
Contact lenses effectively provide a bandage for the cornea and eliminate the rubbing of eyelids across exposed corneal nerve endings. They are particularly useful following a surgical procedure (eg, excision of a sequestrum where a conjunctival pedicle graft has not been placed, or following removal of a corneal foreign body). The employment of a contact lens has advantages over a third eyelid flap, which obstructs visualisation of the cornea and, unless sutured to bulbar conjunctiva, does not move with the globe, thus defeating the purpose of the bandage.
Note that the use of contact lenses is contraindicated in the presence of secondary bacterial infection, keratomalacia or kerato-conjunctivitis sicca. If there is insufficient tear production, the cornea underneath the contact lens may become deprived of oxygen and nutrients. Where neutrophils and bacteria are trapped underneath the contact lens, release of their proteases can increase collagenolytic destruction of the cornea.
Brachycephalic cats may not retain contact lenses well due to their wide palpebral fissures. Placement of temporary tarsorrhaphy sutures to partially close the eyelids can facilitate retention.
Keratoconjunctivitis sicca (in this case secondary to viral conjunctivitis) is a contraindication to the use of therapeutic contact lenses
Tear replacement
Tear replacement should be provided where evidence of keratoconjunctivitis sicca exists (eg, in FHV-1 infection).
Topical atropine
Topical atropine can ablate the painful ciliary muscle spasm that may be present in cats with reflex uveitis, and thereby provide potent analgesia. (Reflex uveitis is a prodromal effect secondary to corneal nerve stimulation causing miosis and increased vascular permeability of uveal blood vessels.) Tear production should be assessed prior to the use of atropine as prolonged suppression of tear production can occur, which may inhibit corneal healing. 2
Topical antibacterial drugs
Topical antibacterial treatment is sufficient in most cases of ulcerative keratitis. The choice of drug and frequency of administration will depend on the depth and progression of the ulcer, as well as cytological smear and bacterial culture results. Additional factors that penetration of the drug, the potential for irritation and toxicity, drug formulation and, lastly, cost.
In practice the choice of antibacterial formulation (solution/suspension or ointment) may depend on what is available. Ointments remain in the conjunctival sac for an extended period compared with solutions or suspensions, giving a longer duration of action per application, and can be of benefit where owner compliance or cat temperament dictates. Ointments should be avoided where a soft contact lens has been placed, or when a corneal perforation is present or imminent (intraocular ointment is uveitogenic).
Frequency of topical application
Rapidly progressive ulceration or keratomalacia should receive hourly topical antimicrobial treatment. Stromal ulceration cases may require topical medications six to eight times daily, while four times daily prophylaxis may suffice for simple superficial ulceration. Ointments are retained within the tear film for longer than solutions and suspensions, and therefore may be given less frequently.
Table 1 summarises the spectrum of activity and other treatment considerations for the available topical antibacterial preparations. Gentamicin, while broad spectrum, is epitheliotoxic.11–13 Ciprofloxacin and tobramycin have also been shown to be cytotoxic to corneal epithelial cells in vitro, 14 although ciprofloxacin was identified as least epitheliotoxic in another in vitro study. 15
Topical antibacterial agents
MIC50 = minimum inhibitory concentration at which growth of 50% of isolates is inhibited
The preservative benzylalkonium chloride, which is present in the majority of commercial multi-dose antibiotic preparations, is also epitheliotoxic. 16 This epitheliotoxicity is partially responsible for enhancing the ocular penetration of antibiotics in these preparations, but should be borne in mind, particularly where the antibiotic is being used prophylactically.
Systemic antibacterial drugs
Systemic antibiotics may be indicated where accompanying injuries are present (eg, perforating foreign body injury) leading to suspected endophthalmitis, if there is a full thickness corneal perforation, or where the cat's temperament will not support frequent topical treatment. Good vascularisation needs to be present for systemic antibiosis to be effective.
Tetracyclines
Some systemic antibiotics reach the tear film at greater concentrations than others, and this is an important factor in the choice of drug. Tetracyclines are believed to concentrate within the fatty secretions of meibomian glands and are regularly used in human patients suffering from blepharoconjunctivitis.17–19 However, some studies suggest that tear concentrations of tetracyclines may not be as high as meibomian gland concentrations after oral administration.19–22 The beneficial effects of tetracyclines may extend beyond their antibacterial activity; hypotheses include the inhibition of extracellular enzymes released by ocular bacteria, improvement of meibomian gland function and anti-inflammatory effects.17–19 Doxycycline has also been shown to reduce tear film matrix metallo-proteinases in human and equine patients.20,23,24
Doxycycline has a broad spectrum of antibacterial activity, although resistance is not uncommon. 25 Efficacy against Chlamydophila species has been demonstrated both in vitro and in vivo. Gastric irritation and oesophageal stricture formation have been reported in human and feline patients following oral doxycycline medication.26–29 Growth retardation and tooth discoloration are also recognised side effects when tetracyclines are administered to young or pregnant animals. 30
Azithromycin
Azithromycin has good bioavailability in cats following oral administration, and has been proposed as a treatment for Chlamydophila felis infections by some authors, 31 although others found poor clearance of Chlamydophila species with one dosage protocol. 32 A further study, comparing oral azithromycin with oral amoxicillin as an empirical treatment for suspected upper respiratory tract bacterial infections in cats, found no statistically significant difference in clinical efficacy. 33
Fluoroquinolones
The use of parenteral and/or oral enrofloxacin should be limited to cats in which culture and sensitivity testing has shown it to be the only indicated drug. The dosage must be strictly adhered to, with a dosage reduction for renally impaired individuals. Toxic irreversible retinal degeneration has been recorded in cats receiving oral and/or intravenous enrofloxacin, and in one case this occurred at less than the recommended dosage of 5 mg/kg q24h. 34 It would be wise to consider the use of a fluoroquinolone with a wider safety margin with respect to retinotoxic effects in cats.
Subconjunctival antibacterial drugs
Subconjunctival antibiotics are reserved for severe corneal infections when sufficient therapeutic levels cannot be achieved by topical therapy alone. Some drugs (eg, chloramphenicol, tetracycline) are capable of inducing sub-conjunctival granuloma formation, scarring or irritation and should be avoided. 35 Others (eg, vancomycin), while efficacious for certain infections, should be avoided due to the risks of promoting resistance, given the importance of such drugs in multiresistant infections in human patients. 31
Keratomycosis
Although fungi are present in the conjunctival flora of normal cats, keratomycosis is rarely reported. A recent case study described the successful treatment of Aspergillus fumigatus keratomycosis with topical 1% voriconazole. 36 In this case, the fungal infection was considered secondary to primary FHV-1 keratitis.
Long-term topical corticosteroid treatment or immunosuppression may increase the risk of fungal colonisation of corneal ulcers in endemic areas.37,38 Corneal trauma with plant material is also recognised as a risk factor in other species.39–42 After invasion of compromised corneal epithelium, fungi can migrate to the deep stroma by hyphal tip elongation. An affinity to the glycosaminoglycans of the deep stroma is postulated for the apparent tropism of fungal organisms to Descemet's membrane.
Topical antifungal agents used in the treatment of keratomycosis have included amphotericin B, natamycin, nystatin, clotrimazole, miconazole, itraconazole, ketoconazole, econazole, fluconazole, voriconazole, povidone-iodine and silver sulfadiazine.37,38,43 Surgical intervention with aggressive debridement (to remove fungal organisms and neutrophils) or superficial keratectomy with conjunctival pedicle grafting (to provide vascular support, see accompanying article at
Keratomalacia
Proteases and collagenases may be released by neutrophils, corneal epithelial cells, fibro-blasts, some bacteria (notably Pseudomonas species) and fungi. In the majority of cases of keratomalacia (corneal melting) the most significant contribution is believed to originate from neutrophils.
Topical anticollagenases and antiproteases are indicated for cats with keratomalacia, and should be applied frequently (every 1–2 h). A number of drugs have been suggested as being efficacious, including acetylcysteine, EDTA, heparin, tetracycline compounds, plasma and serum.
One in vitro study on equine corneas found no significant difference in anticollagenase activity between equine serum, acetylcysteine and equine tetanus antitoxin. 44 Serum should not be mixed with other topical agents as this is believed to inactivate the anticollagenases. 45 Autologous (from the same patient) serum should be used in cats due to the risk of retro-virus transmission. Heterologous (from a different species, eg, dog) serum has been used in feline patients; although there is no literature to support this practice, clinical experience has been encouraging.
Surgical debridement of malacic tissue can assist in removal of protease load. Support of a fragile cornea and the provision of a blood supply is often indicated in keratomalacia. The choice of a conjunctival pedicle graft (see accompanying article,
FHV-1 keratitis
As described in the earlier article on the aetiology of corneal ulcers, 1 FHV-1 is an alphaher-pesvirus with a short reproductive cycle that shows rapid replication and spread in culture, efficient lysis of infected cells and the capacity to establish latency, particularly in sensory ganglia. Lifelong latency develops in approximately 80% of infected cats. 46 An estimated 50% of these latently infected cats later shed FHV-1 intermittently, and reportedly 29% may shed virus spontaneously. 47
Supportive treatment of primary FHV-1 infections in kittens is usually sufficient as most infections are self-limiting in 3–4 weeks.48,49 Broad-spectrum antibiosis, administered systemically (if respiratory disease is present) and/or topically (if corneal ulceration is present), is indicated to treat secondary bacterial infections. 48 Severe cases may require specific antiviral therapy, particularly where conjunctival and corneal ulceration is present that is likely to progress to symblepharon formation, or severe respiratory disease develops.9,50 Mechanical breakdown of early symblepharon adhesions under topical local anaesthesia using a cotton-tipped swab has been advocated to prevent permanent adhesions. 51 Attention to the respiratory signs is paramount in kittens as viral pneumonia, while uncommon, is associated with high mortality. 50
Recrudescent disease in adult cats can range from a brief episode of conjunctivitis to chronic ulcerative keratitis with progression to sequestrum formation or stromal keratitis. 52 Mild cases of conjunctivitis or dendritic ulceration may heal in a few days with no, or only supportive, treatment. Where there is progression to geographical ulceration, which is often prolonged, antiviral therapy should be considered. Effective treatment with rapid resolution of the ulceration is the best prevention for sequestrum formation or development of stromal keratitis. Early referral should be considered if an adequate response to treatment is not seen.
Redundant loose epithelium at the edge of chronic superficial ulcers can be debrided under topical local anaesthesia. Note that neither chemical cauterisation with phenol nor grid keratotomy should be used in cats; grid keratotomy will increase the risk of sequestrum formation. 53 Debridement of stromal ulceration is not warranted or appropriate.
Specific therapies
Ophthalmic antiviral agents are virostatic and therefore require very frequent application for efficacy. 35 Many cats will not tolerate this intensity of treatment and, given that stress plays an important role in the pathogenesis of the disease, treatment may be counterproductive.
Tables 2 and 3 outline the available topical and systemic antiviral therapies and their relative efficacies in cats. Owners need to be counselled on the recurrent nature of the disease, as well as the impact of stress on reactivation of latent virus.
Topical antiviral therapy for FHV-1 infection
IC50 = inhibitory concentration at which plaque numbers (measure of viral replication) are reduced by 50%
Systemic antiviral therapy for FHV-1 infection
Acyclic nucleoside analogues
The acyclic nucleoside analogues' virostatic effect is a result of their interference with viral replication. Their mode of action is competitive inhibition of DNA polymerase and chain termination of replicating DNA. Their antiviral activity is therefore largely limited to DNA viruses. The affinity of the acyclic nucleoside analogues for viral DNA polymerase is significantly greater than the natural nucleosides.
Most of these drugs require three phosphorylation steps for conversion to their active triphosphate form. The first step is catalysed by a viral kinase (thymidine kinase) within a virally infected host cell; the subsequent two steps are catalysed by host cellular kinases. Certain kinase-deficient strains of herpes viruses, as well as many non-herpesviruses, are predictably resistant to the acyclic nucleoside analogues. 54
Cidofovir differs from other acyclic nucleoside analogues in that it requires only two phosphorylation steps for activation and does not require a viral kinase-mediated phosphorylation step, making it attractive for use against viruses lacking thymidine kinase (or showing poor thymidine kinase efficiency). In an in vivo study, twice daily application significantly decreased viral shedding and clinical disease in experimentally induced ocular FHV-1 infection. 55
In an in vitro study on feline kidney cells infected with FHV-1, the inhibitory concentrations at which plaque numbers (a measure of viral replication) were reduced by 50% (IC50) by antiviral compounds were trifluorothymidine (trifluridine) 0.67 μM, idoxuridine 6.8 μM, vidarabine 21.4 μM, bromovinyl-deoxyuridine 30.1 μM and aciclovir 85.6 μM. 56 In this study the cells were incubated in varying concentrations of antiviral drug 1 h after inoculation with feline herpesvirus.
Relative efficacy against FHV-1
Trifluorothymidine > idoxuridine ≫ ganciclovir > cidofovir > penciclovir > vidarabine ≫ acyclovir = foscarnet
Another in vitro study comparing antiviral therapies on feline kidney cells found penciclovir to have a lower IC50 than trifluorothymidine, in contrast to earlier studies. However, this study also found aciclovir to be the least potent antiviral therapy for FHV-1. 57
No controlled in vivo studies of topical antiviral therapies for feline herpetic keratitis exist; however, a retrospective study suggested a poor overall response to topical anti-virals. 58 In this study, all vidarabine-treated cats (n=4) showed an improvement in clinical signs, but not complete resolution. In the seven cases treated with idoxuridine, three improved or resolved, while four became worse. Of three cases treated with trifluorothymidine, one improved, one failed to improve and one became worse (developed a sequestrum). The author conceded that inadequate frequency of antiviral therapy may have resulted in the poor overall response. 58
The systemic antiviral famciclovir (prodrug of penciclovir) has been reported to have good efficacy in cats. 59 Other systemic antivirals, such as aciclovir and valaciclovir (prodrug of aciclovir), have significant toxicity limitations (including bone marrow suppression, leukopenia, anaemia and renal tubular necrosis) and should not be used in cats.60,61 Famciclovir is metabolised by the liver and excreted by the kidney so caution should be exercised when treating cats with hepatic or renal disease.62,63 A recent study obtained similar peak plasma famciclovir and penciclovir concentrations (Cmax), and times to peak plasma concentrations (Tmax), following doses of 40 mg/kg and 90 mg/kg of famciclovir in cats. 64 This suggests the higher dose offers no advantage.
Pyrophosphate analogue (foscarnet)
Foscarnet is a non-competitive inhibitor of the pyrophosphate binding site on DNA polymerase. It does not require phosphorylation for activation and its efficacy is therefore not dependent on viral kinases. In vitro studies have demonstrated poor efficacy in FHV-1 infections.54,69
Interferon
Interferons (eg, IFN-Ω and IFN-α) are members of a cytokine family mediating innate (non-specific) immunity with antiviral, anti-proliferative and immune regulatory functions. Both oral and topical treatment for FHV-1 infections has been described. As IFNs are degraded by the gastrointestinal system, absorption is postulated to occur via the mucosal membranes of the oropharnyx. 65
In vitro studies with feline kidney cell lines demonstrated significant reductions in FHV-1 plaque size (a measure of cell resistance to viral infection) when pre-incubated with IFN-Ω (feline) or -α (human). Reduction in plaque numbers was only seen at concentrations of 100,000 U/ml and 500,000 U/ml for IFN-Ω (not IFN-α). 66 Overall, IFN-Ω may be the most appropriate of the interferons, at high doses (500,000 U/ml), early in the disease course and as an adjunct to other antiviral treatments, rather than as sole therapy. 66
An experimental study demonstrated that the response to topical or oral IFN-Ω depends on the dose and route of administration. 67 Mx protein expression (a marker of the biological response to IFN-Ω) was identified in circulating white blood cells, but not in conjunctival cells, following oral administration of IFN-Ω. Topical administration of IFN-Ω at doses less than 10,000 U did not result in Mx protein expression, suggesting this is the minimum topical dose. 67
An in vivo study using both topical and oral IFN-Ω prior to experimental infection with FHV-1 demonstrated no benefit, either clinically or in terms of viral shedding. In fact, in this study the IFN treatment groups displayed higher scores for epithelial keratitis than the control group.
The successful treatment of FHV-1 dermatitis in a cat has been reported with intralesional and subcutaneous injections of IFN-Ω on six occasions. 70 Subcutaneous injections of IFN-Ω may be an area of further research with respect to FHV-1 keratitis (previous studies have tested oral and topical administration only).
Adjunctive therapy (L-lysine)
Supplementation of the diet with L-lysine has received considerable attention for prophylaxis and treatment of latently infected cats. It is postulated that viral replication can be competitively inhibited by L-lysine, which competes with arginine, an essential amino acid for viral replication.71,72 Studies suggest that restriction of arginine is required for this effect and that such restriction in cats is rapidly toxic.72–74 Supplementation with L-lysine prior to infection has been shown to be effective in limiting clinical disease, although not viral shedding. Clearly, however, pre-infection supplementation is not realistic in the clinical situation. 71
One study found no benefit of oral L-lysine supplementation in shelter cats with respect to the development of upper respiratory tract infections, compared with those not receiving L-lysine. 75 Another study of cats with upper respiratory tract infections demonstrated increased severity of clinical signs, increased FHV-1 DNA detection, increased plasma lysine and decreased plasma arginine levels in those cats receiving dietary supplementation of L-lysine. 76 A recent study also suggested that respiratory signs were more severe, and the rate of detection of FHV-1 DNA in oro-pharyngeal or conjunctival mucosal swab specimens was greater, in a population of shelter cats receiving L-lysine oral supplementation compared with a similar population not receiving supplementation. 77
Dietary supplementation with L-lysine has been studied in normal cats. While plasma lysine levels in treated cats increased, there was no significant change in plasma arginine levels in these cats. In addition, at the highest levels of dietary L-lysine supplementation (111 g/kg and 131 g/kg), there was a reduction in food consumption, although this did not reach a statistically significant level within the 2 week study period. The authors postulated that this reduction in food consumption was due to an amino acid toxicity. 78
Stromal keratitis
Viral replication may be absent or low grade in stromal keratitis (Fig 1) and, therefore, antiviral agents are unlikely to be effective as the sole therapeutic strategy. Anti-inflammatory therapy is usually required, and evidence suggests a similar mechanism may be involved in feline chronic rhinosinusitis.88,89

Stromal keratitis in an Exotic Shorthair at presentation
The judicious use of topical steroid preparations, in combination with antiviral medication, may be indicated in some cases (Fig 2). 90 Topical non-steroidal medications may be less deleterious to an underlying FHV-1 infection, and have been shown to be beneficial in a mouse model of herpes simplex virus (HSV-1) stromal keratitis. 91

Stromal keratitis after 2 months of treatment with topical dexamethasone, trifluorothymidine and oral prednisolone (1 mg/kg q48h)
FHV-1 vaccination
In general, FHV-1 vaccination protects against disease but not infection. 50 One study demonstrated that titres of serum neutralising antibodies develop after vaccination against FHV-1 in 70% of cats given either a subcutaneous or intranasal vaccination. 79 The authors concluded that more than one immunisation should be given to naive cats. However, cell-mediated immune responses also play a role in inducing protection against FHV-1 and are arguably more important than humoral responses. 80 Cell-mediated mucosal immunity is important in resistance to infection; however, mild clinical disease may be encountered more commonly following intranasal vaccination. In experimental studies using intranasal vaccines a more rapid protective immunity was reported; 80 however, intranasal vaccine virus is reportedly able to establish latency. 81 Modfied-live intranasal vaccines are no longer available in Europe. 50
In FHV challenge studies all cats (vaccination and control groups) developed signs characteristic of feline infectious rhinotracheitis (FHV-1). The cats in the vaccinated groups had less severe signs than those in the control groups.82,83 Vaccinated cats still shed FHV-1 after challenge with virulent virus, although in one study this was shown to be at a lower level than in control cats. 82
A further study of serum neutralising antibody protection gained after vaccination of feral cats at the time of neutering demonstrated that an inactivated virus vaccine gave greater protection than a modified-live vaccine. 84 This is in contrast to the generally accepted view that modified-live vaccines provide a greater duration of immunity than killed vaccines, 85 and earlier protection. 50 Modified-live vaccines retain some virulence and may induce clinical signs if administered incorrectly (eg, accidental aerosolisation). 50
Recent (European Advisory Board on Cat Diseases, ABCD) guidelines on the prevention and management of FHV-1 infection, recommend that kittens should have primary vaccinations at 9 and 12 weeks of age, with the first booster vaccination at 1 year of age. In the ABCD guidelines, annual boosters are recommended for all cats except those in low risk situations (eg, indoor only, less than five cats in household — 3 year intervals recommended). 50 This is in contrast to AAFP and WSAVA guidelines that a booster should be administered 1 year after primary vaccinations and then every 3 years. 86 Protection against clinical signs is not complete (approximately 90%), 87 and, in the face of a large viral challenge or immunosuppression, this may be further eroded. 50
Topical ciclosporin has been shown to be efficacious in reducing stromal infiltrate in human patients with stromal keratitis secondary to HSV-1. 92 The efficacy of ciclosporin in feline stromal keratitis has not yet been evaluated; 90 however, it has been shown to be beneficial and relatively well tolerated (2/35 cats developed blepharitis, one of which had discomfort on application of ciclosporin) in cases of feline eosinophilic keratoconjunctivitis. 93 It might be considered prudent to use ciclosporin in conjunction with antiviral therapy to ameliorate any recrudescent disease associated with the use of an immunomodulatory drug.
Uveitis with concurrent ulceration
Treatment of cats with both uveitis and ulceration is problematic as topical cortico-steroids are contraindicated. The use of oral prednisolone in combination with antiviral therapy can be successful (Fig 3), although vigilance for potential side effects of systemic steroids is required (eg, diabetes mellitus, pancreatitis). Systemic non-steroidal drugs can also be successful in managing some uveitis cases, and cautious use of topical non-steroidal drugs can be considered in cases of concurrent ulceration. Keratomalacia associated with the use of topical NSAIDs has been reported in humans and dogs, so frequent monitoring while receiving medication is recommended.94,95

Domestic shorthair cat with chronic uveitis controlled by oral prednisolone treatment (0.5 mg/kg q48h); topical steroid medication in this case caused recrudescence of epithelial keratitis (suspected sequela of FHV-1 infection)
Topical atropine can alleviate painful ciliary spasm associated with uveitis, and is an important tool in the provision of analgesia for these patients. Topical atropine has also been reported to stabilise the blood-aqueous barrier. 96 Again, monitoring of tear production both before and after treatment is warranted in the face of concurrent ulceration. Due to systemic absorption via mucosal surfaces (including via nasolacrimal drainage), both eyes may experience reduced tear production. 2
Feline uveitis associated with FHV-1 infection has been described. 97 The virus is best treated with systemic antiviral therapy since, although some topical antiviral medications can penetrate intraocularly, they may not achieve the therapeutic levels required for FHV-1. 98
Eosinophilic keratitis
Eosinophilic keratitis has been postulated as a sequela to FHV-1 infection. 99 Some authors therefore recommend antiviral medication in addition to the mainstay of corticosteroid therapy. Topical steroid is effective in the majority of cases (Fig 4) and can be gradually withdrawn as clinical signs abate. 100 The efficacy of steroid treatment somewhat contradicts the hypothesis of an FHV-1 aetiology.

(a) Eosinophilic keratitis in a domestic shorthair cat at presentation. (b) Appearance after 1 week of topical prednisolone treatment
Ciclosporin has also recently been reported to be efficacious in the treatment of eosinophilic keratitis (Fig 5). 93 Recalcitrant cases, or those where topical treatment is not possible, have been treated with oral corticosteroids with good effect. 100

(a) Severe stromal ulceration associated with eosinophilic keratitis in a domestic shorthair cat. (b) Appearance 11 days after ciclosporin and trifluorothymidine treatment
Chemical corneal injury
In cases of suspected chemical injury the nature of the chemical should be ascertained. Acids tend to denature protein on contact, limiting their penetration through tissue; alkaline agents tend to penetrate far more deeply, resulting in greater tissue injury (Fig 6). 101 Copious flushing of the cornea and conjunctival sacs with sterile lactated Ringer's (Hartmann's) solution is essential and should be continued for a minimum of 15 mins as soon after the injury as possible. Sterile saline and water are hypotonic to the cornea and may cause further damage. 101 However, the most important treatment of the chemically injured cornea is dilution of the chemical agent. If Hartmann's solution is not available for flushing, the use of sterile saline or water is still beneficial.

Right eye of a cat 2 months after alkaline (caustic soda) injury. Note the trichiasis (a), as upper eyelid cicatrisation has caused hair to be drawn onto the corneal surface, and (b) obliteration of the conjunctival fornix and eyelid margin
Keratomalacia is a recognised sequela of chemical injury and thus the provision of topical anticollagenases is highly recommend-ed. 101 If at all possible the cat should be referred at the earliest opportunity after initial flushing has been undertaken for more specialist care and treatment of potential sequelae.
Destruction of stem cell populations can result in conjunctivalisation of the corneal surface and may require stem cell transplantation from the fellow eye (if unaffected). 102 Amniotic membrane transplantation and corneal transplantation have been reported in human patients following chemical corneal trauma.102,103
Due to obliteration of lacrimal ductule openings in damaged conjunctiva, tear production may be severely compromised in these cases and artificial tears should be implemented. 104
Corneal damage may be sufficient to affect endothelial function and severe corneal oedema may result. 105 The formation of corneal bullae (vesicles of fluid within the corneal stroma; bullous keratopathy), and subsequent rupture of these bullae, can result in further ulceration. The use of hypertonic sodium chloride drops or ointment may help to alleviate corneal hydration sufficiently to ameliorate bulla formation but is rarely able to clear the corneal oedema.106,107 Thermal keratoplasty or penetrating keratoplasty (corneal transplant) may be required where medical treatment is not successful. 106
KEY POINTS
Topical antibiotics are sufficient in most cases of ulcerative keratitis.
Systemic antibiotics should be administered if corneal perforation is present.
Topical antiviral medications need to be given frequently as they are virostatic (not virucidal).
Trifluorothymidine is the most efficacious topical antiviral.
If topical therapy is likely to cause stress to the cat (risking progression of viral infection), systemic antiviral treatment may be more appropriate.
Famciclovir has been shown to be safe and efficacious as a systemic antiviral treatment in cats.
