Investigation of feline uveitis is often viewed as complicated and unrewarding. Despite this, investigation of feline uveitis is essential due to the number of potentially life-threatening diseases proposed to be causally associated with feline uveitis (Figure 1) and the potential health implications for in-contact cats when infectious disease is identified.

Uveitis may be classified on anatomical location (e.g., anterior vs. posterior), duration (acute vs. chronic) or aetiopathogenesis (exogenous vs. endogenous). In the context of this discussion division of uveitis based on pathogenesis is useful in determining which diagnostic tests should be considered. Exogenous causes of uveitis, including trauma (blunt or penetrating) and corneal ulceration, are usually diagnosed on ophthalmic examination. Endogenous uveitis develops from within the eye or from the bloodstream. The known and proposed causes of feline uveitis are listed in Figure 1.

Figure 1. Feline uveitis: Known or proposed causes.

 Infectious

 Protozoal:

 Toxoplasma gondii

 Leishmania donovani

 Viral:

 Feline leukaemia virus (FeLV)

 Feline immunodeficiency virus (FIV)

 Feline infectious peritonitis virus (FIP)

 Feline herpesvirus 1 (FHV-1)

 Bacterial:

 Bartonella henselae

 Mycobacterium spp.

 Bacterial infections causing septicaemia, endotoxaemia, exotoxaemia or immune complex deposition (e.g., endocarditis, pyelonephritis and discospondylitis)

 Mycoses:

 Cryptococcus neoformans

 Histoplasma capsulatum

 Blastomyces dermatitides

 Coccidioides immitis

 Candida albicans

 Parasitic:

 Ophthalmomyiasis interna posterior (Diptera sp.)

 Ocular filariasis (Dirofilaria immitis)

 Ocular larval migrans (Toxocara sp.)

 Immune-mediated

 Poorly defined group likely to include many autoimmune diseases to a variety of endogenous antigens (e.g., retinal-S antigen) that either initiate or perpetuate uveitis

 Lens

 Cataracts (lens-induced uveitis)

 Lens trauma (phacoclastic uveitis)

 Neoplasia

 Primary (melanoma)

 Secondary neoplasia (lymphosarcoma most common)

 Non-ocular neoplasia (paraneoplastic syndromes)

 Metabolic

 Hyperlipidaemia

 Vascular

 Hypertension

 Toxaemia

 Anaemia

 Hyperviscosity syndromes

 Bleeding diathesis

The initial assessment of a feline uveitis patient includes a thorough history, and physical and ophthalmic examinations. The latter will identify most causes of exogenous uveitis. It is worth noting that occasionally penetrating injuries are not obvious, especially if the wound has occurred through the conjunctiva/sclera rather than the cornea or has involved the posterior aspects of the globe. These cases are unilateral and frequently present with severe panuveitis. In such cases the history may be suggestive, e.g., recent maxillary molar tooth extraction followed by severe unilateral uveitis, or recent history of a catfight or gunshot wounds in which adnexal injury may have been noted but no overt corneal penetration was apparent. In such cases the ocular media are frequently opaque, preventing examination of the posterior segment. Initial investigations are therefore usually directed towards imaging the ocular, orbital and adnexal structures; radiography and ultrasonography being the most commonly utilised in the primary setting.

The majority of feline uveitis cases will have no history or clinical findings suggestive of an exogenous cause, and consideration of the endogenous causes of uveitis (Figure 1) will be necessary. Excluding cases of uveitis associated with intraocular neoplasia, lens disease, lipid aqueous or the rare situation where an intraocular parasite is directly visualised (e.g., ophthalmomyiasis interna), there are no pathognomonic clinical ophthalmic changes associated with endogenous uveitis in the cat. Independent of aetiology, aqueous flare, iritis, keratic precipitates, hyphaema and hypopyon commonly develop in the anterior chamber, whilst posterior uveitis presents with signs of choroidal and retinal inflammation including exudates, haemorrhage and retinal detachment.

Feline diffuse iris melanoma is the most common primary neoplasm of cats and usually presents as a diffuse darkening of the entire iris although nodular forms may occur. Ophthalmic signs which increase the index of suspicion that an iridal pigmentary change represents a diffuse iris melanoma, rather than benign melanosis include: the presence of iridal thickening or mass lesions, dyscoria, anisocoria, concurrent uveitis or intraocular haemorrhage, and free pigment in the anterior chamber or pigment accumulation in the iridocorneal angle (± secondary glaucoma). In the early stages, diffuse iris melanoma differentiation from benign melanosis may be difficult, and close monitoring of the ophthalmic changes is advised (serial photographs are useful). Fine needle aspiration biopsy (FNAB) from the iris is usually unrewarding in distinguishing iris melanoma from melanosis but small 'snip' biopsies may be useful if the pigmentary change affects the pupillary zone of the iris.

Primary post-traumatic intraocular sarcoma is thought to originate from lens epithelial cells following lens trauma. The intraocular mass lesion may be visible ophthalmoscopically, or ultra-sonographically in cases where uveitis or intraocular haemorrhage obscures examination. These tumours may invade locally (e.g., optic nerve/ chiasma) or metastasise to distant sites. Therefore preoperative imaging to determine any local and/or systemic involvement of the tumour is indicated if there is an index of suspicion for this disease (e.g., elderly cat with chronic unilateral uveitis, evidence of an intraocular mass and a history or clinical evidence of previous ocular trauma (e.g., corneal scar, synechia)).

Ocular involvement in feline lymphoma is common and often precedes systemic manifestations. These cases may present with pale intraocular mass lesions or, in the earlier stages, more subtle nodular iris lesions may be evident. Cytology on aqueocentesis samples or FNAB specimens from an anterior uveal mass may be rewarding in these cases although the diagnosis is usually achieved by sampling of other affected organs or lymph nodes.

The presence of large capillary beds within the uveal tract makes it a potential site for haematogenous metastasis from any distant neoplasia. One particular neoplasm which appears to have a characteristic ophthalmic presentation is bronchogenic carcinoma. An ischaemic chorioretinopathy occurs due to invasion of the choroidal vasculature leading to wedge-shaped areas of tan discoloration within the tapetal fundus and serous subretinal exudation. These cats present with unilateral or bilateral blindness and may also have painful ischaemic necrosis of the distal extremities. Diagnosis in such cases relies on biopsy of the primary or non-ocular metastatic lesions.

The discussion so far has considered causes of feline uveitis in which the clinical presentation either identifies, or is highly suggestive of, a particular causation. In other cases further investigation of causes of endogenous uveitis including infectious, immune-mediated and vascular disease, is indicated. The minimum database obtained should be a complete blood count, serum biochemistry profile, urinalysis, thoracic radiography and abdominal ultrasonography. Selection of additional diagnostic tests may be affected by other clinical findings (e.g., coagulation studies and blood pressure assessment if intraocular haemorrhage is present) and knowledge of the diseases endemic in a given region (e.g., blastomycosis, histoplasmosis and leishmaniasis not reported in the UK but could occur in a cat which has travelled from endemic areas).

In the UK, testing for Toxoplasma gondii, feline leukaemia virus (FeLV), feline immunodeficiency virus (FIV), feline infectious peritonitis virus (FIP) and Bartonella henselae are recommended. Aqueocentesis and vitreocentesis may be considered in cases in which granuloma formation is evident (e.g., mycotic or mycobacterial infection). Vitreocentesis is more likely to yield a diagnosis even if there is anterior chamber involvement. Cryptococcus neoformans is the most commonly reported feline mycotic infection. Frequent sites of infection are the nasal passages, skin and central nervous system; granulomatous uveitis may also be a feature. The diagnosis of cryptococcosis can be established on the basis of cytological identification of the organism aspirated from affected tissue (non-ocular in most cases), and detection of cryptococcal antigen in serum, aqueous or vitreous using latex agglutination kits.

When performing serological tests for infectious diseases it is important to remember that they indicate exposure to an infectious agent (as may occur in healthy cats) and therefore do not definitively establish a direct causal link between the infectious agent and uveitis in an individual cat. Furthermore no test has a 100% sensitivity and specificity, and the reliability of a test (i.e., its predictive value) decreases as the prevalence of the disease in the population decreases.

Uveitis has been induced in cats following inoculation with T. gondii. Serological testing of cats for T. gondii should include both T. gondii-specific IgM and IgG. Evidence of recent or reactivated infection includes high T. gondii-specific IgM titres or increasing (four-fold or greater) T. gondii-specific IgG titres in the serum. However, since seropositivity to T. gondii-specific IgM and IgG can occur in clinically normal cats they may not correlate to clinical ocular disease in an individual case. Calculation of indices which indicate evidence of intraocular T. gondii-specific immunoglobulin production (e.g., Goldmann-Witmer coefficient (C-value)) have been evaluated in an attempt to improve the accuracy of testing. Unfortunately studies have shown that ocular antibody production for T. gondii-specific immunoglobulins increased in some experimentally inoculated healthy cats and chronically infected cats by non-specific immune stimulation. Therefore the positive predictive value of testing for T. gondii involvement in ocular disease may not be improved. Testing for T. gondii antigens or DNA in the aqueous humour may potentially assist in the diagnosis of T. gondii-induced uveitis.

Initial testing for FeLV is usually performed using an 'in-house' ELISA (enzyme-linked immuno-sorbent assay) or ICGA (immunochromatographic assay) to detect soluble protein (FeLV--p27) in the plasma or serum. This indicates FeLV viral infection but does not distinguish transient or permanent viraemia. A confirmatory test such as a different type of ELISA or ICGA testing kit or an IFA (immunofluorescent antibody assay) is necessary to confirm the positive result. A positive IFA detects FeLV p27 antigen within white blood cells (WBCs) and is likely to signify permanent FeLV infection. Testing should be repeated at 12 weeks to confirm a positive viraemia. A direct causal link between FeLV and uveitis has not been established. Therefore even in a persistently viraemic patient, direct causation between FeLV and feline uveitis cannot be assumed. Ocular involvement in feline lymphoma is common and probably represents metastasis of neoplastic cells to the uveal tract (haematogenously).

As with FeLV, initial testing for FIV is usually performed using an 'in-house' ELISA or ICGA to detect antibodies to FIV in the serum. Positive results need to be confirmed using an alternative testing methodology such as IFA or western blot (which also detect antibodies to FIV) or polymerase chain reaction (PCR). In one study all FIV-seropositive cats with uveitis were also seropositive for T. gondii emphasising the need to screen for multiple infectious agents.

Diagnosis of FIP-induced feline uveitis is hindered because serology does not distinguish between enteric and FIP-inducing strains of feline coronavirus. It is therefore important to interpret the serological findings in conjunction with other parameters to increase the accuracy of testing. In one report a combination of a coronavirus antibody titre>1:160,withconcurrentlymphopenia and hypergammaglobulinaemia had a positive predictive value of 88.9% in cases with suspected systemic FIP.

The cat is the major persistent reservoir host for numerous Bartonella spp. including B. henselae and B. clarridgeiae. There is increasing evidence that these agents may be involved with feline uveitis, but interpretation of serological findings is difficult due to the frequency of asymptomatic carriers (in one study 41% of cats were persistently bacteraemic but showed no clinical signs). In the future testing for Bartonella spp. antigens or DNA in the aqueous humour may potentially assist in the diagnosis of feline bartonellosis.

Immune-mediated (including autoimmune) uveitides, although probably common in cats (as in other species), are poorly defined and are likely to represent a significant percentage of cases which are classified as idiopathic. The diagnosis is largely one of exclusion. In human medicine there are strong associations between an increasing numbers of uveitides with specific MHC haplotypes; responses to specific endogenous ocular antigens are now recognised to either initiate or perpetuate uveitis. An example of this is cancer-associated retinopathy (CAR) in people, which is believed to be driven at least in part by antibodies against the retinal protein recoverin. Feline uveitis may occur in association with non-ocular tumours and some of these may represent immune-mediated paraneoplastic uveitides akin to the human CAR syndrome.

References

1.  Lappin MR. Feline infectious uveitis. Journal of Feline Medicine and Surgery 2000; 2, 159-163.

2.  Stiles J, Townsend WM. Feline ophthalmology. In: Gelatt, KN. ed. Veterinary ophthalmology (fourth edition). Iowa: Blackwell Publishing, 2007; 1095-1164.