Royal College of Ophthalmologists Guidelines (Focus)

Uveitis - When to investigate?

Uveitis, a term correctly used to describe inflammation of the uveal tract (iris, ciliary body, choroid) alone, in reality comprises a large group of diverse diseases affecting not only the uvea but also the retina, optic nerve and vitreous. Uveitis is a major cause of severe visual impairment and has been estimated to account for 10-15% of all cases of total blindness in the USA. In surveys of the causes of blindness uveitis has usually not been included and is probably underestimated.

Sight threatening complications include: cystoid macular oedema, secondary glaucoma, secondary cataract, vitreous opacities and retinal scars. The majority of patients are of working age and so this condition may cause potentially serious economic consequences with many days off work or job losses as well as interfere with education and prevent driving.

The International Uveitis Study Group classification separates uveitis by anatomical localisation of the disease, according to the major visible signs: anterior, posterior, pan and intermediate. The course of the disease can be described as acute, chronic (> 3 months duration) and recurrent. In the majority of cases of endogenous uveitis, the aetiology is unknown although some cases are a manifestation of a systemic disease, such as sarcoidosis or Behçet's disease, whilst others are associated with the HLA-B27 related group of diseases. Although one can attach a label to a number of uveitis syndromes, such as Fuchs' heterochromic cyclitis or Vogt-Koyanagi-Harada (VKH) disease their underlying cause is unknown. Despite this, classifying uveitis patients into different subgroups is important, even if an aetiology is never found, as evidence exists on how best to manage these distinct uveitis entities and their likely prognoses.

One of the most pressing questions that arises in the mind of every ophthalmologist who sees a new case of uveitis is "what is the cause of this disease?" In evaluating patients with uveitis, the ophthalmologist must consider that a lengthy list of infections, autoimmune systemic diseases, distinctive inflammatory conditions and masquerade syndromes may all cause uveal inflammation. Despite this array of potential diagnoses, the vast majority of patients have disease that defies categorisation.

In most uveitis patients routine serological and radiological investigations are usually unhelpful. There are no serological markers of disease activity as can be found in patients with systemic vasculitis. Also, any abnormalities found in peripheral blood are unlikely to reflect what is going on inside the eye. Nevertheless, because uveitis has a puzzling nature and it may form part of a systemic disease process, many patients are frequently over-investigated by being subjected to a battery of unnecessary tests.

The management of uveitis should be a systematic approach tailored to each patient's particular type of uveitis. It is essential that a detailed history is taken and direct questioning should include asking about back / joint problems, skin disease, respiratory disease, neurological disease, gastrointestinal disease, mouth and genital ulcers and sexually transmitted disease. When seeing a patient with chronic or recurrent disease it is important to pay attention to previous findings in the notes for any clues to diagnosis, such as an absence of posterior synechiae in a white and painless eye (Fuchs' heterochromic cyclitis), or unilateral uveitis often associated with raised intraocular pressure recurring within a few months of stopping topical steroids (herpes simplex or varicella zoster uveitis). A careful ocular examination should be performed as details such as conjunctival or iris nodules or iris atrophy may point to a specific diagnosis. It is often the history and examination findings that are far more informative than any laboratory investigations and may save the patient undergoing unnecessary investigations. Most patients do not have an underlying systemic disease.

Should one investigate?

Unfortunately, no large studies exist which demonstrate the value of investigating patients with uveitis. A number of specific uveitis entities can often be diagnosed solely on clinical examination; these include Fuchs' heterochromic cyclitis and recurrence of presumed congenital ocular toxoplasmosis, neither of which require further investigations.

Debate exists as to whether patients with the commonest type of uveitis (acute anterior uveitis - AAU) should be investigated. It is well recognised that approximately 50% of patients with AAU are HLA-B27 positive. A number of these patients will give a history of an associated HLA-B27 disease. In the others, HLA typing is unnecessary because the result will not help in the future management of the patient. Also, HLA-B27-associated AAU often presents with a number of clinical clues which help in diagnosis: it is usually recurrent, unilateral but alternating, with severe anterior chamber inflammation (posterior synechiae, fibrin and hypopyon).

It is important to understand the reasons for ordering any investigation:

• will it identify any underlying systemic disease process or association?
• will it provide a 'definitive' aetiology?
• will it confirm or reject a diagnosis?
• will it help in the management of the patient?

Interpretation of results is also very important, particularly with regards to false negatives and false positives. The latter are not uncommon with regards to syphilis serology and the Mantoux test.

A recent retrospective review of patients with various types of uveitis showed the following abnormal results: full blood count: 23/113 (20.3%), plasma viscosity / ESR: 37/108 (34.2%), VDRL/TPHA: 3/70 (4.3%), angiotensin converting enzyme (ACE): 9/77 (10.8%) and chest x-ray (CXR): 15/103 (14.6%). Sarcoidosis was diagnosed in eight patients who had an abnormal CXR ± raised ACE. None of the other abnormal results helped in establishing an underlying cause for the uveitis or assisted in the further management of the patients. All patients with symptoms of other organ system dysfunction or general malaise should be investigated to rule out under-lying systemic disease.

What to order

Ordering large numbers of tests in the hope that one may turn out to be positive should be actively discouraged. If one does enough investigations on any patient there is the chance that something will turn up abnormal but it may have no relevance to the uveitis.

Attention should be paid to the sensitivity and specificity of each test:

• sensitivity - measures how well the presence of a disease is predicted by a diagnostic test.
• specificity - measures how well the absence of a disease is predicted by a diagnostic test.

In the clinical setting, the minimum number of investigations should be performed that will give the maximum information regarding the management of the patient. There are a number of general tests that would be common to most uveitis patients, and specific tests that might be relevant to a particular type of uveitis.

General Investigations

full blood count (eosinophilia in parasitic infections, raised white count in bacterial infections, relative lymphocytosis in viral infections or tuberculosis)

• plasma viscosity/ESR (underlying systemic disease)
• syphilis serology
• urinalysis (diabetes mellitus)
• chest x-ray (CXR) (sarcoidosis and tuberculosis)

Any other tests that need to be ordered should depend on the clinical findings and the ophthalmologist's index of suspicion for a particular diagnosis.

Specific Investigations

• sacroiliac joint x-ray (HLA-B27 related disease)
• angiotensin converting enzyme (ACE) (sarcoidosis)
• toxoplasma dye test / IgG antibodies (if negative in undiluted serum to exclude congenital toxoplamosis)
• toxocara ELISA
• HLA-A29 typing (birdshot retinochoroidopathy)
• anti-neutrophil cytoplasmic antibody (Wegener's granulomatosis)
• Mantoux test (tuberculosis, may be negative in sarcoidosis)
• Kveim test (sarcoidosis)
• fundus fluorescein angiography (AMMPE, geographic choroidopathy)
• electrophysiology (birdshot retinochoroidopathy)
• CT scan of chest (sarcoidosis)
• CT scan of orbits / B-scan ultrasound (posterior scleritis)
• MRI head scan (demyelination, non-Hodgkins lymphoma,
• neurosarcoidosis)
• CSF studies (demyelination, non-Hodgkins lymphoma, VKH)
• conjunctival biopsy (sarcoidosis-'blind' biopsies should be discouraged)
• polymerase chain reaction of intraocular fluid (herpesviral DNA, propionibacter DNA)
• vitreous biopsy (non-Hodgkins lymphoma, amyloid)
• choroidal biopsy (non-Hodgkins lymphoma)

Routine "immunological" tests are often of little help. Out of a series of 893 uveitis patients screened for a variety of non-ocular specific autoantibodies, the only significant finding was in patients with juvenile chronic arthritis in whom 10/13 (77%) were antinuclear antibody positive.

In some patients with recurrent or chronic uveitis, repeating investigations 3-5 years later may increase the yield of positive results. Those patients in whom an underlying systemic disease is suspected should be referred to a Physician as they may require more detailed/invasive investigations.

Conclusion

Uveitis is a puzzling and potentially sight threatening disease. Do not expect to find a 'definitive' aetiology in the vast majority of patients. A detailed history and thorough clinical examination remains essential in establishing a diagnosis of underlying systemic disease in these patients. Baseline screening investigations should be avoided as they do not contribute to finding a cause or help in management, and are often expensive. Tests should be tailored to the clinical findings and ordered only if there is a strong suspicion of systemic disease.

Philip I. Murray PhD FRCS FRCOphth
Professor of Ophthalmology
University of Birmingham

References

Barile GR, Flynn TE. Syphilis exposure in patients with uveitis. Ophthalmology 1997;104:1605-9

Marr JE, Stavrou P, Moradi P, Murray PI. Should we investigate patients with uveitis? Invest Ophthalmol Vis Sci 1998;39:S608.

Murray P. Serum autoantibodies and uveitis. Br J Ophthalmol 1986;70:266-8.

Murray PI, Stavrou P, Marr JE, Moradi P. Pattern of visual loss in patients with uveitis. Invest Ophthalmol Vis Sci 1998;39:S607.

Rosenbaum JT, Wernick R. The utility of routine screening of patients with uveitis for systemic lupus erythematosus or tuberculosis. A Bayesian analysis. Arch Ophthalmol 1990;108:1291-3.

Rothova A, Suttorp-van Schulten MS, Frits Treffers W, Kijlstra A. Causes and frequency of blindness in patients with intraocular inflammatory disease. Br J Ophthalmol 1996;80:332-6.

Suttorp-van Schulten MSA, Rothova A. The possible impact of uveitis in blindness: a literature survey. Br J Ophthalmol 1996;80:844-8.