Article Text

Steroid responsive disc neovascularisation in uveitis associated with juvenile chronic arthritis
  1. Moorfields Eye Hospital, City Road, London
  2. Prince Charles Eye Unit, King Edward VII Hospital, Windsor
  3. Moorfields Eye Hospital, City Road, London EC1V 2PD
  1. Moorfields Eye Hospital, City Road, London
  2. Prince Charles Eye Unit, King Edward VII Hospital, Windsor
  3. Moorfields Eye Hospital, City Road, London EC1V 2PD
  1. Moorfields Eye Hospital, City Road, London
  2. Prince Charles Eye Unit, King Edward VII Hospital, Windsor
  3. Moorfields Eye Hospital, City Road, London EC1V 2PD
  1. Professor Susan Lightman.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Editor,—Childhood uveitis is most frequently associated with juvenile chronic arthritis (JCA).1 The uveitis associated with JCA is anterior with no posterior involvement, although cystoid macular oedema may occur after cataract surgery. We describe disc neovascularisation (NVD) in two children with JCA associated chronic anterior uveitis. The possible aetiology of this and subsequent management are presented.


Case 1

Left NVD was diagnosed in a white 13 year old girl with a 3 year history of bilateral anterior uveitis associated with pauciarticular onset JCA. There were no clinical or laboratory features to suggest any other disease process and she was completely asymptomatic apart from the involved joint. At presentation and throughout follow up she had a mild right anterior uveitis and moderate left anterior uveitis despite the constant use of topical steroids. No other form of immunomodulation had been used. A unilateral secondary cataract reduced left visual acuity to 6/12. There was no diabetes, pars planitis, vitritis or retinal vasculitis, or any other ocular pathology which may be associated with neovascularisation. The presence of unilateral left NVD without retinal ischaemia or retinal vascular leakage was confirmed angiographically (Fig 1).

Figure 1

Left fluorescein angiogram demonstrating leakage from disc new vessels before systemic steroid treatment.

Increased activity of her left iridocyclitis, and absence of ischaemia prompted the use of systemic steroids to control the neovascularisation. Oral prednisolone (1 mg/kg/day starting dose, reducing by 10 mg each week) was successful in causing significant regression of the NVD after 2 weeks of treatment (Fig 2). Attempted withdrawal resulted in increased NVD.

Figure 2

Fluorescein angiogram demonstrating less fluorescein leakage after 2 weeks of oral steroids. This photograph is of a later fluorescein circulation time than Figure 1 (by 9 seconds). Note clearer view resulting from reduced inflammatory media opacities.

Case 2

Bilateral burnt out NVD was observed in a 4 year old Kuwaiti boy. At presentation the patient had severe bilateral anterior uveitis, secondary cataracts, and ocular hypotony. Subsequently, he developed pauciarticular JCA and there was nothing to suggest any other systemic or ocular disease. When the cataracts were removed, bilateral NVD with gliosis was found. No other cause for neovascularisation was detected; in particular, there was no evidence of pars planitis, retinal vascultiis, or ischaemia. Before cataract extraction, no form of topical or systemic immunosuppression had been used.


This is the first report of posterior segment neovascularisation in patients with juvenile chronic iridocyclitis and JCA. The uveitis associated with JCA is entirely anterior and not associated with posterior segment disease.2 Chevalley et alhave described “complete regression” of both “subretinal and preretinal neovascularisation” in other types of uveitis after treatment with systemic steroids.3

It is presumed (but not proved) that the NVD found in these two cases of JCA associated chronic anterior uveitis have a causal relation. Certainly in case 1, the anterior segment inflammatory activity had been very difficult to control. In the absence of posterior segment ischaemia or inflammation, it would appear that angiogenic factors released from the anterior segment led to disc neovascularisation.

Angiogenesis (neovascularisation) occurs under both physiological (for example, wound healing, placental maturation) and pathological (for example, tumour growth, rubeosis) conditions. Regardless of aetiology, new vessel formation occurs in identical stages in response to various angiogenic factors. These angiogenic factors form an ever increasing group of compounds.4

Present experimental evidence of neovascularisation associated with uveitis would appear to support the angiogenic role of prosta glandins.5-9 Primate immunological ocular inflammation produced by serum albumin5 produces a clinical picture similar to that of proliferative diabetic retinopathy. Anti-inflammatory drugs that inhibit the effects of prostaglandins (for example, indomethacin6 and methylprednisolone7) and irradiation (which induces leucopenia8), all reduce experimental neovascularisation in this primate model. Methylprednisolone is also well known for its potent antiangiogenic activity. This activity is common to a group of steroids collectively known as angiostatic steroids, including several glucocorticoid and mineralocorticoid steroids. Neither the glucocorticoid nor mineralocorticoid function is necessary for antiangiogenic activity.9

The regression of NVD after exposure to oral prednisolone in case 1 may therefore be a clinical example of the angiostatic role of steroids in uveitic neovascularisation.