Uveitis in Juvenile Idiopathic Arthritis

Inflammatory eye diseases comprise some of the most devastating complications of childhood rheumatic diseases, especially juvenile idiopathic arthritis (JIA). Chronic (initially asymptomatic) uveitis accompanying JIA is one of the most common causes of uveitis in childhood. It is predominantly anterior, nongranulomatous inflammation affecting the iris and ciliary body (iridocyclitis) of insidious onset ( Fig. 22-1 ). Acute (symptomatic) anterior uveitis is characteristic of human leukocyte antigen-B27 (HLA-B27)–associated diseases such as enthesitis-related arthritis. The posterior uveal tract—the choroid—is rarely affected in rheumatic diseases of childhood.

Classification of Uveitis

It has been customary to classify anterior uveitis as acute or chronic , terms that are used imprecisely, and often synonymously, with symptomatic and asymptomatic, respectively. The Standardization of Uveitis Nomenclature (SUN) Working Group has proposed a framework for the classification of uveitis, a standardized grading system, and definitions of other terminology used to describe uveitis ( Tables 22-1 and 22-2 ). The SUN working group suggests the use of the terms insidious or sudden to describe the onset of uveitis, and limited (if the duration is 3 months or less) or persistent (if the duration exceeds 3 months) to describe its course (which may be influenced by treatment). It recommends that the term acute be applied to those instances in which the onset was sudden and the duration limited (as seen in HLA-B27–associated acute anterior uveitis), and that the term chronic be used to describe persistent disease with prompt (within 3 months) relapses after discontinuation of therapy ( Table 22-2 ). Most uveitis in children with oligoarthritis, polyarthritis, or psoriatic JIA has an insidious onset, with a chronic and frequently recurrent course. Intermediate uveitis with inflammation predominantly in the vitreous humor is occasionally described in JIA.

TABLE 22-1

The SUN Working Group Classification of Uveitis

TYPE	PRIMARY SITE OF INFLAMMATION	INCLUDES
Anterior uveitis	Anterior chamber	Iritis
		Iridocyclitis
		Anterior cyclitis
Intermediate uveitis	Vitreous	Pars planitis
		Posterior cyclitis
		Hyalitis
Posterior uveitis	Retina or choroid	Focal, multifocal, or diffuse choroiditis
		Chorioretinitis
		Retinochoroiditis
		Retinitis
		Neuroretinitis

SUN , Standardization of Uveitis Nomenclature.

TABLE 22-2

The SUN Working Group Descriptors of Uveitis

CATEGORY	DESCRIPTOR	COMMENT
Onset	Sudden
	Insidious
Duration	Limited	<3 months’ duration
	Persistent	>3 months’ duration
Course	Acute	Episode of sudden onset and limited duration
	Recurrent	Repeated episodes separated by periods of inactivity without treatment >3 months in duration
	Chronic	Persistent uveitis with relapses in <3 months after discontinuing treatment

SUN , Standardization of Uveitis Nomenclature.

Uveitis in Juvenile Idiopathic Arthritis

The uveitis associated with JIA accounts for 20% to 40% of cases of uveitis in some large series that characterize uveitis in childhood. Interpretation of data on this subject is hindered by the use of different classifications to describe chronic arthritis in children (see Chapter 15 ), different types of uveitis, different definitions of response and remission of uveitis, and different types of study populations. For these reasons, care must be taken in interpreting and generalizing conclusions from published information.

History: The Association of Arthritis and Uveitis

Ohm first described chronic uveitis and band keratopathy in 1910 in a child with arthritis. The association of ocular disease and juvenile arthritis was confirmed by several authors. In Sury’s large series of children with chronic arthritis, chronic uveitis was found in 15% of the total, and two thirds of those patients with uveitis had band keratopathy. The majority of his patients had an insidious onset of uveitis with little or no early disturbance of vision; diagnosis was often delayed until slit-lamp examination was performed. The occurrence of “chronic, asymptomatic, nongranulomatous anterior uveitis” became recognized as an important complication of what was called juvenile rheumatoid arthritis, particularly the limited joint disease referred to as pauciarthritis, equivalent to oligoarthritis, a subset of juvenile idiopathic arthritis in the International League of Associations for Rheumatology (ILAR) classification, as well as rheumatoid factor (RF)-negative polyarthritis and psoriatic arthritis.

Epidemiology

The reported frequency of chronic uveitis in children with chronic arthritis has varied considerably, from 2% in Costa Rica to 11.6 % in the United States, 13% in Canada, and 16% in the Nordic countries. It appears to be particularly uncommon in Asian and African populations but has been reported worldwide. A recent population-based study of 2636 children with JIA in Taiwan reported uveitis in 4.6%. There is some evidence that, although the prevalence of JIA may be increasing, there may be a decreasing frequency of uveitis in children with JIA.

The frequency of chronic uveitis is related to the subtype of JIA: it occurs in 15% to 20% of children with oligoarthritis (in up to 30% of those with extended oligoarthritis), 10% of those with psoriatic arthritis, and 14% of those with polyarthritis (RF negative). It is extremely uncommon in children with systemic JIA or RF-positive polyarthritis.

Early studies have consistently shown that the frequency of uveitis in JIA is highest in young girls who have oligoarthritis with an early age at onset (younger than 7 years of age) and who are antinuclear antibody (ANA) positive ( Table 22-3 ). However, in a recent study of 1047 patients with JIA, the age-associated risk of uveitis was seen only in young girls but not in young boys. A large study from Germany failed to demonstrate gender as an independent risk factor for the development of uveitis in JIA. In a recent study of 56 Greek children with JIA, the high frequency of ANA positivity in children with JIA, with or without uveitis, limited the value of ANA positivity as an indication of risk for uveitis. Reports from India document a low frequency of ANA positivity in children with JIA, including those with uveitis.

TABLE 22-3

Characteristics of Children with Chronic Uveitis and Arthritis

CHARACTERISTIC	OVERALL AVERAGE
Female-to-male ratio	4.4 : 1
Mean age at onset of arthritis (years)	4
Arthritis subtype (% with uveitis)
Oligoarthritis	30
Polyarthritis (RF negative)	15
Psoriatic arthritis	10
Polyarthritis (RF positive)	<1
Systemic arthritis	<1
Enthesitis-related arthritis	<1
Serology (%)
RF positive	<1
Antinuclear antibody positive	80

Acute uveitis ~7%.

Data summarized from published studies.

Etiology and Pathogenesis

The pathogenesis of chronic uveitis and the basis of its association with JIA are not known, although it is evident that T lymphocytes and their products are vital participants in the process, and that the innate immune system plays a role. Involvement of the immune response and perturbation of inflammatory cytokines are documented in animal and human studies, although there is limited direct evidence from studies of uveitis in JIA. Ooi and colleagues have reviewed the evidence that interleukin (IL)-1β, IL-2, IL-6, interferon-γ, and tumor necrosis factor (TNF)-α are present in ocular fluids and tissues of adults with ankylosing spondylitis, and that higher levels of these cytokines are associated with more severe uveitis. Levels of TNF-α were especially elevated in the aqueous of patients with ankylosing spondylitis–associated uveitis. The clinical responsiveness of uveitis to the administration of some, but not all, anti–TNF-α agents provides strong support for the premise that TNF-α is at least in part responsible for the inflammation. Serum levels of IL-6 and IL-8 are higher when uveitis is active in adults with posterior uveitis, anterior uveitis, or panuveitis. IL-6 influences the maturation of Th17 CD4 ⁺T cells, which participate in autoimmune diseases. In a murine model, primarily of posterior uveitis or retinitis, Yoshimura and colleagues demonstrated that absence of both IL-6 and IL-23 prevented the production of Th17 cells and reduced the ocular inflammation. Keino and colleagues demonstrated that the oral administration of an inhibitor of IL-12/IL-23 prevented autoimmune uveoretinitis in animals by reducing production of IL-17–producing cells. Other animal studies have shown that anti-Th17 can block the development of uveitis in uveitis-susceptible mice, but that administration of the cytokine itself into uveitis-susceptible rats has a mitigating effect on disease development. Anti-IL-17 treatment of human uveitis (principally posterior disease) has not demonstrated any significant effect, however. Despite the implication of IL-17 in uveitis, some models involve either Th1 or Th2 cells exclusively. Animal models may not be entirely relevant to understanding human disease, however. In addition, in the most widely studied T cell–dependent mouse model of uveitis, experimental autoimmune uveitis (EAU) is predominantly a chorioretinitis whose pathogenesis might differ significantly from anterior uveitis. Sijssens and colleagues have demonstrated age-related differences in aqueous humor cytokines in patients, including some with JIA.

Parikh and colleagues have suggested, on the basis of the characteristics of the cellular infiltrate seen in histological studies of enucleated eyes, that the pathogenesis probably involved B lymphocytes as well. In addition to the high frequency of ANAs of undetermined specificity, autoantibodies to ocular antigens have been described in children with arthritis and uveitis. There have been unconfirmed reports that children with uveitis and JIA have a higher frequency of immunity to soluble retinal antigen (S antigen) than do children with arthritis alone. It is not clear whether immunity to these ocular antigens is pathogenic or merely reflects the inflammatory process.

The basis of the association between inflammatory joint disease and inflammatory ocular disease is unexplained, but it is likely that a shared genetic predisposition plays a part.

Genetic Background

Genetic susceptibility to uveitis in JIA is complex. Although there is limited evidence for the familial occurrence of this disorder, a few case reports have documented the occurrence of oligoarthritis and chronic anterior uveitis in siblings.

A number of sometimes contradictory studies have reported an array of HLA alleles associated with chronic anterior uveitis in JIA. The strongest and most consistent associations appear to be with genes in the class II region. Studies from the United States have documented a strong association with DRB1*1104 (formerly called DR5), and associations with DQA1*0501 and DQB1*0301, which are in linkage disequilibrium with DRB1*1104. The association of uveitis in JIA with DRB1*11 noted in American children was confirmed in one Italian study but not in another. An association with DRB1*13 has been reported in British and Greek children. An increased prevalence of HLA DR9 (DRB1*09) (odds ratio [OR] 2.33) was also noted in one study.

Acute anterior uveitis is associated with the class I gene HLA-B27. Other associations between class I genes and chronic uveitis are less convincing. Single studies have alleged associations with HLA-A19 (OR 2.87), HLA-B22 (OR 4.52), and A2*06, but these studies have not yet been confirmed.

Genetic polymorphisms of non-HLA genes have been less well studied. TNF gene polymorphisms (−238 GA and −308 GA) have been linked with HLA B*27–associated uveitis. In patients with acute anterior uveitis, an association was found with another single nucleotide polymorphism (SNP) (857T) in the TNF gene. The TNF locus is located in the HLA/major histocompatibility complex (MHC), and large studies are needed to determine whether TNF associations reflect linkage disequilibrium with class I or class II loci.

Clinical Manifestations

Insidious Onset, Chronic Uveitis

Insidious onset, chronic uveitis is characteristic of oligoarticular JIA (persistent or extended), and most children with psoriatic arthritis or RF-negative polyarthritis develop uveitis (see Table 22-3 ). The onset of chronic uveitis is usually insidious and often entirely asymptomatic, although up to half of the children have some symptoms attributable to the uveitis (pain, redness, headache, photophobia, change in vision) later in the course of their disease ( Table 22-4 ). Uveitis is detected in less than 10% of patients before the onset of arthritis, usually in the course of a routine ophthalmic examination. In almost half of all patients with uveitis, it occurs just before arthritis is diagnosed, at the time of diagnosis, or shortly thereafter. A Finnish study of children with JIA and uveitis documented the appearance of uveitis within the first 3 months after onset of arthritis in 49%; 90% developed uveitis within the first 4 years after onset of arthritis. The risk is never entirely absent, however ( Fig. 22-2 ). The disease is bilateral in 70% to 80% of children. Patients with unilateral disease are unlikely to develop bilateral involvement after the first year of disease; however, there are exceptions, and unilateral uveitis may persist for many years in a few children before the other eye is involved.

TABLE 22-4

Ocular Signs and Symptoms in Children with Chronic Uveitis and Arthritis

CHARACTERISTIC	PERCENT AFFECTED (RANGE)
Bilateral uveitis	25-89
Symptoms
Ocular pain and or redness	0-25
Change in vision	0-30
Photophobia	0-8
Headache	0-6
None	51-97

Data summarized from published studies.

The early detection of chronic uveitis requires slit-lamp biomicroscopy, which should be performed at the time of diagnosis in every child with JIA and repeated at prescribed intervals during the first few years of the disease. The recommended frequency of ophthalmological examinations is influenced by the level of risk of uveitis ( Table 22-5 ). It is recommended that slit-lamp examinations be performed every 3 months for the first 2 years in children in the high-risk group (early age at onset, oligoarthritis or polyarthritis, ANA positivity) and every 4 to 6 months thereafter for a period of 7 years at a minimum. In children with ANA-negative disease, slit-lamp examinations should be done initially at 4- to 6-month intervals. Children with psoriatic arthritis are also at considerable risk for the development of uveitis and should be followed at the same frequencies as children with oligoarticular or polyarticular JIA. In children with systemic onset JIA, examinations once a year are probably sufficient. Subsequent to the publication of these guidelines, there have been suggestions for their modification. Heiligenhaus et al. proposed guidelines that conformed to the ILAR classification of JIA ( Table 22-6 ). Any child who has had uveitis should be considered to be at high risk, even if it has remitted, and continued surveillance is essential.

TABLE 22-5

American Academy of Pediatrics Guidelines for the Ophthalmological Screening of Children with JIA

JIA ONSET TYPE	ANA	SCREENING SCHEDULE
JIA ONSET TYPE	ANA	ONSET <7 YEARS *	ONSET >7 YEARS ^†
Oligoarthritis	Positive	Every 3-4 months ^‡	Every 4-6 months
Oligoarthritis	Negative	Every 4-6 months	Every 4-6 months
Polyarthritis	Positive	Every 3-4 months ^‡	Every 4-6 months
Polyarthritis	Negative	Every 4-6 months	Every 4-6 months
Systemic	Negative or positive	Every 12 months	Every 12 months
High risk: Screen every 3 months Moderate risk: Screen every 4-6 months Low risk: Screen every 12 months

Adapted from [No authors listed], American Academy of Pediatrics Section on Rheumatology and Section on Ophthalmology: Guidelines for ophthalmologic examinations in children with juvenile rheumatoid arthritis. Pediatrics 93 (1993) 295–296.

* All patients are considered to be at low risk 7 years after onset of arthritis; they should have yearly ophthalmological examinations indefinitely.

† All patients are considered to be at low risk 4 years after onset of arthritis; they should have yearly ophthalmological examinations indefinitely.

‡ All high-risk patients are considered to be at medium risk 4 years after onset of arthritis.

TABLE 22-6

Recommendations for Screening Based on the ILAR Classification of JIA

JIA SUBGROUP	ANA	AGE AT ONSET OF JIA	DURATION OF JIA	SCREENING
Oligoarthritis	+	<7 years	<5 years	3 months
RF-negative polyarthritis
Psoriatic arthritis
Undifferentiated arthritis
	+	<7 years	>4 years	6 months
	+	<7 years	>7 years	12 months
	+	>6 years	<3 years	6 months
	+	>6 years	>2 years	12 months
	—	<7 years	<5 years	6 months
	—	<7 years	>4 years	12 months
	—	>6 years	NA	12 months
Enthesitis-related arthritis		NA	NA	12 months
RF-positive polyarthritis		NA	NA	12 months
Systemic arthritis		NA	NA	12 months
Patients in any category with uveitis		NA	NA	According to uveitis course

Adapted from Heiligenhaus et al.

The diagnostic signs of anterior uveitis on slit-lamp examination are the presence of inflammatory cells and increased protein concentration (“flare”) in the aqueous humor of the anterior chamber of the eye ( Fig. 22-3 ). Deposition of inflammatory cells on the inner surface of the cornea (keratic precipitates) may be detected at presentation or develop later.

Complications of chronic anterior uveitis are frequent and increase with increasing duration of active disease. In recent series, the frequency of complications is somewhat lower than in earlier reports, presumably due to earlier treatment. Posterior synechiae, inflammatory adhesions between the iris and anterior surface of the lens, result in an irregular or poorly reactive pupil ( Table 22-7 ; Fig. 22-4 ). This abnormality may be the first obvious clue to the presence of uveitis on ophthalmoscopic examination, but it is often a sign of disease of considerable duration or severity. Synechiae that are circumferential prevent the free flow of aqueous humor between the posterior and anterior chambers, resulting in bulging of the iris (iris bombé) and increased intraocular pressure.

TABLE 22-7

Frequency of Complications of Chronic Uveitis in Reported Cases