The "disease" called juvenile rheumatoid arthritis (JRA) in the United States today was first well described in the English language by George Frederick Still in 1 89697.' He described in detail 19 children with chronic arthritis who had come to his attention during his training in pediatrics and pathology at the Hospital for Sick Children, Great Ormond Street, London. Dr. Still noted the diversity of disease manifestations in these children, and proposed that chronic childhood arthritis represented more than one disease. Since that time, there has been considerable confusion and controversy over "JRA"; its definition, its nomenclature, and its relationship to other forms of arthritis. Most observers agree that "JRA" is characterized by the presence of chronic synovitis andlthe absence of other identifiable diseases known to be associated with synovitis or joint complaints in children2,3 (Table 1). What is called JRA in the United States is referred to by a number of different names in other countries, including juvenile chronic polyarthritis, juvenile chronic arthritis, juvenile arthritis, and Still's disease.
THE ARTHRITIS OF JRA
The term "arthritis" denotes inflammation in joints (Greek: arthron - joint; itis - inflammation), specifically inflammation of synovial tissues. Synovial inflammation is signaled by joint swelling, pain, loss of motion, warmth or erythema; of these, joint swelling is the most specific indicator of arthritis. Joint pain alone is properly designated arthralgia rather than arthritis. Pain and dysfunction at musculoskeletal sights other than joints, such as bones, muscles, ligaments and nerves, should not be mistaken for arthritis. In evaluation of musculoskeletal dysfunction in a child, care must be taken to pinpoint the sight of trouble, lest nonarthritic conditions be confused with arthritis.'
True objective arthritis must be present - and for at least six consecutive weeks - to permit a diagnosis of JRA. Rheumatoid joints in children are almost invariably swollen because of joint effusion or synovial hypertrophy; stiffness after inactivity, loss of joint motion, and mild to moderate degrees of pain and motion, tenderness and warmth are also usually present. The arthritis of JRA is characteristically chronic and not transient; as noted previously, to permit diagnosis, arthritis must have been present for six consecutive weeks in the same sites. In systemic onset JRA, classic systemic disease with rash, fever and other systemic manifestations may be associated with transient arthralgia, myalgia and synovitis during early disease; the question of permitting a firm diagnosis of systemic Onset J RA in the absence of chronic synovitis remains moot.
THE SUBGROUPS OF JRA
Most observers now agree that, among children with chronic synovial inflammation, several distinct subgroups of disease exist that can be separated by both clinical and laboratory findings. Some investigators recognize three groups, others recognize more. This discussion will describe five subgroups of JRA (Table 2).
CONDITIONS CAUSING OR SIMULATING CHILDHOOD ARTHRITIS
SYSTEMIC ONSET DISEASE
Systemic onset JRA occurs in about 20 percent of all JRA patients.2'3 Boys and girls are equally affected, or there may be a male preponderance. The disease may begin at any age during childhood, and there is no significant correlation with family history of associated rheumatic disease (Table 2).
High intermittent fevers6'7 with elevations to 1030F or higher for several consecutive weeks are the hallmark of systemic onset JRA. Temperature elevations occur once or twice daily, most often in the evening hours; they are followed by a rapid return to normal or even subnormal levels. Patients frequently have shaking chills at the time of temperature elevation. Most patients with systemic onset JRA also have a characteristic rheumatoid rash.8'9 This rash may occur anywhere on the body, is extremely evanescent, and appears most frequently during febrile periods. Individual· lesions are small (about 10mm in diameter), pale, red macules often with a central area of clearing. Occasionally, the rash is pruritic. Other extra-articular manifestations are also frequent. Lymphadenopathy and hepatosplenomegaly may be very marked; lymph node histology may resemble that of lymphoma.2'3 Mildly abnormal liver function tests may occur; liver histology may show nonspecific collections of inflammatory cells in the portal areas. I0 It is important to differentiate the primary liver involvement of the disease from the hepatic effects of salicylates." Pleuritis2 and pericarditis12'11 occur in about half of patients; although generally mild, severe pericarditis or even myocarditis may occur. A few patients have severe abdominal pain, probably related either to mesenteric adenopathy or peritonitis. Leukocytosis is usually present; white blood counts may exceed 50,000/mm3, with a general predominance of early white cell forms. Profound anemia may also occur. Iridocyclitis is conspicuous by its absence.2'3 Systemic manifestations generally continue for several consecutive months, but rarely exceed six consecutive months' duration. More than half of the children with systemic onset JRA have one or more subsequent episodes of systemic disease months to years later; however, febrile periods rarely recur in adulthood (Table 2)..
Musculoskeletal complaints of systemic onset JRA are also a prominent part of the disease, although they may be overlooked at the onset in the face of the severe systemic illness. Most affected children have severe myalgia, arthralgia, or transient arthritis during febrile periods. Such transient musculoskeletal complaints may resolve dramatically during afebrile periods. Nearly all affected, children develop persistent arthritis of multiple joints during the first few months of disease. The pattern of joint involvement ultimately resembles that of polyarticular disease as described below. About a quarter of children with systemic onset disease have chronic polyarthritis, which exceeds the period of systemic disease manifestations."'1 The question of classification of patients with classic systemic manifestations but no chronic arthritis remains unresolved and differs from center to center.
SUBGROUPS OF JUVENILE RHEUMATOID ARTHRITIS
Children with systemic onset JRA are seronegative for both antinuclear antibodies and rheumatoid factors. As noted above, leukocytosis is regularly present during active systemic disease, and severe anemia is common. Early in the disease, radiographs show only soft tissue swelling about affected joints, and perhaps evidence of either pericarditis or pleuritis. If chronic arthritis persists for more than a year or so, joint radiographs may begin to show evidence of cartilage loss and joint destruction.
Diagnosis of systemic onset JRA rests entirely on clinical grounds. Other diseases associated with fever and arthritis- notably malignancies, systemic lupus erythematosus, inflammatory bowel disease, and Kawasaki disease - must be appropriately excluded.
The ultimate morbidity of systemic onset JRA lies in the relatively high risk of severe destructive chronic arthritis. About 25 percent of children with systemic onset disease will ultimately incur severe joint disability, although this may not occur for some years after disease onset.3 The systemic manifestations rarely recur after the individual reaches adulthood, and are generally more of a nuisance than life-threatening. Occasionally, however, severe heart involvement or severe anemia may demand heroic therapy. Growth retardation, described in the next section, may occur in some patients.
POLYARTICULAR SERONEGATIVE JRA
About 25 percent of children with JRA have polyarticular seronegative disease. Girls are affected much more frequently than boys (nine girls to one boy).3'14 The disease may begin at any time during childhood, often during the early childhood years. Family history is generally negative for rheumatic diseases (Table 2).
Any synovia! joints of the body may be affected, except perhaps those of the lumbodorsal spine. The most characteristic involvement is a symmetrical polyarthritis of small joints of the hands, particularly the proximal interphalangeal and metacarpophalangeal joints; small joints of the feet are similarly involved. Distal interphalangeal joints are also occasionally affected. Knees, ankles and elbows are involved in the majority of patients, and about half of patients have arthritis of temporomandibular joints, neck, hips or shoulders. Mild systemic complaints such as low grade fever, mild anemia, and modest hepatosplenomegaly may be associated, but the striking systemic complaints of systemic onset disease are, by definition, absent. Rheumatoid nodules are rarely associated; iridocyclitis and pericarditis are uncommon.2'' Generalized growth retardation may occur as a function of long-active inflammatory disease.15,16
By definition, tests for rheumatoid factors are negative at onset; they remain negative throughout the course of disease, regardless of duration of disease activity.1'17 About a quarter of patients have positive tests for antinuclear antibodies. Leukocytosis is not striking, and anemia, if present, is generally mild. Radiographs of joints late in disease may reveal evidence of joint destruction in severe cases (Table 2).
Diagnosis is based on clinical signs and symptoms, and assumes that tests for rheumatoid factors are negative.
The majority of patients with seronegative polyarthritis have a favorable prognosis; only about 10 percent suffer severe joint destruction. However, the disease may be long-active, and appropriate therapy is important to maintain good function. Growth retardation may resolve with remission of disease; or if active disease continues through the time of the ordinary pubertal growth spurt, it may be permanent.
POLYARTICULAR SEROPOSITIVE JRA
About 10 percent of children with JRA have polyarthritis with positive agglutination tests for rheumatoid factors (IgM rheumatoid factors).1*4 Girls are predominantly affected, and the age at onset is generally after eight years. Family history is occasionally positive for seropositive rheumatoid arthritis (Table 2).
The disease closely resembles severe adult onset seropositive rheumatoid arthritis. Joint involvement does not differ from the pattern seen in seronegative polyarthritis (described above), except, perhaps, for less frequent involvement of distal interphalangeal joints of the hands. However, the course of disease does differ in many instances from that of seronegative disease, with more rapidly progressive and destructive arthritis occurring in more than half of patients. Associated systemic complaints may include malaise, weight loss, low grade fever, and mild hepatosplenomegaly. Occasionally, patients develop rheumatoid vasculitis, as described in adult arthritis, splenomegaly with leukopenia (Felly's syndrome), or Sjogren's syndrome. Rheumatoid nodules are a frequent occurrence, as in adult disease.
By definition, agglutination tests for rheumatoid factor are positive. These tests are positive at disease onset and generally remain positive throughout the course of disease. Titers are generally greater than one to 40, and are often very high. Caution must be taken not to misinterpret transiently positive low-liter rheumatoid factor lesls as signifying the presence of seropositive disease; such transient "false positive" tests occur in the course of some viral infections, after certain immunizations, or as a result of faulty laboratory techniques.18 About 75 percent of patients with seropositive JRA also have positive tests for antinuclear antibodies.3 There is an association with Dw4 and HLA-DRw419 as in adult disease.20'"1 Radiographs frequently show evidence of joint destruction after six to 12 months of disease.
Diagnosis is made on clinical grounds and the presence of agglutination tests positive for rheumatoid factor.
Unfortunately, the majority of children with seropositive JRA have severe destruction and long-lasting arthritis. At least 50 percent will ultimately suffer significant joint destruction that interferes with function.
About 45 percent of children with chronic arthritis have arthritis which, rather than affecting multiplejoints, is limited to only a few joints (Latin: pauci - few; aniculus- joint).2'3'22 The distribution of affected joints is often asymmetrical. Large joints such as the knee, ankle and elbow are most often involved; isolated finger and toe joints may occasionally be involved, and also the cervical spine and the wrists. Patients may conform to the pattern and type of joint involvement, but may have as many as eight or nine total joints affected. Thus, pauciarticular disease is difficult to define solely on the total number of joints involved. Current American Rheumatism Association criteria define four or fewer joints as pauciarticular arthritis.23
Two distinct subgroups of disease are included under the designation of pauciarticular disease:22 pauciarticular arthritis of young onset patients associated with chronic iridocyclitis and positive tests for antinuclear antibodies (pauciarticular disease type I); and pauciarticular arthritis of older onset patients associated with sacroiliitis and subsequent spondyloarthropathy (pauciarticular disease type II).
* Pauciarticular Disease Type I
This group includes about 25 percent of all JRA patients.3,22 Patients are generally under the age of five years at the onset of disease. Girls are affected more often than boys. Family history is rarely positive for rheumatic disease (Table 2).
Knees, ankles and elbows are the most frequently affected joints. Affected joints are swollen, warm, stiff and painful as in polyarticular disease; indeed, synovial histology resembles that of classic rheumatoid arthritis. Hip involvement rarely occurs, nor do patients have sacroiliitis. Although the arthritis may be long-lasting, destructive arthritis is unusual and the prognosis for ultimate joint function is good. However, an estimated 50 percent of patients will have chronic iridocyclitis occurring sometime during the first 10 years of disease.24"28 (This type of eye inflammation affects the iris and ciliary body, is insidious and potentially scarring. It is associated with few early symptoms or signs, although occasional patients complain of photophobia, eye pain, or mild eye redness. Diagnosis can be made early only through routine slit-lamp examinations; optimally these should be performed four times a year for the first ten years of disease. Such slit-lamp examinations allow detection of increased cells and protein in the anterior chamber of the eye. Late sequelae of chronic iridocyclitis include posterior synechiae, band keratopathy, cataract formation, secondary glaucoma, visual loss, and even blindness. Early recognition and therapy of iridocyclitis are vital to preservation of vision [Table 2].)
There are no striking findings in peripheral blood counts in patients with pauciarticular arthritis; occasionally mild leukocytosis or anemia is present. Sedimentation rates may be normal or elevated. Tests for rheumatoid factors are negative. Antinuclear antibodies are strongly associated with iridocyclitis, and 50 percent or more of affected children have positive tests for them.28 There is an association with HLA-DR5, and perhaps DR8.29 Joint radiographs show soft tissue swelling around affected joints; joint destruction is often minimal, even with chronic disease. There maybe epiphyseal overgrowth about affected joints.
Diagnosis rests on the clinical picture and the exclusion of other diseases. It is particularly important to exclude septic arthritis early in the course of disease.
Most children with pauciarticular disease type I have a favorable prognosis concerning joint function. Significant joint destruction and disability are unusual, even though the arthritis may persist for years. The ultimate morbidity lies chiefly in the chronic iridocyclitis and the amount of eye scarring that occurs. Subsequent spondyloarthropathy and sacroiliitis are not associated with this type of disease.14,22
* Pauciarticular Disease Type II
This group includes about 15 percent of patients with "JRA".3,22 Males are predominantly affected; the age at onset is generally past eight years; and the family history is frequently positive for ankylosing spondylitis, Reiter's disease, pauciarticular arthritis, acute iridocyclitis, or inflammatory bowel disease (Table 2).
Affected joints resemble those of pauciarticular disease type I, but the distribution is somewhat different. Most patients have predominantly lower limb involvement of knees, ankles, or foot joints. Patients frequently have hip girdle involvement early in disease, and are often found to have radiographie sacroiliitis either at disease onset or sometime duringthefollow-upperiod. Peripheral arthritis may be of very acute onset with inability to walk due to hip, knee or foot involvement; on the other hand, some patients have indolent arthritis of long duration and few associated symptoms. In some patients, arthritis waxes and wanes, leading to misdiagnosis of acute rheumatic fever. Chronic iridocyclitis is not associated, but acute iridocyclitis sometimes occurs during their childhood years, resembling that of ankylosing spondylitis and Reiter's syndrome (Table 2).3,27,28
There are no striking findings in peripheral blood counts. Sedimentation rates may be elevated or normal. Tests for antinuclear antibodies and rheumatoid factors are negative; HLA-B27 can be found in 75 percent of patients.14 Radiographs of peripheral joints may show soft tissue and joint swelling; destructive arthritis in peripheral joints occasionally occurs with chronic disease. Destructive hip disease is present in some patients. Radiographie sacroiliitis can be found at onset or sometime during the period of followup in more than half of patients.
Diagnosis rests on the clinical picture and the exclusion of other diseases.
The ultimate morbidity for children with pauciarticular disease type II depends on the number who develop chronic spondyloarthropathy. It is well known that some of these patients will continue to have progressive ankylosing spondylitis as adults.30'31 However, it is quite possible that some of these children have mild disease that will remit without causing permanent damage. The overall prognosis for this group remains to be determined.
THERAPY OF JUVENILE RHEUMATOID ARTHRITIS
Therapy of juvenile rheumatoid arthritis should be geared to the manifestations and extent of disease in the individual patient. Several aspects of disease require therapy: arthritis, extra-articular manifestations, and the whole child. Various anti-inflammatory agents are employed in the treatment of arthritis. Of these, salicylates currently remain the safest and best; in sufficient dosages for sufficient periods of time, salicylates provide satisfactory control of disease in the majority of patients.11 An adequate course of salicylate therapy entails enough drug to maintain blood levels between 20 and 30 mgm% for several months. Aspirin, lOOmg/kgin four daily divided doses, is generally sufficient to achieve such blood levels for children of 25 kg body weight or less; for children heavier than 25 kg, doses of between 300 and 450 mg of aspirin daily generally suffice. Hazards of salicylate therapy include sah'cylism from overdosage and gastric irritation. If an adequate trial of salicylates fails to control synovitis, gold is generally the second drug of choice." Gold therapy has side effects similar in children and adults, and appears to be relatively safe if carefully prescribed. It is imperative that patients be examined prior to each gold injection; a complete blood count, platelet count, and urinalysis should also be done. The appropriate dosage is 1 mg/kg per week up to a body weight of 25 kg; the course of therapy is similar to that in adults - injections of maintenance dosage are given for 20 consecutive weeks and then spaced out to every two, three, and four weeks to maintain remission. Chloroquine is a useful antirheumatic drug in some children," but must be given carefully. There is possible ocular toxicity, and it is a poison without antidote if taken in overdose. Other nonsteroidal anti-inflammatory agents, such as indotnethacin, Butazolidin, ibuprofen and tolmetin, may prove to be useful adjunctive agents, but none has been shown clearly superior to aspirin in double blind studies; only tolmetin is approved for use in children in the United States.34'" Corticosteroids should be used as rarely as possible to control the arthritis of JRA; if absolutely necessary for symptomatic control of disease, they should be given in the lowest possible dosages.36 Ample evidence has accumulated that corticc steroid s do not prevent joint destruction in severe JRA and, indeed, may even hasten it. There is little rationale at the present time for the use of various anticancer drugs in the treatment of JRA.37
Systemic manifestations of systemic-onset JRA are usually controlled by salicylates in the doses described above. If an adequate trial of salicylate therapy is not effective, a short course of therapy with corticosteroids is indicated to bring symptomatic relief to the patient. Such corticosteroid therapy should never be prolonged since systemic manifestations of JRA rarely exceed a duration of six months. Patients can be maintained on salicylates as corticosteroids are being withdrawn. In management of patients with systemic-onset JRA, it is important to define whether joint disease, systemic manifestations, or both require therapy at the moment.
The iridocyclitis of JRA should be managed in consultation with an ophthalmologist.24'27 Topical steroids and dilating agents are appropriate initial therapy and will adequately control the eye disease in many patients. If active ocular inflammation persists, however, steroids - either locally injected or systemically administered - should be used in an effort to preserve vision. Steroid doses can be regulated by observing the degree of inflammation seen on slit-lamp examination; the lowest possible dose to achieve suppression of the process should always be used. Alternate-day dosage schedules may prove effective.
Concerning treatment of the whole child, great care should be taken that patients avoid the image of chronic invalidism. Patients need a physician who is comfortable with the disease and is able to work closely with the patient and the family. Children should not be needlessly restricted from activities, but rather should be encouraged to do as much as possible. Bed rest is generally contraindicated, and all but the most severely handicapped children should attend regular schools.
1. Still GF. On a form of chronic joint disease in children. Med Chir Trans 80:47-59. 1897.
2. Schaller J and Wedgwood RJ. Is juvenile rheumatoid arthritis a single disease? A review. Pediatr 50:940-953, 1972.
3. Schaller JG. The diversity of JRA: A 1976 look at the subgroups of chronic childhood arthritis. Arthritis Rheum (Suppl) 20:S52-S6I, 1977.
4. Ansell BM. Chronic arthritis in childhood. Ann Rheum Dis 37:107-120, 1978.
5. Schaller JG. An h r it is and infections of bones and joints in children. In: Common Orthopedic Problema. Staheli L (ed). Pediatr Clin North Am 24:755-789, 1977.
6. McMinn FJ and By waters EGL. Differences between the fever of Still's disease and that of rheumatic fever. Ann Rheum Dis 18:293-297, 1959.
7. Ca labro JM and Marchesane JM. Fever associated with juvenile rheumatoid arthritis, N Engl J Med 276:11-18, 1967.
8. Isdale IC and By waters EGL. The rash of rheumatoid arthritis and Still's disease. Quart J Med 25:377-387, 1956.
9. Calabro JJ and Marchesane J M. Rash associated with juvenile rheumatoid arthritis. J Pediatr 72:61 1-619, 1968.
10. Schaller J, Beck wit h B, Wedgwood RJ. Hepatic involvement in juvenile rheumatoid arthritis. J Pediatr 77:203-210, 1970.
11. Schaller JG. Chronic salicylate administration in juvenile rheumatoid arthritis. Aspirin 'hepatitis" and its clinical significance. Pediatrics (Suppl) 62:916-925, 1978.
12. Lietman PS and By waters EGL. Pericarditis in juvenile rheumatoid arthritis. Pediatrics 32:855-860. 1963.
13. Bernstein B. Ta ka has hi M, Hanson V. Cardiac involvement in juvenile rheumatoid arthritis. J Pediatr 85:313-317, 1974.
14. Schaller JG. et al. Histocompatibiliiy antigens in childhood onset arthritis. J Pedialr 88:926-930, 1976.
15. Ansell BM. Bywaters EGL. Growth in Still's disease. Ann Rheum Dis 15:295-319, 1956.
16. Bernstein BH. et al. Growth retardation in juvenile rheumatoid arthritis (JRA). Arthritis Rheum 20:212-216, 1977.
17. Jeremy R, et al. Juvenile rheumatoid arthritis persisting into adulthood. Am J Med 45:419-434, 1968.
18. Bartfield H. Distribution of rheumatoid factor activity in nonrheumatoid states. Ann NY Acad Sci 168:30-40, 1969.
19. Schaller J and Hansen J. In preparation.
20. Stastny P. HLA-D typing in rheumatoid arthritis. Arthritis Rheum (Suppl 6) 20:S45-S49, 1977.
21. Stastny P. Association of the B-cell alloantigen DRw4 with rheumatoid arthritis. N Engl J Med 298:869-871, 1978.
22. Schaller JG and Wedgwood RJ. Pa uc iarticular childhood arthritis: Identification of two distinct subgroups. Arthritis Rheum 19:820-821. 1976.
23. Brewer EJ. et al. Current proposed revision of JRA criteria. Arthritis Rheum 20:195-199, 1977.
24. Smiley WK, May E, By walers EGL. Ocular presentations of Still's disease and their treatment. Ann Rheum Dis 16:371-382, 1957.
25. Schaller J, Kupfer C, Wedgwood RJ. Iridocyclitis in juvenile rheumatoid arthritis. Pediatrics 44:92-100, 1969.
26. Chylack LT Jr. The ocular manifestations of juvenile rheumatoid arthritis. Arthritis Rheum 20:217-223, 1977.
27. Kanski JJ. Anterior uveitis in juvenile rheumatoid arthritis. Arch Ophthalmol 95:1794-1797, 1977.
28. Schaller JG, et al. The association of antinuclear antibodies with the chronic iridocyclitis of juvenile arthritis (Still's disease). Arthritis Rheum 17:409-416. 1974.
29. Suciu-Foca N, et al. Increased frequency of DRw5 in pauciarlicular JRA. In: Histocompatbiliir Typing i980. Terasaki PI (ed). Los Angeles: UCLA Tissue Typing Laboratory, 1980, p. 953.
30. Schaller J, Bilnum S, Wedgwood RJ. Ankylosing spondylitis with childhood onset. J Pediatr 74:505-516. 1969.
31. Bywaters EGL. Ankytosing spondylitis in childhood. Clin Rheum Dis 2:387-396, 1976.
32. Levinson JE, BaIiGP, Bondi S. Gold therapy. Arthritis Rheum 20:531-535, 1977.
33. Stillman JS, et al. Antimalarial therapy in JRA. In press, 1978.
34. Leunson JE, et al. Comparison of tolmetin sodium and aspirin in the treatment of juvenile rheumatoid arthritis. J Pediatr 91:799-804. 1977.
35. Brewer EJ. Nonsteroidalantiinflammatory agents. Arthritis Rheum 20:5 13-525. 1977.
36. Schaller JG. Con ico stero ids in juvenile rheumatoid arthritis. Arthritis Rheum 20:537-543. 1977.
37. Hollister JR. lmmunosuppressant therapy of juvenile rheumatoid arthritis. Arthritis Rheum 20:544-547. 1977.
CONDITIONS CAUSING OR SIMULATING CHILDHOOD ARTHRITIS
SUBGROUPS OF JUVENILE RHEUMATOID ARTHRITIS