Ocular Manifestations of Juvenile Rheumatoid Arthritis

OCULAR MANIFESTATIONS O F J U V E N I L E RHEUMATOID A R T H R I T I S LEO T.

CHYLACK, JR., M.D.,

DON C. BIENFANG, M.D.,

AND J. SYDNEY STILLMAN,

A. ROBERT BELLOWS,

M.D.,

M.D.

Boston, Massachusetts

Juvenile rheumatoid arthritis (JRA) is a chronic, progressive, crippling disease of childhood. Diamantberger first described the arthritic and systemic manifestations in 1891,1 and George Frederick Still published a more widely known paper in 1897.2 How­ ever, it was not until 1910 that'Ohm 3 first noted the occurrence of iridocyclitis in this disease. Since then, in varying detail, the clin­ ical individuality of the iridocyclitis of JRA was described.4"18 While most authors em­ phasized the need for early detection and treatment, there has been no published study of the effect of regular periodic slit-lamp biomicroscopy and early treatment on the course of the ocular manifestations of JRA. We de­ scribe our experience with 210 patients with JRA who were followed closely and regu­ larly since 1963. Each patient participated in a long-term prospective study of the articu­ lar and systemic, as well as ocular, manifesta­ tions of JRA.

form of JRA has been recognized for many years7'9·11·19 ; therefore, all cases of mono- or pauciarticular JRA, as well as all patients with a history of uveitis, were examined at least every three months. Patients with active iridocyclitis were seen as frequently as neces­ sary. Children with the polyarticular form were seen at least once a year, or as needed. Examination consisted of best corrected vis­ ual acuity at near and distance, color vision testing with H-R-R pseudoisochromatic plates, tactile, Schlitz, or applanation tonometry, external examination, slit-lamp biomicroscopy, direct or indirect ophthalmoscopy or both, and scierai depression. Perimetry studies, psychophysical and electroretinographic testing was done, if indicated. We followed the progress of children in the Bos­ ton area while local ophthalmologists followed those outside the area. At least once a year, all patients received a complete clinical and laboratory evaluation.

SUBJECTS AND METHODS

RESULTS

Approximately 325 patients with possible JRA were admitted to the long-term prospec­ tive study. Of these, 210 patients had JRA, according to the criteria established by the American Rheumatism Association. Each pa­ tient underwent a regular ophthalmological examination and the data were subjected to computer-type analysis. The higher incidence of iridocyclitis in the mono-pauciarticular

There were 36 (17.2%) cases of iridocy­ clitis in 210 cases of definite JRA (Table 1). There was a 5:1 preponderance of females to males in the uveitis group compared to the 2.5:1 ratio in the nonuveitis group ; the size TABLE 1 INCIDENCE OF IRIDOCYCLITIS IN JUVENILE RHEUMATOID ARTHRITIS

Female Male From the Robert Breck Brigham Hospital, Har­ vard Medical School, Boston, Massachusetts. This No. % No. %* study was funded by National Arthritis Foundation, Mabel Louise Reilly Charitable Trust, Peabody No uveitis 49 22 125 78 Foundation, Charles E. Merrill Trust, Jerry Fund, Uveitis 30 83 6 17 and the Mayo-Smith Fund. 155 74 Total 55 26 Reprint requests to Leo T. Chylack, Jr., M.D., Peter Bent Brigham Hospital, 721 Huntington Ave., Boston, MA 02115. * % refers to row subtotal. 1026

Total No.

%

174 36

100 100

210

100

VOL. 79, NO. 6

1027

JUVENILE RHEUMATOID ARTHRITIS TABLE 2 TYPE OF ARTHRITIS IN POPULATION

Onset Insidious Insidious Insidious Sudden Uncertain Total

Course Monoarticular Oligoarticular Polyarticular Systemic

With Uveitis

Without Uveitis

Entire Group

No.

%

No.

%

No.

%

28

78

55

32

83

39.5

5 2 1

14 6 2

76 34 9

44 19 5

81 36 10

38.6 17.1 4.8

36

100

174

100

210

of the male and female samples was not con­ sidered. Eleven percent of the male popula­ tion (55), and 19% of the female population (155) had uveitis, with a ratio of 1.8:1 cor­ rected for the size of the samples. Charac­ teristically, the onset of juvenile rheumatoid arthritis was insidious (78.1%) (Table 2 ) . Patients with the mono-pauciarticular form of the disease comprised 78% of all patients with uveitis. Those patients with the poly­ articular or systemic onset were less likely to develop iridocyclitis. Mono-pauciarticular patients represented only 39.5% of the total population, but comprised 78% of the uveitis cases. Sixty-one percent of the uveitis population entered the study with a noncontributory oc­ ular history (Table 3). A positive history of uveitis (39%) reflected the number of estab­ lished JRA cases with diagnosed uveitis re­ ferred to the study. However, 67% of those patients who had uveitis before or during the study had evidence of uveitis on entry. Thus, in 28% of the patients uveitis was detected for the first time on entry. The ophthalmic evaluation was an integral part of the first evaluation of the patient, supported by the fact that 42% of the patients had active dis­ ease on entry. Seventeen of the 21 cases that were inactive on entry became active during the course of the study. Ophthalmologists outside the hospital detected uveitis in only three of these patients. The high rate of de­ tection of previously unsuspected uveitis at

100

the time of initial examination, as well as the equally high rate of discovery of the new cases of uveitis and flares of activity in quies­ cent cases, points out the desirability of the hospital-based, multispecialty approach to the disease and the need for regular frequent oc­ ular examinations. Sixty-nine percent of the children with uveitis developed arthritis before 4 years of age (Table 4 ) . This was in striking contrast to the nonuveitis group in which there was no single age group between birth and 16 years with a higher attack rate of arthritis. TABLE 3 TIME OF DIAGNOSIS OF UVEITIS

History

No.

%

22 14

61 39

12 24

33 67

21 15

58 42

21 17 13 3 4

58 47 36 8 11

At Entry into Study Negative for any ocular disease Positive for uveitis Evidence of uveitis on initial ocular examination Negative Positive Activity of uveitis Inactive Active After Entry into Study Inactive on entry Active later Detected in clinic Detected elsewhere Inactive throughout

JUNE, 1975

AMERICAN JOURNAL OF OPHTHALMOLOGY

1028

TABLE 4 AGE AT ONSET

Age (yrs) Birth-2 2-4 4-6 6-8 8-10 10-12 12-14 14-16 16-20 20 + Total

Arthritis in Patients Who Developed Uveitis

Arthritis in Patients Without Uveitis

Uveitis in Patients by Age

No.

%

No.

%

No.

%

17 8 1 1 4 2 0 3 0 0

47 22 3 3 11 6 0 8 0 0

8 26 19 22 13 32 27 19 7 0

5 IS 11 13 7 18 16 11 4 0

3 4 5 2 3 5 0 3 6 S

8 11 14 7 8 14 0 8 16 14

36

100

173

100

36

100

Only 19% of the patients developed uveitis before the age of 4 years ; therefore, arthritis will bring these children into a medical com­ munity. Most of the children who developed uveitis were preschoolers and not subjected to regular school vision tests. They were not likely to verbalize visual disturbances. Their parents were usually prepresbyopics who would not contact an ophthalmologist. Thirty percent of the patients with definite JRA developed their uveitis after the age of 16 (Table 4). There was no mention in pub­ lished reports of the risk of developing uve­ itis late in this disease. Our data suggest that late appearance of uveitis is not infrequent. Most published reports emphasized the asymptomatic nature of the onset of ocular inflammation; in our series, 56% fell into this group. However, 44% had one or more identifiable signs or symptoms (Table 5). In order of decreasing frequency, red eye, ocular pain, decreased visual acuity, and photophobia were identified. The iridocyclitis was as often monocular as binocular. Most published reports about uveitis in JRA stressed the importance of early detec­ tion ; however, the inflammation was difficult to control once detected. Many papers cited an incidence of serious complications and se­ vere visual acuity loss as high as 50%.8'13,15,1T

We rapidly controlled 39% of the episodes, with complete clearing of signs and symp­ toms of ocular inflammation (Table 6 ) . How­ ever, an equal number (41%) required more than six months of intensive treatment. In these cases, we administered dexamethasone, 0.1% in both eyes, every hour during waking hours, plus the ointment form at bedtime, as well as atropine, twice daily. In a few cases, we administered subtenon's injections of corticosteroids or systemic corticosteroids with no radical change in the manifest degree of inflammation. These results were certainly consistent with many published reports of TABLE 5 MODE OF PRESENTATION OF UVEITIS

Symptom Pain Decreased vision Photophobia Headache Sign Red eye Anisocoria Completely asymptomatic One symptom or sign More than one symptom or sign Monocular Binocular

No.

%

8 6 3 2

22 17 8 6

9 1 20 8 8 19 17

25 3 56 22 22 53 47

VOL. 79, NO. 6

JUVENILE RHEUMATOID ARTHRITIS

1029

TABLE 6

TABLE 7

RESPONSE OF UVEITIS TO CORTICOSTEROID AND MYDRIATIC-CYCLOPLEGIC TREATMENT

INCIDENCE OF UVEITIS COMPLICATIONS

Clearing Inflammation

No. of Episodes

/0

„

Rapid (less than 3 mo) Slow (3-6 mo) None (more than 6 mo) Uncertain

16 4 17 4

39 10 41 10

treatment failures in this disease but our re­ sults were not discouraging. The complica­ tions encountered in this group (Table 7) included posterior synechiae, visual acuity loss, and cataract as the most frequent. Many of these eyes had more than one complica­ tion ; cataract, band-shaped keratopathy, vis­ ual acuity loss, and posterior synechiae were present concurrently in many eyes. However, the incidence and severity of band keratopathy and cataract in this group were less than the often published 50% figure in many re­ ports.8·13'17 The visual acuity loss was greater than 20/200 in only ten eyes. Our close, regular patient care approach led to a significant de­ crease in the incidence and severity of ocular complications when compared with other series (Tables 7 and 8 ) . Our regimen of corticosteroids and mydriatics, although not the ideal treatment, was effective. In addition to the inflammatory ocular manifestations of JRA, there were drugrelated ocular problems. Hydroxychloroquine and chloroqiiine were often used (Table 9). Prior to the availability of hydroxychloro­ quine, chloroquine was used; later, all these patients received hydroxychloroquine. We found one case of hydroxychloroquine chorioretinopathy among the 36 uveitis patients, an 18-year-old woman who received chloro­ quine for one year. Over the next 5.3 years, she received 424 g of hydroxychloroquine. At age 24, macular pigmentation with a crescent of pigment rarefaction was noted and we found broken ring scotomata in the central field.

Complication

No. of Eyes

%

Posterior synechiae Visual acuity loss Cataract Glaucoma Band keratopathy Phthisis bulbi Anterior synechiae Iris atrophy Papillitis

27 16 16 10 9 7 2 2 1

38 22 22 14 13 10 3 3 1

Thirty percent of all JRA patients received systemically administered corticosteroids. Fifty-seven percent of the patients with uve­ itis received corticosteroids systemically, but only 17% for uveitis treatment; the rest of the group received them for JRA treatment. Use of systemic corticosteroids in the JRA treatment group was associated with posterior subcapsular lens opacification in many pa­ tients. The results of surgical treatment of uve­ itis complications were compiled (Table 10). We performed none of the surgery. Despite the lack of detail about technique, the uni­ formly poor results suggest that no one tech­ nique of cataract extraction was clearly su­ perior. Because many of the patients had biTABLE 8 LOSS OF VISION DUE TO UVEITIS COMPLICATIONS DURING THE STUDY

Visual Acuity

No. of Eyes

20/40-20/60 20/60-20/80 20/80-20/100 20/100-20/200 20/200-20/400 Counting fingers Hand movements Light projection Light perception No light perception

4 1 1 0 2 4 1 1 0 2

Total

16 (22%)

JUNE. 1975

AMERICAN JOURNAL OF OPHTHALMOLOGY

1030

TABLE 9 DRUGS USED IN ARTHRITIS TREATMENT

Uveitis

No Uveitis Female

Male

Prednisone Corticc tropin Hydroxychloroquine Chloroquine Gold Au 198

Female

Male

No.

%

No.

%

No.

%

No.

%

1 0 4 1 1

17 0 67 17 17

12 3 17 3 4

40 10 57 10 13

10

19.6

52

34.9

29 6 8

56.9 11.8 15.7

89 24 29

59.7 16.1 19.5

lateral uveitis complications, this group prob­ ably had particularly severe disease. These patients needed help, but frequently benefited little from surgery. Three of 36 patients with uveitis had symp­ tomatic keratoconjunctivitis sicca with posi­ tive Schirmer tests. This association with JRA was described previously.20 DISCUSSION

The iris and ciliary body are the principal tissues involved in the uveitis of JRA. This consensus, however, is based on clinical im­ pression, not histopathology, since there is no published study of the histopathology of this ocular disease. Only one of our patients had posterior segment involvement; in this par­ ticularly acute iridocyclitis, papillitis devel­ oped. We administered large doses of sys­

temic corticosteroids, and hospital bed rest, before the inflammation subsided. Fortunate­ ly, this patient regained visual acuity of 20/20. We dilated and examined several eyes of our patients with indirect ophthalmoscopy and scierai depression during active anterior segment inflammation. No foci of posterior inflammation were found. One study10 de­ scribed inflammatory cells in the retrolental vitreous body, another14 a single choroidal focus of inflammation, and another12 a cen­ tral retinal scar. All other studies implied that the posterior segment was not involved in JRA uveitis. The incidence of iridocyclitis in JRA varies only slightly from one study to another. Eight to 24% is the range in several studieSj6-io,i2,is w j t | 1 m o s t favoring i o % . The in­ cidence of uveitis in our patients was some-

TABLE 10 SURGICAL RESULTS

Case No., Eye

Surgical Procedure

1, ? 2, R.E. L.E. 3, R.E. L.E. 4, R.E.

Cataract extraction, 1971 Cataract extraction, 1971 Cataract extraction, 1972 Cataract extraction, 1965 Cataract extraction, 1966 Iridectomy, 1961 Cataract extraction, 1967 Iridectomy, 1961 Cataract extraction, 1967 Cataract extraction, 1950 Filtering surgery, 1953 "Scraping of cornea," 1940

L.E. 5, R.E. L.E.

Best Postoperative Visual Acuity Light projection, 1973 20/100, 1973 20/50, 1973 20/200, 1973 15/200, 1973 Light perception, 1973 No light perception, 1973 13/200, 1973 No light perception

VOL. 79, NO. 6

JUVENILE RHEUMATOID ARTHRITIS

what high (17.2%). This is only partially explained by the high incidence (39.5%) of mono-pauciarticular arthritis in our series. Other investigators reported a 34.3 % 8 and 30.8% 21 incidence of mono-pauciarticular dis­ ease. Recent studies6"8·10*18 stressed that the incidence of iridocyclitis is much higher (29%) in the mono- and pauciarticular form than in thé polyarticular form of the illness. This reverses an earlier clinical impression16 that patients with chronic polyarthritis had more ocular complications. Arthritis preceded the uveitis in 32 of 36 cases (89%). This is similar to published reports6·'·15*16 showing that 75 to 8 8 % of the patients developed iridocyclitis two months to 12 years after the onset of the arthritis. In three of our 36 cases, the uveitis and the arthritis appeared simultaneously, but the uveitis preceded the arthritis in only one patient. Although un­ usual, several case reports documented the occurrence of uveitis prior to the arthri£ j s 7,9,18,17,22

In most children, the iridocyclitis is a quiet, chronic, insidious anterior segment in­ flammation that produces few symp­ toms.7·8'13·17 In published reports,7·8 only 12 to 25% of the patients with uveitis presented with symptoms of severe pain, redness, and photophobia. However, in our series, 44% of the patients had one or more recognizable presenting signs or symptoms (Table 5). There was no parallel between the activ­ ity of the iridocyclitis and the joint disease. This confirms similar findings in earlier re­ ports.7·8·13,17 Laboratory studies were not helpful in detecting or managing the irido­ cyclitis of JRA. 7 · 8 In some series, complicated cataracts8·13·17 and band keratopathy15·17 were present in 50% of the children with iridocyclitis and JRA. In most series,11·16 these complications were present in slightly less than 50% of children. In our series (Table 7), 22% of afflicted eyes developed cataracts and only 13% developed band keratopathy. In view of the long follow-up period for most of

1031

our patients (mean follow-up time for pa­ tients with uveitis is 6.3 years), these data are significantly better than other published data. The incidence and degree of visual loss is difficult to ascertain from most published studies ; in 1950,17 52% of the patients with iridocyclitis and JRA were largely incapaci­ tated, with visual acuities of 0 to 20/120 in their better eyes. Forty-eight percent had full functional visual capacity. A more recent study7 indicated that "visual loss did not oc­ cur in patients referred without visual loss," but data as to length of follow-up were miss­ ing. In our series, 78% of the eyes had no loss of vision during the study; of the pa­ tients (22% ) who experienced loss of visual acuity during the study, ten of 72 eyes had visual acuity less than 20/200. The lower in­ cidence of complication in our group is a significant improvement. However, the fact that a significant number of eyes still de­ velop incapacitating visual acuity loss indi­ cates that further improvement in treatment is necessary. Treatment of iridocyclitis in JRA centers about the use of corticosteroids and mydriatics. Corticosteroids are most frequently used topically but they have been adminis­ tered by means of a subtenon's injection. They are also given orally. In most studies,8·11'15·17 there is little treatment detail. In our group, 39% of the eyes responded rap­ idly to treatment but 4 1 % showed no sig­ nificant response over a six-month period, despite steadily increasing the dose of corticosteroid. These results indicate the need for an alternate therapeutic modality. Corticotropin ( A C T H ) 1 5 and cyclophosphamide28 have been used successfully to treat this form of iridocyclitis. Neither of these drugs could be used as an alternative to cor­ ticosteroids in this young population. Until a safe, more effective drug is found, careful follow-up of each case and use of cortico­ steroids topically, and occasionally systemically, is the best therapy, since it can prevent

1032

AMERICAN JOURNAL OF OPHTHALMOLOGY

serious visual acuity loss and other complica­ tions in better than 8 0 % of the patients with iridocyclitis. T h e community physician is vital to the success of the management of J R A by a multispecialty group. Since a hospital-based pro­ gram alone cannot meet the needs for con­ tinuing care of nonlocal patients, the family physician or pediatrician and the local com­ munity ophthalmologic specialist provide much of the day-to-day medical care. This is as true for ophthalmologic needs as it is for medical and orthopedic needs. Such an asso­ ciation offers the patient a significantly better chance for good health. SUMMARY

W e followed 210 cases of juvenile rheuma­ toid arthritis closely for eleven years. Thirtysix of the 210 patients ( 1 7 . 2 % ) developed iridocyclitis. Iridocyclitis was seen most fre­ quently in young female patients ( 0 to 4 years) with the monoarticular or pauciarticular form of the arthritis. However, 3 0 % of the patients developed uveitis after 16 years of age. Although 6 1 % of patients had a noncontributory ocular history on entry, 4 2 % had active uveitis on entry. O u r approach was effective in detecting uveitis in new cases and exacerbations of uveitis in established cases. Forty-four percent of patients with uveitis had one or more identifiable signs or symp­ toms, such as red eye, ocular pain, decreased visual acuity, or photophobia, in order of de­ creasing frequency. Even after early detec­ tion and prompt treatment, 4 1 % of cases of uveitis did not respond to more than six months of intensive topical treatment with corticosteroids and mydriatics. Despite this, there was a dramatic decrease in the 5 0 % in­ cidence of blinding complications of uveitis cited in earlier studies. Cataract and band keratopathy occurred in only 22 and 1 3 % of our group, respectively. W e used chloroquine or hydroxychloroquine in 173 of 210 cases and found only one case of chorioretinopathy attributable to these drugs. Systemically ad­

JUNE, 1975

ministered corticosteroids were used in 75 of 210 cases ; a significant number of posterior subcapsular cataracts was found. Typical keratoconjunctivitis sicca devel­ oped in three of the uveitis cases. This asso­ ciation with uveitis and J R A was not noted previously. Surgical treatment of cataracts, band kera­ topathy, and glaucoma achieved uniformly discouraging results. ACKNOWLEDGMENTS

We thank Beverly Hoffman for assistance with case reviews, and Ray Neff, Harvard School of Public Health, for assistance with statistical evalua­ tion of the data. REFERENCES

1. Diamantberger, M. S. : Du Rhumatisme Noueux (Polyarthrite Déformante), Chez les En­ fants, Lecrosmier et Badé, Libraire, Editeures, Paris, 1891, pp. 1-148. 2. Still, G. F. : On a form of chronic joint dis­ ease in children. Med. Chir. Trans. 80:47, 1897. 3. Ohm, J. : Bandförmige Hornhauttrübung bei einem neunjährigen Mädchen und ihre Behandlung mit subkonjunktivalen Jodkaliumeinspritzungen. Klin. Monatsbl. Augenheilkd. 48:243, 1910. 4. Schaller, J., and Wedgwood, R. J. : Juvenile rheumatoid arthritis. A review. Pediatrics 50:940, 1972. 5. Calabro, J. J., Katz, R. M., and Maltz, B. A. : A critical reappraisal of juvenile rheumatoid arthritis. Clin. Orthop. 74:101, 1971. 6. Lods, F., and Ziegler, M. G.: Contributions à l'étude des syndromes oculo-articulaires de l'enfant (4 cas d'uveitis au cours des monoarthrities du genous). Bull. Soc. Ophtalmol. Fr. 70:490, 1970. 7. Calabro, J. J., Parrino, G. R., Atchoo, P. D., Marchesano, J. M., and Goldberg, L. S. : Chronic iridocyclitis in juvenile rheumatoid arthritis. Ar­ thritis Rheum. 13:406, 1970. 8. Schaller, J., Kupfer, G, and Wedgwood, R. J. : Iridocyclitis in juvenile rheumatoid arthritis. Pedi­ atrics 44:92, 1969. 9. Pietrowa, N., and Duchowska, H. : Eye changes in the course of rheumatoid arthritis in children. Pediatrics 43:1489, 1968. 10. Kohoutek, J., and Havelka, S. : The results of the ophthalmic examination in rheumatoid arthri­ tis in children. The eye in juvenile rheumatoid arthritis. Rheumatologia 5:47,1967. 11. Steward, A. J., and Hill, B. M.: Ocular mani­ festations in juverilè rheumatoid arthritis. Can. J. Ophthalmol. 2:58, 1967. 12. Vasseler, V., and Ivanor, N. : Clinical and paraclinical research to establish the role of rheu­ matism and of rheumatoid arthritis in the etiology of iridocyclitis. Folia Med. 9:291, 1967.

VOL. 79, NO. 6

JUVENILE RHEUMATOID ARTHRITIS

13. Smiley, W. K. : Iridocyclitis in Still's dis­ ease. Trans. Ophthalmol. Soc. U.K. 85:351, 1965. 14. Edstrom, G. : Band-shaped keratitis in juve­ nile rheumatoid arthritis. Acta Rheum. Scand. 7: 169, 1961. 15. Smiley, W. K., May, E., and Bywaters, E. G. L. : Ocular presentations of Still's disease. Its complications and treatment. Ann. Rheum. Dis. 16 : 371, 1957. 16. Franceschetti, A., Blum, J. D., and Bamatter, F. : Diagnostic value of ocular symptoms in juvenile chronic polyarthritis (Still's disease). Trans. Oph­ thalmol. Soc. U.K. 71:17, 1951. 17. Vesterdal, E., and Surry, B. : Iridocyclitis and band-shaped corneal opacity in juvenile rheumatoid arthritis. Acta Ophthalmol. 28:321, 1950. 18. Blevad, O. : Iridocyclitis and disease of the joints in children. Acta Ophthalmol. 19:219, 1941.

1033

19. Brattstrom, M., and Sundberg, J. : Juvenile rheumatoid gonarthritis. Acta Rheum. Scand. 11 : 266, 1965. 20. Jackson, J., Anderson, L., Schur, P. H., and Stillman, J. S. : Sjögren's syndrome in juvenile rheumatoid arthritis (JRA). Arthritis Rheum. 16: 122, 1973. 21. José, D. G., and Good, R. A. : Iridocyclitis and pauciarticular juvenile rheumatoid arthritis. J. Pediatr. 78:910, 1971. 22. Pietrowa, N., Sowinska, J., and Strzelecka, M. : Early ophthalmic changes as the first mani­ festation of rheumatoid arthritis in children. Rheumatologia 4:305, 1966. 23. Buckley, C. E , III, and Gills, J. P., Jr.: Cyclophosphamide therapy of peripheral uveitis. Arch. Intern. Med. 124:29, 1969.

Fnophthal mos P—xophthalmos Prepared by Virginia Howard Louisiana State University Medical Center New Orleans