Arthritis and Rheumatism DECEMBER, 1958 VOL. I, NO. 6 The Relationship to Rheumatoid Arthritis of Its So-Called Variants By CURFUER MCEWEN,MORRISZIFP, PHILIPCARMEL,DOMENICK DITATA AND MARTIN TANNER The principal manifestations of certain syndromes presumed to be variants of rheumatoid arthritis are reviewed and the results of serologic studies presented. The entities investigated were ankylosing spondylitis, Still's disease, psodatic arthritis, the arthritis of ulcerative colitis and Reiter's syndrome. Serologic observations suggested that Still's syndrome is truly juvenile rheumatoid arthritis but that the other variants may be separate diseases. Es passate in revista le manifestationes principal de certe syndromes que presumitemente es variantes de arthritis rheumatoide. Le resultatos de studios serologic es presentate. Le entitates investigate es spondylitis ankylosante, morbo de Still, arthritis psoriatic, le arthritis de colitis ulcerative, e syndrome de Reiter. Le\ constatationes serologic suggere que le morbo de Still es de facto un forma juvenil de arthritis rheumatoide, durante que le altere variantes es plus probabilemente morbos separate. I N THE CLASSIFICATION of rheumatic diseases adopted by the American Rheumatism Association in 1941* ankylosing sponclylitis and Still's disease are listed as variants of rheumatoid arthritis.' Many physicians in the United States and England also consider arthritis accompanying psoriasis and that accompanying ulcerative colitis to be variants of rheumatoid arthritis; indeed, psoriatic arthritis was so listed in the classification prepared by a committee" of the New York Rheumatism Association in 1940.* By a smaller group Reiter's disease is also suspected of being a modified form of rheumatoid arthritis. In Europe, on the other hand, these types of joint disease (with the usual exception of Still's disease) are thought by most rheumatologists to be distinct from rheumatoid arthritis. At the time the From the Department of Medicine and the Rheumatic Diseases Study Group, New York University College of Medicine, the Third Medical Division, BeUeoue Hospital Center, the New York Veterans Administration Hospital, and the Brooklyn Outpatient Clinic and Medical Diuision, New York Regbnul Ofice, Veterans Administration. Aided by grants from tlre Nutionul Institute of Arthritis and Metabolic Diseases, U.S.P.H.S. (A-l431(Cl), 2A-S064(C) 6 (CS) ( A ) , A-679(C3) ), The Arthritis and Rheumatism Foundation and the New York State Chapter of The Lrtlrritis nnd Rheumatism Foundation. One of the authors ( M . 2.) is Senior Intwstigatm, Arthritis nnd Rheumatism Foundation. *The senior author was a member of the committees of Imth the New York Rheumatism Association and the American Rheumatism Association which drew lip these classifications. 481 482 MCEWEN, ZIFi!', CARMEL, DITATA AND TANNER American Rheumatism Association's committee for the adoption of a classification of arthritis was deliberating, there was considerable doubt in the minds of some members of the committee that ankylosing spondylitis and arthritis accompanying psoriasis should be considered variants of rheumatoid arthritis. In the case of psoriatic arthritis these doubts prevailed, and it was separately listed. In the case of ankylosing spondylitis, however, pathologic studies had revealed histologic lesions similar to those of characteristic rheumatoid arthritis?~~ and the majority opinion was that it should be considered merely a special form of the latter. Now, in the light of advancing knowledge of clinical, pathologic and other laboratory aspects of rheumatic diseases, it is timely to reconsider the question of the relationship of these forms of joint disease to rheumatoid arthritis. This paper is written for that purpose. The various clinical, radiologic and pathologic similarities and differences will be reviewed briefly, and particular attention will be devoted to information provided by results of sheep erythrocyte agglutination tests in the various types of joint diseases under discussion. In this paper the abbreviation S.E.A. will be used to designate the hemagglutination test in general. RESULTS O F S.E.A. TESTS Still's Disease Stillj originally considered the disease of children which subsequently bore his name to be distinct from childhood and adult rheumatoid arthritis. Today, however, it is rather generally agreed that Still's disease is merely rheumatoid arthritis occurring in children. Nevertheless, this opinion has not beell accepted by and various differences have been pointed out. Among these may be mentioned the more frequent occurrence of high fever and severe systemic and visceral manifestations, the occurrence in some patients of a rather distinctive maculopapular cmiption,n the extremely common involvement of the cervical spine,Rthe frequency of comparative freedom from pain in spite of marked swelling, the great rarity of subcutaneous nodules and the differences in character of such nodules when they do occur.lo On the whole, however, the clinical features of Still's disease are similar to those of adult rheumatoid arthritis, and the gross and histologic changes in the joints are essentially the ~arne.~Jl'~:' In the present study the sheep erythrocyte agglutination test was done with serum from 31 patients with Still's disease. The method used for these tests (and for those to be reported throughout this paper) was that of ZifE,l4 using the euglobulin fraction of the test sera for both direct agglutination and the more sensitive inhibition technic. Twenty-four of the patients were under the age of sixteen years, and seven were adults past the age of twenty-five. In all, the arthritis had begun before the age of twelve years. The results of the tests on these patients are shown in table 1. It is seen that 58 per cent of the children gave positive tests by direct agglutination and that all but one (95.8 per cent) were positive by the more sensitive inhibition technic. Of the seven adults, four were positive by agglu- 483 RHEUMATOID ARTHRITIS AND ITS SO-CALLED VARIANTS TABLEl.-Results of Sheep Erythrocyte Agglutinution and Inhibition Tests in Juvenile Rheumatoid Arthritis Negative Age group Children Adultst Total Number of Patients 24 7 31 Agglutination Positive Number Number Per cent 10 14 4 58 3 13 18 57 58 Negative Number lo 1# 2 Inhibition Positive Number Per cent 23 6 29 95.8 80. 93.5 OTiter in this patient, sick less than one year and tested only once, was 7 by agglutination and 14 by inhibition; a somewhat equivocal result. t Disease started in childhood but test performed after patients had reached adulthood. Atypical history in this patient. tination and six by inhibition. Considering children and adults together, more than half were positive by agglutination and 93.5 per cent by inhibition, It should be noted also that the adult who was negative by the inhibition technic had originally had an attack of rheumatic fever with polyarthritiq, carditis and chorea at the age of eight. Several years later, however, the joint disease returned and this time gradually assumed a character which lead to the diagnosis of Still’s disease. The S.E.A. test was first performed when she was 26 years old, by which time her arthritis was chronic though mild. The one child recorded as negative in table 1 had been ill less than a year, and the single test performed gave a borderline negative result. Ankylosing Spondylitis The view that ankylosing spondylitis is rheumatoid arthritis involving the sacroiliac joints and spine, which is so widely held in the United States, is based chiefly on the relatively common occurrence of peripheral joint involvement like that of rheumatoid arthritis in many spondylitic patients,11J5-18 and on early reports of pathologic similarities in the two types of joint disea~e.3>*J1J0-1~ More recently, however, pathologic studies of bone and joint tissues obtained by biopsy at various stages in the development of ankylosing spondylitis have shown features which were considered distinct from those of rheumatoid arthritis.20-24In spite of the official acceptance of the term rheumatoid spondylitis by the American Rheumatism Association, there are many rheumatologists in the United States who doubt that the Marie-Striimpell type of ankylosing spondylitis is a variant of rheumatoid arthritis, and, as has been already noted, this doubt is very commonly held in Europe. In general there are three points of view regarding the relationship: first, that ankylosing spondylitis is a variant of rheumatoid art h r i t i ~ , ~ ~ l l . 1 6 - 1 8 , 2 6 - second, 86 that they are distinct disease and third, that “true” ankylosing spondylitis involving only sacroiliac joints and spine, with or without involvement of hips and shoulders, is a distinct disease, but that when more peripheral joints are affected the disease is rheumatoid arthritis with involvement of the spine. Favoring the view that ankylosing spondylitis is distinct from rheumatoid arthritis is the sharp difference in the frequency with which me ! and 484 MC EWEN, ZIFF, CARMEL, DI TATA A N D TANNER TABLE2.--Results of Sheep Erythrocyte Agglutination and lnhibition Tesfs in Ankylosing Spondylitis T y p e of case Number of Pntients Without peripheral joint involvement With peripheral joint involvement Atypical Total Negative Agglutination Positive Per cent Number Number 92 91 1 1.1 18 10' 18 10 119 0 0 1 0 0.8 120 0 Negative Number Inhibition Positive Number Per cent 91 1 1.1 18 10 119 0 0 1 0 0 0.8 'Seven of these patients had ulcerative colitis. women are affected by the two diseases; for ankylosing spondylitis affects approximately nine men for every woman, whereas rheumatoid arthritis occurs two or three times as often ill women as in men. Other important differences are the lack of subcutaneous nodules in ankylosing spondylitis, and the relative rarity with which classic rheumatoid arthritis affects sarroiliac joints and any part of the spine except that of the cervical region. Furthermore, even when involvement of other parts of the spine occurs in patients with rheumatoid arthritis, the sacroiliacs usually are unaffected, and there rarely is restriction of chest expansion."g A fourth significant difference is the commonly observed tendency for the hip joint in ankylosing spondylitis to become rapidly immobile again due to soft-tissue contracture following arthroplasty, in contrast to the sustained mobility after such operations in rheumatoid arthritis. Other distinguishing features which are somewhat less convincing but which are frequently cited are the differences in response to therapy. Chrysotherapy, for example, is rather generally considered to be of little if any value in ankylosing spondylitis, in contrast to its reported usefulness in rheumatoid arthritis. Conversely, x-ray therapy and phenylbutazone are widely regarded as helpful in ankylosing spondylitis, yet of no, or cornparatively little, value in rheumatoid arthritis. We have applied the sheep erythrocyte agglutination test in the study of 120 patients with ankylosing spondylitis, with the results shown in table 2. In 92 of these patients involvement was limited to the sacroiliac joints and spine with or without involvement of the shoulders or hips. In 18 there was, in addition, involvement of more peripheral joints. Ten patients were considered to have atypical spondylitis using the criteria of Sharp.53 As the table shows, only one patient gave a positive test, and this was a man with characteristic ankylosing spondylitis without involvement of peripheral joints. Psoriatic Arthritis Opinions with regard to the relationship between rheumatoid arthritis and arthritis accompanying psoriasis are, as in the case of ankylosing spondylitis, of three general types. According to one view arthritis with psoriasis is a special form of joint disease which is distinct from rheumatoid arthriti~.'~*"-~~ At the other extreme is the concept that, barring some patients with degenerative joint disease or other obviously coincidental arthritis, all such cases are examples of rheumatoid arthritis occurring in patients with RHEUMATOID ARTHRITIS AND ITS SO-CALLED VARIANTS TABLE3.-Results T y p e of diaeane of 485 Sheep Erythroqte Agglutination und lnhibition Tests in Arthritis Accompanying Psorhis Number of Patients Negative Number Agglutination Positive Number Per cent Negative Number Inhibition Positive Number Per cent With involvement of distal inter51 50 lo 2 51 lo 2 phalangeals Without involvement of distal interphalangeals 25 25 0 0 25 0 0 76 75 1 1.3 76 1 1.3 Total 'This is the same patient whose titer on a single occasion was 14 by agglutination and 7 by inhibition, an equivocal result. In view of the fact that joint disease occurs in patients with psoriasis too frequently to be explained by coincidence," this view carries with it the belief that psoriasis and rheumatoid arthritis, for some reason, tend to occur together. Thirdly, there is the compromise opinion that, whereas most cases are indeed examples of rheumatoid arthritis in patients who also have psoriasis, there is a separate entity with very definite features, which should be considered psoriatic arthritis or a r t h r ~ p a t h y . Among ~ ~ - ~ ~ the special features described as characteristic of the latter type of joint disease are: involvement of the distal interphalangeal joints of fingers and toes (which are rarely the site of rheumatoid arthritis), psoriasis of the nails, asymmetiic rather than symmetric involvement of joints, the peculiar type of bone destruction, the tendency for the arthritis and the skin lesions to flare and subside together, the greater frequency in men than in women, the comparatively poor response to corticosteroid therapy, the relative lack of osteoporosis and of muscular atrophy in many patients, the absence of subcutaneous nodules, and certain pathologic differences in the joint lesions, notably the lack of pannus and, in late stages, fibrous replacement of normal joint structures with only mild inflammatory rea~tion.'*"~*~~*~ The authors have had the opportunity to perform sheep erythrocyte agglutination and inhibition tests on sera from 76 patients with psoriasis and arthritis (table 3). In 51 of these patients there was "characteristic" involvement of the distal interphalangeal joints of the fingers and often of the toes, with psoriatic changes in the nails. Approximately half of these patients also had involvement of other joints as well. The remaining 25 patients had arthritis in various joints without involvement of the distal interphalangeals. As is shown in table 3, only one of these patients gave a positive S.E.A. test, and the result in this instance was equivocal. This patient had characteristic involvement of the distal interphalangeal joints. None of the patients without distal interphalangeal involvement was positive. Arthritis Accompanying Ulceratioe Colitis Although it has been reportedRgthat arthritis is the most common complication of ulcerative colitis outside the intestinal tract, comparatively little has been written about it. HenchRSfound 60 instances of arthritis among 486 M C EWEN, ZIFF, CARMEL, DI TATA AND TANNER TABLE&--Results of Sheep Erythrocyte Agglutination and Znhibition Tests in Arthritis Accompanying Ukeratit;e Cok% Type of disease Peripheral joint involvement Spondylitis Total Number of Patients 15 7 22 Negative Agglutination Positive Inhibition Positive Number Per cent Number Number Per cent Negative Number 15 0 0 0 15 0 0 7 7 22 0 0 0 0 0 0 0 22 1,500 patients with ulcerative colitis, a frequency of 4 per cent. In other series, figures up to 20 per cent have been reported, but most authors place The opinion of Short, the frequency at from 4 per cent to 7 per cent.Do-loU Bauer and Reynolds” that this form of joint disease is probably merely rheumatoid arthritis occurring in patients with ulcerative colitis is rather generally held. Hench!””and Ansell and Bywaters:’ on the other hand, have cited evidence supporting their view that this type of arthritis is distinct and a manifestation of ulcerative colitis. The liistologic lesions in synovial biopsies are similar to those of rheumatoid arthritis,llft’Ubut clinically the close temporal relationship between attacks of colitis and of arthritis, and the improvement in arthritis following colectomy suggest an etiologic relationship between the articular and intestinal lesions. In the present series 22 patients with ulcerative colitis and arthritis have been studied. In 15 the joints involved have been peripheral ones, and there has been close parallelism between episodes of colitis and arthritis. Seven patients, however, had quite characteristic ankylosing spondylitis. In six of these, colitis preceded arthritis by brief periods but in the other, a young woman, spondylitis had preceded the colitis by six years. As is shown in table 4, the S.E.A. tests were negative in all of these patients by both agglutination and inhibition methods. Reiter’s Disease During recent years there has been an increasing interest in the form of arthritis associated with “nonspecific urethritis” and conjunctivitis, to which attention was drawn by Fiessinger and Leroyl”l and by Reiter1(J2in 1916 and first reported in the American literature by Bauer and Engleman in 1942.1°” While this “triad has been emphasized, many patients also have diarrhea as an early symptom, and others present keratodermia blennorrhagica, balanitis or other lesions. It would be inappropriate for the purposes of this paper to discuss the various theories as to the nature of the disease, its relationship to dysenteric arthritis, or the validity of including under the heading of Reiter’s disease the relatively large number of cases in which either urethritis or conjunctivitis may be lacking.l0’112 On the other hand, it is appropriate to note that these cases are sometimes considered to be examples of rheumatoid arthritis. As was remarked at the start of this paper, fully developed Reiter’s disease is not held by many to be related to rheumatoid arthritis.l12 On the other hand, since the reference by Hench and Boland1I3 to cases of rheumatoid arthritis precipitated by acute genital gonorrhea ( “postgonor- 487 RHEUMATOID ARTHRITIS AND ITS SO-CALLED VARIANTS rhea1 rheumatoid arthritis”) some examples of ‘Xeiter’s disease” undoubtedlv are so diagnosed, especially when conjunctivitis is absent or goes unnoticed. For this reason patients with arthritis similar to that of Reiter’s disease following urethritis but without conjunctivitis were included in the series on whom sheep erythrocyte agglutination tests were done. Data on 41 patients are shown in table 5, of whom 21 presented the full picture of Reiter’s disease with or without diarrhea or keratodermia blennorrhagica, and 20 had wholly comparable arthritis following diarrhea or nongonococcal urethritis but lacked conjunctivitis and keratodermia. It will be seen that a positive S.E.A. test was obtained in only one patient, a man with the full picture of Reiter’s disease including keratodermia blennorrhagica. Summary of Restrlts of Sheep Erythrocyte Agglutination Tests Before summarizing the results of sheep erythrocyte agglutination tests in the various forms of arthritis discussed thus far, it will be useful to present our experience with this test in patients with rheumatoid arthritis and in controls. The data are shown in table 6. All patients with rheumatoid arthritis in this table had classic involvement of peripheral joints with or without involvement of the cervical spine but with no clinical or radiographic evidence of spondylitis elsewhere nor of sacroiliac damage. The controls included a large number of apparently normal people of various ages, as well as patients with degenerative joint diseases, gout, rheumatic fever and miscellaneous nonrheumatic diseases. TABLE 5.-Results of Number of Patients Manifestations Urethritis and arthritis Above two pliis eye involvement Above three plus keratodcrma blennorrhagica Total Sheep Erythrocyte Agglutiriation and Inhibition Tests in Reiter’s Disease Nemtive Agglutination Positive Inhibition Positive Number Per cent Number Number Per cent Negative Number 209 20 0 0 20 0 0 16 16 0 0 16 0 0 5 41 4 40 1 1 - 4 1 - 40 1 2.4 2.4 OOne of these patients had severe diarrhea and no urethritis. TABLE0.-Results of Sheep Eqthrocyte Agglutination and Inhfbition Tests in Rheumatoid Arthritis and Controls Groups tested ~~ ~ Rheumatoidarthritis Controls Whole Serum Agglutination Number of Patients % Positive - Euglobulin Fraction Agglutination Inhibition Niimter of Number of Patients Yo Positive Patients Positive ~~ 103 104 78 13 144 342 88 2 140 321 96.4 3.7 ‘Osteoarthritis, gout, rheumatic fever, miscellaneous nonrheumatic diseases and normal individuals. 388 MC EWEN, ZIFP, CARMEL, DI TATA AND TANNER TABLE7.--Resultu of Sheep Erythrocyte Agglutination Inhibition Tests Various Arthritides and Crintrnls Disease Rheumatoid arthritis Juvenile rheumatoid arthritis (children only) Ankylosing spondylitis Psoriatic arthritis Arthritis accompanying ulcerative colitis Reiter's disease Systemic lupus erythematosus Controls' iti Number of Patients Number Positive Per cent Positive 140 135 96.4 24 23 1 1 95.8 0 1 0.0 2.4 28.0 3.7 119 76 22 41 18 32 1 5 12 0.8 1.3 'Osteoarthritis, gout, rheumatic fever, miscellaneous nonrheumatic diseases and normal individuals. Whole serum was used for agglutination tests in 103 rheumatoid arthritic patients and 104 controls. The euglobulin fraction was tested by the agglutination method14 in 144 rheumatoid arthritic patients and in 342 controls, and by the inhibition methodI4 in 140 rheumatoid arthritic patients and in 321 controls. It will be seen in table 6 that the positive results increased progressively from 78 per cent using whole serum to 88 per cent using the euglubulin fraction for agglutination, and to 96 per cent using the euglobulin fraction for testing by the inhibition technic. Conversely, the number of what were taken to be false positive results in the control patients decreased appreciably from 13 per cent to 2 per cent when the euglobulin fraction was used instead of whole serum for agglutination, and rose merely to 3.7 per cent when the very highly sensitive inhibition technic was employed. The results shown in tables 1 to 6 are summarized in table 7. Here only the results obtained by the inhibition technic using the euglobulin fraction of the serum are shown, since this method was the most discriminatory. The sharp difference between the results in patients with classic rheumatoid arthritis and Still's disease on the one hand and those with ankylosing spondylitis, arthritis accompanying psoriasis, arthritis accompanying ulcerative colitis, and Reiter's disease on the other hand, is striking. Table 7 also includes results obtained in 18 patients with systemic lupus erythematosus. It is not the purpose of this paper to discuss this disease and, furthermore, the number of cases is not large; nevertheless, they are included in the table for comparison because this is the only group of patients in our series to date, save for those with straightforward adult rheumatoid arthritis and Still's disease, with a higher percentage of positive agglutination and inhibition tests than the controls. DISCUSSION It was mentioned at the beginning of this paper that one of the important reasons for listing ankylosing spondylitis as a variant of rheumatoid arthritis in the classification adopted by the American Rheumatism Association in 1941 was the pathologic similarity reported by a number of observers3~4~11,1"-'B RHEUMATOID ARTHRITIS AND ITS SO-CALLED VARIANTS 489 between the lesions in the latter disease and those in peripiheral and apophyseal joints of patients with spondylitis. More recently, however, a number of pathologic studies of ankylosing spondylitis have concluded that the gross and histologic characteristics are difierent from those of rheumatoid a r t h i t i s , 2 u 14 In any case, most pathologists who have made a special study of arthritis now emphasize the frequent inconclusiveness of histologic criteria in attempting to distinguish among various types of joint diseases."Ld.114-11T Thus, this argument for considering the diseases the same no longer has the weight it once did. Turning to a consideration of the clinical and radiologic features, one finds the differences between classic rheumatoid arthritis and the Marie-Striimpell type of ankylosing spondylitis to be striking. Indeed the only clinical reason for considering them the same would appear to be the peripheral joint involvement which occurs in some patients with ankylosing spondylitis. It is in this connection that the results of the sheep erythrocyte agglutin,'i t'ion tests are of particular interest, because they were negative in 119 of 120 patients with ankylosing spondylitis irrespective of the presence or absence of peripheral joint involvement. The situation with regard to the arthritis which accompanies psoriasis is similar. Both pathologically and clinically there is now strong evidence that at least those cases with characteristic involvement limited to distal interphalangeal joints represent a disease which is distinct from rheumatoid arthritis. As in the case of ankylosing spondylitis, the principal reason for considering this form of joint disease merely rheumatoid arthritis occurring in patients who also have psoriasis is the large number of cases in which involvement of some or all joints does closely resemble that of rheumatoid arthritis. Here again the results of the S.E.A. tests are of particular interest, because, with a single exception, they were consistently negative in the 76 patients studied, irrespective of the presence or absence of distal interphalangeal joint involvement. Philosophic consideration of the problem posed by patients with psoriasis and arthritis is consistent with these S.E.A. results. As has been pointed out by Short, Bauer and Reynolds, I the relationship between psoriasis and arthritis cannot be merely one of coincidence. Furthermore, there is an increasing agreement with the point of view of Hench,h:' Bauer, Bennett and ZellerH" and others, that the cases with distinctive involvement limited to distal interphalangeal joints do differ from rheumatoid arthritis. If, then, one accepts the existence of a separate arthritic entity in some patients with psoriasis, it does not appear logical to conclude that, nevertheless, the majority of cases of arthritis occurring in psoriatics represents merely an unexplained tendency for psoriasis and rheumatoid arthritis to occur in the same patients. Rather, it would be more probable that rheumatoid arthritis and psoriasis would occur in the same patient no more often than would be expected on the h x i s of chance. The number of p,itieiits with arthritis ilccoI1~p:ln~i1lgulcerative colitis reported here is rather small. Nevertheless, the consistently negative results warrant the suggestion that this type of joint disease, too, is distinct from 490 MCEWEN, ZIFF, CARMEL, DITATA AND TANNEA rheumatoid arthritis. Certainly this interesting form of arthritis has received much less attention than it deserves; further studies in a larger series of patients are in progress. The questions raised by Reiter’s disease are different, for the most part, from those discussed in relation to the other arthritides which are the subject of this paper because there is no large body of opinion linking it with rheumatoid arthritis. Of special interest are patients with indolent arthritis: clinically similar to that of typical Reiter’s disease (and consistent also with that of rheumatoid arthritis ) which follows urethritis, but which is not associated with conjunctivitis or keratodermia blennorrhagica and not responsive to antibiotic therapy even though gonococci may have been cclltured from the urethral exudate. The data presented indicate that the great majority of patients of this type do not give a positive sheep erythrocyte agglutination test even when done by the inhibition method. It is suggested that in this type of case the diagnosis of atypical Reiter’s disease be considered before one of “postgonorrheal rheumatoid arthritis” is made. The results of the S.E.A. tests in the patients with Still’s diseuse in this series are, of course, completely in keeping with the almost universally accepted view that this form of joint disease is rheumatoid arthritis occurring in children. The term “Still’s disease” was used throughout this pitper merely so as not to “beg the question” before the data were given. The evidence from all sources is overwhelming that the disease is truly juvenile rheumatoid arthritis, and the latter is the preferred term. Comment must be made on differences between results of S.E.A. tests repcrted here and those obtained by some other investigators. Other writers have reported positive results in from 0 to 61 per cent of children with juvenile rheumatoid arthritis,55J1*.122 ccmpared with the 96 per cent presented here. it will be noted, however (table l ) ,that the latter result was obtained using the highly sensitive inhibition test and the euglobulin fraction of serum, and with sensitized sheep erythrocytes as the agglutinable particle. Ccnversely, although most workers have obtained negative results in patients with ankylosing spondylitis and arthritis accompanying psoriasis comparable to the results in the present series, a few have reported positive tests in up to Again it must be pointed out that more than 50 per cent of cases.55J1H-122 different technics have been employed and that in the authors’ hands the use of the euglobulin fraction rather than whole serum for the test with sensitized sheep erythrocytes has been found to eliminate a considerable niimber of false positive results (table 6 ) . It must be emphasized that it would be untenable to make decisions regarding the interrelationships of these various joint diseases solely on the basis of a laboratory test, the full significance of which is not known. In the opinion of the ;iuthors, however, the results cf the sheep erythrocyte agslutination tests which havc: lwcn reported in this paper do have considerable weight when considered in relation to the available clinical, patholcgic and mcliologic evidence. One might reasonably question also the wisdom of attempting to separate these arthritides from rheumatoid arthritis since the latter is itself a somewhat ill-defined concept which has gradually emerged RHEUMATOID ARTHRITIS A N D ITS SO-CALLED VARIANTS 491 over more than a century of slowly sharpening clinical and pathologic understanding. The definition of rheumatoid arthritis is largely a clinical one, and probably must remain so until its cause and pathogenesis are known, or until a diagnostic test is developed which can be universally accepted as valid. Whether some technic based on the mechanism underlying the sheep erythrocyte agglutination test may prove to be usable in this way it is too early to predict. Meanwhile most physicians will agree that even on purely clinical grounds the concept of rheumatoid arthritis is sufficiently definite to establish it as an entity. In the opinion of the authors, that concept can and should be still further sharpened by separating from it ankylosing spondylitis, arthritis accompanying psoriasis and ulcerative colitis, and Reiter’s disease. Finally, it is of interest to point out that at just the time when confusion regarding the relationship of rheumatoid arthritis to its so-called variants appears to be approaching clarification, new confusion has emerged regarding its relationship to systemic lupus erythematosus, polyarteritis nodosa and others of the “collagen diseases.”123 SUMMARY AND CONCLUSIONS In this paper a number of joint diseases which have been thought to be special forms or variants of rheumatoid arthritis have been discussed, and the clinical, pathologic and radiologic features which support such a relationship and those which raise doubts have been briefly reviewed. Particular attention has been devoted to results of sheep erythrocyte agglutination tests done by the inhibition technic on the euglobulin fraction of sera to be tested. On the basis of these data it is concluded that Still’s disease is truly juvenile rheumatoid arthritis, but that ankylosing spondylitis, Reiter’s disease, arthritis accompanying psoriasis, and arthritis accompanying ulcerative colitis are distinct from rheumatoid arthritis. ACKNOWLEDGhlENTS The authors acknowledge with thanks the help given by the following physicians who contributed sera and data for some of the patients with psoriatic arthritis, ulcerative colitis and juvenile rheumatoid arthritis: Dr. Barbara Ansell, British Postgraduate Medical School, London; Dr. Alfred J. Bollet, Wayne University College of Medicine, Detroit; Drs. Joseph L. Hollander and Ralph A. Jessar, University of Pennsylvania College of Medicine, Philadelphia; Dr. Ann Kuttner, Department of Pediatrics, New York University College of Medicine, New York; Dr. Frank Schmid, Northwestern University College of Medicine, Chicago; Dr. John W. Sigler, Henry Ford Hospital, Detroit. REFERENCES 1. Hench, P. S. et al.: Rheumatism nnd arthritis-Review of American and English literature for 1940 (Eighth Rheumatism Review). Ann. Int. Med. 151002, 1941. 2. - et al.: Rheumatism and arthritisReview of American and English literature of recent years (Ninth Rheumatism Review). Ann. Int. Med. 28:68, 1948. 3. Allison, N. and Ghorniley, R. K.: Diagnosis in Joint Disease: a Clinical and Pathological Study of Arthritis. New York, William Wood, 1931. 4. Collins, D. H.: The Pathology of Articular and Spinal Diseases. London, Edward Arnold & Co., 1949. 492 MCEWEN, ZIFF, CARMEL, DITATA AND TANNER 5. Still, G. F.: On a form of chronic joint disease in children. Med. Chir. l'r. S0:47. published by the Royal Mcdical and Chirurgical Society of London, 1897. b. Tnussig, A. E.: Still's disease with hypcrglobulin. J. Lab. & Clin. hled. 23'3833, 1938. 7. Hench, P. S. et al.: Rheumatism and arthritis-Review of American and English literature for 1938 (Sixth Rheumatism Review). Ann. Int. Med. 13~1655and 1837, 1940. (Editorial comment, p. 1878.) 8. Isdale, I. C. and Bywaters, E. G. L.: Thc rash of rheumatoid arthritis and Still's disease. Quart. J. Med. 25: 377, 1956. 9. Potter, T. A., Barkin, R. E. and Stillman, J. s.: Occurrence of spondylitis in juvenile rheumatoid arthritis. Ann. Rheumat. Dis. 13:364, 1954. 10. Bywaters, E. G . L., Glynn, L. E. and Zeldis, A.: Subciitaneous nodiilc of Still's disease. Ann. Rheumat. Dis. 1958. 11. Short, C. L., Bailer, W. and Reynolds, W. E.: Rheumatoid Arthritis: A Definition of the Disease and a Clinical Description Based on a Nuincrical Study of 293 Patients and Controls. Cambridge, Mass., Harvard University Press, 1957. 12. l'ortis, R. B.: Pathology of chronic arthritis in Children. Am. J. Dis. Child. 55:1OCO, 1938. 13. Angevine, I>. M.: Rheumatoid arthritis in children. Clinics 1:582, 1943. 14. Ziff, hi., Brown, P., Lospahto, J., Badin, J. and McEwcn, C.: Agglutination and inhibition by serum globrilin in the sensitized sheep cell agglutination reaction in rheumatoid arhtritis. Am. J. Med. 20:500, 19-50. 15. Boland, E. W. and Present, A. J.: Rheumatoid spondylitis. J.A.M.A. m a 4 3 , 194s. 16. Polley, H. F.: A Study of 1035 Cases of Rheumatoid Arthritis. Thesis, Graduate School, University of Minnesota, 1945. 17. Baucr, W., Warren, C. F., hlote, J. R. and Jones, T. D.: Streptococcus immunological stiidies in patients with rheumatoid arthritis and comparable 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. control groups. Trans. A. Am. Phys. 51:106, 1936. Tyson, T. L.: Spondylitis ankylopoietica. Med. Clin. N. America, 21: 1755, 1937. Freund, E.: A contribution to the pathogenesis of spondylitis ankylopoietica. Edinburgh Med. J. 49~91, 1942. Cruickshank, B.: Histopathology of diarthrodial joints in ankylosing spondylitis. Ann. Rheumat. Dis. 10: 393, 1951. Van Swaay, H.: Spondylosis Ankylopoietica, een pathogentische Studie, Thesis, Leiden, 1950. Engfeldt, B., Romanus, R. and Y d h , S.: Histological studies of pelvospondylitis ossificans ( ankylosing spondylitis ) correlated with clinical and radiological findings. Ann. Rheumat. Dis. 13:219, 1954. Gibson, H. J.: Ankylosing spondylitis, aetiology and pathology, J. Fac. Radiol. 5:193, 1957. Forestier, J., Jacqueline, F. and RotesQuerol, J.: Ankylosing Spondylitis, Clinical Considerations, Roentgenology, Pathologic Anatomy, Treatment. Springfield, Ill., Charles C Thomas, 1951. Hench, P. S. et al.: Rheumatism and arthritis-Review of American and English literature for 1933 (First Rheumatism Review). Ann. Int. Med. 8:1495, 1935, (Editorial comment, p. 1547). - et al.: Rheumatism and arthritisReview of American and English literature for 1935 (Third Rheumatism Review). Ann. Int. Med., 10: 754, 1936 (Editorial comment, p. 853). HEFEHENCE 1: Editorial comment on p. 1061. Cominittec of the American Rheumatism Association: Primer on Rheumatic Diseases. J.A.M.A. 152:323; 405; 552, 1953. Spangler, D.: Osteoarthritis of the spine. Texas State J. Med. 31:709, 1936. Taylor, G. D., Ferguson, A. B., Kasbach, Haig and Dawson, M. H.: Roentgenologic observations on var- RHEUMATOID ARTHRITIS AND ITS SO-CALLED VARIANTS 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. ious types of chronic arthritis. A.M.A. Arch. Int. bled. 42:979, 1936. Oppenheimer, A.: Diseases affecting the intervertebral foramina. Radiology 283582, 1937. Edstrom, G.: Is spondylitis ankylopoietica an independ-nt disease or a rheumatic syndrome? Acta med. scand. 104:396, 1940. Hnre, H. F.: The diagnosis of MarieStrunipell arthritis with certain aspects of treatment. New England J. Mcd. 223:702, 1940. Hench, P. S., Slocumb, C. H. and Polley, H. F.: Rheumatoid spondylitis; Questions and answers. Med. Clin. N. America, 31:879, 1947. Lennon, W. and Chalmers, I. S.: Ankylosing spondylitis. Lancet 1:12, 1948. Buckley, C. W.: Ankylosing spondylitis. I n Reports on Chronic Rheumatic Diseases, London, H. K. Lewis & Co. Ltd., Number 1, 1935, p. 77. Golding, F. C.: Spondylitis ankylopoietica. Brit. J. Surg., 23:484, 1936. Gordon, R. G.: The metabolic factor in chronic rheumatism with special reference to fibrositis. Brit. M. J. 2:1213, 1936. Crave, H. W.: Chronic rheumatic arthritis, Med. Press 194:426, 1937. Scott, S. G.: Spondylitis adolescens with associated pathological changes in the sacro-iliac joints. Charterhouse Rheum. Clinic, Original Papers 1937, Oxford University Press, 1:169. van Dam, G.: Radiography in Rheumatism, in a Survey of Chronic Rheumatic Diseases. London, Oxford University Press, 1938, pp. 241. Osgood, R. B.: T h e medical and social approaches to the problem of chronic rheumatism. Am. J. Med. Sc. 200: 429, 1940. Baker, L. D.: Marie-Strumpell arthritis and the undiagnosed low back patient. Nebraska Med. J. 33:331, 1948. Law, W. A.: Surgery in treatment of anklyosing spondylitis. Proc. Roy. SOC.Med. 41~251,1948. Hart, F. D., Robinson, K. C., Allchin, F. M. and Maclagan, N. F.: Anky- 493 losing spondylitis. Quart. J. Med. 18:217, 1949. 46. Mowbray, R., Latner, A. L., Allchin, F. M. and Maclagan, N. F.: Ankylosing spondylitis. Quart. J. Med. 18:187, 1949. 47. Williams, A. J.: Rheumatoid (MarieStriimpell) spondylitis technique of examination and importance of the costal joints. California Med. 70:257, 1949. 48. Copeman, W. S. C. and Mason, R. M.: The medical treatment of chronic b a c k a c h c . Med. Press. 228:292, 1952. 49. Hart, F. D.: Cortisone and ACTH in treatment of ankylosing spondylitis. Brit. Med. J. 1:188, 1952. 50. Parr, L. J. A., White, P. and Shipton, E.: Some observations on 100 cases of ankylosing spondylitis. Med. J. Australia. 1:544, 1951. 51. Turney, H. F.: Ankylosing spondylitis. Proc. Roy. SOC.Med. 45:57, 1952. 52. Kellgren, J. H.: Some concepts of rheumatic disease. Brit. Med. J. 1: 1093 and 1152, 1952. 53. Sharp, J.: Differential diagnosis of ankylosing spondylitis. Brit. Med. J. 1:975, 1957. 54. Mason, R. M., Murray, R. S., Oates, J. K. and Young, A. C.: Prostatitis and ankylosing spondylitis. Brit. Med. J. 1:748, 1958. 55. Ziff, M.: The agglutination reaction in rheumatoid arthritis. J. Chronic Dis. 5:644, 1957. 56. Zellner, E. : Arthropathia psoriatica und Arthritis bei Psoriatibern. Wiener Arch.f.Innere Med. 15435, 1928. 57. Bauer, J. and Vogl, A.: Psoriasis und Gelenkleiden, Zugleich ein Beitrag zur Kenntnis des hereditiiren Hydrops articulorum intermittens. Klin. Wchnschr. 10:1700, 1931. 58. Jeghers, H. and Robinson, L. J.: Arthropathia psoriatica. J.A.M.A. 108: 949, 1937. 59. Leczinsky, C. G.: The incidence of arthropathy in a ten-year series of psoriasis cases. Acta derm. vener. 28:483, 1948. 60. Sherman, M. S.: Psoriatic arthritis, observations on the clinical, roentgenographic and pathological changes. 491 hlCEWEN, ZIFF, CARMEL, DITATA AND TANNER J. Bone &Joint Surg. 34-A:831, 1952. 61. Fischel: Inaug. Dissertat, Berlin 1897. As cited by Balzer and Burnier in Bull. Soc. franc. de dermat. et syph. 223182, 1911. 62. Garrod, A. and Evans, G.: Arthropathia psoriatica. Quart. J. Med. 17:171, 1923-24. 63. Civatte, A.: Psoriasis and seborrhoeic eczema: Pathological anatomy and diagnostic histology of the two dermatoses. Brit. J. Derm. 36:461, 1921. 04. Hunt, E.: Psoriasis and rheumatism: Comparison. Lancet 2351, 1933. 65. Boots, R. H.: Rheumatoid (atrophic) arthritis and other diseases with joint manifestations, Int. Clin. 3:154, 1935. 66. Dawson, M. H.: Chronic Arthritis. Nelson’s Loose-Leaf Medicine, 5: 605, N.Y., Thos. Nelson and Sons, 1935. 67. Guzman, 1.: Pathology and symptoms of arthropnthic psoriasis: Question of common etiology. Orvosi hetil. 79:1331 1935. 68. Nordin, G.: Fatal case of psoriasis arthropathica. Acta derm. vener. 15: 221, 1934. 69. Comroe, B. I. and Hollander, J. L., Eds.: Arthritis and Allied Conditions, Philadelphia, Lea & Febiger, 1949, Ch. 11, p. 156. 70. Dawson, M. H.: A note on the occurrence of psoriasis in rheumatoid arthritis. Med. Clh. N. America. 21: 1807, 1937. 71. - and Tyson, T. I,.: Psoriasis arthropathica with observations on certain features common to psoriasis and rheumatoid arthritis. Trans. A. Am. Phys. 53:303, 1938. 72. Hunt, E.: Psori,isis and rheumatism. Proc.Roy.Soc.Med. 31 308, 1938. 73. Bauer, W.: The diagnosis of the various arthritides. New England J. Med. 221 5 3 4 , 1939. 74. Weissenbach, R. J. et Bouwens, G.: Etude radiologique du rhumatisine psoriatique. Ann. de Derm:it et Syph., 1:5, 1941. 75. Romanus, T.: Psoriasis from a prognostic and hereditary point of view. Acta derm. vener. 26(12):1, 1945. 76. Pinol-Aguade, J., Barcelo, P. and Rotes- 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 58. 89. 90. Querol, J.: Poliartritis ch6nica y psoriasis. Rev. Esp. Rheum. y enferm. ost.-artic. 2:373, 1948. Cecil, R. L.: Psoriatic arthritis. Chicago Med. SOC.Bull. 51:747, 1949. Wassniann, K.: Rheumatoid arthritis and psoriasis; Statistical statements. Ann. Rheumat. Dis. 8:70, 1949. Lewis-Faning, E.: Report on an enquiry into the aetiological factors associated with rheumatoid arthritis. Ann. Rheumat. Dis. 9, suppl. 94 pp.. 1950. Vilanova, X. and Pinol, J.: Psoriasis arthropathica. Rhcumatism. 7:197, 1951. Gribble, M. de G.: Rheumatoid arthritis and psoriasis. Ann. Rheumat. Dis. 14:198, 1955. Lievre, J. A. and Breuzard, J.: Le Problhme du Rhumatisme Psoriasique. Rev. du Rhumatisme. 23:549, 1956. Hench, P. S.: Acute and Chronic Arthritis. Nelson’s Loose-Leaf Living Surgery. New York, Thomas Nelson and Sons, pp. 104-175, 1935. Lane, C. G. and Crawford, G. M.: Psoriasis: Statistical Study of 231 Cases. A.M.A. Arch. Derm. & Syph. 351051, 1937. Barber, H. W.: The skin manifestations in rheumatism. Proc. Roy. SOC. Med. 31:701, 1938. Bauer, W., Bennett, G. A. and Zeller, J. W.: The pathology of joint lesions in patients with psoriasis and arthritis. Trans. A. Am. Phys. 56:349, 1941. Wright, V.: Psoriasis and arthritis, A study of the radiographic appearances. Brit. J. Radio]. 30:113, 1957. Wilens, S.: Personal communication. Bargen, J. A., Jackman, R. J. and Kerr, J. G.: Studies on the life histories of patients with chronic ulcerative colitis ( thrombo-ulcerative colitis ) with some suggestions for treatment. Ann. Int. Med. 12:339, 1938. McKittrick, L. S. and Miller, R. H.: Idiopathic ulcerative colitis: Review of 149 cases, with particular reference to the value of, and indications for, surgical treatment. Ann, Surg. 102:656, 1935, RHEUMATOID ARTHRJTIS AND ITS SO-CALLED VARIANTS 91. Jackman, R. J., Bargen, J. A. and Helmholz, H. F.: Life histories of ninety-fivc children with chronic ulcerative colitis. Am. J. Dis. Child. 593459, 1940. 92. Jankelson, I. R., McClnre, C. W. and Sweetsir, F. N.: Chronic ulcerative colitis; Complications outside the digestive tract. Rev. Gastroenterol. 9:99, 1942. 93. Ricketts, W. E. and Palmer, W. I,.: Complications of chronic non-specific ulcerative colitis. Gastroenterol. 7: 55, 1946. 94. Sloan, W. P., Bargen, A. J. and Gage, R. P.: Life histories of patients with chronic ulcerative colitis: A review of 2,000 cases. Gastroenterol. 16:25, 1950. 95. Crohn, B. B., Garlock, J. H. and Yarnis, H.: Right sided (regional) colitis. J.A.M.A. 134:334, 1947. 96. Rice-Oxley, J. M. and Truelove, S.: Complications of ulcerative colitis. Lancet 1:607, 1950. Ulrerative colitis, course and prognosis. Lancet 1: 863, 1950. 97. Brown, M. L., Kasich, A. M. and Weingartcn, B.: Complications of chronic ulcerative colitis. Am. J. Dig. Dis. 18:52, 1951. 98. Brooke, B. N.: Outcome of surgery for ulcerative colitis. Lancet 2532, 1956. 99. Bywaters, E. G. L. and Ansell, B. M.: Arthritis associated with ulcerative colitis; A clinical and pathological study. Ann. Rheumat. Dis. 17:169, 1958. 100. Bacon, H. E.: Ulcerative colitis, Philadelphia, F. B. Lippincott Co., 1958. 101. Fiessingcr, M. N. and Leroy, h4. E.: Contribution B 1’Etude d’une Bpidemie de hysenterie dans La somme (juillet-octobre 1918). SOC. Med. des Hopitaux de Paris, Bulletins et Memoires, 3:2032, 1916-17. 102. Reiter, €I.: Uber ein bisher unerkannte Spirochaetinfektion ( Spirochaetosis Arthriticn ) Deutsrh Med. Wchnschr. 42:1535, 1916. 103. Bauer, W. and Engleman, E. P.: A syndrome of unknown etiology characterised by urethritis, conjunctivitis and arthritis (so-called Reiter’s dis- . 495 ease). Trans. A. Am. Phys. 57:307, 1942. 104. Hollander, J. L., Fogarty, C. W., Abrams, N. R. and Kydd, D. M.: Arthritis resembling Reiter’s syndrome. J.A.M.A. 1293593, 1945. 105. Young, R. H. and McEwen, E. G.: Bacillary dysentery as a cause of Reiter’s syndrome. J.A.M.A. 134: 1456, 1947. 108. Paronen, I.: Reiter’s Disease: A study of 344 cases observed in Finland. Acta med. scand. (Suppl. 212) 131: 1, 1948. 107. McEwen, C.: Unpublished data. 108. Marche, J.: Le maladie rhumatismale consbcutive aux infections entbrourinaires. Giornale di hlalattie Infettive e Parassitarie, Vol. 9, 1957. 109. Ford, D. K.: Human tissue culture studies of non-gonococcal urethritis. Brit. J. Ven. Dis. 32(3):184, 1956. 110. Reiter, H.: Reiter’s disease. German Med. Monthly (English edition of Deutsch Med. Wchnschr.) 2346, 1957. 111. Ford, D. K.: Reiter’s syndrome. Bull. Rheumat. Dis. 8:159, 1958. 112. Littler, T. R.: The changing pattern of r h e u m a t o i d arthritis. Ann. Rheumat. Dis. 10:405, 1951. 113. Hench, P. S. and Boland, W.: The management of chronic arthritis and other rheumatic diseases among soldiers of the Unitecl States Army. Ann. Int. Med. 26:808, 1948. 114. Sherman, M. S.: The non-specificity of synovial reactions. Bull. Hosp. Joint Dis. 12110, 1951. 115. Sokoloff, L.: Personal Communication. 116. Cooper, N.: Personal communication. 117. Collins, D. H.: Personal communication. 118. Alexander, R. and de Forest, G. K.: The sensitized sheep cell agglutination reaction in rheumatoid arthritis. Am. J. Med. 16:191, 1954. 119. Shichikawa, K., Asano, N., Orihara, M., Mayeda, A., Tamatani, Y. and Yasuda, J.: Studies on the WaalerRose hcmagglutination test. Acta rheum. scand. 2:34, 1958. 120. Jacobson, A. S., Kammerer, W. H., Wolf, J., Epstein, W. V. and Heller, 496 MCEWEN, ZIFF, CARMEL, DITATA AND TANNER G.: The hemagglutination test for rheumatoid arthritis. 111. Clinical evaluation of the sheep erythrocyte agglutination (S.E.A.) test and the gamma globulin ( F u ) tests. Am. J. Med. 20:490, 1956. 121. Ball, J.: Sheep cell agglutination for rheumatoid arthritis. Ann. Rheuniat. Dis. 11: 1952. 122. Hall, A. P., Mednis, A. D. and Bayles, J. B.: The latex agglutination and inhibition reactions-clinical experience in the diagnosis of rheumatoid arthritis. New England J. Med. 258: 731, 1958. 123. hfcEwen, C.: Current status of the socalled collagen diseases. Maryland State Med. J., In press. Currier McEwen, M.D., D.Sc. (Hon.), Associate Professor of Medicine and Chairman, Rheumatic Diseases Study Group, New York University College of Medicine, New York, N. Y . Morris Zifl, M.D., Ph.D., Professor of Medicine, Department of Internal Medicine, University of Texas Southwestern Medical School, Dallas, Tex. Philip D. Carmel, M . D., Attending Rheamatologist, Lutheran Medical Center, Brooklyn; Attending Physician, Veterans’ Out-Patient Depnrtment, Veterans’ Hospital, New York; Physician in Chnrge, Arthritis Clinic, Cone9 Island Hospital, Brooklyn, N. Y . Domenick DiTata, M.D., Chief, Arthritis Clinic, Medical Division, New York Regional Office, Veterans Administration, New York, N . Y. Martin Tanner, Senior Research Assistant, Department of Medicine, New York University College of Medicine, New York, N. Y.