Chronic Lyme Arthritis Clinical and Immunogenetic Differentiation from Rheumatoid Arthritis ALLEN C. STEERE, M.D.; ALLAN GIBOFSKY, M.D.; MANUEL E. PATARROYO, M.D.; ROBERT J. WINCHESTER, M.D.; JOHN A. HARDIN, M.D.; and STEPHEN E. MALAWISTA, M.D.; New Haven, Connecticut; and New York, New York
Ten patients with Lyme arthritis have developed chronic involvement of one or both knees. Lyme arthritis was diagnosed by onset with erythema chronicum migrans (six patients); residence in Lyme, Connecticut (eight); seasonal onset in summer and early fall (nine); early periods of short recurrent attacks (nine); absence of rheumatoid factor (nine); and absence of symmetrical polyarthritis, morning stiffness, subcutaneous nodules, or antinuclear antibodies (in all). Five patients had synovectomies; pannus formation and underlying cartilage erosion were present in all. Seven of the 10 patients had the same B-cell alloantigen, DRw2 (frequency in normal control subjects, 2 2 % [ P < 0.005]), but did not have an increased frequency of the alloantigens associated with rheumatoid arthritis. Chronic Lyme arthritis, the result of an apparent tick-transmitted infection, resembles rheumatoid arthritis pathologically but generally differs from it in both prearticular and immunogenetic characteristics. L Y M E ARTHRITIS is characterized by brief but recurrent
attacks of asymmetric, oligoarticular swelling and pain affecting primarily large joints, or by periods of migratory polyarthritis in both large and small joints (1-3). The illness was first recognized as new because of close geographic clustering of cases. It then became apparent that a preceding annular skin lesion, erythema chronicum migrans, is the best clinical marker for the illness (2, 3) and that neurologic (4) or cardiac conduction abnormalities may occur. There is strong epidemiologic evidence for transmission of an infectious agent by a tick vector, specifically, by Ixodes scapularis (1-3, 5, 6). We report here 10 cases of Lyme arthritis in patients whose joint involvement in knees has lost its typical intermittent character and become chronic. In this study, "chronic Lyme arthritis" is differentiated from rheumatoid arthritis by both clinical (prearticular) and immunogenetic characteristics. Materials and Methods
Our case definitions for erythema chronicum migrans and Lyme arthritis have been reported previously (2, 4). Chronic Lyme arthritis was defined clinically by unremitting joint swelling and pain for at least 1 year, or pathologically by pannus formation and underlying cartilage erosion. In addition to tests outlined previously for prospective study of patients with all forms of the disease (2), those with chronic Lyme arthritis had roentgenograms of affected joints taken yearly and complete immunogenetic profiles done. Alleles of the HLA-A, B, and C loci were identified by the • From the Departments of Internal Medicine, Epidemiology, and Public Health, Yale University School of Medicine, N e w Haven, Connecticut; and The Rockefeller University, N e w York, N e w York.
896
standard lymphocyte cytotoxicity test. B-cell alloantisera used in this study for the determination of DR specificities had been previously assayed for reactivity against a panel of 40 B-cell lymphoblastoid lines derived from persons homozygous for HLA-D alleles (7-9). The B-cell alloantigens of each patient were identified by a modified Amos two-stage cytotoxic activity on purified B cells and by indirect immunofluorescence of mononuclear cells cultured 6 days with pokeweed mitogen. At least five alloantisera were used to test for each allele; a positive test was defined by a reaction in both assays with at least four of five antisera. Results
Ten of 102 patients (10%) who had onset of Lyme arthritis from 1974 through 1976 developed chronic involvement in one or both knees (Table 1). Their ages ranged from 9 to 57 years (median, 32 years), eight were male and two female, and eight were from Lyme, Connecticut. During the summer months before onset of arthritis, six had erythema chronicum migrans, sometimes accompanied by fatigue, fever, headache, stiff neck, myalgias, backache, or sore throat. Two others had such symptoms during the summer months without noting the skin lesion. When headache and stiff neck were prominent, symptoms were suggestive of aseptic meningitis, and two of the patients later developed meningoencephalitis and cranial nerve palsies. From 5 days to 2 months after onset of erythema chronicum migrans, five of the six patients had either several brief but recurrent attacks of oligoarticular arthritis or periods of migratory polyarthritis often without manifest swelling, patterns typical of other patients with Lyme arthritis. The four patients without the skin lesion had similar brief attacks. After 4 to 24 months of intermittent arthritis, nine patients developed chronic involvement in one or both knees; arthritis in the remaining patient began with such involvement. Although the degree of swelling fluctuated (strenuous physical activity generally seemed to aggravate it), the knees were often intensely swollen, out of proportion to the pain, and hypertrophied synovium was frequently palpable. Three patients developed popliteal cysts that ruptured; dissection of synovial fluid into the calf simulated thrombophlebitis. Chronic knee involvement was usually not accompanied by other signs or symptoms; a few patients still had occasional fever, fatigue, or continued lymphadenopathy. One subject (Patient J) later in her course had severe low back pain, particularly in the morning. (This patient was HLA-B27 negative and did not have roentgenographs abnormalities of the lumbosacral spine or sacroiliac joints
Annals of Internal Medicine 90:896-901, 1979
Downloaded From: https://annals.org/ by a University of Kentucky User on 08/24/2018
© 1979 American College of Physicians
Figure 1 Left. The knee of Patient C (see Tables 1 and 2) at the time of synovectomy. The patient had erythema chronicum migrans in June 1 9 7 4 followed by intermittent migratory polyarthritis. Two years later, he developed persistent swelling of the right knee unresponsive to medical therapy. After 7 months, he underwent a synovectomy. Parts of the articular surfaces of the femur and patella can be seen, the latter having been pulled laterally and inverted. Arrows indicate what was the advancing border of pannus, now stripped away f r o m the underlying eroded cartilage. A similar eroded area can be seen along the margin of the patella. Right. The tissue of the pannus of Patient C. There is marked synovial proliferation, vascularization, and patchy perivascular infiltration (upper right) with lymphocytes and plasma cells (lower right). (Hematoxylin and eosin stain; original magnification, approximately x l O O [upper right] and x 4 0 0 [lower right].)
or reflex changes in the lower extremities.) N o patient had symmetrical polyarthritis, morning stiffness, subcutaneous nodules, conjunctivitis, or urethritis. When chronic knee involvement was present, five patients had elevated erythrocyte sedimentation rates, but only one had a positive test for rheumatoid factor, in low titer (1:160), and none had antinuclear antibodies (Table 2). Joint fluid leukocyte counts ranged from 500 to 76 800 cells/mm 3 with predominantly polymorphonuclear leukocytes; total protein levels were elevated, and C3 levels were one half to two thirds of serum levels. Cryoglobulins were present in all joint fluids but in only three sera. Only one subject (Patient E) had roentgenographic evidence of bony erosions (in the femoral notch, confirmed at surgery) 1 year after onset of persistent involvement. Seven of the 10 patients had the same B-cell alloantigen, D R w 2 , compared with 2 2 % of a normal population (10) ( P < 0.005) (Table 2). Another patient reacted reproducibly with two of the five alloantisera used for typing D R w 2 . In contrast, the frequency of other H L A - A , B, C, or D alleles was not significantly different from the normal population. Four of 10 subjects (Patients D , E, G, and H) had DRw4-7-10 (normal frequency, 3 1 % [9]), and only one (Patient I) had HLA-B27. The major functional impairment in these patients was difficulty in walking, which in some cases interfered with work and in others, with life-style. Five patients with marked functional difficulty had anterior synovectomies 6 to 18 months after onset of chronic knee involvement. At surgery, a thickened, inflamed synovium was seen in
all, with pannus formation and underlying cartilage erosion (Figure 1). Histologically, the synovium was hypertrophic and hyperplastic (six to 10 cells thick), with focal necrosis and surface deposits of fibrin and polymorphonuclear leukocytes (2). Beneath the synovial cells, there was vascular proliferation and a marked infiltration of plasma cells and small lymphocytes, sometimes with focal nodular clusters suggestive of early lymphoid follicles. After surgery, four of the five patients required manipulation of adhesions under anesthesia followed by 1 to 6 months of physical therapy. However, in all five, the operated knees have remained uninflamed since surgery, a period of 4 to 32 months. Three of the patients have subsequently had transient pain in other joints. Of the five patients who did not have surgery, three had remission of chronic involvement within 12 to 16 months of its onset, but the remaining two (Patients F and I) continue to have swelling and pain, now more than 2 years after onset. Discussion
Ten patients with Lyme arthritis developed chronic involvement of one or both knees, usually after periods of intermittent arthritis. Arthritis of such persistence blurs the major feature of joint involvement in typical Lyme arthritis—its usual brevity—that allows one to distinguish it from other rheumatoid diseases. Clinical differentiation then rests primarily on associated findings—erythema chronicum migrans, neurologic abnormalities, or residence in an endemic area—that may be absent in an individual patient. In children, chronic Lyme arthritis Steereetal. • Chronic Lyme Arthritis
Downloaded From: https://annals.org/ by a University of Kentucky User on 08/24/2018
897
Table 1. Chronic Lynie Arthritis: C linical Chanicteristics Patient
Age
Sex
Residence in Lyme Area
Early Events Month of Onset
ECM* Nervous System at Onset Involvement
Brief Attacks of Arthritis Number
Joints Affected
yr A B C D E F G H I
14 41 29 57 28 35 9 14 43
M M M M M M M F M
+ + + + + + + +
Sep Sep Jun Apr Jul Jun Jul Jul Jun
J Total Median (range)
37
F
—
Jun
—
8
_ —
+
— -
— — — —
+ + + +
+ +
+6
—
—
2 1 1 0 3 6
Shoulder, wrist, elbow Wrist, elbow Migratory polyarthritis Knee Migratory polyarthritis Shoulder, both knees Migratory polyarthritis Shoulder, elbow, both knees, temporomandibular Migratory polyarthritis
2
1.5(0-6)
* ECM = erythema chronicum migrans.
closely resembles the pauciarticular form of juvenile rheumatoid arthritis, except that iridocyclitis or antinuclear antibodies, or both, which occur in 20% to 25% of those with pauciarticular juvenile rheumatoid arthritis (11-13), are not present in Lyme arthritis. Of adult disorders, joint involvement in chronic Lyme arthritis has similarities to the rheumatoid variants, particularly Reiter's syndrome, a disease that primarily affects weight-bearing joints in adult men (14). At least in this sample, chronic Lyme arthritis affected knees and was also more common in men; other patients of ours have had conjunctivitis (but associated with erythema chronicum migrans, not with arthritis); and one of the current patients, a woman, developed severe low back pain, without roentgenographic abnormalities, that appeared to be inflammatory in nature. However, the definitive features of Reiter's syndrome—urethritis, conjunctivitis or iritis, and other mucocutaneous lesions—are not features of chronic Lyme arthritis. Neither are the characteristics of classic rheumatoid arthritis: symmetrical polyarthritis, morning stiffness, subcutaneous nodules, and a positive test for rheumatoid factor. However, the chronic Lyme arthritis does meet the requirements for possible rheumatoid arthritis (15), a diagnosis of exclusion. Differentiation may then depend upon appreciation of the prearticular signs and symptoms. Joint abnormalities of these entities are remarkably similar (2, 14, 16). Joint fluid leukocyte counts usually range between 10 000 and 25 000 cells/mm 3 (counts may range from 500 to 100 000 cells/mm 3 ) with predominantly polymorphonuclear leukocytes; total protein is elevated (2, 17). In both Lyme and rheumatoid arthritis, cryoglobulins are uniformly present in joint fluid but uncommonly in serum (18-20). However, compared with serum, joint fluid complement levels tend to be lower in rheumatoid arthritis than in Lyme arthritis or Reiter's syndrome (21, 22). In all three disorders, histologically there is ear898
ly hypertrophy of synovial lining cells, vascular proliferation, and a predominantly mononuclear cell infiltrate (2, 14, 16). Later, the synovium becomes thickened, scarred, and has focal nodular clusters of lymphocytes (lymphoid follicles). Chronically inflamed, adherent synovium (pannus) is associated with erosion of underlying cartilage (Figure 1). Thus, in an individual patient, joint abnormalities are nonspecific; they reflect synovial inflammation of variable intensity and duration. Because of the increased frequency of particular alleles of the major histocompatibility complex in certain rheumatic diseases, we suspected that immunogenetic factors may also be important in chronic Lyme arthritis. In Reiter's syndrome, the increase is in HLA-B27 (23, 24); in classic rheumatoid arthritis, in HLA-Dw4 (25) (or in DRw4-7-10 if identified by B-cell alloantisera [8, 9, 26]); and in systemic lupus erythematosus, in DRw2 and DRw3 (8, 9, 27). We now know that at least seven of the 10 patients with chronic Lyme arthritis have the B-cell alloantigen DRw2, compared with 22% of a normal population (P < 0.005). The frequency of this alloantigen is also increased in multiple sclerosis (28), a disease that has similar neurologic abnormalities to those of certain patients with Lyme disease (4). The only patient with chronic Lyme arthritis who had a positive test for rheumatoid factor had DRw4-7-10, the alloantigen associated with rheumatoid arthritis. Although he did not have erythema chronicum migrans, his course—intermittent and then chronic involvement of one knee—and exposure history in Lyme, Connecticut, support the diagnosis of Lyme arthritis rather than rheumatoid arthritis. The increased frequency of DRw2 in chronic Lyme arthritis has two major implications: Patients must have distinct immunogenetic characteristics to develop chronic Lyme arthritis, and those characteristics are different from the ones associated with rheumatoid arthritis and Reiter's syndrome. We do not yet know whether DRw2 is
June 1979 • Annals of Internal Medicine • Volume 90 • Number 6
Downloaded From: https://annals.org/ by a University of Kentucky User on 08/24/2018
Table 1. (Continued) Patient
Localized Chronic Joint Disease Interval from Onset of Disease to Onset of Chronic Involvement
Duration of Chronic Involvement
mos
mos
12 4 24 12 20 11 12 4 18
9 8 7 18 12 24 12 16 26
Knee Knee Knee Knee Knee Knee Both knees Both knees Both knees
5
13
Both knees
12 (4-24)
Synovectomy
Joints Affected
+ + + + + — — —
A B C D E F G H I
5
J Total Median (range)
12.5 (7-26)
associated only with chronicity or also with other manifestations of Lyme arthritis. Because of marked functional difficulties and lack of response to medical therapy, five of our 10 patients had synovectomies. Although the follow-up period is still short (4 to 32 months), the patients seem to have benefited from the procedure. They have returned to work and recreational activities without recurrence of synovitis. On the other hand, we know little about the natural history of chronic Lyme arthritis. Three of the five patients who did not have surgery had remission of their knee involvement within 12 to 16 months. Two patients who continue to have chronic involvement (Patients F and I), now for more than 2 years, have engaged intermittently in strenu-
ous physical activity that may have affected their course. In rheumatoid arthritis, early synovectomy of knees may reduce pain and swelling, but it is questionable whether the procedure prevents subsequent damage to joints (2931). Thus, we currently reserve synovectomy in Lyme arthritis for the patient with marked functional difficulty (such as inability to work) whose arthritis remains unresponsive to medical therapy—salicylates, crutch walking, and intra-articular injection of corticosteroid esters—for at least 6 to 12 months. Chronic Lyme arthritis appears to fit within a pathogenetic framework suspected to be important in other rheumatic diseases. In rheumatoid arthritis, systemic lupus erythematosus, and Reiter's syndrome, infectious
Table 2. Crironic Lyme Arthri tis: Laboratory 1Findings Patient
Blood* ESR
RF
mm/h A B C D E F G H
It J Total positive Median (range)
4 27 10 54 40 20 33 30 12 20
C3
62 99 94 281 98 133 113 221 70 73
i 24 (4-54)
Cryoglobulin
Cryoglobulin
mg/dL
— — — 1:160 — — — — — —
106(62-281)
B-Cell AlloamUgens
Joint Fluid
— — + — — + + — —
+ + + + + + + + + +
3
10
C3
Leukocyte Count
Total protein
mg/dL
Xl0 3 /mm 3
g/dL
25 50 44 158 58 60 55 124 37 34
5.3 11.6 17.8 76.8 7.7 5.9 40.3 3.8 0.5 31.7
5.6 5.7 4.5 5.7 4.0 4.3 4.6 3.8 3.6 6.5
Ial
— + — — — — — — + 2
53(25-158)
8.7 (0.5-76.8)
Ia2
Ia3 Ia4- Ia5 7-10
+ (+)t + + + + — + +
+ — + — — — + —
_ — + + — + + — —
_ +
7
3
4
1
— — — —
4.6 (3.6-5.7)
* ESR = erythrocyte sedimentation rate; RF = rheumatoid factor. t This patient's serum reacted with only two Ia2 typing sera; all the other Ia2-positive sera reacted with at least four of five. % Only patient positive for HLA-B27. Steere et al. • Chronic Lyme Arthritis
Downloaded From: https://annals.org/ by a University of Kentucky User on 08/24/2018
899
agents or endogenous immune stimulation may trigger, in genetically susceptible persons, a disordered or inappropriate immune response that leads to chronicity (3239). Evidence for an infectious cause is best in Reiter's syndrome; patients, primarily those with HLA-B27, may develop the syndrome after certain sexual exposures or diarrheal illnesses (37, 38), including a large outbreak of shigellosis in which there was clustering of cases (39). However, there are no reports of participation of the immune system in the pathogenesis of Reiter's syndrome. Lyme arthritis is thought to be caused by an infectious agent transmitted by Ixodes scapularis ticks (5, 6). By a characteristic expanding skin lesion, erythema chronicum migrans, patients may often be identified 1 to 3 weeks after exposure and before the onset of arthritis. When the skin lesion is present, most patients have circulating immune complexes (40, 41). At that time, those with high serum IgM levels, cryoglobulins containing IgM, and low IgG levels are the ones at risk of developing arthritis within months; those with high IgG levels usually do not (18, 42). Later, immune complexes and cryoglobulins are found uniformly in joint fluid but uncommonly in serum (18, 41). At least in patients with DRw2, arthritis in knees may become chronic, a process that may involve pannus formation and cartilage erosion. Thus, in genetically susceptible persons, a tick-transmitted agent appears to trigger an inappropriate immune response that may localize and persist in joints, with pathologic consequences similar to those seen in other rheumatoid diseases.
cum migrans and Lyme arthritis: field study of ticks. Am J Epidemiol 108:322-327, 1978 7. WINCHESTER RJ, F U SM, W E R N E T P, K U N K E L HG, D U P O N T B, JER-
SILD C: Recognition by pregnancy serums of non-HL-A alloantigens selectively expressed on B lymphocytes. J Exp Med 141:924-929, 1975 8. GIBOFSKY A, WINCHESTER RJ, H A N S E N J, PATARROYO M, D U P O N T B, PAGET S, LAHITA R, HALPER J, FOTINO M, Y U N I S E, K U N K E L HG:
Contrasting patterns of newer histocompatibility determinants in patients with rheumatoid arthritis and systemic lupus erythematosus. Arthritis Rheum 21(suppl):134-138, 1978 9. GIBOFSKY A, WINCHESTER RJ, PATARROYO M, FOTINO M, K U N K E L
HG: Disease associations of the la-like human alloantigens. Contrasting patterns in rheumatoid arthritis and systemic lupus erythematosus. / Exp Med 148:1728-1732, 1978 10. D U P O N T B, Y U N I S EJ, DUQUESNOY R, POLLACK M, N O R E E N H, H A N SEN JA, REINSMOEN N, A N N E N K, GREENBERG L, L E E TD, W H I T -
SETT C, ANTONELLI P, BRAUN D: HLA-D and la alloantigen in North American Caucasians, in Histocompatibility Testing 1977, edited by BODMER WF, BATCHELOR JR, BODMER JG, FESTENSTEIN F, MORRIS
PJ. Copenhagen, Munksgaard, 1978, p. 603 11. CALABRO J J, HOLGERSON WB, SONPAL GM, K H O U R Y MI: Juvenile
rheumatoid arthritis: a general review and report of 100 patients observed for 15 years. Semin Arthritis Rheum 5:257-298, 1976 12. PETTY RE, STORM PB, CASSIDY JT, B U R T A, SULLIVAN DB: Immuno-
logic correlates of antinuclear antibody in juvenile rheumatoid arthritis (abstract). Arthritis Rheum 12:323-324, 1969 13. CALABRO JJ, PARRINO GR, ATCHOO PD, MARCHESANO JM, G O L D -
BERG LS: Chronic iridocyclitis in juvenile rheumatoid arthritis. Arthritis Rheum 13:406-413, 1970 14. WEINBERGER HW, ROPES MW, K U L K A JP, B A U E R W: Reiter's syn-
drome, clinical and pathologic observations. A long term study of 16 cases. Medicine (Baltimore) 41:35-91, 1962 15. ROPES MW, B E N N E T T GA, COBB S, JACOX R, JESSAR RA: 1958 revi-
sion of diagnostic criteria for rheumatoid arthritis. Bull Rheum Dis 9:175-176, 1958 16. SOKOLOFF L: The pathology of rheumatoid arthritis and allied disorders, in Arthritis and Allied Conditions: A Textbook of Rheumatology, 8th ed., edited by HOLLANDER JL, MCCARTY DJ JR. Philadelphia, Lea & Febiger, 1972, pp. 309-311 17. JESSAR RA: The study of synovial fluid. See Reference 16, p. 72 18. STEERE AC, H A R D I N JA, R U D D Y S, M U M M A W JG, M A L A W I S T A SE:
Lyme arthritis: correlation of serum and cryoglobulin IgM with activity, and serum IgG with remission. Arthritis Rheum, May 1979, in press
ACKNOWLEDGMENTS: The authors thank the physicians in the surveillance system, Drs. James A. Albright, David N. Romond, and Henry Drinker for doing the synovectomies and Dr. Nicholas H. Bartenhagen for helping with patient care; Dr. Marilena Fotino for identifying HLA antigens in certain patients; Ms. Stella Cretella for expert laboratory assistance; and Ms. Elise DeSanna for preparation of the manuscript. These investigations were supported in part by U.S. Public Health Service grants AM-20358, AM-10493, AM-07107, AM-5639, RR-05443, RR-00125, A1-00216, and CA-20107; the Arthritis Foundation and its Connecticut Chapter; the Kroc Foundation of Santa Ynez, California; and the Colombian National Science Foundation (Col Sciencias 140-3-37-77). Parts of this work have appeared in abstract form in Clin Res 27:338A,
KANTOR GL, MCINTOSH RM: Cryoglobulinemia and disease. Ann Intern Med 73:95-107, 1970 20. MARCUS RL, TOWNES AS: The occurrence of cryoproteins in synovial fluid; the association of a complement-fixing activity in rheumatoid synovial fluid and cold-precipitable protein. / Clin Invest 50:282-293, 1971 21. PEKIN TJ JR, ZVAIFLER NJ: Hemolytic complement in synovial fluid. / Clin Invest 43:1372-1382, 1964 22. FOSTIROPOULOS G, AUSTEN KF, BLOCH KJ: Total hemolytic complement (CH50) and second component of complement (C'2ku) activity in serum and synovial fluid. Arthritis Rheum 8:219-230, 1965
April 1979 (STEERE AC, GIBOFSKY A, WINCHESTER RJ, MALAWISTA, SE:
23. BREWERTON DA, CAFFREY M, NICHOLLS A, WALTERS D , OATES JK,
Persistent Lyme arthritis with pannus formation and cartilage erosion: clinical and immunogenetic differentiation from rheumatoid arthritis).
24. MORRIS R, METZGER AL, BLUESTONE R, TERASAKI PI: HL-A W27—
• Requests for reprints should be addressed to Allen C. Steere, M.D.; Department of Internal Medicine, Yale University School of Medicine, 333 Cedar Street; New Haven, CT 06510. Received 29 September 1978; revision accepted 8 March 1979.
19. BARNETT EV, BLUESTONE R, CRACCHIOLO A III, GOLDBERG LS,
JAMES DCO: Reiter's disease and HL-A 27. Lancet 2:996-998, 1973 a clue to the diagnosis and pathogenesis of Reiter's syndrome. AT Engl J Med 290:554-556, 1974 25. STASTNY P: Mixed lymphocyte culture typing cells from patients with rheumatoid arthritis. Tissue Antigens 4:571-579, 1974 26. STASTNY P: Association of the B-cell alloantigen DRw4 with rheumatoid arthritis. N Engl J Med 298:869-871, 1978 27.
References 1. STEERE AC, M A L A W I S T A SE, S N Y D M A N DR, SHOPE RE, A N D I M A N
WA, Ross MR, STEELE FM: Lyme arthritis: an epidemic of oligoarticular arthritis in children and adults in three Connecticut communities. Arthritis Rheum 20:7-17, 1977 2. STEERE AC, MALAWISTA SE, H A R D I N JA, R U D D Y S, A S K E N A S E PW,
ANDIMAN WA: Erythema chronicum migrans and Lyme arthritis. The enlarging clinical spectrum. Ann Intern Med 86:685-698, 1977 3. STEERE AC, HARDIN JA, MALAWISTA SE: Lyme arthritis: related problems recently recognized in Connecticut. Conn Med 42:353-357, 1978 4. REIK L, STEERE AC, BARTENHAGEN NH, SHOPE RE, M A L A W I S T A
REINERTSEN JL, KLIPPEL JH, JOHNSON AH, STEINBERG A D , D E C K -
ER JL, M A N N DL: B-lymphocyte alloantigens associated with systemic lupus erythematosus. N Engl J Med 299:515-518, 1978 28. WINCHESTER RJ, EBERS G, F U SM, ESPINOSA L, ZABRISKIE J, K U N K -
EL HG: B-cell alloantigen Ag 7a in multiple sclerosis (letter). Lancet 2:814, 1975 29. M C E W E N C, O'BRIAN WB: A multicenter evaluation of early synovectomy in the treatment of rheumatoid arthritis (abstract). / Rheumatol l(suppl 1):107, 1974 30. THOMPSON M, DOUGLAS G, D A V I S O N EP: Synovectomy of the meta-
carpophalangeal joints in rheumatoid arthritis. Proc R Soc Med 66:197199, 1973 31. ARTHRITIS A N D RHEUMATISM COUNCIL A N D BRITISH ORTHOPAEDIC
migrans and Lyme arthritis: epidemiologic evidence for a tick vector. Am J Epidemiol 108:312-321, 1978
ASSOCIATION: Controlled trial of synovectomy of knee and metacarpophalangeal joints in rheumatoid arthritis. Ann Rheum Dis 35:437-442, 1976 32. CHRISTIAN CL: Rheumatoid arthritis—etiologic considerations. Arthritis Rheum 7:455-466, 1964
6. WALLIS RC, B R O W N SE, KLOTER KO, M A I N AJ JR: Erythema chroni-
33. WILKES RM, SIMSARIAN JP, HOPPS HE, ROTH H, DECKER JL, A P T E -
SE: Neurologic abnormalities of Lyme disease. Medicine July 1979, in press
(Baltimore),
5. STEERE AC, B RODE RICK TF, MALAWISTA SE: Erythema chronicum
9 0 0
June 1979
• Annals of Internal Medicine • Volume 90 • Number 6
Downloaded From: https://annals.org/ by a University of Kentucky User on 08/24/2018
34. 35. 36. 37. 38.
KAR RG, M E Y E R H M J R : Virologic studies on rheumatoid arthritis. Arthritis Rheum 16:446-454, 1973 Z I F F M: Viruses and the connective tissue diseases. Ann Intern Med 75:951-958, 1971 FERGUSON RH, W O R T H I N G T O N JW: Recent advances in rheumatic diseases: 1967 through 1969. Ann Intern Med 73:109-124, 1970 PHILLIPS PE: The virus hypothesis in systemic lupus erythematosus. Ann Intern Med 83:709-715, 1975 F O R D DK: The etiology of non-gonococcal urethritis and Reiter's syndrome. Excerpta Med Int Congr Ser 165:227-229, 1969
39. C A L I N A, FRIES JF: An "experimental" epidemic of Reiter's syndrome revisited. Follow-up evidence on genetic and environmental factors. Ann Intern Med 84:564-566, 1976
A H O K, A H V O N E N P, A L K I O P, LASSUS A, S A I R A N E N E, SIEVERS K,
42.
40.
H A R D I N JA, W A L K E R LC, S T E E R E AC, T R U M B L E TC, T U N G KSK, W I L L I A M S R C J R , R U D D Y S, M A L A W I S T A SE: Circulating immune
complexes in Lyme arthritis: detection by 125 I-Clq binding, C l q solid phase and Raji cell assays. / Clin Invest 63:468-477,1979 41. H A R D I N J A, S T E E R E AC, C A R B O N I J M , SPAGNOLA A M , S P I E L E R PN,
M A L A W I S T A SE: Lyme arthritis: immune complexes correlate with stage and type of disease (abstract). Arthritis Rheum 21:564, 1978
TIILIKAINEN A: HL-A 27 in reactive arthritis following infection. Ann Rheum Dis 34(suppl l):29-30, 1975
Annals of Internal Medicine 90:901-904, 1979
Downloaded From: https://annals.org/ by a University of Kentucky User on 08/24/2018
S T E E R E AC, H A R D I N JA, M A L A W I S T A SE: Erythema chronicum mi-
grans and Lyme arthritis: cry ©immunoglobulins and clinical activity of skin and joints. Science 196:1121-1122, 1977
901
Ten patients with Lyme arthritis have developed chronic involvement of one or both knees. Lyme arthritis was diagnosed by onset with erythema chronicum migrans (six patients); residence in Lyme, Connecticut (eight); seasonal onset in summer and early fall (nine); early periods of short recurrent attacks (nine); absence of rheumatoid factor (nine); and absence of symmetrical polyarthritis, morning stiffness, subcutaneous nodules, or antinuclear antibodies (in all). Five patients had synovectomies; pannus formation and underlying cartilage erosion were present in all. Seven of the 10 patients had the same B-cell alloantigen, DRw2 (frequency in normal control subjects, 2 2 % [ P < 0.005]), but did not have an increased frequency of the alloantigens associated with rheumatoid arthritis. Chronic Lyme arthritis, the result of an apparent tick-transmitted infection, resembles rheumatoid arthritis pathologically but generally differs from it in both prearticular and immunogenetic characteristics. L Y M E ARTHRITIS is characterized by brief but recurrent
attacks of asymmetric, oligoarticular swelling and pain affecting primarily large joints, or by periods of migratory polyarthritis in both large and small joints (1-3). The illness was first recognized as new because of close geographic clustering of cases. It then became apparent that a preceding annular skin lesion, erythema chronicum migrans, is the best clinical marker for the illness (2, 3) and that neurologic (4) or cardiac conduction abnormalities may occur. There is strong epidemiologic evidence for transmission of an infectious agent by a tick vector, specifically, by Ixodes scapularis (1-3, 5, 6). We report here 10 cases of Lyme arthritis in patients whose joint involvement in knees has lost its typical intermittent character and become chronic. In this study, "chronic Lyme arthritis" is differentiated from rheumatoid arthritis by both clinical (prearticular) and immunogenetic characteristics. Materials and Methods
Our case definitions for erythema chronicum migrans and Lyme arthritis have been reported previously (2, 4). Chronic Lyme arthritis was defined clinically by unremitting joint swelling and pain for at least 1 year, or pathologically by pannus formation and underlying cartilage erosion. In addition to tests outlined previously for prospective study of patients with all forms of the disease (2), those with chronic Lyme arthritis had roentgenograms of affected joints taken yearly and complete immunogenetic profiles done. Alleles of the HLA-A, B, and C loci were identified by the • From the Departments of Internal Medicine, Epidemiology, and Public Health, Yale University School of Medicine, N e w Haven, Connecticut; and The Rockefeller University, N e w York, N e w York.
896
standard lymphocyte cytotoxicity test. B-cell alloantisera used in this study for the determination of DR specificities had been previously assayed for reactivity against a panel of 40 B-cell lymphoblastoid lines derived from persons homozygous for HLA-D alleles (7-9). The B-cell alloantigens of each patient were identified by a modified Amos two-stage cytotoxic activity on purified B cells and by indirect immunofluorescence of mononuclear cells cultured 6 days with pokeweed mitogen. At least five alloantisera were used to test for each allele; a positive test was defined by a reaction in both assays with at least four of five antisera. Results
Ten of 102 patients (10%) who had onset of Lyme arthritis from 1974 through 1976 developed chronic involvement in one or both knees (Table 1). Their ages ranged from 9 to 57 years (median, 32 years), eight were male and two female, and eight were from Lyme, Connecticut. During the summer months before onset of arthritis, six had erythema chronicum migrans, sometimes accompanied by fatigue, fever, headache, stiff neck, myalgias, backache, or sore throat. Two others had such symptoms during the summer months without noting the skin lesion. When headache and stiff neck were prominent, symptoms were suggestive of aseptic meningitis, and two of the patients later developed meningoencephalitis and cranial nerve palsies. From 5 days to 2 months after onset of erythema chronicum migrans, five of the six patients had either several brief but recurrent attacks of oligoarticular arthritis or periods of migratory polyarthritis often without manifest swelling, patterns typical of other patients with Lyme arthritis. The four patients without the skin lesion had similar brief attacks. After 4 to 24 months of intermittent arthritis, nine patients developed chronic involvement in one or both knees; arthritis in the remaining patient began with such involvement. Although the degree of swelling fluctuated (strenuous physical activity generally seemed to aggravate it), the knees were often intensely swollen, out of proportion to the pain, and hypertrophied synovium was frequently palpable. Three patients developed popliteal cysts that ruptured; dissection of synovial fluid into the calf simulated thrombophlebitis. Chronic knee involvement was usually not accompanied by other signs or symptoms; a few patients still had occasional fever, fatigue, or continued lymphadenopathy. One subject (Patient J) later in her course had severe low back pain, particularly in the morning. (This patient was HLA-B27 negative and did not have roentgenographs abnormalities of the lumbosacral spine or sacroiliac joints
Annals of Internal Medicine 90:896-901, 1979
Downloaded From: https://annals.org/ by a University of Kentucky User on 08/24/2018
© 1979 American College of Physicians
Figure 1 Left. The knee of Patient C (see Tables 1 and 2) at the time of synovectomy. The patient had erythema chronicum migrans in June 1 9 7 4 followed by intermittent migratory polyarthritis. Two years later, he developed persistent swelling of the right knee unresponsive to medical therapy. After 7 months, he underwent a synovectomy. Parts of the articular surfaces of the femur and patella can be seen, the latter having been pulled laterally and inverted. Arrows indicate what was the advancing border of pannus, now stripped away f r o m the underlying eroded cartilage. A similar eroded area can be seen along the margin of the patella. Right. The tissue of the pannus of Patient C. There is marked synovial proliferation, vascularization, and patchy perivascular infiltration (upper right) with lymphocytes and plasma cells (lower right). (Hematoxylin and eosin stain; original magnification, approximately x l O O [upper right] and x 4 0 0 [lower right].)
or reflex changes in the lower extremities.) N o patient had symmetrical polyarthritis, morning stiffness, subcutaneous nodules, conjunctivitis, or urethritis. When chronic knee involvement was present, five patients had elevated erythrocyte sedimentation rates, but only one had a positive test for rheumatoid factor, in low titer (1:160), and none had antinuclear antibodies (Table 2). Joint fluid leukocyte counts ranged from 500 to 76 800 cells/mm 3 with predominantly polymorphonuclear leukocytes; total protein levels were elevated, and C3 levels were one half to two thirds of serum levels. Cryoglobulins were present in all joint fluids but in only three sera. Only one subject (Patient E) had roentgenographic evidence of bony erosions (in the femoral notch, confirmed at surgery) 1 year after onset of persistent involvement. Seven of the 10 patients had the same B-cell alloantigen, D R w 2 , compared with 2 2 % of a normal population (10) ( P < 0.005) (Table 2). Another patient reacted reproducibly with two of the five alloantisera used for typing D R w 2 . In contrast, the frequency of other H L A - A , B, C, or D alleles was not significantly different from the normal population. Four of 10 subjects (Patients D , E, G, and H) had DRw4-7-10 (normal frequency, 3 1 % [9]), and only one (Patient I) had HLA-B27. The major functional impairment in these patients was difficulty in walking, which in some cases interfered with work and in others, with life-style. Five patients with marked functional difficulty had anterior synovectomies 6 to 18 months after onset of chronic knee involvement. At surgery, a thickened, inflamed synovium was seen in
all, with pannus formation and underlying cartilage erosion (Figure 1). Histologically, the synovium was hypertrophic and hyperplastic (six to 10 cells thick), with focal necrosis and surface deposits of fibrin and polymorphonuclear leukocytes (2). Beneath the synovial cells, there was vascular proliferation and a marked infiltration of plasma cells and small lymphocytes, sometimes with focal nodular clusters suggestive of early lymphoid follicles. After surgery, four of the five patients required manipulation of adhesions under anesthesia followed by 1 to 6 months of physical therapy. However, in all five, the operated knees have remained uninflamed since surgery, a period of 4 to 32 months. Three of the patients have subsequently had transient pain in other joints. Of the five patients who did not have surgery, three had remission of chronic involvement within 12 to 16 months of its onset, but the remaining two (Patients F and I) continue to have swelling and pain, now more than 2 years after onset. Discussion
Ten patients with Lyme arthritis developed chronic involvement of one or both knees, usually after periods of intermittent arthritis. Arthritis of such persistence blurs the major feature of joint involvement in typical Lyme arthritis—its usual brevity—that allows one to distinguish it from other rheumatoid diseases. Clinical differentiation then rests primarily on associated findings—erythema chronicum migrans, neurologic abnormalities, or residence in an endemic area—that may be absent in an individual patient. In children, chronic Lyme arthritis Steereetal. • Chronic Lyme Arthritis
Downloaded From: https://annals.org/ by a University of Kentucky User on 08/24/2018
897
Table 1. Chronic Lynie Arthritis: C linical Chanicteristics Patient
Age
Sex
Residence in Lyme Area
Early Events Month of Onset
ECM* Nervous System at Onset Involvement
Brief Attacks of Arthritis Number
Joints Affected
yr A B C D E F G H I
14 41 29 57 28 35 9 14 43
M M M M M M M F M
+ + + + + + + +
Sep Sep Jun Apr Jul Jun Jul Jul Jun
J Total Median (range)
37
F
—
Jun
—
8
_ —
+
— -
— — — —
+ + + +
+ +
+6
—
—
2 1 1 0 3 6
Shoulder, wrist, elbow Wrist, elbow Migratory polyarthritis Knee Migratory polyarthritis Shoulder, both knees Migratory polyarthritis Shoulder, elbow, both knees, temporomandibular Migratory polyarthritis
2
1.5(0-6)
* ECM = erythema chronicum migrans.
closely resembles the pauciarticular form of juvenile rheumatoid arthritis, except that iridocyclitis or antinuclear antibodies, or both, which occur in 20% to 25% of those with pauciarticular juvenile rheumatoid arthritis (11-13), are not present in Lyme arthritis. Of adult disorders, joint involvement in chronic Lyme arthritis has similarities to the rheumatoid variants, particularly Reiter's syndrome, a disease that primarily affects weight-bearing joints in adult men (14). At least in this sample, chronic Lyme arthritis affected knees and was also more common in men; other patients of ours have had conjunctivitis (but associated with erythema chronicum migrans, not with arthritis); and one of the current patients, a woman, developed severe low back pain, without roentgenographic abnormalities, that appeared to be inflammatory in nature. However, the definitive features of Reiter's syndrome—urethritis, conjunctivitis or iritis, and other mucocutaneous lesions—are not features of chronic Lyme arthritis. Neither are the characteristics of classic rheumatoid arthritis: symmetrical polyarthritis, morning stiffness, subcutaneous nodules, and a positive test for rheumatoid factor. However, the chronic Lyme arthritis does meet the requirements for possible rheumatoid arthritis (15), a diagnosis of exclusion. Differentiation may then depend upon appreciation of the prearticular signs and symptoms. Joint abnormalities of these entities are remarkably similar (2, 14, 16). Joint fluid leukocyte counts usually range between 10 000 and 25 000 cells/mm 3 (counts may range from 500 to 100 000 cells/mm 3 ) with predominantly polymorphonuclear leukocytes; total protein is elevated (2, 17). In both Lyme and rheumatoid arthritis, cryoglobulins are uniformly present in joint fluid but uncommonly in serum (18-20). However, compared with serum, joint fluid complement levels tend to be lower in rheumatoid arthritis than in Lyme arthritis or Reiter's syndrome (21, 22). In all three disorders, histologically there is ear898
ly hypertrophy of synovial lining cells, vascular proliferation, and a predominantly mononuclear cell infiltrate (2, 14, 16). Later, the synovium becomes thickened, scarred, and has focal nodular clusters of lymphocytes (lymphoid follicles). Chronically inflamed, adherent synovium (pannus) is associated with erosion of underlying cartilage (Figure 1). Thus, in an individual patient, joint abnormalities are nonspecific; they reflect synovial inflammation of variable intensity and duration. Because of the increased frequency of particular alleles of the major histocompatibility complex in certain rheumatic diseases, we suspected that immunogenetic factors may also be important in chronic Lyme arthritis. In Reiter's syndrome, the increase is in HLA-B27 (23, 24); in classic rheumatoid arthritis, in HLA-Dw4 (25) (or in DRw4-7-10 if identified by B-cell alloantisera [8, 9, 26]); and in systemic lupus erythematosus, in DRw2 and DRw3 (8, 9, 27). We now know that at least seven of the 10 patients with chronic Lyme arthritis have the B-cell alloantigen DRw2, compared with 22% of a normal population (P < 0.005). The frequency of this alloantigen is also increased in multiple sclerosis (28), a disease that has similar neurologic abnormalities to those of certain patients with Lyme disease (4). The only patient with chronic Lyme arthritis who had a positive test for rheumatoid factor had DRw4-7-10, the alloantigen associated with rheumatoid arthritis. Although he did not have erythema chronicum migrans, his course—intermittent and then chronic involvement of one knee—and exposure history in Lyme, Connecticut, support the diagnosis of Lyme arthritis rather than rheumatoid arthritis. The increased frequency of DRw2 in chronic Lyme arthritis has two major implications: Patients must have distinct immunogenetic characteristics to develop chronic Lyme arthritis, and those characteristics are different from the ones associated with rheumatoid arthritis and Reiter's syndrome. We do not yet know whether DRw2 is
June 1979 • Annals of Internal Medicine • Volume 90 • Number 6
Downloaded From: https://annals.org/ by a University of Kentucky User on 08/24/2018
Table 1. (Continued) Patient
Localized Chronic Joint Disease Interval from Onset of Disease to Onset of Chronic Involvement
Duration of Chronic Involvement
mos
mos
12 4 24 12 20 11 12 4 18
9 8 7 18 12 24 12 16 26
Knee Knee Knee Knee Knee Knee Both knees Both knees Both knees
5
13
Both knees
12 (4-24)
Synovectomy
Joints Affected
+ + + + + — — —
A B C D E F G H I
5
J Total Median (range)
12.5 (7-26)
associated only with chronicity or also with other manifestations of Lyme arthritis. Because of marked functional difficulties and lack of response to medical therapy, five of our 10 patients had synovectomies. Although the follow-up period is still short (4 to 32 months), the patients seem to have benefited from the procedure. They have returned to work and recreational activities without recurrence of synovitis. On the other hand, we know little about the natural history of chronic Lyme arthritis. Three of the five patients who did not have surgery had remission of their knee involvement within 12 to 16 months. Two patients who continue to have chronic involvement (Patients F and I), now for more than 2 years, have engaged intermittently in strenu-
ous physical activity that may have affected their course. In rheumatoid arthritis, early synovectomy of knees may reduce pain and swelling, but it is questionable whether the procedure prevents subsequent damage to joints (2931). Thus, we currently reserve synovectomy in Lyme arthritis for the patient with marked functional difficulty (such as inability to work) whose arthritis remains unresponsive to medical therapy—salicylates, crutch walking, and intra-articular injection of corticosteroid esters—for at least 6 to 12 months. Chronic Lyme arthritis appears to fit within a pathogenetic framework suspected to be important in other rheumatic diseases. In rheumatoid arthritis, systemic lupus erythematosus, and Reiter's syndrome, infectious
Table 2. Crironic Lyme Arthri tis: Laboratory 1Findings Patient
Blood* ESR
RF
mm/h A B C D E F G H
It J Total positive Median (range)
4 27 10 54 40 20 33 30 12 20
C3
62 99 94 281 98 133 113 221 70 73
i 24 (4-54)
Cryoglobulin
Cryoglobulin
mg/dL
— — — 1:160 — — — — — —
106(62-281)
B-Cell AlloamUgens
Joint Fluid
— — + — — + + — —
+ + + + + + + + + +
3
10
C3
Leukocyte Count
Total protein
mg/dL
Xl0 3 /mm 3
g/dL
25 50 44 158 58 60 55 124 37 34
5.3 11.6 17.8 76.8 7.7 5.9 40.3 3.8 0.5 31.7
5.6 5.7 4.5 5.7 4.0 4.3 4.6 3.8 3.6 6.5
Ial
— + — — — — — — + 2
53(25-158)
8.7 (0.5-76.8)
Ia2
Ia3 Ia4- Ia5 7-10
+ (+)t + + + + — + +
+ — + — — — + —
_ — + + — + + — —
_ +
7
3
4
1
— — — —
4.6 (3.6-5.7)
* ESR = erythrocyte sedimentation rate; RF = rheumatoid factor. t This patient's serum reacted with only two Ia2 typing sera; all the other Ia2-positive sera reacted with at least four of five. % Only patient positive for HLA-B27. Steere et al. • Chronic Lyme Arthritis
Downloaded From: https://annals.org/ by a University of Kentucky User on 08/24/2018
899
agents or endogenous immune stimulation may trigger, in genetically susceptible persons, a disordered or inappropriate immune response that leads to chronicity (3239). Evidence for an infectious cause is best in Reiter's syndrome; patients, primarily those with HLA-B27, may develop the syndrome after certain sexual exposures or diarrheal illnesses (37, 38), including a large outbreak of shigellosis in which there was clustering of cases (39). However, there are no reports of participation of the immune system in the pathogenesis of Reiter's syndrome. Lyme arthritis is thought to be caused by an infectious agent transmitted by Ixodes scapularis ticks (5, 6). By a characteristic expanding skin lesion, erythema chronicum migrans, patients may often be identified 1 to 3 weeks after exposure and before the onset of arthritis. When the skin lesion is present, most patients have circulating immune complexes (40, 41). At that time, those with high serum IgM levels, cryoglobulins containing IgM, and low IgG levels are the ones at risk of developing arthritis within months; those with high IgG levels usually do not (18, 42). Later, immune complexes and cryoglobulins are found uniformly in joint fluid but uncommonly in serum (18, 41). At least in patients with DRw2, arthritis in knees may become chronic, a process that may involve pannus formation and cartilage erosion. Thus, in genetically susceptible persons, a tick-transmitted agent appears to trigger an inappropriate immune response that may localize and persist in joints, with pathologic consequences similar to those seen in other rheumatoid diseases.
cum migrans and Lyme arthritis: field study of ticks. Am J Epidemiol 108:322-327, 1978 7. WINCHESTER RJ, F U SM, W E R N E T P, K U N K E L HG, D U P O N T B, JER-
SILD C: Recognition by pregnancy serums of non-HL-A alloantigens selectively expressed on B lymphocytes. J Exp Med 141:924-929, 1975 8. GIBOFSKY A, WINCHESTER RJ, H A N S E N J, PATARROYO M, D U P O N T B, PAGET S, LAHITA R, HALPER J, FOTINO M, Y U N I S E, K U N K E L HG:
Contrasting patterns of newer histocompatibility determinants in patients with rheumatoid arthritis and systemic lupus erythematosus. Arthritis Rheum 21(suppl):134-138, 1978 9. GIBOFSKY A, WINCHESTER RJ, PATARROYO M, FOTINO M, K U N K E L
HG: Disease associations of the la-like human alloantigens. Contrasting patterns in rheumatoid arthritis and systemic lupus erythematosus. / Exp Med 148:1728-1732, 1978 10. D U P O N T B, Y U N I S EJ, DUQUESNOY R, POLLACK M, N O R E E N H, H A N SEN JA, REINSMOEN N, A N N E N K, GREENBERG L, L E E TD, W H I T -
SETT C, ANTONELLI P, BRAUN D: HLA-D and la alloantigen in North American Caucasians, in Histocompatibility Testing 1977, edited by BODMER WF, BATCHELOR JR, BODMER JG, FESTENSTEIN F, MORRIS
PJ. Copenhagen, Munksgaard, 1978, p. 603 11. CALABRO J J, HOLGERSON WB, SONPAL GM, K H O U R Y MI: Juvenile
rheumatoid arthritis: a general review and report of 100 patients observed for 15 years. Semin Arthritis Rheum 5:257-298, 1976 12. PETTY RE, STORM PB, CASSIDY JT, B U R T A, SULLIVAN DB: Immuno-
logic correlates of antinuclear antibody in juvenile rheumatoid arthritis (abstract). Arthritis Rheum 12:323-324, 1969 13. CALABRO JJ, PARRINO GR, ATCHOO PD, MARCHESANO JM, G O L D -
BERG LS: Chronic iridocyclitis in juvenile rheumatoid arthritis. Arthritis Rheum 13:406-413, 1970 14. WEINBERGER HW, ROPES MW, K U L K A JP, B A U E R W: Reiter's syn-
drome, clinical and pathologic observations. A long term study of 16 cases. Medicine (Baltimore) 41:35-91, 1962 15. ROPES MW, B E N N E T T GA, COBB S, JACOX R, JESSAR RA: 1958 revi-
sion of diagnostic criteria for rheumatoid arthritis. Bull Rheum Dis 9:175-176, 1958 16. SOKOLOFF L: The pathology of rheumatoid arthritis and allied disorders, in Arthritis and Allied Conditions: A Textbook of Rheumatology, 8th ed., edited by HOLLANDER JL, MCCARTY DJ JR. Philadelphia, Lea & Febiger, 1972, pp. 309-311 17. JESSAR RA: The study of synovial fluid. See Reference 16, p. 72 18. STEERE AC, H A R D I N JA, R U D D Y S, M U M M A W JG, M A L A W I S T A SE:
Lyme arthritis: correlation of serum and cryoglobulin IgM with activity, and serum IgG with remission. Arthritis Rheum, May 1979, in press
ACKNOWLEDGMENTS: The authors thank the physicians in the surveillance system, Drs. James A. Albright, David N. Romond, and Henry Drinker for doing the synovectomies and Dr. Nicholas H. Bartenhagen for helping with patient care; Dr. Marilena Fotino for identifying HLA antigens in certain patients; Ms. Stella Cretella for expert laboratory assistance; and Ms. Elise DeSanna for preparation of the manuscript. These investigations were supported in part by U.S. Public Health Service grants AM-20358, AM-10493, AM-07107, AM-5639, RR-05443, RR-00125, A1-00216, and CA-20107; the Arthritis Foundation and its Connecticut Chapter; the Kroc Foundation of Santa Ynez, California; and the Colombian National Science Foundation (Col Sciencias 140-3-37-77). Parts of this work have appeared in abstract form in Clin Res 27:338A,
KANTOR GL, MCINTOSH RM: Cryoglobulinemia and disease. Ann Intern Med 73:95-107, 1970 20. MARCUS RL, TOWNES AS: The occurrence of cryoproteins in synovial fluid; the association of a complement-fixing activity in rheumatoid synovial fluid and cold-precipitable protein. / Clin Invest 50:282-293, 1971 21. PEKIN TJ JR, ZVAIFLER NJ: Hemolytic complement in synovial fluid. / Clin Invest 43:1372-1382, 1964 22. FOSTIROPOULOS G, AUSTEN KF, BLOCH KJ: Total hemolytic complement (CH50) and second component of complement (C'2ku) activity in serum and synovial fluid. Arthritis Rheum 8:219-230, 1965
April 1979 (STEERE AC, GIBOFSKY A, WINCHESTER RJ, MALAWISTA, SE:
23. BREWERTON DA, CAFFREY M, NICHOLLS A, WALTERS D , OATES JK,
Persistent Lyme arthritis with pannus formation and cartilage erosion: clinical and immunogenetic differentiation from rheumatoid arthritis).
24. MORRIS R, METZGER AL, BLUESTONE R, TERASAKI PI: HL-A W27—
• Requests for reprints should be addressed to Allen C. Steere, M.D.; Department of Internal Medicine, Yale University School of Medicine, 333 Cedar Street; New Haven, CT 06510. Received 29 September 1978; revision accepted 8 March 1979.
19. BARNETT EV, BLUESTONE R, CRACCHIOLO A III, GOLDBERG LS,
JAMES DCO: Reiter's disease and HL-A 27. Lancet 2:996-998, 1973 a clue to the diagnosis and pathogenesis of Reiter's syndrome. AT Engl J Med 290:554-556, 1974 25. STASTNY P: Mixed lymphocyte culture typing cells from patients with rheumatoid arthritis. Tissue Antigens 4:571-579, 1974 26. STASTNY P: Association of the B-cell alloantigen DRw4 with rheumatoid arthritis. N Engl J Med 298:869-871, 1978 27.
References 1. STEERE AC, M A L A W I S T A SE, S N Y D M A N DR, SHOPE RE, A N D I M A N
WA, Ross MR, STEELE FM: Lyme arthritis: an epidemic of oligoarticular arthritis in children and adults in three Connecticut communities. Arthritis Rheum 20:7-17, 1977 2. STEERE AC, MALAWISTA SE, H A R D I N JA, R U D D Y S, A S K E N A S E PW,
ANDIMAN WA: Erythema chronicum migrans and Lyme arthritis. The enlarging clinical spectrum. Ann Intern Med 86:685-698, 1977 3. STEERE AC, HARDIN JA, MALAWISTA SE: Lyme arthritis: related problems recently recognized in Connecticut. Conn Med 42:353-357, 1978 4. REIK L, STEERE AC, BARTENHAGEN NH, SHOPE RE, M A L A W I S T A
REINERTSEN JL, KLIPPEL JH, JOHNSON AH, STEINBERG A D , D E C K -
ER JL, M A N N DL: B-lymphocyte alloantigens associated with systemic lupus erythematosus. N Engl J Med 299:515-518, 1978 28. WINCHESTER RJ, EBERS G, F U SM, ESPINOSA L, ZABRISKIE J, K U N K -
EL HG: B-cell alloantigen Ag 7a in multiple sclerosis (letter). Lancet 2:814, 1975 29. M C E W E N C, O'BRIAN WB: A multicenter evaluation of early synovectomy in the treatment of rheumatoid arthritis (abstract). / Rheumatol l(suppl 1):107, 1974 30. THOMPSON M, DOUGLAS G, D A V I S O N EP: Synovectomy of the meta-
carpophalangeal joints in rheumatoid arthritis. Proc R Soc Med 66:197199, 1973 31. ARTHRITIS A N D RHEUMATISM COUNCIL A N D BRITISH ORTHOPAEDIC
migrans and Lyme arthritis: epidemiologic evidence for a tick vector. Am J Epidemiol 108:312-321, 1978
ASSOCIATION: Controlled trial of synovectomy of knee and metacarpophalangeal joints in rheumatoid arthritis. Ann Rheum Dis 35:437-442, 1976 32. CHRISTIAN CL: Rheumatoid arthritis—etiologic considerations. Arthritis Rheum 7:455-466, 1964
6. WALLIS RC, B R O W N SE, KLOTER KO, M A I N AJ JR: Erythema chroni-
33. WILKES RM, SIMSARIAN JP, HOPPS HE, ROTH H, DECKER JL, A P T E -
SE: Neurologic abnormalities of Lyme disease. Medicine July 1979, in press
(Baltimore),
5. STEERE AC, B RODE RICK TF, MALAWISTA SE: Erythema chronicum
9 0 0
June 1979
• Annals of Internal Medicine • Volume 90 • Number 6
Downloaded From: https://annals.org/ by a University of Kentucky User on 08/24/2018
34. 35. 36. 37. 38.
KAR RG, M E Y E R H M J R : Virologic studies on rheumatoid arthritis. Arthritis Rheum 16:446-454, 1973 Z I F F M: Viruses and the connective tissue diseases. Ann Intern Med 75:951-958, 1971 FERGUSON RH, W O R T H I N G T O N JW: Recent advances in rheumatic diseases: 1967 through 1969. Ann Intern Med 73:109-124, 1970 PHILLIPS PE: The virus hypothesis in systemic lupus erythematosus. Ann Intern Med 83:709-715, 1975 F O R D DK: The etiology of non-gonococcal urethritis and Reiter's syndrome. Excerpta Med Int Congr Ser 165:227-229, 1969
39. C A L I N A, FRIES JF: An "experimental" epidemic of Reiter's syndrome revisited. Follow-up evidence on genetic and environmental factors. Ann Intern Med 84:564-566, 1976
A H O K, A H V O N E N P, A L K I O P, LASSUS A, S A I R A N E N E, SIEVERS K,
42.
40.
H A R D I N JA, W A L K E R LC, S T E E R E AC, T R U M B L E TC, T U N G KSK, W I L L I A M S R C J R , R U D D Y S, M A L A W I S T A SE: Circulating immune
complexes in Lyme arthritis: detection by 125 I-Clq binding, C l q solid phase and Raji cell assays. / Clin Invest 63:468-477,1979 41. H A R D I N J A, S T E E R E AC, C A R B O N I J M , SPAGNOLA A M , S P I E L E R PN,
M A L A W I S T A SE: Lyme arthritis: immune complexes correlate with stage and type of disease (abstract). Arthritis Rheum 21:564, 1978
TIILIKAINEN A: HL-A 27 in reactive arthritis following infection. Ann Rheum Dis 34(suppl l):29-30, 1975
Annals of Internal Medicine 90:901-904, 1979
Downloaded From: https://annals.org/ by a University of Kentucky User on 08/24/2018
S T E E R E AC, H A R D I N JA, M A L A W I S T A SE: Erythema chronicum mi-
grans and Lyme arthritis: cry ©immunoglobulins and clinical activity of skin and joints. Science 196:1121-1122, 1977
901
