Scand J Rheumatology, Suppl. 74: 89-93, 1988

Implications from the Occurrence of Reiter‘s Syndrome and Related Disorders in Association with Advanced HIV Infection

ROBERT WINCHESTER, LENORE BRANCATO, SILVIU ITESCU, MARY LOUISE SKOVRON and GARY SOLOMON Hospital for Joint Diseases, Orthopedic Institute, Department of Rheumatology, 301 E. 17 Street, New York, NY 10003, USA

There is considerable evidence that the immune system is involved in the pathogenesis of Reiter’s syndrome. This includes induction by specific microorganisms with a characteristic time course suggesting an immune response. Moreover, the widely recognized effectiveness of immunosuppressive therapy, the presence of lymphocytes in lesional sites, and the central observation that susceptibility is associated with HLA-B27 (1) support this view. However, the absence of immune complexes and autoantibodies sets the pathogenesis of the disease apart from disorders such as rheumatoid arthritis and systemic lupus erythematosus.

Accordingly, we were initially unprepared for the finding that Reiter’s syndrome occurred in individuals with profound failure of the immune system in acquired immune deficiency syndrome and other stages of advanced infection with HIV (2, 3). A broader relationship of the spondyloarthropathies as a group to HIV infection was suggested by the finding of other arthritic syndromes that exhibited features of psoriatic arthritic or were undifferentiated forms of seronegative arthritis and enthesopathy (3-7). The relationship to the spondylo- arthropathies was reinforced by the finding of various forms of psoriasis and psoriasiform le- sions among HIV patients (8, 9). The observations, taken together, suggested that the pathogenesis of these disorders proceeded without impairment among individuals devoid of functional CD4 lineage helper T cells. Accordingly, the hypothesis was formulated (3,4) that critical events in the pathogenesis of these disorders could primarily involve the CD8 lineage of suppressor-cytotoxic T cells that is largely unaffected by direct HIV infection.

While the arthritic and cutaneous disorders clinically resemble their counterparts in non- HIV-infected individuals, considerable additional information is required on the natural his- tory, epidemiology and immunologic features of these illnesses to determine whether the dis- orders as they occur in the setting of HIV infection are equivalent in pathogenesis to those occurring in the general population.

Findings of a prevalence study. A beginning toward this objective was made in a small prev- alence study (6). Table I illustrates the prevalence of three categories of disorders relevant to rheumatic disease in a study of 65 individuals attending an HIV clinic. Ascertainment was through examination by a rheumatologist . What appeared to be psoriasiform lesions were met with most frequently. They were not documented by biopsy. Reiter’s syndrome was found in 3 patients, with a point prevalence of 4.6%. This was a proportional increase of more than 140-fold over the inferred prevalence of Reiter’s syndrome in Minnesota or reac- tive arthritis in Finland (10). This proportional increase was numerically equal to the relative risk or odds ratio, but these latter terms should not be used to describe this observation be- cause the disease and control populations are not comparable.

Among the 7 cases of arthritis other than Reiter’s syndrome encountered in this series, two met the criteria of psoriatic arthritis and five had various combinations of enthesopathy, dac- tylitis and oligoarthritis (6) (Table 11). There was a strong correlation between skin and nail

Scan

d J

Rhe

umat

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Fr

eie

Uni

vers

itaet

Ber

lin o

n 11

/13/

14Fo

r pe

rson

al u

se o

nly.

90 R. Winchester et al. Scand J Rheumatology Suppl74

disease (including seborrheic dermatitis, psoriasiform rashes, onychodystrophy or onycho- lysis) and arthritic involvement (p=0.002). However, while the two individuals clearly meet criteria for psoriatic arthritis, it remains to be established whether this was the coincidence of the two prevalent conditions of psoriasiform rash and arthritis or whether it is a single clinicopathologic entity.

Certain diseases appear to be significantly decreased in prevalence among HIV infected in- dividuals. These include rheumatoid arthritis. systemic lupus erythematosus, a n d - o f in- terest because it is a spondyloarthropathy-ankylosing spondylitis. While the manifestations of the latter disorder require a lengthy time period to evolve fully, it is significant that axial symptoms of any sort. including pain or stiffness of the spine, other than sacroiliitis, are dis- tinctly uncommon among these patients. The implications of this preliminary observation is that the disease mechanism leading to ankylosing spondylitis is different from that leading to Reiter’s syndrome.

Determinants of susceptibility to HIV-associated Reiterj. syndrome are similar to those in conventional ReiterS syndrome. Independent of the prevalence study, a total of twenty-one patients with HIV-associated Reiter’s syndrome studied in some detail revealed that HLA- B27 was present in 71 YO. a percentage well within the range in which HLA-B27 was encoun- tered among non-HIV Reiter’s syndrome (1, 11). Similarly, among 30% of the individuals with HIV infection, the Reiter’s syndrome followed infection with S. typhimurium, S. flexneri and C . jejuni. These findings emphasize the epidemiologic similarity of Reiter’s syndrome developing in the setting of HIV infection with that occurring in non- HIV-infected individuals. Taken together with their clinical similarities, we conclude that the two forms of Reiter’s syndrome are equivalent and proceed by similar disease mechanisms. In addition, the appearance of infection with M. avium species was temporally related to the

Table I. Prevalence of skin and joint manifestations in HIVpositive outpatients (n=65)

95% confidence Prevalence interval

n (”/.I ( % )

Psoriasiform lesions 13 20 1 O.4-29.6 Keiter’s syndrome 3 4.6 2.G13.0 Psoriatic and other forms of arthritis 7 11 3.6-18.4

Table 11. Prevalence of A I D S associated arthritis syndromes in HIV-positive outpatients (n=651

Diagnosis Prevalence

Number ( O h 1 Reiter’s syndrome 3 Psoriatic arthritis 2

Guttate psoriasis+onycholysis+dactylitis Onycholvsisl-pitting+dactylitis

Enthesopathy Enthesopathy+psoriasis vulgaris Enthesopathy+sebomheic dermatitis Oligoarthritis+dactylitis +onycholysis

Unclassifiable arthritic syndromes .5 8

Scan

d J

Rhe

umat

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Fr

eie

Uni

vers

itaet

Ber

lin o

n 11

/13/

14Fo

r pe

rson

al u

se o

nly.

Scand J Rheumatology Suppl 74

onset of arthritis in an additional 25%, raising the question of whether the syndrome could develop as a reaction to this organism.

It remains to be determined what factor associated with HIV infection accounts for the greatly increased risk of developing Reiter’s syndrome. Possibilities range from abnor- malities of immune regulation, a response to virus tropic for certain synovial cells such as the macrophage, to the trivial one that the individual’s behavior leading to the acquisition of HIV also results in an enhanced likelihood for the acquisition of Reiter’s syndrome.

Enhanced penetrance of HLA-B27 related susceptibility of Reiter’s syndrome susceptibility. A detailed study of the prevalent disease in these patients offers an important opportunity to define disease mechanisms and risk factors as they are influenced by the events associated with the development and response to HIV infection. HLA-B27 is a critical determinant of susceptibility to Reiter’s syndrome. Since HLA-B27 is present in 6-8?’0 of Caucasian and lower percentages of Blacks and Hispanics, adjustment of the expected frequency of HLA- B27 for ethnically mixed population at risk puts the value between 4% and 5%. Since 71% of affected individuals in the present case study are HLA-B27, and if a prevalence of 4.6% can be extrapolated from the value found in the prevalence study, one can infer that approx- imately three-quarters of HLA-B27+ individuals will develop Reiter’s syndrome. The pro- portion of HLA-B27+ individuals acquiring reactive arthritis or Reiter’s syndrome after ex- posure to a documented arthritogenic organism has been estimated to be no higher than 20% (12, 13), and is viewed to be considerably lower by some workers (14). Should these assump- tions and inferences be verified in a formal genetic epidemiologic study, it suggests that some event in these HIV’ individuals greatly enhances their liability to develop Reiter’s syn- drome. The similar proportion of HLA-B27 negative Reiter’s syndrome also emphasized that this event acts homogeneously, directing attention to the fine structure of the HLA alleles in those that develop Reiter’s syndrome in the absence of HLA-B27.

One interpretation of these results is that cells of the CD4 lineage normally inhibit the de- velopment of Reiter’s syndrome. Thus, by extension, the occurrence of Reiter’s syndrome in a non-HIV infected individual would signify a transient diminution of CD4 lineage T cells, or an equivalent disturbance in immune regulation.

Susceptibility to psoriasiform rashes in HZV infection is not HLA-associated. In contrast to Reiter’s syndrome, psoriasis and psoriatic arthropathy are not generally thought to primarily proceed from an immunologic mechanism. Nevertheless, susceptibility to the latter two dis- eases in the general population is associated with particular HLA class I alleles of the B and C loci (15). Since a 20% prevalence of psoriasiform lesions was found among HIV patients (6), the question was addressed of whether the HLA class I alleles associated with suscep- tibility in the general population would be identified in HIV-associated psoriasis or psoriatic arthropathy. In complete contrast to the findings in Reiter’s syndrome, no significant associ- ation was found between psoriasis and HLA-Cw6, nor with other alleles associated with sus- ceptibility to psoriasis or psoriasiform lesions in these HIV patients. This raises the strong possibility that the pathogenesis of the psoriasiform lesions in HIV patients proceed through a different disease mechanism which does not involve specific polymorphic class I molecules, and could be directly induced by a response to viral infection of the Langerhans cells.

The lack of an HLA association with a psoriasiform rash raises several questions, among them the relation of the skin lesions of keratoderma blenorrhagicum in Reiter’s syndrome to morphologically similar lesions of pustular psoriasis. Another question is the nature of the arthritis syndromes associated with some skin or nail disease discussed above. Does this suggest that the critical event in the development of psoriatic arthritis is the presence of one of several skin diseases that each have the appearance of psoriasis but result from distinct etiologies? A similar situation exists with enteropathic arthritis where several intestinal dis- eases induce a similar arthritic syndrome.

Reiter’s syndrome and H N infection 91

Scan

d J

Rhe

umat

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Fr

eie

Uni

vers

itaet

Ber

lin o

n 11

/13/

14Fo

r pe

rson

al u

se o

nly.

92 R. Winchester et al. Scand J Rheumatology Suppl74

New therapeutic approaches? The therapy of the Reiter‘s syndrome and allied arthro- pathies has proven difficult. A considerable proportion of patients do not respond to the usual non-steroidal anti-inflammatory agents, nor to steroids. Phenylbutazone, 100 mg taken 2 or 3 times a day, has been particularly helpful in relieving enthesopathic and arthritic symptoms in these patients. A trial of sulfasalazine is also perhaps warranted (3). Metho- trexate and other immunosuppressive agents should be avoided. While clearly effective, their use has resulted in either the development or worsening of the manifestations of frank AIDS in HIV-infected patients (3,4). Nearly all patients in this series have been treated with AZT for various indications, and from these data it is not possible to determine if there is a salutary effect of this agent on the course of Reiter’s syndrome or psoriasiform manifesta- tions.

Since the action of many immunosuppressive drugs mimics certain of the effects of HIV infection, and since agents such as methotrexate have suppressed severe Reiter’s syndrome both in AIDS patients and in the general population (16, 17) the question arises of whether these drugs have their effect in this disease on cells outside the CD4 lineage. Given the view that the pathogenesis of Reiter’s syndrome proceeds through a CD8 lineage Tcell, one would expect that novel therapeutic agents could be developed that act only on functions of this T cell lineage and should be selectively effective in Reiter’s syndrome and perhaps other spon- dyloarthropathies.

Specific immune recognition events in Reiter’s syndrome. This article began with a consid- eration of the participation of the immune system in the pathogenesis of Reiter’s syndrome. We would suggest that the occurrence of Reiter’s syndrome in a setting of HIV infection ad- vances the understanding of this participation by emphasizing the potential role of CD8 lineage T cells. Taken together with the association of susceptibility with the class I MHC al- lele HLA-B27, we would further suggest that a specific immune recognition event underlies the development of Reiter’s syndrome. This event can be visualized as the interaction of a specific antigen fragment of currently unknown character with an HLA-B27 molecule in an antigen presentation event to a CD8 lineage T cell characterized by a particular clonal recep- tor. Consistent with this interpretation of the pathogenesis of Reiter’s syndrome centered around CD8 lineage T cells is the absence of evidence of significant participation of helper T cells, such as by the lack of autoantibodies and findings of a delayed type hypersensitivity response. Findings supporting this view have been recently presented (18). Should this view be supported by additional evidence, it would emphasize that beneath the layers of recruit- ment of non-antigen-specific cells into the immune response in Reiter’s syndrome that a specific set of molecule interactions centered on presentation of a defined antigen fragment underlies the development of the disease. The nature of the antigen and the events leading to the development of the T cell clone of course remain the challenging questions.

SUMMARY

Information bearing on the prevalence and character of Reiter’s syndrome and allied dis- orders as they occur in a setting of HIV infection was reviewed. Based on the frequency of infections by organisms capable of inducing reactive arthritis and the presence of HLA-B27 in 71% of the individuals, it was concluded that the disorder strongly resembled Reiter’s syn- drome occurring in the non-HIV infected group. Preliminary evidence suggested an en- hanced penetrance of susceptibility among HLA-B27 individuals. In view of the preservation of CD8 lineage T cells and functional loss of CD4 lineage T cells in HIV-infected patients, it was suggested that a specific immune recognition event is at the center of the pathogenesis of this syndrome which involves preservation of an unknown antigen in the context of HLA- B27 to CD8 lineage suppressorkytotoxic T cells.

Scan

d J

Rhe

umat

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Fr

eie

Uni

vers

itaet

Ber

lin o

n 11

/13/

14Fo

r pe

rson

al u

se o

nly.

Scand J Rheumatology Suppl 74 Reiter’s syndrome and H N infection 93

ACKNOWLEDGEMENT This research was supported by NIH grant AR39626.

REFERENCES 1. Brewerton DA, Hart FD, Nicholls A, Caffrey M, James DCO, Sturrock RD. Ankylosing spon-

dylitis and HL-A27. Lancet 1973; 1: 904. 2. Fischer H , Solomon G, Enlow R , Winchester R. Reiter’s syndrome and the acquired immune defi-

ciency syndrome. Arthritis Rheum 1985 (abstr). 3. Winchester R, Bernstein DH, Fischer HD, Enlow R, Solomon G. The co-occurrence of Reiter’s syn-

drome and acquired immunodeficiency. Ann intern Med 1987; 106: 19. 4. Winchester R, Solomon G. Reiter’s syndrome and acquired immunodeficiency. In: Espinoza L, Gol-

denberg D, Arnett F, Alarcon G , eds. Infections in the rheumatic diseases. Orlando, Fla.: Grune & Stratton, 1988: 343.

5. Rynes RI, Goldenberg D , Olsen R, Hussain M, Veazey J, DiGiacomo R. Acquired immunodefi- ciency syndrome associated arthritis, Arthritis Rheum 1987; 30 (Suppl 4): S37 (abstr).

6. Brancato LJ, Itescu S, Skovron ML, Mildvan D , Winchester RJ. Arthritis, psoriasis and related syn- dromes associated with HIV infection. Clin Res 1988; 36: 531A (abstr).

7. Espinoza LR, Berman A, Vasey FB, Cahalin C, Nelson R , Germain BF. Psoriatic arthritis and ac- quired immunodeficiency syndrome. Arthritis Rheum 1988; 31: 1034.

8. Johnson TM, Duvic M, Rapini RP, Rios A. AIDS exacerbates psoriasis (Letter). N Engl J Med 1985; 313: 1415.

9. Kaplan MH, Sadick N, McNutt S, Meltzer M, Sarngadharan MG, Pahwa S. Dermatologic findings and manifestations of acquired immunodeficiency syndrome (AIDS). J Am Acad Dermatol 1987; 16: 485.

10. Michet CJ, Machado EBV, Ballard DJ, McKenna CH. Epidemiology of Reiter’s syndrome in Rochester, Minnesota: 1950-1980. Arthritis Rheum 1988; 31: 428.

11. Aho K, Ahvonen PI, Alkio P et al. HLA-B27 in reactive arthritis following infection. Ann Rheum Dis 1975; 34 (Suppl 1): 29.

12. Calin A, Fries JE An “experimental” epidemic of Reiter’s syndrome revisited: follow-up evidence on genetic and environmental factors. Ann Intern Med 1976; 84: 564.

13. Paronen I. Reiter’s disease: a study of 344 cases observed in Finland. Acta Med Scand 1948; 131

14. Khan MA, Kushener I , Ballou SP, Braun WF. Familial rheumatoid arthritis and HLA-DRw4. Lan- cet 1979; i: 921.

15. Terasaki PI, ed. Histocompatibility testing 1980. Los Angeles: UCLA Tissue Typing Laboratory, 1980: 711.

16. Mullins JF, Maberry JD, Stone OJ. Reiter’s syndrome treated with folk acid antagonists. Arch Der- matol 1966; 94: 335.

17. Farber GA, Forshner JG, OQuinn SE. Reiter’s syndrome: treatment with methotrexate. JAMA 1967; 20: 171.

18. Santavirta S, Nordstrom D , Konttinen YT, Bergroth V, Saari H. Cell-mediated immune response in reactive arthritis. Scand J Rheumatol 1988; Suppl 67: 22.

(Suppl 212): 1.

Scan

d J

Rhe

umat

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Fr

eie

Uni

vers

itaet

Ber

lin o

n 11

/13/

14Fo

r pe

rson

al u

se o

nly.