Article Text

Download PDFPDF

Is Behçet’s syndrome associated with infection?
  1. A P Rozin1
  1. 1B Shine Department of Rheumatology, Rambam Medical Centre and B Rappaport Faculty of Medicine, Israel-Technion Institute of Technology, Haifa, Israel
  1. Correspondence to:
    Dr A Rozin
    Department of Rheumatology, Rambam Medical Center, PO Box 9602, Haifa 31096, Israel;
  1. G Hatemi2,
  2. H Yazici2
  1. 2Division of Rheumatology, Department of Medicine, Cerrahpasa Medical Faculty, University of Istanbul, Aksaray 34303, Ýstanbul, Turkey

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

I read with interest that the pustular skin lesions in Behçet’s syndrome (BS) had been thought aseptic, were found to be not sterile, and that the microbiology of these lesions is different from ordinary acne.1 I would like to report my observation of a patient with refractory pustulosis of Behçet’s disease, who fulfilled the international study group criteria, was HLA-Bw51 positive, and had a family history of BS. The patient’s skin rash disappeared after a 6 week course of co-trimoxazole (sulfamethoxazole-trimetoprim).

The patient, a 31 year old man had had recurrent oral and genital ulcers since childhood. Inflammatory joint disease developed 4 years ago, affecting shoulders, ankles, and knees, relapsing every 2–3 months. Recurrent knee effusions caused serious knee dysfunction. Skin pustulosis, which was episodic at onset, became persistent and massive during the past 4 years, affecting the body, back, and limbs (fig 1A). A skin vesicle was observed 24–48 hours after taking blood for analysis from the knee at the point of needle entry. Polyarthritis and skin pustulosis became refractory to local, systemic, and intra-articular corticosteroids and colchicine. The pustular lesions thought to be sterile in BS were not cultured. Salazopyrin, methotrexate given orally and parenterally at maximal dose of 25 mg/week, and azathioprine failed to control the knee effusions, shoulder and ankle arthritis, skin rash, and oral ulcers. Invasive knee procedures were planned.

Figure 1

 Pustulosis of Behçet’s disease localised in the right suprascapular area: (A) before co-trimoxazole treatment; (B) on the 10th day of co-trimoxazole treatment (960 mg twice a day).

At this point the failed second line treatment was stopped, and we made the decision to start co-trimoxazole treatment. The rationale for this was the anti-inflammatory properties of the drug reported previously2 and its effectiveness for some patients with Wegener’s granulomatosis3 associated with severe neutrophillic activation,4 which is also seen in skin lesions of patients with BS.5

BS was reported to be associated with a higher incidence of Streptococcus mediated tonsillitis, and its adjuvant action to autoimmune disease cannot be excluded.6 Circulating immune complexes are thought to precipitate a neutrophilic vascular reaction, resulting in mucocutaneous lesions.7,8 A decrease in serum IgG and IgM was noted during co-trimoxazole treatment.2

Co-trimoxazole treatment was started with a daily dose of 50 mg/kg given in four divided doses (960 mg×4/day) for the first 3 days. Then the dose was reduced to 960 mg×3/day given for 1 week, followed by two double strength tablets a day until 6 weeks of treatment. The pustular rash gradually disappeared (figs 1A and B). After 6 weeks of the co-trimoxazole treatment the drug was stopped and weekly methotrexate injections were restarted at the previous dose. Knee effusion has relapsed only once during 1 year of follow up.

Evidence, that infection is the most probable environmental trigger of inflammatory joint disease is controversial, but interest in the topic is growing. The relationship between infection and collagen disease may be more subtle and complex than one of simply responding to Koch’s postulates.9 Multiple infectious triggers which attack at an unknown rate, the delayed interval between infection and disease onset, and a role for primary, secondary, and persistent infection in the perpetuation of collagen disease are the substance of the microbiology of rheumatic diseases. Bacteria are not only a source of exogenous antigens, which potentially cross react with those of the host, but can also exert adjuvant effects and release self antigens. Lipopolysaccharides, peptidoglycans, and bacterial DNA activate the innate immune system through specialised pattern recognition receptors of the Toll-like receptor family.10 Such microbial determinants are referred to as “pathogen associated molecular patterns”. These patterns, together with the self antigens, activate the production of polyreactive antibodies, cytokine, and chemokines. Infected pustules of Behçet’s disease might cause severe activation of the autoimmune response. Co-trimoxazole may be a promising treatment for controlling the microbial inductors and autoimmune reactions.


Authors’ reply

We thank Dr Rozin for his interest in our article and sharing his experience about a patient with Behçet’s syndrome (BS) who improved with co-trimoxazole. We also recently had a patient with BS who had severe pustulosis, arthritis, oral and genital ulcers and who similarly did well with antibiotics. Staphylococcus aureus grew from both the dermal pustules and the pustular pathergy lesions.

Thus far there have been few formal studies of antibiotic use in BS. Çalgüneri et al reported that penicillin treatment was beneficial for the mucocutaneous lesions1 and arthritis.2 A similar beneficial effect was observed with minocycline, which reduced both the frequency of clinical symptoms and the production of inflammatory cytokines by peripheral blood mononuclear cells stimulated by streptococcal antigens.3

The issue of an infectious aetiology in BS has also long been discussed. Behçet himself proposed a viral aetiology.4 It has been suggested that viruses, such as herpes simplex virus5 and parvovirus,6 and bacteria including various streptococcal strains7 and staphylococci8 have a role.

In one study peripheral γδ+CD8+T cells of patients with BS showed a significantly proliferative response to the Streptococcus sanguis strain KTH-1.7 In another, T cells from patients with BS produced interferon γ when stimulated with staphylococcal superantigens.8 Clinical evidence for the role played by an infectious agent in pathogenesis includes the presence of a higher incidence of chronic tonsillitis and dental caries in patients with BS,9 observation of exacerbations of BS symptoms after acute episodes of infection with Streptococcus agalactiae vaginitis,10 and gingival infections with methicillin resistant Staphylococcus aureus.11

There are also reports from our group showing the association of papulopustular lesions with arthritis in BS, suggesting a reactive type of arthritis.12,13 Lehner and colleagues suggested that a common antigen such as a stress protein might be involved.14 A significant increase of IgA antibodies to mycobacterial 65 kDa heat shock protein (HSP) in the serum of patients with BS was shown. Owing to the significant homology between mammalian and microbial HSPs, it is suggested that recurrent exposure to HSP may cause bacterial HSP responsive T cells to stimulate autoreactive T cells by cross reactivity mechanisms. In turn, these T cells might produce Th1-like proinflammatory and/or inflammatory cytokines, leading to tissue injury.

Whatever the precise pathogenic pathways will turn out to be, it is clear that further controlled trials with antibiotics in BS are warranted.