Developing a Disease Activity IndexBy M. Anne CHAMBERLAIN
& Alan TENNANT (U.K.)
The Behçet's Disease Activity Form is now in widespread use. At the recent meeting discussions about its use led to considering the desirability of a Behçet's Disease Activity Index (BDAI) derived from the form. This task was given to the Rasch Studies Centre (RSC) in the Rheumatology & Rehabilitation Research Unit at the University of Leeds in the UK. At the meeting a request was made for contributions of data to the analysis and to-date four countries have donated data. Many of those who prefer to buy book reports can use our contact form at https://specialessays.com/buy-book-reports/ to leave a message with all the details.
The basic requirement for measurement laid down by the RSC was for a set of symptoms which could be summated to provide a simple index of severity. With both physicians' and patients' overall severity scores available such an index should clearly discriminate across these summary measures. An Index should also be free from Differential Item Functioning (DIF) by age and gender, and most important by country. What this means is that the probabilistic relationship between any proposed set of items should hold constant across countries.
The approach used is that of Item Response Theory (IRT) which is a general statistical theory about item (symptom) and scale performances and how that performance relates to the levels of disease activity.
The data from 242 patients have been analysed for the first stage of this exploratory analysis. With a mean age of 37 years, 47% were male and the mean perception of disease activity of patients and clinicians being 3.89 and 3.32 respectively on a simple 1-7 scale (smiley faces). The most common symptoms reported were mouth ulcers (68%), fatigue (46%), joint arthralgia (36%), headaches (36%) and eye problems (26%). There were some significant variations in prevalence between countries. For example, Skin Lesions (EN/ST) were reported by 13% of patients in Korea, but 53% of patients from China. Headaches were common in the UK (71%), but rare in Korea (15%). These variations cause problems with the basic requirement of a consistent probabilistic relationship across countries.
Perhaps surprisingly therefore, 16 symptoms did maintain such an order. These included mouth ulcers, joints arthralgia, joints arthritis, any red eye or painful eye, balance loss, arm pain/swelling and breathlessness. The symptom leg/pain swelling was the least commonly reported problem in this set of 16, arguably the upper level of severity.
However, this finding might be modified if a larger data base were available, as some impairments on the BDAF were not represented in the data analysed. Mouth ulcers were the most common and thus can be considered to mark the lowest level defined by adequate fit to the one parameter IRT (Rasch) model. DIF was largely absent, although skin lesions (pustules) did show some variation by country. The sixteen item scale showed strong discrimination across both patients' and physicians summary scales of disease activity (Mann Whitney U, p<.001). For example, an average score of 0.6 was found for the most smiley face for patients, compared to 6.3 for the most miserable face.
In summary a 16 item (symptom) scale has been identified which appears to offer a unidimensional scale for disease activity, which is largely free of DIF and discriminates across self- of physician reported disease activity states. A second set of data is currently being collected to independently validate these results.
By Haner DIRESKENELI (Turkey)
Vascular involvement in BD is predominantly venous in contrast to SV. The rare presence of pulmonary emboli is suggestive of a "sticky" thrombi and in histopathological studies, in addition to thrombi, inflammatory infiltrates in vessel walls point to a vasculitic process. Antibodies against endothelial cells (AECA) have been described in BD, but specific target antigens of AECA, such as adhesion molecules, proteoglycans or co-stimulatory molecules (as recently described CTLA-4 of anti-lymphocyte antibodies) are still unknown. Increased Factor V Leiden mutation, mainly in patients with venous thrombosis and retinal occlusive disease also show the role of increased propensity to thrombosis in BD.
Immune effector mechanisms
Th1/Th2 balance is suggested to be critical in the induction and regulation of autoimmunity and inflammation. A Th1 type; IL-2, IFN-g and IL-12 dominant cytokine profile is present in active BD, whereas Th2 activity (IL-4, IL-13), except for IL-10, is controversial. A pro-inflammatory mileu with serum and local elevations of monocyte-macrophage derived cytokines such as IL-1b, IL-6, TNF-a and chemokines such as IL-8 and MCP-1 is also demonstrated. This profile might be responsible for the increased O2 release, enzymatic activity, phagocytosis and chemotaxis of neutrophils, reported by some denied by others.
An infectious etiology
Although herpetiform ulcers are not common, Herpes simplex virus (HSV) is currently the only virus possibly associated with BD. HSV DNA has been demonstrated in the genital and intestinal ulcers of BD patients and HSV infection increases T lymphocyte adhesion to endothelium. Thirty percent of ICR mice inoculated with HSV developed Behçet's disease-like symptoms.
Among bacterial agents mainly streptoccocci have been implicated. A Th1 type cytokine response and increased gd+ T cells are observed after in vitro streptococcia stimulation of BD lymphocytes. Cytokine responses against St. salivarius are suppressed by minocyline treatment.
Molecular mimicry and 65 kD HSP
65 kD heat-shock protein (HSP) is a candidate antigen of streptoccocal infection in BD. Studies with mycobacterial 65 kD HSP which has >90% homology with streptoccocal HSP and its 60 kD human homologue have shown increased T cell and humoral responses to HSPs in patients from UK, Japan and Turkey. BD patients with uveitis, in addition to streptoccocal and human HSPs, also have increased antibody responses to retinal HSP, suggesting a cross-reactivity of these antibodies. Mucosal or systemic inoculation of human 60 kD HSP immunodominant epitope 336-51 causes uveitis in rats. In contrast to the situation in BD, T cell responses to human 60 kD HSP are thought to be protective in autoimmune diseases such as juvenile rheumatoid arthritis and further characterization of HSP responsive T cells, such as their Th1/Th2 cytokine profile, is required.
Superantigens and hyperreactivity
In addition to streptoccocal antigens, Escherichia coli and Staphylococus aureus also activate BD lymphocytes to release increased amounts of IFN-g and IL-6. A further observation of increased IFN-? response of Behçet's T lymphocytes to very low (1-10 pg/ml) doses of staphyloccocal superantigens SEB and SEC1 is proposed as an evidence of T cell hyperreactivity in BD, which might relate to abnormalities in signal transduction triggered through the T cell receptor. A broader intra-cellular signaling abnormality of all immune cells (including neutrophils) which lowers the threshold of inflammatory responses to external stimuli, as proposed for familial Mediterranean fever with decreased pyrine expression of neutrophils, might also be an interesting hypothesis.
gd+ T cells, found to be increased both in peripheral blood and tissue infiltrations, can also be a part of this antigen non-specific immune activation. In addition to conventional peptide antigens such as HSPs, gd+ T cells recognize non-peptidic small molecules and activate polyclonally after infections, giving them a role close to innate immunity.
Immune responses to organ-specific antigens such as dermal keratins in psoriasis, collagen type II in rheumatoid arthritis or retinal-S antigen in uveitis were previously demonstrated. Similarly, different organ-specific antigens might be relevant for BD subgroups with muco-cutaneous, vascular or articular involvement such as mucosal, endothelial or synovial antigens. HLA-B derived, uveitogenic peptide 125-38 might be an example of these organ-specific antigens in BD. These need further scrutiny.