Results: Mean BD activity score fell from 7.2 to 1.5, posterior uveitis score from 4.43 to 1.08 and visual acuity rose from 0.27 to 0.5. There was one non-responder (3.6%). Two patients had to stop therapy because of intolerable side effects (hair loss in one, depression and diarrhea in the other). Other side effects were development of fibromyalgia (n = 4), of thyroid antibodies (n = 4), in one case with hyperthyroidism, of antinuclear antibodies (n = 6) and exacerbation of psoriasis (n = 2). Fever, arthralgia and headache occured in all patients, but only during the first week of therapy. One patiient was lost for follow up because of noncompliance. Thus, 25 patients are still followed up with a mean observation period of 42 months (range 6 to 86 months).

Objective: To determine the frequency of acneiform skin lesions (comedones, papules, and pustules) and whether or not there is a relationship between arthritis and a specific type of skin lesion in Behçet patients with arthritis.

Methods: We studied 44 Behçet patients with arthritis, 32 males and 12 females (group A); 42 Behçet patients without arthritis, 31 males and 11 females (group B); 21 patients with active rheumatoid arthritis, 5 males and 16 females (group C); and 33 healthy volunteers, 5 males and 16 females (group D), all prospectively. The mean age was 37.8 ± 8.9 in group A, 35.5 ± 6.4 in group B, 48.8 ± 14.1 in group C, and 40.1 ± 8.1 in group D. All patients and healthy volunteers were examined by a rheumatologist and a dermatologist. Patients with Behçet syndrome were also evaluated for eye involvement by an ophthalmologist. Skin lesions, including comedones, papules, and pustules were counted and scored (0:absent, 1:1-5, 2:6-10, 3:11-15, 4:16-20, 5: >20).

Results: Both Groups A and B were similar for age, sex, medications, oral and genital ulcers, erythema nodosum, thrombophlebitis, eye involvement, pathergy, and HLA-B5 positivity. Although there was no significant difference among the four groups in the frequency of comedones, the number of papules and pustules were higher in Behçet patients with arthritis. Mean scores for papules and pustules for each patient in group A, B, C, and D were 1.27, 0.57, 0.67, 0.91 and 1.3, 0.48, 0.14, 0.12, respectively (p < 0.05 by Kruskal-Wallis One-Way ANOVA). In the two Behçet groups, papules and pustules were more predominant in patients with arthritis (p < 0.05 by Mann-Whitney-U Test). The frequency of skin lesions in Behçet patients with arthritis did not correlate with the pattern of joint involvement (mono-oligo and polyarthritis), the number of previous arthritis attacks, and the number of arthritis attacks in the last three months. There was no significant difference between eye involvement and the number of skin lesions in Behçet patients with/without arthritis.

Conclusion: Certain types of acneiform skin lesions (papules and pustules) appear to be more frequent in Behçet patients with arthritis. This brings up the consideration that the arthritis seen in Behçet's syndrome may have similar pathogenic mechanism(s) to that of the acne-associated arthritis.

HLA- B51 allele is associated with Behçet's disease in several populations but its importance in disease complications is not well studied. We analyzed the HLA typing in 46 Lebanese patients diagnosed to have Behçet's disease according to the International Study Group criteria. Twenty eight percent were HLA-B51 positive compared to 15.5% in our general population (p = 0.04). Anterior and/or posterior uveitis was present in 15/31 (48%). In the HLA-B51 positive group, the frequency of uveitis was 87% (7/8) compared to 35% (8/23) in the HLA-B51 negative group (p = 0.01) suggesting a strong association. No statistically significant difference was seen in venous, arterial or neurological complications between the two groups. In addition, patients with HLA-B51 allele have at least one major complication (uveitis, venous, arterial or neurological) compared to HLA- B51 negative patients (p = 0.036). This study provides additional evidence that Behçet's disease is associated with HLA- B51 allele in the Lebanese population and suggests that HLA- B51 allele may have a prognostic value in patients with Behçet's disease.

Triplet repeat polymorphism in the transmembrane region (TM) of the MHC class I chain-related A (MICA) gene has been recently strongly associated with the occurrence of BD in Japanese patients. To address the possibility that MICA, rather than HLA-B gene is responsible for the genetic disposition to BD in caucasoid patients, we studied the triplet repeat polymorphism in the TM of the MICA gene in 104 caucasoid patients fulfilling the International Criteria for BD and compared it to 77 healthy caucasoid blood donors. Fisher's exact test was used for statistical analysis.

The gene frequency of the A6 allele was significantly increased in BD, as compared with controls (44.6% vs 28.8% respectively, P = 0.003). No significant differences was found between BD patients and controls for the gene frequencies of the A4, A5, A5.1 and A9 alleles of the MICA gene. The phenotype frequency of the A6 allele was significantly higher in BD than in controls (67.3% vs 48%, P = 0.0057). Forty-one BD patients were HLA-B51 positive (39.4%) and 38 of them (95.1%) possessed the MICA A6 allele. In addition, 32 HLA-B51 negative BD had the MICA A6 allele, indicating the association of MICA A6 allele with BD disease was stronger than that of HLA-B51 (P < 0.0001).

Immune response to retinal autoantigens may play a central role in the pathogenesis of uveitis. A synthetic peptide from a common sequence of various HLA-B molecules associated with uveitis such as HLA-B27 and B51 (BPD), which shares aminoacid homologies with a retinal-S antigen derived peptide (PDS) was shown to be immunogenic in human and experimental uveitis.

In this study, we investigated T cell responses to these 2 synthetic peptides in Behçet's disease patients with posterior uveitis (BD, n = 26) and compared with non-Behçet anterior uveitis (AU, n = 14) and healthy controls (HC, n = 26), in a 6-day proliferation assay.

BD patients had higher responses (stimulation index, SI: 2.6 ± 1.2) than AU (SI: 1.3 ± 0.6) and HC (SI:1.2 ± 0.7) for BPD (p < 0.001). Responses to PDS were also higher in BD patients (SI: 3.3 ± 1.8) than AU (SI: 1.6 ± 0.5) and HC (SI: 1.3 ± 0.6) (p < 0.001). A significant correlation between the responses to BPD and PDS was also observed (r = 0.6, p < 0.001).

These results suggest that cellular immunity to cross-reactive ret-S and HLA-B derived peptides might play a role in the pathogenesis of posterior uveitis in BD.

In conclusion our data represent the first Italian collaborative study on pBD. Despite the relatively low number of patients, the prevalence of GI symptoms in Italian pBD seems to be frankly increased with respect to other pediatric series from Mediterranean basin. This latter finding confirms the clinical heterogeneity of pBD among different ethnic group.