OBJECTIVE: We have previously reported distinct symptom clusters among our patients with Behçet's syndrome (BS). The presence of such clusters suggests that more than one pathogenetic mechanism might be operative in BS. Increases in the frequency of certain clusters in familial BS cases, if present, would further support this notion. To test this hypothesis, we compared the frequency of symptom clusters between familial (group F) and non-familial (group NF) cases of BS. METHODS: We identified 380 BS patients who had reported a first-degree relative by reviewing 6031 patient charts. We were able to contact 186 (Group F). From the same initial pool, 500 patients were randomly selected. Of those, patients who did not report a family history of BS and who had attended our clinic during the previous 3 months made up group NF (n = 221). Both groups were questioned about their symptoms within the previous 3 months. Data were analysed using factor analysis, cluster analysis and χ2 tests. RESULTS: The make-up of the symptom clusters were very similar for the factor and the cluster analyses. The frequency of papulopustular lesions and joint involvement cluster was significantly higher in group F (39.2 vs. 21.5%, P < 0.001). Furthermore, the same cluster was shared in 5/17 related pairs from group F and in only 5/110 unrelated pairs from group NF [29 vs. 4.5%, P = 0.004; risk ratio (RR) = 6.47, 95% CI 2.15, 18.89]. CONCLUSION: The papulopustular lesions and arthritis cluster in BS appears to cluster in familial BS as well. This further supports the notion that the pathogenesis of BS may entail several distinct mechanisms resulting in separate phenotype clusters.
OBJECTIVE: We have previously reported distinct symptom clusters among our patients with Behçet's syndrome (BS). The presence of such clusters suggests that more than one pathogenetic mechanism might be operative in BS. Increases in the frequency of certain clusters in familial BS cases, if present, would further support this notion. To test this hypothesis, we compared the frequency of symptom clusters between familial (group F) and non-familial (group NF) cases of BS. METHODS: We identified 380 BS patients who had reported a first-degree relative by reviewing 6031 patient charts. We were able to contact 186 (Group F). From the same initial pool, 500 patients were randomly selected. Of those, patients who did not report a family history of BS and who had attended our clinic during the previous 3 months made up group NF (n = 221). Both groups were questioned about their symptoms within the previous 3 months. Data were analysed using factor analysis, cluster analysis and χ2 tests. RESULTS: The make-up of the symptom clusters were very similar for the factor and the cluster analyses. The frequency of papulopustular lesions and joint involvement cluster was significantly higher in group F (39.2 vs. 21.5%, P < 0.001). Furthermore, the same cluster was shared in 5/17 related pairs from group F and in only 5/110 unrelated pairs from group NF [29 vs. 4.5%, P = 0.004; risk ratio (RR) = 6.47, 95% CI 2.15, 18.89]. CONCLUSION: The papulopustular lesions and arthritis cluster in BS appears to cluster in familial BS as well. This further supports the notion that the pathogenesis of BS may entail several distinct mechanisms resulting in separate phenotype clusters.