Pos0924 Cluster Analysis for Identification of Clinical Phenotypes in Behçet’s Disease in a Cohort From Northern Spain

Background: Behçet's disease (BD) is characterized by heterogeneity of clinical manifestations due to involvement of variable blood vessels. Objectives: to identify clinical phenotypes using cluster analysis. Methods: a model-based clustering relaying on 16 clinical variables was performed in a retrospective cohort of 120 BD patients, diagnosed and follow-up from January 1, 1980 to December 31, 2019 in 3 hospital at Northern Spain (Cantabria). Chi-square test and ANOVA were used to compare categorical and continuous variables among groups. Two-sample t-tests and the partition of Pearson's chi-square statistic were used in pairwise comparisons. Results: Four groups were identified in this study: C1 (n=47; 39.2%), C2 (n=33; 27.5%), C3 (n=33; 27.5%) and C4 (n=7; 5.8%). The clusters were defined as follows: C1 as mucosal involvement, C2 as joint involvement and pseudofolliculitis, C3 as neurological and vascular involvement and C4 as uveitis and erythema nodosum. No baseline demographic differences between clusters and no differences in the application of classification criteria were observed (Table 1). Oral ulcers were predominant in all clusters, ranging from 85.7% (C4) to 95-97% (C1-C2), with no significant differences. However, a higher frequency of genital ulcers was observed in C1 compared to the other clusters. Similarly, a significantly higher frequency of erythema nodosum was observed in C4 (85.7%) and of pseudofolliculitis in C2 (93.9%) (pFigure 1). Conclusion: we have defined 4 BD phenotypes in our study. This cluster approach may be useful for better patient management due to early identification of clinical patterns. REFERENCES: NIL. Acknowledgements: NIL. Disclosure of Interests: None declared.