fetching data ...

AB0055 (2024)
DISEASE PROFILES IN A BEHÇET’S DISEASE MONOCENTRIC COHORT - A NEURAL NETWORK ANALYSIS
Keywords: Skin, Artificial Intelligence, Gastrointestinal tract, Uveitis, Observational studies/ registry
F. DI Cianni1, V. Lorenzoni2, N. Italiano3, D. Marinello1, M. Mosca3, R. Talarico3
1Azienda Ospedaliero-Universitaria Pisana, Pisa, Italy
2Scuola Superiore di Studi Universitari e di Perfezionamento Sant’Anna, Institute of Management, Pisa, Italy
3University of Pisa, Pisa, Italy

Background: Behçet’s syndrome (BS) is a rare systemic vasculitis of unknown aetiology. The onset of disease typically involves young adults, the clinical spectrum is variable and characterized by a relapsing-remitting course. Notably, BS organ involvements are frequently clustered rather than discrete manifestations. Indeed, recently the concept has emerged that BS may not be a single nosological entity but a multi-system complex disorder of distinct clinical profiles.[1,2] Clinical profiles analysis might help to clarify the pathogenesis of single disease manifestations and to classify the profiles according to the clinical outcome.


Objectives: The aim of the present work was to identify disease profiles of a mono-centric cohort of patients affected by BS, evaluating the associations between clinical, epidemiological and therapeutical variables.


Methods: Patients referring to the Behçet Clinic of a tertiary care centre were included in the study. Demographics, clinical features and pharmacological treatment at last evaluation were retrospectively and prospectively collected. The presence of HLA-B51 was also included in the analysis. A multiple correspondence analysis (MCA) was performed to evaluate the patterns of association between the variables, obtaining distinct disease profiles.


Results: 202 patients were included in the analysis, most of the patients were female (67%), Caucasian (92%) and negative for HLA-B51 (67%, 76/113 patients for whom HLA-B51 was known). Mean age at disease onset was 28 years and oral and genital ulcerations were the most common manifestations (91,6% and 57% respectively), while neurological and vascular involvement were less frequent (8,4% and 11,8% respectively). Three disease profiles were obtained applying MCA: A) male and Caucasian subjects with oral ulcerations plus any other organ involvement, B) Caucasian subjects with bipolar ulcerations, papulopustular lesions, arthralgias and gastrointestinal involvement, without vascular and neurological involvement, and C) male subjects with bipolar ulcerations and ocular involvement, without neurological involvement.


Conclusion: The disease profiles from our cohort agree with the well-known clinical heterogeneity of BS. However, in profile A) oral ulcerations are associated with the presence of any other organ involvement, apparently without a clearly recognizable pattern of association. This profile may capture borderline clinical presentations which could constitute a distinct pathological entity, rather than a real BS disease profile. Future research is necessary to evaluate the clinical outcomes according to distinct disease profiles.


REFERENCES: [1] Yazici, H., Seyahi, E., Hatemi, G. & Yazici, Y. Behçet syndrome: a contemporary view. Nat Rev Rheumatol 14 , 107–119 (2018).

[2] Hatemi, G. et al. Behçet’s syndrome: one year in review 2022. Clinical and Experimental Rheumatology (2022) doi:10.55563/clinexprheumatol/h4dkrs.

Cohort distribution of patterns applying MCA


Acknowledgements: NIL.


Disclosure of Interests: None declared.


DOI: 10.1136/annrheumdis-2024-eular.904
Keywords: Skin, Artificial Intelligence, Gastrointestinal tract, Uveitis, Observational studies/ registry
Citation: , volume 83, supplement 1, year 2024, page 1255
Session: Behcet`s disease (Publication Only)