Phenotype-based risk stratification for deep vein thrombosis in Behçet's disease: a global 10-year longitudinal study

Vascular involvement, particularly deep vein thrombosis (DVT), is a major cause of morbidity and mortality in Behçet's disease (BD). This study aimed to identify independent risk factors for DVT and establish a robust framework for phenotype-based risk stratification.

We conducted a retrospective cohort study using the TriNetX global health research network and identified 19,739 patients with BD. Three independent propensity score-matched cohorts were established to evaluate the impact of intestinal involvement, thrombophlebitis, and antiphospholipid syndrome (APS). A multivariable Cox proportional hazards model was used to identify independent predictors of DVT over a 10-year follow-up. To ensure the robustness of our findings and minimise peri-diagnostic detection bias, a 90-day landmark analysis was performed as a sensitivity analysis.

Multivariable analysis identified APS (adjusted hazard ratio [aHR] 6.28, 95% confidence interval [CI] 4.56-8.64) and a history of thrombophlebitis (aHR 5.92, 95% CI 4.72-7.41) as the strongest independent predictors of DVT. In the 90-day landmark analysis, these associations remained significant, with persistent risk observed for APS (aHR 5.10) and thrombophlebitis (aHR 2.50), confirming their long-term predictive value. Additional independent predictors included intestinal involvement (defined as non-infective inflammatory colitis; aHR 1.51), obesity (aHR 1.98), glucocorticoid use (aHR 1.78), and male sex (aHR 1.77).

APS and thrombophlebitis, markers of heightened inflammatory and thrombotic activity at diagnosis, also appear to be persistent indicators of elevated DVT risk in BD. These findings support phenotype-based risk stratification and highlight the need for intensive monitoring and individualised preventive strategies to improve long-term outcomes.