Development of a clinical prediction model for the postthrombotic syndrome in a prospective cohort of patients with proximal deep vein thrombosis

Abstract

Background Postthrombotic syndrome (PTS) is a chronic complication that develops in 20-50% of patients after deep vein thrombosis (DVT). Although individual risk factors for PTS have been characterized, the ability to predict which DVT patients are likely to develop PTS remains limited. Objective To develop a clinical prediction score for PTS in patients with DVT. Methods The derivation cohort consisted of participants in the SOX Trial, a randomized double-blind placebo-controlled trial of elastic compression stockings versus placebo stockings worn for 2years after DVT to prevent PTS in patients with a first proximal DVT, enrolled in 24 community and tertiary-care hospitals from 2004 to 2010. Multivariable logistic regression analysis of baseline characteristics was performed. The outcome was the occurrence of PTS, diagnosed starting from 6months or later according to Ginsberg's criteria. Results Seven hundred and sixty-two patients were included in the analysis. The median follow-up was 728days. The model includes three independent predictors, and has a range of possible scores from 0 to 5. High-risk predictors were: index DVT in the iliac vein; body mass index of ≥35kgm-2 ; and moderate-severe Villalta severity category at DVT diagnosis. As compared with patients with a score of 0, those with a score of ≥4 had an odds ratio of 5.9 (95% confidence interval 2.1-16.6) for developing PTS. Conclusions To our knowledge, this is the first clinical prediction score for PTS. We identified three independent predictors that, when combined, predicted PTS risk after a first proximal DVT. The SOX-PTS score requires external validation before it can be considered for clinical use.