D-dimer levels over time after anticoagulation and the association with recurrent venous thromboembolism

Abstract

Introduction: D-dimer measured shortly after discontinuation of anticoagulation by an immunoturbidimetric assay predicts the risk of recurrent venous thromboembolism (VTE). We assessed the performance of this assay over time and its association with recurrent VTE.

Materials and methods: We followed 556 patients with a first VTE for a median of 9.6 years. The study end point was recurrent VTE. D-dimer was measured 3 weeks, and 3, 9, and 15 months after discontinuation of anticoagulation in plasma using an immunoturbidimetric assay (INNOVANCE D-Dimer). To estimate the effect of longitudinal D-dimer on the recurrence risk, we used a dynamic prediction Cox model with landmark times (3 weeks, and 3, 9, 15 months) as a stratification factor.

Results: 135 patients had recurrent VTE. D-dimer levels varied between patients but without a consistent pattern. Levels increased slightly over time [0.7% increase (95% CI: 0.5-0.9; p < 0.001)/month]. D-dimer levels were positively correlated with body mass index (BMI) [2% (95% CI: 1.1-2.9; p < 0.001) increase/1 unit BMI increase], and were 14.8% (95% CI: 5.1-25.3; p = 0.002) higher in women than in men. The recurrence risk with doubling D-dimer levels was higher after 3 weeks, 3, 9 and 15 months [hazard ratios 1.4 (1.06-1.84), 1.37 (1.06-1.77), 1.31 (1.04-1.65) and 1.26 (1.01-1.57), respectively].

Conclusions: In patients with a first VTE, immunoturbidimetric D-dimer levels are associated with the risk of recurrence at multiple times points from 3 weeks up to 15 months after discontinuation of oral anticoagulation.

Keywords: D-dimer; Recurrence risk; Venous thromboembolism.