D-Dimer Measured at Diagnosis of Venous Thromboembolism is Associated with Risk of Major Bleeding.

Sigrid Brækkan,Vânia Morelli,Ellen Brodin,Håkon Johnsen,Esben Bjøri,Kristian Hindberg,John-Bjarne Hansen

doi:10.1055/s-0039-1683395

Abstract

Identification of patients at risk of major bleeding is pivotal for optimal management of anticoagulant therapy in venous thromboembolism (VTE). Studies have suggested that D-dimer may predict major bleeding during anticoagulation; however, this is scarcely investigated in VTE patients. We aimed to investigate the role of D-dimer, measured at VTE diagnosis, as a predictive biomarker of major bleeding. The study population comprised 555 patients with a first community-acquired VTE (1994–2016), who were identified among participants from the Tromsø study. Major bleeding events were recorded during the first year after VTE and defined according to the criteria of the International Society on Thrombosis and Haemostasis. Cox-regression was used to calculate hazard ratios (HRs) with 95% confidence intervals (CIs) adjusted for age, sex, and duration of anticoagulant therapy. In total, 29 patients experienced major bleeding (incidence rate: 5.7/100 person-years, 95% CI: 4.0–8.2). The major bleeding risk was highest during the first 3 months, especially in patients with D-dimer ≥8.3 µg/mL (upper 20th percentile), with 28.8 major bleedings/100 person-years (95% CI: 13.7–60.4). Patients with D-dimer ≥8.3 µg/mL had a 2.6-fold (95% CI: 1.1–6.6) higher risk of major bleeding than patients with D-dimer ≤2.3 µg/mL (lower 40th percentile). Major bleeding risk according to D-dimer ≥8.3 versus ≤2.3 µg/mL was particularly pronounced among those with deep vein thrombosis (HR: 4.6, 95% CI: 1.3–16.2) and provoked events (HR: 4.2, 95% CI: 1.0–16.8). In conclusion, our results suggest that D-dimer measured at diagnosis may serve as a predictive biomarker of major bleeding after VTE, especially within the initial 3 months.

Highlights

Anticoagulant therapy (AT) is the cornerstone in the treatment of venous thromboembolism (VTE)
D-dimer levels were in the ranges of 2.3, 2.4 to 8.2, and !8.3 μg/mL in the lowest (Q1–2), middle (Q3–4), Table 1 Baseline characteristics of venous thromboembolism (VTE) cases across categories of D-dimer
Sensitivity analyses restricted to the time on anticoagulant treatment showed essentially similar results (►Supplementary Table S3 and ►Supplementary Fig. S1). In this population-based cohort study of patients with a first lifetime community-acquired VTE, we found that patients with high D-dimer levels (!8.3 μg/mL) had 2.6-fold higher risk of a major bleeding (MB) during the 1 year of follow-up compared with those with D-dimer levels 2.3 μg/mL

Summary

Introduction

Anticoagulant therapy (AT) is the cornerstone in the treatment of venous thromboembolism (VTE). Extended AT effectively prevents recurrent events, but at the cost of bleeding complications.[1,2,3] The reported annual risk of major bleeding (MB) varies in the range of 1 to 4%4–6 and is dependent on the choice of anticoagulant, intensity of anticoagulation, and duration of treatment.[6,7,8] The MB risk is high within the first months of AT,[2] with a case-fatality rate of 11% during the initial 3 months of anticoagulation.[9] received September 24, 2018 accepted after revision February 1, 2019. Even though initially promising in derivation studies,[10,11,12] prediction models developed to stratify risk of MB in VTE patients have demonstrated inconsistent discriminative powers in validation studies.[13,14,15,16] The existing prediction models mainly apply the same traditional predictors for bleeding such as age, history of bleeding, previous stroke, and cancer.[10,11,12] Identification of novel predictors for bleeding in VTE patients is an essential step for the development of a more accurate prediction score for MB, capable of guiding clinical decision-making in the future

Objectives

Methods

Conclusion