Daniel M. Witt, PharmD, FCCP, BCPS, CACP
In patients with a first unprovoked venous thromboembolism (VTE), do D-dimer levels after cessation of anticoagulation therapy predict risk for recurrence? Is prediction affected by timing of testing, test cutpoint, or patient age?
Included studies evaluated patients who had a first symptomatic VTE, included some patients with unprovoked VTE (proximal venous thrombosis, pulmonary embolism, or both), treated patients with standard anticoagulation (heparin for 5 to 10 d and vitamin K antagonist for ≥ 3 mo), and measured D-dimer levels after anticoagulation was stopped. Patients with unprovoked VTE were included in the individual-patient data analysis. Exclusion criteria were distal venous thrombosis, random allocation to a group restarting anticoagulation, antithrombin deficiency, or presence of antiphospholipid antibodies. Outcome was symptomatic, objectively confirmed recurrent VTE.
MEDLINE, EMBASE/Excerpta Medica, CINAHL, and Cochrane Central Register of Controlled Trials were searched up to July 2010 for randomized trials or prospective cohort studies. Investigators were contacted and provided individual patient data. 2 randomized controlled trials and 5 prospective cohort studies met the selection criteria; 1818 (mean age 59 y, 51% men) of 1863 patients with first unprovoked VTE were included in the individual patient data meta-analysis. Mean follow-up after cessation of anticoagulation was 27 months. Cutpoints for positive D-dimer results were defined by individual studies.
Individual patient data meta-analyses showed that positive D-dimer results after cessation of anticoagulation were associated with increased risk for recurrent VTE (Table); results did not differ by patient age or timing of D-dimer test (Table). In a sensitivity analysis, results did not differ by D-dimer test cutpoint (250 µg/L or 500 µg/L).
In patients with a first unprovoked venous thromboembolism, positive D-dimer test results after cessation of anticoagulation predict recurrence, regardless of test timing or cutpoint, or patient age.
Individual patient data analysis of association of D-dimer results after cessation of anticoagulation with risk for recurrence in patients with a first unprovoked venous thromboembolism*
*CI defined in Glossary.
†Adjusted for study.
‡Stratified by study and adjusted for age, body mass index, sex, use of hormonal therapy prior to index event, thrombophilia status, timing of post-anticoagulation D-dimer testing, and duration of anticoagulant therapy.
§Corrected data provided by author.
Witt DM. Review: Positive D-dimer results predict recurrence after first unprovoked VTE regardless of test timing or cutpoint, or patient age. Ann Intern Med. ;154:JC1–5. doi: 10.7326/0003-4819-154-2-201101180-02005
Download citation file:
Published: Ann Intern Med. 2011;154(2):JC1-5.
Results provided by:
Copyright © 2018 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use