Value of D‐dimer testing to decide duration of anticoagulation after deep vein thrombosis: not yet

T. Baglin

DOI: https://doi.org/10.1111/j.1538-7836.2006.02248.x

2006-12-01

Journal of Thrombosis and Haemostasis

Abstract:Following a first episode of venous thromboembolism (VTE), patients are usually treated with a finite period of oral anticoagulation of between six weeks and six months [1,2]. When treatment is stopped, the overall incidence of recurrence is between 5% and 20% after 2 years [3–6]. The case-fatality rate for recurrent VTE is about 5% [7]: recurrent deep vein thrombosis (DVT) increases the risk of post-thrombotic syndrome [8] and recurrent pulmonary embolus increases the risk of thromboembolic pulmonary hypertension [9]. Continued treatment with oral anticoagulant therapy will prevent most episodes of recurrence, but there is a significant risk of major bleeding associated with prolonged treatment [10,11]. In theory, anticoagulant therapy should be continued until the risk outweighs the benefit. However, the optimal duration of treatment for each patient is uncertain, because the risk of bleeding associated with anticoagulation and the risk of recurrent VTE after stopping treatment are not easily predicted on an individual basis. For this reason, the series of studies performed by Palareti et al., which examine the risk of recurrent VTE in relation to D-dimer measurement, is of considerable interest and importance. The studies, from observation to patient management, show that the measurement of D-dimer levels following cessation of anticoagulant therapy predicts the risk of recurrent VTE; see Palareti et al. [12,13] and PROLONG. An association between the D-dimer level after stopping anticoagulant therapy and the subsequent risk of VTE recurrence has been confirmed independently [14]. The current default position for treating patients after a first episode of VTE is a finite period of anticoagulation [1,2]. Therefore, if the measurement of D-dimer is going to inform clinical decision-making, and change clinical practice, it must have a useful positive predictive value (PPV). The negative predictive value (NPV) is less important as the default is to stop treatment. The crucial question is, therefore, how high does the PPV have to be? In 2002, Palareti and colleagues reported on the predictive value of D-dimer measurement (VIDAS ELISA; bioMérieux Inc., Durham, NC, USA) for the risk of VTE recurrence after withdrawal of oral anticoagulant therapy after a first episode of VTE. A D-dimer level of £ 500 ng mL measured 3 months after stopping therapy had a highNPVof 95.6%.However, the PPV was only 12.7% [12]. In a subsequent study, Palareti et al. [13] reported a PPV of 16.2%. In a separate study, Eichinger et al. [14] reported an 11.5% cumulative probability of recurrence at two years in patients with a D-dimer level of ‡ 250 ng mL, compared with 3.7% in patients with lower levels (measurement by Asserachrom ELISA; Boehringer Mannheim, Germany, 3 weeks after discontinuation of therapy). Recently, Palareti et al. have completed PROLONG, a multicenter, prospective randomized interventional management study. The aim of the study was to assess the risk of recurrence using the D-dimer measurement (Simplify D-dimer; Agen Biomedical, Brisbane, Australia) one month after completion of oral anticoagulant therapy after a first episode of VTE. Patients with a D-dimer level of> 500 ng mL were randomized to remain off therapy or to restart oral anticoagulant treatment. The cumulative probability of recurrence at two years in patients with a high D-dimer level who remained off therapy was 15%, compared with 6.2% in patients who restarted oral anticoagulation. PROLONG will also provide information on the overall benefit and risk of anticoagulation, as adverse bleeding events were recorded as well as recurrent VTE. From these studies there appears to be a consistent PPV of about 15%. Is this high enough? Acquired risk factors are often identified in patients presenting with VTE, and simple assessment of clinical risk factors associated with the first episode of VTE predicts risk of recurrence [3,4,12,15–18]. For example, in the prospective Cambridge cohort study of unselected patients with a first episode of VTE the cumulative probability of recurrent VTE at two years in patients with a first event precipitated by surgery was 0%. In contrast, the recurrence rate in patients with Correspondence: Trevor Baglin, Department of Haematology, Addenbrooke’s Hospital, Cambridge CB2 2QQ, UK. Tel.: +44 1223 216748; fax: +44 1223 256168; e-mail: trevor.baglin@addenbrookes.nhs.uk Journal of Thrombosis and Haemostasis, 4: 2530–2532