Duration of primary/secondary treatment to prevent recurrent venous thromboembolism: a systematic review and meta-analysis

Abstract

Antithrombotic therapy can prevent recurrent deep vein thrombosis (DVT) and pulmonary embolism (PE). It is, however, associated with an increased risk for major bleeding. This meta-analysis systematically reviewed the evidence regarding the duration of antithrombotic therapy to assess benefits and harms. We systematically searched for randomized controlled trials (RCTs) that compared shorter (3-6 months) with longer (>6 months) courses of anticoagulation for the primary treatment of venous thromboembolism (VTE) or that compared discontinued with indefinite antithrombotic therapy for the secondary prevention of VTE. Pairs of reviewers screened the eligible trials and collected data. This study included 22 RCTs (11 617 participants). Pooled estimates showed that, for the primary treatment of unprovoked VTE, VTE provoked by chronic risk factors or transient risk factors, treating patients with a longer course (>6 months) of anticoagulation, as opposed to a shorter course (3-6 months), probably reduced recurrent PE (risk ratio [RR], 0.66; 95% confidence interval [CI], 0.42-1.02) and DVT (RR, 0.85; 95% CI, 0.63-1.14), but it was associated with increased mortality (RR, 1.43; 95% CI, 0.85-2.41) (moderate certainty) and a higher risk for major bleeding (RR, 2.02; 95% CI, 1.02-3.98; high certainty). For the secondary prevention of unprovoked VTE and VTE provoked by chronic risk factors, when compared with discontinuing treatment, indefinite anticoagulation therapy was associated with decreased mortality (RR, 0.54; 95% CI, 0.36-0.81), a reduction in recurrent PE (RR, 0.25; 95% CI, 0.16-0.41) and DVT (RR, 0.15; 95% CI, 0.10-0.21), and an increase in the risk for bleeding (RR, 1.98; 95% CI, 1.18-3.30), all supported by high certainty. Indefinite antiplatelet therapy may be associated with decreased mortality (RR, 0.95; 95% CI: 0.53-1.68; low certainty), probably a reduction in recurrent PE (RR, 0.65; 95% CI, 0.41-1.03) and DVT (RR, 0.44; 95% CI, 0.17-1.13) (moderate certainty), and may increase the risk for bleeding (RR, 1.28; 95% CI, 0.48-3.41; low certainty). In summary, for the primary treatment of all types of VTE, shorter (3-6 months) duration of anticoagulation is more beneficial. For the secondary prevention of unprovoked VTE or VTE provoked by chronic risk factors, indefinite antithrombotic treatment is more beneficial.

Figure 1.

Time frame for making decisions regarding anticoagulant treatment. Time frame of the decisions. Initial management (yellow box) spans the first 5 to 21 days following a diagnosis of a new VTE and includes issues concerning whether the patient can be treated at home or it they require admission to the hospital, the use of thrombolytic therapy, whether an inferior vena cava filter needs to be placed, and the initial anticoagulant therapy. During primary treatment, anticoagulant therapy is continued for 3 to 6 months total and represents the minimal duration of treatment for the VTE. After completion of the primary treatment, the next decision concerns whether anticoagulant therapy will be discontinued or if it will be continued for secondary prevention of recurrent VTE. Typically, secondary prevention is continued indefinitely, although patients should be reevaluated on a regular basis to review the benefits and risks of continued anticoagulant therapy. Our choice of terminology reflects the distinct clinical intentions of the different phases of VTE management, linking them to important clinical decisions addressed in the guidelines rather than using terms that reflect the relative duration of therapy. Adapted from the ASH 2020 guidelines for the management of VTE.

Figure 2.

Primary treatment: shorter duration (3-6 months) vs longer duration (>6 months) of anticoagulation. Risk of bias legend: A, random sequence generation (selection bias); B, allocation concealment (selection bias); C, blinding of participants and personnel (performance bias); D, blinding of outcome assessment (detection bias); E, incomplete outcome data (attrition bias); F, selective reporting (reporting bias); G, other bias.

Figure 2.

Primary treatment: shorter duration (3-6 months) vs longer duration (>6 months) of anticoagulation. Risk of bias legend: A, random sequence generation (selection bias); B, allocation concealment (selection bias); C, blinding of participants and personnel (performance bias); D, blinding of outcome assessment (detection bias); E, incomplete outcome data (attrition bias); F, selective reporting (reporting bias); G, other bias.

Figure 3.

Secondary prevention: indefinite vs discontinued antithrombotic therapy. ∗Trials using DOAC. Risk of bias legend: A, random sequence generation (selection bias); B, allocation concealment (selection bias); C, blinding of participants and personnel (performance bias); D, blinding of outcome assessment (detection bias); E, incomplete outcome data (attrition bias); F, selective reporting (reporting bias); G, other bias.

Figure 3.

Secondary prevention: indefinite vs discontinued antithrombotic therapy. ∗Trials using DOAC. Risk of bias legend: A, random sequence generation (selection bias); B, allocation concealment (selection bias); C, blinding of participants and personnel (performance bias); D, blinding of outcome assessment (detection bias); E, incomplete outcome data (attrition bias); F, selective reporting (reporting bias); G, other bias.