Long-term anticoagulation after first unprovoked VTE associated with considerable bleeding risk

Patients with a first unprovoked venous thromboembolism (VTE) have significant long-term risks and consequences of major bleeding related to anticoagulation, a systematic review and meta-analysis found.

Researchers reviewed the literature through July 2021 for randomized controlled trials and prospective cohort studies that looked at major bleeding among patients who had a first unprovoked VTE and received oral anticoagulants for at least six months after finishing at least three months of anticoagulant treatment. The primary outcome was a first major bleeding event, and secondary outcomes were intracranial bleeding and fatal bleeding. Results were published Sept. 14 by Annals of Internal Medicine.

Fourteen randomized trials and 13 cohort studies were included in the analysis. Overall, 9,982 patients received a vitamin K antagonist (VKA) and 7,220 received a direct oral anticoagulant (DOAC). The incidence of major bleeding was 1.74 per 100 person-years (95% CI, 1.34 to 2.20 per 100 person-years) with VKAs and 1.12 per 100 person-years (95% CI, 0.72 to 1.62 per 100 person-years ) with DOACs. The five-year cumulative incidence of major bleeding with VKAs was 6.3% (95% CI, 3.6% to 10.0%). The researchers noted that data were insufficient to estimate incidence of major bleeding beyond one year with DOACs. The case-fatality rate from major bleeding was 8.3% (95% CI, 5.1% to 12.2%) with VKAs and 9.7% (95% CI, 3.2% to 19.2%) with DOACs.

Among patients receiving either type of anticoagulant, the incidence of major bleeding was statistically significantly higher in those who were older than age 65 years (incidence rate ratio [IRR], 1.84 [95% CI, 1.32 to 2.57] with VKAs and 2.92 [95% CI, 1.50 to 5.70] with DOACs). Risk was also increased in those with a creatinine clearance less than 50 mL/min, a history of bleeding, concomitant use of antiplatelet therapy, or a hemoglobin level less than 100 g/L.

The incidence of major bleeding was statistically significantly higher among women than men in those receiving a VKA (IRR, 1.55; 95% CI, 1.17 to 2.06). Patients receiving a DOAC had no statistically significant difference by sex (IRR, 1.00; 95% CI, 0.51 to 1.95). The five-year cumulative incidence of major bleeding with VKAs was 8.5% (95% CI, 4.0% to 15.0%) in women and 6.5% (95% CI, 4.2% to 9.2%) in men, 8.3% (95% CI, 4.0% to 14.3%) among patients older than age 65 years, and 4.4% (95% CI, 2.5% to 6.7%) in those ages 65 years or younger.

There was no statistically significant difference in the incidence of major bleeding between patients with initial isolated proximal deep venous thrombosis (DVT) and those with isolated pulmonary embolism (PE) or between patients with isolated PE and those with concomitant PE and DVT.

“Taken together, our results provide clinicians, patients, and policymakers with a management framework in which to consider the long-term risks for and consequences of major bleeding if anticoagulation is continued beyond the initial 3 to 6 months,” the authors wrote. It has been proposed that patients with a major bleeding risk of 3% or higher per year be classified as at high risk for bleeding high risk and that they not be considered for indefinite anticoagulation, regardless of their risk for recurrent VTE, the authors noted. However, no standardized approach currently exists to identify such a subgroup of patients with unprovoked VTE, they said.

“When weighed against the long-term risks for and consequences of recurrent VTE if anticoagulation is discontinued, our results could be used to balance the benefits and harms of extended anticoagulation for unprovoked VTE,” the authors wrote.