Lack of Evidence to Support Thromboprophylaxis in Hospitalized Medical Patients with Cancer

Background

The administration of anticoagulant thromboprophylaxis for all patients with cancer who are hospitalized for acute medical illness is considered standard practice and strongly recommended in clinical guidelines. These recommendations are extrapolated from randomized controlled prophylaxis trials not specifically conducted in cancer cohorts. Because hospitalized patients with cancer constitute a unique population with increased risk of venous thromboembolic events and major hemorrhage, validation of the efficacy and safety of primary thromboprophylaxis in this population is critical. We sought to summarize the rates of venous thromboembolic events and major bleeding episodes among hospitalized patients with cancer who were receiving anticoagulant therapy compared with placebo.

Methods

A systematic literature search strategy was conducted using MEDLINE, EMBASE, and the Cochrane Register of Controlled Trials. Two reviewers independently extracted data onto standardized forms. The primary end points were all venous thromboembolic events. Secondary end points included major bleeding episodes and symptomatic venous thromboembolic events. Pooled analysis with relative risk using a random effect model was used as the primary measurement.

Results

A total of 242 citations were identified by the literature search. Of these, 3 placebo-controlled randomized trials included venous thromboembolic events as a primary outcome and were analyzed according to cancer subgroups. The pooled relative risk of venous thromboembolic events was 0.91 (95% confidence interval, 0.21-4.0; I²: 68%) among hospitalized patients with cancer who were receiving thromboprophylaxis compared with placebo. None of the trials reported the rates of symptomatic venous thromboembolic events or major bleeding episodes according to cancer status.

Conclusions

The risks and benefits of primary thromboprophylaxis with anticoagulant therapy in hospitalized patients with cancer are not known. This is especially relevant because numerous Medicare-type pay-for-performance incentives mandate prophylaxis specifically in patients with cancer.