Clinical Question

What is the optimal treatment strategy for acute venous thromboembolism?

Bottom Line

This complex network meta-analysis of eight treatment regimens for acute venous thromboembolism found that a combination of unfractionated heparin and vitamin K antagonists is associated with the least effective strategy with the highest risk of recurrent events. Oral rivaroxaban (Xarelto) and apixaban (Eliquis) may be associated with the lowest risk of bleeding, but no overall significant differences occurred for effectiveness and safety compared with the combination of low-molecular-weight heparin (LMWH; Lovenox) and vitamin K antagonists. Rivaroxaban and apixaban have been compared head-to-head only with the traditional LMWH–vitamin K antagonist combination in three manufacturer-sponsored clinical trials (two of rivaroxaban; one of apixaban). (Level of Evidence = 1a)

Synopsis

These investigators performed a meta-analysis comparing the clinical outcomes and safety associated with eight different treatment regimens for acute venous thromboembolism, including deep venous thrombosis or pulmonary embolism. Multiple databases were searched, including Medline, Embase, the Cochrane Registry, the Health Technology Assessment, and references of included studies, for randomized trials that compared at least two of any of the eight various regimens with each other, but not with placebo. No language restrictions were applied. Two individuals independently evaluated potential studies for inclusion and assessed methodologic quality using a standard risk-of-bias scoring tool. Differences were resolved by consensus agreement. The primary outcomes measured included recurrent venous thromboembolism events and major bleeding episodes of clinical significance.

A total of 45 articles (N = 44,989 patients) met study inclusion criteria, including 22 trials that compared an unfractionated heparin–vitamin K antagonist combination with an LMWH–vitamin K antagonist combination; 12 that compared an unfractionated heparin–vitamin K antagonist combination with LMWH alone; three that compared an LMWH–vitamin K antagonist combination with LMWH alone; two that compared a fondaparinux (Arixtra)–vitamin K antagonist combination with an LMWH–vitamin K antagonist combination or an unfractionated heparin–vitamin K antagonist combination; and six that compared an LMWH–vitamin K antagonist combination with one of the direct oral anticoagulants: two with dabigatran (Pradaxa), one with apixaban, one with edoxaban, and two with rivaroxaban.

Follow-up occurred for a median of three months. Compared with an LMWH–vitamin K antagonist combination, all treatment strategies except an unfractionated heparin–vitamin K antagonist combination resulted in a similarly lower rate of recurrent venous thromboembolism events. The unfractionated heparin–vitamin K antagonist combination was associated with a significantly increased rate of recurrent venous thromboembolism events (number needed to treat to harm = 188) compared with an LMWH–vitamin K antagonist combination. Compared with an LMWH–vitamin K antagonist combination, the risk of a major bleeding episode was statistically lower with rivaroxaban (number needed to treat = 258) and apixaban (number needed to treat = 165). All of the other treatment regimens were associated with a similar risk of adverse bleeding events compared with an LMWH–vitamin K antagonist combination. Apixaban was associated with the greatest overall probability of being the least harmful therapy, although it was evaluated in only one manufacturer-sponsored trial.

Study design: Meta-analysis (randomized controlled trials)

Funding source: Foundation

Setting: Various (meta-analysis)

Reference: CastellucciLACameronCLe GalGet alClinical and safety outcomes associated with treatment of acute venous thromboembolism: a systematic review and meta-analysis. JAMA.2014; 312( 11): 1122– 1135.