Review: Extended therapy with DOACs, but not VKAs, reduces mortality after VTE treated with initial anticoagulation

Abstract

Question In patients with acute venous thromboembolism (VTE) treated with initial anticoagulation, what are the effects of extended therapy with vitamin K antagonists (VKAs) and direct oral anticoagulants (DOACs)? Review scope Included studies compared extended VKA or DOAC therapy with placebo, aspirin, or no treatment after initial anticoagulation for 3 months in adults with deep venous thrombosis (DVT) and/or pulmonary embolism (PE). Studies of VKA therapy with a target international normalized ratio of 2.0 to 3.0 and low- or standard-dose DOACs were included. Outcomes included all-cause mortality, VTE mortality, VTE recurrence, major bleeding, and a composite of recurrent symptomatic VTE or major bleeding (net clinical benefit). PROSPERO CRD42018088739. Review methods MEDLINE, EMBASE/Excerpta Medica, Cochrane Library (all 1990 to Sep 2018); reference lists; and gray literature were searched for randomized controlled trials (RCTs). 16 RCTs (n =12458, mean age 54 to 69 y, 40% to 68% women, 41% to 100% unprovoked VTE, mean treatment duration 12 mo when reported), with 34 to 3396 patients, met the selection criteria. Mean follow-up was 12 to 38 months. Random sequence generation and allocation concealment were adequate in 14 RCTs, and blinding of patients and clinicians was adequate in 15 trials. Main results The results are in the Table. Conclusion In patients with acute VTE treated with initial anticoagulation, extended therapy with DOACs, but not VKAs, reduces all-cause mortality. Vitamin K antagonists (VKAs) or direct oral anticoagulants (DOACs) vs control after initial anticoagulation in patients with acute venous thromboembolism (VTE)* Outcomes Number of trials reporting outcome events and included in analysis (n) Events/100 patient-y Mean 12 to 38 mo VKA Control RRR (95% CI) NNT (CI) VKA vs control VTE recurrence 9 (2654) 2.45 11.46 78% (59 to 88) 8 (5 to 12) Recurrent symptomatic VTE or major bleeding 11 (3518) 4.43 11.32 54% (28 to 70) 16 (10 to 51) RRI (CI) NNH (CI) All-cause mortality 5 (1353) 1.42 1.88 17% (36 to 113) Not significant VTE mortality 3 (1298) 1.63 0.78 42% (64 to 467) Not significant Major bleeding 9 (2799) 2.89 1.66 167% (28 to 460) 69 (34 to ) DOAC vs control DOAC Control RRR (CI) NNT (CI) VTE recurrence 4 (8432) 1.49 7.94 80% (69 to 87) 14 (10 to 24) Recurrent symptomatic VTE or major bleeding 4 (8432) 1.91 8.16 75% (61 to 84) 15 (11 to 26) All-cause mortality 4 (8432) 0.52 0.84 52% (14 to 73) 196 (112 to 833) VTE mortality 4 (8432) 0.20 0.46 64% (11 to 85) 278 (164 to 833) RRI (CI) NNH (CI) Major bleeding 4 (8432) 0.48 0.35 59% (58 to 495) Not significant *Abbreviations defined in Glossary. RRR, RRI, and CI calculated from relative risks in article using a random-effects model. This composite is reported as net clinical benefit in the article. Commentary VTE is a common and potentially fatal disease with a high rate of recurrence in the first 3 months. The 2016 CHEST guidelines (1) have strong (Grade 1b) recommendations for anticoagulant therapy during this 3-month period. How to manage patients after the 3-month period is less clear. The systematic review by Mai and colleagues assessed the effects of extended anticoagulation beyond 3 months in VTE. The review included most of the same trials considered by the guideline writers in 2016. However, unlike the meta-analyses that informed the guideline, the review pooled the 4 DOAC trials (that included 3 different medications), producing a larger sample size and allowing identification of a survival benefit for the drug class. The review also made indirect comparisons of DOACs with VKAs. Such comparisons should be relied on only if the populations are similar, but the DOAC and VKA trials in the review may have included different populations. For example, some DOAC trials excluded use of nonsteroidal antiinflammatory drugs, dual-antiplatelet therapy, and aspirin >165 mg and patients with anemia. Moreover, all-cause mortality may be higher in the VKA trials than the DOAC trials (1.88 vs 0.84 deaths/100 person-y in the control groups), suggesting that the populations included in trials of each drug category may have differed. Most patients in the included trials had unprovoked VTE. Such patients, particularly those with other risk factors for recurrence, should be strongly considered for indefinite therapy. Patients with high bleeding risk were excluded from many RCTs. Thus, we must carefully weigh the risks and benefits, incorporate patient values and preferences into treatment decisions, and regularly reevaluate anticoagulation as clinical conditions change.