Review: Self-testing or self-management of oral anticoagulation reduces thromboembolic events and mortality more than usual care

Is patient self-testing (PST), alone or combined with self-adjustment of drug doses (patient self-management [PSM]), more effective and safer than usual care in patients receiving long-term oral anticoagulant therapy?

Included studies evaluated PST or PSM in adult outpatients who were receiving vitamin K antagonists for > 3 months. Outcomes were thromboembolic events, mortality, major bleeding, and percentage of time within international normalized ratio (INR) therapeutic range (TTR).

MEDLINE (2005 to Oct 2010), Cochrane Central Register of Controlled Trials, reference lists of relevant articles, and a 2007 Health Technology Assessment report were searched for English-language, randomized, controlled trials (RCTs) published in peer-reviewed journals after 1995. 22 RCTs (n = 8413, mean age 65 y, mean 75% men) met the selection criteria. 5 RCTs met 4 quality criteria (adequate allocation concealment, blinding of endpoint assessors, number of withdrawals and dropouts reported, and use of intention-to-treat analysis); 12 were fully or partially funded by industry, and 4 did not report funding sources; 13 had follow-up < 12 months. 11 RCTs used warfarin in all patients; 7 used phenprocoumon, acenocoumarol, or fluindione; 1 used both warfarin and phenprocoumon; and 3 did not report the drug used. 5 RCTs compared PST with usual care in a physician’s office or anticoagulation clinic, 14 compared PSM with usual care, 1 compared PST with PSM, 1 included 4 treatment groups (PST, PSM, and usual care with or without education), and the intervention was unclear in 1 RCT.

Meta-analysis showed that PST or PSM reduced major thromboembolic events and all-cause mortality more than usual care (Table); groups did not differ for major bleeding (Table) or percentage of TTR (pooled weighted mean 66% vs 62%, P = 0.17).

In patients receiving long-term oral anticoagulant therapy, patient self-testing or self-management reduces thromboembolic events and mortality more than usual care and does not increase major bleeding.