Daniel M. Witt, PharmD, FCCP, BCPS, CACP
Is weekly international normalized ratio (INR) self-testing better than monthly testing in a high-quality clinic for reducing a composite of stroke, major bleeding, or death?
Randomized controlled trial (The Home INR Study [THINRS]). ClinicalTrials.gov NCT00032591.
Blinded (endpoint adjudication group for primary and most secondary endpoints).*
≥ 2 years (mean 3 y).
28 Veterans Affairs medical centers in the USA that had anticoagulation clinics with guideline-consistent, high-quality anticoagulation management and were treating ≥ 400 patients.
2922 patients (mean age 67 y, 98% men) who had atrial fibrillation and/or a mechanical heart valve, needed long-term warfarin therapy, and were evaluated as competent to use INR self-tests (ProTime Microcoagulation System) after training.
Weekly home-based self-testing of INRs using an approved device, interactive voice–response reporting, and Web-based local monitoring (n = 1465); or monthly INR testing in a high-quality clinic (n = 1457).
Composite endpoint of stroke, major bleeding, or death. Other outcomes included minor bleeding, time within INR target range, and quality of life (Health Utilities Index Mark 3). 3200 patients and 363 primary events were needed to detect a 32% relative reduction from 5.5% in annual rate of the composite endpoint (90% power, α = 0.05).
99% (intention-to-treat analysis).
Weekly INR self-testing and monthly clinic INR testing did not differ for the composite endpoint; INR self-testing increased risk for minor bleeding (Table). Patients in the self-test group spent more time within the target therapeutic INR range (mean 66% vs 62%, P < 0.001) and had better quality of life at 2 years (mean cumulative gain in health utilities, 1.2 y vs 1.0 y, P < 0.001).
In patients receiving warfarin, weekly INR self-testing did not reduce a composite of stroke, major bleeding, or death more than monthly clinic INR testing.
Weekly INR self-testing vs monthly clinic INR testing in patients receiving warfarin†
†Abbreviations defined in Glossary. RRR, RRI, NNT, NNH, and CI calculated from data in article.
‡For self-testing vs clinic testing: death (10% vs 11%, P = 0.41), stroke (2.12% vs 2.13%, P = 0.83), and major bleeding (10% vs 9.8%, P = 0.83). Analyses of stroke and major bleeding included death as a competing risk.
Daniel M. Witt. Weekly INR self-testing did not reduce stroke, major bleeding, or death more than monthly clinic testing. Ann Intern Med. 2011;154:JC1–4. doi: 10.7326/0003-4819-154-2-201101180-02004
Download citation file:
Published: Ann Intern Med. 2011;154(2):JC1-4.
Results provided by:
Copyright © 2017 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use