The Effect of Warfarin Administration Time on Anticoagulation Stability (INRange)

A Pragmatic Randomized Controlled Trial

Scott R. Garrison, MD, PhD; Lee Green, MD, MPH; Michael R. Kolber, MD, MSc; Christina S. Korownyk, MD; Nicole M. Olivier, RVT; Balraj S. Heran, PhD; Mary E. Flesher, RD, MA; G. Michael Allan, MD

Disclosures

Ann Fam Med. 2020;18(1):42-49. 

In This Article

Abstract and Introduction

Abstract

Purpose: Without supporting evidence, clinicians commonly recommend that warfarin be taken in the evening. We conducted a randomized controlled trial to evaluate the effect of administration time (morning vs evening) on the stability of warfarin's anticoagulant effect.

Methods: A total of 236 primary care physicians serving 54 western Canadian communities mailed letters of invitation to all their warfarin-using patients. Eligible patients were community-dwelling warfarin users (any indication) with at least 3 months of evening warfarin use and no plans for discontinuation. Participants were randomized (by web-based allocation) to morning vs continued evening warfarin ingestion. We used the Rosendaal method to determine the proportion of time within therapeutic range (TTR) of the international normalized ratio (INR) blood test month 2 to 7 postrandomization vs the 6 months prerandomization. The primary outcome was the percent change in proportion of time outside target INR range (with an a priori minimum clinically important difference of ±20%). All analyses were intention to treat.

Results: Between March 8, 2015 and September 30, 2016, we randomized 109 participants to morning and 108 to evening warfarin use. TTR rose from 71.8% to 74.7% in the morning group, and from 72.6% to 75.6% in the evening group, for a change in TTR of 2.9% in the former vs 3.0% in the latter (difference, −0.1%; P = .97; 95% CI for the difference, −6.1% to 5.9%). The difference in percent change in proportion of time outside the therapeutic INR range (obtained via Hodges-Lehmann estimation of the difference in medians) was 4.4% (P = .62; 95% CI for the difference, −17.6% to 27.3%).

Conclusions: Administration time has no statistically significant nor clinically important impact on the stability of warfarin's anticoagulant effect. Patients should take warfarin whenever regular compliance would be easiest.

Introduction

Stroke and pulmonary embolism have devastating, often lifelong health consequences, and conditions that predispose to these events (atrial fibrillation, deep vein thrombosis, and mechanical heart valves) are common. Warfarin substantially reduces the risk of such thromboembolic events.[1] The safety and effectiveness of warfarin depends greatly on the proportion of time in the therapeutic range (TTR) of the international normalized ratio (INR) blood test, however.[2–5]

Most physicians and pharmacists recommend warfarin be taken in the early evening.[6–10] This strategy likely shortens the interval between learning of the need to make a dose adjustment (typically communicated to patients in the late afternoon following a morning blood test) and being able to make that dosing change. Hence, if evening warfarin use means quicker dose adjustments, it might conceivably lead to better TTR. Although the hypothesis is reasonable, there is no evidence to support this practice and other factors could meaningfully influence optimal administration time. For instance, dietary vitamin K (with which warfarin interacts) has an ultrashort 2.5-hour half-life and is found in foods (green leafy vegetables) having highly variable intake and rarely ingested in the morning.[11,12] Conceivably, ingesting warfarin in the morning, when vitamin K levels are more consistent, might lead to greater INR stability. Although patients are commonly advised to take warfarin in the evening, it is unclear whether administration time matters, and if it does, which time is best.

In this pragmatic primary care study, The Effect of Warfarin Administration Time on Anticoagulation Stability (INRange), we randomized established warfarin users taking the medication in the evening either to switch to morning warfarin use or to continue evening use, and examined the TTR to detect differences in the stability of warfarin's anticoagulant effect.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....