OVWAC VII

In Review

Boonyawat K, et al. "The effect of low-dose oral vitamin K supplementation on INR stability in patients receiving warfarin. A randomised trial". Thromb Haemost. 2016. 116(3):epub 1-6.
PubMed • Full text • ClinicalTrials.gov

Clinical Question

In patients receiving warfarin with a goal INR of 2-3, does the addition of low-dose oral vitamin K supplementation increase INR stability compared to placebo?

Bottom Line

Low-dose vitamin K (LDVK) supplementation does not increase the mean time in therapeutic range (TTR) compared to placebo, but was associated with a decreased number of extreme INR values (<1.5 or >4.5) without increasing the risk of adverse events.

Major Points

Previously, in three different studies, it has been shown that LDVK improves INR stability in patients taking a vitamin K antagonist by comparing the standard deviation of INRs among patients. These were small randomly controlled trials performed in Europe, and either did not assess time in therapeutic range (TTR) or reported no significant difference compared to placebo. A 2014 Cochrane review found insufficient data to support benefit of LDVK for INR stability.

Published in 2016, this study by Boonyawat and colleagues set out to demonstrate an improvement in TTR in a large population. However, this study found that LDVK does not significantly increase INR stability using mean TTR. Though, it was shown that LDVK did reduce the number of extreme INR values. The study was performed over 6 months, and no significant changes in adverse events were found.

Design

• Multicenter, double-blind, randomized, placebo controlled trial
• N=253
  • LDVK (n=126)
  • Placebo (n=127)
• Setting: 4 hospitals in Ontario, Canada
• Enrollment: August 2010 to August 2013
• Mean follow-up: 6 months after a 1 month run-in period to allow for dose adjustment required when starting vitamin K
• Analysis: As treated
• Primary endpoint: Comparison of mean TTRs over 6 months
• Secondary endpoints:
  • All extreme INR readings (>4.5 or <1.5)
  • Number of times a patient's INR was measured over 6 months
  • Bleeding and thrombotic events that require medical intervention

Population

Inclusion Criteria

• Receiving warfarin therapy for at least the past 3 months
• Goal INR of 2-3

Exclusion Criteria

• Planned duration of warfarin therapy less than 6 months
• Anticipated length of survival less than 6 months
• INR on screening day outside range (1.8-3.4)
• Severe liver disease
• Excess alcohol consumption
• Malabsorption syndrome
• In ability to take oral medications
• Administration of medications known to interfere with warfarin and whose dose was likely to change over the course of the study
• Allergy to vitamin K
• Inability or unwillingness to follow study procedures
• Prior participating in this study
• Participation in a competing trial that may impact INR control

Baseline Characteristics

Presented in the order of LDVK vs Placebo

• Mean age (years): 66.9 vs 65.7
• Male sex (%): 56.4 vs 50
• Indication for warfarin (no.):
  • Atrial fibrillation: 44 vs 45
  • Treatment of arterial thromboembolism: 7 vs 7
  • Treatment of VTE: 66 vs 61
  • Valve replacement: 14 vs 17
• History of warfarin associated bleeding (no.): 3 vs 8
• Chronic renal disease (no.): 3 vs 4
• Active malignancy (no.): 2 vs 3
• Prior mean TTRs (%): 51.8 vs 54.6
• Mean number of INRs prior to the study/days of follow up: 9.9/165 vs 9.8/169
• Mean number of INRs during follow up/days of follow up: 9.9/168 vs 9.3/170

Interventions

• Randomization to receive either a daily dose of 150 mcg of vitamin K or a placebo for a total of 7 months (1 month run-in period followed by 6 months of study follow-up)
• INRs were checked at a minimum of 1 week after initiation of treatment and then at the discretion of the anti-coagulation clinics.

Outcomes

Comparisons are low-dose vitamin K vs. placebo.

Primary Outcomes

Mean TTRs over 6 months

65.1% vs. 65.9% (AD -0.8%, 95% CI -7.0% to 5.4%, P=0.8)

Secondary Outcomes

All extreme INR readings (<1.5 or >4.5)/total INR readings

59/1086 (5.4%) vs. 80/1029 (7.8%) (AD 2.4%, 95% CI 0.2% to 4.0%, p=0.03)

Mean number of times a patient's INR was measured over 6 months

9.9 vs. 9.3 (Not calculated due to lack of significance)

Bleeding and thrombotic events that require medical intervention

Due to the low number of adverse events, a multivariable logistic regression was not performed.

Subgroup Analysis

Mean TTR improvement from baseline in patients who had mean TTR values <50% prior to enrollment

28.6% vs. 28.7% (No significant difference)

Adverse Events

Due to the lack of adverse events, a multivariable logistic regression was not performed. However the number of adverse events reported are presented below as LDVK vs placebo.

• Thromboembolic events: 0 vs 0
• Non-major bleeding events: 3 vs 1
• Deaths: 3 vs 1
  • Congestive Heart Failure Deaths: 1 vs 1
  • Deaths of unknown cause: 2 vs 0

Criticisms

• A patient's dietary intake of vitamin K is an important variable that this study did not attempt to eliminate or even monitor. This could significantly effect the results of the study.
• Due to the established INR lowering effects of vitamin K, there is a possibility that the physicians and/or patients in the LDVK group may have become unblinded upon seeing the first INR after treatment was initiated.
• Due to lack of study drug supply, the trial was discontinued prematurely which could underestimate the effectiveness of interventions.

Funding

Funding provided by the Heart and Stroke Foundation of Canada.

No conflicts of interest were disclosed.