Oral Vitamin K Versus Placebo to Correct Excessive Anticoagulation in Patients Receiving Warfarin - Oral Vitamin K Versus Placebo to Correct Excessive Anticoagulation in Patients Receiving Warfarin


The goal of the trial was to evaluate treatment with oral vitamin K compared with placebo in nonbleeding patients with an elevated international normalized ratio (INR).


Oral vitamin K would be more effective in decreasing bleeding events.

Study Design

  • Placebo Controlled
  • Randomized
  • Blinded
  • Parallel

Patients Enrolled: 724
Mean Follow Up: 90 days
Mean Patient Age: 70 years
Female: 47%

Patient Populations:

  • Nonbleeding ambulatory patients with an INR 4.5-10


  • Warfarin was planned to be discontinued
  • Age less than 18 years
  • Life expectancy less than 10 days
  • Indication for urgent normalization of INR, such as surgery
  • Severe liver disease
  • Major bleeding event in the prior month
  • Known bleeding disorder
  • Thrombolytic therapy within the prior 48 hours
  • Thrombocytopenia
  • Inability to take oral medications
  • Allergy to vitamin K
  • Inability to return for follow-up

Primary Endpoints:

  • Any bleeding event

Secondary Endpoints:

  • Thromboembolism
  • Major bleeding event
  • Death

Drug/Procedures Used:

Nonbleeding patients with INR values of 4.5-10 were randomized to 1.25 mg of vitamin K (n = 355) versus placebo (n = 369).

Principal Findings:

Overall, 724 patients were randomized. There was no difference in baseline characteristics between the groups. The mean age was 70 years, 47% were women, mean INR was 6.0, treatment of arterial or venous thromboembolism was the indication for warfarin in 39%, and 18% had a previous bleeding event. INR decreased from 5.95 at enrollment to 3.17 the following day with vitamin K, and from 5.75 to 4.40 with placebo.

The incidence of the primary outcome, any bleeding event at 90 days, was 15.8% in the vitamin K group versus 16.3% in the placebo group (p = 0.86). Major bleeding was 2.5% versus 1.1% (p = 0.22), thromboembolism was 1.1% versus 0.8% (p = 0.72), and death was 2.0% versus 1.9% (p = 0.94), respectively for vitamin K versus placebo.


Among nonbleeding patients with elevated INR, the use of oral vitamin K was not effective in reducing bleeding events compared with placebo. There was also no signal of increased thromboembolism or mortality with the use of vitamin K; however, this agent was effective in reducing INR the day after administration.

Due to a low incidence of major bleeding events, this trial was markedly underpowered to detect a difference between treatment arms. Among nonbleeding patients with elevated INR, it appears that either administration of oral vitamin K or simply holding warfarin is reasonable.


Crowther MA, Ageno W, Garcia D, et al. Oral vitamin K versus placebo to correct excessive anticoagulation in patients receiving warfarin: a randomized trial . Ann Intern Med 2009;150:293-300.

Clinical Topics: Anticoagulation Management, Pulmonary Hypertension and Venous Thromboembolism, Vascular Medicine, Anticoagulation Management and Venothromboembolism

Keywords: Vitamin K, International Normalized Ratio, Warfarin, Venous Thromboembolism

< Back to Listings