Skip to main content Back to Top


Warfarin Withdrawal May Not Be Enough If INR > 6

Kate Traynor

A patient's age and initially high International Normalized Ratio (INR) may influence the likelihood that cessation of warfarin therapy will quickly lower that value to an acceptable level, say researchers.

Two days after warfarin therapy was stopped in response to an INR greater than 6.0, 37 percent of 633 patients at an outpatient anticoagulation clinic had an INR of at least 4.0—a well-recognized threshold for increased risk of major bleeding during warfarin therapy.

Researchers found that, for each INR unit above 6.0, patients had a 25 percent increased likelihood of having an INR of 4.0 or more after not taking warfarin for two days. Also, for each additional decade of age beyond 59 years, patients had an 18 percent increased likelihood of having an INR of 4.0 or more on day 2.

The findings of this retrospective study, which included only patients who had had an INR of 6.0 or more, were reported in the Sept. 18 Annals of Internal Medicine.

According to the report, the patients with decompensated congestive heart failure (CHF) were about 1.8 times as likely as those without the disorder to have an INR of 4.0 or higher on day 2. Cancer patients were about 1.5 times as likely as the other patients to have a high INR two days after stopping warfarin therapy.

The authors attempted to identify the patients whose INR drop by day 2 was slow enough to have merited phytonadione therapy to reverse the effects of "excessive anticoagulation." A complication of phytonadione therapy, said the researchers, is that a patient's INR can be overcorrected to a dangerously low level—less than 2.0.

For patients 80 years of age or older, especially those who used no more than 15 mg of warfarin each week, the researchers found that the INR drop was slow enough to justify phytonadione treatment. Among patients of all ages who had decompensated CHF, the authors said that the benefits of phytonadione treatment to quickly lower an INR likely outweigh the risk of overcorrecting an INR.

The INR of most of the younger patients, especially those who took at least 50 mg of warfarin each week, had dropped enough by day 2 to make phytonadione use too risky, the researchers said.

The study participants were among 7,279 patients who had been treated at a Massachusetts anticoagulation clinic between August 1993 and September 1998. Only patients who had been treated for 60 days or more were eligible for the study. None of the patients received phytonadione during the initial two days on which warfarin was not taken.

An editorial (PDF) by Henry I. Bussey, Pharm.D., noted that the study provided valuable insights into which patients might best benefit from phytonadione treatment to lower a high INR. But, in Bussey's opinion, the researchers did not adequately examine the influence of warfarin-potentiating drugs on the patients' INR values. One in five patients who participated in the study had reported "recent" use of a medication known to amplify the anticoagulant effects of warfarin.

Bussey also noted that the researchers had not confirmed the participants’ INR on day 2 but relied, instead, on a single determination of this value. In his experience, Bussey said, 10–20 percent of laboratory INR measurements are inaccurate.