BOSTON—Serious bleeding may occur in patients who were given excessive warfarin anticoagulation; prolonged delay in the return of the international normalized ratio (INR) to a safe range likely increases this risk. It would therefore be extremely helpful to be able to predict a delay in INR normalization after warfarin is stopped.

That may now be possible, suggests a retrospective study by Elaine M. Hylek, MD, and coworkers.[1] “We found that warfarin sensitivity, advanced age, and extreme INR elevation independently predict a prolonged delay in the INR’s return to a safe range,” Dr. Hylek, an Associate Physician at Massachusetts General Hospital, told PULMONARY REVIEWS.

Patients 80 years and older are most at risk, she said, especially if they are sensitive to warfarin (that is, they require 15 mg or less of the drug per week). Furthermore, risk rises linearly with each 1.0-unit increase in INR, she and her colleagues found. Decompensated congestive heart failure and active cancer greatly increase the risk in all age groups.

“Clearly, the findings of a retrospective study must be viewed with caution until confirmed by prospective controlled trials,” commented Henry I. Bussey, PharmD, in an editorial accompanying the Hylek study.[2] In the absence of more definitive data, however, the study may help clinicians estimate the rate of INR decline after excessive anticoagulation, decide about vitamin K use to reverse the INR, and plan follow-up INR measurements, stated Dr. Bussey, a Professor in the College of Pharmacy at the University of Texas, San Antonio.

The study included 633 outpatients ages 25 to 95 years who had been receiving warfarin and who had an INR of 6.0 or higher. A prosthetic heart valve, atrial fibrillation, and cerebrovascular disease were among the most common indications for anticoagulation in these patients.

To be included in the study, patients had to have been receiving warfarin for more than 60 days to ensure that they had been given a stable maintenance dose of the anticoagulant. Once the INR elevation was identified, the warfarin was withheld for two consecutive days; a follow-up INR was measured on the second day. No patient was given vitamin K.

For statistical analysis, patients were stratified according to whether their INRs were above or below 4.0 on the second day without warfarin; 4.0 is the threshold for an increased risk of major hemorrhage during warfarin therapy, said Dr. Hylek, who is also an Instructor of Medicine at Harvard Medical School in Boston.

Two days after warfarin cessation, 232 patients (37%) still had an INR of 4.0 or greater. Compared to those with lower INRs, these patients were older, required lower maintenance doses of warfarin, and had higher initial INRs.

The likelihood that the INR would be 4.0 or higher on day 2 rose by 18% for each decade of life and by 25% for each unit increase in the index INR. Conversely, the likelihood dropped by 13% for each 10-mg increase in the weekly warfarin dose, reflecting more rapid INR normalization among patients on higher maintenance doses. The odds ratios that the INR would be 4.0 or higher on day 2 were 2.79 in patients with decompensated congestive heart failure and 2.48 in those with active cancer.

IMPLICATIONS FOR PRACTICE

Early intervention with vitamin K may be most beneficial for excessively anticoagulated patients 80 years or older (who are at the highest risk of major hemorrhage), particularly those requiring lower maintenance doses of warfarin, the study authors concluded. However, immediate administration of vitamin K may not be advisable for younger patients (those younger than 60 years) who have been receiving high warfarin maintenance doses.

“Many of these patients have a very rapid rate of spontaneous INR decay, so intervening with vitamin K may put their INR into a subtherapeutic range,” Dr. Hylek cautioned. Twelve percent of patients in this study spontaneously experienced subtherapeutic INR values (ie, below 2.0) after two doses of warfarin were withheld.

The authors acknowledge that an important limitation of the study was their inability to quantify the effects of dietary vitamin K on the rate of INR decline. However, vitamin K intake probably did not substantially bias the results, they asserted, because patients were advised not to alter their diet in response to INR changes.

—Timothy Begany

References
1. Hylek EM, Regan S, Go AS, et al. Clinical predictors of prolonged delay in return of the international normalized ratio to within the therapeutic range after excessive anticoagulation with warfarin. Ann Intern Med. 2001;135:393-400.
2. Bussey HI. Managing excessive warfarin anticoagulation. Ann Intern Med. 2001;135:460-463.

Return to table of contents