Dietary vitamin K variability affects International Normalized Ratio (INR) coagulation indices

Background: Changes in daily vitamin K intake may contribute to marked variations in the International Normalized Ratio (INR) coagulation index in patients receiving oral warfarin anticoagulant therapy, with potentially serious adverse outcomes. Thus, patients receiving warfarin therapy are routinely counseled regarding this drug-nutrient interaction and are instructed to maintain consistent vitamin K intakes, though little quantitative information about this relationship is available. Objective: To determine the quantitative impact of variability in dietary vitamin K-1 (phylloquinone) intake, assessed by a validated patient self-monitoring instrument, on weekly INR in patients receiving warfarin anticoagulant therapy. Methods: A prospective dietary assessment study was conducted at the Massachusetts. General Hospital in Boston. Sixty outpatients (37 males and 23 females) were selected with a mean age 60.3 +/- 16.8 years, who beg gan oral warfarin anticoagulant therapy within 14 days prior to their first clinic-visit to an outpatient anticoagulation therapy unit. Exclusion criteria included more than 2 drinks of alcohol per day, inability to speak English, and concurrent disease states affecting warfarin therapy such as liver disease and terminal illness. Over the five-week study period, participants recorded daily intakes in specified amounts of all-food items appearing on a validated dietary self-assessment tool. Concomitant use of prescription and/or non-prescription medications was also obtained. Concurrent daily warfarin dose and adherence to the drug regimen, concomitant use of prescription and/or non-prescription medications known to interact with warfarin, and weekly INR were obtained. Week-to-week changes in vitamin K intake, warfarin dose, and INR were determined and cross-correlated. Results: Forty-three patients (28 males and 15 females) completed the study and 17 dropped out. Pearson's correlation coefficient revealed the variability in INR and changes in vitamin K intake were inversely correlated (r = -0.600, p < 0.01). Multiple regression analysis (r = 0.848) indicated that a weekly change of 714 mu g dietary vitamin K significantly altered weekly INR by 1 unit (p < 0.01) and a weekly change of 14.5 mg warfarin significantly altered weekly INR by 1 unit (p < 0.01) after adjustment for age, sex, weight, height, and concomitant use of medications known to interact with warfarin. Conclusions: Patients taking warfarin and consuming markedly changing amounts of vitamin K, may have, a variable weekly INR with potentially unstable anticoagulant outcomes.