Patients who take long-term daily oral anticoagulation medications, such as warfarin, are at risk for excessive bleeding during minor surgical or dental procedures. However, the risk of a recurrent thromboembolic event is also possible if the patient stops taking anticoagulation medication, even for a short time. Patients at lower risk can be taken off anticoagulation medication for the procedure and resume therapy a few days later, while patients at higher risk must be hospitalized, taken off anticoagulation therapy and treated with intravenous heparin during surgery. Both starting and stopping anticoagulant therapy are considered complicated and dangerous. Wentzien and colleagues examined the safety and efficacy of partially reducing the anticoagulant effect of warfarin by giving patients a single small dose of oral vitamin K₁.

Patients who were scheduled to undergo a minor surgical or dental procedure or those found to have an International Normalized Ratio (INR) of 5.0 and no current bleeding were eligible for the study. Exclusion criteria included clinical bleeding at the time of the screening or a history of a thromboembolic event after warfarin therapy was discontinued. The dosage of vitamin K was calculated by using the following formula: 16 − (17 × desired INR/initial INR), rounded to the nearest 1.25 mg.

Two groups of patients were studied: the first group was scheduled to undergo a dental or minor surgical procedure; the second group had an excessively high INR and no bleeding. Patients in the surgery group were given oral vitamin K₁ 36 hours before their scheduled procedure and were instructed to continue their same dosage of warfarin. Their target INR was 1.75; the INR was measured four hours before the procedure and again 32 hours after ingestion of the vitamin K₁. After the procedure, their INR was tested every one to two days until it exceeded 1.9. For patients in the high-INR group, the goal of vitamin therapy was to return their INR levels to a therapeutic range of 2 to 3 (or 3 to 4 in patients with prosthetic heart valves). These patients did not take their warfarin dose on the first day of the study and were given a dose of oral vitamin K₁, which was calculated by using the same formula used for the first group.

Fifty-one patients were enrolled in the study, with 30 in the surgery group and 21 in the high-INR group. Mean patient age was 55 years, and the distribution between men and women was almost equal. The average dose of vitamin K₁ was 5 mg for the surgery group and 10 mg for the high-INR group. At baseline, the mean INR for patients in the surgery group was 2.6; the mean INR was 8.4 in the high-INR group. After ingestion of vitamin K₁, patients' INRs corrected to a mean of 1.6 and 1.7, respectively. Fifty of 51 patients attained a desired INR after taking oral vitamin K₁. Prothrombin times for patients in the surgery group returned to a therapeutic level within an average of six days, and no patient experienced a thromboembolic event.

The authors conclude that giving patients a single dose of oral vitamin K₁ is a safe and effective method of reversing the anticoagulant effect of warfarin without decreasing the warfarin dosage or discontinuing the oral anticoagulant.