Despite its early acceptance as an approved treatment for excessive anticoagulation, oral vitamin K₁ seems to have fallen into disuse. Some recent guidelines on anticoagulant therapy [2, 11, 12] mention only the parenteral use of vitamin K₁ to correct excessive anticoagulation. These guidelines may reflect, at least in part, a published statement that oral vitamin K₁ produces a variable response [13]. This statement was made on the basis of an earlier pharmacologic study [14] that measured plasma concentrations of vitamin K₁ rather than prothrombin time.

We have frequently given 2.5 mg of oral vitamin K₁ to patients who had an INR greater than 5.0 when they visited the anticoagulant clinics at the medical centers of the University of California, San Diego, and University of California, San Francisco. Data from the records of both clinics support the use of oral vitamin K₁ in the management of excessive anticoagulation.

Methods

Top Methods Results Discussion Author & Article Info References

Data taken from records at the anticoagulation clinic of the University of California, San Diego, Medical Center are presented for 66 patients who had an episode of excessive anticoagulation (INR >5.0) between August 1989 and January 1996. Only the first episode of excessive anticoagulation that was treated with vitamin K was included in the data. Seven patients had the following types of minor bleeding: hematuria (3 patients), gingival bleeding (2 patients), hemoptysis (1 patient), and vaginal bleeding (1 patient). No patient had evidence of major bleeding (gastrointestinal, retroperitoneal, or intracranial bleeding or a decrease in hemoglobin concentration 2 g/dL). Also presented are data for 15 patients who were treated at the anticoagulation clinic of the University of California, San Francisco, Medical Center.

Each patient received 2.5 mg of oral vitamin K₁ and had either 1 or 2 doses of warfarin withheld. Fifty-eight patients who returned for another INR measurement 24 hours later had 1 dose of warfarin withheld. Twenty-three patients who could not return to the clinic the next day had 2 doses of warfarin withheld and had their INR measured after 48 hours. Warfarin therapy was started again when the INR was less than 5.0 and minor bleeding had apparently stopped. The previous dose of warfarin received by each patient was either reduced or continued, depending on an evaluation of the cause of the excessive anticoagulation. A third INR measurement was obtained 4 to 7 days after warfarin therapy had resumed.

International normalized ratios were determined from prothrombin times that were measured 1) in the anticoagulant clinic by using a portable capillary fingerstick monitor (Coumatrak, DuPont Pharma, Wilmington, Delaware) with an International Sensitivity Index [ISI] of 2.06 for thromboplastin or 2) in the clinical laboratory by an automated photo-optical coagulometer (MLA-1000, Medical Laboratory Automation, Inc., Mount Vernon, New York). The thromboplastins used in the laboratory at the University of California, San Diego, were Dade Thromboplastin C, with an ISI of 2.88, until August 1993 and then Innovin (Baxter Diagnostics, Deerfield, Illinois), with an ISI of approximately 1.0. In the laboratory at the University of California, San Francisco, the thromboplastins used were Simplastin Excel (Organon Teknika, Durham, North Carolina), with an ISI of approximately 2.0, or Thromboplastin C (Baxter Diagnostics), with an ISI of 2.04. The appropriate ISI values were used to convert prothrombin time ratios to INRs.

Results

Top Methods Results Discussion Author & Article Info References

Of the 81 study patients, 49 were men and 32 were women. The mean age was 56 years (range, 3 to 83 years). Thirty-four of the patients were white, 24 were Hispanic, 13 were Asian, 7 were black, and 3 were Arabic. Indications for warfarin therapy were venous thromboembolism (34 patients), heart valve prosthesis (19 patients), atrial fibrillation (15 patients), angioplasty or coronary arterial stent (4 patients), arterial thromboembolism or grafts (4 patients), acute myocardial infarction with left ventricular thrombosis or dilated cardiomyopathy (3 patients), and cerebrovascular accident (3 patients).

The effect of giving 2.5 mg of oral vitamin K₁ and withholding warfarin on the INRs of patients is shown in Figure 1 and Figure 2. Data are divided into groups according to initial INRs. The INRs obtained after 24 hours (that is, after one dose of warfarin had been withheld) are shown in Figure 1. In 42 of the 58 patients (72%) who returned to the clinic for an INR measurement after 24 hours, the values fell within the target range of 2.0 to 5.0. In 6 patients (10%), the 24-hour INR exceeded 5.0 (range, 5.6 to 11.6). Four of these patients had an initial INR greater than 10.0. In 10 patients (17%), the 24-hour INR was less than 2.0 (1.9 in 5 patients and between 1.6 and 1.9 in 5 patients).

View larger version (19K): [in this window] [in a new window]   Figure 1. The effect on the international normalized ratio (INR) 24 hours after patients received 2.5 mg of oral vitamin K1 and had 1 dose of warfarin withheld. Data are grouped according to the initial INR for each patient. Open symbols (B columns) are INRs before patients received vitamin K1; closed symbols (A columns) are INRs 24 hours after patients received vitamin K1. The dashed lines delineate the target range for correcting excessive anticoagulation (an INR between 2.0 and 5.0). The solid lines delineate geometric mean INRs.

View larger version (17K): [in this window] [in a new window]   Figure 2. The effect on the international normalized ratio (INR) 48 hours after patients received 2.5 mg of oral vitamin K1 and had 2 doses of warfarin withheld. Data are grouped according to the initial INR for each patient. Open symbols (B columns) are INRs before patients received vitamin K1; closed symbols (A columns) are INRs 48 hours after patients received vitamin K1. The dashed lines delineate the target range for correcting excessive anticoagulation (an INR between 2.0 and 5.0). The solid lines delineate geometric mean INRs.

The INRs that were obtained after 48 hours (that is, after 2 doses of warfarin had been withheld) are shown in Figure 2. In 17 of the 23 patients (74%), the INR after 48 hours was between 2.0 and 5.0. In 1 patient, the INR increased from 6.2 to 9.7, but we determined that this patient had cholecystitis. In another patient, whose initial INR was 11.6, the INR after 48 hours was 5.4. In the remaining 4 patients (17%), the INR after 48 hours was between 1.5 and 1.9. The minor bleeding in 7 patients seemed to have subsided when these patients returned to the clinic for their second visit (3 of the 7 patients returned after 24 hours, and 4 of the 7 returned after 48 hours).

Warfarin therapy was resumed when the INR for each patient was less than 5.0. The dose of warfarin was based on test results and the cause of excessive anticoagulation. The original mean daily dose of warfarin was 5.2 ± 3.0 mg (n = 81), which was reduced to a mean daily dose of 3.8 ± 2.7 mg (n = 76). Results of a third INR measurement, which was obtained from 4 to 7 days later in 76 patients, were as follows: between 2.0 and 5.0 in 63 patients (82.9%), 5.4 in 1 patient (1.3%), between 1.8 and 2.0 in 7 patients (9.2%), and between 1.3 and 1.7 in 5 patients (6.6). No patient required a follow-up dose of warfarin that substantially exceeded the dose of warfarin given before the oral vitamin K₁ was administered, and no patient had clinically evident thromboembolism.

Discussion

Top Methods Results Discussion Author & Article Info References

Because the data compiled by Cosgriff [8] were published more than 40 years ago, they are unavailable to persons who review the literature through a standard computer search. Cosgriff's research established the efficacy of oral vitamin K₁ in correcting excessive anticoagulation. Cosgriff recommended giving 5 to 10 mg of oral vitamin K₁ to patients with excessive anticoagulation. He noted, however, that in 14 of 15 patients who had received 2.5 mg of oral vitamin K₁, the prothrombin times were below the upper therapeutic limit after 24 hours [8].

We had two objectives for giving 2.5 mg of oral vitamin K₁ to patients with excessive anticoagulation: to reduce the INR to less than 5.0 within 24 to 48 hours and to prevent the INR from falling below 2.0. The first objective was achieved in 90% of the patients. If we exclude data from the 10 patients in whom the initial INR was greater than 10.0, the dose of 2.5 mg corrected the INR to a value less than 5.0 in 96% of the remaining 71 patients.

The second objective was achieved in 83% of the patients. Of the 14 patients in whom the INR was reduced to less than 2.0, the INR decreased to less than 1.8 in only 5 patients (6%) and did not decrease to less than 1.5 in any patient. When warfarin therapy was resumed, vitamin K₁ had not induced resistance to anticoagulation.

From epidemiologic data, it can be estimated that the risk for thrombosis is less than 0.1% in patients who have an INR below the therapeutic range for 2 to 3 days [12]. Thus, we view the data presented here as evidence of both the efficacy and the safety of a management plan in which warfarin is withheld for 1 or 2 doses and 2.5 mg of oral vitamin K₁ is given to patients who do not have major bleeding episodes and in whom the INR is greater than 5.0 but less than 10.0. Although our data on patients with an INR greater than 10.0 are limited, they suggest that a minimum dose of 5 mg of oral vitamin K₁ and reevaluation (done by repeating the INR the next day) might be more appropriate than a dose of 2.5 mg.

Because we have no data from a comparison group, our results cannot be used to conclude that giving patients 2.5 mg of oral vitamin K₁ and temporarily withholding warfarin is better than only withholding warfarin. Moreover, some clinicians may prefer the latter approach in selected patients with an INR between 5.0 and 6.0 for whom even a brief period of increased risk for thromboembolism resulting from subtherapeutic anticoagulation is viewed as more serious than the risk for hemorrhage. However, evidence for substantially delayed correction of excessive anticoagulation in patients who do not receive vitamin K can be readily found in the literature [5, 8, 10, 15, 16]. Cosgriff [8] reported that of 232 patients who had excessive anticoagulation and did not receive vitamin K, 33% had returned to a safe level of anticoagulation within 24 hours, another 34% had returned to this level after 48 hours, and an additional 21% had returned to this level after 72 hours. In a study in which warfarin therapy was withheld in patients with test values within the therapeutic range, 48 to 72 hours were required before values decreased to below the lower limit of the therapeutic range [16]. In a recent report [15], 5 of 12 patients with an INR between 5.0 and 8.0 had an INR greater than 5.0 on the day after warfarin therapy was discontinued for 1 day.

In our case series, the decision to give oral vitamin K₁ was made by the clinician who managed the oral anticoagulant therapy; not all clinic patients who had excessive anticoagulation received vitamin K₁. Oral vitamin K₁ would not be appropriate for patients with disorders that affect absorption of vitamin K. When absorption is impaired or significant bleeding exists, parenteral vitamin K is required. Also, as already discussed, if subtherapeutic anticoagulation may produce an extraordinary risk for thrombosis, then withholding warfarin without giving vitamin K₁ may be more appropriate. Additional controlled studies are needed to define optimum strategies for managing excessive anticoagulation.

In the past, oral vitamin K₁ played an important role in the management of patients who received oral anticoagulants. Its use seems to have waned, however, and oral vitamin K₁ is not mentioned in recent guidelines on anticoagulation therapy [2, 11, 12]. The data reported here should temper the reluctance to use oral vitamin K₁ [13, 14]. Administering 2.5 mg of oral vitamin K₁ and withholding warfarin was found to be a safe, painless, and inexpensive treatment for excessive anticoagulation in 96% of 71 patients who did not have a major bleeding episode and had an INR greater than 5.0 and less than 10.0.

Drs. Le and Rapaport: University of California, San Diego, School of Medicine, 9500 Gilman Drive (0612), La Jolla, CA 92093.

Dr. Kayser: University of California, San Francisco, School of Pharmacy, Room C-152, San Francisco, CA 94143.