Comparison of 5-mg and 10-mg Loading Doses in Initiation of Warfarin Therapy

Methods

Patients with no contraindications to warfarin who required anticoagulation (target INR, 2.0 to 3.0) were invited to participate in the trial between June and November 1994 at The Hamilton Civic Hospitals (Ontario, Canada). Enrollment was targeted at 50 patients. All patients gave written, informed consent, and the study was approved by the local internal review board.

Patients were assigned by random number table to receive warfarin at an initial dose of 5 or 10 mg. Subsequent doses were determined on the basis of nomograms that were developed before the study began through an iterative process. Because the study used laboratory outcomes, patients and health care professionals were not blinded to treatment assignments.

Patients were followed for a maximum of 108 hours, during which time they received five doses of warfarin. Blood samples were taken before warfarin therapy was initiated, 12 hours after the first dose of warfarin, and at 24-hour intervals for 5 days. Blood was collected into vacuum-sealed containers that contained 0.102 mol of buffered sodium citrate per liter. To obtain platelet-free plasma, the sample was double centrifuged at 1700 g and frozen at −70°C for batch assays [13].

The prothrombin time was measured using Thromborel S (International Sensitivity Index 1.06, Behring Diagnostics, Kanata, Ontario, Canada) and reported as an INR [14]. Heparin does not increase the prothrombin time in patients receiving both warfarin and heparin if this thromboplastin reagent is used. Factors II, VII, IX, and X were assayed by using the one-stage clotting method [15]. Protein C levels were measured by using a functional clotting assay (Dade International, Mississauga, Ontario, Canada) [16]. Treatment of supratherapeutic INRs with vitamin K was left to the discretion of the attending physician.

The outcome measures were the time required to achieve an INR of 2.0 to 3.0, the proportion of INR values greater than 3.0, the time course for reductions in levels of factors II and X, and the time course for reduction in protein C levels. Additional data included levels of factors VII and IX.

Results

Fifty-one consecutive patients were enrolled in the trial. Two were excluded after one dose of warfarin: One died, and one required cardiac catheterization. Twenty-five patients were randomly assigned to receive a 10-mg loading dose of warfarin; 24 were assigned to receive a 5-mg loading dose. The two groups did not differ in age; weight; or frequency of acute thromboembolism, cancer, or surgery during or immediately before the study period. Forty-eight patients received concomitant heparin therapy for all or part of the study period.

Beginning on study day 2, the dose of warfarin was adjusted using a nomogram. Warfarin doses in the two groups were similar except on days 1 and 2. The 10-mg group achieved an INR greater than 2.0 statistically significantly sooner than did the 5-mg group (Table 1). At 36 hours, 11 of 25 patients (44% [CI, 34% to 54%]) in the 10-mg group had INRs of 2.0 or greater compared with 2 of 24 patients (8% [CI, 3% to 14%]) in the 5-mg group (P = 0.005). At 60 hours, 9 of 25 patients in the 10-mg group (36% [CI, 17% to 54%]) had INRs greater than 3.0 compared with none of 23 patients in the 5-mg group (P = 0.002). Five patients (4 in the 10-mg group and 1 in the 5-mg group) received vitamin K. These patients had INRs of 4.8 to 9.3 and received 0.5 to 2.0 mg of vitamin K subcutaneously. No patient bled.

Levels of factors II and X declined slowly, and no substantial differences were seen between the 5-and 10-mg groups. In contrast, levels of factor VII and protein C decreased more rapidly and were statistically significantly lower in the 10-mg group than in the 5-mg group at 36 and 60 hours (Figure 1).

View larger version:

• In this window
• In a new window

• Download as PowerPoint Slide

Figure 1. Levels of factor II ( ) decreased at a similar rate in the 5- and 10-mg groups. Levels of factor VII ( ) and protein C ( ) decreased more rapidly in the 10-mg group than in the 5-mg group. Error bars represent 95% Cls. Plasma levels of coagulation factors in patients receiving a 10-mg (striped bars) or 5-mg (white bars) loading dose of warfarin for each of six time points assessed.topmiddlebottom

Discussion

We compared the relative effects of 5- and 10-mg loading doses of warfarin. The time course of reduction in levels of factor II was used as a surrogate end point for clinical efficacy, and excessive elevation of the prothrombin time and unopposed reduction in protein C levels were used as surrogate end points for safety.

Patients who received a 10-mg loading dose of warfarin achieved INRs greater than 2.0 more rapidly than did patients who received a 5-mg loading dose. However, because this change in the INR was entirely caused by the early reduction of factor VII levels in the 10-mg group, it may not reflect an antithrombotic effect of warfarin, which is thought to result from a reduction in factor II levels. Two separate experimental observations in rabbits support this idea: The first is the early report of a 5-day delay in achieving an antithrombotic effect with warfarin, although the prothrombin time was first prolonged into the therapeutic range and the level of factor VII was markedly reduced after 2 days of treatment [17]. This observation provides the rationale for overlapping heparin and warfarin therapy for 5 days during the initiation of anticoagulant therapy [17, 18]. The second is the finding that the antithrombotic effect of warfarin is abrogated by the infusion of factor II and (to a lesser extent) factor X [18]. In our study, the rate of reduction in levels of factor II and factor X activity were similar in the 5- and 10-mg groups.

On the other hand, the 10-mg loading dose was associated with a significantly more rapid decrease in protein C activity (which decreased before levels of factors X and II were substantially reduced) than that seen in the 5-mg group and an excessive prolongation of the INR. The combination of markedly reduced protein C levels and near-normal levels of factors II and X over the first 2 days of warfarin therapy could produce a hypercoagulable state, and excessive prolongation of the INR could create a higher risk for bleeding.

Our study is limited because we used surrogate markers for efficacy and safety. A much larger trial using clinical outcome measures is needed to determine whether the surrogate markers that we used are clinically relevant. Nevertheless, our findings suggest that both regimens result in a therapeutic INR in most patients by day 5 of treatment.

From Hamilton Civic Hospital, McMaster University, and The Hamilton Civic Hospitals Research Centre, Hamilton, Ontario, Canada.