Assessment of adherence to Warfarin Anticoagulation using the ratio of vitamin K dependent factors in a teaching hospital in South Africa

Abstract

Background: Warfarin has both anticoagulant and antithrombotic effects, as reflected by the short and long half-lives of the Vitamin K dependent clotting factors. A therapeutic International Normalized Ratio (INR) may reflect the anticoagulant effect, but not the antithrombotic effect with consequent suboptimal assessment of thrombotic risk. A potential approach to monitor adherence to warfarin is to measure the INR in relation to short half-life and long half-life clotting factors. Objectives: To evaluate adherence to warfarin in participants with a therapeutic INR by measuring clotting factor II and VII levels. Patients and Methods: This was a prospective cross-sectional study approved by the University ethics committee and participants gave written informed consent. The study included participants ≥18 years of age on steady state (≥ 3 months) warfarin anticoagulation. Participants were stratified into a therapeutic INR (study group) and a sub-therapeutic INR group (control group). A 5ml venous blood sample was collected in a trisodium citrate (0.109M, 3.2%) tube and factor II and factor VII plasma levels were measured on a STA-R coagulation analyser in both the study and control groups. Data on patient demographics, concurrent medication and monthly INR results was collected from the Anticoagulation Clinic records. Results: Of the 350 participants enrolled in the study, 45 (12.8%) had INRs values above target the therapeutic range, whilst 213 (60.8%) had therapeutic INR values, and 92 (26.2%) had a sub-therapeutic INR. Overall non-adherence levels as judged by clotting factor levels were 9.4%. Of the participants with a sub-therapeutic INR, measured factor II and VII levels indicated adequate anticoagulation in 51(64.5%) participants and inadequate anticoagulation in 28 (35.4%) participants. In participants with a therapeutic INR, clotting factor levels indicated inadequate anticoagulation in 2 participants (0.9%). Time in therapeutic range (TTR) was assessed over a 4 month period and overall 71.6 % showed appropriate TTR. Subgroup analysis indicated that Caucasians spent more time in TTR whilst Asians spent the least TTR (p<0.0001; Cramer’s V=0.24). In relation to the INR, the clotting factor measurements showed a 79.8% sensitivity and 93.3% specificity in predicting adherence and nonadherence to warfarin anticoagulation respectively.

Conclusions: In this study, clotting factor II and VII levels measured in steady warfarin therapy warfarin did not depict significant levels of non-adherence and are therefore not useful for routine monitoring of warfarin adherence. However, specific indications for clotting factor testing would include patients with a sub-therapeutic INR at increased risk of bleeding, those suspected of being non-adherent on warfarin therapy, those with a hypercoagulable state, and those on drugs that interfere with warfarin metabolism . This study also showed that in a number of patients with a sub-therapeutic INR, anticoagulation may well be adequate based on clotting factor measurements. In the absence of clotting factor levels, the participant’s warfarin dose may be inappropriately increased with associated increased bleeding risk.