Factors affecting the safe use of oral anticoagulants

Abstract

The aim of this PhD was to elucidate the factors that influence anticoagulation response to the oral anticoagulants, warfarin and non-vitamin K antagonist oral anticoagulants (NOACs). Identification of the factors which influence clinical response to these agents could lead to improvements in patient management and treatment safety. In a cross-sectional study, the mean TTR over a 12 months period was significantly higher in hospital monitored patients (78%) (P=0.001) compared to those monitored in general practice (71%) or at their homes (68%). Domiciliary monitored patients had the least TTR among the three groups although they had the highest numbers of dose changes and INR monitoring events. In a further longitudinal study of patients on warfarin for up to 14 years, the TTR according to age was significantly lower in home monitored patients, with this group having a higher probability of having poor anticoagulation control (TTR ≤65%). In an ex-vivo study, rivaroxaban in vitamin K deficient older subjects produced a greater prolongation of both PT and modified PT, and a greater suppression in the rate and amount of thrombin generation compared to vitamin K replete younger subjects. Therefore, poor vitamin K intake could play a role in the reported incidences of bleeding associated with NOACs We had previously demonstrated that daily vitamin K supplementation causes an increase in warfarin daily dose requirement which varies between different patients and is related to VKORC1 genotype. Based upon these observations we set out to test our hypothesis that patients on warfarin therapy with the VKORC1(-1639)GG polymorphism could have poorer anticoagulation control than those with GA or AA genotypes as a result of variable dietary intake of vitamin K. However, the study results failed to confirm this hypothesis. Patients on warfarin therapy scheduled for an invasive surgery have to stop taking the drug for a fixed number of days to avoid peri-operative bleeding. However, there is variance in the rate by which INR falls after stopping warfarin. I found that patients with CYP2C9 variant alleles (CYP2C9*2*2 or CYP2C9*2*3) were >8 times (95% CI = 2.25–33.25) more likely to have INR of ≥ 1.5 before the planned day of surgery than those with wild-type genotype. In a further study, patient age and CYP2C9 *2*2 & *2*3 ii genotype were found to significantly influence the time required to reach an INR of 1.5 following warfarin cessation. VKORC1 genotype had no effect on the rate of INR decline.