Background: Acenocoumarol dose is normally determined via step-by-step adjustment process based on International Normalized Ratio (INR) measurements. During this time, the risk of adverse reactions is especially high. Several genotype-based acenocoumarol dosing algorithms have been created to predict ideal doses at the start of anticoagulant therapy.
Methods: Nine dosing algorithms were selected through a literature search. These were evaluated using a cohort of 63 patients with atrial fibrillation receiving acenocoumarol therapy.
Results: None of the existing algorithms could predict the ideal acenocoumarol dose in 50% of Russian patients. The Wolkanin-Bartnik algorithtm based on European population was the best-performing one with the highest correlation values (r=0.397), mean absolute error (MAE) 0.82 (±0.61). EU-PACT also managed to give an estimate within the ideal range in 43% of the cases. The two least accurate results were yielded by the Indian population-based algorithms. Among patients receiving amiodarone, algorithms by Schie and Tong proved to be the most effective with the MAE of 0.48±0.42 mg/day and 0.56±0.31 mg/day, respectively.
Conclusions: Patient ethnicity and amiodarone intake are factors that must be considered when building future algorithms. Further research is required to find the perfect dosing formula of acenocoumarol maintenance doses in Russian patients.
Keywords: CYP2C9; CYP4F2; GGCX; VKORC1; acenocoumarol; dosing algorithms.