Aim: To develop and validate an algorithm for the prediction of therapeutic dose of acenocoumarol in Romanian patients.
Methods: The inclusion criteria for entry to the study was age ≥ 18 years and starting acenocoumarol treatment for at least one of the following clinical indications: acute deep vein thrombosis of the lower limbs, persistent or permanent atrial fibrillation, and/or the presence of valvular prostheses requiring prolonged oral anticoagulant therapy. The patients were followed up for 3 months. Patients admitted to the internal medicine, cardiology, and geriatrics wards of the Municipal Clinical Hospital, Cluj-Napoca and "Niculae Stăncioiu" Heart Institute between October 2009 and June 2011 who fulfilled the inclusion criteria were included in the study. Clinical and demographic data that could influence the acenocoumarol stable dose were recorded for each patient. Genetic analysis included the genotyping the CYP2C9*2 and *3, and the VKORC1 -1693 G > A polymorphisms. The patients were randomly divided into two groups: (1) the main group on which the development of the clinical and genetic algorithms for acenocoumarol dose prediction was based; (2) the validation group.
Results: The study included 301 patients, of whom 155 were women (51.5 %) and 146 were men (48.5 %). The median age of the patient cohort was 66 (women, 57; men, 73) years. After randomization the main group comprised 200 patients (66.4 %) and the validation group 101 patients (33.6 %). Age and body mass index explained 18.8 % (R (2)) of the variability in acenocoumarol weekly dose in patients in the main group. When the genetic data were added to the algorithm, the CYP2C9*2 and *3 polymorphisms and the VKORC1 -1693 G > A polymorphism accounted for 4.7 and 19. 6 % of acenocoumarol dose variability, respectively. For the main group, we calculated a mean absolute error of 5 mg/week (0.71 mg/day). In the validation group, clinical parameters explained 22.2 % of the weekly acenocoumarol dose variability. Genetic polymorphisms increased the R(2) coefficient to 32.8 %.
Conclusion: We have developed and validated an accurate algorithm for prediction of the stable therapeutic dose of acenocoumarol in a Romania population.