Pharmacoepidemiology and Drug Utilization
Clinical Pharmacology & Therapeutics (2001) 70, 159–164; doi: 10.1067/mcp.2001.117444
Interindividual variability in sensitivity to warfarin-Nature or nurture?
Ronen Loebstein MD1,*, Hagith Yonath MD1,*, Daria Peleg MSc1, Shlomo Almog PhD1, Michal Rotenberg PhD1, Aharon Lubetsky MD1, Joseph Roitelman PhD1, Dror Harats MD1, Hillel Halkin MD1 and David Ezra MD1
1Division of Clinical Pharmacology and Toxicology and the Anticoagulation Clinic, Department of Medicine, The Chaim Sheba Medical Center, Tel Hashomer Israel.
Correspondence: Hillel Halkin, MD, Division of Clinical Pharmacology and Toxicology, The Chaim Sheba Medical Center, Tel Hashomer 52621, Israel E-mail: halkin@netvision.net.il
*Ronen Loebstein and Hagith Yonath contributed equally to this work.
Received 1 March 2001; Accepted 23 May 2001.
Abstract
Background: Interindividual variability in responses to warfarin is attributed to dietary vitamin K, drug interactions, age, or genetic polymorphism in the cytochrome P4502C9 enzyme (CYP2C9) (allelic variants 2C9*2 and 2C9*3) linked with impaired metabolism of the potent enantiomere S -warfarin.
Patients and Methods: We quantified the relative effects of age and of simultaneously determined CYP2C9 genotype, plasma warfarin and vitamin K concentrations, and concurrent medications on warfarin maintenance doses in 156 patients at optimized stable anticoagulation.
Results: Allele frequencies for CYP2C9*1, CYP2C9*2, and CYP2C9*3 were 0.84, 0.10, and 0.06. Warfarin doses were 6.5 
3.2, 5.2 2.4, and 3.3 2.0 mg/d in the 3 genotype groups (P < .0001). Warfarin doses decreased with age as follows: 7.7 3.7 versus 4.9 2.9 mg/d at <50 years and >66 years (P < .001), mainly as a result of decreased plasma warfarin clearance (2.8 1.4 mL/min versus 1.9 0.8 mL/min; P < .001). Vitamin K (1.6 1.1 ng/mL) did not differ among the age or genotype groups. Patients 
66 years old with the CYP2C9*3 allele required only 2.2 1.2 mg/d compared with 7.9 3.7 mg/d in those 
65 years old bearing the CYP2C9*1 allele (P < .001). On multiple regression, warfarin maintenance doses were independently associated with plasma warfarin (reflecting its metabolic clearance) (r 2 = 0.26), age (possibly reflecting increased intrinsic sensitivity) (r 2 = 0.12), and genotype (reflecting S -warfarin levels) (r 2 = 0.10) but not with plasma vitamin K.
Conclusions: At optimized steady state, individual sensitivity to warfarin is determined by CYP2C9 genotype and age with no effect of vitamin K. Prospective studies will determine the impact of these findings in clinical practice.
