Effect of Body Weight on Dose of Vitamin K Antagonists

Timothy H. Self, PharmD; Jessica L. Wallace, PharmD, BCPS; Sami Sakaan, PharmD, BCPS; Christopher W. Sands, MD


South Med J. 2015;108(10):637-643. 

In This Article


Studies With a Specific Goal of Assessing Body Weight or Size, Among Other Factors

Warfarin. Studies that initially evaluated the effect of body weight on maintenance warfarin dose requirements were conducted using prothrombin time before the availability of the international normalized ratio (INR). Eccles[2] reported no effect of body weight on maintenance warfarin doses in 80 patients older than 65 years. These patients were assessed in the first 17 days of warfarin therapy. Routledge et al[3] detected a small effect of body weight on response to stable maintenance doses of warfarin in 228 patients (mean age 55 years). In this study, age accounted for 10.4% of the variation of warfarin dose versus a 3.77% variation attributable to weight (P < 0.01).

Kamali et al[4] studied patients with stable warfarin doses and INRs to evaluate the contribution of age, body size, and CYP2C9 genotype to the response to warfarin. Regarding body size, the investigators assessed body weight and body surface area (BSA). Among 121 patients (66 men, 55 women) in this study, the mean age was 72 years. Simple correlation analysis revealed a positive correlation with warfarin dose and body weight (r = 0.25; P = 0.005) and BSA (r = 0.21; P = 0.02); however, in the multiple linear regression model, body weight and BSA did not make a significant contribution to the warfarin dose requirement.[4] Age and CYP2C9 polymorphism had significant effects on the dose requirement of warfarin in this study.

Wallace et al[5] assessed the effect of obesity on the initial response to warfarin in hospitalized patients (N = 211, mean age 60.6, 45% men) of different body mass index (BMI) categories (morbidly obese [n = 37], obese [n = 71], overweight [n = 48], normal weight [n = 45], and underweight [n = 10]) in a retrospective study. For inclusion, patients had to have been newly initiated on warfarin and managed by a pharmacy dosing service with 4 consecutive days of inpatient therapy. Among 1386 patients screened, the most common exclusions were prior warfarin therapy at home (n = 622) and ≤4 days of warfarin therapy in the hospital (n = 318). Patients also were excluded if significant drug–drug or drug–disease interactions were present or if therapy was interrupted during hospitalization for ≥2 days for reasons other than bleeding or supratherapeutic INR. The percentage of patients who achieved a therapeutic INR before discharge differed across BMI categories (P = 0.0004), with 71.1 % of normal-weight patients with a therapeutic INR compared with 42.3% of obese and 38% of morbidly obese patients. Obese and morbidly obese patients compared with normal-weight patients required a higher average daily dose (6.6 ± 0.3, obese P = 0.0004 and 7.6 ± 0.5, morbidly obese P < 0.0001 vs 5 ± 0.3 mg, normal weight), a longer time to achieve therapeutic INR (8 and 10 days vs 6 days, P < 0.0001), and a higher discharge dose (6.7 ± 0.5, obese P = 0.007 and 6.7 ± 0.7, morbidly obese P = 0.0062 vs 4.4 ± 0.5 mg, normal weight).

In another retrospective study, Mueller et al[6] sought to determine the association between BMI and the total weekly dose (TWD) of warfarin in a total of 831 patients of different BMI categories. All of the patients included in the study were on a stable dose of warfarin, defined as having a therapeutic INR without a change in warfarin dose for at least 30 days. Exclusion criteria included starting warfarin treatment within the prior month or missing data (demographics, baseline or target INR, or current weekly dose). The weekly dose of warfarin varied between 3.5 and 140 mg with an average TWD of 32 mg and the BMI range varied between 13.4 and 63.1 kg/m2. The mean TWD of warfarin in obese patients in classes I, II, and III (34.26, 35.9, and 45.77 mg, respectively) was higher than in patients with normal body weight (27.7 mg). The study found a positive correlation between BMI and TWD in which for each 1-point increase in BMI the weekly warfarin dose increased by 0.69 mg (P < 0.001). The authors concluded that patients with a higher BMI would require a higher TWD of warfarin to maintain a stable therapeutic INR.

Moffett and Bomgaars[7] evaluated the response to warfarin therapy in obese pediatric patients newly initiated on warfarin and dosed according to institutional guidelines in an inpatient setting. A total of 30 patients were included in the study in which obese patients were matched in a 1:2 ratio by age and sex with nonobese patients. No differences were noted in the obese and nonobese patient groups with regard to percentage of patients who had a supratherapeutic INR value or patients discharged with a therapeutic INR; however, the time to reach a therapeutic INR in obese patients was twice as long as nonobese (6 vs 3 days, respectively, P < 0.01). The milligram per kilogram warfarin dose was lower in obese versus nonobese patients (0.06 mg/kg vs 0.11 mg/kg, respectively; P < 0.01). The maximum milligram dose was higher in obese patients (7.2 vs 5.1 mg; P < 0.01), however. From the results of the study, the authors concluded that obese pediatric patients initiated on traditional warfarin dosing schemes have an increased time to therapeutic INR compared with nonobese patients and suggested modifying warfarin dosing algorithms for pediatric patients to exclude a maximum initial dose when dosed by weight.[7]

Yoo et al[8] evaluated potential variables such as weight, age, and comorbidities in a program to individualize warfarin doses. In this group of 321 stroke patients, the effect of these factors on maintenance warfarin doses was assessed. Maintenance doses were evaluated in patients who had INRs in the therapeutic range for a minimum of three successive clinic visits. Data analysis revealed a highly significant effect of age (P < 0.001) and weight (P = 0.001) on the warfarin maintenance dose.[8] In addition, stepwise multiple linear regression found a lower warfarin dose requirement associated with lower body weight added to age (P = 0.008). Based on these results an individualized initial warfarin dosing program was developed. For example, patients older than 80 years or patients <55 kg older than 55 years would receive lower initial warfarin maintenance doses.

Yoo et al[9] assessed their warfarin dosing program (including age and weight) in a subsequent study of 104 acute stroke patients. The initial warfarin dosing nomogram included high (10 mg; 55 years and younger, weight ≥55 kg), standard (7 mg; age 55 years and younger, weight <55 kg, or older than 55 but younger than 80 years, weight ≥55 kg), and low (5 mg; older than 55 but younger than 80 years; weight <55 kg or age 80 years and older) doses for days 1 and 2, with further doses based on INR. These patients were compared with an historical control group of patients (n = 96) with physician-determined warfarin doses. More patients in the warfarin dosing program achieved therapeutic INR at 5 days (P = 0.002) versus the physician-determined warfarin dosing group. Excessive INR (≥3.5) during hospitalization also was lower in the warfarin dosing program (P = 0.024).

Miao et al[10] studied the effect of body weight and age and the effect of CYP2C9 and VKORC1 genotypes on response to warfarin in Chinese patients. Among 178 patients who had stable warfarin dose requirements for at least three clinic visits for 3 months, body weight and age were significant factors on warfarin dose (P < 0.001). CYP2C9 and VKORC1 genotypes also were important factors affecting warfarin dose.

Phenprocoumon. Meyer zu Schwabedissen et al[11] found a slower initial response to phenprocoumon associated with obesity. Patients (n = 260) who were initiated on this VKA in the hospital were evaluated for several factors that may affect response to VKA, including CYP2C9 genotype, BMI, diet, comorbidities, and concomitant drug therapy. Endpoints for this investigation included time required to achieve the target INR and the total amount of phenprocoumon needed to reach the target INR.

Study results revealed that the lower the BMI, the less time it took to achieve the target INR.[11] For example, in patients with BMI <22, the mean time required to reach the target INR was 5.48 days versus 8.5 days for BMI >30. The mean dose required to reach target INR for BMI <22 was 23.8 mg versus 35.8 mg for BMI >30. Results for the studies including a specific goal of assessing body size are summarized in Table 1.

Other Studies Evaluating Body Weight or Size, Among Various Factors

Absher et al[12] conducted a prospective study of 146 patients to identify patient-specific factors predictive of maintenance warfarin dosage requirements >5 mg/day. Patients were stratified from an inpatient community hospital and outpatient clinic into two groups based on the stable daily warfarin maintenance doses: ≤5 mg/day (n = 109) and >5 mg/day (n = 37). Using multiple logistic regression, five factors were identified as independent factors for increased daily warfarin dose. Body weight ≥91 kg was one of the independent factors, in addition to age younger than 55 years, male sex, African American ethnicity, and vitamin K intake >400 μg/day.[12]

Ageno et al[13] conducted a retrospective study of the initial phase of warfarin therapy in outpatients with deep venous thrombosis. In this study of 55 patients (mean age 61.4 years, 29 men), the investigators evaluated predictors of a slow achievement of a therapeutic INR and the concurrent duration (minimum 5 days) of low-molecular-weight heparin treatment. Among these 55 patients, 47.3% were overweight (BMI ≥25 kg/m2) and 14.5% were obese (BMI ≥30 kg/m2). All of the patients were initiated on 5 mg/day warfarin. Patients received follow-up in the clinic for INR checks and warfarin dose adjustments in 3 ± 1 days and then again at 6 ± 1 days. Patients were excluded from the study if warfarin was not started on day 1 with low molecular weight heparin or if they were receiving drugs well documented to affect the response to warfarin. A therapeutic INR was achieved within 7 days in 54.4% of patients, whereas the remaining 45.6% of patients took an average of 10.5 days to achieve the goal INR. Although a statistical correlation was found only between age and warfarin dose among the total population, the authors reported that men younger than 65 years with a BMI >28 kg/m2 required the highest doses, whereas women older than 65 years with a BMI <25 kg/m2 required the lowest doses of warfarin. No statistical difference was found based only on BMI.

Sakaan et al[14] conducted a retrospective study to determine the effect of chronic kidney disease (CKD) on response to warfarin in 210 hospitalized patients. In addition to normal kidney function (n = 49), patients with CKD stage 3 (n = 44), stages 4/5 (n = 27), and end-stage renal disease (n = 90) were included. The average daily dose of the VKA to maintain a therapeutic INR was determined. In patients with CKD/end-stage renal disease, the average daily dose of warfarin was approximately 20% lower compared with patients with normal renal function. Of note, across all stages of kidney function, as weight increased by 10 kg, the average daily dose of warfarin increased by 0.28 ± 0.08 mg.[14]

Shine et al[15] conducted a randomized trial comparing initial fixed doses of warfarin (5 mg) with doses based on simple clinical criteria. Among 90 patients (fixed dose, n = 46, mean weight 194 lb; calculated dose, n = 44, mean weight 182 lb), 25% to 30% of the variance in dose was affected by weight, age, serum albumin, and presence of malignancy.[15] Analysis based on weight alone was not reported. Patients with the calculated dose reached therapeutic anticoagulation quicker (4.2 days vs 5.0 days; P = 0.007).

Jiang et al[16] assessed the warfarin dose requirement in patients with pulmonary embolism complicated by concomitant obstructive sleep apnea hypopnea syndrome (OSAHS). In this prospective study, 97 patients with pulmonary embolism were enrolled consecutively, and 32 of these patients had OSAHS. Patients with OSAHS had a higher warfarin dose requirement compared with the non-OSAHS group (P < 0.001). Logistic analysis revealed that OSAHS was an independent risk factor for a higher warfarin dose. Body weight also significantly affected warfarin dose (P = 0.001). Table 2 summarizes these studies.

Investigations Primarily Evaluating Pharmacogenetics

Several VKA pharmacogenetic studies have included body weight or size in the data analyses.[17–35] Among these investigations of warfarin pharmacogenetics, some were focused on Japanese,[20] Turkish,[21] Sudanese,[22] Chinese,[24–27] and African American[29,30] populations. Body weight or size consistently was found to have an effect on warfarin dose. One study in Iran did not find an effect of body weight on the warfarin dose requirement.[31] Likewise, pharmacogenetic investigations of acenocoumarol, phenprocoumon, and fluindione have included body weight.[32–35] In each of these studies, body weight was a significant factor in the VKA dose requirement.