Therapeutic use of warfarin

Therapeutic use of warfarin

Karen A Valentine, MD, PhD
Russell D Hull, MBBS, MSc
Section Editor
Lawrence LK Leung, MD
Deputy Editor
Stephen A Landaw, MD, PhD
Last literature review version 19.3: Fri Sep 30 00:00:00 GMT 2011 | This topic last updated: Mon Oct 17 00:00:00 GMT 2011 (More)

INTRODUCTION — Warfarin and other vitamin K antagonists (VKAs, eg, acenocoumarol, phenprocoumon, fluindione) are the standard oral anticoagulants used in a variety of clinical settings. The general principles underlying the clinical use of VKAs, including their complications and laboratory monitoring will be reviewed here [1]. Other aspects of VKA therapy are discussed separately:


  • The outpatient management of the patient taking warfarin, including the importance of strict INR control. (See “Outpatient management of oral anticoagulation”, section on ‘Importance of strict INR control’.)
  • Correction of excess anticoagulation on warfarin therapy. (See “Correcting excess anticoagulation after warfarin”.)
  • Management of anticoagulation in patients who have had, or are at risk for, intracranial bleeding. (See ‘Intracranial bleeding’ below.)
  • A patient information document on warfarin therapy. (See “Patient information: Warfarin (Coumadin®)”.)



Warfarin — Commercially-available warfarin is a racemic mixture of S and R enantiomers. The more potent S form of the drug is metabolized primarily by the CYP2C9 hepatic microsomal enzyme system. This enzyme system is inducible by many drugs and has a number of genetic variants, both of which may profoundly alter warfarin’s in vivo activity (see ‘Genetic interactions’ below).

Warfarin is strongly protein-bound, primarily to albumin; only the non-protein-bound fraction is biologically active. Accordingly, any agent that is also bound to albumin may displace warfarin from its albumin binding sites and increase its biological activity.

Warfarin is water soluble and completely absorbed after oral administration. The majority of the drug is absorbed in the proximal small bowel, although successful use of the sublingual route has been described in two patients [2,3]. Excretion is via the urine, primarily as drug metabolites.

Other vitamin K antagonists — Warfarin has a biological half-life of 36 to 42 hours, considerably longer than that of the closely related anticoagulant acenocoumarol (8 to 11 hours) and shorter than that of phenprocoumon (3 to 5 days) [4] and fluindione (69 hours) [5]. Thus:


  • Moderate overdosage of acenocoumarol may be more easily treated by omitting a single dose of this agent, in comparison to a similar overdosage of warfarin [6]. (See “Correcting excess anticoagulation after warfarin”, section on ‘INR 5 to 9 without bleeding’.)
  • Because of its longer-half life, phenprocoumon produces more stable anticoagulation than acenocoumarol [4] and may be less susceptible to the effects of CYP2C9 polymorphisms [7].


Tecarfarin — Tecarfarin (ATI-5923) is a novel vitamin K antagonist currently in clinical trials; it is metabolized by esterases rather than the cytochrome P450 system [8]. Accordingly, it is not susceptible to the drug-drug, drug-food, and genetic alterations common to the other vitamin K antagonists.

Mechanism of action — The anticoagulant effect of warfarin is mediated through inhibition of the vitamin K-dependent gamma-carboxylation of coagulation factors II, VII, IX, and X [2,9]. (See “Vitamin K, gamma carboxyglutamic acid, and the function of coagulation and other proteins”.)

This effect of warfarin results in the synthesis of immunologically detectable but biologically inactive forms of these coagulation proteins. Warfarin also inhibits the vitamin K-dependent gamma-carboxylation of proteins C and S, which have anticoagulant properties through their inhibition of activated factors VIII and V [10]. (See “Overview of hemostasis”.)

Because of these competing effects, vitamin K antagonists such as warfarin create a biochemical paradox by producing an anticoagulant effect due to the inhibition of procoagulants (factors II, VII, IX, and X) and a potentially thrombogenic effect by impairing the synthesis of naturally occurring inhibitors of coagulation (proteins C and S) [10].

The ultimate anticoagulant effect of warfarin is delayed until the normal clotting factors, especially prothrombin, are cleared from the circulation. The peak effect does not occur until 36 to 72 hours after drug administration (figure 1) [11], especially because the plasma half-life of factor II (prothrombin) is approximately 3 days (table 1).

During the first few days of warfarin therapy, prolongation of the prothrombin time mainly reflects its action on the extrinsic coagulation pathway, through the depression of factor VII, which has a half-life of four to six hours (table 1 and figure 2). The other vitamin K-dependent factors within the common and intrinsic coagulation pathways (ie, factors II, IX, and X) remain relatively unchanged during the first few days. Accordingly, the patient is not fully anticoagulated with warfarin until these other components are also reduced.

Equilibrium levels of factors II (prothrombin), IX, and X, approximately 10 to 35 percent of normal at therapeutic INR levels [12], are reached about one week after the initiation of therapy. For this reason, parenteral anticoagulants and warfarin should OVERLAP by four to five days when warfarin is initiated in patients with acute thrombotic disease (eg, venous thromboembolism, heparin-induced thrombocytopenia) [2]. (See “Treatment of lower extremity deep vein thrombosis”, section on ‘Initial anticoagulation regimen’ and “Heparin-induced thrombocytopenia”, section on ‘Warfarin’.)

Laboratory monitoring — The test most commonly used to measure the effects of warfarin is the one-stage prothrombin time (PT). The PT is sensitive to reduced activity of factors II, VII, and X, but is insensitive to reduced activity of factor IX. Confusion about the appropriate therapeutic range has occurred because the different tissue thromboplastins used for measuring the PT vary considerably in their sensitivity to the vitamin K-dependent clotting factors and in response to warfarin [13]. (See “Clinical use of coagulation tests”, section on ‘Prothrombin time’.)

In order to promote standardization of the PT for monitoring oral anticoagulant therapy, the World Health Organization (WHO) developed an international reference thromboplastin from human brain tissue and recommended that the PT ratio be expressed as the International Normalized Ratio or INR [14].

Use of the INR — The INR is the PT ratio obtained by testing a given sample using the WHO reference thromboplastin. For practical clinical purposes, the INR for a given plasma sample is equivalent to the PT ratio obtained using a standardized human brain thromboplastin known as the Manchester Comparative Reagent, which has been widely used in the United Kingdom [14]. (See “Clinical use of coagulation tests”, section on ‘Measurement of INR’.)

Serial monitoring of the INR will detect many patients who are overanticoagulated before they have had a bleeding episode. However, monitoring is not completely protective because, among patients who have a bleeding episode associated with a high INR, there may be only a brief warning period during which a slightly elevated INR predicts for an imminent bleeding event. This was illustrated in a review of 32 patients with a warfarin-related hemorrhage [15]. The mean INR at the time of the bleeding event was 5.9; in contrast, the mean INR at the last prior measurement, which was obtained an average of 12 days before the bleeding event, was 3.0.


Genetic interactions — Polymorphisms in the genes for the following two enzymes have been associated with altered sensitivity to warfarin [16-22]:


  • Hepatic cytochrome P-450 2C9 (CYP2C9), involved in the metabolic clearance of warfarin. (See “Overview of pharmacogenomics”, section on ‘CYP isoenzymes and drug metabolism’.)
  • Vitamin K epoxide reductase complex 1 (VKORC1), which recycles vitamin K and is required for gamma carboxylation of vitamin K-dependent coagulation factors. (See “Vitamin K and the synthesis of gamma carboxyglutamic acid”, section on ‘Recycling of vitamin K’.)


Cytochrome P-450 2C9 — A well-studied cause for individuality in patient responses to warfarin is the presence of genetic variation (polymorphisms) in the hepatic cytochrome P-450 2C9 (CYP2C9) isoenzyme, which inactivates both warfarin [23,24] and acenocoumarol [25,26].

These polymorphisms (wild-type CYP2C9*1 and variants CYP2C9*2 and CYP2C9*3) were studied in 561 warfarin-treated patients; mean maintenance doses required to reach a target INR of 2.5 were, according to genotype [27]:


  • *1*1 wild type (70 percent of group) — 5.0 mg
  • *1*2 heterozygote (19 percent) — 4.3 mg
  • *1*3 heterozygote (9 percent) — 4.0 mg
  • *2*3 compound heterozygote (1 percent) — 4.1 mg
  • *2*2 homozygote (0.5 percent) — 3.0 mg


In this and an additional study, patients whose daily warfarin requirement was ≤1.5 mg/day were five to six times more likely to have a variant allele [23,27,28]. During induction of therapy, they were six times more likely to have an INR >4.0 and four times more likely to have major bleeding. In addition, those with at least one variant allele required a significantly longer time (median difference 95 days) to reach a stable dose, effects seen in other studies [29-31].

The effect of CYP2C9 polymorphisms on the time to the first INR within the therapeutic range, to the first INR >4, and the average dose of warfarin after 29 days was studied in 297 patients undergoing warfarin anticoagulation with an average starting dose of 4.8 ± 0.8 mg/day. Results included [19]:


  • The CYP2C9 genotype was not a significant predictor of the time to first INR within the therapeutic range, but was a significant predictor of the time to the first INR >4.
  • The CYP2C9 genotype was a significant predictor of the average dose of warfarin required after day 29 to achieve the desired INR. The dose was highest in those with the *1/*1 genotype (5.2 mg/day), intermediate in those with the *1/*2 or *1/*3 genotype (4.2 mg/day) and lowest in those with the *2/*2, *3/*3, or *2/*3 genotypes (3.4 mg/day).


In a separate study, a significant interaction between the presence of local bleeding sources and the CYP2C9*2 and/or *3 alleles was seen [32].

Vitamin K epoxide reductase complex 1 — Polymorphisms in the gene encoding vitamin K epoxide reductase complex 1 (VKORC1) affect the patient’s response to warfarin and acenocoumarol [17,26,33-37]. (See “Vitamin K and the synthesis of gamma carboxyglutamic acid”, section on ‘Mutations and polymorphisms of the VKOR complex’.)

In one study, a low-dose haplotype group (group A) and a high-dose haplotype group (non-A or group B) were identified, with the following mean warfarin maintenance doses [33]:


  • Group A/A — 2.7 ± 0.2 mg/day
  • Group A/B — 4.9 ± 0.2 mg/day
  • Group B/B — 6.2 ± 0.3 mg/day


Asian-Americans had a higher proportion of group A haplotypes, while African-Americans had a higher proportion of group B haplotypes. It was estimated from this study that genetic alterations in CYP2C9 and VKORC1 accounted for 6 to 10 percent and 21 to 25 percent of the variance in warfarin dose, respectively. Similar studies in patients from France, Slovenia, and Sweden have estimated that polymorphisms in these two enzymes can account for up to 60 percent of interindividual variability in response to treatment with warfarin [22,38] or acenocoumarol [26].

The effect of VKORC1 polymorphisms on the time to the first INR within the therapeutic range, to the first INR >4, and the average dose of warfarin after 29 days to achieve the desired INR was studied in 297 patients undergoing warfarin anticoagulation with an average starting dose of 4.8 ± 0.8 mg/day. Results included [19]:


  • The VKORC1 genotype was a significant predictor of both the time to first INR within the therapeutic range as well as the time to the first INR >4. Presence of the A allele was associated with an accelerated and greater sensitivity to warfarin.
  • Initial variability in the INR response to warfarin was more strongly associated with genetic variability in VKORC1 than with CYP2C9.
  • The VKORC1 genotype was a significant predictor of the average dose of warfarin required after day 29 to achieve the desired therapeutic INR. The dose was highest in those with the non-A/non-A genotype (5.7 mg/day), intermediate in those with the non-A/A genotype (4.4 mg/day) and lowest in those with the A/A genotype (3.7 mg/day).


A reduced starting dose may be considered in patients with A haplotypes, in Asian-Americans, and in selected patients known to require lower doses of warfarin (eg, age >65, presence of liver disease, malnourished patient, or taking a medication that potentiates warfarin) [18]. (See ‘Initial dose’ below and ‘Genotype testing’ below.)

Cytochrome P-450 4F2 — Although the exact mechanism by which this enzyme affects vitamin K or warfarin metabolism is not known, a DNA variant in cytochrome P-450 4F2 (CYP4F2 V433M) was associated with warfarin dose in three independent cohorts of patients taking stable doses of warfarin [39]. The difference amounted to a 4 to 12 percent increase in the required warfarin dose per T allele, as compared with that needed in those having the more common CC variant.

Similarly, CYP4F2 V433M polymorphism has been shown to play a relevant role in the earliest response to treatment with acenocoumarol and the required maintenance dose [40].

Pharmacokinetic and pharmacogenetic dosing — Given the many variables affecting the maintenance dose of vitamin K antagonists, as well as their narrow therapeutic range, efforts are underway to develop computer-assisted dosing programs for selecting appropriate initial dosing for warfarin [18,41-51], and for acenocoumarol and phenprocoumon [52]. While some of these programs have aided selection of initial dosing of warfarin, especially for those requiring the lowest (≤21 mg/week) or highest (≥49 mg/week) stable doses [49,53], randomized, controlled studies using these programs have not yet resulted in a significant improvement in clinical outcomes (eg, longer time in therapeutic range, lesser incidence of adverse events); initial studies also differ as to whether or not these programs are cost-effective [50,54,55]. Some of these studies are reviewed below.


  • In a report of 369 patients taking a stable dose of warfarin, an algorithm that included multiple factors (eg, age, race, sex, body surface area, CYP2C9 polymorphisms, target INR, use of amiodarone or simvastatin) explained 39 percent of the variance in the maintenance warfarin dose [45]. It was estimated that significantly fewer (6.5 percent) patients in this study would have been over-dosed via the use of pharmacogenetic dosing compared with the 16 percent of patients who would have been overdosed if they had been prescribed an empirical dose of 5 mg/day. This algorithm is available online at www.WarfarinDosing.org/Source/Home.aspx, and needs to be validated for other clinical settings [56].
  • In a prospective study, 206 patients were randomly assigned to start warfarin therapy using either a standard dosing regimen or pharmacogenetic-guided therapy (PGT) using a regression equation including CYP2C9 and VKORC1 genotypes, age, sex, and weight [47]. While PGT predicted stable warfarin doses more accurately than the standard regimen, the percent out-of-range INRs (31 versus 33 percent), the primary end point of the study, did not differ significantly between treatment arms.
  • An international randomized study of oral anticoagulant dosing using two commercial computer-assisted dosage programs (PARMA 5 and DAWN AC) versus manual medical staff dosage was performed in 13,052 patients from 32 centers [48]. The time in target INR range was significantly improved by computer assistance as compared with medical staff dosage, with the greatest advantage being seen at those medical centers with fewer patient-years of experience. However, the overall number of adverse clinical events (ie, bleeding, thrombosis, death) was not significantly reduced (5.5 versus 6.0 events/100 patient-years for the computer-assisted and manual dosage groups, respectively; adjusted incidence rate ratio 0.90; 95% CI 0.80-1.02).
  • In a study of 266 acenocoumarol-treated patients, 89 of whom had gastrointestinal bleeding, the presence of a VKORC1 or cytochrome P450 polymorphism did not constitute a significant bleeding risk [57]. The risk of bleeding in such carriers was significantly increased only if the weekly amiodarone dose exceeded 15 mg, or the patient was also taking amiodarone or aspirin.
  • In a study of 1496 Swedish patients starting warfarin therapy and genotyped for 183 polymorphisms in 29 candidate genes, homozygosity for CYP2CP or VKORC1 variant alleles significantly increased the risk of over-anticoagulation (ie, INR >4) [22]. However, no significant association was found between CYP2CP or VKORC1 genotypes and clinical bleeding during the study.
  • In a non-randomized study, the 896 subjects who received warfarin genotyping for CYP2C9 and VKORC1 had a significantly lower six-month risk for hospitalization for bleeding or thromboembolism than a matched historical group of 2688 controls (HR 0.72; 95% CI 0.53-0.97) [58].


The accuracy of several warfarin dose prediction tables and algorithms was compared in a cohort of 1378 patients taking a stable dose of warfarin, who also had complete data for dose prediction by all dosing methods. Methods included a clinical algorithm, dosing according to a newly revised warfarin label, dosing according to a genotype mean table, empiric dosing of 35 mg/week, and a pharmacogenetic algorithm [59]. While the pharmacogenetic algorithm was significantly more accurate in predicting (within ± 20 percent) the actual therapeutic warfarin dose than all of the other methods, 48 percent of the patients were ultimately not dosed accurately even by this method.

The National Heart, Lung, and Blood Institute will be sponsoring a randomized trial comparing trial and error dosing, clinical algorithm dosing, and clinical plus genetic algorithm dosing. However, pharmacogenetic-based dosing was not recommended for routine use in the 2008 ACCP Guidelines [2], and a cost-effectiveness study of using pharmacogenetic information for warfarin dosing concluded that, at its current cost (see below), routine genotyping before warfarin dosing is unlikely to be cost-effective for typical patients with nonvalvular atrial fibrillation (ie, an estimated marginal cost-effectiveness of testing exceeding $170,000 per quality-adjusted life-year gained) [54].

In May 2009 the Centers for Medicare and Medicaid Services (CMS) announced a decision to decline payment for warfarin genetic testing unless administered as part of a clinical trial comparing outcomes in tested and untested patients.

Genotype testing — The United States Food and Drug Administration has approved a genetic test for detecting variants of the CYP2C9 and VKORC1 genes, which will help clinicians assess whether a patient may be especially sensitive to warfarin. Information on this test (the Nanosphere Verigene Warfarin Metabolism Nucleic Acid Test, Nanosphere, Inc., Northbrook, IL), which has been estimated to cost approximately 300 to 400 US Dollars per patient [54], is available on the United States FDA website at: www.accessdata.fda.gov/cdrh_docs/pdf7/K070804.pdf [60]. (See “Overview of pharmacogenomics”.)

Drug interactions — A large and increasing number of drugs interact with warfarin. Critical appraisal of the literature reporting such interactions indicates that the evidence substantiating some of the claims is limited [61]. Nonetheless, interactions leading to overanticoagulation, underanticoagulation, or increased bleeding independent of changes in the INR are well described for a large number of drugs. Mechanisms that can be associated with such interactions include [62]:


  • Altered platelet function (eg, aspirin, clopidogrel [63])
  • Gastrointestinal injury (eg, NSAIDs)
  • Altered vitamin K synthesis in the GI tract (eg, antibiotics)
  • Alterations in warfarin metabolism (eg, amiodarone, rifampin, simvastatin, gemfibrozil) [64]
  • Interference with vitamin K metabolism (eg, acetaminophen [65], although the effect is variable [66,67])


As an example of how often these medication interactions occur, in one study performed in a primary care setting, nearly one-third of patients taking warfarin had also been prescribed a medication known to interact with warfarin (eg, acetaminophen, non-steroidal anti-inflammatory drugs, fluconazole, metronidazole, sulfamethoxazole) [68].

Accordingly, patients must be warned against taking any new drugs, including herbal products, over-the-counter medications, and even cutaneous application of large amounts of potentially interfering drugs (eg, topical econazole or bifonazole [69]) without the knowledge of the responsible clinician (table 2 and table 3). (See “Overview of herbal medicine and dietary supplements”, section on ‘Herb-drug interactions’.)

Antibiotics — A population-based cohort study evaluated the risk of overanticoagulation, as defined by an INR >6.0, caused by concomitant antibiotic use in outpatients taking the coumarin anticoagulants acenocoumarol or phenprocoumon [70]. Overall, 351 of the 1124 subjects (31 percent) developed an INR >6.0, with an incidence of 6.9 per 10,000 treatment days. The risk of overanticoagulation was most strongly increased by amoxicillin, clarithromycin, norfloxacin, and trimethoprim-sulfamethoxazole, often within the first three days of antibiotic usage.

The relationship between hospitalization for upper gastrointestinal (UGI) hemorrhage and the use of antibiotics was studied in a population-based nested case-control study of Ontario residents ≥66 years of age who had been taking warfarin for at least 180 days [71]. Concomitant use of trimethoprim-sulfamethoxazole (OR 3.84; 95% CI 2.33-6.33) or ciprofloxacin (OR 1.94; 95% CI 1.28-2.95) was associated with a significant increase in hospital admission for UGI hemorrhage. Admission for UGI hemorrhage was not associated with the use of nitrofurantoin, amoxicillin, ampicillin, or norfloxacin in the 14 days prior to admission. Approximately 10 percent of those hospitalized for this complication died before hospital discharge.

SEE MORE:  Treating Chronic Fatigue Syndrome with Acupuncture: Let the Energy Flow

Proton pump inhibitors — A population-based cohort study evaluated the risk of overanticoagulation, as defined by an INR >6.0, caused by concomitant use of proton pump inhibitors in outpatients taking acenocoumarol [72]. Overall, 887 of the 2755 subjects (32 percent) developed an INR >6.0. The risk of overanticoagulation was most strongly increased by use of esomeprazole (HR 1.99; 95% CI 1.55-2.55) and lansoprazole (HR 1.49; 95% CI 1.05-2.10). The authors did not detect a modification of these results according to the subjects’ CYP2C19*2 genotype.

Use of NSAIDs and antiplatelet agents — Use of aspirin, nonsteroidal antiinflammatory drugs (NSAIDs), and antiplatelet agents (eg, clopidogrel, dipyridamole) increases the risk of bleeding in vitamin K antagonist-treated patients via multiple mechanisms, including gastrointestinal toxicity, enhancing the effect of the antagonist by increasing the INR, as well as interference with platelet function [73-76]. (See “NSAIDs: Mechanism of action”, section on ‘Cyclooxygenase inhibition’ and ‘Bleeding’ below and “Nonselective NSAIDs: Overview of adverse effects”.)

Use of COX-2-selective NSAIDs may lead to fewer bleeding complications, as these agents do not interfere with platelet function. (See “Overview of selective COX-2 inhibitors”, section on ‘Lack of effect upon platelets’.)

The lower bleeding risk with COX-2 selective NSAIDs was demonstrated in a nested case-control study from the Netherlands in patients taking phenprocoumon or acenocoumarol along with NSAIDs. Over a two-year period there were 1,491 bleeding episodes, 15 percent of which involved the use of NSAIDs [74]. On multiple regression analysis, there was a significantly increased risk for bleeding in the following settings:


  • Use of a nonselective versus a COX-2 selective NSAID — odds ratio (OR) 3.1
  • NSAID use for >1 month (versus ≤1 month) — OR 3.0
  • Last INR >4.0 (versus ≤4.0) — OR 1.9


Low-dose aspirin — The use of therapeutic doses of warfarin along with low-dose aspirin (ie, 81 to 100 mg/day) is quite common. In one study of Kaiser Permanente Colorado members ≥18 years of age, the prevalence of this treatment combination was 328 per 1000 patients anticoagulated with warfarin, and was most commonly employed in patients with heart failure, coronary heart disease, stroke, or transient ischemic attack [75]. The overall benefits and risks of this combination of low-dose aspirin and warfarin are discussed in depth elsewhere. (See “Antithrombotic therapy in patients with prosthetic heart valves”, section on ‘Warfarin plus aspirin’ and “Benefits and risks of aspirin in secondary and primary prevention of cardiovascular disease”, section on ‘Aspirin induced bleeding’.)

In one large study using records from the United Kingdom, warfarin plus low-dose aspirin, as well as its combination with an NSAID or Cox-2 inhibitor, was associated with a risk for gastrointestinal bleeding greater than that for either agent alone, as follows [63]:


  • No drugs — adjusted risk (AR) 1.00 (comparator)
  • Aspirin alone — AR 1.4
  • Cox-2 inhibitors alone — AR 1.6
  • NSAID alone — AR 1.8
  • Warfarin alone — AR 1.9
  • Cox-2 inhibitor plus warfarin — AR 4.6
  • NSAID plus warfarin — AR 4.8
  • Aspirin plus warfarin — AR 6.5


Similar results were obtained from a large primary care database, in which the adjusted relative risks for development of upper gastrointestinal bleeding were 1.00, 1.77, 1.82, and 3.62 for those taking no drugs (comparator), oral anticoagulants alone, low-dose aspirin alone, or both agents [77].

Influence of diet — The influence of dietary intake of vitamin K on the stability of control of anticoagulation with vitamin K antagonists is discussed separately. (See “Outpatient management of oral anticoagulation”, section on ‘Influence of diet’.)

Influence of smoking — A systematic review and meta-analysis has concluded that smoking may potentially cause significant interaction with warfarin by increasing warfarin clearance, requiring a 12 percent (95% CI 7-17) increase in warfarin dosage compared with nonsmokers [78].


Loading doses — The use of warfarin with initial (“loading”) doses in excess of 5 mg/day has several potential complications, including a transient hypercoagulable state due to a precipitous decline in protein C levels in first 36 hours (figure 1). These higher doses of warfarin also produce a precipitous decline in factor VII, resulting in an initial prolongation of the INR before onset of the full antithrombotic effect, which does not occur until there has been significant reduction in factor II. This also illustrates the necessity for overlap of heparin and warfarin therapy for four to five days, in order to have the benefit of anticoagulation from heparin until the full effect of warfarin has been achieved.

The safety and efficacy of two different loading doses of warfarin were evaluated in a trial of 49 patients who were randomly assigned to receive initial warfarin doses of either 5 or 10 mg [79]. This trial used surrogate laboratory markers for safety (prothrombin time elevation and protein C levels) and efficacy (time course of factor II reduction). The following observations were noted:


  • There was no significant difference between the two regimens with respect to the reduction in factor II levels as a marker of efficacy.
  • The 5 mg loading dose was less likely to produce excess anticoagulation (ie, INR greater than 3.0) than the 10 mg dose; it also produced a less marked fall in protein C and factor VII levels.


A similar conclusion was reached in a second study involving 53 patients, which investigated the ability of initial loading doses of 5 or 10 mg to reach both of the following goals: attainment of a therapeutic INR (2.0 to 3.0) on two consecutive daily determinations on days 3, 4, or 5 of the study; and avoidance of an INR >3.0 at any point [80]. These goals were reached in 66 percent of patients randomly assigned to receive the 5 mg/day initial doses and in only 24 percent of patients receiving the 10 mg/day doses.

In an analogous study, the 66 subjects randomly assigned to receive the 10 mg/day initial dose achieved a therapeutic INR significantly earlier than the 66 receiving the 5 mg/day initial dose [81]. Mean INRs on the fifth to seventh day of treatment were in the excessive range (ie, 3.0 to 5.0) for 58 to 61 percent of those in the 10 mg/day group as compared with 8 to 12 percent for those in the 5 mg/day group. Bleeding and thrombotic events were similar in the two treatment groups.

In a study in 201 patients randomly assigned to receive initial doses of either 5 or 10 mg/day, those in the 10 mg/day arm achieved therapeutic INR levels at a mean of 1.4 days earlier than those in the 5 mg/day arm; significantly more patients in the 10 mg/day arm were in the therapeutic range at five days (83 versus 46 percent) [82]. There were no differences between the two arms in terms of recurrent thrombotic events, major bleeding, survival, or number of INR measurements exceeding 5.0.

Initial dose — Because of variability in the rate of drug metabolism and vitamin K dietary status [83-86], the selection of the initial dose of warfarin must be individualized, especially in older patients and others at increased risk for bleeding. (See ‘Genetic interactions’ above and ‘Drug interactions’ above and ‘Bleeding’ below and “Anticoagulation in older adults”, section on ‘Initiation and monitoring’.)

Suggested initial oral doses of warfarin for the first two days have ranged from a low of 2 mg/day to as high as 10 mg/day, with the daily dose subsequently adjusted up or down according to the INR. Higher initial (“loading”) doses of warfarin are not recommended for reasons presented elsewhere. (See ‘Loading doses’ above and “Correcting excess anticoagulation after warfarin”, section on ‘Treatment’.)

The following issues are of importance in the selection of the initial dose within this suggested range:


  • Initial doses at the lower end of this range (ie, 2 to 5 mg/day) are suggested for patients with hepatic impairment, poor nutrition, debilitation, congestive failure, the elderly, those with severe chronic kidney disease [87], and those at high risk of bleeding. Patients with certain genetic polymorphisms or those taking a medication that decreases the metabolic degradation rate of warfarin may also require lower starting doses [2,88]. (See ‘Genetic interactions’ above.)
  • Initial doses at the high end of this range (ie, ≤10 mg/day) may be reasonable in highly selected patients, such as stable, reliable patients with a low bleeding risk in whom prior treatment with warfarin required higher maintenance doses. However, as noted above, such higher doses have resulted in overanticoagulation in three randomized trials. (See ‘Loading doses’ above.)


As a result of available studies, we suggest that initial doses of warfarin in excess of 5 mg/day not be employed. This suggestion is more conservative than the 2008 ACCP Guidelines which suggest initial doses in the range of 5 to 10 mg/day for most individuals, but is consistent with their suggestion of reduced initial doses (ie, ≤5 mg/day) in elderly patients, those that are debilitated or malnourished, have congestive failure, liver disease, recent major surgery, or are taking medication known to increase sensitivity to warfarin (eg, amiodarone) [2]. (See “Anticoagulation in older adults”.)

The role of testing for genetic polymorphisms for determination of the starting dose of warfarin is not clear and was not suggested in the 2008 ACCP Guidelines [2]. The United States Food and Drug Administration (FDA) is involved in a Critical Path project related to the pharmacogenetics of warfarin therapy in an effort to resolve this issue [89,90].

Adjusting warfarin dosing — No protocols for initial dosing of warfarin and for adjusting subsequent doses have been uniformly accepted [2,80,82,88,91-93]. Despite this, the UpToDate authors and editors feel that we should provide some general guidance on this issue. The suggested dosing protocol (table 4) is one currently available at the Beth Israel Deaconess Medical Center in Boston, MA, which also contains all of the elements of the most commonly cited schedule for a 5 mg/day starting dose of warfarin [82]. While this suggested protocol requires confirmation through further testing, if used in conjunction with clinical judgement, it should be of clinical value to our readers.

Maintenance therapy — Maintenance doses of warfarin vary significantly from patient to patient, ranging from <2 mg/day to ≥10 mg/day, depending upon a number of factors, such as the patient’s nutritional status, hepatic and renal function, rate of intestinal absorption, genetic factors altering warfarin pharmacokinetics, degree of compliance, and the presence of drug interactions [94-97]. The goal INR varies with the clinical state, ranging from 2.0 to 3.0 in venous thromboembolism to somewhat higher ranges in patients with mechanical heart valves. (See “Treatment of lower extremity deep vein thrombosis”, section on ‘Warfarin’ and “Antithrombotic therapy in patients with prosthetic heart valves”.)

Frequent INR determinations are required initially to establish that therapeutic anticoagulation levels have been achieved. (See “Outpatient management of oral anticoagulation”, section on ‘Importance of strict INR control’.)

Once the anticoagulant effect and patient’s warfarin dose requirements have been stabilized for at least one to two weeks, the INR can be monitored less frequently, at intervals in the range of every two to four weeks, throughout the course of warfarin therapy for venous thromboembolism [2,98]. The INR should be monitored more frequently if there are factors that may produce an unpredictable response to warfarin (eg, concomitant drug therapy, other medical conditions, variable intake of vitamin K) [2,61,99]. (See ‘Excessive anticoagulation’ below and “Outpatient management of oral anticoagulation”, section on ‘Influence of diet’.)

The INR should also be monitored more frequently when substitution of one warfarin preparation for another has occurred, in order to screen for differences in drug availability [100-102].

Transitioning to or from dabigatran — Since dabigatran has been approved for use in patients with non-valvular atrial fibrillation in a number of countries, there may be a desire to change treatment from warfarin (or another vitamin K antagonist, VKA) to dabigatran, or vice versa. Recommendations on how to accomplish this are as follows (see “Anticoagulants other than heparin and warfarin”, section on ‘Dabigatran’):


  • From warfarin or other VKAs to dabigatran ─ Discontinue VKA and initiate dabigatran when the INR is <2.0.
  • From dabigatran to warfarin ─ Initiate warfarin three, two, or one day before discontinuation of dabigatran in patients with a creatinine clearance of >50, 31 to 50, or 15 to 30 mL/minute, respectively. This recommendation does not cover transition from dabigatran to the other VKAs.


Warfarin resistance — Maintenance doses of warfarin and other vitamin K antagonists vary significantly from patient to patient, depending upon a number of factors, such as the patient’s nutritional status, vitamin K intake, hepatic function, rate of intestinal absorption, genetic factors altering warfarin pharmacokinetics, degree of compliance, and the presence of drug interactions [94-96,103].

Hereditary warfarin resistance occurs in rats and has been rarely reported in humans. Patients with genetic warfarin resistance require doses that are fivefold to 20-fold higher than average to achieve an anticoagulant effect [2]. A number of hereditary factors may be associated with this resistance, including an inability to absorb this agent from the gastrointestinal tract, altered affinity of the warfarin receptor, unusually high CYP2C9 activity, and certain missense mutations or single nucleotide polymorphisms within the gene for vitamin K epoxide reductase complex subunit 1 (VKORC1) [104-107]. (See “Vitamin K and the synthesis of gamma carboxyglutamic acid”, section on ‘Warfarin resistance’.)

Outpatient management — A number of options are available for the outpatient management of anticoagulation with warfarin. These include supervision by hospital-based or community-based clinicians, anticoagulation clinics, and self-monitoring and self-management programs. This subject is discussed separately. (See “Outpatient management of oral anticoagulation”.)

Guidelines — Guidelines for the safe use and quality control of oral anticoagulation have been published and updated regularly by the British Committee for Standards in Hematology (available at: <www.bcshguidelines.com>) [1,108] and have been updated periodically by the American College of Chest Physicians, with the latest version appearing in June 2008 [2].

BLEEDING — Warfarin is among the top 10 drugs with the largest number of serious adverse event reports submitted to the United States FDA during the 1990 and 2000 decades [109]. Anticoagulants also ranked first in 2003 and 2004 in the number of total mentions of death for drugs “causing adverse effects in therapeutic use” and are a common cause of emergency department visits. Accordingly, the FDA has required that a “black box” warning about warfarin’s bleeding risk be added to US product labeling, that a Medication Guide be provided to the patient with each prescription [110], and that clinicians should counsel patients about preventive measures to minimize bleeding and to report immediately signs and symptoms of bleeding [109].

The risk of major bleeding episodes in patients treated with warfarin is related to the degree of anticoagulation as well as the presence in the patient of pre-existing risk factors for bleeding. An exception to this general rule may be retroperitoneal hemorrhage. This complication is common in patients taking anticoagulants, even when INR levels are within the therapeutic range [111]. (See “Approach to the adult patient with anemia”, section on ‘Blood loss’ and “Anticoagulation in older adults”, section on ‘Risk of bleeding’.)

Major risk factors — There are many patient characteristics that have been associated with an increased risk of bleeding following the use of vitamin K antagonists (VKAs). The ones listed below have been associated with a significantly increased risk of bleeding on one or more multivariate analyses [63,64,87,112-127]:


  • Increased age (variously given as >60, >65, >75, or >80 years)
  • Female sex
  • Diabetes mellitus
  • Presence of malignancy
  • Hypertension (ie, systolic >180 or diastolic >100 mmHg)
  • Acute or chronic alcoholism, liver disease
  • Severe chronic kidney disease; elevated creatinine
  • Anemia
  • Poor drug compliance or clinic attendance
  • Prior stroke or intracerebral hemorrhage
  • Presence of bleeding lesions (eg, gastrointestinal blood loss, peptic ulcer disease)
  • Bleeding disorder (coagulation defects, thrombocytopenia)
  • Concomitant use of aspirin, nonsteroidal antiinflammatory drugs (NSAIDs), antiplatelet agents (eg, clopidogrel), antibiotics, amiodarone, statins, fibrates
  • Instability of INR control and INR >3.0
  • Pre-treatment INR >1.2
  • Previous severe hemorrhage during treatment with warfarin with an INR in the therapeutic range


Patients with one or more of these risk factors for bleeding have generally been excluded from the very same pivotal trials that established the effectiveness of VKAs for their clinical condition (eg, atrial fibrillation, VTE, coronary heart disease). Despite this, many patients with an increased bleeding risk still receive treatment with VKAs. This was shown in a case-control study of 1986 hospitalized patients receiving VKAs, one-half of whom were admitted for bleeding and the other half for infection (control group). Findings of this study included [128]:


  • Subjects admitted with bleeding, when compared with the control group, had a significantly higher incidence of an INR ≥5.0 (30 versus 4.5 percent), creatinine >1.7 mg/dL (>150 microM, 5.9 versus 2.7 percent), and liver failure (2.4 versus 1.0 percent).
  • Twenty-three percent of the controls and 40 percent of the patients presenting with bleeding had one or more risk factors that would have excluded them from participation in these pivotal trials.
  • The relative risks for bleeding when one, two, or three or more of these exclusion criteria were present were 2.9, 3.8, and 14.9, respectively.


Use of bleeding risk scoring systems — A number of studies have provided retrospectively and/or prospectively derived indices for estimating the risk of bleeding following the use of oral anticoagulants [115,117,120,123,125,129,130], although no one index can reliably predict bleeding risk in a particular patient [131]. The best way to measure risk in an individual patient is unclear; one suggestion is to determine risk based on at least two different bleeding indices, such as those presented below.

An important caveat is that bleeding risk in a younger patient anticoagulated for an episode of deep vein thrombosis might result in a relatively benign episode of gastrointestinal bleeding following the concomitant use of aspirin or a nonsteroidal antiinflammatory agent, whereas anticoagulant-associated bleeding in an elderly patient with hypertension and atrial fibrillation might result in a fatal episode of cerebral hemorrhage.

Outpatient bleeding risk index — In an attempt to estimate the probability of major bleeding in outpatients treated with warfarin, an outpatient bleeding risk index was derived in a retrospective cohort of 556 patients, and then prospectively validated in a separate cohort of 264 outpatients [115]. The index included the following four adverse risk factors:


  • Age ≥65 years
  • History of stroke
  • History of gastrointestinal bleeding
  • One or more of the following: recent myocardial infarction, hematocrit <30 percent, serum creatinine >1.5 mg/dL (>133 micromol/L), diabetes mellitus


The cumulative incidence of major bleeding at 48 months in the low (no risk factors), intermediate (one to two), and high (three or more) risk groups was 3, 12, and 53 percent, respectively (table 5). Of the 18 episodes of major bleeding occurring in high-risk patients, 17 were potentially preventable (eg, avoidance of overanticoagulation and avoidance of NSAID use) [74].

HEMORR2HAGES risk index — Three earlier risk models, including the outpatient bleeding risk index noted above, were compared with a newer risk index in a validation cohort of 3932 elderly patients with atrial fibrillation (AF) receiving treatment with oral anticoagulants [123]. This study population was larger than any of the derivation or validation cohorts from the earlier risk models and offered the potential for more generalizable results. This index, called, HEMORR2HAGES, included a large number of risk factors drawn from the earlier models, including the following:

SEE MORE:  Diarrhea


  • Hepatic or renal disease
  • Ethanol abuse
  • Malignancy
  • Older age (>75 years)
  • Reduced platelet count or function, including aspirin therapy
  • Rebleeding risk (history of prior bleed)
  • Hypertension
  • Anemia
  • Genetic factors
  • Excessive fall risk
  • Stroke


Due to similar average relative risks of bleeding in prior studies, all risk factors were assigned one point, except for history of a prior bleed (rebleeding risk), which was assigned two points, for a total score ranging from zero to 12. The validation cohort was drawn from a multistate registry of AF patients. Among 1604 patients prescribed warfarin, the risks of a major bleed per 100 patient-years of warfarin therapy were as follows:


  • 0 points — 1.9
  • 1 point — 2.5
  • 2 points — 5.3
  • 3 points — 8.4
  • 4 points — 10.4
  • ≥5 points — 12.3


Two-thirds of the major bleeding events were gastrointestinal and 15 percent were intracranial. Although the HEMORR2HAGES model identified subgroups at elevated bleeding risk, the predictive value for individual patients was limited and similar to other models.

HAS-BLED bleeding risk score — The HAS-BLED bleeding risk score was designed to include risk factors either readily available from the clinical history or routinely tested in patients with atrial fibrillation [130,132,133] (table 6). In an unselected nationwide cohort of hospitalized patients with atrial fibrillation, the HAS-BLED score performed similarly to the HEMORR2HAGES risk score in predicting for bleeding risk, but appeared to be simpler and easier to use in everyday clinical practice [134].

The HAS-BLED risk score has been proposed for use in making clinical decisions about whether or not to use anticoagulation and/or antiplatelet therapy in various cardiac disorders [132], and has been recommended within the European Society of Cardiology and Canadian guidelines for assessing the risk of bleeding in atrial fibrillation management [135,136].

ATRIA risk score — A risk score for predicting warfarin-associated bleeding in patients with atrial fibrillation was developed from the results of the Anticoagulation and Risk Factors in Atrial Fibrillation (ATRIA) Study. Five independent variables were included in the final model, with weighting as follows [137]:


  • Anemia ─ 3 points
  • Severe renal disease (estimated glomerular filtration rate <30 mL/min or dialysis-dependent) ─ 3 points
  • Age ≥75 years ─ 2 points
  • Any prior hemorrhage ─ 1 point
  • Diagnosed hypertension ─ 1 point


Subjects were considered low risk (score zero to 3 points), intermediate risk (4 points), or high risk (5 to 10 points), with major bleeding rates of 0.76, 2.62, and 5.76 events/100 patient-years, respectively.

Thrombomodulin — High-risk patients may have bleeding episodes at INR values in the therapeutic or subtherapeutic range. In order to improve the ability to predict major bleeding, measurements in addition to the INR have been evaluated. As an example, in three reports, the risk of hemorrhage was associated with increased levels of thrombomodulin, an endothelium-derived antithrombotic cell-surface glycoprotein that is mainly present on the luminal surface of endothelial cells [138-140]. The anticoagulant properties of thrombomodulin result from its binding to thrombin and subsequent activation of the natural anticoagulant protein C.

Other causes for bleeding — Not all bleeding episodes in anticoagulated patients are due to the anticoagulation. In one report of 243 patients treated with warfarin who were prospectively followed for two years, the incidence of hematuria was similar to that of a control group not receiving warfarin [141]. Furthermore, evaluation of patients who developed hematuria revealed a genitourinary cause in 81 percent of cases. Infection was most common, but papillary necrosis, renal cysts, and several malignancies of the bladder were also found. (See “Etiology and evaluation of hematuria in adults”.)

Thus, hematuria in a patient taking chronic stable warfarin therapy requires evaluation rather than assuming it is related to the anticoagulation. A probable exception is that specific evaluation of the urinary tract may not be necessary in patients with hematuria and an excessively elevated INR who are also bleeding from other sites.

Intracranial bleeding — Discussions of the risk of intracranial bleeding following the use of warfarin in patients with advanced age, hypertension, and/or prior or potential intracranial bleeding lesions (eg, aneurysm, brain tumor, prior intracranial bleeding, infective endocarditis) are presented separately. (See “Anticoagulation in older adults” and “Anticoagulant and antiplatelet therapy in patients with an unruptured intracranial aneurysm” and “Anticoagulant and antiplatelet therapy in patients with brain tumors” and “Anticoagulant and antiplatelet therapy in patients with an acute or prior intracerebral hemorrhage” and “Anticoagulant and antiplatelet therapy in patients with infective endocarditis” and “Risk of intracerebral hemorrhage in patients treated with warfarin” and “Management of warfarin-associated intracerebral hemorrhage”.)

Excessive anticoagulation — Patients treated with warfarin frequently become excessively anticoagulated, even those who have been stable for many months. The most common causes are interactions between warfarin and other drugs and superimposed diseases (eg, liver disease, malabsorption) that may interfere with warfarin ingestion, absorption, or metabolism (see ‘Drug interactions’ above).

Studies in patients with atrial fibrillation indicate that the risk of bleeding increases substantially at INR values ≥5.0 (figure 3 and table 7) [142-144], although other studies have shown that this risk increases for patients given oral anticoagulation for a number of indications when the INR is >3.0 or >3.5 (figure 4). (See “Antithrombotic therapy to prevent embolization in nonvalvular atrial fibrillation”, section on ‘Bleeding risk’ and “Outpatient management of oral anticoagulation”, section on ‘Importance of strict INR control’.)

A more complete discussion of other causes underlying excessive anticoagulation as well as its treatment and prevention are presented separately. (See “Correcting excess anticoagulation after warfarin”.)

Activity limitations — Patients taking warfarin are at increased risk of spontaneous as well as trauma-induced bleeding. Reasonable precautions which can be taken concerning avoidance of falls, cuts, and other trauma are detailed elsewhere. (See “Patient information: Warfarin (Coumadin®)”, section on ‘Other recommendations’.)

Participation in sports — Guidelines are not available to assist patients taking warfarin in determining which sports activities are to be avoided and which are acceptable. Common sense must prevail, however, and individuals taking anticoagulants should not participate in contact or collision sports, or in activities in which there is an increased risk of serious trauma [145,146].

OTHER COMPLICATIONS — In addition to excess bleeding following the use of oral anticoagulants, there are a number of other concerns, including that of skin necrosis as well as special patient situations (factor IX mutation, pregnancy). These are discussed below.

Skin necrosis — Skin necrosis has been reported in some patients within the first few days of receiving large (loading) doses of warfarin [147]. The skin lesions may occur on the extremities, breasts, trunk, and penis and marginate over a period of hours from an initial central erythematous macule (picture 1 and picture 2 and picture 3). Biopsies demonstrate fibrin thrombi within cutaneous vessels with interstitial hemorrhage.

Skin necrosis appears to be mediated by the rapid reduction of protein C levels on the first day of therapy, which induces a transient hypercoagulable state. Approximately one-third of patients have underlying protein C deficiency; however, among patients with protein C deficiency, skin necrosis is an infrequent complication of warfarin therapy [148]. Case reports have also described skin necrosis or thrombosis at other sites in association with an acquired functional deficiency of protein C or S [149], heterozygous protein S deficiency, and factor V Leiden. (See “Protein C deficiency”, section on ‘Warfarin-induced skin necrosis’.)

Factor IX mutation — A rare mutation in factor IX, found in less than 1.5 percent of the population, causes bleeding following warfarin treatment without excessive prolongation of the prothrombin time. In these patients factor IX levels, which are not reflected in the prothrombin time (figure 2), are markedly reduced following warfarin therapy, while levels of the other vitamin K-dependent coagulation factors are reduced to about 30 to 40 percent. (See “Vitamin K and the synthesis of gamma carboxyglutamic acid”, section on ‘Recognition site mutations’.)

Pregnancy — Warfarin derivatives are generally contraindicated during at least the first trimester of pregnancy because of their teratogenic effects. This subject is discussed in detail separately, along with guidelines for anticoagulation during pregnancy using alternative agents (eg, low molecular weight heparin). (See “Anticoagulation during pregnancy”, section on ‘Warfarin’ and “Anticoagulation during pregnancy”, section on ‘General recommendations’.)

Cholesterol embolization — Embolization of cholesterol crystals (cholesterol microembolism) is a rare complication of anticoagulation with warfarin. Typically, this occurs after several weeks of therapy, and may present as a dark, purplish, mottled discoloration of the plantar and lateral surfaces of the lower extremities. This condition has been variously called “blue toe syndrome” or “purple toe syndrome” [150-152]. (See “Embolism from atherosclerotic plaque: Atheroembolism (cholesterol crystal embolism)”, section on ‘Skin’ and “Acute arterial occlusion of the lower extremities (acute limb ischemia)”, section on ‘Blue toe syndrome’.)

Vascular calcification — A number of studies have suggested that the chronic use of warfarin may lead to arterial calcification (eg, aortic valve, coronary arteries, femoral artery) [153-155]. This effect was evaluated in a study in which coronary artery calcium scans were performed in 157 patients with low-risk atrial fibrillation but without significant cardiovascular disease [156]. Mean coronary calcium scores increased significantly with the duration of vitamin K antagonist (VKA) use, and were 53, 90, and 236 in those not treated with VKA, treated with a VKA for 6 to 60 months, and treated with a VKA for >60 months, respectively.

This effect may be the result of inhibition of the vitamin K-dependent matrix Gla protein, important in inhibiting calcification. (See “Vitamin K, gamma carboxyglutamic acid, and the function of coagulation and other proteins”, section on ‘Gla-containing proteins of mineralized tissue’.)

Interference with thrombophilia testing — One cannot reliably test for congenital deficiencies of proteins S or C when patients are taking warfarin (table 8). (See “Evaluation of the patient with established venous thrombosis”, section on ‘Technical screening issues’.)

Allergic reactions — Cross-allergy between coumarin derivatives has been described, but details are sketchy:


  • Known hypersensitivity to coumarin derivatives, including warfarin, is listed as a contraindication on the UK-EU labeling of acenocoumarol.
  • In a case report, a single patient developed a maculopapular rash following treatment with three different coumarin derivatives [157].


The paucity of information concerning the safety of switching from one coumarin derivative to another because of an allergic reaction is such that the clinician will need to make clinical decisions on a case-by-case basis.

SURGERY — Depending upon the type of surgery planned, anticoagulation with warfarin may or may not have to be temporarily interrupted. This subject is discussed separately. (See “Management of anticoagulation before and after elective surgery”.)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, “The Basics” and “Beyond the Basics.” The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on “patient info” and the keyword(s) of interest.)


  • Beyond the Basics topics (see “Patient information: Warfarin (Coumadin®)” and “Patient information: Deep vein thrombosis (DVT)”)


SUMMARY AND RECOMMENDATIONS — Vitamin K antagonists (VKAs; eg, warfarin, acenocoumarol, phenprocoumon, fluindione) have a high incidence of adverse effects, mainly bleeding, a narrow therapeutic window, and a high propensity to be adversely affected by changes in diet and medications. Accordingly, all patients should be fully informed of the risks and benefits of VKA therapy before such treatment is initiated

Estimation of bleeding risk — The major factors increasing the risk of bleeding following the use of VKAs include increased age, presence of bleeding lesions, coagulation disorders, and the concomitant use of antiplatelet agents. Several bleeding risk indices are available for determination of such risks. (See ‘Bleeding’ above.)

Treatment recommendations

Starting dose — For initial therapy with warfarin, we recommend a starting oral dose of approximately 5 mg/day for most patients (Grade 2B).


  • Doses of ≤5 mg/day may be more appropriate in the elderly, debilitated, or malnourished patient, or those with congestive failure, liver disease, severe chronic kidney disease, recent major surgery, or are taking medication known to increase sensitivity to warfarin (eg, amiodarone). (See ‘Initial dose’ above and “Anticoagulation in older adults”.)
  • Initial doses >5 mg/day can be employed in selected patients deemed to be at low risk for bleeding and/or in those previously treated with warfarin at maintenance doses >5 mg/day. Loading doses >10 mg/day should be avoided. (See ‘Loading doses’ above.)
  • There is not sufficient information available at this time to support the use of pharmacogenetic dosing. (See ‘Genetic interactions’ above.)
  • If the patient is to be transitioned to warfarin from unfractionated heparin, low molecular weight heparin, or fondaparinux, the two agents should be given simultaneously for a minimum of five days. The heparin or fondaparinux can be discontinued on day 5 or 6 if the INR has been in the therapeutic range for at least 24 hours or two consecutive days. (See “Treatment of lower extremity deep vein thrombosis”, section on ‘Initial anticoagulation regimen’.)


Initial monitoring — We suggest that the patient have a baseline INR determination before initiation of treatment, receive two daily doses of warfarin, and have the INR checked again on the following day (Grade 2C). Subsequent doses are modified, up or down (table 4), in order to achieve a target INR of 2.5 (target range: 2.0 to 3.0) for most conditions requiring anticoagulation (figure 4).

Once the anticoagulant effect and patient’s warfarin dose requirements have been stabilized for at least one to two weeks, the INR can be monitored less frequently, at intervals in the range of every two to four weeks.

Long-term follow-up — Ideally, the patient’s anticoagulation status should be followed in a systematic and coordinated outpatient setting which covers such items as patient education, self-testing, diet, and the importance of strict INR control. This subject is covered separately. (See “Outpatient management of oral anticoagulation”.)

Correction of excess anticoagulation — This subject is covered separately. (See “Correcting excess anticoagulation after warfarin”.)


  1. Keeling D, Baglin T, Tait C, et al. Guidelines on oral anticoagulation with warfarin – fourth edition. Br J Haematol 2011; 154:311.
  2. Ansell J, Hirsh J, Hylek E, et al. Pharmacology and management of the vitamin K antagonists: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest 2008; 133:160S.
  3. Batke-Hastings S, Carman TL. Sublingual administration of warfarin: a novel form of delivery. Vasc Med 2008; 13:123.
  4. Fihn SD, Gadisseur AA, Pasterkamp E, et al. Comparison of control and stability of oral anticoagulant therapy using acenocoumarol versus phenprocoumon. Thromb Haemost 2003; 90:260.
  5. Mentré F, Pousset F, Comets E, et al. Population pharmacokinetic-pharmacodynamic analysis of fluindione in patients. Clin Pharmacol Ther 1998; 63:64.
  6. Ageno W, Crowther M, Steidl L, et al. Low dose oral vitamin K to reverse acenocoumarol-induced coagulopathy: a randomized controlled trial. Thromb Haemost 2002; 88:48.
  7. Beinema M, Brouwers JR, Schalekamp T, Wilffert B. Pharmacogenetic differences between warfarin, acenocoumarol and phenprocoumon. Thromb Haemost 2008; 100:1052.
  8. Ellis DJ, Usman MH, Milner PG, et al. The first evaluation of a novel vitamin K antagonist, tecarfarin (ATI-5923), in patients with atrial fibrillation. Circulation 2009; 120:1029.
  9. Freedman MD. Oral anticoagulants: pharmacodynamics, clinical indications and adverse effects. J Clin Pharmacol 1992; 32:196.
  10. Clouse LH, Comp PC. The regulation of hemostasis: the protein C system. N Engl J Med 1986; 314:1298.
  11. O’Reilly RA, Aggeler PM. Studies on coumarin anticoagulant drugs. Initiation of warfarin therapy without a loading dose. Circulation 1968; 38:169.
  12. Lind SE, Callas PW, Golden EA, et al. Plasma levels of factors II, VII and X and their relationship to the international normalized ratio during chronic warfarin therapy. Blood Coagul Fibrinolysis 1997; 8:48.
  13. Hull R, Delmore T, Genton E, et al. Warfarin sodium versus low-dose heparin in the long-term treatment of venous thrombosis. N Engl J Med 1979; 301:855.
  14. Hirsh J, Poller L. The international normalized ratio. A guide to understanding and correcting its problems. Arch Intern Med 1994; 154:282.
  15. Kucher N, Connolly S, Beckman JA, et al. International normalized ratio increase before warfarin-associated hemorrhage: brief and subtle. Arch Intern Med 2004; 164:2176.
  16. Shikata E, Ieiri I, Ishiguro S, et al. Association of pharmacokinetic (CYP2C9) and pharmacodynamic (factors II, VII, IX, and X; proteins S and C; and gamma-glutamyl carboxylase) gene variants with warfarin sensitivity. Blood 2004; 103:2630.
  17. D’Andrea G, D’Ambrosio RL, Di Perna P, et al. A polymorphism in the VKORC1 gene is associated with an interindividual variability in the dose-anticoagulant effect of warfarin. Blood 2005; 105:645.
  18. Sconce EA, Khan TI, Wynne HA, et al. The impact of CYP2C9 and VKORC1 genetic polymorphism and patient characteristics upon warfarin dose requirements: proposal for a new dosing regimen. Blood 2005; 106:2329.
  19. Schwarz UI, Ritchie MD, Bradford Y, et al. Genetic determinants of response to warfarin during initial anticoagulation. N Engl J Med 2008; 358:999.
  20. Takahashi H, Wilkinson GR, Nutescu EA, et al. Different contributions of polymorphisms in VKORC1 and CYP2C9 to intra- and inter-population differences in maintenance dose of warfarin in Japanese, Caucasians and African-Americans. Pharmacogenet Genomics 2006; 16:101.
  21. Cooper GM, Johnson JA, Langaee TY, et al. A genome-wide scan for common genetic variants with a large influence on warfarin maintenance dose. Blood 2008; 112:1022.
  22. Wadelius M, Chen LY, Lindh JD, et al. The largest prospective warfarin-treated cohort supports genetic forecasting. Blood 2009; 113:784.
  23. Aithal GP, Day CP, Kesteven PJ, Daly AK. Association of polymorphisms in the cytochrome P450 CYP2C9 with warfarin dose requirement and risk of bleeding complications. Lancet 1999; 353:717.
  24. Leung AY, Chow HC, Kwong YL, et al. Genetic polymorphism in exon 4 of cytochrome P450 CYP2C9 may be associated with warfarin sensitivity in Chinese patients. Blood 2001; 98:2584.
  25. Hermida J, Zarza J, Alberca I, et al. Differential effects of 2C9*3 and 2C9*2 variants of cytochrome P-450 CYP2C9 on sensitivity to acenocoumarol. Blood 2002; 99:4237.
  26. Bodin L, Verstuyft C, Tregouet DA, et al. Cytochrome P450 2C9 (CYP2C9) and vitamin K epoxide reductase (VKORC1) genotypes as determinants of acenocoumarol sensitivity. Blood 2005; 106:135.
  27. Taube J, Halsall D, Baglin T. Influence of cytochrome P-450 CYP2C9 polymorphisms on warfarin sensitivity and risk of over-anticoagulation in patients on long-term treatment. Blood 2000; 96:1816.
  28. Joffe HV, Xu R, Johnson FB, et al. Warfarin dosing and cytochrome P450 2C9 polymorphisms. Thromb Haemost 2004; 91:1123.
  29. Higashi MK, Veenstra DL, Kondo LM, et al. Association between CYP2C9 genetic variants and anticoagulation-related outcomes during warfarin therapy. JAMA 2002; 287:1690.
  30. Schalekamp T, van Geest-Daalderop JH, de Vries-Goldschmeding H, et al. Acenocoumarol stabilization is delayed in CYP2C93 carriers. Clin Pharmacol Ther 2004; 75:394.
  31. Meckley LM, Wittkowsky AK, Rieder MJ, et al. An analysis of the relative effects of VKORC1 and CYP2C9 variants on anticoagulation related outcomes in warfarin-treated patients. Thromb Haemost 2008; 100:229.
  32. Margaglione M, Colaizzo D, D’Andrea G, et al. Genetic modulation of oral anticoagulation with warfarin. Thromb Haemost 2000; 84:775.
  33. Rieder MJ, Reiner AP, Gage BF, et al. Effect of VKORC1 haplotypes on transcriptional regulation and warfarin dose. N Engl J Med 2005; 352:2285.
  34. Montes R, Ruiz de Gaona E, Martínez-González MA, et al. The c.-1639G > A polymorphism of the VKORC1 gene is a major determinant of the response to acenocoumarol in anticoagulated patients. Br J Haematol 2006; 133:183.
  35. Loebstein R, Dvoskin I, Halkin H, et al. A coding VKORC1 Asp36Tyr polymorphism predisposes to warfarin resistance. Blood 2007; 109:2477.
  36. Wang D, Chen H, Momary KM, et al. Regulatory polymorphism in vitamin K epoxide reductase complex subunit 1 (VKORC1) affects gene expression and warfarin dose requirement. Blood 2008; 112:1013.
  37. Limdi NA, Wadelius M, Cavallari L, et al. Warfarin pharmacogenetics: a single VKORC1 polymorphism is predictive of dose across 3 racial groups. Blood 2010; 115:3827.
  38. Herman D, Peternel P, Stegnar M, et al. The influence of sequence variations in factor VII, gamma-glutamyl carboxylase and vitamin K epoxide reductase complex genes on warfarin dose requirement. Thromb Haemost 2006; 95:782.
  39. Caldwell MD, Awad T, Johnson JA, et al. CYP4F2 genetic variant alters required warfarin dose. Blood 2008; 111:4106.
  40. Pérez-Andreu V, Roldán V, Antón AI, et al. Pharmacogenetic relevance of CYP4F2 V433M polymorphism on acenocoumarol therapy. Blood 2009; 113:4977.
  41. Tham LS, Goh BC, Nafziger A, et al. A warfarin-dosing model in Asians that uses single-nucleotide polymorphisms in vitamin K epoxide reductase complex and cytochrome P450 2C9. Clin Pharmacol Ther 2006; 80:346.
  42. Wu AH. Use of genetic and nongenetic factors in warfarin dosing algorithms. Pharmacogenomics 2007; 8:851.
  43. Zhu Y, Shennan M, Reynolds KK, et al. Estimation of warfarin maintenance dose based on VKORC1 (-1639 G>A) and CYP2C9 genotypes. Clin Chem 2007; 53:1199.
  44. Carlquist JF, Horne BD, Muhlestein JB, et al. Genotypes of the cytochrome p450 isoform, CYP2C9, and the vitamin K epoxide reductase complex subunit 1 conjointly determine stable warfarin dose: a prospective study. J Thromb Thrombolysis 2006; 22:191.
  45. Gage BF, Eby C, Milligan PE, et al. Use of pharmacogenetics and clinical factors to predict the maintenance dose of warfarin. Thromb Haemost 2004; 91:87.
  46. Millican EA, Lenzini PA, Milligan PE, et al. Genetic-based dosing in orthopedic patients beginning warfarin therapy. Blood 2007; 110:1511.
  47. Anderson JL, Horne BD, Stevens SM, et al. Randomized trial of genotype-guided versus standard warfarin dosing in patients initiating oral anticoagulation. Circulation 2007; 116:2563.
  48. Poller L, Keown M, Ibrahim S, et al. An international multicenter randomized study of computer-assisted oral anticoagulant dosage vs. medical staff dosage. J Thromb Haemost 2008; 6:935.
  49. International Warfarin Pharmacogenetics Consortium, Klein TE, Altman RB, et al. Estimation of the warfarin dose with clinical and pharmacogenetic data. N Engl J Med 2009; 360:753.
  50. Jowett S, Bryan S, Poller L, et al. The cost-effectiveness of computer-assisted anticoagulant dosage: results from the European Action on Anticoagulation (EAA) multicentre study. J Thromb Haemost 2009; 7:1482.
  51. Gong IY, Tirona RG, Schwarz UI, et al. Prospective evaluation of a pharmacogenetics-guided warfarin loading and maintenance dose regimen for initiation of therapy. Blood 2011; 118:3163.
  52. van Schie RM, Wessels JA, le Cessie S, et al. Loading and maintenance dose algorithms for phenprocoumon and acenocoumarol using patient characteristics and pharmacogenetic data. Eur Heart J 2011; 32:1909.
  53. Moyer TP, O’Kane DJ, Baudhuin LM, et al. Warfarin sensitivity genotyping: a review of the literature and summary of patient experience. Mayo Clin Proc 2009; 84:1079.
  54. Eckman MH, Rosand J, Greenberg SM, Gage BF. Cost-effectiveness of using pharmacogenetic information in warfarin dosing for patients with nonvalvular atrial fibrillation. Ann Intern Med 2009; 150:73.
  55. Rosove MH, Grody WW. Should we be applying warfarin pharmacogenetics to clinical practice? No, not now. Ann Intern Med 2009; 151:270.
  56. Gage BF, Eby C, Johnson JA, et al. Use of pharmacogenetic and clinical factors to predict the therapeutic dose of warfarin. Clin Pharmacol Ther 2008; 84:326.
  57. Montes R, Nantes O, Alonso A, et al. The influence of polymorphisms of VKORC1 and CYP2C9 on major gastrointestinal bleeding risk in anticoagulated patients. Br J Haematol 2008; 143:727.
  58. Epstein RS, Moyer TP, Aubert RE, et al. Warfarin genotyping reduces hospitalization rates results from the MM-WES (Medco-Mayo Warfarin Effectiveness study). J Am Coll Cardiol 2010; 55:2804.
  59. Finkelman BS, Gage BF, Johnson JA, et al. Genetic warfarin dosing: tables versus algorithms. J Am Coll Cardiol 2011; 57:612.
  60. www.accessdata.fda.gov/cdrh_docs/pdf7/K070804.pdf (Accessed on July 18, 2011).
  61. Holbrook AM, Pereira JA, Labiris R, et al. Systematic overview of warfarin and its drug and food interactions. Arch Intern Med 2005; 165:1095.
  62. Juurlink DN. Drug interactions with warfarin: what clinicians need to know. CMAJ 2007; 177:369.
  63. Delaney JA, Opatrny L, Brophy JM, Suissa S. Drug drug interactions between antithrombotic medications and the risk of gastrointestinal bleeding. CMAJ 2007; 177:347.
  64. Schelleman H, Bilker WB, Brensinger CM, et al. Fibrate/Statin initiation in warfarin users and gastrointestinal bleeding risk. Am J Med 2010; 123:151.
  65. Thijssen HH, Soute BA, Vervoort LM, Claessens JG. Paracetamol (acetaminophen) warfarin interaction: NAPQI, the toxic metabolite of paracetamol, is an inhibitor of enzymes in the vitamin K cycle. Thromb Haemost 2004; 92:797.
  66. Kwan D, Bartle WR, Walker SE. The effects of acetaminophen on pharmacokinetics and pharmacodynamics of warfarin. J Clin Pharmacol 1999; 39:68.
  67. Mahé I, Bertrand N, Drouet L, et al. Paracetamol: a haemorrhagic risk factor in patients on warfarin. Br J Clin Pharmacol 2005; 59:371.
  68. Feldstein AC, Smith DH, Perrin N, et al. Reducing warfarin medication interactions: an interrupted time series evaluation. Arch Intern Med 2006; 166:1009.
  69. Alexandra JF, Pautas E, Gouin-Thibault I, et al. Overanticoagulation with coumarin and cutaneous azole therapy. Ann Intern Med 2008; 148:633.
  70. Visser LE, Penning-van Bees FJ, Kasbergen AA, et al. Overanticoagulation associated with combined use of antibacterial drugs and acenocoumarol or phenprocoumon anticoagulants. Thromb Haemost 2002; 88:705.
  71. Fischer HD, Juurlink DN, Mamdani MM, et al. Hemorrhage during warfarin therapy associated with cotrimoxazole and other urinary tract anti-infective agents: a population-based study. Arch Intern Med 2010; 170:617.
  72. Teichert M, van Noord C, Uitterlinden AG, et al. Proton pump inhibitors and the risk of overanticoagulation during acenocoumarol maintenance treatment. Br J Haematol 2011; 153:379.
  73. van Dijk KN, Plat AW, van Dijk AA, et al. Potential interaction between acenocoumarol and diclofenac, naproxen and ibuprofen and role of CYP2C9 genotype. Thromb Haemost 2004; 91:95.
  74. Knijff-Dutmer EA, Van der Palen J, Schut G, Van de Laar MA. The influence of cyclo-oxygenase specificity of non-steroidal anti-inflammatory drugs on bleeding complications in concomitant coumarine users. QJM 2003; 96:513.
  75. Johnson SG, Witt DM, Eddy TR, Delate T. Warfarin and antiplatelet combination use among commercially insured patients enrolled in an anticoagulation management service. Chest 2007; 131:1500.
  76. Johnson SG, Rogers K, Delate T, Witt DM. Outcomes associated with combined antiplatelet and anticoagulant therapy. Chest 2008; 133:948.
  77. García Rodríguez LA, Lin KJ, Hernández-Díaz S, Johansson S. Risk of upper gastrointestinal bleeding with low-dose acetylsalicylic acid alone and in combination with clopidogrel and other medications. Circulation 2011; 123:1108.
  78. Nathisuwan S, Dilokthornsakul P, Chaiyakunapruk N, et al. Assessing evidence of interaction between smoking and warfarin: a systematic review and meta-analysis. Chest 2011; 139:1130.
  79. Harrison L, Johnston M, Massicotte MP, et al. Comparison of 5-mg and 10-mg loading doses in initiation of warfarin therapy. Ann Intern Med 1997; 126:133.
  80. Crowther MA, Ginsberg JB, Kearon C, et al. A randomized trial comparing 5-mg and 10-mg warfarin loading doses. Arch Intern Med 1999; 159:46.
  81. Farahmand S, Saeedi M, Seyed Javadi HH, Khashayar P. High doses of warfarin are more beneficial than its low doses in patients with deep vein thrombosis. Am J Emerg Med 2010.
  82. Kovacs MJ, Rodger M, Anderson DR, et al. Comparison of 10-mg and 5-mg warfarin initiation nomograms together with low-molecular-weight heparin for outpatient treatment of acute venous thromboembolism. A randomized, double-blind, controlled trial. Ann Intern Med 2003; 138:714.
  83. Cushman M, Booth SL, Possidente CJ, et al. The association of vitamin K status with warfarin sensitivity at the onset of treatment. Br J Haematol 2001; 112:572.
  84. Lubetsky A, Dekel-Stern E, Chetrit A, et al. Vitamin K intake and sensitivity to warfarin in patients consuming regular diets. Thromb Haemost 1999; 81:396.
  85. Khan T, Wynne H, Wood P, et al. Dietary vitamin K influences intra-individual variability in anticoagulant response to warfarin. Br J Haematol 2004; 124:348.
  86. Franco V, Polanczyk CA, Clausell N, Rohde LE. Role of dietary vitamin K intake in chronic oral anticoagulation: prospective evidence from observational and randomized protocols. Am J Med 2004; 116:651.
  87. Limdi NA, Beasley TM, Baird MF, et al. Kidney function influences warfarin responsiveness and hemorrhagic complications. J Am Soc Nephrol 2009; 20:912.
  88. Gage BF, Fihn SD, White RH. Management and dosing of warfarin therapy. Am J Med 2000; 109:481.
  89. www.fda.gov/oc/initiatives/criticalpath/warfarin.html (Accessed on July 18, 2011).
  90. Warfarin label change falls short of recommending genetic testing. “The Pink Sheet”, F-D-C Reports 2007; 69:8. August 20, 2007.
  91. Barnes GD, Froehlich JB. Anticoagulation: a pathway to clinical effectiveness. Am J Med 2009; 122:126.
  92. Kim YK, Nieuwlaat R, Connolly SJ, et al. Effect of a simple two-step warfarin dosing algorithm on anticoagulant control as measured by time in therapeutic range: a pilot study. J Thromb Haemost 2010; 8:101.
  93. Le Gal G, Carrier M, Tierney S, et al. Prediction of the warfarin maintenance dose after completion of the 10 mg initiation nomogram: do we really need genotyping? J Thromb Haemost 2010; 8:90.
  94. Lefrère JJ, Guyon F, Horellou MH, et al. [Resistance to vitamin K antagonists. 6 cases]. Ann Med Interne (Paris) 1986; 137:384.
  95. Hallak HO, Wedlund PJ, Modi MW, et al. High clearance of (S)-warfarin in a warfarin-resistant subject. Br J Clin Pharmacol 1993; 35:327.
  96. Hulse ML. Warfarin resistance: diagnosis and therapeutic alternatives. Pharmacotherapy 1996; 16:1009.
  97. Limdi NA, Limdi MA, Cavallari L, et al. Warfarin dosing in patients with impaired kidney function. Am J Kidney Dis 2010; 56:823.
  98. Rose AJ, Ozonoff A, Berlowitz DR, et al. Reexamining the recommended follow-up interval after obtaining an in-range international normalized ratio value: results from the Veterans Affairs study to improve anticoagulation. Chest 2011; 140:359.
  99. Sconce E, Avery P, Wynne H, Kamali F. Vitamin K supplementation can improve stability of anticoagulation for patients with unexplained variability in response to warfarin. Blood 2007; 109:2419.
  100. Halkin H, Shapiro J, Kurnik D, et al. Increased warfarin doses and decreased international normalized ratio response after nationwide generic switching. Clin Pharmacol Ther 2003; 74:215.
  101. Milligan PE, Banet GA, Waterman AD, et al. Substitution of generic warfarin for Coumadin in an HMO setting. Ann Pharmacother 2002; 36:764.
  102. Witt DM, Tillman DJ, Evans CM, et al. Evaluation of the clinical and economic impact of a brand name-to-generic warfarin sodium conversion program. Pharmacotherapy 2003; 23:360.
  103. Pérez-Andreu V, Roldán V, López-Fernández MF, et al. Pharmacogenetics of acenocoumarol in patients with extreme dose requirements. J Thromb Haemost 2010; 8:1012.
  104. O’Reilly RA, Pool JG, Aggeler PM. Hereditary resistance to coumarin anticoagulant drugs in man and rat. Ann N Y Acad Sci 1968; 151:913.
  105. Alving BM, Strickler MP, Knight RD, et al. Hereditary warfarin resistance. Investigation of a rare phenomenon. Arch Intern Med 1985; 145:499.
  106. Bodin L, Perdu J, Diry M, et al. Multiple genetic alterations in vitamin K epoxide reductase complex subunit 1 gene (VKORC1) can explain the high dose requirement during oral anticoagulation in humans. J Thromb Haemost 2008; 6:1436.
  107. Schmeits PC, Hermans MH, van Geest-Daalderop JH, et al. VKORC1 mutations in patients with partial resistance to phenprocoumon. Br J Haematol 2010; 148:955.
  108. Baglin TP, Cousins D, Keeling DM, et al. Safety indicators for inpatient and outpatient oral anticoagulant care: [corrected] Recommendations from the British Committee for Standards in Haematology and National Patient Safety Agency. Br J Haematol 2007; 136:26.
  109. Wysowski DK, Nourjah P, Swartz L. Bleeding complications with warfarin use: a prevalent adverse effect resulting in regulatory action. Arch Intern Med 2007; 167:1414.
  110. Jaffer A, Bragg L. Practical tips for warfarin dosing and monitoring. Cleve Clin J Med 2003; 70:361.
  111. González C, Penado S, Llata L, et al. The clinical spectrum of retroperitoneal hematoma in anticoagulated patients. Medicine (Baltimore) 2003; 82:257.
  112. Fitzmaurice DA, Blann AD, Lip GY. Bleeding risks of antithrombotic therapy. BMJ 2002; 325:828.
  113. Veeger NJ, Piersma-Wichers M, Tijssen JG, et al. Individual time within target range in patients treated with vitamin K antagonists: main determinant of quality of anticoagulation and predictor of clinical outcome. A retrospective study of 2300 consecutive patients with venous thromboembolism. Br J Haematol 2005; 128:513.
  114. Odén A, Fahlén M. Oral anticoagulation and risk of death: a medical record linkage study. BMJ 2002; 325:1073.
  115. Beyth RJ, Quinn LM, Landefeld CS. Prospective evaluation of an index for predicting the risk of major bleeding in outpatients treated with warfarin. Am J Med 1998; 105:91.
  116. Adjusted-dose warfarin versus low-intensity, fixed-dose warfarin plus aspirin for high-risk patients with atrial fibrillation: Stroke Prevention in Atrial Fibrillation III randomised clinical trial. Lancet 1996; 348:633.
  117. Kuijer PM, Hutten BA, Prins MH, Büller HR. Prediction of the risk of bleeding during anticoagulant treatment for venous thromboembolism. Arch Intern Med 1999; 159:457.
  118. Hylek EM, Singer DE. Risk factors for intracranial hemorrhage in outpatients taking warfarin. Ann Intern Med 1994; 120:897.
  119. van Walraven C, Hart RG, Singer DE, et al. Oral anticoagulants vs aspirin in nonvalvular atrial fibrillation: an individual patient meta-analysis. JAMA 2002; 288:2441.
  120. Wells PS, Forgie MA, Simms M, et al. The outpatient bleeding risk index: validation of a tool for predicting bleeding rates in patients treated for deep venous thrombosis and pulmonary embolism. Arch Intern Med 2003; 163:917.
  121. Shireman TI, Howard PA, Kresowik TF, Ellerbeck EF. Combined anticoagulant-antiplatelet use and major bleeding events in elderly atrial fibrillation patients. Stroke 2004; 35:2362.
  122. Douketis JD, Arneklev K, Goldhaber SZ, et al. Comparison of bleeding in patients with nonvalvular atrial fibrillation treated with ximelagatran or warfarin: assessment of incidence, case-fatality rate, time course and sites of bleeding, and risk factors for bleeding. Arch Intern Med 2006; 166:853.
  123. Gage BF, Yan Y, Milligan PE, et al. Clinical classification schemes for predicting hemorrhage: results from the National Registry of Atrial Fibrillation (NRAF). Am Heart J 2006; 151:713.
  124. Warfarin Antiplatelet Vascular Evaluation Trial Investigators, Anand S, Yusuf S, et al. Oral anticoagulant and antiplatelet therapy and peripheral arterial disease. N Engl J Med 2007; 357:217.
  125. Ruíz-Giménez N, Suárez C, González R, et al. Predictive variables for major bleeding events in patients presenting with documented acute venous thromboembolism. Findings from the RIETE Registry. Thromb Haemost 2008; 100:26.
  126. Nieto JA, Solano R, Ruiz-Ribó MD, et al. Fatal bleeding in patients receiving anticoagulant therapy for venous thromboembolism: findings from the RIETE registry. J Thromb Haemost 2010; 8:1216.
  127. Zhao HJ, Zheng ZT, Wang ZH, et al. “Triple therapy” rather than “triple threat”: a meta-analysis of the two antithrombotic regimens after stent implantation in patients receiving long-term oral anticoagulant treatment. Chest 2011; 139:260.
  128. Levi M, Hovingh GK, Cannegieter SC, et al. Bleeding in patients receiving vitamin K antagonists who would have been excluded from trials on which the indication for anticoagulation was based. Blood 2008; 111:4471.
  129. Shireman TI, Mahnken JD, Howard PA, et al. Development of a contemporary bleeding risk model for elderly warfarin recipients. Chest 2006; 130:1390.
  130. Lip GY, Frison L, Halperin JL, Lane DA. Comparative validation of a novel risk score for predicting bleeding risk in anticoagulated patients with atrial fibrillation: the HAS-BLED (Hypertension, Abnormal Renal/Liver Function, Stroke, Bleeding History or Predisposition, Labile INR, Elderly, Drugs/Alcohol Concomitantly) score. J Am Coll Cardiol 2011; 57:173.
  131. Dahri K, Loewen P. The risk of bleeding with warfarin: a systematic review and performance analysis of clinical prediction rules. Thromb Haemost 2007; 98:980.
  132. Pisters R, Lane DA, Nieuwlaat R, et al. A novel user-friendly score (HAS-BLED) to assess 1-year risk of major bleeding in patients with atrial fibrillation: the Euro Heart Survey. Chest 2010; 138:1093.
  133. Lip GY. Implications of the CHA(2)DS(2)-VASc and HAS-BLED Scores for thromboprophylaxis in atrial fibrillation. Am J Med 2011; 124:111.
  134. Olesen JB, Lip GY, Hansen PR, et al. Bleeding risk in ‘real world’ patients with atrial fibrillation: comparison of two established bleeding prediction schemes in a nationwide cohort. J Thromb Haemost 2011; 9:1460.
  135. European Heart Rhythm Association, European Association for Cardio-Thoracic Surgery, Camm AJ, et al. Guidelines for the management of atrial fibrillation: the Task Force for the Management of Atrial Fibrillation of the European Society of Cardiology (ESC). Eur Heart J 2010; 31:2369.
  136. Cairns JA, Connolly S, McMurtry S, et al. Canadian Cardiovascular Society atrial fibrillation guidelines 2010: prevention of stroke and systemic thromboembolism in atrial fibrillation and flutter. Can J Cardiol 2011; 27:74.
  137. Fang MC, Go AS, Chang Y, et al. A new risk scheme to predict warfarin-associated hemorrhage: The ATRIA (Anticoagulation and Risk Factors in Atrial Fibrillation) Study. J Am Coll Cardiol 2011; 58:395.
  138. Jansson JH, Boman K, Brännström M, Nilsson TK. High concentration of thrombomodulin in plasma is associated with hemorrhage: a prospective study in patients receiving long-term anticoagulant treatment. Circulation 1997; 96:2938.
  139. van der Heijden JF, Rekké B, Hutten BA, et al. Non-fatal major bleeding during treatment with vitamin K antagonists: influence of soluble thrombomodulin and mutations in the propeptide of coagulation factor IX. J Thromb Haemost 2004; 2:1104.
  140. Lind M, Boman K, Johansson L, et al. Thrombomodulin as a marker for bleeding complications during warfarin treatment. Arch Intern Med 2009; 169:1210.
  141. Culclasure TF, Bray VJ, Hasbargen JA. The significance of hematuria in the anticoagulated patient. Arch Intern Med 1994; 154:649.
  142. Optimal oral anticoagulant therapy in patients with nonrheumatic atrial fibrillation and recent cerebral ischemia. The European Atrial Fibrillation Trial Study Group. N Engl J Med 1995; 333:5.
  143. Torn M, van der Meer FJ, Rosendaal FR. Lowering the intensity of oral anticoagulant therapy: effects on the risk of hemorrhage and thromboembolism. Arch Intern Med 2004; 164:668.
  144. Palareti G, Leali N, Coccheri S, et al. Bleeding complications of oral anticoagulant treatment: an inception-cohort, prospective collaborative study (ISCOAT). Italian Study on Complications of Oral Anticoagulant Therapy. Lancet 1996; 348:423.
  145. Meyering C, Howard T. Hypercoagulability in athletes. Curr Sports Med Rep 2004; 3:77.
  146. Mellwig KP, van Buuren F, Gohlke-Baerwolf C, Bjørnstad HH. Recommendations for the management of individuals with acquired valvular heart diseases who are involved in leisure-time physical activities or competitive sports. Eur J Cardiovasc Prev Rehabil 2008; 15:95.
  147. Bauer KA. Coumarin-induced skin necrosis. Arch Dermatol 1993; 129:766.
  148. Teepe RG, Broekmans AW, Vermeer BJ, et al. Recurrent coumarin-induced skin necrosis in a patient with an acquired functional protein C deficiency. Arch Dermatol 1986; 122:1408.
  149. Haran MZ, Lichman I, Berebbi A, et al. Unbalanced protein S deficiency due to warfarin treatment as a possible cause for thrombosis. Br J Haematol 2007; 139:310.
  150. Bols A, Nevelsteen A, Verhaeghe R. Atheromatous embolization precipitated by oral anticoagulants. Int Angiol 1994; 13:271.
  151. Rauh G, Spengel FA. Blue toe syndrome after initiation of low-dose oral anticoagulation. Eur J Med Res 1998; 3:278.
  152. Hyman BT, Landas SK, Ashman RF, et al. Warfarin-related purple toes syndrome and cholesterol microembolization. Am J Med 1987; 82:1233.
  153. Koos R, Mahnken AH, Mühlenbruch G, et al. Relation of oral anticoagulation to cardiac valvular and coronary calcium assessed by multislice spiral computed tomography. Am J Cardiol 2005; 96:747.
  154. Koos R, Krueger T, Westenfeld R, et al. Relation of circulating Matrix Gla-Protein and anticoagulation status in patients with aortic valve calcification. Thromb Haemost 2009; 101:706.
  155. Rennenberg RJ, van Varik BJ, Schurgers LJ, et al. Chronic coumarin treatment is associated with increased extracoronary arterial calcification in humans. Blood 2010; 115:5121.
  156. Weijs B, Blaauw Y, Rennenberg RJ, et al. Patients using vitamin K antagonists show increased levels of coronary calcification: an observational study in low-risk atrial fibrillation patients. Eur Heart J 2011; 32:2555.
  157. Kruis-de Vries MH, Stricker BH, Coenraads PJ, Nater JP. Maculopapular rash due to coumarin derivatives. Dermatologica 1989; 178:109.
SEE MORE:  Anticoagulation during pregnancy