Good anticoagulant practice

Learn about starting anticoagulants and how to conduct a systematic assessment of the potential risk of thromboembolism.


Since publication of this Medicinewise News, apixaban has been approved by the TGA, and rivaroxaban, dabigatran and apixaban have all been PBS-listed for the prevention of stroke and systemic embolism in patients with non-valvular atrial fibrillation and at least one additional risk factor for stroke.


Key points

  • Stroke risk in people with non-valvular atrial fibrillation (AF) is under-treated even when CHADS2 scores indicate that anticoagulation is required.
  • People with non-valvular AF well controlled on warfarin may not benefit clinically from switching to a newer oral anticoagulant.
  • At present, there are no readily available and validated methods for monitoring or reversing the activity of the newer oral anticoagulants.
  • Coagulation monitoring of patients taking newer oral anticoagulants is not currently possible but routine clinical monitoring is essential.

The availability of newer oral anticoagulants, dabigatran, rivaroxaban and apixaban, is an opportunity to review and individualise treatment and prevention of thromboembolism and to improve current warfarin management.


Starting treatment

  • Start anticoagulants based on a systematic assessment of a person’s potential risk of thromboembolism using the CHADS2 score (Tables 1 and 2).
  • Identify risk factors for bleeding as well as patients who need increased clinical monitoring (those with three or more risk factors) using the HAS-BLED tool (Table 3).
  • People at high risk of bleeding are not precluded from anticoagulation but regular clinical review is recommended.1
  • Treat correctable risk factors for bleeding, and consider a Home Medicines Review2 for people prescribed anticoagulant therapy. Twice-yearly reviews may benefit patients with three or more risk factors for bleeding.3
  • Choose an anticoagulant in consultation with the patient; discuss the potential benefits and harms of treatment options.
  • Consider warfarin as first-line anticoagulant treatment for non-valvular AF.4 Aspirin is no longer recommended for people with non-valvular AF who are at moderate risk of stroke unless anticoagulation is contraindicated.4

Table 1: CHADS2: stroke risk assessment in people with non-valvular AF5
Risk factor Score
Congestive heart failure history 1
Hypertension (including well-controlled hypertension) 1
Diabetes mellitus

Stroke or transient ischaemic attack history


Table 2: Stroke risk based on CHADS2 score4,5
Total CHADS2 score Risk of stroke Adjusted stroke ratea
0 Low 1.9%
1 Moderate 2.8%

a Per 100 patient-years without treatment.


Treat non-valvular AF based on stroke risk

Assess people with non-valvular AF, who are not already taking anticoagulants, for risk of stroke at least annually, using a tool such as the CHADS2 scoring system.6 

People with non-valvular AF and one additional risk factor for stroke are considered to be at moderate risk, and people with non-valvular AF and two risk factors are considered to be at high risk of stroke.4 Therapeutic Guidelines recommends anticoagulation for these people (CHADS2 ≥ 1).4 

When the CHADS2 score is zero, consider using the CHA2DS2-VASc score (Table 4) to determine stroke risk and the need for anticoagulation more precisely. Some international guidelines recommend anticoagulation for people with non-valvular AF and a CHA2DS2-VASc score ≥ 1 for males and ≥ 2 for females.1

Table 3: HAS-BLED: a Tool to identify risk factors for bleeding in people with AF7

Clinical characteristic
H Hypertension (systolic blood pressure > 160 mmHg)
AAbnormal renal or liver function
SStroke (history of)
BBleeding (history of, or predisposition to bleeding)
LLabile INRs (< 6 in 10 INRs in therapeutic range)
EElderly (eg, age > 65 years)
DDrugs (antiplatelet agents, NSAIDs, or alcohol ≥ 8 units per week)
Table 4: CHA2DS2VASc: A tool to further risk stratify low-risk people with non-valvular AF1
Risk factor Score

Congestive heart failure or left ventricular dysfunction

Hypertension 1
Age 75 years or older
Diabetes mellitus
Stroke or transient ischaemic attack history
Vascular disease (prior myocardial infarction, peripheral artery disease, aortic plaque)
Age 65–74
Sex category (female gender)

A recent cohort study of over 132,000 people with non-valvular AF showed that for people with a CHADS2 score ≥ 1 the benefits of warfarin anticoagulation for reducing risk of stroke outweighed the increased risk of intracranial haemorrhage (ICH) regardless of HAS-BLED score and that warfarin was of no benefit for people truly at low risk of stroke (CHA2DS2-VASc score = 0).8 

A meta-analysis showed that aspirin does not significantly reduce incidence of stroke in people with non-valvular AF compared with placebo or no treatment.9 In this study, warfarin reduced the incidence of stroke in people with non-valvular AF by 37% compared with antiplatelet therapy (95% confidence interval (CI) 23% to 48%) and is associated with only a small increase in absolute risk of ICH; 0.2% per year.9


Age and risk of falls: current barriers but not contraindications to warfarin

A 2008 survey of around 600 Australian GPs reported that around 30% are often unsure whether to prescribe warfarin and 38% found it hard to decide if the benefits of warfarin outweigh the risks.10 

The decision to prescribe warfarin is influenced both by true contraindications and by factors for which the benefits of anticoagulation outweigh the bleeding risk.10

Risk of falls

Having a risk of falls is not a direct contraindication to anticoagulant therapy although great care should be taken when treating such people.6 

High risk of falls increases the risk of ICH (hazard ratio [HR] 1.9, 95% CI 1.3 to 2.9)11 and has a significant influence on whether anticoagulants are prescribed.10,12-15 

A recent cohort study showed that for people with AF, risk of stroke is significantly higher in people at high risk of fallsb than in others (stroke rate per 100 patient–years 13.7 [95% CI 11.6 to 16.3] vs 6.9 [95% CI 6.5 to 7.3], respectively).11 

In this study warfarin demonstrated significant benefit for people at high risk of falls and high risk of stroke (CHADS2 score ≥ 2) despite the increased risk of ICH.11 However, warfarin should not be prescribed for people after ICH and risk of falls should be reduced when possible.11


People aged over 70 years are less likely to be prescribed warfarin than people aged under 70 years.13 

A recent meta-analysis of patient data showed risk increases for both serious haemorrhage (HR 1.61, 95% CI 1.47 to 1.77) and ischaemic stroke (HR 1.45, 95% CI 1.26 to 1.66) with each increasing decade of age in people with non-valvular AF.16 

The study shows that older people with non-valvular AF benefit from warfarin therapy and that the relative benefits of oral anticoagulation for preventing ischaemic stroke do not change significantly with age.16 

Antiplatelet therapy significantly reduces the risk of ischaemic stroke in younger people (HR 0.40, 95% CI 0.22 to 0.72 at age 50 years); however, the benefits decrease with age, and antiplatelet therapy is no longer of benefit in people > 77 years.16

b Defined as people with notes of 'frequent falls', 'history of falls', 'multiple falls' or 'tendency for falls' in their medical records. Notes of 'a single fall' did not constitute high risk.


Warfarin contraindications

Warfarin is contraindicated in people with severe active bleeding or with conditions associated with increased risk of severe bleeding6 (eg, severe uncontrolled hypertension,6 recent gastrointestinal [GI] bleeding, genitourinary bleeding, active ulceration or severe thrombocytopenia6) as well as people with compliance problems (includes lack of access to INR monitoring as well as non-adherence).

Those at high risk of bleeding are also at high risk of stroke

HAS-BLED (Table 3) is a simple and predictive tool17 for assessing the risk of bleeding in patients with AF and is recommended in some international guidelines.18,19 Each clinical characteristic in the HAS-BLED table increases a person’s bleeding risk.7 

HAS-BLED was designed to systematically assess annual risk of a major bleed in people with AF before they started warfarin.7 Many of the risk factors in the tool are also relevant when considering the newer oral anticoagulants.7


The newer oral anticoagulants

The place in therapy of the newer oral anticoagulants is currently uncertain. In Australia rivaroxaban (once a day)20 and dabigatran (twice a day)21 are approved by the Therapeutic Goods Administration (TGA) for use in prevention of stroke or systemic embolism in people with non-valvular AF.

Therapeutic Guidelines recommends these as second-line treatment options for people with non-valvular AF at moderate-to-high risk of stroke.4 Some international guidelines recommend both drugs as second-line,22 some as first-line alongside23,24 or even before warfarin.1,18 

Apixaban is not yet approved by the TGA but is approved by the US Food and Drug Administration and for use in Europe for the prevention of stroke or systemic embolism in people with non-valvular AF . 

Dabigatran, rivaroxaban and apixaban are listed on the Pharmaceutical Benefits Scheme (PBS) for the prevention of venous thromboembolism following total knee or total hip replacement.25-27 

Rivaroxaban is also PBS-listed for the treatment of symptomatic deep vein thrombosis without symptomatic pulmonary embolism and to prevent recurrent venous thromboembolism.28

Warfarin is not suitable for all; neither are the newer drugs

Warfarin is underused for many reasons, including the perceived risk of bleeding,10,29 the need for regular monitoring and the risk of adverse interactions with other drugs and foods.23 

The newer oral anticoagulants address some of these concerns for people with non-valvular AF (eg, INR monitoring is not required, lower risk of ICH) but some drug interactions remain a concern.6,20,21,30 The newer oral anticoagulants have not yet been tested in people who are unsuited to warfarin.31-33 

Dabigatran, rivaroxaban and apixaban are contraindicated in people with severe renal impairment.6,20,21,30

When to consider switching from warfarin

  • A patient is not able to maintain a consistent INR within 2–3 (60–70% of the time) despite all correctable factors having been addressed and adherence confirmed. Those not able to maintain a consistent INR within 2–3 due to poor adherence are not suitable for newer oral anticoagulants and should not be switched.
  • A patient is not able to tolerate warfarin or is no longer able to undergo routine INR monitoring.
  • It is the patient’s preference.

Refer to the Product Information for details about switching between therapies.20,21


People well-controlled on warfarin may not benefit clinically from switching

In key trials, dabigatran 150 mg (RE-LY trial) and apixaban (ARISTOTLE trial) both reduced the incidence of stroke/systemic embolism compared with warfarin based on analysis of the intention-to-treat trial populations.31,32 

Rivaroxaban (ROCKET-AF trial) and dabigatran 110 mg (the indicated dose in Australia for people > 75 years)21 were non-inferiorc to, but not better than, warfarin for the incidence of stroke/systemic embolism based on analysis of the intention-to-treat trial populations.31,33 

Participants in the warfarin arms of the rivaroxaban, apixaban and dabigatran key trials had a mean TTRd of 55%, 62.2% and 64%,  respectively.31-33 When warfarin was well controlled (≥ 66% TTR34) dabigatran (110 mg and 150 mg) and apixaban were non-inferior to warfarin for the incidence of stroke/systemic embolism.31,32 

Non-inferiority was not shown by rivaroxaban for incidence of stroke/systemic embolism when warfarin was well controlled.33,35,36 Only apixaban was better than warfarin for reducing major bleeding when warfarin was well controlled.32,34,e

Participants in the key trials all had non-valvular AF. Participants in the rivaroxaban study had at least two additional risk factors for stroke and a mean CHADS2 score of ~3.5 whereas participants in the dabigatran and apixaban trials had at least one additional risk factor for stroke and an average CHADS2 score of 2.1–2.2 and 2.1, respectively.31-33 

Concomitant use of anticoagulant and aspirin was permitted in both arms of the trials, which may have affected safety outcomes.31-33

The newer oral anticoagulants are associated with reduced incidence of ICH compared with warfarin, but absolute risk reductions in trials were small and rivaroxaban and dabigatran (150 mg) had increased incidences of GI bleeds (Table 5).31-33 Incidence of death after ICH was similar with warfarin and dabigatran in trials.37

c Non-inferior: the effect of a new drug is not worse than that of another drug by more than a specified margin.

d Time in therapeutic range (TTR): the percentage of time when a person has an INR within the prescribed range, when stroke and bleeding risk is lowest. For people with non-valvular AF this is 2–3.

e In the ARISTOTLE trial, bleeding outcomes were assessed in patients who received at least one dose of the study drug and events that occurred from the time the patients received the first dose of the study drug through 2 days after they received the last dose.

Table 5: Newer oral anticolagulants compared with warfarin31,32,36,f

Intracranial haemorrhage: number needed to treat Major gastrointestinal bleed: number needed to harm
Dabigatran 150 mg228 204
Dabigatran 110 mg 197N/Ag

f Number needed to treat to prevent one intracranial haemorrhage or number needed to harm to cause one extra major gastrointestinal bleed in 1 year by using newer oral anticoagulants compared with warfarin.

g Differences in events compared with warfarin non-significant.


Ongoing INR monitoring

Point-of-care testing reduces some of the burden of laboratory-based INR testing and may be useful in rural or remote areas where access to laboratory testing is difficult.38 A recent meta-analysis showed that point-of-care testing (where patients self-tested or self-managed anticoagulation):39 

  • significantly reduced incidence of thromboembolism
  • resulted in similar or better INR control (TTR) 
  • was associated with a similar incidence of major haemorrhage and death compared with primary care or anticoagulation clinic INR testing.

Subset analysis for patients with AF showed no significant difference for incidence of thromboembolism or major haemorrhage.39 

A previous meta-analysis comparing anticoagulation monitoring using a point-of-care testing device with laboratory analysis showed similar results, although use of point-of-care testing devices significantly reduced mortality and improved TTR.40 

Not all patients with non-valvular AF will be suitable for point-of-care testing (patients need visual acuity, manual dexterity and cognitive abilities) and how well its accuracy compares with laboratory results is uncertain.41 Medicare does not currently fund point-of-care testing.

Newer oral anticoagulants do not need routine clinical monitoring but the people taking them do

Bleeding risk assessment is especially important in these people because there is no readily available, validated method of measuring or reversing the activity of the newer oral anticoagulants. 

Ensure good adherence because the newer drugs have a much shorter half-life than warfarin and the risk of stroke is greater if a dose is missed.1

Assess renal function before starting anticoagulant treatment and at least annually thereafter (more frequently in people with renal impairment).1,18 

Assess hepatic function before starting dabigatran.21


Peri-procedural management of oral anticoagulants

The need for anticoagulant substitution with heparin depends on the patient’s thromboembolic risk.

In people at low risk of stroke, stop warfarin 3–5 days before procedures when there is a high risk of bleeding and resume warfarin once adequate haemostasis is achieved.4 

For people with high risk of thromboembolism undergoing procedures associated with a high risk of bleeding, anticoagulant substitution with a therapeutic dose of unfractionated heparin or low molecular weight heparin is often required (see Therapeutic Guidelines).4 

Some procedures associated with a low risk of bleeding (eg, simple dental procedures) do not require warfarin to be stopped if the INR is within 2–3.42

Stop dabigatran and rivaroxaban at least 1–3 days before any surgery, or earlier for people at higher risk of bleeding, those with renal impairment and those undergoing procedures with a high risk of bleeding.6,20,21 Stop apixaban at least 3 days before surgery.6 

Expert opinion suggests that a substitute anticoagulant would not be required if anticoagulation were stopped for periods up to 48 hours for most procedures in people at low bleeding risk and with normal renal function.43 Dabigatran and rivaroxaban can also be restarted when adequate haemostasis has been restored.1

If heparin is needed, dabigatran, rivaroxaban or apixaban can be restarted after the operation when the next dose of heparin would have been given. As they have an immediate onset of action, crossover therapy with the heparin is not needed.4 

No evidence-based guidelines for the dental management of patients receiving these agents are currently available.


Expert reviewers

  • Prof David Brieger, Concord Clinical School, University of Sydney & ANZAC Research Institute, Concord Hospital, Sydney, NSW
  • Dr Jenny Curnow, Staff specialist, ANZAC Research Institute Concord Repatriation General Hospital, Sydney, NSW
  • Assoc Prof John Worthington, Faculty of Medicine, University of NSW & Ingham Institute, Liverpool Health Service, Sydney, NSW


  • Dr John Dowden, Editor, Australian Prescriber
  • Dr Graham Emblen, GP, Toowoomba, QLD
  • Dr Oliver Frank, GP, Hillcrest, SA
  • Benafsha Khariwala, Editor, Therapeutic Guidelines
  • Deborah Norton, QUM Pharmacist, West Vic DGP


  1. Camm AJ, Lip GY, De Caterina R, et al. 2012 focused update of the ESC Guidelines for the management of atrial fibrillation: An update of the 2010 ESC Guidelines for the management of atrial fibrillation. Europace 2012;14:1385\u2013413. [PubMed].
  2. Australian Government Department of Human Services. Home Medicines Review (HMR). 2012.
  3. Roughead EE, Barratt JD, Ramsay E, et al. Collaborative home medicines review delays time to next hospitalization for warfarin associated bleeding in Australian war veterans. J Clin Pharm Ther 2011;36:27\u201332.[PubMed].
  4. Cardiovascular Expert Group. Therapeutic guidelines: cardiovascular version 6. Melbourne: Therapeutic Guidelines Ltd, 2012. [eTG online] (accessed 26 November 2012).
  5. Gage BF, Waterman AD, Shannon W, et al. Validation of clinical classification schemes for predicting stroke: results from the National Registry of Atrial Fibrillation. JAMA 2001;285:2864\u201370. [PubMed].
  6. Australian Medicines Handbook. Adelaide: Australian Medicines Handbook Pty Ltd, 2012.
  7. 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\u2013100. [PubMed].
  8. Olesen JB, Lip GY, Lindhardsen J, et al. Risks of thromboembolism and bleeding with thromboprophylaxis in patients with atrial fibrillation: A net clinical benefit analysis using a 'real world' nationwide cohort study. Thromb Haemost 2011;106:739\u201349. [PubMed].
  9. Hart RG, Pearce LA, Aguilar MI. Meta-analysis: antithrombotic therapy to prevent stroke in patients who have nonvalvular atrial fibrillation. Ann Intern Med 2007;146:857\u201367. [PubMed].
  10. Gattellari M, Worthington J, Zwar N, et al. Barriers to the use of anticoagulation for nonvalvular atrial fibrillation: a representative survey of Australian family physicians. Stroke 2008;39:227\u201330. [PubMed].
  11. Gage BF, Birman-Deych E, Kerzner R, et al. Incidence of intracranial hemorrhage in patients with atrial fibrillation who are prone to fall. Am J Med 2005;118:612\u20137. [PubMed}.
  12. Rosenman MB, Baker L, Jing Y, et al. Why is warfarin underused for stroke prevention in atrial fibrillation? A detailed review of electronic medical records. Curr Med Res Opin 2012;28:1407\u201314. [PubMed].
  13. Pugh D, Pugh J, Mead GE. Attitudes of physicians regarding anticoagulation for atrial fibrillation: a systematic review. Age Ageing 2011;40:675\u201383. [PubMed].
  14. Scottish Intercollegiate Guidelines Network (SIGN). Antithrombotics: indications and management: NHS Quality Improvement Scotland, 2012.
  15. Baczek VL, Chen WT, Kluger J, et al. Predictors of warfarin use in atrial fibrillation in the United States: a systematic review and meta-analysis. BMC Fam Pract 2012;13:5. [PubMed].
  16. van Walraven C, Hart RG, Connolly S, et al. Effect of age on stroke prevention therapy in patients with atrial fibrillation: the atrial fibrillation investigators. Stroke 2009;40:1410\u20136. [PubMed].
  17. Apostolakis S, Lane DA, Guo Y, et al. Performance of the HEMORR(2)HAGES, ATRIA, and HAS-BLED Bleeding Risk-Prediction Scores in Patients With Atrial Fibrillation Undergoing Anticoagulation: The AMADEUS (Evaluating the Use of SR34006 Compared to Warfarin or Acenocoumarol in Patients With Atrial Fibrillation) Study. J Am Coll Cardiol 2012;60:861\u20137. [PubMed].
  18. Skanes AC, Healey JS, Cairns JA, et al. Focused 2012 update of the Canadian Cardiovascular Society atrial fibrillation guidelines: recommendations for stroke prevention and rate/rhythm control. Can J Cardiol 2012;28:125\u201336. [PubMed].
  19. Camm AJ, Kirchhof P, Lip GY, 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). Europace 2010;12:1360\u2013420.
  20. Bayer Australia Ltd. Xarelto (Rivaroxaban) Product Information. 3 April 2012.
  21. Boehringer Ingelheim Pty Ltd. Dabigatran Product Information. 9 November 2011.
  22. Canadian Agency for Drugs and Technologies in Health. CADTH Therapeutic Review Recommendations: new oral anticoagulants for the prevention of thromboembolic events in patients with atrial fibrillation, 2012.
  23. National Institute for Health and Clinical Excellence. Atrial fibrillation \u2013 dabigatran etexilate: guidance 2012.
  24. National Institute for Health and Clinical Excellence. TA256 Atrial fibrillation (stroke prevention) \u2013 rivaroxaban: guidance, 2012.
  25. NPS MedicineWise. Dabigatran (Pradaxa) for preventing venous thromboembolism after hip or knee replacement surgery. NPS RADAR 2010. [Online].
  26. NPS MedicineWise. Rivaroxaban (Xarelto) for preventing venous thromboembolism after hip or knee replacement surgery. NPS RADAR, 2009. [Online].
  27. Apixaban (Eliquis) for preventing venous thromboembolism after knee or hip replacement surgery. NPS RADAR 2012. [Online].
  28. NPS MedicineWise. Rivaroxaban (Xarelto) for stroke prevention in non-valvular atrial fibrillation. NPS RADAR, 2012. [Online].
  29. Lip GYH, Andreotti F, Fauchier L, et al. Bleeding risk assessment and management in atrial fibrillation patients: Executive summary of a position document from the European Heart Rhythm Association [EHRA], endorsed by the European Society of Cardiology [ESC] working group on thrombosis. Thromb Haemost 2011;106:997\u20131011. [PubMed].
  30. Bristol-Myers Squibb Australia Pty. Ltd. Apixaban Product Information, 2011.
  31. Connolly SJ, Ezekowitz MD, Yusuf S, et al. Dabigatran versus warfarin in patients with atrial fibrillation. [Erratum appears in N Engl J Med. 2010 Nov 4;363(19):1877]. N Engl J Med 2009;361:1139\u201351. [PubMed].
  32. Granger CB, Alexander JH, McMurray JJV, et al. Apixaban versus warfarin in patients with atrial fibrillation. N Engl J Med 2011;365:981\u201392. [PubMed].
  33. Patel MR, Mahaffey KW, Garg J, et al. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med 2011;365:883\u201391. [PubMed].
  34. Canadian Agency for Drugs and Technologies in Health. Safety, effectiveness, and cost-effectiveness of new oral anticoagulants compared with warfarin in preventing stroke and other cardiovascular events in patients with atrial fibrillation. 2012. [Online] (accessed 30 August 2012).
  35. Australian Government Department of Health and Ageing. March 2012 PBAC public summary documents by meeting rivaroxaban, tablet, 15 mg and 20 mg (Xarelto). [Australian Government Web Archive online] (accessed 24 August 2012).
  36. U.S. Food and Drug Administration. FDA Draft Briefing Document for the Cardiovascular and Renal Drugs Advisory Committee (CRDAC), UCM270797. FDA: Washington, 2011.
  37. Hart RG, Diener HC, Yang S, et al. Intracranial hemorrhage in atrial fibrillation patients during anticoagulation with warfarin or dabigatran: the RE-LY trial. Stroke 2012;43:1511\u20137.
  38. Cardiac Society of Australia and New Zealand. Submission to the Review of Anticoagulation Therapies in Atrial Fibrillation, 2012. [Online] (accessed 7 September 2012).
  39. Heneghan C, Ward A, Perera R, et al. Self-monitoring of oral anticoagulation: systematic review and meta-analysis of individual patient data. Lancet 2012;379:322\u201334. [PubMed].
  40. Wells PS, Brown A, Jaffey J, et al. Safety and effectiveness of point-of-care monitoring devices in patients on oral anticoagulant therapy: a meta-analysis. Open Medicine 2007;1:e131\u201346. [PubMed].
  41. Bubner TK, Laurence CO, Gialamas A, et al. Effectiveness of point-of-care testing for therapeutic control of chronic conditions: results from the PoCT in General Practice Trial. Med J Aust 2009;190:624\u20136. [PubMed].
  42. Baker RI, Coughlin PB, Gallus AS, et al. Warfarin reversal: consensus guidelines, on behalf of the Australasian Society of Thrombosis and Haemostasis. Med J Aust 2004;181:492\u20137. [PubMed].
  43. Schulman S, Crowther MA. How I treat with anticoagulants in 2012: new and old anticoagulants, and when and how to switch. Blood 2012;119:3016\u201323.