This module considers the clinical use of anticoagulants: the initiation of anticoagulation with vitamin K antagonists and practical issues around prescribing and monitoring; the use of direct oral anticoagulants (DOACs) and their effect on laboratory tests; interruption of anticoagulation for invasive procedures and in patients undergoing cardioversion; along with management of bleeding issues. The therapeutic indications for anticoagulants in acute coronary syndromes and non-valvular atrial fibrillation (AF), as well as their use in cardioversion for AF, are covered in module 3, which also considers the choice of anticoagulant for stroke prevention. The use of anticoagulants in heart valve disease is covered in module 7 of our heart valve disease learning programme.
Practical issues associated with vitamin K antagonists
Vitamin K antagonists (VKAs) are effective anticoagulants but are considered high risk medications, on account of their narrow therapeutic index and poor safety profile. The risk of major haemorrhage is 2–5% per year and is higher in the elderly or where other risk factors for bleeding co-exist. Bleeding risk can be estimated using formal bleeding risk scores, such as the HAS-BLED score (see module 3). Good anticoagulant management is important in reducing risk. Warfarin is the VKA of choice in the UK and comes in four colour-coded doses (see figure 1)2
The level of anticoagulation is assessed using the International Normalised Ratio, or INR. The INR is derived from the prothrombin time, with a correction to adjust for the sensitivity of each laboratory’s reagents (see below). The INR therefore gives a standardised measure of the degree of anticoagulation with warfarin: it provides consistency between laboratories. The target INR depends on the indication for anticoagulation but for atrial fibrillation (AF) and most non-valvular cardiac conditions it is 2.5 (range 2.0–3.0).
Table 1 shows the various indications for anticoagulant treatment, as well as INR target and range and the duration of treatment.
There is great biological variability in response to VKAs. Most of this is genetically determined but there are also many interactions with drugs and foodstuffs. Two common polymorphisms in the cytochrome P450 2C9 system are associated with increased warfarin sensitivity, and mutations in the VKORC1 gene complex can cause either sensitivity or resistance to VKAs.
Drugs can interact with warfarin in a number of ways:
- antibiotics can reduce the synthesis of vitamin K, which is usually performed by gut bacteria
- some drugs can affect warfarin absorption (e.g. cholestyramine)
- Inducers or inhibitors of the cytochrome P450 system can have substantial effects on warfarin levels. Commonly-used drugs which can increase warfarin levels include several antibiotics, omeprazole, statins and amiodarone. Over-the-counter remedies can also affect levels (e.g. glucosamine can increase levels, St John’s Wort can reduce levels), as can some foods (e.g. grapefruit and cranberry can increase warfarin levels; acute excessive alcohol consumption may reduce warfarin metabolism). Significantly increasing intake of green leafy vegetables, rich in vitamin K, can lead to a reduction in INR.
Always consult the full list of potential interactions in the British National Formulary. Great care should be taken when making changes to medication in patients taking warfarin, including more frequent INR checks.
Starting anticoagulation with warfarin
Anticoagulation should be initiated only when appropriate follow up has been arranged. Frequent checks of the INR are required until the dosage is stabilised, at least twice in the first week, and if the patient is to attend an anticoagulant clinic it is often safer to defer initiation until first attendance at the clinic.
Discuss the risks and benefits of anticoagulation with the patient and ensure they have consented to the treatment. If the patient is a younger woman, check that she is not pregnant or planning pregnancy as warfarin can cause embryopathy between six and 12 weeks gestation.
Check for bleeding history and other medication that may interact with warfarin (see table 2). Also check baseline INR, full blood count, liver and renal function. A bleeding risk assessment should be used to direct optimisation of modifiable bleeding risk factors, such as uncontrolled hypertension and alcohol excess.
Complete the locally available Warfarin anticoagulant record booklet and dosage record (“yellow book”) and give this to the patient. Keep separate clinical notes including INR results and dosage.
Ensure the patient understands:
- the rationale for treatment
- the importance of regular blood tests
- variable warfarin doses and tablet strengths
- what to do and who to contact if bleeding or bruising or any other complications occur
- the importance of extra checks during periods of intercurrent illness, or when starting or stopping other medication, including self-medication
- the teratogenicity of warfarin in pregnancy where appropriate
- the possible effects of dietary changes and alcohol intake
- any limitations to sports and leisure activities
- the importance of informing health care professionals about anticoagulant use.
Warfarin induction regimes
Locally approved induction protocols should be followed in all patients. A typical induction dose for acute venous thromboembolism (VTE) is 5–10 mg on day 1 and day 2, with subsequent doses dependent on INR. For an indication of AF stroke prevention, where there is less urgency to achieve therapeutic anticoagulation, a slow-loading regimen (e.g. 1–3 mg per day) is generally safer, particularly in the outpatient setting.4 Risk factors such as age >80 years, heart failure, liver disease or other bleeding risk should also prompt a reduced starting dose.
Patients requiring cardioversion
These patients should have a therapeutic INR (>2.0) for at least three weeks before and four weeks after cardioversion to reduce the risk of embolisation. INR checks should be performed at least weekly during this period. As INR values do tend to fluctuate it is useful to aim for target INR between 2.5 and 3.0 to avoid dips below 2.0 that result in cancellation of the procedure.
If the patient has been in AF for a short period only, and cardioversion is performed after transoesphageal echocardiography has shown no evidence of intra-atrial thrombus, anticoagulation is not mandatory.
High INR values and bleeding in warfarinised patients
The British Committee for Standards in Haematology Guidelines (BCSH) on oral anticoagulation with warfarin (4th edition)3 has guidance on high INR and bleeding to which the reader is directed for more complete information (see also table 3). This is provided for educational purposes only – specialist advice should always be sought when required. Unexpected bleeding at therapeutic levels of warfarin should prompt investigation for a possible underlying cause.
Interruption of anticoagulation for invasive procedures
For dental procedures, including extraction, warfarin should not be stopped, but the INR should be <4.0. Other low risk procedures, such as skin biopsy and endoscopy, without biopsy can also be performed without stopping warfarin. The assessment of bleeding risk associated with any given procedure should be led by the operating practitioner or dentist.
If a reduction in INR to <1.5 is required, warfarin should be stopped three to five days (depending on INR and target range) before the procedure. Depending on thrombotic risk, (including CHADS2 or CHA2DS2-VASc score), bridging anticoagulation with heparin or low molecular weight heparin may be considered, but needs to be balanced against potential bleeding risk. The BCSH have made a strong recommendation (grade 1A) that bridging should not be given to AF patients who have a CHADS2 score ≤4 and who have not had a stroke/transient ischaemic attack (TIA) within three months.8 This is based on the BRIDGE trial,12 a randomised placebo-controlled trial that showed an increased risk of major bleeding from bridging this group of AF patients, without a reduction in arterial thromboembolism. In higher risk patients who were not represented within the trial (stroke/TIA within three months or CHADS2 ≥5), bridging should still be considered. With regards to metallic heart valves, the BCSH recommends consideration of bridging, except in patients with bileaflet aortic valves and no additional risk factors.8
Overall, this is a specialist area and it is best to seek advice from a haematologist for an appropriate regime, and to ensure a clearly documented plan is available in a timely manner and communicated fully to the patient and relevant healthcare professionals.
Patient self-testing is considered a cost-effective option that can improve both the quality of anticoagulation management and patient satisfaction. The CoaguChek® series of devices (Roche -see figure 2) are approved by the National Institute for Health and Care Excellence (NICE), and have built in quality control material that gives good correlation with venous blood samples in patients who are stably anticoagulated. Patients on long-term anticoagulation may wish to purchase such a machine. The disposables required can be provided on NHS prescription. Patients who wish to self-test should discuss this with their anticoagulant clinic before purchase of a machine, and should remain under the overall care of an anticoagulant service or their GP. The performance of the machine should be checked in the clinic every six months. Long-term, patients may also have the option of progressing to self-management, where they are trained to make dose adjustments themselves.
|Reproduced from Roche Diagnostics with permission|
Practical issues associated with direct oral anticoagulants (DOACs)
In those who are suitable for a DOACs, the choice of drug must be selected carefully (See table 45 below taken from North Central London DOAC prescribing support)
Prior to starting patients on a DOAC, patients must be adequately informed of the importance of taking their medication at the same time every day and with instructions in the event of bleeding complications.
DOACs remain contraindicated in patients with mechanical heart valves.
Laboratory measurement of anticoagulant effect
This is not required in routine use but it may be useful to know DOAC levels in certain situations such as:6
- In the presence of spontaneous or traumatic haemorrhage
- Following suspected overdose
- When patients are taking another interacting drug
- To monitor efficacy in patients presenting with new thrombosis whilst on the anticoagulant
- When emergency surgery is required
- In patients due to have neuraxial anaesthesia for elective or emergency procedures or surgery
- In patients requiring elective surgery and in whom the drug may still be present
- In patients with renal impairment
- When bridging from one anticoagulant to another
- To assess compliance
- At the extremes of body weight
- In subjects with prior intestinal surgery where it is unclear if absorption will be affected
- Trough levels may be useful to assess potential accumulation in very elderly patients.
Such testing, however, should only be performed under the guidance of a coagulation expert, recognising potential uncertainties in the interpretation of results, and the absence of good quality randomised data to support this strategy in clinical practice.
Although there is currently no routine measurement of anticoagulant effect with DOACs, it is important to monitor renal function and weight regularly (e.g every three to 12 months depending on baseline creatinine clearance) to ensure suitability and correct dosing.
Routine clotting tests and DOACs
When interpreting the results of screening clotting tests, it is useful to have the coagulation cascade in mind. Figure 3 shows the coagulation cascade, marked to show which sections are tested by the three screening clotting tests: PT (prothrombin time), APTT (activated partial thromboplastin time), and TT (thrombin time).
As might be predicted from figure 3, and a knowledge of the mechanism of action of DOACs, these anticoagulants may affect the results of screening clotting tests. However, results of these tests with DOACs are not consistent or standardised, and they therefore cannot be used for monitoring of these agents. Perhaps most importantly, results of standard clotting tests can be normal, even at therapeutic levels of anticoagulation with DOACs. Each lab will get different results depending on reagents, and results should be discussed with lab staff or a clinical haematologist. Table 5 is a guide showing the influence of DOACs on screening coagulation tests.3,7
Table 5. Influence of DOACs on screening coagulation tests
|PT||Prolonged +||Prolonged ++
If PT normal, suggests low drug levels (provided sensitive reagent used)
|May be normal||Usually prolonged but may be normal. Normal PT makes high drug levels unlikely|
|APTT||Prolonged ++||Prolonged +||May be normal||Usually prolonged, but may be normal|
|Thrombin time||Prolonged ++. If TT normal, dabigatran concentration likely very low||Normal||Normal||Normal|
|Data compiled from Keeling D3 and Dale BJ7
Key: APTT = activated partial thromboplastin time; DOAC = direct oral anticoagulant; PT = prothrombin time
For accurate measurement of drug levels, specialist assays have been developed (table 6). These may not be available in all laboratories.
Cessation of DOACs before invasive procedures
This is a confusing area. The manufacturers of each of the DOACs give recommendations, but pharmacokinetic modelling suggests that longer may be needed, especially in patients with impaired renal function, and where ensuring adequate haemostasis is crucial. Bridging anticoagulation with low molecular weight heparin is not recommended. As ever, decisions should be individualised based on thrombotic and bleeding risk, and specialist advice sought. We recommend that the European Heart Rhythm Association (EHRA) guidance9 published in 2021 is consulted for how to cease DOACs before invasive procedures. The British Society of Haematology have also recently published a useful addendum to a previous guideline in this area (https://b-s-h.org.uk/media/19825/addendum-to-the-guideline-on-peri-operative-management-of-anticoagulation-and-antiplatelet-therapy.pdf).
Management of bleeding with DOACs
Specific reversal agents exist for all DOACs besides edoxaban (idarucizumab is available for dabigatran; andexanet alpha for rivaroxaban and apixaban – these are discussed in more depth in module 3).
The indications for when to use this include:10
- Life threatening bleeding including intracranial haemorrhage
- Bleeding in a closed space or critical organ such as intraspinal or intraocular area
- Persistent major bleeding despite local haemostatic measures or risk of recurrent bleeding because of delayed DOAC clearance or DOAC overdose
- Need for urgent intervention that is associated with a high risk of bleeding that can not be delayed to await drug clearance
- Emergency surgery or intervention in patients at high risk for procedural bleeding, for example – neurosurgery.
It should be noted, however, that in latest NICE guidance, andexanet alfa is only approved for use in the setting of life-threatening gastointestinal bleeding.13
The specific reversal agents should not be used in those having elective surgery, gastrointestinal bleeding that is responding to supportive measures, those with high drug levels without bleeding, and those who need surgery that can be delayed until the DOAC has cleared.
General haemostatic supportive measures are key to the management of DOAC-related bleeding, and should not be neglected even in circumstances where a specific reversal agent is used. The European Heart Rhythm Association has published guidance9 for management of mild, non-life threatening major, and life-threatening/critical site bleeding. Please refer to guidelines at your local hospital and discuss with a haematologist when appropriate.
1. National Institute for Health and Care Excellence. Key therapeutic topic: Anticoagulants, including direct-acting oral anticoagulants (KTT16). Last updated: September 2019. Available at: https://www.nice.org.uk/advice/ktt16
2. Warfarin Anticoagulant Record Booklet. ADVANZ Pharma. Available from https://www.medicines.org.uk/emc/rmm/1081/Document [last accessed 16th May 2021]
3. Keeling D, Baglin T, Tait C et al. British Committee for Standards in Haematology Guidelines on oral anticoagulation with warfarin – fourth edition. Br J Haematol 2011;154:1365–2141. http://doi.org/10.1111/j.1365-2141.2011.08753.x
4. National Institute for Health and Care Excellence. NICE Clinical Knowledge Summaries. Anticoagulation – oral. Available at http://cks.nice.org.uk/anticoagulation-oral [last accessed 16th May 2021]
5. Drebes A, Gates C for the North Central London Joint Formulary Committee. DOAC prescribing support for NCL, AF and VTE. https://www.ncl-mon.nhs.uk/wp-content/uploads/Guidelines/9_DOAC_prescribing_support.pdf
6. Kitchen S, Gray E, Mackie I, Baglin T, Makris M and the BCSH committee. Measurement of non‐Coumarin anticoagulants and their effects on tests of haemostasis: Guidance from the British Committee for Standards in Haematology. Br J Haematol 2014;166:830–41. https://doi.org/10.1111/bjh.12975
7. Dale BJ, Chan NC, Eikelboom JW. Laboratory measurement of the direct oral anticoagulants. Br J Haematol 2015 (published online 22nd October 2015). http://doi.org/1111/bjh.13810
8. Keeling D, Tait RC, Watson H, the British Committee of Standards for Haematology. Peri-operative management of anticoagulation and antiplatelet therapy. Br J Haematol. 2016;175(4):602-613. https://doi.org/10.1111/bjh.14344
9. Steffel J, Collins R, Antz M, et al. 2021 European Heart Rhythm Association Practical Guide on the Use of Non-Vitamin K Antagonist Oral Anticoagulants in Patients with Atrial Fibrillation. Published online 25th April 2021. https://doi.org/10.1093/europace/euab065
10. Levy JH, Ageno W, Chan NC, Crowther M, Verhamme P, Weitz J and Subcommittee on Control of Anticoagulation. When and how to use antidotes for the reversal of direct oral anticoagulants: Guidance from the SSC of the ISTH. J Thromb Haemost 2016;14:623–7. https://doi.org/10.1111/jth.13227
11. Cuker A, Siegal D. Monitoring and reversal of direct oral anticoagulants. Hematology Am Soc Hematol Educ Program 2015;1117–24. https://doi.org/10.1182/asheducation.V2015.1.117.3916182
12. Douketis JD, Spyropoulos AC, Kaatz S, et al; BRIDGE Investigators. Perioperative Bridging Anticoagulation in Patients with Atrial Fibrillation. N Engl J Med. 2015;373(9):823-33. https://doi.org/10.1056/NEJMoa1501035
13. National Institute of Health and Care Excellence. Andexanet alfa for reversing anticoagulation with apixaban and rivaroxaban. Technology appraisal guidance [TA697]. Published 12th May 2021. https://www.nice.org.uk/guidance/TA697 [last accessed 18th August 2021]
Hicks T, Stewart F, Eisinga A. NOACs versus warfarin for stroke prevention in patients with AF: a systematic review and meta-analysis. Open Heart 2016;3:e000279. http://doi.org/10.1136/openhrt-2015-000279