1

Atrial Fibrillation and Anticoagulation Slough CCG Bracknell & Ascot CCG Windsor, Ascot & Maidenhead CCG Detection and appropriate treatment of atrial fibrillation (AF) will ultimately decrease disease burden and cardiovascular spend on AF related disease such as heart failure and stroke. This paper is designed to:

Improve patient outcomes by: Improved screening and detection Targeting patients with long term conditions Treating with appropriate anticoagulation

Provide clear guidance and defined local pathway: Education – understanding stroke prevalence, stroke and bleeding risks Resources to support uptake

Effect of Atrial Fibrillation on Health

AF affects 835,000 people in England and through increasing susceptibility to stroke, it is a major cause of both morbidity and mortality. The QOF 2011/12 estimate of AF was 1.48%, but The NICE Commissioning Guide estimates the prevalence of AF as 1.6%, although accepts that with targeted opportunistic screening, the true prevalence of AF for England is of the order of 2.0%. Stroke is the third largest cause of death in England and costs the NHS £2.8 billion each year, yet an estimated 200,000 strokes per year are preventable. AF confers a 5-fold risk of stroke, and 1 in 5 of all strokes are attributed to this arrhythmia. Ischaemic strokes in association with atrial fibrillation are often fatal, and those patients who survive are left more disabled by their stroke and more likely to suffer a recurrence than patients with other causes of stroke. What changes could be made to improve health outcomes? The 2006 atrial fibrillation NICE Guideline laid down criteria for anticoagulation, yet amongst patients with known AF, only 55% of those fulfilling the 2006 criteria for anticoagulant therapy currently receive it. It is thought that the attitude of healthcare professionals and perceived risks of anticoagulation may be a major factor limiting uptake. When considering both diagnosed and undiagnosed AF patients, it is estimated that less than a quarter in need of stroke risk reduction gets adequate treatment. Of 11,939 patients admitted with stroke to hospitals in UK (excl Scotland) in the first 3 months of 2013, about one fifth were in AF on admission. Only 36% were receiving an anticoagulant, yet 38% were on an antiplatelet drug as sole antithrombotic therapy and 26% were on no antithrombotic treatment. This low rate highlights not just the large undiagnosed population but widespread ineffective treatment with antiplatelets such as aspirin or clopidogrel1.

2

Over 15,000 strokes and 5,500 patient deaths would be prevented every year if AF patients in the UK were effectively identified and treated with oral anticoagulation. Earlier detection will also reduce cases of heart failure, frequently a result of long term AF. The importance of recognising and treating AF increases with age - the incidence of resulting strokes rises from 1.5% amongst people aged 50-59 to 23.5% amongst people aged 80-89. There is wide variation in detection (when compared with expected prevalence), exception reporting and choice of treatment. The aim is to reduce this variation thereby reducing the incidence and disease burden of strokes. In those patients with atrial fibrillation whose latest record of a CHADS2 score is greater than 1, the percentage of patients who are currently treated with anti-coagulation therapy

AF WAM B&A Slough England

CCG Population 143,343 135,934 150,554

Prevalence AF 1.4% 1.2% 0.7% 1.5%

Number of AF patients per practice

5 – 322 Total CCG = 2249

11 – 253 Total CCG = 1618

11 – 199 Total CCG = 1063

849,407

% pts with CHADS >1 on anticoagulant (net of exceptions)

68.3 – 100% (Average 85.1%)

73.1 – 100% (Average 83.2%)

66.7 – 100% (Average 79.4%)

81.3%

Exception rate 0 – 45.2% (Average 17.0%)

0 – 50% (Average 19.2%)

0 – 64.3% (Average 21.8%)

19.96%

Actual % pts with CHADS >1 on anticoagulant (excepted patients included)

54.8 – 100% (Average 70.1%)

50 – 81.8% (Average 66.6%)

35.7 – 90.9% (Average 61.3%)

65.1%

Stroke or Transient Ischaemic Attacks (TIA) Prevalence

1.5% 1.3% 1.2% 1.7%

Annual TIA/CVA NEL / 1000 weighted patients

0.59 – 4.56 (Average 2.67)

0.82 – 4.8 (Average 2.92)

0 – 4.98 (Average 2.47)

There are 4930 patients in Berkshire East on the AF register and there are 134 TIA/stroke /1000 weighted patients in Berkshire East annually (2012/13). For every 100 patients on an AF register who are not adequately treated, 5 stokes each year are estimated. Detection Common causes of AF include hypertension, ischaemic heart disease, mitral valve disease, heart failure (cause and effect) and may occur post operatively. Screen all patients presenting with palpitations, breathlessness/dyspnoea, syncope/dizziness, chest discomfort, TIA/stroke. Patients with persistent high alcohol intake, obstructive sleep apnoea, and those who are obese are also at higher risk of AF. Genetic causes represent <1%. Other innovative ways of increasing AF detection include pulse taking at vaccination and long term condition clinics, e.g. falls clinics, flu, shingles, pneumonia vaccination; diabetes, hypertension, cardiovascular and heart failure clinics, mindful that prevalence is likely to be higher in patients with long term uncontrolled hypertension and >20% in patients >80 years. SAFE trial compared active vs passive AF screening and showed mass ECG screening is not cost effective and does not get better results. However, when a pulse was taken opportunistically and a 12 lead ECG run on those with an irregular pulse, 1 case was diagnosed per 4-5 ECGs.

3

Exception reporting GRASP-AF is a query and risk stratification tool which is free and available for use with all GP clinical systems in England. GRASP-AF provides a set of MIQUEST queries to identify, for each practice, patients with a diagnosis of AF who are not on warfarin. GRASP–AF calculates stroke risk using CHADS2 / CHA2DS2–VASc scoring system and highlights patients with a CHADS2 score of 2 or more who are not on warfarin. By using a GP / patient friendly information leaflet summarising treatment options, place in therapy and the risks vs benefit of AF treatment, those who have either declined or not been offered anticoagulation in the past can be targeted. A sample patient letter to call this patient group in for review can be found Appendix 1. The detail aid (Appendix 2) and new oral anticoagulant drugs (NOACs) patient counselling sheet (Appendix 3) may be used to facilitate a discussion at review regarding the benefits vs risks of anticoagulation and the choice of treatments. Choice of treatment (NICE Clinical Guideline 180 August 2014) 1. Use the CHA2DS2–VASc stroke risk score to assess stroke risk in people with:

symptomatic or asymptomatic paroxysmal, persistent or permanent AF

atrial flutter

a continuing risk of arrhythmia recurrence after cardioversion back to sinus rhythm. 2. Offer anticoagulation to people with a CHA2DS2–VASc score of 2 or above, taking bleeding risk into account. 3. Consider anticoagulation for men with a CHA2DS2–VASc score of 1 taking bleeding risk into account. Bleeding Risk assessment: Use the HAS-BLED score to assess the risk of bleeding in people who are starting anticoagulation. When discussing the benefits and risks of anticoagulation people should be told:

for most people the benefit of anticoagulation outweighs the bleeding risk

for people with an increased risk of bleeding, the benefit of anticoagulation may not always outweigh the bleeding risk, and careful monitoring of risk is important.

By treating more patients appropriately with anticoagulation, the incidence and cost of serious and non-serious bleeding will also increase, but the savings from stroke prevention far out way any increase. Use the HAS-BLED score to assess the risk of bleeding in people starting or have started anticoagulation and highlight, correct and monitor the following modifiable risk factors:

Uncontrolled hypertension

Poor control of INR (“labile INRs”) – exclude measurements taken in the first 6 weeks.

Concurrent medication, eg use of aspirin or NSAID

Harmful alcohol consumption

4

Anticoagulation

1. WARFARIN

Calculate the person’s time in therapeutic range (TTR) at each visit. Calculate TTR over period of at least 6 months (exclude measurements taken in the first 6 weeks.)

Reassess anticoagulation for a person with poor anticoagulation control shown by: o 2 INR values higher than 5 or 1 INR value higher than 8 within the past 6 months o 2 INR values less than 1.5 within the past 6 months o TTR less than 65%

Take into account and if possible correct the following factors that may contribute to poor control:

o Cognitive function o Adherence to prescribed therapy o Illness o Interacting drug therapy o Lifestyle factors including diet and alcohol

If poor control cannot be improved, evaluate the risks and benefits of alternative stroke prevention strategies and discuss them with the patient. (Appendix 3 may be helpful)

2. Offer NOAC in accordance with NICE TA and local priority triangle

3. Do not offer aspirin monotherapy solely for stroke prevention to people with AF

4. Review the need for anticoagulation and quality of control at least annually, or more

frequently if clinically relevant.

5. For people not taking an anticoagulant because of bleeding risk or other factors, review stroke and bleeding risk annually and ensure all reviews and decisions are documented.

Choices of anticoagulant include warfarin and three new oral anticoagulants (NOACs), dabigatran, rivaroxaban and apixaban. At present there are no head-to-head trials comparing the relative efficacy and safety of apixaban, rivaroxaban and dabigatran. Individual drug characteristics can be found at appendix 4.

Antiplatelets such as aspirin and clopidogrel are only needed for patients who do not want, or cannot have, oral anticoagulants. Antiplatelets have only a small effect in protecting against stroke, but have a similar risk to OACs of severe bleeding, including intracranial haemorrhage. Choice of anticoagulant offered should reflect current evidence, benefit and patient preference in context of the East Berkshire health economy, where the maximum health benefit can be obtained for the population from the treatments available.

5

Berkshire East CCGs: Priority criteria for anticoagulation for stroke prevention in AF (figures based on approximately 5000 patients on AF register in East Berkshire). Nationally it is estimated that 15% of patients are unable to take warfarin due to allergy (Appendix 5) or inability to manage monitoring. Up to 40% of patients with AF are not controlled adequately within therapeutic range (TTR) which is highly centre dependent, and approximately 45% of people with AF receive no anticoagulation. High Priority: Warfarin allergy or inability to monitor Consider NOAC as high priority. Use of NOACs rather than warfarin in this group of about 207 patients is estimated to cost about £166,000 per year and may save up to 7 strokes annually. Cost per stroke saved £23,714 Medium Priority (higher): Out of range despite warfarin compliance (<65% TTR) Consider NOAC as medium (higher) priority. Use of NOACs rather than warfarin in this group of between 252 and 502 patients is estimated to cost between £141,000 and £282,000 per year and may save up to 8 – 16 strokes annually. Cost per stroke saved £17,625 Medium Priority (lower): Diagnosed AF patients currently prescribed aspirin or nothing Consider warfarin or NOAC (Discussion with patient using counselling sheet may be useful). Use of NOACs rather than warfarin in this group of about 1,257 patients is estimated to cost between £505,000 and £1,010,000 and may save up to 20 - 40 strokes annually. Cost per stroke saved £25,250 Low Priority: Newly diagnosed AF patients and warfarin stable patients. Switching warfarin stable patients to NOAC is low priority. Use of NOACs rather than warfarin in these 2 groups of between 1,026 and 1,266 patients is estimated to cost between £572,000 and £712,000 and may save up to 3-5 strokes annually. Cost per stroke saved between £142,400 and £190,667 In order to record patient discussion regarding choice of treatment, Reason for NOAC initiation (record of decision making process) document can be found at Appendix 6

6

Cardioversion: Warfarin is given for 3 - 4 weeks before cardioversion and for at least 4 weeks afterwards to minimise the chance of having a stroke. In patients with AF of confirmed duration of less than 48 hours undergoing cardioversion, anticoagulation following successful restoration of sinus rhythm is usually not required. If cardioversion is successful, warfarin may be stopped, although anticoagulation may be continued long term in patients with AF who have undergone cardioversion where there is a high risk of AF recurrence or where it is there is a high stroke risk defined as:

previous ischaemic stroke/TIA or thromboembolic event

age =75 with hypertension, diabetes or vascular disease (coronary artery disease or peripheral artery disease)

clinical evidence of valve disease, heart failure, or impaired LV function on echocardiography

Patients awaiting DC cardioversion or urgent AF ablation Warfarin is the only anticoagulant licensed for this indication, although patients can stay on dabigatran or apixaban while being cardioverted (Summaries of Product Characteristics). If anticoagulation is required long term post cardioversion (regardless of rhythm), the decision about whether to continue NOAC treatment or convert to warfarin should be made after an informed discussion between the clinician and the patient about the risks and benefits of anticoagulants, and in particular NOACs compared with warfarin. (Refer to Patient Counselling Sheet – New Oral Anticoagulants) Falls: A patient with an annual stroke risk of 5% (CHADS2 score 2-3) would need to fall 295 times for fall risk to outweigh stroke reduction benefit of warfarin or other anticoagulant. Do not withhold anticoagulation solely because the person is at risk of having a fall.

7

Post myocardial Infarction patients who have other conditions – anticoagulation. (NICE clinical guideline 172: November 2013). Take into account bleeding, thromboembolic and cardiovascular risks when thinking about treatment for people who have had an MI and who have an indication for anticoagulation: Unless there is a high risk of bleeding, continue anticoagulation and add aspirin to treatment in people who have had an MI who otherwise need anticoagulation and who: - have had their condition managed medically or - have undergone balloon angioplasty or - have undergone CABG surgery. Continue anticoagulation and add clopidogrel to treatment in people who have had an MI, who have undergone percutaneous coronary intervention with bare-metal or drug-eluting stents and who otherwise need anticoagulation. Offer clopidogrel with warfarin to people with a sensitivity to aspirin who otherwise need anticoagulation and aspirin, and who have had an MI. Do not routinely offer warfarin in combination with prasugrel or ticagrelor to people who need anticoagulation who have had an MI. After 12 months post MI, continue anticoagulation and take into consideration the need for ongoing antiplatelet therapy, taking into account all of the following: - Indication for anticoagulation - Thromboembolic risk - Bleeding risk - Cardiovascular risk - Patient’s wishes. Consider using warfarin and discontinuing treatment with a new oral anticoagulant (rivaroxaban, apixaban or dabigatran) in people who otherwise need anticoagulation and who have had an MI, unless there is a specific clinical indication to continue it. Do not add a new oral anticoagulant (rivaroxaban, apixaban or dabigatran) in combination with dual antiplatelet therapy in people who otherwise need anticoagulation, who have had an MI.

8

Anticoagulation INR Services It is important to understand the detail and full cost of anticoagulation services provided, particularly as the NOACs are often presented as enabling savings to be made in clinic and pathology costs. What are the current costs? Up until 2014, anticoagulation and pathology services were part of a block contract therefore reducing activity did not reduce costs. Additionally, GP practices can provide anticoagulation services under a LES. From April 2014, the block contract has been broken down and is charged as follows. INR Hospital clinic: £30 per appointment + £7 test INR GP service: £120 per patient per year (Test strips provided by CCG) Estimated total annual cost per patient: Hospital INR clinic £450 GP level 4 INR service £170 NOAC cost £767 - £802 (+ additional tests as appropriate according to SPC) Does East Berkshire need a local INR anticoagulation service? An INR service is not only for non valvular AF; it has many other indications where NOACs are not suitable or desirable (e.g. undesirable side effects, pregnancy, haematology, cancer, valvular patients). If the core business of warfarin INR testing for AF was to be quickly and substantially reduced, unless actively managed, there is the potential for the cost per patient to substantially increase.

9

Appendix 1 – Sample patient letter Dear Patient Due to an update to both national and local cardiac guidelines, we have been reviewing our patients who have a diagnosis of atrial fibrillation. Atrial fibrillation means the heart does not beat in a regular manner, which is called an arrhythmia. One in five of all strokes is attributed to this arrhythmia and those patients who survive are left more disabled by their stroke and more likely to suffer a recurrence than patients with other causes of stroke. For each patient we are able to assess the long term risk of stroke which depends on a number of factors. The risk of stroke is reduced by taking a tablet called an anticoagulant which thins the blood and reduces the risk of clotting. You currently take ASPIRIN and/or CLOPIDOGREL. This/these are antiplatelet agents which reduce the risk of stroke by about 20%. Until now the most effective anticoagulant has been warfarin which can reduce the risk of stroke by over 60%, but there are now other agents available which have a similar affect. We invite you to make a non urgent appointment where we can discuss your risk of stroke versus the risks and benefits of available treatments Yours sincerely

Appendix 2 – Detail Aid for anticoagulation treatment decisions

Stroke Prevention in Atrial Fibrillation: Warfarin is highly effective at reducing stroke risk but is greatly under used.

AF as a cause of stroke

National costs

The annual risk of stroke is 5-6

times greater in AF patients than in

people with normal rhythm.

19% patients presenting with stroke

are in AF. This equates to 16,000

strokes in England of which 12,000

are thought to be directly

attributable to AF.

Warfarin is highly effective in

preventing stroke in AF (NNT 25),

reducing risk of stroke by 64%

compared to placebo. Aspirin

reduces stroke risk by 22%.

Only 12 patients at high risk of

stroke need to be treated with

warfarin for 1 year to prevent 1

stroke, compared to 40 patients

taking aspirin.

There is no statistical difference in

the risk of major haemorrhage for

warfarin and aspirin in elderly

patients (1.9% vs 2.0%) (BAFTA).

If annual stroke risk is >2.0%/year

the benefit of warfarin exceeds

bleeding risk as long as well

controlled (TTR>60%).

NICE 2005 guidance on AF

concluded that 46% of patients who

should have been receiving warfarin

were not.

Assessing stroke risk in AF patients

Assess CHADS2 SCORE FIRST

CHADS2 Score Stroke Risk % p.a.

0 1.9

1 2.8

2 4.0

3 5.9

4 8.5

5 12.5

6 18.2

Risk Stratification

CHA2DS2-

VASc

Stroke Risk %

p.a.

0 0

1 1.3

2 2.2

3 3.2

4 4.0

5 6.7

6 9.8

7 9.6

8 6.7

9 15.2

If CHADS2 = 0 or 1 or uncertain of risk

USE CHADS2-VASC

GRASP-AF Query and risk stratification tool is FREE and available for use with

all GP clinical systems in England.

GRASP-AF provides a set of MIQUEST queries to identify, for your practice, patients

with a diagnosis of AF who are not on warfarin.

It calculates their risk of stroke using the validated CHADS2 scoring system and

highlights patients with a CHADS2 score of 2 or more who are not on warfarin and

would benefit from a review to assess anticoagulation.

To access the tool contact the Medicines Optimisation Team or find more information

via http://www.primis.nottingham.ac.uk/index.php/services/information/grasp-suite

HAS-BLED Major Bleeding Risk Score Letter Clinical Characteristic Points HAS_BLED Score

(total points)

Major Bleed Risk % p.a.

H Hypertension 1 0 1.13

A Abnormal renal & liver

function

1 or 2 1 1.02

S Stroke 1 2 1.88

B Bleeding diathesis 1 3 3.74

L Labile INR 1 4 8.70

E Elderly 1 5 to 9 Insufficient t data

D Drugs/Alcohol 1 or 2

Treatment

Warfarin should be considered first line for all patients including the elderly unless absolutely

contraindicated. Contraindications include recent life threatening haemorrhage, pregnancy, and known

hypersensitivity to warfarin.

(See BNF for complete list)

Target INR 2.5 (range 2-3)

Assess bleeding risk using HAS-BLED (see below) and exercise caution and review regularly if score >3.

In a very few patients and on specialist advice, concomitant use of antiplatelet agents and warfarin may be

appropriate.

Falls are not a major risk factor for bleeding in anticoagulated patients A patient with an annual stroke

risk of 5% (CHADS2 score 2-3) would need to fall 295 times for fall risk to outweigh stroke reduction

benefit of warfarin.

Use of new oral anticoagulants (NOACs)

High Priority: Allergy to warfarin or inability to monitor.

Medium priority: Patients on aspirin or nothing & patients out of range despite warfarin compliance

Low Priority: warfarin stable & new patients

HAS-BLED

Hypertension: systolic BP > 160mmHg:

Renal function: creatinine >200 or

dialysis

Liver function: chronic liver disease

(e.g. cirrhosis) >2x ULN + ASP/ALKP

> 3x upper limit normal)

Bleeding: previous bleeding, bleeding

diathesis or unexplained anaemia.

Labile INR. TTR <65% (over 6m).

Drugs: concomitant drugs e.g.

antiplatelet agents and NSAIDs.

Alcohol: excess alcohol

Risk category CHA2DS2-

VASc Score

Recommended Stroke

Prevention Therapy

High ≥2 Offer oral anticoagulation

(OAC)

Moderate (age

>65 & vascular

disease)

1 Consider OAC for men

.

No risk factors

and age < 65

0 No therapy

11

Risk Stratification. CHADS2 Scoring Scheme

CHA2DS2-VASc Scoring Scheme Total stroke (haemorrhagic & ischaemic) reduced by approximately 64% by warfarin (NNT 25) & 22% by aspirin (Relative risk reduction) The total risk of major bleeds is increased by about 50% with warfarin and about 17% with aspirin. Degree of Risk of Stroke CHADS2

score Risk of stroke

Antiplatelet (aspirin or clopidogrel) Treatment

Anticoagulant (warfarin) treatment

Low Risk Baseline risk 10 per 1000/12m

- 1% 2.2 in 1000 “saved”

6.4 in 1000 “saved”

Low Risk Baseline risk 19 per 1000/12m

0 1.9% 4.2 in 1000 “saved”

12.2 in 1000 “saved”

Moderate Risk Baseline risk 28 per 1000/12m

1 2.8% 6.2 in 1000 “saved”

17.9 in 1000 “saved”

Moderate Risk Baseline risk 40 per 1000/12m

2 4.0% 8.8 in 1000 “saved”

25.6 in 1000 “saved”

High Risk Baseline risk 59 per 1000/12m

3 5.9% 13 in 1000 “saved”

37.8 in 1000 “saved”

High Risk Baseline risk 85 per 1000/12m

4 8.5% 18.7 in 1000 “saved”

54.4 in 1000 “saved”

High Risk Baseline risk 125 per 1000/12m

5 12.5% 27.5 in 1000 “saved”

80 in 1000 “saved”

High Risk Baseline risk 182 per 1000/12m

6 18.2% 40 in 1000 “saved”

116.5 in 1000 “saved”

HARMS: Degree of risk of major extra cranial bleed

HAS-BLED score

Risk of bleed

Aspirin Treatment

Warfarin treatment

Baseline risk of 11.3 per 1000/12m 0 1.13% 1.9 in 1000 “bleed”

5.65 in 1000 “bleed”

Baseline risk of 10.2 per 1000/12m 1 1.02% 1.7 in 1000 “bleed”

5.1 in 1000 “bleed”

Baseline risk of 10 per 1000/12m 2 1.00% 1.7 in 1000 “bleed”

5 in 1000 “bleed”

Baseline risk of 37.4 per 1000/12m 3 3.74% 6.4 in 1000 “bleed”

18.7 in 1000 “bleed”

Baseline risk of 87 per 1000/12m 4 8.70% 14.8 in 1000 “bleed”

43.5 in 1000 “bleed”

Feature Score

C Congestive heart failure 1

H Hypertension 1

A Age > 75 years 1

D Diabetes Mellitus 1

S2 Prior Stroke or TIA 2

Feature Score

C Congestive Heart Failure or LVEF ≤ 40% 1

H Hypertension 1

A2 Age >75 years 2

D Diabetes mellitus 1

S2 Stroke/TIA/Thromboembolism 2

V Vascular disease (previous MI, peripheral arterial disease or aortic plaque)

1

A Age between 65 and 74 years 1

Sc Female 1

12

Key points regarding risk stratification. CHADS2 scoring is used as a guide to assessing individual stroke risk, however, this score does not discriminate particularly well between very low-risk patients and intermediate-risk patients, therefore use CHADS2 VASc if uncertain of risk or if CHADS2 is 0 or 1.

Congestive heart failure must be moderate to severe to count.

Female sex is an independent risk factor in the presence of other risk factors, but not on its own. This means that female patients aged <65 years with lone AF (CHADS2 VASc 1) should not need to be offered an OAC. In the presence of other risk factors, such as hypertension and diabetes, female sex does confer additional risk.

Male patients with a CHADS2 VASc score of 1 should normally be considered for an OAC.

Vascular disease includes the presence of coronary disease, as well as peripheral vascular, aortic and carotid disease.

Patients with severe renal failure and AF have not been well studied. They have an increased risk of stroke as well as bleeding, but whether they should be treated in the same way as other patients is not known.

Bleeding risk

OACs are successful in reducing stroke risk by up to 60-70% in patients with AF (higher time in TTR for warfarin equates to a greater reduction in stroke risk), but their use will always be tempered by concerns about severe bleeding, particularly intracranial haemorrhage.

Net clinical benefit analysis of OACs, balancing the reduction in stroke risk against risk of intracranial haemorrhage, has found that most categories of patient with AF benefit from OACs, except the lowest stroke risk group. This is because patients with a high bleeding risk also tend to have an even higher stroke risk and therefore continue to accrue a net overall benefit on an OAC.

HAS-BLED is used to assess bleeding risk in patients with AF and helps to focus attention on modifiable risk factors for bleeding and to encourage surveillance in higher-risk patients. It should not be used as a reason for not offering OACs to patients who might still benefit from them.

So, for example, tighter management of hypertension, avoidance of NSAIDs or closer monitoring of renal function or INR might result from the decision to anticoagulate a patient who has high scores for both CHADS2 VASc and HAS-BLED.

Concomitant aspirin with an anticoagulant was the most important modifiable independent risk factor for intracranial haemorrhage. NOACs have all been associated with slightly lower absolute rates of intracranial haemorrhage, but also a possible increase in gastrointestinal bleeding.

A HAS-BLED score of 3 or more is considered to reflect a high risk of severe bleeding on OACs and use of OACs in these patients is not contraindicated, but should be approached with caution.

13

Appendix 3 Patient Counselling Sheet – New Oral Anticoagulants The decision about whether to start treatment with a NOAC should be made after an informed discussion between the clinician and the patient about the risks and benefits of anticoagulants, and in particular NOACs compared with warfarin. For people who are taking warfarin, the potential risks and benefits of switching to a NOAC should be considered in light of their level of international normalised ratio (INR) control. NOACs as a cost-effective choice of anticoagulant is highly dependent upon drug cost, anticoagulation control for warfarin, and anticoagulation service cost. NOACs therapy was found to be cost-effective when time in therapeutic range (TTR) of warfarin therapy was 60 % or below.2 Good anticoagulation control is associated with a greater reduction in the risk of stroke. In the large ACTIVE W clinical trial, it was found that the benefits of oral anticoagulation therapy were most pronounced in patients in centres with a mean % of time spent within therapeutic range above 65%. 3 Patients who spent at least 70% of time within therapeutic range had a 79% reduced risk of stroke compared to patients with 30% of time in range. In NOAC trial data, mean time in treatment range varied considerably and importantly the relationship between events and INR control was not reported. We do not know how poor control may have affected reported event rates.

Warfarin: Time in Therapeutic Range (TTR) [% and (mean TTR)] vs Event Rates (%/yr)4

Poor < 60% (48%) n = 1190

Average. 60 – 75% (69%) n = 1207

Good > 75% (83%) n = 1190

Combined (estimated) n = 3587

Stroke + Systemic Embolic Event 2.1 1.34 1.07 1.5

Major Bleed 3.85 1.96 1.58 2.46

Mortality 4.2 1.84 1.69 2.58

Myocardial Infarction 1.38 0.89 0.62 0.96

Total 11.53* 6.03 4.96* 7.5

*Number needed to treat for 1 year to prevent 1 event would be 15 patients in “good” vs “poor” control groups.

Discontinuation of warfarin & dabigatran in RE-LY was 10% and 17% of warfarin-treated patients stopping warfarin at 1 and 2 years, respectively. The rates of discontinuation of dabigatran were higher at 15% and 21% at 1 and 2 years, respectively.

Warfarin NOAC

Tried and tested treatment. Local expertise is high

New drug. Local and national experience is comparably limited

Regular monitoring required, which is widely available, accessible and expertise is high

Occasional safety monitoring needed

Compliance with treatment is measurable Compliance with treatment is not measurable

Variable dose Standard dose

Antidote widely available No readily available antidote: trauma or emergency surgery, and overdose more hazardous. Product information should be consulted for advice in the event of bleeding complications, or overdose.

Many drug / food interactions, but well understood Limited information on drug / food interactions

Narrow therapeutic index – may be times when patient is sub therapeutic or supra therapeutic

Wide therapeutic index

Dabigatran, rivaroxaban and apixaban have all been associated with slightly lower rates of intracranial haemorrhage but also a possible increase in gastrointestinal bleeding5.

Appendix 4 Consult individual SPCs for full details and complete list of interactions Drug Warfarin Rivaroxaban Dabigatran Apixaban

Dose Variable dose once a day generally at teatime. Dose dependent on INR and target range

20mg taken orally once a day (taken with food) In patients with moderate (creatinine clearance 30 - 49 ml/min) or severe (creatinine clearance 15 - 29 ml/min) renal impairment the following dosage recommendations apply: the recommended dose is 15 mg once daily (see section 5.2).

150mg taken orally twice a day (under 75 year olds)

5 mg taken orally twice a day

150mg taken orally twice a day (75-80 year olds) or if bleeding risk is high and thromboembolic risk low consider 110mg BD

2.5 mg taken orally twice a day in patients with at least two of the following characteristics: age ≥ 80 years, body weight ≤ 60 kg, or serum creatinine ≥ 1.5 mg/dl (133 micromol/l) 110mg twice a day (over 80 year olds)

Patients who receive concomitant verapamil

Monitoring INR monitoring Visible monitoring of patient for signs of bleeding and anaemia

Visible monitoring of patient for signs of bleeding and anaemia Renal function should be assessed by calculating the CrCL prior to initiation of treatment to exclude patients with CrCL< 30 mL/min Renal function should also be re-assessed at least annually

Visible monitoring of patient for signs of bleeding and anaemia

Contra-indications See SPC for full details

Haemorrhagic stroke

Clinically significant bleeding

Within 72 hours of major surgery with risk of severe bleeding

Within 48 hours postpartum

Clinically significant active bleeding.

Hepatic disease associated with coagulopathy and clinically relevant bleeding risk

Condition leading to significant risk of major bleeding

Patients with severe renal impairment (CrCL< 30 ml/min)

Active clinically significant bleeding

Organic lesion at risk of bleeding

Spontaneous or pharmacological impairment of haemostasis

Hepatic impairment or liver disease expected to have any impact on survival

Patients with prosthetic heart valves requiring anticoagulant treatment

Clinically significant active bleeding.

Hepatic disease associated with coagulopathy and clinically relevant bleeding risk

Condition leading to significant risk of major bleeding

Cautions See SPC for full details

Thyroid disorders – closely monitor patients The following may ↑bleeding risk: weight loss, acute illness, smoking cessation (doses may need to be reduced)

Renal impairment: Moderate: (CrCl 30-49ml/min) – 15mg ONCE A DAY Severe: (CrCl 15-29ml/min) – 15mg ONCE A DAY (use cautioned in these patients) Very severe: CrCl<15ml/min – not

Patients with elevated liver enzymes > 2 ULN should not be prescribed dabigatran

Renal impairment:

Very severe: CrCl<15ml/min – not recommended in this group of patients

The following may reduce efficacy: weight gain, D & V (doses may need to be ↑)

recommended in this group of patients N.B. Safety and efficacy of rivaroxaban has not been studied in patients with prosthetic heart valves; therefore, there are no data to support that rivaroxaban 20 mg (15 mg in patients with moderate or severe renal impairment) provides adequate anticoagulation in this patient population and is therefore not recommended for these patients

Patients with elevated liver enzymes > 2 ULN should be treated cautiously

Drug interactions. See individual SPC

Avoid or use with caution the following: NSAIDs Heparins Thrombin inhibitors Drugs inhibiting haemostasis, clotting or platelet action Azole antifungals SSRI/SNRI Glucosamine Statins Fibrates Methylphenidate Zafirlukast

Interactions below lead to an ↑ bleeding risk – patients should be monitored closely for signs of bleeding and anaemia. Concomitant use of systemic azole antimycotics e.g. ketoconazole, itraconzaole, voriconazole, posaconazole and HIV protease inhibitors e.g. ritonavir Drugs inhibiting one of the rivaroxaban elimination pathways, either CYP3A4 or P-gp, are expected to increase rivaroxaban plasma concentrations to a lesser extent e.g. clarithromycin, erythromycin. Caution with ketoconazole Concomitant use of dronedarone should be avoided Caution with concomitant other anticoagulants Caution with concomitant NSAIDs and platelet aggregation inhibitors (e.g. clopidogrel) Interactions below lead to a ↓ in anticoagulant concentration therefore treatment may be

Generally interactions lead to an ↑ bleeding risk – patients should be monitored closely for signs of bleeding and anaemia. Caution with concomitant other anticoagulants Systemic azole-antimycotics Caution with concomitant strong P-gp inhibitors e.g. amiodarone, quinidine, verapamil. For patients who receive concomitant verapamil the recommended daily dose of dabigatran is 220 mg taken as one 110 mg capsule twice daily. Caution concomitant use of amiodarone (amiodarone has a long half-life and interaction may exist for weeks after discontinuation, especially in patients with mild-moderate renal impairment) Caution concomitant use of quinidine especially in patients with mild-moderate renal impairment The following strong P-gp inhibitors are contraindicated: systemic ketoconazole, cyclosporine, itraconazole and dronedarone (see section 4.3). Concomitant treatment with tacrolimus is not recommended.

Concomitant use of

azole-antimycotics& HIV protease inhibitors ↑ plasma levels as dodiltiazem, naproxen, amiodarone, verapamil, quinidine

Inducers of CYP3A4 and P-gp ↓plasma levels of apixaban e.g. include rifampicin,phenytoin, carbamazepine, phenobarbital or St. John's Wort

Co administration with anticoagulants, platelet aggregation inhibitors and NSAIDs↑risk of bleed

suboptimal Caution concomitant use of strong CYP3A4 inducers e.g. phenytoin, cbz, phenobarbital, St. John’s Wort)

Other key points

Adherence to therapy can be checked by INR monitoring

Half-life 5-13 hours means any missed doses can lead to loss of anticoagulation

Half-life 12-14 hours means any missed doses can lead to loss of anticoagulation. Use of fibrinolytic medicinal products for the treatment of acute ischemic stroke The use of fibrinolytic medicinal products for the treatment of acute ischemic stroke may be considered if the patient presents with a dTT, ECT or aPTT not exceeding the ULN according to the local reference range.

Half-life 12 hours means any missed doses can lead to loss of anticoagulation

Annual prescribing cost

Approximately £15 £767 £802 £802

17

Appendix 5 – Contraindications to the initiation of oral anticoagulants and anti platelets in patients with atrial fibrillation in primary care As a patient’s relative stroke & bleeding risk can change, it is essential that all AF patients are reviewed at least annually for a re-assessment of their stroke versus bleeding risk & the anti-thrombotic treatment option of choice. Contraindications listed below apply to BOTH anti-platelet agents (e.g. aspirin, clopidogrel, dipyridamole) & ALL oral anticoagulants (e.g. warfarin, phenindione, dabigatran, rivaroxaban) except where indicated. N.B. Poor compliance with any oral anticoagulant agent will reduce benefits and may increase risks associated with use. Absolute Contraindications

Known large oesophageal varices.

Significant thrombocytopenia (platelet count < 50 x 109/L) - refer to haematologist.

Within 72 hours of major surgery with risk of severe bleeding - defer & reassess risk postoperatively.

Previously documented hypersensitivity to either the drug or excipients – consider cardiology

opinion.

Acute clinically significant bleed - defer & re-assess stroke versus bleeding risk within 3 months.

Decompensated liver disease or deranged baseline clotting screen (INR>1.5) – refer to Gastroenterology /Hepatology. Contraindication applies to oral anticoagulants only

Pregnancy or within 48 hours post partum - seek urgent haematological advice. Contraindication applies

to oral anticoagulants only. Severe renal impairment (GFR < 30 mL/min/1.73 m2 or on dialysis). Contraindication applies to

dabigatran only. Relative Contraindications

Previous history intracranial haemorrhage - as some AF patients especially those considered at higher

stroke risk (i.e. CHADS2 score ≥3) may benefit from anti-thrombotic therapy, seek the opinion of a stroke specialist. Recent major extracranial bleed within the last 6 months where the cause has not been

identified or treated – decision for oral anti-thrombotic therapy should be deferred.

Recent documented peptic ulcer (PU) within last 3 months– decision for oral anti-thrombotic therapy

should be deferred until treatment for PU completed. In all cases with history PU give PPI cover whilst on anti-thrombotic. Recent history recurrent iatrogenic falls in patient at higher bleeding risk.

A patient at higher bleeding risk is assessed by having 3 or more of the following risk factors:-

Known large oesophageal varices. age > 65 years previous history bleed or predisposition to bleeding (e.g. diverticulitis) uncontrolled hypertension severe renal impairment (i.e. serum creatinine > 200umol/L, GFR < 30 mL/min/1.73 m2 or

on dialysis) acute hepatic impairment (e.g. bilirubin > 2xULN + LFTS > 3x ULN), chronic liver disease

(e.g.cirrhosis) low platelet count < 80 x 109/L or a thrombocytopenia or anaemia of undiagnosed cause on concomitant drugs associated with an increased bleeding risk e.g. SSRIs, oral steroids,

NSAIDs, methotrexate or other immune-suppressant agents.

Dementia or marked cognitive impairment with poor medicines compliance & no access to carer support.

Chronic alcohol abuse – especially if associated with binge drinking. Poor compliance with any oral anticoagulant agent will reduce benefits and may increase risks associated with use.

18

This information has been drawn from clinical opinion together with established documented clinical evidence where available. Key Supporting References European Society of Cardiology (ESC) Guidelines for the management of atrial fibrillation. Eur Heart J. Aug 29 2010. http://www.escardio.org/guidelines-surveys/esc-guidelines/GuidelinesDocuments/guidelines-afib-FT.pdf (Recommendation that selection of anti-thrombotic therapy should be based upon the absolute risks of stroke/thrombo-embolism and bleeding and the relative risk and benefit for a given patient. Highlights the use of the ‘ HAS-BLED’ bleeding risk score as a tool to assess bleeding risk in AF patients.) Keeling D, Baglin T et al; Guidelines on oral anticoagulation with warfarin – 4th Ed 2011; British Journal of Haematology 1365-2141. http://www.bcshguidelines.com/documents/warfarin_4th_ed.pdf (Latest updated BCSH guidance- includes statement re concomitant use of anticoagulants & anti-platelets) Mant J Hobbs FDR, et al, Warfarin versus aspirin for stroke prevention in an elderly community population with atrial fibrillation (AF),(BAFTA RCT study): Lancet 2007;370: 493-503. (BAFTA study showed clear superiority of warfarin over aspirin with no increase in risk of major haemorrhage. Mean age of population was 81.5 years.) Bailey RD, Hart RG, Benavante O, Pearce LA. Recurrent brain haemorrhage is more frequent than ischaemic stroke after intracranial haemorrhage. Neurology 2001;56:773-7. (Recurrent stroke among survivors of primary intracranial haemorrhage (ICH) occurs at a rate of about 4% per patient year and most are recurrent ICH. Survivors of ICH likely to have a higher risk of recurrent ICH than of ischaemic stroke with CHADS 2 score < 3). (Adjusted annual stroke rate risk with CHADS2 2 score 3 is 5.9%) Man-Son Hing M, Nichol G, Lau A, Laupacis A. Choosing antithrombotic therapy for elderly patients with atrial fibrillation who are at risk of falls. Arch Intern Med 1999; 159: 677-85. (Showed a calculated risk of a subdural haemorrhage from falling in patients with annual stroke risk 5% would require a patient to fall 295 times for falls risk to outweigh stroke reduction benefit of warfarin.) The ACTIVE Writing Group on behalf of the ACTIVE investigators. Lancet 2006;367:1903–12 (Study found incidence of bleeding was significantly greater with aspirin + clopidogrel compared with warfarin) (19.3% vs. 16.5%; NNH 35; RR=1.21, 95% CI 1.08–1.35, P=0.001). Wehinger C, Stollberger C, Lamger T et al. Evaluation of risk factors for stroke/embolism & of complications due to anticoagulant therapy in atrial fibrillation. Stroke 2001;32(10):2246-2252 (Study found significant difference in bleeding complications between those patients prescribed at least three additional medicines & those prescribed less than three.) Shireman TI, Howard PA, Kresowik TF et al. Combined anticoagulant –antiplatelet use and major bleeding events in elderly atrial fibrillation patients. Stroke.2004:35(10)2362-2367. (Found history of bleeding to be a significant independent predictor of future bleeding events.) PROGRESS Collaborative Group. RCT of a perindopril-based BP-lowering regimen among 6,105 individuals with previous stroke or transient ischaemic attack. Lancet. 2001 Sep 29;358(9287):1033-41 (study showed importance of BP control in patients with cerebrovascular disease in significantly lowering risk of first ICH.) Laupacis A, Boysen G, Connolly S et al. Risk factors for stroke & efficacy of antithrombotic therapy in AF: analysis of pooled data from five randomised controlled trials. Archives of Internal Medicine. 1994:154(13):1449-1457. (Study found both systolic and diastolic BP to be significantly higher in those patients with bleeding complications than in those without bleeding complications.) Dite PD. Labrecque D et al. World Gastroenterology Organisation Practice Guideline Oesophageal Varices 2008. (Oesophageal varices develop in patients with cirrhosis at an annual rate of 5–8%, but varices large enough to pose a risk of bleeding occur in only 1–2% of cases. Approx 4–30% of pts with small varices will develop large varices each year & therefore be at risk of bleeding. Mortality resulting from bleeding depends on the severity of the underlying liver disease.) UK Teratology Information/Toxicology database report on use of Warfarin in Pregnancy. National Poisons Information service commissioned by HPA. March 2011 http://www.toxbase.org Summary Product Characteristics for Marevan (warfarin) Pradaxa (Dabigatran); Plavix (Clopidogrel). Electronic Medicines Compendium @ http://www.medicines.org.uk Acknowledgements NHS Buckinghamshire; Atrial Fibrillation Association and Dr Matthew Fay, GP & NHS Heart Improvement Clinical Lead and Dr Paul Guyler, Lead Stroke Consultant Southend University Hospital NHS Foundation Trust and Stroke Improvement Programme Associate, with contributions received from various clinical healthcare professionals in cardiology, neurology, haematology and gastroenterology.

19

Appendix 6

Reason for NOAC initiation (record of decision making process)

The decision about whether to start treatment with a NOAC should be made after an informed discussion between the clinician and the patient about the risks and benefits of anticoagulants, and in particular NOACs compared with warfarin.

PATIENT NAME: PATIENT DOB:

Please Tick

East Berkshire Priority criteria for NOAC use for stroke prevention in AF

High Medium Low

Patient has warfarin allergy

Patient has a warfarin specific contraindication

Patient unable to tolerate previously prescribed warfarin due to adverse effects

Patient unable to achieve satisfactory INR after adequate trial of warfarin despite compliance with therapy and correction of factors that may contribute to poor control. Excluding first 6 weeks of treatment and assessed over 6 months: state % TTR

INR ≥ 8.0 on 1 occasion or INR ≥ 5.0 on 2 occasions over a period of 6 months (once anticoagulation is established), with a high likelihood of recurrence i.e. warfarin deemed unsafe

Patient unable to comply with specific monitoring requirements of warfarin Please state reason:

Patient is aware of the benefits and risks of NOAC therapy

Y / N

Patient is aware there is no specific antidote to NOAC therapy

Y / N

Anticoagulation and MI: NICE November 2013. Consider using warfarin and discontinuing treatment with a new oral anticoagulant (rivaroxaban, apixaban or dabigatran) in people who otherwise need anticoagulation and who have had an MI, unless there is a specific clinical indication to continue it. Do not add a new oral anticoagulant (rivaroxaban, apixaban or dabigatran) in combination with dual antiplatelet therapy in people who otherwise need anticoagulation, who have had an MI. Patients awaiting DC cardioversion or urgent AF ablation: Warfarin is the only anticoagulant licensed for this indication, although patients can stay on dabigatran or apixaban while being cardioverted (Summaries of Product Characteristics). If anticoagulation is required long term post cardioversion (regardless of rhythm), the decision about whether to continue NOAC treatment or convert to warfarin should be made after an informed discussion between the clinician and the patient about the risks and benefits of anticoagulants, and in particular NOACs compared with warfarin. (Refer to Patient Counselling Sheet – New Oral Anticoagulants)

20

REFERENCES 1 CLOPIDOGREL + ASPIRIN AS AN ALTERNATIVE TO WARFARIN: Bruce Soloway, MD reviewing The ACTIVE Writing Group on

behalf of the ACTIVE Investigators. Lancet 2006 Jun 10. Verheugt FWA. Lancet 2006 Jun 10. 2 J Gen Intern Med. 2013 Oct 17. [Epub ahead of print]

Novel Oral Anticoagulants Versus Warfarin Therapy at Various Levels of Anticoagulation Control in Atrial Fibrillation-A Cost-Effectiveness Analysis. 3 Connolly SJ, Pogue J, Eikelboom J, et al.; ACTIVE W Investigators. Benefit of oral anticoagulant over antiplatelet therapy in atrial

fibrillation depends on the quality of international normalized ratio control achieved by centers and countries as measured by time in therapeutic range. Circulation 2008; 118: 2029–2037. 4 White HD, Gruber M, Feyzi J, Kaatz S, et al. Comparison of outcomes among patients randomized to warfarin therapy according to

anticoagulation control; results from SPORTIF III and V. Arch Intern Med 2007; 167:239-245. NICE Clinical Guideline AF 2014 Warfarin versus aspirin for stroke prevention in an elderly community population with atrial fibrillation (the Birmingham Atrial Fibrillation Treatment of the Aged Study, BAFTA): a randomised controlled trialThe Lancet, Volume 370, Issue 9586, Pages 493 - 503, 11 August 2007 doi:10.1016/S0140-6736(07)61233-1 Gage BF,Waterman AD, ShannonW, Boechler M, RichMW, Radford MJ. Validation of clinical classification schemes for predicting stroke: results from the National Registry of Atrial Fibrillation. JAMA 2001;285(22):2864-70. Lip GY, Nieuwlaat R, Pisters R, Lane DA, Crijns HJ. Refining clinical risk stratification for predicting stroke and thromboembolism in atrial fibrillation using a novel risk factor-based approach: the euro heart survey on atrial fibrillation. Chest. 2010 Feb;137(2):263-72. European Heart Rhythm Association; European Association for Cardio-Thoracic Surgery, Camm AJ, Kirchhof P, Lip GY, Schotten U, Savelieva I, Ernst S, Van Gelder IC, Al-Attar N, Hindricks G, Prendergast B, Heidbuchel H, Alfieri O, Angelini A, Atar D, Colonna P, De Caterina R, De Sutter J, Goette A, Gorenek B, Heldal M, Hohloser SH, Kolh P, Le Heuzey JY, Ponikowski P, Rutten FH. 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 Oct;31(19):2369-429. Van Staa TP, Setakis E, Di Tanna GL, Lane DA, Lip GY (October 2010). A comparison of risk stratification schema for stroke in 79884 atrial fibrillation patients in general practice.. J Thromb Haemost.: no. doi:10.1111/j.1538-7836.2010.04085 5 Rates of GI belleding in RE-LY: Dabigatran 150mg 1.56% - p value 0.001 vs warfarin; dabigatran 110mg 1.15% - p

value 0.052 vs warfarin. 150mg dose had increase in GI bleeding vs warfarin; 110mg dose was non inferior.