The risk of ischemic stroke is increased 5-fold in patients with atrial fibrillation (AF).1 While this risk has been recognized for more than 20 years, several therapeutic options available to reduce the risk of stroke in patients with nonvalvular AF are relatively new. For the past several years, therapeutic options for lowering the risk of stroke in these patients consisted of warfarin and antiplatelet agents, which have a relative risk reduction of approximately 60% and 20%, respectively.2
In recent years, 3 new oral anticoagulants (OACs) (ie, apixaban,* dabigatran, and rivaroxaban) have been introduced for this indication. However, because these novel agents have only been studied in select patient populations, many questions remain: Should these drugs be selected over warfarin, a medication with which we have almost 60 years of experience? Do the potential benefits of these new agents outweigh the unknowns? At what point should these new drugs be started as first-line therapy?
The objective of this article is to review the OACs recommended for stroke prevention in patients with AF and provide clinicians with a systematic, practical approach for weighing the risk of stroke versus the risk of bleed in patients with nonvalvular AF.
Case description
A 62-year-old man, Mr G.R., presents to you complaining of having had mild palpitations for the past 2 weeks. At times he feels light-headed but he has never fainted. He denies any history of chest pain at rest or with exertion, orthopnea, and paroxysmal nocturnal dyspnea. His medical history includes hypertension (HTN), which is currently controlled with 160/25 mg of the valsartan-hydrochlorothiazide combination daily. He is a non-smoker and has an alcohol intake of 1 to 2 ounces of whiskey every week.
On physical examination he is in no obvious distress. Blood pressure is 134/82 mm Hg. His pulse is irregularly irregular at 110 beats per minute. He has no edema, jugular venous distention, or heart murmur on auscultation. There is no audible bruit, and his extremities are warm with palpable bilateral peripheral pulses. Results of his recent blood tests reveal normal complete blood count, thyroid and renal function, and liver enzymes. He has no history of bleeding involving transfusion or a decrease in hemoglobin greater than 20 g/L.
Electrocardiogram confirms AF with a heart rate of 120 beats per minute.
Bringing evidence to practice
The following stepwise approach can be used to tailor antithrombotic therapy to an individual. The risk of stroke for paroxysmal AF is similar for those who have persistent or permanent AF.3–5 The approach below is applicable to all 3 types of AF.
Step 1: Determine your patient’s risk of stroke using the CHADS2 (congestive heart failure, HTN, age ≥ 75 years, diabetes mellitus, prior stroke or transient ischemic attack [TIA]) score or the CHA2DS2-VASc (congestive heart failure; HTN; age ≥ 75 years; diabetes mellitus; stroke or TIA; vascular disease [prior myocardial infarction, peripheral artery disease, or aortic plaque]; age 65–74 years; sex category [ie, female]) score. The Canadian Cardiovascular Society (CCS), the European Society of Cardiology (ESC), and the American College of Chest Physicians guidelines recommend the CHADS2 score be used to predict risk of stroke and guide antithrombotic therapy in patients with AF owing to its simplicity, extensive validation, and widespread use (Table 1).3–5 Other risk factors for stroke should be considered for patients with CHADS2 scores of less than 2. The 2012 CCS AF guidelines list female sex, vascular disease, and age older than 65 years as additional stroke risk factors to consider.3 The 2010 ESC guidelines recommend the CHA2DS2-VASc score (Table 24) be subsequently calculated to further estimate the risk of stroke and guide therapy in lower-risk individuals.4
The CHADS2 score for estimating risk of stroke in patients with AF: Points are allocated by the CHADS2 risk criteria (congestive heart failure [symptoms in past 3 mo], 1 point; hypertension [diagnosis], 1 point; age ≥ 75 y, 1 point; diabetes mellitus, 1 point; stroke or TIA [prior], 2 points).
The CHA2DS2-VASc score for estimating risk of stroke in patients with AF: Points are allocated by the CHA2DS2-VASc risk criteria (congestive heart failure, 1 point; hypertension, 1 point; age ≥ 75 y, 2 points; diabetes mellitus, 1 point; stroke or TIA [prior], 2 points; vascular disease [prior MI, PAD, or aortic plaque], 1 point; age 65–74 y, 1 point; sex category [ie, female], 1 point).
The CHADS2 and CHA2DS2-VASc scores have similar ability to predict stroke.3–5 The CHADS2 score is easier to remember and use, but the CHA2DS2-VASc score is better at categorizing low- or intermediate-risk individuals.3–7
Step 2: Determine your patient’s risk of major bleeding using the HAS-BLED (HTN [systolic blood pressure > 160 mm Hg], abnormal renal or liver function, stroke [caused by a bleed], bleeding, labile international normalized ratio [INR], elderly [age > 65 years], drugs [acetylsalicylic acid (ASA) or nonsteroidal anti-inflammatory drugs] or alcohol [≥ 8 drinks/week]) score. Antithrombotic therapy increases the risk of both minor bleeding (eg, gingival bleeding, epistaxis) and major bleeding (eg, intracranial or gastrointestinal hemorrhage). When initiating an antiplatelet or anticoagulant agent for stroke prophylaxis, the efficacy of these agents must be balanced against the risk of major hemorrhage. Both the ESC and CCS AF guidelines recommend the HAS-BLED score be used for estimating the risk of major bleeds (Table 3).3,4
The HAS-BLED score for estimating risk of bleeding*: Points are allocated by the HAS-BLED criteria (hypertension [SBP > 160 mm Hg], 1 point; abnormal renal† or liver‡ function [1 point each], 1 or 2 points; stroke [caused by a bleed], 1 point; bleeding, 1 point; labile INR,§ 1 point; elderly [age > 65 y], 1 point; drugs [ASA or NSAIDs] or alcohol [≥ 8 drinks/wk] [1 point each], 1 or 2 points).
Step 3: Balance the risk of stroke versus the risk of bleeds. Collaborate with the patient to determine which antithrombotic, if any, is best suited for his or her needs. Oral anticoagulants (apixaban,* dabigatran, rivaroxaban, warfarin) are a recommended option for patients with a CHADS2 or CHA2DS2-VASc score of 1 or more.3–5 The CHADS2 and CHA2DS2-VASc risk factors that are assigned 1 point all increase the risk of stroke but differ in the degree of risk.3,5,6 As such, for patients with a CHADS2 or CHA2DS2-VASc score equal to 1, ASA might be an alternative to an OAC, depending on which risk factor is present and the patient’s preference.3–5 For example, ASA might be considered in patients with a single CHA2DS2-VASc point based on vascular disease or female sex.3 (Tables 13–5 and 24 summarize the recommendations of the guidelines for therapy in relation to the CHADS2 or CHA2DS2-VASc scores, associated recommendation strength, and levels of evidence.)
If a patient’s HAS-BLED score is greater than his or her CHADS2 or CHA2DS2-VASc score, antithrombotic therapy needs to be used cautiously with close monitoring and follow-up.3 A HAS-BLED score of 3 or more indicates the patient is at an increased risk of a major bleed.4 However, the effects of stroke are considerable, with up to 70% being either fatal or resulting in severe residual deficit.3 Major bleeding is less often fatal and is less likely to cause serious residual effects.3 In patients with a CHADS2 score of 2 or more, the benefit of using an anticoagulant to prevent stroke often outweighs the risk of bleeding while taking therapy.3–5,8,9
The preference for one OAC over another has been the subject of great debate. Based on current published evidence, the new OACs have been shown to be as good as or better than warfarin in stroke prevention.10–12 However, these findings apply only to the patient populations included in the trials (Table 4),10–16 and real-world experience with these new agents continues to offer insight into their possible advantages, disadvantages, effectiveness, etc. Guideline recommendations vary. The 2011 AF guidelines by the American College of Cardiology Foundation, American Heart Association, and Heart Rhythm Society and the 2010 ESC guidelines recommend dabigatran as an alternative to warfarin.4,17 In 2012, the ESC Working Group on Thrombosis stated all 3 new OACs are attractive alternatives to warfarin.18 The CCS guidelines state the new OACs are preferred over warfarin,3 and the American College of Chest Physicians guidelines recommend dabigatran over warfarin.5
Inclusion criteria, exclusion criteria, and highlights of landmark trials
Of note, these documents also list several exceptions where warfarin would be better suited (eg, patients with valvular heart disease, patients at risk of dyspepsia or gastrointestinal bleeding, patients well controlled with warfarin and who have no concerns regarding laboratory monitoring, patients with poor renal function, patients who meet exclusion criteria from landmark trials, and those concerned with direct medication cost).3–5,17,18
The Canadian Agency for Drugs and Technologies in Health recommends that the new OAC agents should only be used in patients who are unable to achieve adequate anticoagulation with warfarin and who have a CHADS2 score of 2 or more.19 The agency’s review on dabigatran highlighted the potential increase in myocardial infarction (MI) with dabigatran seen in the RELY trial. A post-hoc subgroup analysis concluded the risk of MI was smaller than originally thought; however, an association between dabigatran and MI cannot be ruled out.20
The RxFiles Trial Summaries of ARISTOTLE (Apixaban for Reduction in Stroke and Other ThromboemboLic Events in AF), RELY (Randomized Evaluation of Long-term Anticoagulation Therapy), and ROCKET-AF (Rivaroxaban Once Daily Oral Direct Factor Xa Inhibition Compared with Vitamin K Antagonist for Prevention of Stroke and Embolism Trial in AF) are available from CFPlus.† Table 410–16 provides summaries of these landmark trials pertaining to the new agents. Table 510–12,19,21–23 compares the OAC options available to reduce the risk of stroke in people with AF.
Warfarin compared with the new oral anticoagulants in AF
The decision as to which OAC to use for stroke prevention in AF should be individualized for each patient by looking at the current evidence, the risks and benefits for the patient, and the available local resources (eg, anticoagulation management services, provincial formulary status).
Step 4: Re-assess the patient’s risk of stroke and risk of bleeds annually or sooner if his or her risk criteria change, as AF is a chronic, recurrent, and progressive condition.
Back to Mr G.R
Now we will apply the stepwise approach to Mr G.R.
Step 1: Calculate Mr G.R.’s risk of stroke using the CHADS2 score (Table 13–5). Mr G.R.’s CHADS2 score is 1 (history of HTN). Because Mr G.R.’s CHADS2 score is less than 2, the CHA2DS2-VASc score should be calculated (Table 2). His CHA2DS2-VASc score is 1 (history of HTN).
Step 2: Calculate Mr G.R.’s risk of major bleed using the HAS-BLED score (Table 3). His HAS-BLED score is 0.
Step 3: Balance the predicted risk of stroke versus the predicted risk of bleeds. Mr G.R.’s risk of stroke based on CHA2DS2-VASc is 1.3% per year (risk of stroke based on CHADS2 is 2.8% per year) and his risk of bleeding is 1.13% per year.
With a CHA2DS2-VASc score of 1, Mr G.R. can be started on an OAC or ASA. In a discussion with Mr G.R., you outline his therapeutic options and compare warfarin with the new OACs. (The online stroke prevention in AF risk calculator at www.vhpharmsci.com/sparc/ can assist in explaining the risks and benefits of antithrombotic therapy to patients with AF and compares the effectiveness of the OACs and antiplatelets. A stroke prevention in AF tool is also available at the Canadian Cardiovascular Pharmacists Network website, www.ccpn.ca.) Mr G.R. travels a lot for work and, as such, thinks the required INR monitoring with warfarin is impractical. He also does not qualify for provincial drug coverage with the new OACs and is unable to afford one of these agents. Despite the lower effectiveness of ASA compared with the OACs, he believes ASA best fits his lifestyle.
Mr G.R. is started on 81 mg/d of enteric-coated ASA for stroke prevention and 2.5 mg/d of bisoprolol for rate control. His valsartan and hydrochlorothiazide dosages are continued with a plan to monitor his blood pressure frequently. (For additional information on rate versus rhythm options, visit www.rxfiles.ca to see the RxFiles AF chart.21)
Step 4: Re-assess the patient’s risk of stroke and risk of bleed annually or sooner if his or her risk criteria change.
Eleven years later (Mr G.R. is now 73 years old)
Mr G.R.’s risk of stroke and bleeding is reassessed annually over several years. Today he presents to the emergency department with numbness and weakness in his right arm, along with speech difficulty. A computed tomography scan, a carotid ultrasound, and an echocardiogram are completed. He is diagnosed with a TIA.
His current medical history includes AF, HTN (systolic blood pressure below 160 mm Hg), dyslipidemia, and degenerative joint disease. His medications include 81 mg/d of enteric-coated ASA, 10 mg/d of bisoprolol, 160 mg of valsartan and 12.5 mg of hydrochlorothiazide daily, and 20 mg/d of atorvastatin. His laboratory test results (complete blood count, renal and liver enzymes) are normal.
Reevaluation of therapy to prevent further TIA or stroke
-
Step 1: Mr G.R.’s CHADS2 score is 3 (HTN, TIA).
-
Step 2: His HAS-BLED score is 1 (age > 65 years).
-
Step 3: His risk of stroke is 5.9% per year and his risk of bleeds is 1.02% per year.
The OAC options and a summary of the risks versus benefits for each option are discussed with Mr G.R. (Tables 410–16 and 510–12,19,21–23). He prefers warfarin over the newer agents for the following reasons:
-
Mr G.R. finds comfort in knowing warfarin has been used for decades and there is an available antidote, unlike the new OACs, if he does experience a bleed due to a supratherapeutic INR. To him, this is more important than the convenience of the other OACs (eg, no need for INR monitoring, less potential for drug-drug and drug-food interactions).
-
Financially, he still cannot afford to pay for one of the newer OACs, and he does not meet provincial drug formulary criteria.
-
The once-daily dosing of warfarin appeals to him.
Mr G.R. starts taking warfarin. His ASA dosage is discontinued once his INR reaches the therapeutic range (2 to 3) and he has received at least 5 days of warfarin therapy.
Seven years later (Mr G.R. is now 80 years old)
Despite several years with well-controlled INRs, it has been difficult to maintain Mr G.R.’s INR in target range over the past few months (only 5 out of 10 INRs were within therapeutic range). His daughter brings him to see you to discuss his warfarin therapy. His wife passed away 3 months ago. Mr G.R. is still mourning the loss of his wife and is struggling to take care of himself. He has lost 15 pounds owing to a decreased appetite and has been drinking alcohol daily. His daughter has been assisting with his care but is finding this difficult, and sometimes she cannot drive him to his appointments for INR testing. There are no new concerns with his laboratory test results, or signs of bleeding or another stroke. His serum creatinine level is 112 μmol/L, with an estimated calculated creatinine clearance (CrCl) of 48 mL/min.
Reevaluation of therapy to prevent further TIA or stroke
-
Step 1: Mr G.R.’s CHADS2 score is 4 (HTN, TIA, age > 75 years).
-
Step 2: His HAS-BLED score is 3 (age, labile INRs, alcohol).
-
Step 3: Mr G.R.’s risk of stroke is 8.5% per year and his risk of bleeds is 3.74% per year.
Mr G.R. is no longer a good candidate for warfarin therapy owing to his drinking, his labile INRs, and the challenges of getting to his appointments for INR testing. Currently, other OAC options include rivaroxaban or dabigatran.*
Bringing evidence to practice
-
Mr G.R. has a CHADS2 score of 4. Approximately one-third of the patients included in the landmark trials had a baseline CHADS2 score similar to Mr G.R.’s—32.5% in RELY (dabigatran, CHADS2 score of 3 to 6) and 29% in ROCKET-AF (rivaroxaban, CHADS2 score of 4) (Table 410–16).
-
The dose of the new OACs should be reduced in patients who are elderly or who have a low body mass index, hepatic dysfunction, or an increased risk of bleeding. These agents should not be used in patients with a CrCl rate less than 30 mL/min. Dabigatran should be reduced to 110 mg twice daily in patients 80 years of age or older, or in those older than 75 years with 1 or more bleeding risks.
-
Neither rivaroxaban nor the 110-mg dose of dabigatran regimen was superior to warfarin for reducing the risk of stroke and systemic embolism; however, the regimen of 150 mg of dabigatran twice daily was superior. For Mr G.R., 110 mg of dabigatran twice daily would be appropriate owing to decreased renal clearance of the drug and his age. In theory, Mr G.R.’s reduced clearance of the regimen of 110 mg of dabigatran twice daily might correlate better with the regimen of 150 mg of dabigatran twice daily used in the RELY trial.
-
He now meets provincial drug formulary criteria for dabigatran (ie, inadequate anticoagulation with warfarin after a reasonable trial). Rivaroxaban is not listed on Mr G.R.’s provincial formulary for AF.24
Mr G.R. is willing to try dabigatran and has agreed to obtain help to cope with the loss of his wife. Home-care services are also arranged to assist with his activities of daily living and medication administration, as dabigatran cannot be pill or compliance packaged. Warfarin is discontinued, and he will start taking 110 mg of dabigatran twice daily once his INR is less than 2 (approximately 2 to 5 days). His renal function should be assessed at least annually to ensure dabigatran is still appropriate (CrCl > 30 mL/min).
Footnotes
-
This article is eligible for Mainpro-M1 credits. To earn credits, go to www.cfp.ca and click on the Mainpro link.
-
↵* To date, apixaban has only been approved by Health Canada for venous thromboembolism prophylaxis after total hip or knee replacement surgery. Approval for prevention of stroke and systemic embolism in patients with atrial fibrillation is pending.3
-
↵† The RxFiles Trial Summaries of the ARISTOTLE, RELY, and ROCKET-AF trials are available at www.cfp.ca. Go to the full text of the article online, then click on CFPlus in the menu at the top right-hand side of the page.
-
Competing interests
RxFiles and contributing authors do not have any commercial competing interests. RxFiles Academic Detailing Program is funded through a grant from Saskatchewan Health to Saskatoon Health Region; additional “not for profit; not for loss” revenue is obtained from sales of books and online subscriptions.
- Copyright© the College of Family Physicians of Canada