The Indian consensus guidance on stroke prevention in atrial fibrillation: An emphasis on practical use of nonvitamin K oral anticoagulants

Abstract

The last ten years have seen rapid strides in the evolution of nonvitamin K oral anticoagulants (NOACs) for stroke prevention in patients with atrial fibrillation (AF). For the preparation of this consensus, a comprehensive literature search was performed and data on available trials, subpopulation analyses, and case reports were analyzed. This Indian consensus document intends to provide guidance on selecting the right NOAC for the right patients by formulating expert opinions based on the available trials and Asian/Indian subpopulation analyses of these trials. A section has been dedicated to the current evidence of NOACs in the Asian population. Practical suggestions have been formulated in the following clinical situations: (i) Dose recommendations of the NOACs in different clinical scenarios; (ii) NOACs in patients with rheumatic heart disease (RHD); (iii) Monitoring anticoagulant effect of the NOACs; (iv) Overdose of NOACs; (v) Antidotes to NOACs; (vi) Treatment of hypertrophic cardiomyopathy (HCM) with AF using NOACs; (vii) NOACs dose in elderly, (viii) Switching between NOACs and vitamin K antagonists (VKA); (ix) Cardioversion or ablation in NOAC-treated patients; (x) Planned/emergency surgical interventions in patients currently on NOACs; (xi) Management of bleeding complications of NOACs; (xii) Management of acute coronary syndrome (ACS) in AF with NOACs; (xiii) Management of acute ischemic stroke while on NOACs.

1. Introduction

Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia characterized by uncoordinated atrial activation with subsequent deterioration of the atrial mechanical function.¹ It is an important risk factor for cardioembolic stroke.2, 3 Nonvalvular AF (NVAF) is associated with a 5-fold increased risk of stroke,⁴ with stroke being more fatal and disabling than in those without AF.⁵ The risk of stroke in AF increased from 1.5% at 50–59 years of age to 23.5% at 80–89 years.² Valvular AF on the other hand is associated with a 17-fold increased risk of stroke.⁶ Importantly, in India, valvular AF patients comprise not just those with mechanical heart valves, but also those with rheumatic mitral stenosis. The group of mild rheumatic valvular disease with AF patients needs to be tested with NOACs for stroke prevention in AF, even though 21.8% of patients in RE-LY did have patients with mild to moderate valvular heart disease with AF and were still classified as NVAF.⁷ The prevalence of AF in the general population in North America and Europe is 1–2%.⁸ The United Kingdom (UK)-based West Birmingham Atrial Fibrillation project showed a prevalence of 0.6% in the Indian subset.⁹ However, there is paucity of epidemiological data to determine the true incidence and prevalence of AF in India. This paucity seems to be addressed to some extent by the Indian Heart Rhythm Society-Atrial Fibrillation (IHRS-AF) registry, REALIZE-AF, and Indian subgroup analysis (unpublished) of pivotal randomized controlled trials of the NOACs.10, 11 The prevalence of paroxysmal AF as reported in the RE-LY (Randomised Evaluation of Long term anticoagulant therapY), REALIZE, and IHRS-AF study was 38%, 43%, and 19.5%, respectively. Permanent AF was reported in 18.6%, 34.3%, and 33.7% of participants in the RE-LY, REALIZE, and IHRS-AF study, respectively.¹⁰

Every year, 20 million people worldwide experience a stroke. In the Asia-Pacific region, China and India have the maximum number of deaths resulting from stroke.⁹ Approximately 15% of all strokes are associated with AF.¹⁰

Importantly, longitudinal community-based studies conducted worldwide have shown that there has been a steady increase in AF incidence over the last two decades. This trend is likely to continue over the next few decades with an ageing population and higher occurrence of the associated risk factors, including cardiac diseases. Despite major advances in its management, AF still remains a significant cause of cardiovascular morbidity and mortality, especially that arising from stroke and heart failure (HF).¹²

2. Stroke risk assessment

In patients with NVAF, prior stroke or transient ischemic attack (TIA) is the strongest independent predictor of stroke.¹³ Heart failure, hypertension, increasing age, and diabetes mellitus have consistently emerged as independent risk factors for ischemic stroke associated with NVAF.14, 15 Numerous scoring schemes have been devised to predict stroke risk based on the risk factors identified in nonwarfarin arms of randomised clinical trial cohorts. However, these scores have been developed nearly two decades ago and need to be revisited with evidence from new epidemiological studies. Two of the most commonly used stroke risk assessment tools are discussed below.

The CHADS₂ (congestive heart failure, hypertension, age ≥75 years, diabetes mellitus, prior stroke, or TIA or thromboembolism [doubled]) score has been validated in numerous cohorts.16, 17 Though the scoring is simple, most researchers now agree that it does not include many of the common stroke risk factors and has several limitations and hence should not be used in practice (Table 1).18, 19

Table 1.

Definition of major and clinically relevant nonmajor risk factors for stroke in nonvalvular atrial fibrillation.⁸

Major risk factors	Clinically relevant nonmajor risk factors
Previous stroke, TIA, or systemic embolism, age >75 years	Heart failure or moderate to severe LV systolic dysfunction (e.g., LV EF <40%) Hypertension – Diabetes mellitus Female sex – Age 65–74 years Vascular diseasea

TIA, transient ischemic attack; LV, left ventricle; EF, ejection fraction.

^aPrior myocardial infarction, peripheral artery disease, aortic plaque.

The CHA₂DS₂-VASc (congestive heart failure, hypertension, age ≥75 [doubled], diabetes, stroke [doubled], vascular disease [e.g., past myocardial infarction (MI), peripheral arterial disease, or aortic atherosclerosis], age 65–74 years, and sex category [female]) score are inclusive of the most common stroke risk factors in everyday clinical practice. The CHADS₂ score can be used as a simple initial means of assessing stroke followed by the comprehensive risk factor-based approach using the CHA₂DS₂-VASc score.⁶ Table 2 describes the scoring pattern of the CHA₂DS₂-VASc score along with the adjusted stroke rate per year. The net effect of the CHA_2-DS₂-VASc score is to increase the proportion of appropriate AF patients for whom anticoagulation is recommended.²⁰

Table 2.

Risk factors in the CHA₂DS₂-VASc score and the adjusted stroke rate.²⁰

Risk factor	Score assigned
Congestive heart failure/left ventricle dysfunction	1
Hypertension	1
Age >75 years	2
Diabetes mellitus	1
Stroke/TIA/thromboembolism	2
Vascular diseasea	1
Age 65–74 years	1
Sex category (i.e. female sex)	1

CHA2DS2-VASc score (total)	Adjusted stroke rate (%/year)
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%

TIA, transient ischemic attack; LV, left ventricle; EF, ejection fraction.

^aPrior myocardial infarction, peripheral artery disease, aortic plaque.

The overall goal of stroke risk assessment score is to separate patients at ‘true low-risk’ and identify those who need treatment. Individuals with a CHA₂DS₂-VASc score of 0 (age <65 years with ‘lone AF’ [individuals without clinical or echocardiographical evidence of cardiopulmonary disease, including hypertension]) and with a CHA₂DS₂-VASc score of 1 (female patients aged <65 years and with ‘lone AF’) can be safely considered as ‘truly low-risk’ patients and antithrombotic therapy should not be considered.²⁰ Antithrombotic therapy should be considered for those with CHA₂DS₂-VASc score of equal to or more than 1 in males and equal to or more than 2 in females.

3. Bleeding risk assessment

Thromboprophylaxis with antithrombotic agents is associated with an increased risk of bleeding and requires individual risk assessment before initiation. Many of the risk factors for bleeding overlap with the risk factors for stroke.21, 22 Several bleeding risk assessment tools are available but only three have been derived and validated in patients with AF. These include HEMORR₂HAGES (hepatic or renal disease, ethanol abuse, malignancy, older age [≥75 years], reduced platelet count or function, rebleeding risk, hypertension [uncontrolled], anemia, genetic factors, excessive fall risk, and stroke), HAS-BLED (hypertension, abnormal renal/liver function, stroke, bleeding history or predisposition, labile international normalized ratio [INR], elderly [e.g., age >65 years, frailty, etc.], drugs/alcohol concomitantly), and ATRIA (AnTicoagulation and Risk factors In Atrial fibrillation).²⁰

The HAS-BLED score has been validated in multiple cohorts. It has performed as good as (sometimes better than) the more complex HEMORR₂HAGES and outperformed the less practical ATRIA score in predicting clinically relevant bleeding.23, 24 A high HAS-BLED score (≥3) is predictive of major bleeding during bridging of chronic anticoagulant therapy.²⁵ It is recommended to use HAS-BLED score for the assessment of oral anticoagulant-related bleeding risk in clinical practice. Table 3 provides the details of the HAS-BLED scoring system.

Table 3.

Risk factors of the HAS-BLED scoring system.²³

Risk factor	Score assigned
Hypertensiona	1
Abnormal renal or liver function (one point each)b	1 or 2
Stroke	1
Bleedingc	1
Labile INRd	1
Elderly (e.g. age >65 years)	1
Drugs or alcohol (1 point each)e	1
Maximum total score	9

INR, international normalized ratio.

^aHypertension’ is defined as systolic blood pressure >160 mmHg.

^bAbnormal kidney function is defined as the presence of chronic dialysis or renal transplantation or serum creatinine ≥200 μmol/L. Abnormal liver function is defined as chronic hepatic disease (e.g., cirrhosis) or biochemical evidence of significant hepatic derangement (e.g., bilirubin 2× the upper limit of normal, in association with aspartate aminotransferase/alanine aminotransferase/alkaline phosphatase. 3× upper limit normal, etc.).

^cBleeding refers to previous bleeding history and/or predisposition to bleeding, e.g. bleeding diathesis, anemia, etc.

^dLabile INRs refers to unstable/high INRs or poor time in therapeutic range (e.g. <60%).

^eDrugs/alcohol use refers to concomitant use of drugs, such as antiplatelet agents, nonsteroidal anti-inflammatory drugs, or alcohol abuse, etc.

The CHA₂DS₂-VASc and the HAS-BLED scores have been derived and validated mostly in the Western population. A meta-analysis identified different set of risk factors that are associated with stroke in the Western and the Asian population.26, 27 Another meta-analysis found that the incidence of intracerebral hemorrhage is the highest among the Asians.²⁸ Although validation of these scores is available in a Chinese population, it is recommended to validate both the CHA₂DS₂-VASc and the HAS-BLEED scores in the Indian population for improved management strategies for stroke prevention.²⁹

4. Oral antithrombotic agents

4.1. Vitamin K antagonist (VKA)

Until 2009, vitamin K antagonist (VKA) (such as warfarin) was the only class of oral anticoagulant approved for the prevention of stroke in AF. Warfarin is a coumarin derivative that inhibits vitamin K epoxide reductase responsible for the cyclic interconversion of vitamin K and vitamin K epoxide. Vitamin K is an essential cofactor for the carboxylation of coagulation factors II, VII, IX, and X, and therefore, their biological activation. Antagonism of vitamin K reduces the rate at which these factors are produced by the liver, thereby creating a state of anticoagulation.³⁰

A meta-analysis of data from six randomized clinical trials that compared a VKA with placebo or control found that adjusted-dose warfarin reduced the relative risk (RR) of stroke by 64% (95% confidence interval [CI] 49–74) versus placebo or control. The relative risk reduction for ischemic stroke with adjusted-dose warfarin was 67% (95% CI 54–77) while the reduction in all-cause mortality was 26% (95% CI 3–43).³¹ The Birmingham Atrial Fibrillation Treatment of the Aged (BAFTA) study showed that treatment with VKA (target INR of 2–3) was superior to aspirin 75 mg daily in reducing the primary endpoint of fatal or disabling stroke, intracranial hemorrhage (ICH), or clinically significant arterial embolism by 52%. The risk of major hemorrhage was comparable between the two groups.³²

The target INR should protect from ischemic stroke as well as hemorrhagic complications. Many studies and meta-analyses have reported the risk of stroke and/or major bleeding events in relation to INR, or the time spent in therapeutic range (TTR).33, 34, 35 It is evident that the risk of ischemic stroke with insufficient warfarin anticoagulation (INR < 2) and that of bleeding with over anticoagulation (INR > 3) is significantly higher relative to patients with NVAF maintained within an INR of 2–3. For the primary prevention of stroke in patients above 75 years, a target INR of 2 (range 1.6–2.5) is recommended while a target of 2.5 (range 2–3) is favorable for patients below 75 years (Fig. 1).1, 36

Fig. 1.

Limitations of vitamin K therapy.

The effectiveness and safety of warfarin treatment depends on the extent of time spent in the recommended INR range. A meta-analysis found that patients receiving warfarin spent 61% of time within, 13% of time above, and 26% of time below the therapeutic range (INR range 2–3). Gallagher et al. evaluated the association between TTR, when on warfarin anticoagulation treatment, and the risk of stroke and mortality.³⁷ The average time spent in the therapeutic range was 63.1%. Reduction in stroke was the highest (79%) when the patients spent at least 70% of time within the therapeutic range when compared to patients with ≤30% time in the recommended range. Significant reduction in the risk of stroke was observed when time spent in therapeutic range was more than 61% as compared to those who did not receive any antithrombotic therapy. The risk of mortality was reduced by 81% in warfarin users who spent at least 70% of time in the therapeutic range.

Because warfarin undergoes hepatic metabolism and is highly protein bound, it is particularly prone to drug interactions. Warfarin has two active isomers, the S-isomer being 2–4 times more potent than the R-isomer. The S-isomer is metabolized primarily by the cytochrome P450 2C9 (CYP2C9) and the R-isomer is metabolized by cytochrome P450 1A2 and 3A4 isozymes. The effect on INR is typically observed within 3–5 days for drugs with short half-lives and is delayed further for drugs with longer half-lives. Some important interactions of warfarin are listed in Table 4. In addition, warfarin has several other limitations and challenges, such as a narrow therapeutic window, increased risk of bleeding including ICH, frequent intensive INR monitoring, and dose adjustment.³⁸ Some of the limitations of warfarin therapy are illustrated in Fig. 1.39, 40 Other important limitations of warfarin include high coefficient of interlab variation in INR estimation; INR is not reflective of monthly or long-term control (TTR as measured by the Rosendaal method along with Finn method may be estimated for more appropriate control); certain drugs like amiodarone may themselves affect INR and thus, interpretation of the patient's INR in those taking co-medications becomes difficult; patients with INR between 2 and 3 can still bleed or have a stroke.39, 40

Table 4.

Food and drug interactions with warfarin.

Potentiate the effect of warfarin	Inhibit the effect of warfarin
Acetaminophen	Mercaptopurine
Alcohol (if concomitant liver disease)	Mesalamine
Fenofibrate	Ribavirin
Mango	Trazodone
Miconazole vaginal suppositories	Azathioprine
Quilinggao	Bosentan
Amoxicillin/clavulanate	Ginseng
Azithromycin	Ritonavir
Celecoxib	Sulfasalazine
Clarithromycin	Terbinafine
Danshen	Ubidicarenone
Fluorouracil	Green tea
Fluvastatin	Furosemide
Fluvoxamine	Propofol
Gemcitabine	Furosemide
Grapefruit juice
Interferon
Levamisole/fluorouracil
Levofloxacin
Paclitaxel
Paracetamol
Ritonavir
Ropinirole
Tolterodine
Tramadol
Troglitazone
Acarbose
Amiodarone induced toxicosis
Amoxicillin
Chloramphenicol
Danazol
Miconazole topical gel
Ofloxacin
Trastuzumab

Warfarin-related nephropathy is a newly described entity in those with an acutely increased INR of more than 3 soon after the initiation of warfarin. This, if confirmed, is especially serious in patients with chronic kidney disease (CKD) in whom it is more often associated with an unexplained acute increase in serum creatinine and an accelerated progression of CKD. In 4006 patients with CKD and INR exceeding 3, the one-year mortality was 31.1% compared with 18.9% without warfarin-related nephropathy. Hence, the INR should be kept below 3 in all patients soon after starting warfarin, but essentially in those with CKD or those who use antithrombin inhibitors.⁴¹

Even with the advent of NOACs, in India, there will always be a role for the relatively cheaper VKAs due to the cost factor. VKAs are still the preferred drug in situations including, but not limited to, those AF patients on mechanical heart valves or those with creatinine clearance <15 ml/min.

4.2. Nonvitamin K oral anticoagulants (NOACs)

The NOACs fall into two major categories: direct thrombin (factor IIa) inhibitors (dabigatran) and direct factor Xa inhibitors (rivaroxaban, apixaban, and edoxaban). As compared to warfarin, these NOACs have a predictable pharmacokinetic profile and fewer food-drug and drug-drug interactions, and do not require routine anticoagulant monitoring.²¹ Following rigorous phase III clinical trials, dabigatran received the United States Food and Drug Administration (US FDA) approval to prevent stroke in patients with NVAF in 2010. This was followed by rivaroxaban approval in 2011 and apixaban approval in 2012. All the three drugs are also approved in Europe. Edoxaban received the US FDA approval in 2015 and an application for the marketing authorization has been recently submitted to the European Medicines Agency (EMA).

4.2.1. Dabigatran

Dabigatran etexilate is an oral prodrug that is rapidly converted by serum esterase-mediated hydrolysis to dabigatran, a potent, direct competitive inhibitor of thrombin.⁴² A summary of the pharmacokinetics of dabigatran has been presented in Table 5.⁴³

Table 5.

Summary of pharmacokinetic parameters of NOACs.

	Dabigatran (Pradaxa®)43	Rivaroxaban (Xarelto®)55, 56	Apixaban (Eliquis®)62	Edoxaban (SAVYASA)64
Bioavailability	3–7%	66% without food; almost 100% with food	50%	62%
Time for peak effect	2–3 h	2–4 h	3–4 h	1–2 h
Plasma half-life	12–17 h	5–13 h	12 h	10–14 h
Metabolism	Via P-gp transporter	Via CYP450 and P-gp transporter	Via CYP450 and P-gp transporter	Via CYP450 and P-gp transporter
Clearance nonrenal/renal	20%/80%	73%/27%	50%/50%	65%/35%

Dabigatran offers an advantage over indirect thrombin inhibitors like heparin, as it inhibits both free and fibrin-bound thrombin. The reversible binding of dabigatran is comparable to injectable direct thrombin inhibitor (DTI), bivalirudin. DTIs have an antiplatelet effect as well due to reduced thrombin-mediated activation of platelets. They produce a more predictable anticoagulant response than heparin, as they do not bind to plasma proteins and lack immune-mediated thrombocytopenia.44, 45, 46

The pivotal RE-LY trial compared two blinded doses of dabigatran etexilate (110 mg [D110] or 150 mg [D150] BID, twice daily) with an open-label, adjusted-dose of warfarin.⁷ The study design and baseline characteristics have been elaborated in Table 6.

Table 6.

Study design and baseline characteristics of Phase III pivotal trial with NOACs.

	Dabigatran7	Rivaroxaban58	Apixaban62	Edoxaban65
Study acronym	RE-LY	ROCKET-AF	ARISTOLE	ENGAGE AFTIMI-48
Study design	Randomized, open-label	Randomized, double blind	Randomized double blind	Randomized double blind
No. of patients		14,264	18,201	21,105
Follow-up period, yrs	2	1.9	1.8	2.8
Randomized groups	Dose-adjusted warfarin (W) versus dabigatran 110 mg BID (D110), dabigatran 150 mg BID (D150)	Dose-adjusted warfarin (W) versus rivaroxaban 20 mg OD (R20)	Dose-adjusted warfarin (W) versus apixaban 5 mg BID (A5)	Dose-adjusted warfarin (W) versus Low-dose edoxaban 30 mg OD High-dose edoxaban 60 mg OD
Dose adjustment	None	15 mg OD in CrCl- 30 to 49 ml/min	2.5 mg BID if (any two of) age ≥80 years, body weight <60 kg, serum creatinine level ≥1.5 mg/dl	Dose was halved if estimated CrCl 30–50 ml/min, body weight ≤60 kg, or the concomitant potent P-gp inhibitors (verapamil or quinidine)
Age, yrs	71.5 ± 8.7 (mean ± SD)	73 (65–78) [median (interquartile range)]	70 (63–76) [median (interquartile range)]	72 (64–78) [median (interquartile range)]
Male, sex %	63.6	61.3	64.5	62.5
CHADS2 score (Mean)	2.1	3.5	2.1	2.8

RE-LY, Randomized Evaluation of Long term anticoagulant therapy; ROCKET-AF, Rivaroxaban Once-daily oral direct factor Xa inhibition Compared with vitamin K antagonism for prevention of stroke and Embolism Trial in Atrial Fibrillation; ARISTOLE, Apixaban for Reduction In STroke and Other ThromboemboLic Events in atrial fibrillation; ENGAGE AF-TIMI 48, Effective Anticoagulation with Factor Xa Next Generation in Atrial Fibrillation–Thrombolysis in Myocardial Infarction 48; BID, twice daily; OD, once daily; CrCl, Creatinine clearance; P-gp, P-glycoprotein.

The D150 arm was superior (p < 0.001) while the D110 arm was noninferior (p < 0.001) versus the warfarin arm for the primary efficacy endpoint of stroke or systemic embolism in the trial. Both doses of dabigatran reduced the annual rate of hemorrhagic stroke significantly. The annual rate of major bleeding was significantly lower with D110. Intracranial bleeding in the dabigatran group was observed at less than one-third the rate observed with warfarin, without a reduction in efficacy against ischemic stroke. Gastrointestinal (GI) bleeding was the most important adverse effect of the higher dose of dabigatran. A higher rate of MI was observed with both doses of dabigatran. However, the difference was not statistically significant when compared to warfarin. Overall, dabigatran 110 mg BID was noninferior to warfarin but had lower rates of major bleeding episodes and dabigatran 150 mg BID was superior to warfarin but had similar rates of major bleeding episodes.⁷ The efficacy and safety outcomes of the trial have been elucidated in Table 7.

Table 7.

Summary of efficacy and safety outcomes of NOACs.

Outcome (% per yr)	Dabigatran7 (RE-LY)		Rivaroxaban58 (ROCKET-AF)			Apixaban62 (ARISTOLE)		Edoxaban65 (ENGAGE AF TIMI-48)
	W (n = 6022)	D110 (n = 6015) RR, 95% CI, p value	D150 (n = 6076) RR, 95% CI, p value	W (n = 7133)	R20 (n = 7131) HR (95% CI, p value)	W (n = 9081)	A5 (n = 9120) HR (95% CI, p value)	W (n = 7035)	E30 (n = 7034) HR (95% CI, p value)	E60 (n = 7035) HR (95% CI, p value)
Stroke or systemic embolism	1.71	1.54 (0.90; 0.74–1.10; p < 0.001; NI)	1.11 (0.65; 0.52–0.81; p < 0.001 (NI, Sup)	2.4	2.1 (0.88; 0.75–1.03; p < 0.001; NI)	1.6	1.27 (0.79; 0.66–0.95; p < 0.001; NI, p = 0.01; Sup)	1.50	1.61 (1.07; 0.87–1.31; p = 0.005; NI)	1.18 (0.79; 0.63–0.99; p < 0.001; NI)
Ischemic stroke	1.21	1.34 (1.11; 0.88–1.39; p = 0.35)	0.92 (0.76; 0.59–0.97; p = 0.03)	1.42	1.34 (0.94; 0.75–1.17; p = 0.581)	1.05	0.97 (0.92; 0.74–1.13; p = 0.42)	1.25	1.77 (1.41; 1.19–1.67; p < 0.001)	1.25 (1.00; 0.83–1.19; p = 0.97)
Hemorrhagic stroke	0.38	0.12 (0.31; 0.17–0.56; p < 0.001)	0.10 (0.26; 0.14–0.49; p < 0.001)	0.44	0.26 (0.59; 0.37–0.93; p = 0.024)	0.47	0.24 (0.51; 0.35–0.75; p < 0.001)	0.47	0.16 (0.33; 0.22–0.50; p < 0.001)	0.26 (0.54; 0.38–0.77; p < 0.001)
Major bleeding	3.57	2.87 (0.80; 0.70–0.93; p = 0.003)	3.32 (0.93; 0.81–1.07; p = 0.31)	3.4	3.6 (1.04; 0.90–1.20; p = 0.58)	3.09	2.13 (0.69; 0.60–0.80; p < 0.001)	3.43	1.61 (0.47; 0.41–0.55; p < 0.001)	2.75 (0.80; 0.71–0.91; p < 0.001)
Intracranial bleeding	0.76	0.23 (0.30;0.19–0.45; p < 0.001)	0.32; 0.41 (0.28–0.60) p < 0.001	0.7	0.5 (0.67; 0.47–0.93 p = 0.02)	0.80	0.33 (0.42; 0.30–0.58; p < 0.001)	0.85	0.26 (0.30; 0.21–0.43; p < 0.001)	0.39 (0.47; 0.34–0.63; p < 0.001)
Gastrointestinal bleeding	1.07	1.15; 1.08 (0.85–1.38; p = 0.52	1.56; 1.48 (1.18–1.85; p = 0.001	2.2	3.2 (p < 0.001)	0.86	0.76 (0.89; 0.70–1.15; p = 0.37)	1.23	0.82 (0.67; 0.53–0.83; p < 0.001)	1.51 (1.23; 1.02–1.50; p = 0.03)
Myocardial infarction	0.64	0.82 (1.29; 0.96–1.75; p = 0.09)	0.81 (1.27; 0.94–1.71; p = 0.12)	1.1	0.9 (0.81; 0.63–1.06; p = 0.12)	0.61	0.53 (0.88; 0.66–1.17; p = 0.37)	0.75	0.70 (0.94; 0.74–1.19; p = 0.60)	0.89 (1.19; 0.95–1.49; p = 0.13)
All cause mortality	4.13	3.75 (0.91; 0.80–1.03; p = 0.13)	3.64 (0.88; 0.77–1.00; p = 0.051)	2.2	1.9 (0.85; 0.70–1.02; p = 0.07)	3.94	3.52 (0.89, 0.80–0.99; p = 0.047)	4.35	3.80 (0.87; 0.79–0.96; p = 0.006)	3.99 (0.92; 0.83–1.01; p = 0.08)
Net clinical benefit outcome	7.91	7.34 (0.92; 0.84–1.01; p = 0.09)	7.11 (0.90; 0.82–0.99; p = 0.02)	–	–	7.20	6.13 (0.85; 0.78–0.92; p < 0.001)	8.11	6.79 (0.83; 0.77–0.90; p < 0.001)	7.26 (0.89; 0.83–0.96; p = 0.003)

W, dose-adjusted warfarin; D110, dabigatran 110 mg twice daily; D150, dabigatran 150 mg twice daily; R20, rivaroxaban 20 mg once daily; A5, apixaban 5 mg twice daily; E30, edoxaban 30 mg once daily; E60, edoxaban 60 mg once daily.

In a subgroup analysis of the RE-LY trial for treatment effects, dabigatran was compared with warfarin for secondary prevention in patients with prior stroke or TIA; both doses of dabigatran were associated with lower rates of stroke or systemic embolism than warfarin (RR 0.84 for D110 and 0.75 for D150).⁴⁷ A significant treatment-by-age interaction was also observed. D110 was associated with a lower risk of major bleeding in patients below 75 years of age (1.89% versus 3.04%; p < 0.001) and with similar risk in those aged 75 years and above (4.43% versus 4.37%; p = 0.89; p for interaction < 0.001). Similarly, D150 was associated with a lower risk of major bleeding in those aged <75 years (2.12% versus 3.04%; p < 0.001) and showed a trend towards higher risk of major bleeding in those aged ≥75 years (5.10% versus 4.37%; p = 0.07; p for interaction <0.001).⁴⁸

Real-world evidence on the safety and effectiveness of dabigatran versus warfarin is available for a total of more than 250,000 patients; more than 118,000 of these were new users of dabigatran who were propensity-score matched or propensity-score weighed to new users of warfarin.49, 50, 51, 52, 53 Dabigatran was associated with a reduced risk of ischemic stroke (Hazard ratio HR, 0.80; 95% CI, 0.67–0.96), ICH (HR, 0.34; 95% CI, 0.26–0.46), and death (HR, 0.86; 95% CI, 0.77–0.96), compared with warfarin. Rates of major bleeding (HR, 0.97; 95% CI, 0.88–1.07) and MI (HR, 0.92; 95% CI, 0.78–1.08) were similar with both dabigatran and warfarin; however, the risk for major GI bleeding (HR, 1.28; 95% CI, 1.14–1.44) was increased with dabigatran versus warfarin.⁴⁹ Importantly, these findings from large populations in clinical practice were consistent with the favorable safety and efficacy profile of dabigatran indicated in the pivotal RE-LY study.

Factor Xa inhibitors: Factor Xa (FXa) is an attractive target for novel anticoagulants as it acts at the convergence point of the intrinsic and extrinsic coagulation pathways. One molecule of FXa catalyses the formation of 1000 thrombin molecules together with factor Va (as the prothrombinase complex). Inhibition of FXa activity blocks the amplification of thrombin generation, thereby limiting thrombin-mediated activation of coagulation and platelets without affecting the existing thrombin levels.

4.2.2. Rivaroxaban

Rivaroxaban is a highly selective, reversible direct oral FXa inhibitor.⁵⁴ It is rapidly absorbed after oral administration and attains maximum plasma concentration after 2–4 h. The pharmacokinetic parameters of rivaroxaban have been elucidated in Table 5 and the prominent Drug interactions of Rivaroxaban have been detailed in Table 8.55, 56

Table 8.

Drug interactions with NOACs.

, contraindicated; , dose reduction required; , consider dose reduction if another yellow present; , no dose adjustment.

The Rivaroxaban Once daily oral direct factor Xa inhibition compared with vitamin K antagonism for prevention of stroke and Embolism Trial in Atrial Fibrillation (ROCKET-AF) was a double-blind, double-dummy study conducted in 45 countries worldwide. The study design and patient characteristics of ROCKET-AF have been detailed in Table 6.⁵⁷

The trial results showed that rivaroxaban was noninferior to warfarin in both the primary efficacy endpoint of stroke and systemic embolism prevention (p < 0.001 for noninferiority) as well as the safety endpoint of major and clinically relevant nonmajor bleeding. The trial results have been elucidated in detail in Table 7. Major bleeding from a GI site was significantly higher in the rivaroxaban group (3.2%), as compared to the warfarin group (2.2%, p < 0.001). Though there was no significant difference in the rates of major and clinically relevant nonmajor bleeding between the two groups, intracranial bleeding and fatal bleeding occurred less frequently with rivaroxaban.⁵⁸

A subgroup of patients (20.7% of the enrolled population) with moderate renal impairment (Creatinine clearance, CrCl 30–49 ml/min) received a lower dose of rivaroxaban. For patients with moderate renal impairment, the rate of stroke and systemic embolism, major and clinically nonrelevant bleeding events were higher than those with CrCl ≥50 ml/min. Comparative treatment effects for rivaroxaban versus warfarin were similar for all major outcomes, including bleeding events, for those with and without renal impairment. GI bleeding was more frequent than warfarin in this subgroup of patients (4.1% versus 2.6% for warfarin, p = 0.02).⁵⁹

A subgroup analysis of ROCKET-AF investigated the efficacy and safety of rivaroxaban in patients aged ≥75 years and in those aged <75 years. There was no significant interaction between treatment and age for the primary outcome of stroke or systemic embolism (p = 0.31) or for major bleeding (p = 0.34). Clinically relevant nonmajor bleeding was significantly higher for patients aged ≥75 years compared with patients aged <75 years (p = 0.01).⁶⁰

One real-world study that included atrial fibrillation patients compared 3654 rivaroxaban-treated patients with 14,616 matched warfarin patients. Rivaroxaban was associated with similar rates of major bleeding (HR, 1.08; 95% CI: 0.71–1.64), ICH (HR, 1.17; 95% CI: 0.66–2.05), and GI bleeding (HR, 1.27; 95% CI: 0.99–1.63) when compared with warfarin. Rates of composite stroke and systemic embolism for rivaroxaban and warfarin were also similar (HR, 0.77; 95% CI: 0.55–1.09).⁶¹

4.2.3. Apixaban

Apixaban is a selective, reversible direct oral inhibitor of factor Xa. The important pharmacokinetic parameters have been detailed in Table 5 while the common drug interactions with apixaban have been detailed in Table 8.⁶²

The Apixaban for Reduction In STroke and Other ThromboemboLic Events in atrial fibrillation (ARISTOTLE) trial included patients with NVAF and at least one of the following risk factors for stroke: age of at least 75 years, previous history of stroke, TIA, or systemic embolism, symptomatic heart failure within the previous 3 months or left ventricular ejection fraction of no more than 40%; diabetes mellitus; or hypertension. The ARISTOTLE study design and the characteristics of the enrolled study patients have been detailed in Table 6.⁶² It is important to note that 26.4% of patients had mild to moderate valvular heart disease along with AF in the ARISTOTLE study.⁶²

The ARISTOTLE trial proved the superiority of apixaban over dose-adjusted warfarin in preventing stroke and systemic embolism (p < 0.01 for superiority). The major efficacy and safety results of ARISTOTLE have been detailed in Table 7. The protocol of apixaban defined major bleeding as clinically overt bleeding accompanied by a decrease in the hemoglobin level of at least 2 g/dl or more over a 24-hour period, along with the other clauses of the definition as per the International Society on Thrombosis and Haemostasis (ISTH). However, it remains unclear whether the final results capture the major bleeding using this definition or not. Rates of hemorrhagic stroke and intracranial bleeding were significantly lower (p < 0.001 for superiority) in patients treated with apixaban than with warfarin. GI bleeding was similar between the treatment arms. There was no significant difference in the incidence of ischemic stroke. A predefined subgroup analysis in the ARISTOTLE trial found no significant interaction between the TTR with warfarin treatment and any of the other efficacy or safety outcomes. However, a significant interaction (p = 0.003) was observed for major bleeding between diabetics (3.0% per year) and nondiabetics (1.9% per year) when treated with apixaban. Thus, in patients with NVAF and increased risk of stroke, apixaban was superior to warfarin in preventing stroke or systemic embolism, caused less bleeding, and resulted in lower mortality.⁶² The effect of apixaban in preventing stroke and reducing mortality was significantly better than warfarin across all age groups, and was associated with less major bleeding, less total bleeding, and less intracranial hemorrhage regardless of age (p interaction >0.11 for all).⁶³ However, the ARISTOTLE study did not allow patients to be on dual antiplatelet therapy and the predefined dosing in the study probably ensured that patients with a higher risk of bleeding got a lower dose (2.5 BID).

4.2.4. Edoxaban

Edoxaban is also an oral, selective inhibitor of Factor Xa. The pharmacokinetics of edoxaban has been detailed in Table 5 and important drug interactions of Edoxaban are tabled in Table 8.⁶⁴

The Effective Anticoagulation with Factor Xa Next Generation in Atrial Fibrillation–Thrombolysis in Myocardial Infarction 48 (ENGAGE AF-TIMI 48) was a double-blind, double-dummy trial that compared two doses of edoxaban (60 mg [E60] and 30 mg [E30] once daily) with warfarin (target INR 2.0–3.0). The study characteristics have been detailed in Table 6.⁶⁵ Both once-daily regimens of edoxaban were noninferior (p < 0.005 for E30 group and p < 0.001 for E60 group for noninferiority) to warfarin with respect to the prevention of stroke or systemic embolism and were associated with significantly lower rates of bleeding and death from cardiovascular causes. The key efficacy and safety results of the ENGAGE-AF TIMI48 study have been detailed in Table 7.

4.3. Efficacy and safety of NOACs versus warfarin in NVAF

The NOACs have been evaluated and tested extensively in large trials for their efficacy and safety, including “real life” follow-up data. The pivotal randomized trials were mostly designed as noninferiority studies and thus powered to show that NOACs are at least as good as warfarin in the prevention of stroke in AF. It is evident that dabigatran 150 mg BID and apixaban 5 mg BID were superior to warfarin in reducing stroke (or systemic embolism). Dabigatran reduced stroke (or systemic embolism) by 35% and apixaban reduced it by 21%. More importantly, only dabigatran 150 mg BID showed a significant reduction in the incidence of ischemic stroke. All NOACs reduced the risk of hemorrhagic stroke when compared with warfarin (Fig. 2). In the ROCKET-AF study, patients (n = 1474) with a CrCl of 30–49 ml/min received a lower dose of rivaroxaban 15 mg OD. In the ARISTOTLE study, few patients (n = 428) received half the dose of apixaban (2.5 mg BID). In the ENGAGE AF-TIMI 48 study, the dose of edoxaban was reduced from 60 mg OD to 30 mg OD (n = 1787) in the high-dose arm or from 30 mg OD to 15 mg OD (n = 1784) in the low-dose arm. Thus, all three studies included a dose-adjusted subset of population in the primary efficacy and safety analysis that may add to some bias in the endpoints. Dabigatran was also evaluated at a lower dose; however, no further dose adjustments were made and all patients in each subgroup showed comparable baseline characteristics.

Fig. 2.

Efficacy and safety of NOACs.

Currently, there are no head to head trials comparing the efficacy of NOACs. Several authors have performed meta-analysis of these trials. Differences in trial designs along with the definition of safety and efficacy endpoints pose a challenge to the meta-analysis of these trials. A systematic review evaluated the results of the NOAC versus warfarin trials (RE-LY, ROCKET-AF, and ARISTOTLE) and concluded that overall mortality was decreased in patients with AF receiving NOACs (risk difference estimated to be 8 [95% CI 3–11] fewer deaths per 1000 patients, RR 0.88, 95% CI 0.82–0.96).⁶⁶ In the meta-analysis that also included ENGAGE AF-TIMI, all-cause mortality was also significantly reduced with NOACs (2022 events in 29,292 patients [6.9%]) versus warfarin (2245 events in 29,221 patients [7.7%], RR 0.90, 95% CI 0.85–0.95, p = 0.0003).⁶⁷ In a meta-analysis of 50,578 patients from three randomized trials (RE-LY, ROCKET-AF, and ARISTOTLE), NOACs were found to significantly decrease the rate of stroke or systemic embolism as well lower the rates of intracranial bleeding. NOACs were associated with a significant 18% reduction in the composite of stroke or systemic embolism when compared to warfarin (2.8% versus 3.5%, odds ratio [OR] 0.82, 95% CI [0.74–0.91], p < 0.001; I² = 0% for heterogeneity; p = 0.62). All-cause mortality (6.0% versus 6.3%, OR 0.88, 95% CI [0.82–0.95], p = 0.001; I² = 0% for heterogeneity; p = 0.76) and rate of hemorrhagic stroke (0.3% versus 0.8%, OR 0.79, 95% CI [0.71–0.88], p < 0.001; I² = 59% for heterogeneity; p = 0.09) were significantly lower for NOACs as compared to warfarin. NOACs were associated with lower rates of intracranial bleeding (0.6% versus 1.3%, p < 0.001) and higher rates of GI bleeding (2.3% versus 1.3%, p = 0.036); however, heterogeneity among the trials was high for these endpoints. There was no difference in the rates of myocardial infarction.⁶⁸ Yet another meta-analysis found that the risk of intracranial bleeding with NOACs was lower than with warfarin (RR 0.46; 95% CI 0.33–0.65) but the risk of nonhemorrhagic stroke and systemic embolism was comparable to warfarin (RR 0.93; 95% CI 0.83–1.04).⁶⁹ This meta-analysis also observed the influence of geography on treatment outcomes. Asian patients experienced significantly fewer strokes and systemic embolism in RE-LY and ARISTOTLE studies in comparison to non-Europeans, whereas no significant difference was observed in ROCKET-AF.

4.4. NOACs in Asian population

The Asian subgroup analysis for the major Phase III trials of dabigatran, apixaban, and edoxaban is detailed in Table 9 with respect to the major efficacy and safety endpoints.65, 70, 71, 72 Further analysis of the RE-LY trial revealed that the rates of bleeding outcomes (major, GI major, life-threatening major, minor, total, intracranial, and hemorrhagic stroke) when on warfarin were numerically higher in the Asian subjects than in non-Asians. A significant interaction (p = 0.008) between treatment effect and the geographical region was observed when comparing D150 versus warfarin in Asians and non-Asians.⁷³ A previous report on AF patients treated with warfarin found a 4-fold higher HR for ICH in Asians compared with the whites.⁷⁴ A Japanese subgroup analysis of ROCKET-AF trial is also available and it showed noninferiority of rivaroxaban to warfarin in the primary efficacy endpoint of stroke and systemic embolism (HR for rivaroxaban 0.49, 95% CI 0.24–1.00) and demonstrated no significant differences in the incidence of major hemorrhage between the treatment groups.⁷⁵

Table 9.

Summary of efficacy and safety outcomes of NOACs in Asian population subgroup analysis^a

RR (95% CI)	Dabigatran (RE-LY)		Apixaban (ARISTOLE)	Edoxaban Engage AF-TIMI 48		Rivaroxaban ROCKET-AF
	D15073 (n = 2782)	D11073 (n = 2782)	Study using 5 mgb (n = 1993)62	30 mg once dailyb (n = 3383)65	60 mg once dailyb (n = 3383)65	20 mg once dailyb (n = 932)75
Stroke or systemic embolism	0.45 (0.28–0.72)	0.81 (0.54–1.21)	0.74 (0.50–1.10)	Annualized rate 1.83% versus 2.37% (Warfarin)	Annualized rate 2.43% versus 2.37% (Warfarin)	0.76 (0.42–1.37)
Major bleeding	0.57 (0.38–0.84)	0.57 (0.39–0.80)	0.53 (0.35–0.80)	Annualized rate 1.87% versus 4.12% (Warfarin)c	Annualized rate 3.51% versus 4.12% (Warfarin)c	0.63 (0.37–1.09)
Intracranial bleeding	0.20 (0.07–0.60)	0.41 (0.27–0.63)	0.36 (0.18–0.71)	NA	NA	0.23 (0.08–0.68)
All-cause death	0.90 (0.78–1.04)	0.98 (0.73–1.32)	1.02 (0.70–1.50)	NA	NA	0.70 (0.40–1.25)

^aFrom phase III trials.

^bDose adjustments were made in all trials on the basis of renal function, and in some trials on the basis of body mass and concurrent administration of other drugs.

^cHazard ratio not provided. Abbreviations: NA, not available; RR, relative risk.

4.5. Gastrointestinal bleeding with NOACs

In individual studies, major GI bleeding risk was significantly increased with rivaroxaban, edoxaban (higher dose), and dabigatran (higher dose), albeit there was no increase in GI major bleeds with dabigatran (lower dose) and apixaban. In a meta-analysis by Ruff et al., all NOACs together increased GI major bleeding (OR 1.25, 95% CI 1.01–1.55; p = 0.04).⁶⁷ Upon searching for MedDRA preferred terms for nonadjudicated GI bleeding AEs as reported in www.clinicaltrials.gov, similar or higher rates of GI bleeding events were observed with standard dose NOACs versus warfarin.⁷⁶ Table 10 describes the GI bleeding incidences in NOAC studies.

Table 10.

Nonadjudicated GI bleeding with NOACs.67, 76

Study	Warfarin	NOAC standard dose	NOAC low dose
RE-LY	1.37	1.93	1.42
ROCKET-AF	2.68	3.52	–
ARISTOTLE	1.59	1.93	–
ENGAGE-AF	3.19	3.28	2.33

All values in % patients per year.

4.6. NOACs in elderly patients (>75 years)

A meta-analysis by Sardar P and colleagues suggests that risk of major or clinically relevant bleeding was not significantly different between NOACs and conventional therapy in elderly adults. In AF trials, NOACs were more effective than conventional therapy in prevention of stroke or systemic embolism in an elderly population with AF. Hence, the group recommends that age should not be a limiting factor for use of NOACs.⁷⁷

4.7. NOACs in rheumatic heart disease

NOACs are approved for treatment of nonvalvular AF. Patients with valvular AF, i.e., patients with mechanical prosthetic heart valves or with severe valve disorder causing AF were excluded from all NOAC trials. Atrial fibrillation in patients with valvular problems other than these is defined as ‘nonvalvular’ and such patients were included in the trials. Atrial fibrillation with biological valves or after valve repair constitutes a gray area; however, patients with these were included in some trials on ‘nonvalvular AF’.

5. Treatment with NOACs

5.1. Pharmacokinetics and drug–drug interactions

Treatment with NOACs needs consideration of their pharmacokinetics and interaction with concomitant medication and co-morbidities. Table 5 summarizes the pharmacokinetic profile and Table 8 drug interactions of NOACs.

Absorption of NOACs is dependent on P-glycoprotein (P-gp) and various drugs and food components are P-gp modulators.⁷⁸ The prodrug of dabigatran, dabigatran etexilate, is a P-gp substrate and the bioavailability of dabigatran varies with P-gp modulation. As dabigatran is primarily excreted by the kidneys, P-gp inhibitors when administered in cases of renal insufficiency may increase the bioavailability of dabigatran. Many drugs used in AF are substrates for P-gp (e.g., verapamil, dronedarone, amiodarone, and quinidine) and may increase the bioavailability of both FIIa and FXa inhibitors.⁷⁹ NOACs should be avoided with concomitant administration of strong inducers of P-gp and FXa are contraindicated when used in combination with strong inhibitors of both CYP3A4.⁸⁰

Further details on specific interaction with NOACs are presented in Table 8.

5.2. Dose recommendations

Renal function is one of the most important criteria, which affects the excretion of NOACs, and hence should be assessed at least once a year for patients with normal or mild impairment of renal function. Since most of the NOAC trials used Cockroft Gault formula for calculating the creatinine clearance of the patients, we recommend using the same.

Table 11 describes the dosing recommendations for NOACs.

Table 11.

Dosing recommendation of NOACs for thromboprophylaxis in atrial fibrillation.

	Dabigatran (Pradaxa®)79	Rivaroxaban (Xarelto®)58	Apixaban (Eliquis®)	Edoxaban (SAVYASA)65
Dosing recommendation	150 mg twice daily with or without food	20 mg once daily with evening meal	5 mg twice daily	60 mg once daily
Dose adjustment in renal dysfunction	CrCl 15–30 ml/min consider 75 mg twice daily CrCl 30–50 ml/min + concomitant P-gp inhibitor consider 75 mg twice daily CrCl 30–49 ml/min with high risk of bleeding consider 110 mg twice daily *75 mg approved in US only	CrCl 30–50 ml/min consider 15 mg once daily	2.5 mg twice daily if any 2 of the following are present: age ≥80 years, body weight ≤60 kg, serum creatinine ≥1.5 mg/dl In case of CYP3A4 and P-gp dual inhibitors consider 2.5 mg twice daily	CrCl 30–50 ml/min consider 30 mg once daily
Dose adjustment in hepatic impairment	In patients with moderate hepatic impairment (Child Pugh B) showed no evidence of change in exposure or pharmacodynamics	Avoid use in patients with Child–Pugh B and C hepatic impairment or with any degree of hepatic impairment associated with coagulopathy	Mild hepatic impairment: no dose adjustment needed Moderate hepatic impairment: no dosing recommendation available Severe hepatic impairment: avoid use	Avoid use in patients with Child–Pugh B and C hepatic impairment
Not recommended	CrCl < 30 ml/min	CrCl < 30 ml/min	CrCl < 30 ml/min

CrCl, creatinine clearance; P-gp, P-glycoprotein; CYP, cytochrome P.

5.3. Valvular heart disease in AF patients

It is reasonable to use NOACs in patients having mild to moderate valvular disease. The current data also support the use of NOACs in patients with bioprosthetic valves (not mechanical valves).⁸¹

5.4. Monitoring anticoagulant effect

Anticoagulation therapy with warfarin needs dose adjustment to achieve an INR of 2.0–3.0. Because of significant inter- and intrapatient variability of effective doses and various food and drug interactions, regular anticoagulation monitoring is required to keep all patients in the target INR range. When a patient is started on warfarin, INR monitoring should be performed daily for at least two days until the target INR is achieved.⁸²

NOACs do not require routine monitoring of coagulation. However, quantitative assessment may be required to assess drug exposure and anticoagulant effect in emergency situations. INR is not an effective option for monitoring anticoagulation in NOAC-treated patients. Given the direct anticoagulant activity of NOACs, rapid onset/offset of anticoagulation effects, and relatively short half-lives, the exact time of last dose intake relative to the time of blood sampling is of prime importance.⁸³

The activated partial thromboplastin time (aPTT) and the prothrombin time (PT) may provide a qualitative assessment for the presence of dabigatran and rivaroxaban, respectively.⁸⁴ The relation between dabigatran and aPTT is curvilinear. In patients undergoing chronic therapy with dabigatran, the median peak aPTT was approximately 2-fold that of the control. The median aPTT, 12 h after the last dose, was 1.5-fold that of the control. If the aPTT level at trough (i.e., 12–24 h after ingestion) still exceeds two times the upper limit of normal, there may be a high risk of bleeding. Dabigatran has little effect on the PT and INR assays. Therefore, they are not suitable for quantitative assays.

The ecarin clotting time (ECT) assay provides a direct measure of thrombin inhibition activity. Greater than 3-fold elevation in the ECT value at trough is associated with the risk of bleeding. Hemoclot^® is a diluted thrombin time (dTT) test developed with appropriate calibrators for interpretation in the context of dabigatran use. When dabigatran is used with twice daily dosing, the dTT measured at trough is associated with an increased risk of bleeding.⁸⁵

The choice of measurement methods for direct FXa inhibitors is an anti-Xa assay. A number of in vitro and ex vivo studies indicated that anti-Xa chromogenic assays are more specific and sensitive than routine clotting test-based assays. Commercial anti-Xa amidolytic assays are mainly designed for measurement of anti-Xa activity of low molecular weight heparins (LMWHs) and some may require modifications for use with peak and trough levels of direct Xa inhibitors in treated patient plasma. LMWH reference standard cannot be recommended, as the mechanisms of action of the two are different. Product-specific calibrators must be used for accurate estimation of plasma level expressed in mass concentration (e.g., mg/l). Factor Xa-inhibitors demonstrate a concentration-dependent prolongation of PT. However, it is subjected to huge variation because of differences in PT reagents.⁸⁶ At present, there are no US FDA approved assays for the measurement of Factor Xa inhibitors. Currently, PT and antifactor Xa chromogenic assays (where available) are considered appropriate for qualitative and quantitative measurements of Factor Xa inhibitors, respectively. However, the interpretation of these results is complicated, as no therapeutic ranges exist.87, 88 Also, usage of the appropriate calibrator needs to be ensured if an antifactor Xa chromogenic assay is to be performed (specifically, heparin calibrators are not to be used).

Table 12 provides an overview of the anticoagulant monitoring assay with particular application to the NOACs.

Table 12.

Anticoagulant monitoring assay.82, 83, 84, 85, 86, 87, 88, 89

Parameter	Description	Dabigatran (Pradaxa®)	Rivaroxaban (Xarelto®)	Apixaban (Eliquis®)	Edoxaban (SAVYASA)
Time for peak effect	–	2–3 hrs	2–4 hrs	3–4 hrs	1–2 hrs
Plasma trough level	–	12–24 hrs after ingestion	16–24 hrs after ingestion	12–24 hrs after ingestion	12–24 hrs after ingestion
Activated partial thromboplastin time (aPTT)	Test for the intrinsic system; measures kininogen, prekallikrein, XII, XI, IX, VIII, X, V, and thrombin	Qualitative At trough: >2XULN Suggest excess bleeding risk	Not useful	Not useful	Not useful
Ecarin clotting time (ECT)	Specific assay for thrombin generation	Quantitative At trough: ≥3 X ULN Suggest excess bleeding risk	Not affected	Not affected	Not affected
Prothrombin time (PT)	Test of the extrinsic pathway, measures factor VII, X, V, thrombin and fibrinogen	Not useful	Qualitative	Not useful	Not useful
Thrombin time	Functional test of fibrinogen concentration and fibrin formation	Qualitative	Not useful	Not useful	Not useful
Diluted TT (dTT)	Uses the hemoclot thrombin inhibitor assay	Quantitative analysis	Not useful	Not useful	Not useful
Antifactor Xa	Measures factor X activation directly using a chromogenic substrate	Not affected	Quantitative; cut-off values for bleeding or thrombosis risk not established	Quantitative; cut-off values for bleeding or thrombosis risk not established	Quantitative; cut-off values for bleeding or thrombosis risk not established

ULN, upper limit of normal.

5.5. Overdoses

In terms of management of overdoses of NOACs, it is important to distinguish between an overdose with and without bleeding complications. Overdoses associated with bleeding complications should be managed as discussed in the section on management of complications. Coagulation tests can help to determine the risk of bleeding and its severity.

Activated charcoal may be considered to reduce absorption of any NOAC.⁸³ In addition, dialysis can be used to reverse the effect of dabigatran. In absence of specific reversal agents, a wait and see strategy is recommended since the half-life of NOACs is relatively shorter.

5.6. Antidotes

Though specific antidotes for NOACs are currently unavailable, a few are under development. Aripazine (PER977), a small synthetic molecule, and a potential universal reversal agent, completely reversed the anti-Xa activity of rivaroxaban and apixaban in a dose-dependent manner ex vivo in human plasma. When administered to weight-matched rats overdosed with rivaroxaban, apixaban, and dabigatran, aripazine decreased bleeding by >90% and the reduction was within the normal range for un-anticoagulated rats in a standard tail transaction bleeding model.⁸⁹ Aripazine has undergone first-in-human studies in volunteers pretreated or untreated with edoxaban. In this Phase I study, hemostasis was restored from the anticoagulated state within 10–30 min after administration of 100 to 300 mg of Aripazine and was sustained for 24 h.⁹⁰

Andexanet alpha (PRT064445), a truncated form of enzymatically inactive factor Xa, dose-dependently reversed the inhibitory activity and corrected the prolongation of ex vivo clotting time by factor Xa inhibitors.⁹¹ In phase II double-blind studies, intravenous andexanet alpha (420 mg) neutralized the antifactor Xa effects of apixaban and rivaroxaban by 91% and 53% (as compared to placebo), respectively.92, 93 Both these phase II studies have shown that anticoagulation returns to its pretreatment state within several hours of the bolus infusion, and thus, a constant infusion of this agent may be required for reversing anticoagulation for a longer period of time.92, 93 Edoxaban 60 mg once daily was reversed by 52% after a bolus of 600 mg andexanet and by 73% after a bolus of 800 mg, each followed by an infusion of 8 mg/min for 1 h.⁹⁴

Idarucizumab, a humanized mouse monoclonal antibody fragment (Fab), binds specifically to dabigatran with an affinity that is 350-fold greater than the affinity of dabigatran for thrombin. Rapid reversal of anticoagulant activity of dabigatran was observed in rats administered an intravenous bolus injection of idarucizumab. The first-in-human, single-rising-dose study found that idarucizumab achieved rapid peak plasma concentration, had rapid elimination, had no endogenous thrombin potential, and it did not affect any coagulation parameters.⁹⁵ The interim results of the RE-VERSE AD™ trial are now available, and it was shown that idarucizumab normalized the dTT and ECT test results in 88–98% of the patients with a median maximum percentage reversal of 100% (95% CI 100–100), and this effect was evident within minutes of administration of the IV bolus 2.5 g followed by 2.5 g after 15 min – total 5 g of idarucizumab.⁹⁶

Until these agents reach the market, the ‘wait-and-see’ management can be advocated considering the relatively short half-lives of NOACs in cases without bleeding complications.

5.7. Treatment of hypertrophic cardiomyopathy (HCM) with AF using NOACs

Anticoagulation is indicated in HCM with AF irrespective of the CHA₂DS₂-VASc score.⁹⁷ However, the choice of anticoagulation depends on the patient profile, as there is no data available to support one anticoagulant over another.

5.8. Management of bleeding complications and reversal of anticoagulation

Anticoagulant therapy carries the risk of bleeding, which may be due to dosing errors, hemorrhagic diatheses, or emergency medical procedures. Though the risk of major bleeding, particularly intracranial bleeding and life-threatening bleeding, was significantly lower with the NOACs, as compared to warfarin (Table 7), an effective plan is required for the management of bleeding in a real-world clinical setting.

It is known that the anticoagulant effects of heparins and VKAs can be reversed with protamine sulphate and prothrombin supplementation, respectively.⁹⁸ Administration of protamine sulphate may be associated with the potential for allergic response with ensuing hypotension and bronchoconstriction.⁹⁹ Reversal of the anticoagulant effect of VKAs with oral or parenteral vitamin K has a slow onset (at least 12–24 h) while fresh frozen plasma or coagulation factors may restore coagulation more rapidly.⁹⁸ In the absence of a specific antidote for NOACs, the current recommendation for bleeding management lacks the strength of clinical experience but depends on expert opinion or standardized hospital protocols.

Table 13 summarizes the management strategy recommended by this consensus statement that is also in alignment with the commonly referred guidelines.

Table 13.

Possible measures to take in case of bleeding.⁹⁸

Direct thrombin inhibitors (dabigatran)	FXa inhibitors (apixaban, edoxaban, rivaroxaban)
Nonlife-threatening bleeding
• Record the dosage regimen and time of last dose intake • Delay next dose or discontinue treatment as appropriate • Consider factors influencing homeostasis (concomitant antiplatelet medications) and those affecting plasma concentrations (CYP3A4 and P-gp modulators)	Record the dosage regimen and time of last dose intake Delay next dose or discontinue treatment as appropriate
Estimated time for normalization of hemostasis:  Normal renal function: 12–24 hrs  CrCl 50–80 ml/min: 24–36 hrs  CrCl 30–50 ml/min: 36–48 hrs  CrCl 30 ml/min: ≥48 hrs	Estimated time for normalization of hemostasis: 12–24 hrs
Standard supportive measures • Mechanical compression • Surgical hemostasis • Fluid replacement (colloids if needed) • RBC substitution if necessary • Platelet substitution (in case of thrombocytopenia ≤60 × 109/L or thrombopathy • Fresh frozen plasma as plasma expander	Standard supportive measures • Mechanical compression • Surgical hemostasis • Fluid replacement (colloids if needed) • RBC substitution if necessary • Platelet substitution (in case of thrombocytopenia ≤60 × 109/L or thrombopathy • Fresh frozen plasma as plasma expander
• Tranexamic acid can be considered as an adjuvant • Desmopressin can be considered in special cases (coagulopathy or thrombopathy)	• Tranexamic acid can be considered as an adjuvant • Desmopressin can be considered in special cases (coagulopathy or thrombopathy)
• Maintain adequate diuresis • Consider dialysis (preliminary evidence: −65% after 4 h) • Charcoal hemoperfusion not recommended (no data)
Life-threatening bleeding
All of the above	All of the above
PCC 25 U/kg (may be repeated once or twice) (but no clinical evidence)	PCC 25 U/kg (may be repeated once or twice) (but no clinical evidence)
aPCC (50 IE/kg; max 200 IE/kg/day): no strong data about additional benefit over PCC. Can be considered before PCC if available	a PCC 50 IE/kg; max. 200 IE/kg/day): no strong data about additional benefit over PCC. Can be considered before PCC if available
Activated factor VII (rFVIIa; 90 mg/kg) no data about additional benefit + expensive (only animal evidence)	Activated factor VII (rFVIIa; 90 mg/kg) no data about additional benefit + expensive (only animal evidence)

CrCl, creatinine clearance; PCC, prothrombin complex concentrate; aPCC, activated Prothrombin complex concentrates.

5.9. Switching between anticoagulant regimens

When switching between different anticoagulant therapies, it is of paramount importance to maintain the anticoagulation effect while minimizing the risk of bleeding at the same time.

5.10. Switching to and from VKAs to NOAC

INR monitoring is needed when transitioning patients from VKAs to a NOAC to avoid over anticoagulation. While switching from a NOAC to warfarin, consider bridging with a short-acting parenteral agent or a lower dose of the NOAC.

5.11. Switching to and from parenteral anticoagulants to NOACs

NOAC should be initiated up to 2 h before the next dose of the parenteral agent when transitioning from a parenteral agent to a NOAC. The prescribing information of each of the NOACs describes the strategy for switching between these therapies. Table 14 summarizes the transition between different treatment regimens.

Table 14.

Transition between anticoagulant regimens.

VKA to NOAC	INR <2.0: immediate INR 2.0–2.5: immediate or next day INR >2.5: use INR and VKA half-life to estimate time to INR <2.5
Parenteral anticoagulant to NOAC:  Intravenous unfractioned heparin (UFH)  Low molecular weight heparin (LMWH)	Start once UFH discontinued (t½ = 2 h). May be longer in patients with renal impairment Start when next dose would have been given
NOAC to VKA	Administer concomitantly until INR in appropriate range Measure INR just before next intake of NOAC Retest 24 h after last dose of NOAC Monitor INR in first month until stable values (2.0–3.0) achieved
NOAC to parenteral anticoagulant	Initiate when next dose of NOAC is due

VKA, vitamin K antagonist; NOAC, nonvitamin K oral anticoagulants; INR, international normalized ratio.

5.12. Cardioversion or ablation in NOAC-treated patients

Interventions like ablation increase the bleeding risk and require temporary discontinuation of the NOAC. In patients scheduled to undergo ablation, it is reasonable to perform the procedure 24 h after stopping the NOAC. It is recommended to perform a transesophageal echocardiography before the procedure, to rule out left atrial thrombi, as it is possible to have a left atrial thrombus in spite of adequate oral anticoagulation. Pulmonary vein isolation (PVI) carries a risk of serious bleeding. In practice, PVI is performed in VKA-treated patients without the interruption of VKA treatment and such an approach is associated with not only a reduction in thromboembolic events, but also leads to less bleeding. Comparable rates of thromboembolic events and bleeding rates were observed with NOACs compared to uninterrupted VKA. An individualized approach must be taken to decide on changing patients to uninterrupted VKA, or uninterrupted NOAC therapy, or of a well-planned cessation of NOAC. Studies are ongoing to evaluate the use of uninterrupted NOAC therapy before ablation. NOACs can be restarted 4 h after the sheath removal provided there is no evidence of pericardial effusion and adequate hemostasis has been achieved.⁷⁸

For cardioversion in patients with documented AF >48 h duration or AF of unknown duration, cardioversion should be performed only after 3 weeks of effective oral anticoagulation or if a transesophageal echocardiography (TOE) has ruled out Left Atrial (LA) thrombi. If the TOE detects a LA thrombus, the patient should not be subjected to cardioversion, as it can increase the risk of embolization. If AF duration is of less than 48 h, it is recommended to treat the patient with LMWH, supplement with TOE, and take a call to cardiovert the patient. The patient can then be started on NOACs for at least 4 weeks, irrespective of the patient's CHA₂DS₂-VASc score.

5.13. Planned/emergency surgical intervention

Patient characteristics (age, kidney function, previous bleeding complications, and concomitant medication) and surgical factors should be considered when deciding on the interruption and restart of NOAC drug. Interventions that carry no clinically important risk of bleeding (e.g., dental procedures, cataract, or glaucoma) can be performed at the trough concentration of the NOAC and then restarted after 6 h. For procedures involving minor risk of bleeding, it is recommended to discontinue NOACs 24 h before the elective procedure, provided the kidney functions normal. For procedures involving a major risk of bleeding, cease the NOAC treatment 48 h before the intervention. Though NOAC can be resumed in 6–8 h after the intervention, for a few surgical interventions, resuming full-dose anticoagulation within the first 48–72 h of the procedure may carry a bleeding risk that outweighs the risk for cardioembolism. An emergency intervention should be deferred, if possible, for at least 12 h and ideally 24 h after the last dose of NOAC.

5.14. Management of acute coronary syndrome (ACS) and AF

Coronary artery disease may coexist in approximately 20–30% of patients with AF, which is an indication for continuous antithrombotic treatment.100, 101 A considerable number of these patients are candidates for coronary revascularization to reduce risk of recurrent ischemic events with percutaneous coronary interventions (PCI), with stents implantation. Management of patients with NVAF and acute coronary syndrome (ACS), either as a ST-elevation myocardial infarction (STEMI) or as a non-ST elevation ACS (NSTE-ACS), is often challenging given the multiplicity of therapeutic options.¹⁰²

A meta-analysis was performed to evaluate the efficacy and safety of adding NOAC (apixaban, dabigatran, rivaroxaban, and ximelagatran) to single (aspirin) or dual (aspirin and clopidogrel) antiplatelet therapy in patients presenting with ACS. When compared with aspirin alone, the combination of an oral anticoagulant and aspirin reduced the incidence of major adverse cardiovascular events (MACEs) [HR and 95% CI 0.70; 0.59–0.84], but increased clinically significant bleeding (HR: 1.79; 1.54–2.09). Compared to the dual antiplatelet therapy with aspirin and clopidogrel, adding an oral anticoagulant decreased the incidence of MACE modestly (HR: 0.87; 0.80–0.95), but doubled the incidences of bleeding (HR: 2.34; 2.06–2.66).¹⁰³

The EHRA 2013 guidelines also mention that a triple therapy with dual antiplatelet agents and NOACs is associated with at least doubling the risk of major bleeding.⁸³ The WOEST trial¹⁰⁴ and the nationwide registry from Denmark¹⁰⁵ reported twice the number of bleeding episodes with triple therapy as compared to double therapy with warfarin and clopidogrel rather than aspirin. Hence, it cannot be said that NOACs behave differently from VKAs.

When an AF patient on NOAC presents with ACS, the NOAC treatment should be discontinued temporarily. The dual antiplatelet therapy should be initiated immediately (only aspirin in case of frail patients). In case of a STEMI, primary PCI via radial approach is strongly recommended over fibrinolysis.⁸³

The recommendations for management post-ACS in AF patients as per the ESC task force 2014 are depicted in Fig. 3.¹⁰⁶ It suggests that the period of triple therapy (OAC plus aspirin plus clopidogrel) should be as short as possible post-ACS. This should be followed by OAC plus a single antiplatelet therapy (preferably clopidogrel 75 mg/day, or as an alternative, aspirin 75–100 mg/day) and after a year, management should include only OAC for patients with AF and stable vascular disease (i.e., no acute events or revascularization for >12 months, whether coronary or peripheral artery disease). The OAC can be adjusted-dose VKA or a NOAC.¹⁰⁶ Dabigatran is the only NOAC administered with clopidogrel and/or aspirin in AF patients presenting with ACS at doses used for stroke protection.¹⁰⁷ Other NOACs like apixaban (2.5 mg OD) and rivaroxaban (15 mg OD) used in combination with clopidogrel and/or low-dose aspirin were given as a dose adjustment based on patient characteristics. No data are available indicating the benefit of rivaroxaban in ACS at the dose used for anticoagulation (20 mg OD).¹⁰⁸ Apixaban used in stroke prevention at a dose of 5 mg BID in combination with aspirin plus clopidogrel, in an ACS setting, was associated with excess bleeding.¹⁰⁹

Fig. 3.

Management of acute coronary syndrome in atrial fibrillation.⁸⁴

5.15. Management of acute ischemic stroke while on oral anticoagulants

Intravenous recombinant tissue plasminogen activator (rtPA) is an effective thrombolytic agent for acute ischemic stroke and is approved when administered within 4.5 hours’ time window from onset of stroke symptoms.¹¹⁰ The American Heart Association/American Stroke Association (AHA/ASA) guideline¹¹¹ allows the use of intravenous tissue plasminogen activator in warfarin-treated patients, whose INR ≤1.7 are not associated with an increased risk of symptomatic ICH.¹¹² The plasma half-lives of NOACs lie in the range of 8–17 h, and hence, thrombolytic treatment cannot be administered within 48 h of the last administration of the NOAC. In case of uncertainty about the last administered NOAC, coagulation test (aPTT and PT) should be ordered. A prolonged aPTT in case of dabigatran and a prolonged PT in case of factor Xa inhibitors are indicators of in vivo anticoagulation; thus, thrombolytic should not be administered. The EHRA 2013 guidelines recommend that if NOACs have been administered within 48 h and coagulation tests are not available or are abnormal, mechanical recanalization of occluded vessels may be considered.⁸³

Initiation or resumption of anticoagulation depends mainly on the severity of stroke as assessed by the NIHSS (National Institute of Health Stroke Scale) score. The thumb rule of 1-3-6-12 day may be applied, wherein the anticoagulant treatment may be resumed after a day in patients with transient ischemic attack; after 3 days in case of a small, nondisabling infarct; after 6 days in patients with moderate stroke and not before 2 (or even 3) weeks in case of large infarcts/severe stroke.⁸³ An ESC 2012 focused update suggests considering dabigatran 150 mg BID in patients on rivaroxaban or apixaban treatment who experience an ischemic stroke.¹¹³ Fig. 4 depicts the stroke management flowchart when the patient is on a NOAC. Table 15 has detailed the recommendations of this consensus statement for stroke prevention in AF patients.

Fig. 4.

Stroke management in patients on NOACs: (<4.5 h of symptom onset).⁸⁴

Table 15.

Recommendations on the use of NOACs.83, 113, 115

, preferred; , may be considered; , not recommended.

^* Edoxaban is not yet approved in India.

6. Summary of recommendations

Several committees have reviewed the available data on NOACs and provided recommendations to guide clinical practice. EHRA 2013 offers a practical guide on various treatment-related challenges of NOACs.⁸³ The 2012 focused update of the ESC guidelines also provides recommendation for the prevention on stroke in NVAF and guidance for use of NOACs.¹¹³ The American Academy of Neurology (2014) provides guidance on the diagnosis of NVAF and discusses therapeutic option to prevent stroke in NVAF.¹¹⁴ In view of the observations that individuals with Asian ethnicity are at a disproportionately higher risk of stroke and are more prone to warfarin-associated hemorrhages, Sabir et al. review the use of NOACs in the management of AF in Asian populations.¹¹⁵ Table 16 summarizes the important recommendations on stroke prevention as suggested by various guidelines in patients with AF using NOACs.

Table 16.

Recommendations for stroke prevention in nonvalvular AF (ESC 2012 guidelines) 83, 113

Category	Recommendations
Stroke risk assessment	• Stroke risk assessment should be based on CHA2DS2-VASc score
Bleeding risk assessment	• Bleeding risk assessment should be based on HAS-BLED score
Renal function assessment	• Regular assessment of renal function (by CrCl) is recommended in patients following initiation of any NOAC, which should be done annually but more frequently in those with moderate renal impairment where CrCl should be assessed 2–3 times per year
Antithrombotic therapy	• Choice should be based on the absolute risk of stroke and bleeding and the net clinical benefit for a given patient • Initiated for all patients with AF, except in those patients (both male and female) who are at low risk (aged <65 years and lone AF), or with contraindications.
No antithrombotic therapy	• For patients with a CHA2DS2-VASc score of 0 (i.e., aged <65 years with lone AF) with no risk factors • Female patients aged <65 and have lone AF (but still have a CHA2DS2-VASc score of 1 by virtue of their gender)
CHA2DS2-VASc score ≥2	• OAC therapy with adjusted-dose VKA (INR 2–3) or a direct thrombin inhibitor (dabigatran) or an oral factor Xa inhibitor (e.g. rivaroxaban, apixaban)
CHA2DS2-VASc score of 1	• OAC therapy with adjusted-dose VKA (INR 2–3) or a direct thrombin inhibitor (dabigatran) or an oral factor Xa inhibitor (e.g. rivaroxaban, apixaban) should be considered, based upon an assessment of the risk of bleeding complications and patient references
Use of NOACs	• Clinicians should administer dabigatran, rivaroxaban, or apixaban to patients who have NVAF requiring anticoagulant medication and are at higher risk of intracranial bleeding • Clinicians might offer apixaban to patients with NVAF and GI bleeding risk who require anticoagulant medication • Clinicians should offer dabigatran, rivaroxaban, or apixaban to patients unwilling or unable to submit to frequent periodic testing of INR levels
NOACs contraindication	• NOACs (dabigatran, rivaroxaban, and apixaban) are not recommended in patients with severe renal impairment (CrCl <30 ml/min)

CrCl, creatinine clearance; NOACs, new oral anticoagulants; AF, atrial fibrillation; OAC, oral anticoagulants; VKA, vitamin K antagonist; INR, International normalized ratio; BID, twice daily; OD, once daily.

The recommendations made by this consensus statement are summarized in Table 15. For patients with NVAF, direct thrombin inhibitors (Dabigatran) or factor Xa inhibitors (rivaroxaban or apixaban) may be preferred to all patients and specially those who are unable to maintain the target INR levels with warfarin. Similarly, for patients who are unable or unwilling to submit to the frequent periodic testing of INR levels, dabigatran, rivaroxaban, or apixaban may be offered. Edoxaban may be included in any upcoming guidelines, and as more clinical experience accumulates, the management of stroke prevention in NVAF may see more refined recommendations.

7. Future scope for research

There are limited data on the community prevalence of AF and focused studies are needed to delineate the rural and urban prevalence of AF separately. The utility of the CHA₂DS₂-VASc and HAS-BLED scoring in Indian populations has not been studied and further studies focusing on validating these scores in an Indian population are required. As AF patients in India often have a rheumatic valvular component, the utility of NOACs in this population without mechanical valves needs to be studied in detail in an Indian setting. Different studies are ongoing (currently recruiting), including use of NOAC in special situations like post-PCI (RE-DUAL PCI and AUGUSTUS), postablation (RE-CIRCUIT and VENTURE AF), and other conditions like postcardioversion and in CKD patients. The consensus statement will be updated as and when the results of these trials become available.

Conflicts of interest

The authors are on the advisory panel of Boehringer Ingelheim (India) Pvt. Ltd.

Funding

Supported by an educational grant from Boehringer Ingelheim (India) Pvt. Ltd.