Vitamin K antagonists: relative strengths and weaknesses vs. direct oral anticoagulants for stroke prevention in patients with atrial fibrillation

Journal of Thrombosis and Thrombolysis, Dec 2016

Vitamin K antagonists (VKAs) have been the mainstay of anticoagulation therapy for more than 50 years. VKAs are mainly used for the prevention of stroke in patients with atrial fibrillation (AF) and the treatment and secondary prevention of venous thromboembolism. In the past 5 years, four new agents—the direct factor Xa inhibitors apixaban, edoxaban and rivaroxaban and the direct thrombin inhibitor dabigatran [collectively known as direct oral anticoagulants (DOACs) or non-VKA oral anticoagulants]—have been approved for these and other indications. Despite these new treatment options, the VKA warfarin currently remains the most frequently prescribed oral anticoagulant. The availability of DOACs provides an alternative management option for patients with AF, especially when the treating physician is hesitant to prescribe a VKA owing to associated limitations, such as food and drug interactions, and concerns about bleeding complications. Currently available real-world evidence shows that DOACs have similar or improved effectiveness and safety outcomes compared with warfarin. Treatment decisions on which DOAC is best suited for which patient to maximize safety and effectiveness should take into account not only clinically relevant patient characteristics but also patient preference. This article reviews and highlights real and perceived implications of VKAs for the prevention of stroke in patients with non-valvular AF, with specific reference to their strengths and weaknesses compared with DOACs.

A PDF file should load here. If you do not see its contents the file may be temporarily unavailable at the journal website or you do not have a PDF plug-in installed and enabled in your browser.

Alternatively, you can download the file locally and open with any standalone PDF reader:

https://link.springer.com/content/pdf/10.1007%2Fs11239-016-1446-0.pdf

Vitamin K antagonists: relative strengths and weaknesses vs. direct oral anticoagulants for stroke prevention in patients with atrial fibrillation

Vitamin K antagonists: relative strengths and weaknesses vs. direct oral anticoagulants for stroke prevention in patients with atrial ibrillation Andreas Zirlik 0 Christoph Bode 0 Andreas Zirlik 0 0 Department of Cardiology and Angiology I, University Heart Centre Freiburg , Hugstetter Strasse 55, 79106 Freiburg , Germany Vitamin K antagonists (VKAs) have been the mainstay of anticoagulation therapy for more than 50 years. VKAs are mainly used for the prevention of stroke in patients with atrial ibrillation (AF) and the treatment and secondary prevention of venous thromboembolism. In the past 5 years, four new agents-the direct factor Xa inhibitors apixaban, edoxaban and rivaroxaban and the direct thrombin inhibitor dabigatran [collectively known as direct oral anticoagulants (DOACs) or non-VKA oral anticoagulants]-have been approved for these and other indications. Despite these new treatment options, the VKA warfarin currently remains the most frequently prescribed oral anticoagulant. The availability of DOACs provides an alternative management option for patients with AF, especially when the treating physician is hesitant to prescribe a VKA owing to associated limitations, such as food and drug interactions, and concerns about bleeding complications. Currently available real-world evidence shows that DOACs have similar or improved efectiveness and safety outcomes compared with warfarin. Treatment decisions on which DOAC is best suited for which patient to maximize safety and efectiveness should take into account not only clinically relevant patient characteristics but also patient preference. This article reviews and highlights real and perceived implications of VKAs for the prevention of stroke Anticoagulants; Antithrombins; Atrial ibrillation; Factor Xa inhibitors; Stroke - in patients with non-valvular AF, with speciic reference to their strengths and weaknesses compared with DOACs. Vitamin K antagonists (VKAs) provide efective antico agulation and have been the mainstay of anticoagulation therapy for more than 50  years. VKAs are mostly used as long-term anticoagulant therapy, including for the prevention of stroke in patients with atrial ibrillation (AF) and the treatment of venous thromboembolism (VTE). Warfarin currently remains the most frequently prescribed oral anticoagulant (OAC) for these indications, even with the approval within the past 5  years of four new agents, namely the direct factor Xa inhibitors apixaban, edoxaban and rivaroxaban and the direct thrombin inhibitor dabigatran [collectively known as novel/non-VKA/direct OACs (DOACs)]. On average, patients with AF have a ive-fold higher risk of stroke than patients without AF—irrespective of whether they have paroxysmal or chronic AF [123]. AF-related strokes are associated with an approximately 50% increased risk of disability and a 60% increased risk of death at 3  months compared with strokes of other aetiologies [83]. The number of strokes caused by AF-related thromboembolisms may be even higher than currently thought because data from recent studies have shown that cryptogenic strokes (i.e. those without a well-deined aetiology) account for approximately 30% of ischaemic strokes [112]. Evidence suggests that up to 30% of patients with cryptogenic stroke may have AF [54, 110]. Therefore, efective anticoagulation is vital for the long-term management of patients with AF at an intermediate or high risk of stroke [27, 125]. VKAs reduce the rate of stroke by approximately 60% [65], whereas antiplatelet therapy is much less eica cious (reducing the event rate by approximately 20%) and has almost the same bleeding risk as oral anticoagulation therapy [2, 85]. Despite guideline recommendations and clear evidence that oral anticoagulation therapy is indicated in patients with AF and who have a CHA2DS2-VASc score of ≥1, several studies report that, on average, only 60% of eligible patients receive anticoagulation therapy. The pharmacological characteristics of VKAs, particularly their narrow therapeutic window combined with many drug–drug and drug–food interactions, necessitate regular coagulation monitoring and dose adjustments [3, 5]. An important measure for anticoagulation control with VKAs is the percentage of time a patient spends within the target therapeutic range [i.e. international normalized ratio (INR) of 2.0–3.0]. A number of studies have shown that INR control of VKA therapy is suboptimal in routine clinical practice and, in general, patients spend approximately 40% of time outside the recommended INR range. Poor INR control is associated with increased risk of stroke (INR <2.0) and bleeding (INR >3.0) [88, 122]. Real-life evidence demonstrates that time in therapeutic range (TTR) also varies depending on care setting, such as whether patients are managed by a dedicated anticoagulation clinic or by their general practitioner, cardiologist or neurologist [91]. DOACs have been developed to overcome the limitations of VKA therapy. These agents are now approved in many countries worldwide for the prevention of stroke and systemic embolism in patients with non-valvular AF, as well as for other thromboembolic disorders (Table 1) based on data from phase III, randomized clinical trials [11, 20, 22, 38]. DOACs were at least as good as, if not superior to, warfarin in terms of eicacy for the prevention of stroke in patients with non-valvular AF and also ofered a better safety proile (all four major trials consistently indicated reduced rates of intracranial and fatal or life-threatening bleeding compared with the respective warfarin arm) [32, 52, 58, 102, 109]. Real-world evidence of DOACs is accumulating, and available data support the indings of the phase III clinical studies (for example [13, 15, 16, 18, 26, 117, 121]). In general, current guidelines recommend DOACs in preference to VKAs [27, 114] or as an alternative to VKAs [74, 94] for prevention of stroke or systemic embolism in patients with non-valvular AF. However, VKAs are still regarded as the ‘gold standard’ by some physicians and continue to be prescribed to many patients, including those who have had diiculties maintaining their INR within the target therapeutic range [48, 88]. This article highlights real and perceived implications of VKAs for the prevention of stroke in patients with non-valvular AF, with speciic reference to their strengths and weaknesses compared with DOACs. Furthermore, it provides practical guidance on which patients should be switched from VKA to DOAC therapy, which patients should stay on VKA therapy and which DOAC should be given to which patient. Finally, this paper discusses the most suitable overall approach to reducing the burden of AF-related stroke. Characteristics of vitamin K antagonist therapy: why it works and areas of inadequacy The pharmacological characteristics of diferent VKAs, such as warfarin, phenprocoumon and acenocoumarol (Table 2) are associated with several advantages and practical limitations [3, 44]. VKAs have several inherent advantageous characteristics. They are not eliminated by the kidneys and, therefore, can be used in patients with severe renal impairment. Moreover, the need for regular INR monitoring encourages regular physician–patient contact despite being inconvenient and imposing additional costs. However, although regular physician visits may be beneicial from a medical point of view, poor medication adherence is usually attributable to multiple, interlinked factors and there is no evidence that regular physician visits alone can increase patient adherence [23]. In the case of a missed VKA dose, patients are at less immediate risk of a thrombotic event than patients missing a dose of DOAC, and non-adherent patients may beneit from the slow ofset of action. However, (similar to initia tion of therapy) reinitiating therapy after missing several doses of a VKA may actually result in a profound prothrombotic state [3, 8]. Many physicians are highly familiar with the management and the responsible use of VKAs. Moreover, drug costs of VKAs are signiicantly lower than those of DOACs. Therefore, physicians may be hesitant to prescribe any of the DOACs. On the downside, VKAs have an indirect anticoagulant mechanism of action, impairing the synthesis of several vitamin K-dependent coagulation factors (Fig.  1), which results in a slow onset and ofset of the anticoagulant efect. On initiation, VKAs are inherently prothrombotic (a fact often overlooked by physicians) because they inhibit the natural anticoagulant proteins C and S faster than inhibiting the coagulation factors X, IX, VII and II: this creates a temporary imbalance in favour of procoagulation factors [3, 8]. In a large case-control study in more than 70,000 patients with AF, warfarin was associated with a 71% increased risk of stroke in the irst 30  days of treatment compared with longer periods of treatment [8]. Hence, bridging therapy with a fast-acting, parenteral anticoagulant (e.g. Table 1 Indications and dosing regimen of DOACs in the EU [11, 20, 22, 38] Prevention of stroke and systemic embolism in patients with non-valvular AF with ≥1 risk factors Prevention of atherothrombotic events in patients with elevated cardiac biomarkers after an ACS in combination with antiplatelet therapy 10 mg bid for 7 days followed by 5 mg bid; 2.5 mg bid for prevention of recurrence (following 6 months of treatment) Direct thrombin inhibitor 60 mg od (following paren- 15 mg bid for 3 weeks 150 mg bid (following parteral anticoagulant for at followed by 20 mg enteral anticoagulant for at least 5 days)d ode least 5 days)f ACS acute coronary syndrome, AF atrial ibrillation, bid twice daily, DOAC direct oral anticoagulant, DVT deep-vein thrombosis, od once daily, PE pulmonary embolism, VTE venous thromboembolism aNot recommended in patients with CrCl <15 mL/min bContraindicated in patients with CrCl <30 mL/min cStarted with a half dose 1–4 h after completion of surgery followed by full doses from the next day onwards; reduced dose of 150 mg od (taken as two tablets of 75 mg) in patients with one or more of the following: CrCl 30–50 mL/min; receiving concomitant verapamil, amiodarone or quinidine; aged ≥75 years dReduced dose of 30 mg od in patients with non-valvular AF or VTE plus one or more of the following clinical factors: CrCl 15–50 mL/min; low body weight ≤60 kg; concomitant use of the following P-glycoprotein inhibitors: cyclosporine, dronedarone, erythromycin or ketoconazole eAfter the initial dosing period of 15  mg bid for 3  weeks, a reduced dose of 15  mg od should be considered if the patient’s assessed risk for bleeding outweighs the risk for recurrent VTE fReduced dose of 110  mg bid in patients with non-valvular AF or VTE aged ≥80  years or receiving concomitant verapamil; consider this reduced dose based on individual assessment of thromboembolic risk and bleeding risk in: patients aged 75–80 years, patients with CrCl 30–49 mL/min; patients with gastritis, oesophagitis or gastroesophageal relux, and other patients at increased risk of bleeding gReduced dose of 2.5 mg bid in patients with non-valvular AF and serum creatinine ≥1.5 mg/dL (133 µmol/L) plus age ≥80 years and/or body weight ≤60 kg hReduced dose of 15 mg od in patients with non-valvular AF and CrCl 15–50 mL/min enoxaparin, unfractionated heparin) is necessary on initiation of VKA therapy. For surgery or other interventional procedures, the slow ofset of action may delay the pro cedure. Furthermore, bridging therapy with a fast-acting anticoagulant may also be necessary after the procedure to ensure eicient anticoagulation. VKAs have a narrow therapeutic range (Fig.  2) and, therefore, require regular coagulation monitoring and dose adjustments in some patients to keep the anticoagulation intensity within the therapeutic range [3]. Data show that patients on VKAs are efectively anticoagulated only approximately 60% of the time, or even less in some countries [6, 88]. Keeping patients within the target therapeutic range is further complicated by VKAs having multiple food and drugs interactions [3]. These factors can have a signii cant impact on patients’ daily lives, such as considerable time spent in the clinic for coagulation monitoring and dietary restrictions, all of which may reduce patients’ quality of life [3]. Furthermore, routine anticoagulation monitoring confers additional costs: the cost of the test itself, travel, nurse visits, missed work and the increased workload for physicians and other healthcare staf [ 1]. Characteristics of the direct oral anticoagulants: what renders them so attractive and what are their limitations? The pharmacological characteristics of DOACs provide many practical advantages over VKA therapy (Table  2). Direct targeting of factor Xa or thrombin allows for a much faster efective anticoagulation efect—within 0.5–4 h [ 51, CYP cytochrome P450, NR not reported, P-gp P-glycoprotein, R- (R)-enantiomer, S- (S)-enantiomer, tmax time to reach maximal plasma concentration aRivaroxaban 20 mg: 66% under fasting conditions (mean area under the plasma concentration–time curve increased by 39% when given with food) bThe 15 and 20 mg doses of rivaroxaban should be taken with food to enhance their absorption Table 2 Overview of pharmacological characteristics of direct oral anticoagulants and vitamin K antagonists [3, 59, 66, 71, 106, 119] Dabigatran Apixaban Rivaroxaban Edoxaban Warfarin Factor Xa Factor Xa Fig. 1 Coagulation cascade with sites of inhibitions for VKAs and direct oral anticoagulants indicated. Coagulation factors are indicated using their factor numbers in roman numerals, with ‘a’ indicating an active factor. TF tissue factor, VKA vitamin K antagonist 62, 89, 92]—and a faster ofset of action as opposed to the indirect mode of action via multiple coagulation factors as in the case of VKAs (Fig. 2). Moreover, the kinetics of DOACs (e.g. rivaroxaban [78]) closely mimic those of the low-molecular-weight heparin enoxaparin. Therefore, in contrast to VKAs, bridging with a parenteral anticoagulant is not necessary with the DOACs [66]. DOACs also have a much shorter half-life compared with VKAs, making bridging to interventions or surgery obsolete [67]. DOACs have predictable pharmacokinetics and pharmacodynamics and a lower potential for food and drug interactions [11, 20, 22, 38]. These agents can, therefore, be given at ixed dosing schedules without the need for dietary restrictions or routine coagulation monitoring. However, the lack of the requirement for routine monitoring does not negate the need for regular physician–patient contact and patients should schedule regular visits. Although the frequency of these visits should be determined by bleeding risk (HAS-BLED score), age and renal function, patients are recommended to return every 3 months for a review of their treatment [67]. Measurement of the anticoagulation efect and/or drug levels may be helpful in certain clinical circumstances, such as in the event of suspected overdose, during bleeding events, prior to urgent surgery, in patients with deteriorating renal function or when determining the use of thrombolysis. Unlike VKAs, DOACs are eliminated renally, albeit at diferent rates; renal impairment afects exposure and the associated risk of bleeding. Renal clearance of the absorbed active drug is approximately 27% for apixaban [22], 35% for rivaroxaban [92], 50% for edoxaban [38] and >80% for dabigatran [46]. Based on these characteristics, apixaban, edoxaban and rivaroxaban are not recommended in patients with AF and who have creatinine clearance (CrCl) rs 16 a ye 14 t ien 12 t pa 10 0 0 8 1 r e 6 p tra 4 e ke 2 o tS 0 r INR Fig. 2 Observed and predicted risk of a ischaemic stroke and b haemorrhagic stroke according to INR [5]. Reprinted from European Journal of Internal Medicine, Vol 20, Amouyel P, Mismetti P, Langkilde LK, et al. INR variability in atrial ibrillation: A risk model for cerebrovascular events. Pages 63–69, Copyright 2009, with permissions from Elsevier <15 mL/min [11, 22, 38] and dabigatran is contraindicated in patients with CrCl <30  mL/min [20]. Furthermore, edoxaban should be used with caution in patients with high CrCl because of reduced eicacy [ 38]; In the US, edoxaban should not be used in patients with CrCl >95 mL/min [39]. VKAs can only provide clinical beneit if the antico agulation efect is kept within the therapeutic range (INR 2.0–3.0); poor INR control can lead to an increased risk of thromboembolism (INR <2.0) or bleeding (INR >3.0) [5]. TTR during VKA therapy is higher during controlled clinical studies than in daily practice, owing to their strict study protocols and the regular follow-up with patients. Moreover, TTR control in daily clinical practice is also dependent on whether the patient is managed by a dedicated anticoagulation clinic or elsewhere, such as by a general practitioner, cardiologist or neurologist [91]. The rate of fatal and major bleeding events was low (0.25 and 1.1% per year, respectively) in patients whose anticoagulation with warfarin was managed by an anticoagulation clinic achieving a median TTR of 68% [91, 101]. The inability to maintain TTR is well reported: in the Registry of Canadian Stroke Network, 74% of patients with known AF who were taking warfarin at the time of ischaemic stroke had sub-therapeutic anticoagulation [53]. Additional evidence emphasizes that stroke prevention with a VKA is efective in patients who have a good individual mean TTR (>75%) [91]. Data from the GARFIELD-AF registry indicate that only 29% of VKA-treated patients had good anticoagulation control, deined as a TTR ≥70%, and that heavy alcohol use was associated with poor anticoagulation control (TTR ≤60%) [118]. Patients with poor control had a signiicantly higher risk of death [hazard ratio 2.87; 95% conidence interval (CI) 1.97–4.19] and stroke/ systemic embolism (hazard ratio 1.98, 95% CI 1.13–3.47) than those with a TTR >60% [118]. Data collected outside of anticoagulation clinics (and, therefore, most likely in patients with suboptimal anticoagulation control) suggest that real-world efectiveness fails to reproduce eicacy data for VKAs seen in clinical stud ies. However, warfarin has been shown to prevent stroke and systemic embolism more efectively than placebo or acetylsalicylic acid. A large meta-analysis of clinical study data demonstrated a 62% reduction in the risk of stroke and systemic embolism with warfarin therapy compared with placebo/acetylsalicylic acid therapy [64]. A real-world Canadian study in patients with AF showed that warfarintreated outpatients had a signiicantly lower risk of stroke compared with patients who did not receive any antithrombotic treatment (risk ratio 0.31) [29]. The risk of bleeding is much higher in clinical practice compared with the rates reported in clinical studies. A large cohort study in 125,195 patients with AF demonstrated a high risk of haemorrhage during the irst 30 days of warfarin therapy (11.8% per year): considerably higher than the rates of 1–3% reported in randomized controlled trials [55]. In the GARFIELD-AF registry, treatment at an anticoagulation clinic or thrombosis centre was associated with a better TTR compared with other settings (proportion of patients with TTR >60%: 57.1 vs. 46.2%) [118]. There are eforts to simplify the management of VKAs by way of patient self-testing and self-management. Initial clinical trials had encouraging outcomes [19]; however, patients in these trials had a high level of education, which is not necessarily a true representation of all patients encountered in daily clinical practice. Finally, studies trying to optimize the beneit–risk ratio of VKAs by lowering the INR range to 1.5–2.5 failed, with inferior eicacies but similar bleeding compared with standard-dose VKA therapy [103]. Patient preference and compliance to vitamin K antagonist therapy Limitations and inconveniences that both physicians and patients associate with VKA therapy are contributing to their under-prescription in patients with high risk of stroke and systemic embolism. In the GARFIELD-AF registry, 38.0% of patients with a CHADS2 score ≥2 did not receive anticoagulant therapy; 7.2% of patients with AF and CHADS2 ≥2 had refused treatment for various reasons, including inconvenience of regular blood tests, dietary restrictions, bleeding risk and an under-appreciation or lack of knowledge regarding the risk of stroke. As well as being unwilling to start VKA therapy, many patients with AF who are initiated on VKA therapy discontinue or are non-adherent [47, 56, 77, 98]. For example, of 125,195 patients newly diagnosed with AF in Canada from 1997 to 2008, 9% did not collect their second prescription of warfarin within the irst half year and 32% discontinued therapy within 1 year, rising to 43% at 2 years and 61% at 5 years [56]. Similarly, in a US study, more than one in four new warfarin starters discontinued therapy within a year [47]. In another study, 40% of patients were non-adherent to VKA therapy (>20% of days with missed doses or >10% of days where extra doses were taken in addition to the prescribed dose), and this percentage was signiicantly associ ated with poor anticoagulation control [77]. Large registries published between 2005 and 2009 by the European Heart Survey, the German Competence NETwork on AF (AFNET) and the Canadian Stroke Network suggest that 30–60% of patients with AF who are eligible according to guidelines are not prescribed anticoagulation therapy [53, 93, 97]. In the Registry of Canadian Stroke Network, only 10% of patients with acute stroke with known AF were therapeutically anticoagulated at time of hospital admission [53]. Underuse of anticoagulation in these patients had unfavourable implications: approximately 80% of the resulting strokes were disabling or fatal [53]. The global GARFIELD-AF registry (2009–2011) found that 34% of patients with a CHADS2 score ≥2 received antiplatelet therapy [76]. Of patients with a CHADS2 score ≥2 who received anticoagulation therapy, 62% received a VKA. In addition, 43% of patients with a CHADS2 score of 0 received anticoagulation therapy. Taken together, there appears to be underuse of anticoagulation therapy in patients at moderate to high risk of stroke and systemic embolism and overuse in patients at low risk—demonstrating that, in real-life practice, prescribed therapy is often not based on evidence-based risk schemes and guidelines [76]. In almost half of the cases (48.3%) in which VKA therapy was not prescribed, this was the physician’s choice and not based on guidelines or contraindications to therapy; the physician’s reasons included concerns over bleeding risk (7.4%), concerns over the risk of falling (6.5%), concerns over patient compliance (5.3%) and perceived low risk of stroke (4.1%). Many of the concerns given as reasons for not prescribing VKA are not supported by actual data. For example, a prospective study in patients on OACs at high risk of falls did not have a signiicantly increased risk of major bleeding events [ 40]. Poor TTR in everyday clinical practice, coupled with low rates of adherence or high discontinuation rates, and a general underuse of VKA therapy supports the need for alternative oral anticoagulation options that are easier to manage and more convenient than VKA therapy. In the next sections, we review clinical studies and real-world data, with the practical advantages associated with DOAC therapy compared against the aforementioned limitations associated with VKAs. Vitamin K antagonists vs. direct oral anticoagulants: outcomes of phase III studies and real-life evidence Results from phase III trials of DOACs, with a wide range of patients with AF worldwide, showed that all DOACs are at least as efective as warfarin, with similar or lower rates of major bleeding [32, 34, 52, 58, 102]. Importantly, a direct, head-to-head comparison of these studies is not feasible because the study designs and study populations were diferent. In a meta-analysis of all four DOACs in phase III trials for stroke/systemic embolism prevention in patients with AF vs. warfarin, these agents reduced the risk of haemorrhagic stroke by 51% and the risk of intracranial haemorrhage by 52% [109].This favourable beneit–risk proile extends to many subgroups and ethnicities including the Asian population in which the rate of intracranial bleeding is reduced by up to 80% (as reviewed elsewhere [31, 72, 73]). Conversely, the risk of gastrointestinal bleeding was 25% higher with DOACs than with warfarin, owing to bleeding events with dabigatran (150  mg twice daily), edoxaban (60  mg once daily) and rivaroxaban (20  mg) (Tables 3, 4) [32–34, 43, 52, 58, 102, 109]. There are several studies comparing real-life efective ness and safety of VKAs with DOACs (mainly dabigatran or rivaroxaban vs. warfarin; data for apixaban are emerging; data for edoxaban are currently lacking; Table  S1 in the electronic supplementary material). Published studies to date demonstrate similar or improved efectiveness with DOACs compared with VKAs (Table  S1 in the electronic supplementary material). Recent publications showed discrepancies in real-world efectiveness and safety outcomes with DOACs compared with previously published database analyses or compared with phase III clinical trial results (Table 3). Real-life evidence from the international, non-interventional, observational phase IV XANTUS study demonstrates that rates of stroke and major bleeding were low in patients receiving rivaroxaban [26]. Data from the Dresden NOAC Registry suggest that rates of major bleeding may be lower with rivaroxaban, apixaban and dabigatran therapy compared with VKA therapy [13, 16, 18, 90]. Moreover, these data show that real-life rates of major bleeding with rivaroxaban were similar (Dresden NOAC Registry [16, 102]) or lower (XANTUS [26]) compared with ind ings from ROCKET AF [16, 102]. Other observational studies mainly demonstrate that rivaroxaban and dabigatran have similar or reduced rates of major bleeding compared with VKAs, and relect the decreased incidence of intracra nial haemorrhage and increased incidence of gastrointestinal bleeding [30]. Adherence, persistence and discontinuation Adherence is deined as the extent to which the patient acts in accordance with the prescribed interval and dose of the dosing regimen and can also be deined as the percent of doses taken as prescribed [36]. Persistence measures the duration of drug therapy during which the patient takes medication without exceeding the permissible gap (usually 60  days). Two retrospective US database analyses showed that patients with AF were signiicantly more persistent with rivaroxaban than with warfarin, reporting patient persistence with warfarin dropping to <70% at 6  months of therapy [80, 95]. A retrospective US database analysis demonstrated that persistence was higher with dabigatran than with warfarin at 6  months (72 vs. 53%) and 1  year (63 vs. 39%) [126]. This study also showed that patients with a low-to-moderate stroke risk (CHADS2 <2) or with a higher bleeding risk (HEMORR2HAGES >3) were more likely to discontinue treatment than patients with a high stroke risk or lower bleeding risk [126]. When comparing persistence or adherence among DOACs, two retrospective analyses of diferent US databases demonstrated that use of the once-daily medication rivaroxaban was associated with ARISTOTLE [58] ENGAGE AF [52] (Apixaban vs. warfarin) (Edoxaban vs. warfarin) RE-LY [32–34, 43] (Dabigatran vs. warfarin) 2.92 vs. 3.61b,d 3.40 vs. 3.61b,d 3.6 vs. 3.4 2.13 vs. 3.09d 1.61 vs. 3.43 2.75 vs. 3.43 0.19 vs. 0.33 0.23 vs. 0.33 0.2 vs. 0.5 NR (34 vs. 55 patients) 0.13 vs. 0.38 0.21 vs. 0.38 0.23 vs. 0.76 0.32 vs. 0.76 0.5 vs. 0.7 0.33 vs. 0.80 0.26 vs. 0.85 0.39 vs. 0.85 1.36 vs. 1.25 1.85 vs. 1.25 3.2 vs. 2.2 0.76 vs. 0.86 0.82 vs. 1.23 1.51 vs. 1.23 14.66 vs. 18.23b,e 16.45 vs. 18.23b,e 14.9 vs. 14.5f 4.07 vs. 6.01 7.97 vs. 13.02 11.10 vs. 13.02 Table 3 Main eicacy and safety results from the phase III clinical trials of the direct oral anticoagulants approved for prevention of stroke in patients with non-valvular atrial ibrillation Values in bold indicate a statistically signiicant diference between the direct oral anticoagulant and warfarin GI gastrointestinal, ICH intracranial haemorrhage, NR not reported, NMCR non-major clinically relevant, SE systemic embolism aIntention-to-treat analysis bData with additional events as per [34] or [33] or [43] cPrimary eicacy endpoint in ENGAGE-AF was time to adjudicated stroke or systemic embolic event dPrimary safety outcome in RE-LY and ARISTOTLE eMajor or minor bleeding (minor bleeding was any bleeding not considered to be a major bleeding event) fPrimary safety outcome in ROCKET AF Table 4 Subgroup analyses from the phase III clinical trials of direct oral anticoagulants for prevention of stroke in patients with non-valvular atrial ibrillation (there are currently no subgroup analyses of ENGAGE AF data available for subgroups speciied in the table) ROCKET AF subgroups (Rivaroxaban vs. warfarin) ARISTOTLE subgroups (Apixaban vs. warfarin) RE-LY subgroups (Dabigatran vs. warfarin) Statistically signiicant values ( p ≤ 0.05) are given in bold CrCl creatinine clearance, HF heart failure, HF-pEF heart failure with preserved ejection fraction, HF-LVSD heart failure caused by left ventricular systolic dysfunction, MI myocardial infarction, NR not reported, SE systemic embolism aValues are for CrCl <50 mL/min. bValues are for CrCl ≤50 mL/min. cValues are for coronary artery disease deined as documented coronary artery disease, history of MI and/or history of coronary revascularization signiicantly higher rates of persistence at 1-year follow-up or signiicantly higher adherence (percentage of patients who had a proportion of days covered ≥80% during their follow-up) than with the use of the twice-daily medication dabigatran [37, 96]. A Danish nationwide cohort study in approximately 3000 patients with non-valvular AF reported that over 75% of patients treated with dabigatran adhered to therapy more than 80% of the time (as measured by proportion of days covered) [57]. Published data on reallife adherence with edoxaban and apixaban are not yet available. In the phase III studies (across various follow-up periods), discontinuation rates were: signiicantly lower with apixaban compared with warfarin in ARISTOTLE; similar between rivaroxaban and warfarin in ROCKET AF and between edoxaban and warfarin in ENGAGE-AF; but signiicantly higher with dabigatran compared with warfarin in RE-LY, mainly owing to dyspepsia [32, 52, 58, 102]. VKA discontinuation rates in real-life practice range from 25 to 38% at 1-year follow-up and are higher than those reported in controlled phase III studies (10–35% over a median follow-up period of 1.8–2.8  years) [13, 32, 52, 58, 95, 102, 109]. In an analysis of data collected from patients with AF in the Dresden NOAC Registry, discontinuation rates with dabigatran (25.8% per year) were similar to those observed with VKAs in daily practice, whereas discontinuation rates with rivaroxaban therapy (13.6% per year) were much lower than those with VKA therapy [12, 15]. Persistence probabilities at 1 year were 53.1, 47.3 and 25.5% with rivaroxaban, dabigatran and VKA, respectively, and adherence with a high medication possession ratio (≥80%) was 61.4% for rivaroxaban and 49.5% for dabigatran [14]. Together, real-life data suggest that, in the long-term, patients receiving DOACs have better protection against stroke or systemic embolism than patients receiving a VKA. In the US, 33–69% of all medication-related hospital admissions are estimated to be attributable to poor medication adherence, with the resulting costs of nonadherence being approximately $100  billion/year [100, 115]. An analysis of adverse events based on hospital data identiied warfarin as a medication that was most commonly implicated in hospitalization of adults aged ≥65 years (33.3%) owing to adverse drug efects [ 24]. The true cost of vitamin K antagonist therapy In addition to treatment efectiveness and safety, cost-efec tiveness is another consideration for decision making by healthcare professionals who have several therapy options. VKAs are often perceived to have lower costs; however, although costs for the drug itself are lower when comparing with DOACs, the true cost of VKA treatment needs to take into account the expenses related to the general management of therapy. These include routine coagulation monitoring, adverse clinical outcomes during therapy (such as bleeding and thromboembolic events) and as a result of non-adherence. In clinical practice, the estimated mean numbers of hospitalization days, outpatient visits and AF-related hospitalizations associated with rivaroxaban are reported to be lower than those associated with warfarin [79, 81, 82]. Similar published real-world evidence is not yet available for apixaban, edoxaban or dabigatran. A cost-modelling analysis suggests that, based on the expected number of thrombotic or bleeding events avoided with use of DOACs vs. warfarin, medical costs are reduced when DOACs are used instead of warfarin/placebo for the prevention of stroke in patients with non-valvular AF or for the treatment of VTE [4]. However, a model simulation based on the Slovenian healthcare payer perspective using 2014 costs demonstrated that cost-efectiveness of the DOACs vs. warfarin is highly sensitive to warfarin anticoagulation control [75]. With a TTR of 60%, the probability that warfarin was a cost-efective option was unlikely (probability 1%). This percentage rises with increasing TTR: at a TTR of 70%, warfarin was more cost-efective than DOACs in half of the simulations [75]. Reversal of anticoagulant efect and management of bleeding There is currently limited clinical experience with speciic reversal agents for the DOACs. However, although Table 5 Reversal agents for DOACs Reversal agent, description vitamin K is a direct, efective reversal agent for VKAs, a normal INR is generally only achieved over approximately 24 h, which would not help in the case of clinically important bleeding events such as intracranial haemorrhage [3, 84]. Therefore, coagulation factor concentrates need to be administered in parallel with vitamin K to restore haemostasis quickly [3, 84, 111]. In most clinical situations, the short half-lives of the DOACs obviate the need for reversal, and standard procedures for bleeding management are normally suicient to control bleeding events [28]. In fact, speciic reversal agents for DOACs would be very rarely needed in daily clinical care. In exceptional clinical situations (such as life-threatening bleeding or emergency surgery associated with a high bleeding risk), coagulation factor concentrates such as prothrombin complex concentrate, activated prothrombin complex concentrate or recombinant factor VIIa may be considered [11, 20, 22, 38]. However, there is limited clinical experience with these agents in patients with bleeding events. Haemostatic agents such as prothrombin complex concentrate or recombinant factor VIIa may increase the risk of thromboembolism if they are administered when the plasma concentration of the anticoagulant is low [124]. Therefore, the risk with the use of these agents has to be balanced with their potential for bleeding control. Recent months have seen the clinical approval of idarucizumab, a speciic reversal agent for dabigatran, based on results of a phase III study [104, 105] (Table 5). Moreover, results with a speciic reversal agent, andexa net alfa (http://www.clinicaltrials.gov, NCT02220725 and NCT02329327) for factor Xa inhibitors have shown that it has the potential to quickly and efectively reverse the anticoagulation efect of rivaroxaban and apixaban [113]. Andexanet alfa is expected to be approved in 2017 [107]. PER977 (Perosphere) is being assessed as a reversal agent for edoxaban in clinical trials with promising preliminary results [7, 35] (http://www.clinicaltrials.gov, NCT02207257). Approval status References Dabigatran Idarucizumab (Praxbind®): a fully humanized, monoclonal antibody fragment FDA and EMA [25, 104, 105] designed to speciically reverse the anticoagulant efect of dabigatran Factor Xa inhibitors (rivaroxa- Andexanet alfa: an inactive, recombinant version of the human factor Xa Anticipated [113] ban and apixaban tested) designed to speciically reverse the anticoagulant efect of factor Xa inhibi- approval in tors 2016 All DOACs, UFH and LMWH PER977: a small, synthetic, water-soluble, cationic molecule that is designed In development [7, 35] to bind speciically to UFH and LMWH through non-covalent hydrogen bonding and charge–charge interactions and similarly also binds to edoxaban, rivaroxaban, apixaban and dabigatran DOAC direct oral anticoagulant, EMA European Medicines Agency, FDA US Food and Drug Administration, LMWH low-molecular-weight heparin, UFH unfractionated heparin When to switch and when not to switch from vitamin K antagonists to direct oral anticoagulants Patients who have been initiated on VKA therapy can be switched to a DOAC (see individual Summary of Product Characteristics for further details [11, 20, 22, 38]). This switch should be based on a clinical beneit–risk assess ment. Speciic reasons for switching may include—but are not limited to—poor INR control, stroke/systemic embolism or serious bleeding during VKA therapy, poor compliance (e.g. relating to the inconveniences of VKA therapy), patient preference to switch to a DOAC therapy, reduced long-term costs and fear of bleeding (particularly within the fragile patient population). Switching strategies are reviewed in the updated practical guide of the European Heart Rhythm Association (EHRA) [66]. The efectiveness and safety of switching patients with AF from a VKA to DOAC therapy has been demonstrated in the Dresden NOAC Registry [17, 90]. Data from this registry regarding patients who switched from a VKA to rivaroxaban or dabigatran for stroke prevention or VTE treatment suggest that the potential for bleeding should be monitored carefully in the irst few days after the transi tion, during which residual VKA activity may remain [90]. One study reported that only 75% of VKA patients had an INR measurement documented before they were started on a DOAC; on average, DOAC was started within 2–5  days after the last intake of VKA. At the 30-day follow-up, the rates of major cardiovascular events (0.8%; 95% CI 0.3–1.8) and major bleeding complications (0.3%; 95% CI 0.0–1.0) were low, with a rate of any bleeding of 12.2% (95% CI 9.8–14.8) in patients with and without INR testing of the residual VKA efect [ 17]. A Danish analysis demonstrated the importance of adherence to the switching protocols outlined in the Summary of Product Characteristics for dabigatran [116]. This study evaluated real-world outcomes in patients with AF: there was an increased risk of thromboembolism and bleeding with dabigatran in previous VKA users. The authors of this study cautiously interpreted these unexpected results as relecting patient selection and drugswitching practices. Dabigatran use in VKA-naïve patients was reported to be safe [116]. The EHRA practical guide provides a schematic overview of switching protocols from a VKA to a DOAC and vice versa and also emphasizes the importance of adherence to the established switching strategies [66]. Some patients, especially those with good INR control and TTR, may prefer to continue with VKA therapy instead of switching to a DOAC. Patients may also beneit from continued VKA therapy, including those with contraindications to DOAC therapy. For example, patients with endstage kidney disease (CrCl <15 mL/min) have signiicantly increased risks of stroke and bleeding compared with patients with normal kidney function [99]. End-stage kidney disease is also associated with reduced activity of cytochrome P450 2C9, leading to lower warfarin dosing requirements [41]. Patients with valvular AF as opposed to non-valvular AF (particularly in those with mechanical valves) should be treated with a VKA because DOACs are not approved in these patients [11, 20, 22, 38]. Moreover, the outcomes of the RE-ALIGN trial that assessed dabigatran vs. warfarin in patients with mechanical valves reinforced the recommendations of the current guidelines against the use of DOACs in these patients. This trial had to be terminated prematurely owing to an excess of thromboembolic and bleeding events among patients receiving dabigatran (150, 220 or 300 mg twice daily) [42]. Which direct oral anticoagulant for which patient? In the absence of a head-to-head trial with DOACs (no such trial is planned or ongoing), no direct answer can be provided to this question. The overall aim should be that all patients with AF who are indicated for anticoagulation should receive appropriate therapy. As discussed at the beginning of this article, a substantial proportion of patients with AF who should be receiving anticoagulation are not receiving OAC therapy of any form. DOACs—overcoming several of the limitations of VKAs—ofer alternative and potentially preferred therapy options [27] both in treatment-naïve patients with newly diagnosed AF in need of anticoagulation therapy and in patients with AF at risk of stroke and systemic embolism who are not receiving appropriate therapy or who have poorly controlled VKA therapy. The choice of which DOAC is the right agent for which patient, initially choosing between a direct thrombin inhibitor and factor Xa inhibitor, should be based on the pharmacokinetics/pharmacodynamics and integration of the clinical data with respect to the patient’s characteristics. The following recommendations, based on the EHRA practical guide, can be used for decision making [66]. In patients with renal impairment, factor Xa inhibitors (rivaroxaban, apixaban or edoxaban) should be preferred over dabigatran. Similarly, factor Xa inhibitors have demonstrated no change to the beneit–risk proile in elderly patients and in patients with a pronounced cardiovascular co-morbidity compared with other patient groups (with especially favourable data for rivaroxaban [60, 63, 86]). Patients with a history or high risk of gastrointestinal bleeding may have a lower risk of bleeding complications with apixaban and low-dose edoxaban than with dabigatran, rivaroxaban or high-dose edoxaban; however, dabigatran, rivaroxaban and warfarin may have similar rates of gastrointestinal bleeding in real-life clinical practice [30]. Furthermore, there is some evidence that patients with a high risk for ischaemic stroke may beneit from a direct thrombin inhibitor (i.e. dabigatran) [ 108]. More data from real-life studies will shed light on which agent provides the best beneit–risk ratio for which patient. The availability of DOACs provides an alternative management option for patients with AF, especially when the treating physician is hesitant to prescribe a VKA owing to the associated limitations, such as routine coagulation monitoring and dose adjustments, food and drug interactions and concerns about bleeding complications. Overall, currently available real-world evidence shows that DOACs have similar or improved efectiveness and safety outcomes com pared with warfarin. With regards to which DOAC is best suited for which patient to maximize safety and efective ness, more prospective real-world data are required because database studies show divergent outcomes. Overall, recommendations in the EHRA practical guide suggest  actions taking into account not only clinically relevant patient characteristics but also patient preferences. Adherence to therapy is an important factor to achieve best outcomes, and there is some evidence that patients adhere better to oncedaily medications compared with those taken twice-daily. Acknowledgements The authors would like to acknowledge Diana Selig for editorial assistance with the manuscript. The authors would also like to acknowledge Claudia Wiedemann, who provided editorial assistance with funding from Bayer AG and Janssen Scientiic Afairs, LLC. Compliance with ethical standards Conlict of interest The authors declare no conlicts of interest in this work, and received no funding for it. Funding for editorial assistance was provided by Bayer AG. Ethical approval This article does not contain any studies with human participants or animals performed by any of the authors. Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. 1. Abdelhaiz AH , Wheeldon NM ( 2003 ) Use of resources and cost implications of stroke prophylaxis with warfarin for patients with nonvalvular atrial ibrillation . Am J Geriatr Pharmacother 1 : 53 - 60 2. ACTIVE Writing Group of the ACTIVE Investigators ( 2006 ) Clopidogrel plus aspirin versus oral anticoagulation for atrial ibrillation in the atrial ibrillation clopidogrel trial with Irbe - sartan for prevention of vascular events (ACTIVE W): a randomised controlled trial . The Lancet 367 : 1903 - 1912 3. Ageno W , Gallus AS , Wittkowsky A , Crowther M , Hylek EM , Palareti G ( 2012 ) Oral anticoagulant therapy: antithrombotic therapy and prevention of thrombosis, 9th ed: American college of chest physicians evidence-based clinical practice guidelines . Chest 141 : e44S - e88S 4. Amin A , Bruno A , Trocio J , Lin J , Lingohr-Smith M ( 2015 ) Comparison of diferences in medical costs when new oral anticoagulants are used for the treatment of patients with nonvalvular atrial ibrillation and venous thromboembolism vs. warfarin or placebo in the US . J Med Econ 18 : 399 - 409 5. Amouyel P , Mismetti P , Langkilde LK , Jasso-Mosqueda G , Nelander K , Lamarque H ( 2009 ) INR variability in atrial ibrillation: a risk model for cerebrovascular events . Eur J Intern Med 20 : 63 - 69 6. Ansell J , Hollowell J , Pengo V , Martinez-Brotons F , Caro J , Drouet L ( 2007 ) Descriptive analysis of the process and quality of oral anticoagulation management in real-life practice in patients with chronic non-valvular atrial ibrillation: the international study of anticoagulation management (ISAM) . J Thromb Thrombolysis 23 : 83 - 91 7. Ansell JE , Bakhru SH , Laulicht BE et  al ( 2014 ) Use of PER977 to reverse the anticoagulant efect of edoxaban . N Engl J Med 371 : 2141 - 2142 8. Azoulay L , Dell'aniello S , Simon TA , Renoux C , Suissa S ( 2014 ) Initiation of warfarin in patients with atrial ibrillation: early efects on ischaemic strokes . Eur Heart J 35 : 1881 - 1887 9. Bahit MC , Lopes RD , Wojdyla DM et  al ( 2013 ) Apixaban in patients with atrial ibrillation and prior coronary artery disease: insights from the ARISTOTLE trial . Int J Cardiol 170 : 215 - 220 10. Bansilal S , Bloomgarden Z , Halperin JL et al ( 2015 ) Eicacy and safety of rivaroxaban in patients with diabetes and nonvalvular atrial ibrillation: The rivaroxaban once-daily, oral, direct factor Xa inhibition compared with vitamin K antagonism for prevention of stroke and embolism trial in atrial ibrillation (ROCKET AF Trial) . Am Heart J 170 : 675 - 682 11. Bayer Pharma AG ( 2016 ) Xarelto® (rivaroxaban) Summary of Product Characteristics . http://www.ema. europa.eu/docs/ en_GB/document_library/EPAR_-_Product_Information/ human/000944/WC500057108 .pdf. Accessed 21 Sept 2016 12. Beyer-Westendorf J , Ageno W ( 2015 ) Beneit-risk proile of non-vitamin K antagonist oral anticoagulants in the management of venous thromboembolism . Thromb Haemost 113 : 231 - 246 13. Beyer-Westendorf J , Ebertz F , Förster K et  al ( 2015 ) Efec - tiveness and safety of dabigatran therapy in daily-care patients with atrial ibrillation . Results from the Dresden NOAC Reg - istry. Thromb Haemost 113 : 1247 - 1257 14. Beyer-Westendorf J , Ehlken B , Evers T ( 2016 ) Real-world persistence and adherence to oral anticoagulation for stroke risk reduction in patients with atrial ibrillation . Europace 18 : 1150 - 1157 15. Beyer-Westendorf J , Förster K , Ebertz F et  al ( 2015 ) Drug persistence with rivaroxaban therapy in atrial ibrillation patients-results from the Dresden non-interventional oral anticoagulation registry . Europace 17 : 530 - 538 16. Beyer-Westendorf J , Förster K , Pannach S et  al ( 2014 ) Rates, management, and outcome of rivaroxaban bleeding in daily care: results from the Dresden NOAC registry . Blood 124 : 955 - 962 17. Beyer-Westendorf J , Gelbricht V , Forster K et  al ( 2014 ) Safety of switching from vitamin-K antagonists to dabigatran or rivaroxaban in daily care-results from the Dresden NOAC registry . Br J Clin Pharmacol 78 : 908 - 917 18. Beyer-Westendorf J , Werth S , Tittl L , Michalski F , Marten S , Weiss N ( 2015 ) Real life eicacy and safety of apixaban for stroke prevention in atrial ibrillation-results of the pro - spective NOAC registry (NCT01588119) . J Thromb Haemost 13 : 35 (abstract AS098) 19. Bloomield HE , Krause A , Greer N et  al ( 2011 ) Meta-analy - sis: efect of patient self-testing and self-management of longterm anticoagulation on major clinical outcomes . Ann Intern Med 154 : 472 - 482 20. Boehringer Ingelheim International GmbH ( 2016 ) Pradaxa® (dabigatran etexilate) Summary of product characteristics . http://www.ema. europa.eu/docs/en_GB/document_ librar y/EPAR_-_Product_Infor mation/human/000829/ WC500041059 .pdf. Accessed 21 Sept 2016 21. Brambatti M , Darius H , Oldgren J et  al ( 2015 ) Comparison of dabigatran versus warfarin in diabetic patients with atrial ibrillation: results from the RE-LY trial . Int J Cardiol 196 : 127 - 131 22. Bristol-Myers Squibb , Pizer ( 2016 ) Eliquis® (apixaban) Sum - mary of product characteristics . http://www.ema. europa.eu/ docs/en_GB/document_library/EPAR_-_Product_Information/ human/002148/WC500107728 .pdf. Accessed 21 Sept 2016 23. Brown MT , Bussell JK ( 2011 ) Medication adherence: WHO cares? Mayo Clin Proc 86 : 304 - 314 24. Budnitz DS , Lovegrove MC , Shehab N , Richards CL ( 2011 ) Emergency hospitalizations for adverse drug events in older Americans . N Engl J Med 365 : 2002 - 2012 25. Burness CB ( 2015 ) Idarucizumab: irst global approval . Drugs 75 : 2155 - 2161 26. Camm AJ , Amarenco P , Haas S et al ( 2016 ) XANTUS: a realworld, prospective, observational study of patients treated with rivaroxaban for stroke prevention in atrial ibrillation . Eur Heart J 37 : 1145 - 1153 27. Camm AJ , Lip GYH , De Caterina R et al ( 2012 ) 2012 focused update of the ESC guidelines for the management of atrial ibrillation: an update of the 2010 ESC guidelines for the man - agement of atrial ibrillation . Developed with the special con - tribution of the European heart rhythm association . Eur Heart J 33 : 2719 - 2747 28. Camm AJ , Lip GYH , De Caterina R et al ( 2012 ) 2012 focused update of the ESC guidelines for the management of atrial ibrillation: an update of the 2010 ESC guidelines for the man - agement of atrial ibrillation . Developed with the special con - tribution of the European heart rhythm association . Europace 14 : 1385 - 1413 29. Caro JJ , Flegel KM , Orejuela ME , Kelley HE , Speckman JL , Migliaccio-Walle K ( 1999 ) Anticoagulant prophylaxis against stroke in atrial ibrillation: efectiveness in actual practice . CMAJ 161 : 493 - 497 30. Chang HY , Zhou M , Tang W , Alexander GC , Singh S ( 2015 ) Risk of gastrointestinal bleeding associated with oral anticoagulants: population based retrospective cohort study . BMJ 350:h1585 31. Chen CH , Chen MC , Gibbs H et  al ( 2015 ) Antithrombotic treatment for stroke prevention in atrial ibrillation: the Asian agenda . Int J Cardiol 191 : 244 - 253 32. Connolly SJ , Ezekowitz MD , Yusuf S et  al ( 2009 ) Dabigatran versus warfarin in patients with atrial ibrillation . N Engl J Med 361 : 1139 - 1151 33. Connolly SJ , Ezekowitz MD , Yusuf S , Reilly PA , Wallentin L ( 2010 ) Newly identiied events in the RE-LY trial . N Engl J Med 363 : 1875 - 1876 34. Connolly SJ , Wallentin L , Yusuf S ( 2014 ) Additional events in the RE-LY trial . N Engl J Med 371 : 1464 - 1465 35. Costin J , Ansell J , Laulicht B , Bakhru S , Steiner S ( 2014 ) Reversal agents in development for the new oral anticoagulants . Postgrad Med 126 : 19 - 24 36. Cramer JA , Roy A , Burrell A et al ( 2008 ) Medication compliance and persistence: terminology and deinitions . Value Health 11 : 44 - 47 37. Crivera C , Nelson WW , Bookhart B et  al ( 2015 ) Pharmacy quality alliance measure: adherence to non-warfarin oral anticoagulant medication . Circ Cardiovasc Qual Outcomes 9 : 1889 - 1895 (abstract A366) 38. Daiichi Sankyo Europe GmbH ( 2016 ) Lixiana® (edoxaban) Summary of product characteristics . http://www.ema. europa.eu/ docs/en_GB/document_library/EPAR_-_Product_Information/ human/002629/WC500189045 .pdf. Accessed 21 Sept 2016 39. Daiichi Sankyo Inc ( 2015 ) Savaysa® (edoxaban) Prescribing Information . http://dsi. com/prescribing-information-portlet/get PIContent?productName=Savaysa&inline=true. Accessed 21 Sept 2016 40. Donzé J , Clair C , Hug B et  al ( 2012 ) Risk of falls and major bleeds in patients on oral anticoagulation therapy . Am J Med 125 : 773 - 778 41. Dreisbach AW , Japa S , Gebrekal AB et  al ( 2003 ) Cytochrome P4502C9 activity in end-stage renal disease . Clin Pharmacol Ther 73 : 475 - 477 42. Eikelboom JW , Connolly SJ , Brueckmann M et al ( 2013 ) Dabigatran versus warfarin in patients with mechanical heart valves . N Engl J Med 369 : 1206 - 1214 43. Eikelboom JW , Wallentin L , Connolly SJ et  al ( 2011 ) Risk of bleeding with 2 doses of dabigatran compared with warfarin in older and younger patients with atrial ibrillation: an analysis of the randomized evaluation of long-term anticoagulant therapy (RE-LY) trial . Circulation 123 : 2363 - 2372 44. Eikelboom JW , Weitz JI ( 2010 ) New anticoagulants . Circulation 121 : 1523 - 1532 45. Ezekowitz JA , Lewis BS , Lopes RD et  al ( 2015 ) Clinical outcomes of patients with diabetes and atrial ibrillation treated with apixaban: results from the ARISTOTLE trial . Eur Heart J Cardiovas Pharmacother 1 : 86 - 94 46. Ezekowitz MD , Connolly S , Parekh A et  al ( 2009 ) Rationale and design of RE-LY: randomized evaluation of long-term anticoagulant therapy, warfarin, compared with dabigatran . Am Heart J 157 : 805 - 810 47. Fang MC , Go AS , Chang Y et al ( 2010 ) Warfarin discontinuation after starting warfarin for atrial ibrillation . Circ Cardiovasc Qual Outcomes 3 : 624 - 631 48. Fareed J ( 2015 ) Antithrombotic therapy in 2014: making headway in anticoagulant and antiplatelet therapy . Nat Rev Cardiol 12 : 70 - 71 49. Ferreira J , Ezekowitz MD , Connolly SJ et al ( 2013 ) Dabigatran compared with warfarin in patients with atrial ibrillation and symptomatic heart failure: a subgroup analysis of the RE-LY trial . Eur J Heart Fail 15 : 1053 - 1061 50. Fox KAA , Piccini JP , Wojdyla D et  al ( 2011 ) Prevention of stroke and systemic embolism with rivaroxaban compared with warfarin in patients with non-valvular atrial ibrillation and moderate renal impairment . Eur Heart J 32 : 2387 - 2394 51. Frost C , Wang J , Nepal S et al ( 2013 ) Apixaban, an oral, direct Factor Xa inhibitor: single-dose safety, pharmacokinetics, pharmacodynamics and food efect in healthy subjects . Br J Clin Pharmacol 75 : 476 - 487 52. Giugliano RP , Ruf CT , Braunwald E et  al ( 2013 ) Edoxaban versus warfarin in patients with atrial ibrillation . N Engl J Med 369 : 2093 - 2104 53. Gladstone DJ , Bui E , Fang J et al ( 2009 ) Potentially preventable strokes in high-risk patients with atrial ibrillation who are not adequately anticoagulated . Stroke 40 : 235 - 240 54. Gladstone DJ , Spring M , Dorian P et  al ( 2014 ) Atrial ibril - lation in patients with cryptogenic stroke . N Engl J Med 370 : 2467 - 2477 55. Gomes T , Mamdani MM , Holbrook AM , Paterson JM , Hellings C , Juurlink DN ( 2013 ) Rates of hemorrhage during warfarin therapy for atrial ibrillation . CMAJ 185 : E121 - E127 56. Gomes T , Mamdani MM , Holbrook AM , Paterson JM , Juurlink DN ( 2012 ) Persistence with therapy among patients treated with warfarin for atrial ibrillation . Arch Intern Med 172 : 1687 - 1689 57. Gorst-Rasmussen A , Skjoth F , Larsen TB , Rasmussen LH , Lip GYH , Lane DA ( 2015 ) Dabigatran adherence in atrial ibrilla - tion patients during the irst year after diagnosis: a nationwide cohort study . J Thromb Haemost 13 : 495 - 504 58. Granger CB , Alexander JH , McMurray JJ et al ( 2011 ) Apixaban versus warfarin in patients with atrial ibrillation . N Engl J Med 365 : 981 - 992 59. Gschwind L , Rollason V , Daali Y , Bonnabry P , Dayer P , Desmeules JA ( 2013 ) Role of P-glycoprotein in the uptake/ elux transport of oral vitamin K antagonists and rivaroxaban through the Caco-2 cell model . Basic Clin Pharmacol Toxicol 113 : 259 - 265 60. Halperin JL , Hankey GJ , Wojdyla DM et al ( 2014 ) Eicacy and safety of rivaroxaban compared with warfarin among elderly patients with nonvalvular atrial ibrillation in the rivaroxaban once daily, oral, direct factor Xa inhibition compared with vitamin K antagonism for prevention of stroke and embolism trial in atrial ibrillation (ROCKET AF) . Circulation 130 : 138 - 146 61. Halvorsen S , Atar D , Yang H et  al ( 2014 ) Eicacy and safety of apixaban compared with warfarin according to age for stroke prevention in atrial ibrillation: observations from the ARISTO - TLE trial . Eur Heart J 35 : 1864 - 1872 62. Hankey GJ , Eikelboom JW ( 2011 ) Dabigatran etexilate: a new oral thrombin inhibitor . Circulation 123 : 1436 - 1450 63. Hankey GJ , Patel MR , Stevens SR et  al ( 2012 ) Rivaroxaban compared with warfarin in patients with atrial ibrillation and previous stroke or transient ischaemic attack: a subgroup analysis of ROCKET AF . Lancet Neurol 11 : 315 - 322 64. Hart RG , Benavente O , McBride R , Pearce LA ( 1999 ) Antithrombotic therapy to prevent stroke in patients with atrial ibrillation: a meta-analysis . Ann Intern Med 131 : 492 - 501 65. Hart RG , Pearce LA , Aguilar MI ( 2007 ) Meta-analysis: antithrombotic therapy to prevent stroke in patients who have nonvalvular atrial ibrillation . Ann Intern Med 146 : 857 - 867 66. Heidbuchel H , Verhamme P , Alings M et  al ( 2015 ) Updated European heart rhythm association practical guide on the use of non-vitamin K antagonist anticoagulants in patients with nonvalvular atrial ibrillation . Europace 17 : 1467 - 1507 67. Heidbuchel H , Verhamme P , Alings M et  al ( 2013 ) European heart rhythm association practical guide on the use of new oral anticoagulants in patients with non-valvular atrial ibrillation . Europace 15 : 625 - 651 68. Hijazi Z , Hohnloser SH , Oldgren J et  al ( 2014 ) Eicacy and safety of dabigatran compared with warfarin in relation to baseline renal function in patients with atrial ibrillation: a RE-LY (randomized evaluation of long-term anticoagulation therapy) trial analysis . Circulation 129 : 961 - 970 69. Hohnloser SH , Hijazi Z , Thomas L et  al ( 2012 ) Eicacy of apixaban when compared with warfarin in relation to renal function in patients with atrial ibrillation: insights from the ARISTOTLE trial . Eur Heart J 33 : 2821 - 2830 70. Hohnloser SH , Oldgren J , Yang S et  al ( 2012 ) Myocardial ischemic events in patients with atrial ibrillation treated with dabigatran or warfarin in the RE-LY (randomized evaluation of long-term anticoagulation therapy) trial . Circulation 125 : 669 - 676 71. Holford NH ( 1986 ) Clinical pharmacokinetics and pharmacodynamics of warfarin. Understanding the dose-efect relation - ship . Clin Pharmacokinet 11 : 483 - 504 72. Hori M , Connolly SJ , Zhu J et al ( 2013 ) Dabigatran versus warfarin: efects on ischemic and hemorrhagic strokes and bleed - ing in Asians and non-Asians with atrial ibrillation . Stroke 44 : 1891 - 1896 73. Hori M , Matsumoto M , Tanahashi N et al ( 2012 ) Rivaroxaban vs. warfarin in Japanese patients with atrial ibrillation-the J-ROCKET AF study . Circ J 76 : 2104 - 2111 74. January CT , Wann LS , Alpert JS et al ( 2014 ) 2014 AHA/ACC/ HRS Guideline for the management of patients with atrial ibril - lation: a report of the American college of cardiology/american heart association task force on practice guidelines and the heart rhythm society . J Am Coll Cardiol 64 : e1 - e76 75. Janzic A , Kos M ( 2015 ) Cost efectiveness of novel oral antico - agulants for stroke prevention in atrial ibrillation depending on the quality of warfarin anticoagulation control . Pharmacoeconomics 33 : 395 - 408 76. Kakkar AK , Mueller I , Bassand JP et  al ( 2013 ) Risk proiles and antithrombotic treatment of patients newly diagnosed with atrial ibrillation at risk of stroke: perspectives from the inter - national, observational, prospective GARFIELD Registry . PLoS One 8 : e63479 77. Kimmel SE , Chen Z , Price M et  al ( 2007 ) The inluence of patient adherence on anticoagulation control with warfarin: results from the International normalized ratio adherence and genetics (IN-RANGE) study . Arch Intern Med 167 : 229 - 235 78. Kubitza D , Becka M , Schwers S , Voith B ( 2013 ) Investigation of pharmacodynamic and pharmacokinetic interactions between rivaroxaban and enoxaparin in healthy male subjects . Clinical Pharm in Drug Dev 2 : 270 - 277 79. Laliberté F , Cloutier M , Crivera C et al ( 2015 ) Efect of rivar - oxaban versus warfarin on health care costs among nonvalvular atrial ibrillation patients: observations from rivaroxaban users and matched warfarin users . Adv Ther 32 : 216 - 227 80. Laliberté F , Cloutier M , Nelson WW et  al ( 2014 ) Real-world comparative efectiveness and safety of rivaroxaban and war - farin in nonvalvular atrial ibrillation patients . Curr Med Res Opin 30 : 1317 - 1325 81. Laliberté F , Pilon D , Raut MK et  al ( 2014 ) Hospital length of stay of nonvalvular atrial ibrillation patients who were admin - istered rivaroxaban versus warfarin with and without pretreatment parenteral anticoagulants therapies . Hosp Pract ( 1995 ) 42: 17 - 25 82. Laliberté F , Pilon D , Raut MK et  al ( 2014 ) Hospital length of stay: is rivaroxaban associated with shorter inpatient stay compared to warfarin among patients with non-valvular atrial ibril - lation? Curr Med Res Opin 30 : 645 - 653 83. Lamassa M , Di Carlo A , Pracucci G et  al ( 2001 ) Characteristics, outcome, and care of stroke associated with atrial ibril - lation in Europe: data from a multicenter multinational hospital-based registry (The European Community Stroke Project) . Stroke 32 : 392 - 398 84. Levy JH , Tanaka KA , Dietrich W ( 2008 ) Perioperative hemostatic management of patients treated with vitamin K antagonists . Anesthesiology 109 : 918 - 926 85. Lip GYH , Hart RG , Conway DSG ( 2002 ) Antithrombotic therapy for atrial ibrillation . Br Med J 325 : 1022 - 1025 86. Mahafey KW , Stevens SR , White HD et  al ( 2014 ) Ischaemic cardiac outcomes in patients with atrial ibrillation treated with vitamin K antagonism or factor Xa inhibition: results from the ROCKET AF trial . Eur Heart J 35 : 233 - 241 87. McMurray JJ , Ezekowitz JA , Lewis BS et  al ( 2013 ) Left ventricular systolic dysfunction, heart failure, and the risk of stroke and systemic embolism in patients with atrial ibrillation: insights from the ARISTOTLE trial . Circ Heart Fail 6 : 451 - 460 88. Mearns ES , White CM , Kohn CG et al ( 2014 ) Quality of vitamin K antagonist control and outcomes in atrial ibrillation patients: a meta-analysis and meta-regression . Thromb J 12:14 89. Mendell J , Noveck RJ , Shi M ( 2013 ) A randomized trial of the safety, pharmacokinetics and pharmacodynamics of edoxaban, an oral factor Xa inhibitor, following a switch from warfarin . Br J Clin Pharmacol 75 : 966 - 978 90. Michalski F , Tittl L , Werth S et  al ( 2015 ) Selection, management, and outcome of vitamin K antagonist-treated patients with atrial ibrillation not switched to novel oral anticoagulants . Results from the Dresden NOAC registry. Thromb Haemost 114 : 1076 - 1084 91. Molteni M , Cimminiello C ( 2014 ) Warfarin and atrial ibrilla - tion: from ideal to real the warfarin afaire . Thromb J 12 : 5 92. Mueck W , Stampfuss J , Kubitza D , Becka M ( 2014 ) Clinical pharmacokinetic and pharmacodynamic proile of rivaroxaban . Clin Pharmacokinet 53 : 1 - 16 93. Nabauer M , Gerth A , Limbourg T et al ( 2009 ) The registry of the German competence NETwork on atrial ibrillation: patient characteristics and initial management . Europace 11 : 423 - 434 94. National Institute for Health and Care Excellence ( 2014 ) Atrial ibrillation: management . NICE Clinical guideline 180 . https:// www.nice. org.uk/guidance/cg180. Accessed 21 Sept 2016 95. Nelson WW , Song X , Coleman CI et  al ( 2014 ) Medication persistence and discontinuation of rivaroxaban versus warfarin among patients with non-valvular atrial ibrillation . Curr Med Res Opin 30 : 2461 - 2469 96. Nelson WW , Song X , Thomson E et  al ( 2015 ) Medication persistence and discontinuation of rivaroxaban and dabigatran etexilate among patients with non-valvular atrial ibrillation . Curr Med Res Opin 31 : 1831 - 1840 97. Nieuwlaat R , Capucci A , Camm AJ et al ( 2005 ) Atrial ibrilla - tion management: a prospective survey in ESC member countries: the Euro heart survey on atrial ibrillation . Eur Heart J 26 : 2422 - 2434 98. O'Brien EC , Simon DN , Allen LA et  al ( 2014 ) Reasons for warfarin discontinuation in the outcomes registry for better informed treatment of atrial ibrillation (ORBIT-AF) . Am Heart J 168 : 487 - 494 99. Olesen JB , Lip GYH , Kamper AL et  al ( 2012 ) Stroke and bleeding in atrial ibrillation with chronic kidney disease . N Engl J Med 367 : 625 - 635 100. Osterberg L , Blaschke T ( 2005 ) Adherence to medication . N Engl J Med 353 : 487 - 497 101. Palareti G , Legnani C ( 1996 ) Warfarin withdrawal. Pharmacokinetic-pharmacodynamic considerations . Clin Pharmacokinet 30 : 300 - 313 102. Patel MR , Mahafey KW , Garg J et al ( 2011 ) Rivaroxaban ver - sus warfarin in nonvalvular atrial ibrillation . N Engl J Med 365 : 883 - 891 103. Perret-Guillaume C , Wahl DG ( 2004 ) Low-dose warfarin in atrial ibrillation leads to more thromboembolic events without reducing major bleeding when compared to adjusted-dose-a meta-analysis . Thromb Haemost 91 : 394 - 402 104. Pollack CV Jr, Reilly PA , Bernstein R et al ( 2015 ) Design and rationale for RE-VERSE AD: a phase 3 study of idarucizumab, a speciic reversal agent for dabigatran . Thromb Haemost 114 : 198 - 205 105. Pollack CV Jr, Reilly PA , Eikelboom J et  al ( 2015 ) Idarucizumab for dabigatran reversal . N Engl J Med 373 : 511 - 520 106. Popovic J , Mikov M , Jakovljevic V ( 1994 ) Pharmacokinetic analysis of a new acenocoumarol tablet formulation during a bioequivalence study . Eur J Drug Metab Pharmacokinet 19 : 85 - 89 107. Portola Pharmaceuticals ( 2014 ) Portola Pharmaceuticals announces commercial supply agreement for andexanet alfa with Lonza . http://investors.portola.com/phoenix. zhtml?c=198136& p=irol- newsroomArticle&ID=1978716. Accessed 21 Sept 2016 108. Reilly PA , Lehr T , Haertter S et  al ( 2014 ) The efect of dabi - gatran plasma concentrations and patient characteristics on the frequency of ischemic stroke and major bleeding in atrial ibril - lation patients: the RE-LY trial (randomized evaluation of longterm anticoagulation therapy) . J Am Coll Cardiol 63 : 321 - 328 109. Ruf CT , Giugliano RP , Braunwald E et al ( 2014 ) Comparison of the eicacy and safety of new oral anticoagulants with war - farin in patients with atrial ibrillation: a meta-analysis of ran - domised trials . Lancet 383 : 955 - 962 110. Sanna T , Diener HC , Passman RS et  al ( 2014 ) Cryptogenic stroke and underlying atrial ibrillation . N Engl J Med 370 : 2478 - 2486 111. Sarode R , Milling TJ , Jr., Refaai MA et al ( 2013 ) Eicacy and safety of a 4-factor prothrombin complex concentrate in patients on vitamin K antagonists presenting with major bleeding: a randomized, plasma-controlled, phase IIIb study . Circulation 128 : 1234 - 1243 112. Seet RC , Friedman PA , Rabinstein AA ( 2011 ) Prolonged rhythm monitoring for the detection of occult paroxysmal atrial ibrillation in ischemic stroke of unknown cause . Circulation 124 : 477 - 486 113. Siegal DM , Curnutte JT , Connolly SJ et  al ( 2015 ) Andexanet alfa for the reversal of factor Xa inhibitor activity . N Engl J Med 373 : 2413 - 2424 114. Skanes AC , Healey JS , Cairns JA et  al ( 2012 ) Focused 2012 update of the Canadian Cardiovascular Society atrial ibrillation guidelines: recommendations for stroke prevention and rate/ rhythm control . Can J Cardiol 28 : 125 - 136 115. Song X , Sander SD , Johnson BH , Varker H , Amin AN ( 2012 ) Impact of atrial ibrillation and oral anticoagulation on hospital costs and length of stay . Am J Health Syst Pharm 69 : 329 - 338 116. Sørensen R , Gislason G , Torp-Pedersen C et  al ( 2013 ) Dabigatran use in Danish atrial ibrillation patients in 2011: a nation - wide study . BMJ Open 3:e002758 117. Tamayo S , Peacock FW , Patel M et  al ( 2015 ) Characterizing major bleeding in patients with nonvalvular atrial ibrillation: a pharmacovigilance study of 27 467 patients taking rivaroxaban . Clin Cardiol 38 : 63 - 68 118. ten Cate H , Haas S , Accetta G et al ( 2015 ) Quality of vitamin K antagonist control and 1-year outcomes: a global perspective from the GARFIELD-AF registry . J Thromb Haemost 13(suppl 2):130 (abstract OR096) 119. Ufer M ( 2005 ) Comparative pharmacokinetics of vitamin K antagonists: warfarin, phenprocoumon and acenocoumarol . Clin Pharmacokinet 44 : 1227 - 1246 120. van Diepen S , Hellkamp AS , Patel MR et al ( 2013 ) Eicacy and safety of rivaroxaban in patients with heart failure and nonvalvular atrial ibrillation: insights from ROCKET AF . Circ Heart Fail 6 : 740 - 747 121. Villines TC , Schnee J , Fraeman K et al ( 2015 ) A comparison of the safety and efectiveness of dabigatran and warfarin in nonvalvular atrial ibrillation patients in a large healthcare system . Thromb Haemost 114 : 1290 - 1298 122. White HD , Gruber M , Feyzi J et al ( 2007 ) Comparison of outcomes among patients randomized to warfarin therapy according to anticoagulant control: results from SPORTIF III and V. Arch Intern Med 167 : 239 - 245 123. Wolf PA , Abbott RD , Kannel WB ( 1991 ) Atrial ibrillation as an independent risk factor for stroke: the Framingham study . Stroke 22 : 983 - 988 124. Yank V , Tuohy CV , Logan AC et al ( 2011 ) Systematic review: beneits and harms of in-hospital use of recombinant factor VIIa for of-label indications . Ann Intern Med 154 : 529 - 540 125. You JJ , Singer DE , Howard PA et  al ( 2012 ) Antithrombotic therapy for atrial ibrillation: antithrombotic therapy and pre - vention of thrombosis, 9th ed: American college of chest physicians evidence-based clinical practice guidelines . Chest 141 : e531S - e575S 126. Zalesak M , Siu K , Francis K et al ( 2013 ) Higher persistence in newly diagnosed nonvalvular atrial ibrillation patients treated with dabigatran versus warfarin . Circ Cardiovasc Qual Outcomes 6 : 567 - 574


This is a preview of a remote PDF: https://link.springer.com/content/pdf/10.1007%2Fs11239-016-1446-0.pdf

Andreas Zirlik, Christoph Bode. Vitamin K antagonists: relative strengths and weaknesses vs. direct oral anticoagulants for stroke prevention in patients with atrial fibrillation, Journal of Thrombosis and Thrombolysis, 2017, 365-379, DOI: 10.1007/s11239-016-1446-0