Cardiol Ther (2018) 7:15–24 https://doi.org/10.1007/s40119-018-0106-1 REVIEW Stroke Prevention with Oral Anticoagulants: Summary of the Evidence and Efﬁcacy Measures as an Aid to Treatment Choices Jose ´ Francisco Kerr Saraiva Received: December 1, 2017 / Published online: February 27, 2018 The Author(s) 2018. This article is an open access publication introduction of such agents. As compared with ABSTRACT warfarin, NOACs signiﬁcantly reduce the risk of hemorrhagic stroke, and only dabigatran Atrial ﬁbrillation (AF) is an established risk fac- (150 mg twice daily) was found to signiﬁcantly tor for a ﬁrst or recurrent stroke. Despite proven reduce the risk of ischemic stroke. However, efﬁcacy in preventing stroke in patients with measures of relative beneﬁts from medical AF, warfarin is underused, partly due to safety interventions do not immediately provide the concerns. Recent randomized trials have shown estimated beneﬁt to be derived from an indi- that non-vitamin K antagonist oral anticoagu- vidual patient, something best done by consid- lants (NOACs) such as dabigatran (a direct ering the expected absolute beneﬁt. The thrombin inhibitor) and apixaban, edoxaban, number needed to treat (NNT) is presented for and rivaroxaban (factor Xa inhibitors) are not various outcomes in the phase 3 trials of only non-inferior or superior to warfarin but NOACs. Despite the important progress also demonstrate a decreased risk of cere- achieved with the introduction of NOACs, the brovascular bleeding among patients with AF availability of at least four agents with different and moderate to high risk of stroke. Addition- efﬁcacy and safety performances in comparison ally, NOACs have an advantage of requiring no with warfarin prompts the question of whether monitoring of the international normalized any of these agents is preferable to another. It is ratio compared with warfarin. This review hoped that future studies on the efﬁcacy, safety, summarizes the published literature on NOACs and economic performance of NOACs will fur- for the primary and secondary prevention of ther allow for rational choices within this ischemic strokes, with an emphasis on the important therapeutic class. Meanwhile, the expected absolute beneﬁts from the NNT may be a valid metric to be considered by clinicians faced with the need to make such choices. Enhanced content To view enhanced content for this article go to https://doi.org/10.6084/m9.ﬁgshare.57989 61. Keywords: Atrial ﬁbrillation; Hemorrhagic stroke; Ischemic stroke; Non-vitamin K J. F. K. Saraiva (&) antagonist oral anticoagulants; Stroke Cardiology Discipline, Pontiﬁcal Catholic prevention University of Campinas School of Medicine, Campinas, Brazil e-mail: firstname.lastname@example.org 16 Cardiol Ther (2018) 7:15–24 INTRODUCTION SUMMARY OF RESULTS WITH NEW ORAL ANTICOAGULANTS Atrial ﬁbrillation (AF), the most common sus- tained cardiac arrhythmia, was estimated to Direct Comparisons Between Warfarin affect 33.5 million people worldwide in 2010, and New Oral Anticoagulants with a notable increase in incidence and prevalence in comparison with previous dec- NOACs include direct thrombin inhibitors (e.g., ades . In the next 50 years, the prevalence of dabigatran) and factor Xa inhibitors (e.g., AF, which increases with age, is expected to rivaroxaban, apixaban, and edoxaban). To date, double [2, 3], which may be of concern in a direct comparison of commercially available rapidly developing countries such as Brazil, NOACs in a prospective head-to-head trial has China, India, and Indonesia [1, 4]. not been performed. In a systematic review of AF is an established risk factor for a ﬁrst or 12 randomized trials, each comparing one of recurrent stroke . Theriskofstrokeamong these NOACs against warfarin, the pooled rela- patients with AF has been estimated between 1 and tive risk (RR) of death from any cause was 0.89 20% per year [6, 7]. Despite the efﬁcacy of warfarin [95% conﬁdence interval (CI) 0.83–0.96], thus in preventing a ﬁrst or recurrent stroke among indicating a signiﬁcant reduction of all-cause patients with AF , and notwithstanding the mortality of 11% in favor of NOACs. Likewise, recommendation for stroke prophylaxis in selec- signiﬁcant beneﬁts were demonstrated for ted individuals with AF [6, 9], there is evidence of NOACs in terms of stroke or systemic embolism underuse of anticoagulants in such patients (RR 0.77, 95% CI 0.70–0.86), as well as major [10, 11]. Additionally, underuse of warfarin among bleeding (RR 0.86, 95% CI 0.80–0.93). Con- patients with AF at moderate or high risk for stroke versely, no statistically signiﬁcant beneﬁt was has been reported in various countries [12, 13]. It is found for the pooled risk of ischemic stroke, for conceivable that underuse of warfarin when indi- which separate data were available from 11 tri- cated is, in part, due to the concerns about its als (RR 0.92, 95% CI 0.081–1.04) . These safety proﬁle and need for close monitoring of the pooled relative results include data from eight international normalized ratio (INR). randomized phase 2 trials and four phase 3 tri- Large randomized trials have shown that als, although none of these trials studied direct thrombin and factor Xa inhibitors are edoxaban. In the following paragraphs, the either non-inferior or superior to warfarin in individual results of phase 3 trials of dabigatran, terms of efﬁcacy, with a decreased risk of severe rivaroxaban, apixaban, and edoxaban are brieﬂy bleeding, among patients with AF and moderate discussed in chronological publication order. to high risk of stroke. The use of agents from In the RE-LY (Randomized Evaluation of these classes—also known as non-vitamin K Long-term anticoagulation therapY) trial, a antagonist oral anticoagulants (NOACs)—al- pivotal, open-label, phase 3 trial, two doses of lows for ﬁxed dosing and does not require lab- dabigatran (150 and 110 mg, both twice daily) oratory monitoring. The rapid onset and were compared with warfarin (adjusted to relatively short half-life of anticoagulant action achieve a target INR of 2.0–3.0) in 18,113 offers advantages for NOACs over warfarin in patients with AF plus at least one additional risk some clinical settings. It is expected that the use factor for stroke [15, 16]. All comparisons were of such agents will progressively replace war- made according to the intention-to-treat prin- farin in the future. This review summarizes the ciple, and the participants were randomly published literature with an emphasis on the assigned to receive one of the two doses of expected absolute beneﬁts from the introduc- dabigatran with no prior selection. The dose of tion of such agents. This article is based on dabigatran currently approved by the US Food previously conducted studies, and does not and Drug Administration (FDA; 150 mg twice involve any new studies of human or animal daily) was found to be superior to warfarin in subjects performed by any of the authors. reducing stroke or systemic embolism (RR 0.65, Cardiol Ther (2018) 7:15–24 17 95% CI 0.52–0.81, p\ 0.001), which was the 2.0–3.0) in the per-protocol analysis of the pri- primary trial end point. At a lower dose (110 mg mary end point [hazard ratio (HR) 0.79, 95% CI twice daily) dabigatran was non-inferior to 0.66–0.96, p\ 0.001 for non-inferiority], the warfarin, thus also achieving the protocol- intention-to-treat analysis could not conﬁrm speciﬁed primary objective (RR 0.90, 95% CI the superiority of rivaroxaban (HR 0.88, 95% CI 0.74–1.10, p = 0.30). Dabigatran at 150 mg 0.74–1.03, p\ 0.001 for non-inferiority and twice daily was superior to warfarin for both p = 0.12 for superiority). There was no signiﬁ- ischemic (RR 0.76, 95% CI 0.59–0.98, cant difference between the arms in the rate of p = 0.0351) and hemorrhagic (RR 0.26, 95% CI major and non-major clinically relevant bleed- 0.14–0.49, p\ 0.001) strokes, whereas at 110 mg ing, a major composite safety end point (RR twice daily it was non-inferior to warfarin for 1.03, 95% CI 0.96–1.11, p = 0.44), but there ischemic stroke and superior for hemorrhagic were lower rates of intracranial hemorrhage stroke (RR 0.31, 95% CI 0.17–0.56, p\ 0.001). (p = 0.02) and fatal bleeding (p = 0.003) with There was no signiﬁcant reduction in mortality rivaroxaban. Gastrointestinal bleeding was from the use of dabigatran (p = 0.051 for more common with rivaroxaban versus warfarin 150 mg twice daily and p = 0.13 for 110 mg (3.2 vs. 2.2%, p\ 0.001). Selected results of this twice daily), but the rate of the composite of trial and others are presented in Table 1. There major vascular events, major bleeding, or death was a trend toward an interaction between the (i.e., net clinical beneﬁt) was signiﬁcantly primary end point and a history of prior stroke, reduced by the 150 mg twice-daily dose transient ischemic attack, or systemic embolism (p = 0.02). The rate of major bleeding was lower (p = 0.072), and a signiﬁcant interaction with dabigatran 110 mg twice daily (RR 0.80, between the primary end point and safety 95% CI 0.70–0.93, p = 0.003), but similar for (p = 0.039). Among subjects without a history dabigatran 150 mg twice daily (p = 0.31), in of stroke, transient ischemic attack, or systemic comparison with warfarin. The rates of embolism, rivaroxaban was signiﬁcantly supe- intracranial hemorrhage and fatal bleeding with rior to warfarin in achieving the primary efﬁ- both doses of dabigatran were also lower. The cacy (HR 0.71, 95% CI 0.54–0.94) and safety (HR rates of major gastrointestinal bleeding were 0.59, 95% CI 0.42–0.83) end points. Conversely, signiﬁcantly higher with the dabigatran 150 mg there was no difference between the primary twice-daily dose when compared with warfarin efﬁcacy (HR 0.98, 95% CI 0.8–1.2) or safety (HR (RR 1.50, 95% CI 1.19–1.89, p\ 0.001). Selected 0.91, 95% CI 0.72–1.14) of rivaroxaban and results of this study and others are summarized warfarin in the subgroup with a history of prior in Table 1. In subgroup analyses of patients with stroke, transient ischemic attack, or systemic previous stroke or transient ischemic attack, RR embolism. reductions were consistent with those observed Apixaban was compared with warfarin (tar- in the overall study population—albeit with no get INR 2.0–3.0) in the double-blind ARISTOTLE statistical signiﬁcance, possibly as a result of (Apixaban for Reduction In STroke and Other insufﬁcient power in subgroups . Dyspepsia ThromboemboLic Events in AF) trial, which was the only adverse event that was signiﬁ- enrolled 18,201 patients with AF plus at least cantly more common with dabigatran than one additional risk factor for stroke . The with warfarin . apixaban dose was 5 mg twice daily, unless Rivaroxaban was compared with warfarin in patients met two or more of the criteria for a two phase 3 trials. The larger ROCKET AF 2.5 mg twice-daily dose (age C 80 years, body (Rivaroxaban Versus Warfarin in Nonvalvular weight B 60 kg, or serum creatinine C 1.5 mg/ AF) trial was a double-blind, non-inferiority dl). Apixaban was superior to warfarin for the study in 14,264 patients with AF at moderate to primary end point (stroke or systemic embo- high risk of stroke . The primary end point lism; HR 0.79, 95% CI 0.66–0.95). The relative was the composite of stroke or systemic embo- risk reduction was greater for hemorrhagic lism. Although rivaroxaban (20 mg per day) was (49%) than for ischemic or uncertain types of found to be non-inferior to warfarin (target INR stroke (8%, not statistically signiﬁcant; HR 0.92, 18 Cardiol Ther (2018) 7:15–24 Table 1 Selected patient features and results from pivotal phase 3 trials of four new anticoagulants [15, 18–20] Features and results Dabigatran Dabigatran Rivaroxaban Apixaban Edoxaban Edoxaban a a a 110 mg 150 mg 20 mg 5mg 30 mg 60 mg Patients, N 6015 6076 7131 9120 7034 7035 b b c c c c Age, years 71.4 71.5 73 70 72 72 CHADS score, mean 2.1 2.2 3.5 2.1 2.8 2.8 d d d d Median follow-up, years 2 2 1.9 1.8 2.8 2.8 TTR (%), mean 64 64 55 62 65 65 Primary end point NI Superior NI Superior NI NI Ischemic stroke NI Superior NI NI Inferior NI Hemorrhagic stroke Superior Superior Superior Superior Superior Superior Intracranial bleeding Superior Superior Superior Superior Superior Superior Any bleeding Superior Superior NI Superior Superior Superior Results refer to doses compared head-to-head with warfarin in each trial. All trials had stroke or systemic embolism as a primary end point CHADS Congestive heart failure, Hypertension, Age ([65 = 1 point,[75 = 2 points), Diabetes, and Stroke/TIA (2 points), NI noninferior, TTR time in therapeutic range (for warfarin arms of corresponding trial) Twice-daily doses; for apixaban, the criteria for a 2.5 mg twice-daily dose is discussed earlier in the results section Mean Median Reported in aggregate for all trial arms 95% CI 0.74–1.13, p = 0.42). The secondary end were randomized to receive two once-daily points of death (HR 0.89, 95% CI 0.80–0.99, regimens of edoxaban (60 or 30 mg) or warfarin p = 0.047) and major bleeding (HR 0.69, 95% CI (target INR 2.0–3.0). The primary efﬁcacy end 0.60–0.80, p\ 0.001) favored apixaban. The point was stroke or systemic embolism, whereas rates of gastrointestinal bleeding were similar in the principal safety end point was major the apixaban and warfarin groups (HR 0.89, bleeding. Primary comparisons in a modiﬁed 95% CI 0.70–1.15, p = 0.37). Selected results of intention-to-treat population favored both the this trial are also displayed in Table 1. Although 60-mg dose (HR 0.79, 97.5% CI 0.63–0.99, the results of subgroup analyses were generally p\ 0.001 for non-inferiority) and the 30-mg consistent with treatment effects seen for the dose of edoxaban (HR 1.07, 97.5% CI 0.87–1.31, overall study population, signiﬁcant interac- p = 0.005 for non-inferiority). A prespeciﬁed tions suggested greater reduction in bleeding superiority analysis using the intention-to-treat with apixaban among those without diabetes population showed trends in favor of the 60-mg mellitus and moderate or severe renal dose of edoxaban versus warfarin (HR 0.87, 95% impairment. CI 0.73–1.04, p = 0.08) and against the 30-mg Finally, edoxaban was assessed in ENGAGE dose of edoxaban versus warfarin (p = 0.10). AF-TIMI 48 (Effective aNticoaGulation with Other end points that favored edoxaban signif- factor xA next GEneration in AF—Thrombolysis icantly were the rate of major bleeding (HR In Myocardial Infarction study 48), a double- 0.80, 95% CI 0.71–0.91 for the 60-mg dose and blind trial . In this trial, 21,105 patients HR 0.47, 95% CI 0.41–0.55 for the 30-mg dose) with AF and moderate or high risk of stroke and the rates of death from cardiovascular Cardiol Ther (2018) 7:15–24 19 causes (HR 0.86, 95% CI 0.77–0.97 for the randomized trials disclosed nuances whose 60-mg dose and HR 0.85, 95% CI 0.76–0.96 for practical value remain uncertain, especially the 30-mg dose). The rates of the key secondary with regard to non-composite end points (e.g., end point (a composite of stroke, systemic the rate of ischemic stroke alone) and to sub- embolism, or death from cardiovascular causes) groups with a preferential beneﬁt (e.g., patients favored only the 60-mg dose of edoxaban (HR in primary or secondary prevention of stroke or 0.87, 95% CI 0.78–0.96). Gastrointestinal transient ischemic attack). Unfortunately, there bleeding was signiﬁcantly more frequent with are no published data with direct comparisons the 60-mg dose of edoxaban than with warfarin. between NOACs. Selected results of this trial are presented in An adjusted indirect comparison of apixaban, Table 1. A prespeciﬁed analysis of stroke rates dabigatran (150 mg twice daily), and rivaroxaban indicated that patients receiving 60 mg of using data from four randomized trials on one edoxaban showed signiﬁcantly lower stroke of these agents versus warfarin [15, 18, 19, 23] rates (HR 0.8, 95% CI 0.65–0.98), whereas suggested that dabigatran decreases the risks of patients receiving 30 mg of edoxaban showed stroke or systemic embolism, as well as ischemic similar stroke rates (HR 1.10, 95% CI 0.91–1.32) and hemorrhagic strokes versus rivaroxaban, compared with warfarin . However, both with no differences in mortality. Furthermore, doses of edoxaban were signiﬁcantly more the study suggested that apixaban decreases the effective than warfarin in reducing the rates of risk of major and gastrointestinal bleedings hemorrhagic stroke and other types of versus dabigatran and major bleeding versus intracranial bleeding . rivaroxaban, but increases the risk of systemic In summary, compared with warfarin, the embolism versus rivaroxaban . Another four NOACs discussed above signiﬁcantly indirect meta-analysis comparing the outcomes reduce the risk of hemorrhagic stroke (Table 1). of the three pivotal trials for apixaban, dabiga- Only dabigatran (at the dose of 150 mg twice tran, and rivaroxaban [15, 18, 19] drew very daily) was found to signiﬁcantly reduce the risk similar conclusions . The same authors of of ischemic stroke. More recently, a meta-anal- the previous study assessed the results sepa- ysis included the four larger pivotal trials dis- rately in primary and secondary prevention and cussed above (RE-LY, ROCKET AF, ARISTOTLE, found that, for patients with prior stroke, apix- and ENGAGE AF-TIMI 48) . NOACs were aban, dabigatran, and rivaroxaban had essen- found to signiﬁcantly reduce the risks of stroke tially similar efﬁcacy for the main outcomes; or systemic embolism (RR 0.81, 95% CI however, hemorrhagic stroke, vascular death, 0.73–0.91), all-cause mortality (RR 0.90, 95% CI major bleeding, and intracranial bleeding were 0.85–0.95), and intracranial hemorrhage (RR less common with twice-daily dabigatran 0.48, 95% CI 0.39–0.59). The beneﬁt regarding 110 mg than with rivaroxaban. For patients strokes was mainly driven by a reduction in with no prior stroke, all three drugs showed few hemorrhagic stroke (RR 0.49, 95% CI differences.Apixaban was superior to twice-daily 0.38–0.64). Conversely, NOACs signiﬁcantly dabigatran 110 mg for disabling or fatal stroke, increased the risk of gastrointestinal bleeding but inferior to twice-daily dabigatran 150 mg for (RR 1.25, 95% CI 1.01–1.55). stroke; however, apixaban was superior to twice-daily dabigatran 150 mg for bleeding. No signiﬁcant differences were observed for the Indirect Comparisons Among New Oral efﬁcacy and safetyendpointsbetweentwice-daily Anticoagulants dabigatran 150 mg and rivaroxaban. Finally, no signiﬁcant differences were found in efﬁcacy Given the availability of at least four NOACs, end points between apixaban and rivaroxaban, each with a different efﬁcacy and safety perfor- but the former led to less major bleeding . mance versus warfarin, the question arises as to Ideally, all these hypothesis-generating results whether any of these agents is preferable to should be conﬁrmed in randomized trials others, especially considering that individual directly comparing the NOACs of interest. 20 Cardiol Ther (2018) 7:15–24 Where the NOACs have been evaluated in expected absolute beneﬁt. This can be com- postmarketing studies in routine clinical prac- puted by considering the patient’s estimated tice, the real-world data generally support the baseline risk of an outcome of interest (in the efﬁcacy and safety proﬁles observed in their absence of intervention) and the relative beneﬁt respective pivotal randomized trials [27–29]. derived from clinical trials or meta-analyses. One of the largest independent real-world The absolute risk reduction (ARR) is the differ- studies was conducted by the US FDA with an ence in the rates of the outcome of interest in analysis of the Medicare database , which the presence and in the absence of the inter- included over 134,000 Medicare beneﬁciaries, vention in clinical trials or meta-analyses. The all aged C 65 years, with 37,587 person-years of number needed to treat (NNT) is the number of follow-up. The study showed that dabigatran patients who must be treated in order to pre- was associated with reduced risks of ischemic vent one case of the outcome of interest, thus stroke, intracranial hemorrhage, and death, and representing the expected absolute beneﬁt . an increased risk of gastrointestinal hemor- Mathematically, the NNT is the reciprocal of the rhage, when compared with warfarin in elderly ARR (NNT = 1/ARR). An absolute measure of patients with non-valvular AF. Overall, the harm [the number needed to harm (NNH)] may independent results of the US FDA Medicare be computed in the same way as the NNT if the study are consistent with the results of the ARR is replaced by the absolute risk increase. pivotal RE-LY trial, indicating that dabigatran Thus, the NNH is the reciprocal of the absolute has a favorable risk proﬁle, thus leading to no risk increase. The computation of the NNT and changes to the current recommendations for its the NNH only makes sense when there are sta- use. A nationwide cohort study in Denmark tistically signiﬁcant advantages (i.e., superior- (N = 61,678) assessed safety and efﬁcacy of ity) or disadvantages (i.e., inferiority), apixaban, dabigatran, rivaroxaban, and war- respectively, for the use of intervention regard- farin. No signiﬁcant differences were observed ing outcomes of interest in comparison with no in the rates of stroke between NOACs and the use of the intervention (i.e., control). warfarin groups. However, the rates of major Considering the pooled results of the meta- bleeding and mortality were signiﬁcantly lower analysis discussed previously, the NNT to pre- in the dabigatran and apixaban groups com- vent one death from the use of an NOAC pared with the warfarin group . A retro- instead of warfarin is 244, and the correspond- spective cohort study analyzed the safety data of ing NNT for cardiovascular mortality is 500. NOACs and warfarin in patients (N = 44,057) NNTs are more favorable for the outcomes of with nonvalvular AF from a US administrative stroke or systemic embolism (137) and major claims database. The rates of risk of major bleeding (157). For the latter, there was signiﬁ- bleeding were consistent with the ﬁndings from cant heterogeneity in a random-effects model. A pivotal randomized trials that compared subgroup analysis examining each NOAC sepa- NOACs with warfarin . These real-world rately showed that dabigatran and apixaban led studies are important to support understanding to signiﬁcant reductions in the rate of major of the real drug safety proﬁles of approved bleeding, whereas rivaroxaban and edoxaban medicines in routine clinical practice. did not . No NNHs were computed in this meta-analysis. Table 2 displays selected NNTs and NNHs based on the individual analysis of Absolute Measures of Efﬁcacy each pivotal study of these four agents, with the caveat that such metrics cannot be computed Measures of relative beneﬁts from medical for agents that have not proven a statistically interventions, such as RRs, odds ratios, and HRs, signiﬁcant difference against warfarin in selec- do not immediately provide the estimated ted outcomes. For the sake of comparison, the beneﬁt to be derived by an individual patient, NNT for the use of aspirin as primary something best done by considering the Cardiol Ther (2018) 7:15–24 21 Table 2 Selected absolute measures from pivotal phase 3 trials of four new anticoagulants (yearly NNT or NNH in head-to-head comparisons with warfarin within each trial) [15, 18–20] Features and results Dabigatran Dabigatran Rivaroxaban Apixaban Edoxaban Edoxaban a a a 110 mg 150 mg 20 mg 5mg 30 mg 60 mg b b Primary end point N/A 167 N/A 303 N/A N/A b c Ischemic stroke N/A 175 N/A N/A 192 N/A b b b b b b Hemorrhagic stroke 385 357 333 434 323 476 b b b b b b Intracranial bleeding 189 227 250 213 169 217 c c b c Gastrointestinal bleeding N/A 204 100 N/A 244 357 N/A refers to situations where neither superiority nor inferiority was demonstrated, thus precluding computation of NNTs and NNHs NNH numbers needed to harm, NNT numbers needed to treat Twice-daily doses; for apixaban, see the text for the explanation regarding the criteria for the 2.5-mg dose NNT NNH prevention of non-fatal myocardial infarctions [15, 16, 19]. These rough estimates ignore the has been estimated at 200 . burden of AF in people younger than 60 years of age, the different prevalence of AF between males and females , the potentially different Applying NNT to a Country effect of individual agents on patients at mod- erate and at high risk, and the percentage of The economic factors that are not considered in individuals with moderate and high risk for this review but discussed elsewhere  must stroke who should receive anticoagulants also be taken into account, especially in devel- according to guidelines. oping countries, where warfarin may continue to have a role in selected patients . A simple Limitations exercise using Brazilian data, for example, may illustrate this point. The population of Brazil is approximately 200 million, with nearly 10% of The main limitation of this review is that, in the the population aged C 60 years . A popula- absence of head-to-head trials, all the data are tion-based study of residents aged C 65 years hypothesis generating. In addition, it should be living in an economically deprived area of noted that the use of the NNT is not without Sa ˜o Paulo determined the prevalence of AF at drawbacks or controversy [35, 36]. Two prob- 2.4% . Based on this information, we esti- lems with the NNT are immediately apparent in mate that 480,000 individuals in Brazil have AF. the analysis of the four phase 3 trials of the Considering that nearly 75% of these individu- currently available NOACs [15, 18–20]. The ﬁrst als would have an indication to receive antico- problem relates to the slightly different patient agulation , 360,000 users of warfarin populations investigated in each trial; in that potentially exist. The NNT for NOACs with regard, the direct comparison of NNTs may be proven superiority to warfarin may be used to misleading because this absolute measure compute the annual number of strokes that clearly reﬂects the baseline risk of patients, could be prevented. If dabigatran (150 mg twice which was different across studies. For example, daily) and apixaban were used instead of war- the mean CHADS score [Congestive heart fail- farin in Brazil, they could prevent 2156 and ure, Hypertension, Age ([ 65 = 1 1188 strokes, respectively, every year point,[ 75 = 2 points), Diabetes, and Stroke/ 22 Cardiol Ther (2018) 7:15–24 TIA: a prediction for estimating the risk of Regardless of these limitations, the number strokes] was higher in ROCKET AF (3.5) and of patients with AF in whom a stroke could be ENGAGE AF-TIMI 48 (2.8) than in RE-LY (2.1 for prevented by the use of NOACs instead of war- the lower dose and 2.2 for the higher dose) and farin would only increase, accompanying the ARISTOTLE (2.1). Likewise, the percentage of expected increase in prevalence of AF in Brazil patients with a previous stroke or transient and elsewhere. It is hoped that future studies on ischemic attack (and, in some cases, systemic the efﬁcacy, safety, and economic performance embolism) was higher in ROCKET AF (55%) and of NOACs will further allow for rational choices ENGAGE AF-TIMI 48 (28%) than in RE-LY (20%) within this important therapeutic class. and ARISTOTLE (19%). Patients with a higher baseline risk derive more absolute beneﬁt from an intervention with a given relative impact ACKNOWLEDGEMENTS than patients with a lower baseline risk under- going the same intervention. The second prob- lem relates to different results in the control Funding. No funding or sponsorship was arm across the trials. Notably, the quality of received for this study. The article processing anticoagulation, as indicated by the mean per- charges were funded by Boehringer Ingelheim. centage of time the INR remained in the ther- apeutic range, was higher in ENGAGE AF-TIMI Editorial Assistance. The authors acknowl- 48 (65%) and in RE-LY (64%) than in ARIS- edge editorial assistance provided by PAREXEL, TOTLE (62%) and in ROCKET AF (55%). Since with funding from Boehringer Ingelheim. the NNT is a function of the difference between the rates of the outcomes of interest with Authorship. All named authors meet the NOACs and with warfarin, differing results in International Committee of Medical Journal control arms inﬂuence the computed NNTs. Editors (ICMJE) criteria for authorship for this article, take responsibility for the integrity of the work as a whole, and have given their CONCLUSIONS approval for this version to be published. All authors had full access to all of the data in this Encouraging progress has been achieved with study and take complete responsibility for the the introduction of NOACs that, overall, have integrity of the data and accuracy of the data displayed a superior efﬁcacy and improved analysis. safety in comparison with warfarin [15, 18–20]. As noted earlier, differences exist in the efﬁcacy Disclosures. Jose Francisco Kerr Saraiva is a and safety proﬁle among the four currently principal investigator and national coordinator available NOACs. Careful analysis of trial results leader of clinical trials from Amgen, AstraZe- so far indicates that the choice among the neca, Boehringer Ingelheim, Bristol-Myers available agents is not easy. Despite the limita- Squibb, Daiichi Sankyo, Janssen, Novartis, Novo tions of the NNT, this may be a valid metric to Nordisk, Pﬁzer, and Sanoﬁ. Jose ´ Francisco Kerr be considered by a clinician faced with four Saraiva also serves on the speakers’ bureau of different agents to be used instead of warfarin in AstraZeneca, BMS-Pﬁzer, Boehringer Ingelheim, patients with AF plus moderate or high stroke and Novo Nordisk. risk. In addition, economic as well as health factors should be considered, as they play a Compliance with Ethics Guidelines. This major role. In the example of applying NNT to article is based on previously conducted studies, Brazilian data, around 2156 and 1188 stroke and does not involve any new studies of human events could be prevented per year by using or animal subjects performed by any of the dabigatran or apixaban, respectively, instead of authors. warfarin [6, 32]. Cardiol Ther (2018) 7:15–24 23 9. Camm AJ, Kirchhof P, Lip GY, et al. Guidelines for Open Access. This article is distributed the management of atrial ﬁbrillation: the Task under the terms of the Creative Commons Force for the Management of Atrial Fibrillation of Attribution-NonCommercial 4.0 International the European Society of Cardiology (ESC). Eur Heart License (http://creativecommons.org/licenses/ J. 2010;31:2369–429. by-nc/4.0/), which permits any noncommer- 10. Friberg L, Rosenqvist M, Lindgren A, et al. High cial use, distribution, and reproduction in any prevalence of atrial ﬁbrillation among patients with medium, provided you give appropriate credit ischemic stroke. Stroke. 2014;45:2599–605. to the original author(s) and the source, provide 11. Ogilvie IM, Newton N, Welner SA, Cowell W, Lip a link to the Creative Commons license, and GY. Underuse of oral anticoagulants in atrial ﬁb- indicate if changes were made. rillation: a systematic review. Am J Med. 2010;123:638–45. 12. Mohammed MA, Marshall T, Nirantharakumar K, Stevens A, Fitzmaurice D. Patterns of warfarin use in REFERENCES subgroups of patients with atrial ﬁbrillation: a cross- sectional analysis of 430 general practices in the 1. Chugh SS, Havmoeller R, Narayanan K, et al. United Kingdom. PLoS One. 2013;8:e61979. Worldwide epidemiology of atrial ﬁbrillation: a Global Burden of Disease 2010 Study. Circulation. 13. Wang C, Yang Z, Wang C, et al. Signiﬁcant under- 2014;129:837–47. use of warfarin in patients with nonvalvular atrial ﬁbrillation: results from the China National Stroke 2. Go AS, Hylek EM, Phillips KA, et al. Prevalence of Registry. J Stroke Cerebrovasc Dis. diagnosed atrial ﬁbrillation in adults: national 2014;23:1157–63. implications for rhythm management and stroke prevention: the AnTicoagulation and Risk Factors 14. Dentali F, Riva N, Crowther M, et al. Efﬁcacy and in Atrial Fibrillation (ATRIA) Study. JAMA. safety of the novel oral anticoagulants in atrial 2001;285:2370–5. ﬁbrillation: a systematic review and meta-analysis of the literature. Circulation. 2012;126:2381–91. 3. Lip GY, Brechin CM, Lane DA. The global burden of atrial ﬁbrillation and stroke: a systematic review of 15. Connolly SJ, Ezekowitz MD, Yusuf S, et al. Dabiga- the epidemiology of atrial ﬁbrillation in regions tran versus warfarin in patients with atrial ﬁbrilla- outside North America and Europe. Chest. tion. N Engl J Med. 2009;361:1139–51. 2012;142:1489–98. 16. Connolly SJ, Ezekowitz MD, Yusuf S, Reilly PA, 4. Rahman F, Kwan GF, Benjamin EJ. Global epi- Wallentin L. Newly identiﬁed events in the RE-LY demiology of atrial ﬁbrillation. Nat Rev Cardiol. trial. N Engl J Med. 2010;363:1875–6. 2014;11:639–54. 17. Diener HC, Connolly SJ, Ezekowitz MD, et al. 5. Wolf PA, Abbott RD, Kannel WB. Atrial ﬁbrillation Dabigatran compared with warfarin in patients as an independent risk factor for stroke: the Fram- with atrial ﬁbrillation and previous transient ingham Study. Stroke. 1991;22:983–8. ischaemic attack or stroke: a subgroup analysis of the RE-LY trial. Lancet Neurol. 2010;9:1157–63. 6. Furie KL, Goldstein LB, Albers GW, et al. Oral antithrombotic agents for the prevention of stroke 18. Patel MR, Mahaffey KW, Garg J, et al. Rivaroxaban in nonvalvular atrial ﬁbrillation: a science advisory versus warfarin in nonvalvular atrial ﬁbrillation. for healthcare professionals from the American N Engl J Med. 2011;365:883–91. Heart Association/American Stroke Association. Stroke. 2012;43:3442–53. 19. Granger CB, Alexander JH, McMurray JJ, et al. Apixaban versus warfarin in patients with atrial 7. Gage BF, Waterman AD, Shannon W, et al. Valida- ﬁbrillation. N Engl J Med. 2011;365:981–92. tion of clinical classiﬁcation schemes for predicting stroke: results from the National Registry of Atrial 20. Giugliano RP, Ruff CT, Braunwald E, et al. Edoxa- Fibrillation. JAMA. 2001;285:2864–70. ban versus warfarin in patients with atrial ﬁbrilla- tion. N Engl J Med. 2013;369:2093–104. 8. Risk factors for stroke and efﬁcacy of antithrom- botic therapy in atrial ﬁbrillation. Analysis of 21. Giugliano RP, Ruff CT, Rost NS, et al. Cerebrovas- pooled data from ﬁve randomized controlled trials. cular events in 21 105 patients with atrial ﬁbrilla- Arch Intern Med. 1994;154:1449–57. tion randomized to edoxaban versus warfarin: effective anticoagulation with factor Xa next 24 Cardiol Ther (2018) 7:15–24 generation in atrial ﬁbrillation-thrombolysis in weighted nationwide cohort study. BMJ. myocardial infarction 48. Stroke. 2014;45:2372–8. 2016;353:i3189. 22. Ruff CT, Giugliano RP, Braunwald E, et al. Com- 29. Adeboyeje G, Sylwestrzak G, Barron JJ, et al. Major parison of the efﬁcacy and safety of new oral anti- bleeding risk during anticoagulation with warfarin, coagulants with warfarin in patients with atrial dabigatran, apixaban, or rivaroxaban in patients ﬁbrillation: a meta-analysis of randomised trials. with nonvalvular atrial ﬁbrillation. J Manag Care Lancet. 2014;383:955–62. Spec Pharm. 2017;23:968–78. 23. Ezekowitz MD, Reilly PA, Nehmiz G, et al. Dabiga- 30. Laupacis A, Sackett DL, Roberts RS. An assessment tran with or without concomitant aspirin compared of clinically useful measures of the consequences of with warfarin alone in patients with nonvalvular treatment. N Engl J Med. 1988;318:1728–33. atrial ﬁbrillation (PETRO Study). Am J Cardiol. 2007;100:1419–26. 31. Baigent C, Blackwell L, Collins R, et al. Aspirin in the primary and secondary prevention of vascular 24. Baker WL, Phung OJ. Systematic review and adjus- disease: collaborative meta-analysis of individual ted indirect comparison meta-analysis of oral anti- participant data from randomised trials. Lancet. coagulants in atrial ﬁbrillation. Circ Cardiovasc 2009;373:1849–60. Qual Outcomes. 2012;5:711–9. 32. Bista D, Chalmers L, Bereznicki L, Peterson G. 25. Lip GY, Larsen TB, Skjoth F, Rasmussen LH. Indirect Potential use of NOACs in developing countries: comparisons of new oral anticoagulant drugs for pros and cons. Eur J Clin Pharmacol. efﬁcacy and safety when used for stroke prevention 2014;70:817–28. in atrial ﬁbrillation. J Am Coll Cardiol. 2012;60:738–46. 33. Censo Demograﬁco. 2010. http://www.ibge.gov.br/ home/estatistica/populacao/censo2010/. Accessed 26. Rasmussen LH, Larsen TB, Graungaard T, Skjoth F, 30 Aug 2016. Lip GY. Primary and secondary prevention with new oral anticoagulant drugs for stroke prevention 34. Kawabata-Yoshihara LA, Scazufca M, Santos IS, in atrial ﬁbrillation: indirect comparison analysis. et al. Atrial ﬁbrillation and dementia: results from BMJ. 2012;345:e7097. the Sao Paulo ageing and health study. Arq Bras Cardiol. 2012;99:1108–14. 27. Graham DJ, Reichman ME, Wernecke M, et al. Cardiovascular, bleeding, and mortality risks in 35. Chatellier G, Zapletal E, Lemaitre D, Menard J, elderly Medicare patients treated with dabigatran or Degoulet P. The number needed to treat: a clinically warfarin for nonvalvular atrial ﬁbrillation. Circula- useful nomogram in its proper context. BMJ. tion. 2015;131:157–64. 1996;312:426–9. 28. Larsen TB, Skjoth F, Nielsen PB, Kjaeldgaard JN, Lip 36. Cook RJ, Sackett DL. The number needed to treat: a GY. Comparative effectiveness and safety of non- clinically useful measure of treatment effect. BMJ. vitamin K antagonist oral anticoagulants and war- 1995;310:452–4. farin in patients with atrial ﬁbrillation: propensity
Cardiology and Therapy – Springer Journals
Published: Feb 27, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
All the latest content is available, no embargo periods.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud