VIEWS: 7 PAGES: 12 POSTED ON: 1/25/2013
Review | doi: 10.1111/j.1365-2796.2012.02541.x New anticoagulant drugs for treatment of venous thromboembolism and stroke prevention in atrial ﬁbrillation A. Tripodi1 & G. Palareti2 From the 1Angelo Bianchi Bonomi Hemophilia and Thrombosis Center, Department of Internal Medicine, IRCCS Ca Granda Maggiore Hospital ` Foundation and Universita degli Studi di Milano, Milano, Italy; and 2Divsion of Angiology and Blood Coagulation, S.Orsola-Malpighi University ` Hospital, Bologna, Italy Abstract. Tripodi A, Palareti G (IRCCS Ca Granda Mag- ` drugs such as heparins and vitamin K antago- ` giore Hospital Foundation and Universita degli Studi nists. The use of these drugs is, however, complex di Milano, Milano, Italy; S.Orsola-Malpighi University and demanding for both patients and physicians. Hospital, Bologna, Italy). New anticoagulant drugs Recently, new antithrombotic drugs that act for treatment of venous thromboembolism and stroke directly by inhibiting activated coagulation factors prevention in atrial ﬁbrillation (Review). J Intern Med such as factor X or thrombin have been developed 2012; doi: 10.1111/j.1365-2796.2012.02541.x. and investigated in phase III clinical trials. The aim of this article is to review: (i) the need to de- Venous thromboembolism (including deep vein velop new drugs; (ii) their efﬁcacy ⁄ safety as thrombosis and pulmonary embolism) and atrial demonstrated in clinical trials; (iii) the need for ﬁbrillation are common conditions in Western laboratory monitoring and (iv) the direction countries. The mainstay of treatment and preven- towards the use of these new drugs in the real-life tion for these diseases is fast-acting anticoagulant clinical situation. Introduction treatment and prophylaxis in all the above condi- tions, have recently become or will soon be available. Venous thromboembolism (VTE), including deep vein The aim of this article is to review the current state of thrombosis (DVT) and pulmonary embolism (PE), oc- the art and to present expert opinion on what will be curs frequently in Western countries, with an esti- the practical impact for their application within the mated incidence of about 1.5 per 1000 inhabitants next few years. per year . DVT may progress to potentially fatal PE and the postthrombotic syndrome . VTE is a major cause of morbidity and mortality not only in Western Need to develop new drugs countries , as believed until recently, but also in Asia . The standard treatment for acute VTE is anti- Upon diagnosis, acute VTE must be promptly treated coagulant therapy with fast-acting drugs such as un- with fast-acting drugs to prevent thrombus exten- fractionated heparin (UFH), low molecular-weight sion. In this respect, UFH, LMWH or fondaparinux heparin (LMWH) or the synthetic pentasaccharide have proved to be highly effective and safe. However, fondaparinux, which all prevent thrombus extension the need for dose-adjustment (at least for UFH) based and its consequences, followed by prophylaxis with on laboratory tests and the need for parenteral vitamin K antagonists (VKAs) to prevent recurrence administration (intravenous or subcutaneous infu- . The fast-acting drugs are administered parenter- sion) make the use of these drugs challenging both for ally and VKAs orally . Another condition for which patients and physicians. Conversely, because of their anticoagulation is required is the prevention of stroke oral route of administration, VKAs have been the and systemic embolism in patients with nonvalvular agents of choice for secondary prophylaxis of VTE for atrial ﬁbrillation (AF) . It has been estimated that the last 60 years. AF affects 2.5 million individuals in the USA, includ- ing approximately 5% of those older than 65 years Vitamin K antagonists are also the drugs of choice for . Finally, VKAs are used to prevent thromboembo- the primary prevention of stroke in patients with non- lism in patients with mechanical prosthetic heart valvular AF or prosthetic heart valves. Currently, valves . A number of new drugs that may be suit- AF is the most important indication for VKAs, able alternatives to the established agents, both for accounting for more than 50% of all patients treated ª 2012 The Association for the Publication of the Journal of Internal Medicine 1 A. Tripodi & G. Palareti | Review: New anticoagulants for venous thromboembolism and atrial ﬁbrillation worldwide with these drugs. Patients treated with Coagulation Pathway New Anticoagulant drugs VKAs for any indication represent as many as 1.5% of the general population in Western countries, and Initiation TF/VIIa thus the management of VKA treatment is a challenge Thrombin FXa direct inhibitors for patients, their families and the healthcare sys- generation X IX (rivaroxaban, apixaban) tems. Although the organization of the management IXa VIIIa of VKA treatment has received special attention in many countries such as Italy, the Netherlands and Xa Va the UK with the establishment of specialized antico- II agulation clinics, most patients worldwide are still denied this life-saving therapy because of the per- Thrombin Thrombin direct inhibitors activity IIa (dabigatran) ceived management difﬁculties and the risk of bleed- (thrombin) ing. Furthermore, although very effective, treatment with VKAs is complex, is associated with conspicuous burden for the healthcare systems and is highly Fig. 1 Schematic representation of the coagulation cascade and coagulation factors targeted by the new drugs. TF, tissue demanding for patients. factor. Vitamin K antagonists have several limitations that include a slow onset and offset of action, complex The drugs that are currently close to entering the genetic control of their effect with polymorphisms market are dabigatran, rivaroxaban and apixaban. leading to highly variable individual sensitivity and The aim of this article is to review the current situa- a narrow therapeutic window. In addition, metabo- tion with respect to the results of clinical trials lism of VKAs is affected by many factors, including for treatment of acute VTE and long-term prevention diet, drugs, hepatic dysfunction, other co-morbid of recurrences as well as prevention of stroke in AF conditions and alcohol intake. As a result of these (Tables 2 and 3), and to discuss the methods that may limitations, the dose–response relationship of VKAs be employed for laboratory control of these new is unpredictable, and therefore frequent coagula- agents (Tables 4 and 5). tion monitoring and dose-adjustment is needed to ensure efﬁcacy of treatment and to minimize the Dabigatran risk of bleeding complications. Furthermore, the percentage of time spent in the therapeutic range Dabigatran is an oral direct thrombin inhibitor might be inadequate. It was recently shown that (Fig. 1) that has undergone phase III, randomized, the time within the therapeutic range was between controlled clinical trials for both VTE and stroke pre- 51.5% and 62.0% in community studies, rising to vention in AF. The characteristics of dabigatran are between 59.4% and 73.3% in randomized clinical summarized in Table 1, and the main clinical trial re- trials . These unsatisfactory results are of clinical sults are summarized in Tables 2 and 3. relevance because an increased frequency of ad- verse events is associated with poor anticoagula- Treatment of acute VTE: the RE-COVER study tion control . The RE-COVER study  was a randomized, dou- ble-blind, double-dummy, noninferiority trial that New drugs included a total of 2564 patients with acute VTE The challenges posed by the established drugs (espe- (encompassing proximal DVT and ⁄ or PE). After initial cially VKAs) prompted research to develop new anti- treatment with parenteral anticoagulation therapy coagulants (Fig. 1 and Table 1) that would be equally with LMWH or UFH (for a median of 9 days), patients effective and safe, but not require laboratory monitor- were randomly assigned to receive dabigatran at a ing for dose-adjustment. The results of the clinical tri- dose of 150 mg twice daily (b.i.d.) or dose-adjusted als carried out to date suggest that the new drugs will warfarin to achieve an international normalized ratio fulﬁl these expectations. However, deﬁnitive conclu- (INR) of 2.0–3.0. The duration of treatment was sions can only be drawn after they have been used to 6 months. The primary analysis for efﬁcacy, accord- treat patients in real-life situations outside clinical ing to the intention-to-treat principle, was based on trials. For instance, it is likely that laboratory control the composite end-point of symptomatic VTE or will be required to some extent in selected categories death associated with VTE in the 6 months after ran- of patients. domization. 2 ª 2012 The Association for the Publication of the Journal of Internal Medicine Journal of Internal Medicine A. Tripodi & G. Palareti | Review: New anticoagulants for venous thromboembolism and atrial ﬁbrillation Table 1 Main pharmacological characteristics of the new drugs Dabigatran Rivaroxaban Apixaban Brand name Pradaxa Xarelto Eliquis Mechanism of Direct, selective Direct, selective factor Direct, selective action factor IIa inhibitor Xa inhibitor factor Xa inhibitor Prodrug Yes No No Time to Cmax 2  2–4  1–3  (hours) Half-life (hours) 12–14  9–13  8–15  Elimination 80% renal 1 ⁄ 3 renal 25% renal 20% biliary 1 ⁄ 3 renal (as inactive 75% biliary metabolites) 1 ⁄ 3 biliary Interactions P-glycoprotein P-glycoprotein and P-glycoprotein CYP3A4 (minimal) and CYP3A4 Effect of food Absorption Absorption Not reported delayed delayed Protein binding (%) 35 90 87 Dosing b.i.d. o.d. b.i.d. Dabigatran and rivaroxaban are P-glycoprotein substrates. Amiodarone, verapamil and clarithromycin inhibit P-glycoprotein and therefore increase the anticoagulant effect of these factor IIa ⁄ Xa inhibitors. For the safety analysis, bleeding was classiﬁed as ma- anticoagulation, treatment with dabigatran at the jor, according to Schulman and Kearon , or non- dose of 150 mg b.i.d. is noninferior for efﬁcacy and at major clinically relevant. After central adjudication, least as safe as warfarin at the usual INR level in sub- occurrence of the primary outcome for efﬁcacy was jects with acute VTE (Table 3). conﬁrmed in 2.4% of all patients treated with dabigatran and in 2.1% of patients who received Secondary prevention of VTE: the RE-MEDY and RE-SONATE studies warfarin (time in therapeutic range 60%), establish- ing the noninferiority of dabigatran versus warfarin The preliminary results of these two studies have (P < 0.001). A major bleeding episode occurred in been presented at the XXIII Congress of the Interna- 1.6% and 1.9% of patients in the dabigatran and war- tional Society on Thrombosis and Haemostasis, held farin groups, respectively [hazard ratio (HR) 0.82; in Kyoto in July 2011. In the double-blind, noninferi- 95% conﬁdence interval (CI), 0.45–1.48]. The most ority RE-MEDY study , 2856 patients who had re- frequent major bleeding event in the dabigatran ceived 3–12 months of anticoagulant therapy after a group was gastrointestinal (n = 9). Intracranial VTE episode were randomly assigned to treatment haemorrhage occurred in three patients treated with with dabigatran 150 mg b.i.d. or warfarin (INR 2.0– warfarin and in none receiving dabigatran. Major or 3.0) for an additional period of 6–36 months. clinically relevant nonmajor bleeding was detected in 5.6% of patients in the dabigatran group and in 8.8% Safety outcomes were bleeding events, acute coro- in the warfarin group (HR, 0.63; 95% CI, 0.47–0.84; nary syndromes and other adverse events. Recur- P = 0.002). In total, 9.0% of patients in the dabigatran rent symptomatic VTE occurred in 1.8% and 1.3% group and 6.8% in the warfarin group had an adverse of patients treated with dabigatran and warfarin, event that led to discontinuation of the study drug respectively (HR 1.44; 95% CI 0.78–2.64; P = 0.03 (HR, 1.33; 95% CI, 1.01–1.76; P = 0.05). for noninferiority). Major bleeding occurred in 0.9% and 1.8% of dabigatran- and warfarin-treated pa- In conclusion, the results of the RE-COVER study tients, respectively (HR 0.52; 95% CI, 0.27–1.01). showed that, after an initial few days of parenteral Any bleeding occurred in 19% of patients receiving ª 2012 The Association for the Publication of the Journal of Internal Medicine 3 Journal of Internal Medicine A. Tripodi & G. Palareti | Review: New anticoagulants for venous thromboembolism and atrial ﬁbrillation Table 2 Phase III clinical trials to assess the efﬁcacy and safety VTE was observed in 0.4% of patients treated with da- of treatment with the new drugs bigatran and 5.6% of those treated with placebo (HR 0.08; 95% CI 0.02–0.25; P < 0.0001). There were two Study Clinical condition Reference major bleeding episodes (both gastrointestinal; Dabigatran 0.39%) in the dabigatran group and none in the pla- RE-COVER Treatment of acute VTE  cebo group. Clinically, relevant nonmajor bleeding RE-MEDY and Secondary prevention  occurred in 5.3% and 1.8% of patients treated with RE-SONATE of VTE dabigatran and placebo, respectively (HR 2.9; 95% CI, 1.5–5.6; P = 0.001). The rate of cardiovascular RE-LY Treatment of patients  events was not different in the two groups. It was con- with nonvalvular AF cluded that extended treatment with dabigatran was Rivaroxaban highly effective in reducing the rate of VTE recur- EINSTEIN Treatment of acute DVT  rences compared with placebo, and was associated programme with only a low risk of major bleeding (Table 3). Rocket AF Stroke prevention in  patients with Treatment of patients with nonvalvular AF: the RE-LY study nonvalvular AF The RE-LY study was a noninferiority trial in which Apixaban 18 113 patients with AF were randomly assigned to ARISTOTLE Stroke prevention in  two ﬁxed doses of dabigatran (110 mg or 150 mg patients with b.i.d.), administered in a blinded manner, or to open- label use of warfarin . The median duration of fol- nonvalvular AF low-up was 2 years. The primary efﬁcacy outcome AVERROES Stroke prevention in  was stroke or systemic embolism, and the main out- patients with come for safety was major bleeding. Amongst the ma- nonvalvular AF jor bleeding episodes, other events that were consid- Amplify VTE Treatment of acute VTE (Both studies ered life-threatening were symptomatic intracranial programme (AMPLIFY CV 185056) are ongoing) bleeding, decrease in haemoglobin level of at least 5 g dL)1, haemorrhage requiring transfusion of at Secondary prevention of least four units of blood or inotropic agents or events VTE (AMPLIFY-EXT necessitating surgery. CV 185057) With regard to the study population, the mean patient VTE, venous thromboembolism. AF, atrial ﬁbrillation. DVT, age was 71 years, and 63.6% were men. Half of all pa- deep vein thrombosis. tients had received long-term therapy with VKAs. The mean CHADS2 score was 2.1. Aspirin was used dur- ing the treatment period in about 20% of patients in treatment with dabigatran and in 26% of those all the three groups. The mean percentage of time receiving warfarin (HR 0.71; 95% CI, 0.61–0.83). within the therapeutic range for warfarin (INR 2.0– Higher rates of acute coronary syndromes were ob- 3.0) was 64%. served in patients treated with dabigatran com- pared with those treated with warfarin (0.9% vs. Stroke or systemic embolism occurred at a rate of 0.2%; P = 0.02). It was concluded that dabigatran 1.69% per year in patients receiving warfarin and was as effective as warfarin for the extended treat- 1.53% and 1.11% per year, respectively, in those trea- ment of VTE; dabigatran was associated with a re- ted with 110 or 150 mg dabigatran. Both doses of da- duced risk of bleeding but an increased incidence bigatran were noninferior to warfarin (P < 0.001), but of acute coronary events (Table 3). the latter was also superior (RR versus warfarin 0.66; 95% CI 0.53–0.82; P < 0.001). Amongst other out- In the double-blind RE-SONATE study , 1343 pa- comes, the rate of myocardial infarction was found to tients with VTE who had completed 6–18 months of be signiﬁcantly higher in the 150 mg dabigatran anticoagulant therapy were randomly assigned to group (0.74% per year) than in the warfarin group receive dabigatran 150 mg b.i.d. or placebo for a fur- (0.53% per year; RR 1.38; 95% CI, 1.00–1.91; ther 6 months. Patients with a clear indication to P = 0.048). However, after a subsequent revision of continue anticoagulation were not eligible. Recurrent the database, new events were detected, adjudicated 4 ª 2012 The Association for the Publication of the Journal of Internal Medicine Journal of Internal Medicine A. Tripodi & G. Palareti | Review: New anticoagulants for venous thromboembolism and atrial ﬁbrillation Table 3 Main conclusions of phase III clinical trials Study Conclusion Reference Dabigatran RECO-VER (treatment After parenteral anticoagulation, dabigatran 150 mg b.i.d is [11, 12] of acute VTE) noninferior for efﬁcacy and at least as safe as VKAs RE-MEDY (secondary Dabigatran is as effective as VKAs in the extended treatment of  prevention of VTE) VTE. It is associated with a reduced risk of bleeding, but an increased incidence of acute coronary events RE-SONATE Extended treatment with dabigatran is highly effective in  (secondary prevention reducing the rate of recurrence compared to placebo and is of VTE) associated with a low risk of major bleeding RE-LY (stroke Dabigatran 110 mg b.i.d. is as effective as VKAs with lower rates [15, 16] prevention in AF) of major bleeding. Dabigatran 150 mg b.i.d. is associated with lower rates of stroke and systemic embolism, but similar rates of major haemorrhage. Dabigatran 110 and 150 mg are associated with lower rates of ICH than VKAs Rivaroxaban Acute DVT Study Rivaroxaban 15 mg b.i.d. for the ﬁrst 3 weeks, followed by  (treatment of DVT) 20 mg o.d. thereafter provides an effective and safe, Acute PE Study single-drug approach to the initial and continued (Study ongoing) (treatment of PE) treatment of VTE Continued Treatment  Study (secondary prevention of VTE) ROCKET AF (stroke Rivaroxaban 20 (or 15) mg o.d. is noninferior to VKAs in patients  prevention in AF) with moderate-to-high risk of stroke. There is no difference in rates of major and clinically relevant nonmajor bleeding between treatments. ICH occurs less frequently with rivaroxaban than VKAs Apixaban ARISTOTLE (stroke Apixaban 5 (or 2.5) mg b.i.d. is superior to VKAs and is associated  prevention in AF) with less bleeding and lower mortality. Apixaban 5 (or 2.5) mg is associated with a lower rate of ICH than VKAs AVERROES (stroke Apixaban 5 (or 2.5) mg b.i.d. given to patients who cannot take  prevention in AF) VKAs is more effective than aspirin, without increasing the risk of major bleeding or ICH VTE, venous thromboembolism. AF, atrial ﬁbrillation. DVT, deep vein thrombosis. PE, pulmonary embolism. ICH, intracranial haemorrhage. VKA, vitamin K antagonist. in a blinded fashion and in accordance with the study related to myocardial ischaemia were not increased protocol, and the signiﬁcant difference regarding the . However, it should be noted that treatment with rates of myocardial infarction was no longer evident dabigatran was found to be associated with a signiﬁ- . After a further recent revision of the RE-LY re- cantly increased odds ratio (1.33; 95% CI 1.03–1.71) sults, it was found that there was a nonsigniﬁcant in- for myocardial infarction or acute coronary syndrome crease in myocardial infarction with dabigatran com- in a recent meta-analysis of seven randomized trials pared with warfarin and that other clinical events . ª 2012 The Association for the Publication of the Journal of Internal Medicine 5 Journal of Internal Medicine A. Tripodi & G. Palareti | Review: New anticoagulants for venous thromboembolism and atrial ﬁbrillation Table 4 Situations that may require laboratory control of the 110 mg b.i.d. was equally effective, but was new drugs associated with signiﬁcantly lower rates of major bleeding, including intracranial bleeding. At a dose of Emergency presentation with thrombotic or haemorrhagic 150 mg b.i.d., dabigatran was associated with lower events rates of stroke, systemic embolism and intracranial Need for immediate reversal of anticoagulation bleeding, but similar rates of other major haemor- Previously unrecognized renal failure rhage (Table 3). Liver failure Suspicion of or known interaction with other drugs Rivaroxaban Rivaroxaban is an oral direct factor Xa inhibitor (Fig. 1), and phase III, randomized, controlled clinical trials have been conducted for both VTE and stroke With regard to safety, the rate of major bleeding was prevention in AF. The main characteristics of rivarox- 3.36% per year in the warfarin group, 2.71% per year aban are summarized in Table 1, and the clinical in the 110 mg dabigatran group (RR 0.80; 95% CI, trials results are summarized in Tables 2 and 3. 0.69–0.93; P = 0.003) and 3.11% per year in the 150 mg dabigatran group (RR 0.93; 95% CI, 0.81– Treatment of acute DVT: the EINSTEIN programme 1.07; P = 0.31). Of particular interest are the lower rates of intracranial bleeding recorded in patients The EINSTEIN programme consists of three random- who received dabigatran: 0.74% per year in the war- ized trials of rivaroxaban: the Acute DVT Study and farin group and 0.23% and 0.30% per year in the the Acute PE Study, for the acute treatment of pa- 110 mg and 150 mg dabigatran groups, with an RR tients with symptomatic DVT or PE, respectively, and of 0.31 and 0.40 (both P < 0.001), respectively. The the Continued Treatment Study for prolonged treat- rate of major gastrointestinal bleeding was signiﬁ- ment of patients after therapy for acute DVT or PE. cantly higher with dabigatran at the 150 mg dose The results of the Acute DVT Study and Continued (1.51% per year) than with warfarin (1.02% per year; Treatment Study have been recently reported by P < 0.001). With regard to other adverse events, dys- Bauersachs et al. , whereas the Acute PE Study is pepsia was the most frequent in the dabigatran ongoing. groups (>11% with either dose) and signiﬁcantly more frequent than in the warfarin group (P < 0.001). The Acute DVT Study was an open-label, event-dri- Liver enzymes were not abnormally elevated more ven, randomized, noninferiority trial that compared frequently with dabigatran, at either dose, than with oral rivaroxaban alone [15 mg b.i.d. for 3 weeks, fol- warfarin. lowed by 20 mg once daily (o.d.)] with standard anti- coagulation treatment (subcutaneous enoxaparin In conclusion, the RE-LY study performed in patients followed by VKA therapy to a target INR of 2.0–3.0) for with AF conﬁrmed that, compared with standard 3, 6 or 12 months in patients with acute, symptom- treatment with warfarin, dabigatran at a dose of atic DVT and without symptomatic PE. After the com- Table 5 Main laboratory tests to assess the anticoagulant effect of the new drugs Test Characteristics Reference Dabigatran Prothrombin time Linear dose–response, not very responsive to increasing dosage  Activated partial thromboplastin time Nonlinear dose–response, adequately responsive to increasing dosage  Dilute thrombin time Linear dose–response, adequately responsive to increasing dosage  Ecarin clotting time Linear dose–response, adequately responsive to increasing dosage  Rivaroxaban Prothrombin time Linear dose–response, adequately responsive to increasing dosage  Activated partial thromboplastin time Linear dose–response, adequately responsive to increasing dosage  Anti-factor Xa activity Linear dose–response, adequately responsive to increasing dosage  6 ª 2012 The Association for the Publication of the Journal of Internal Medicine Journal of Internal Medicine A. Tripodi & G. Palareti | Review: New anticoagulants for venous thromboembolism and atrial ﬁbrillation pletion of 6–12 months of treatment, patients with get INR 2.0–3.0) . The median duration of expo- DVT or PE were included in the Continued Treatment sure to treatment was 590 days, and the median fol- Study, a double-blind, randomized, superiority study low-up period was 707 days. The primary efﬁcacy that compared rivaroxaban alone (20 mg o.d.) with end-point was the composite of stroke (either ischae- placebo for an additional 6 or 12 months. The pri- mic or haemorrhagic) and systemic embolism. The mary efﬁcacy outcome for both studies was recurrent principal safety end-point was a composite of major VTE. The principal safety outcome was major bleed- and nonmajor clinically relevant bleeding events. ing or clinically relevant nonmajor bleeding in the ini- tial treatment study and major bleeding in the study The median patient age was 73 years and 60.3% of of continued treatment. participants were male patients. Patients were at moderate-to-high risk of stroke; only those with a In the Acute DVT Study, which included 3449 pa- CHADS2 score of ‡2 (mean score of 3.5) were in- tients, the primary efﬁcacy outcome occurred in cluded in the study. Previous use of VKAs was re- 2.1% of patients who received rivaroxaban and in ported by 62.4% of patients and about one-third of 3.0% of those who received LMWH+VKAs; time in the patients also took aspirin at some time during the therapeutic range for warfarin was 57.7% and the study. In warfarin-treated patients, INR values the HR was 0.68 (95% CI 0.44–1.04; P < 0.001 for were within the therapeutic range, an average of noninferiority). With regard to safety, major or clini- 55% of the time. cally relevant nonmajor bleeding occurred in 8.1% of patients treated with either rivaroxaban or stan- In the per-protocol population, primary outcome dard therapy (HR for rivaroxaban, 0.97; 95% CI, events occurred at a rate of 1.7% per year in the riva- 0.76–1.22; P = 0.77). The outcome of a net clinical roxaban group and 2.2% per year in the warfarin beneﬁt occurred in 2.9% of patients in the rivarox- group (HR 0.79; 95% CI 0.66–0.96; P < 0.001 for non- aban group and 4.2% in the standard therapy inferiority). In the intention-to-treat analysis, pri- group (HR 0.67; 95% CI, 0.47–0.95; P = 0.03). mary efﬁcacy events occurred at a rate of 2.1% per There were no differences in other adverse events year in rivaroxaban-treated patients and 2.4% per between the two groups. year in the warfarin group (HR 0.88; 95% CI 0.74– 1.03; P < 0.001 for noninferiority and P = 0.12 for A total of 1197 patients were enrolled in the Contin- superiority). ued Treatment Study. Outcome events occurred in 1.3% of patients in the rivaroxaban group and 7.1% With regard to safety, the rates of major bleeding in the placebo group (HR 0.18; 95% CI, 0.09–0.39; were not different in the rivaroxaban- and warfa- P < 0.001; relative risk reduction 82%). The principal rin-treated patients (3.6% and 3.4%, respectively); safety outcome of major bleeding occurred in four pa- the rates of major and clinically relevant nonmajor tients (0.7%) in the rivaroxaban group and in none in bleeding were also similar in the two groups (14.9% the placebo group (P = 0.11). However, major or clini- and 14.5% per year). Intracranial haemorrhage oc- cally relevant nonmajor bleeding occurred in 6.0% curred less frequently in rivaroxaban-treated pa- and 1.2% in the rivaroxaban and placebo groups, tients than in those receiving warfarin (0.5% vs. respectively (P < 0.001). 0.7% per year, respectively; HR 0.67; CI 0.47–0.93; P = 0.02). The effect of rivaroxaban when compared From the results of both studies it appears that oral to warfarin did not differ across quartiles of the rivaroxaban, at a dose of 15 mg b.i.d. for the ﬁrst percentage of time of INR within the therapeutic 3 weeks followed by 20 mg o.d. thereafter, can pro- range according to study centre. There were no dif- vide an effective, safe, single-drug approach to the ini- ferences in serious adverse events between the two tial and continued treatment of VTE (Table 3). treatment groups. In conclusion, the study showed that a ﬁxed dose of Treatment of patients with nonvalvular AF: the ROCKET AF study rivaroxaban given on a daily basis was noninferior to The ROCKET AF study was a multicentre, random- adjusted-dose warfarin in patients with nonvalvular ized, double-blind, double-dummy, event-driven trial AF at moderate-to-high risk of stroke. There were no in which 14 264 patients with AF were randomly as- signiﬁcant differences in rates of major and clinically signed to receive ﬁxed-dose rivaroxaban 20 mg o.d. relevant nonmajor bleeding between the two treat- (or 15 mg o.d. in patients with a creatinine clearance ments. Intracranial bleeding occurred less frequently of 30 to 49 mL min)1) or adjusted-dose warfarin (tar- in rivaroxaban-treated patients (Table 3). ª 2012 The Association for the Publication of the Journal of Internal Medicine 7 Journal of Internal Medicine A. Tripodi & G. Palareti | Review: New anticoagulants for venous thromboembolism and atrial ﬁbrillation Apixaban Apixaban in patients with AF: the AVERROES study Apixaban is an oral direct factor Xa inhibitor (Fig. 1) It is well known that VKA therapy is effective for that has undergone phase III randomized clinical tri- prevention of cardioembolism in patients with AF. als for stroke prevention in AF. The main characteris- However, its use is limited by many factors and at tics of apixaban are summarized in Table 1, and the least one-third of AF patients who are at risk of results from these clinical trials are summarized in complications does not receive VKA therapy, or Tables 2 and 3. discontinue treatment . Despite the fact that aspirin is not, or only slightly, effective  in reducing thromboembolic complications in these Treatment of patients with nonvalvular AF: the ARISTOTLE study patients, it is largely used to treat those, especially The ARISTOTLE study was a double-blind, double- the elderly people, who for various reasons are not dummy study in which 18 201 patients with nonval- suitable for VKA therapy . vular AF were randomly assigned to treatment with apixaban 5 mg b.i.d. or dose-adjusted warfarin (INR The AVERROES study  was designed to deter- 2.0–3.0) . A lower apixaban dose (2.5 mg b.i.d.) mine the efﬁcacy and safety of apixaban at a dose of was used in patients with at least two of the following 5 mg b.i.d. (2.5 mg in 6% of patients), compared with criteria: age ‡80 years, body weight £60 kg and ser- aspirin at a dose of 81 to 324 mg o.d., for the treat- um creatinine level ‡1.5 mg dL)1. The median follow- ment of patients with nonvalvular AF for whom VKA up duration was 1.8 years. The median age of study therapy was considered unsuitable. Of the 5599 pa- participants was 70 years; 35.3% were women and tients enrolled, 40% had previously received but dis- the mean CHADS2 score was 2.1. Approximately, continued VKA treatment, 42% of them because the 57% of patients had previously received VKA treat- INR could not be maintained in the therapeutic range. ment. The median time spent within the therapeutic The treating physician had established that INR mea- INR range in the warfarin group was 66.0%. surements could not be regularly obtained in 43% of the other patients enrolled. VKA therapy was consid- Primary outcome events (stroke or systemic embo- ered to be unsuitable for 21% of patients because the lism) occurred at a rate of 1.27% per year in the apix- risk of stroke was only moderate (CHADS2 score of 1). aban group and 1.60% per year in the warfarin group In addition, 15% of patients did not want to take (HR 0.79; 95%; CI 0.66–0.95; P < 0.001 for noninferi- VKAs. The mean duration of follow-up was 1.1 years. ority and P = 0.01 for superiority). The rates of death from cardiovascular and noncardiovascular causes The rates of primary outcome events were 1.6% per and of myocardial infarction were lower (although not year in patients treated with apixaban and 3.7% per statistically signiﬁcantly different) in the apixaban year in those assigned to aspirin (HR 0.45, 95% CI group compared with the warfarin group. Major 0.32–0.62; P < 0.001). The rate of major bleeding was bleeding occurred at a rate of 2.13% per year in the 1.4% per year in patients taking apixaban and 1.2% apixaban group and 3.09% per year in the warfarin per year amongst those taking aspirin (HR 1.13; 95% group (HR 0.69; 95% CI, 0.60–0.80; P < 0.001). The CI, 0.74–1.75; P = 0.57). An intracerebral haemor- rate of intracranial haemorrhage was 0.33% per year rhage occurred in six and nine patients receiving in the apixaban group and 0.80% per year in the war- apixaban and aspirin, respectively. In an on-treat- farin group (HR 0.42; 95% CI 0.30–0.58; P < 0.001). ment analysis, whilst patients were receiving the Furthermore, the rate of any bleeding was 25.8% per study treatment, major bleeding events occurred at a year in the warfarin group and 18.1% per year in the rate of 1.4% and 0.9% per year in the apixaban and apixaban group, with an absolute reduction of 7.7 aspirin groups, respectively (HR 1.54; 95% CI, 0.96– percentage points (P < 0.001). The overall rates of 2.45; P = 0.07). Serious adverse events occurred sig- serious adverse events were not different in the two niﬁcantly less frequently in the apixaban than in the treatment groups. aspirin group (22% vs. 27%; P < 0.001). In conclusion, the study showed that a ﬁxed dose of In conclusion, the study showed that, in comparison apixaban (given b.i.d.) in patients with AF was supe- with aspirin, apixaban substantially reduced the risk rior to warfarin in preventing stroke or systemic of thromboembolic complications in patients with AF embolism and was associated with less bleeding for whom VKA therapy is unsuitable, without (including intracranial bleeding) and lower mortality increasing the risk of major or intracranial bleeding (Table 3). (Table 3). 8 ª 2012 The Association for the Publication of the Journal of Internal Medicine Journal of Internal Medicine A. Tripodi & G. Palareti | Review: New anticoagulants for venous thromboembolism and atrial ﬁbrillation Treatment of VTE: the AMPLIFY programme information is currently available on the effect of these drugs in this setting as the presence of chronic Two clinical trials are ongoing to assess the efﬁcacy liver failure has been one of the exclusion criteria for and safety of apixaban in subjects with VTE. The dou- most of the clinical trials so far carried out. (v) Suspi- ble-blind, double-dummy AMPLIFY CV 185056 cion of or known interaction with other drugs. study is enrolling patients with acute DVT and ⁄ or PE Although very few interactions have hitherto been re- who are randomly assigned to receive apixaban 5 mg ported, the possibility of drug interactions cannot be b.i.d. or warfarin (INR 2.0–3.0) for 6 months. The en- excluded a priori; therefore, laboratory control may rolment of 5400 patients is expected to be complete in occasionally be needed in this situation (Table 4). July 2012. For the AMPLIFY-EXT CV 185057, enrolment was Potentially useful tests completed in July 2011. The study included patients On the basis of limited experience, it appears that with a recent VTE episode who received a ﬁrst stan- most of the global and some speciﬁc tests can be used dard anticoagulation course for 6–12 months. After to evaluate the anticoagulant effect of the new drugs this period, patients were randomly assigned to re- (Table 5). ceive placebo or apixaban for 1 year; those treated with apixaban were further randomly assigned to re- ceive 2.5 or 5 mg b.i.d. apixaban in a double-blind Dabigatran fashion. The follow-up of patients is ongoing. To date, the following tests have been evaluated with regard to dabigatran. (i) Activated partial thrombo- Laboratory control of new drugs plastin time (APTT). This test showed a relatively poor dose–response linearity and an intermediate respon- Clinical trials with dabigatran, rivaroxaban and apix- siveness to increasing dose . Different results aban have been designed with a ﬁxed dosage without have been obtained depending on the reagents used. any laboratory control. Indeed, these drugs have Hence, it is anticipated that standardization across proved to be effective and safe for the investigated laboratories will be difﬁcult. As a consequence, man- conditions (see above) and, therefore, strict labora- agement of patients cannot be generalized on the ba- tory control and ⁄ or dose adjustment are not required sis of the test results. (ii) Thrombin clotting time (TT). [24, 25]. However, it should be recognized that pa- Preliminary evaluations have shown that this test tients enrolled in clinical trials are highly selected, has excellent linearity, but excessive responsiveness and therefore it is possible that they do not represent to increasing dose . It is anticipated that stan- the general population encountered in the real-life dardization across laboratories will be an issue. (iii) clinical situation. Thus, laboratory monitoring might Prothrombin Time (PT). Good linearity but poor be needed, at least to some extent. Although experi- responsiveness has been reported for PT ; it is ence is still limited, as these drugs are not yet used on also anticipated that standardization will be difﬁcult. a large scale, we can start to consider the most appro- (iv) Ecarin clotting time (ECT). This test has shown priate strategy for laboratory control of these new good linearity and excellent responsiveness. How- drugs on the basis of current knowledge . ever, standardization can be a problem due to the type ⁄ concentration of phospholipids and purity of the snake venoms used for testing. (v) Others. In the- Situations requiring laboratory control ory the thrombin generation test could be useful; The following situations may require laboratory con- however, at present the method is too complex for trol. (i) Emergency presentation with thrombotic or daily routine use. haemorrhagic events. The treating physician may need to know whether the adverse events are due to Rivaroxaban and apixaban over- or under-dosage of the drug. (ii) Need for imme- diate reversal of anticoagulation. Laboratory control The following tests may be useful to evaluate the anti- may be needed to judge objectively whether or not coagulant effects of rivaroxaban and apixaban. Most reversal has been achieved. (iii) Previously unrecog- of the currently available information focuses on riva- nized renal failure. The kidney is the main route of roxaban; however, it is likely that results will also be excretion for some of these drugs. Therefore, labora- valid for apixaban. (i) Anti-factor Xa activity. In theory tory control may be needed to identify drug accumu- this should be the test of choice ; however, it is not lation in cases of renal failure. (iv) Liver failure. No readily available in an emergency. As a result of the ª 2012 The Association for the Publication of the Journal of Internal Medicine 9 Journal of Internal Medicine A. Tripodi & G. Palareti | Review: New anticoagulants for venous thromboembolism and atrial ﬁbrillation large between-reagent variability, it is anticipated However, according to both patients and physicians, that standardization will be difﬁcult to achieve. (ii) PT. for many years there has been a real need for a sub- Good linearity and responsiveness have been shown stantial improvement in anticoagulant treatment with PT . Results are dependent on the type of with new drugs, possibly free from some of the limita- thromboplastin used for testing, but evidence has tions of VKAs and heparins. As shown in this review, been provided that standardization across reagents the oral direct thrombin or factor Xa inhibitors are in is feasible by employing an international sensitivity the advanced phase of clinical investigation with a index (ISI) based on plasma supplemented with number of phase III clinical trials having demon- increasing doses of rivaroxaban . This index (ISI- strated their efﬁcacy and safety for prolonged treat- rivaroxaban) was successfully used to convert PT ra- ments. One of these drugs, dabigatran, is already tio into INR calibrated for rivaroxaban (INR-rivarox- available for chronic clinical use in some countries, aban) , which proved effective in minimizing be- whereas others are expected to become available tween-thromboplastin variability . (iii) APTT. This soon for stroke prevention in AF patients and for test has excellent linearity, but poor responsiveness treatment of VTE. to increasing dose , and standardization will be difﬁcult. (iv) HepTest. Good linearity and excellent We believe that the use of these new drugs is an responsiveness have been shown  but standardi- important step towards an easier, effective and safe zation could be an issue. (v) Dilute Russell viper ve- long-term anticoagulant treatment, which will enable nom test. This test also has good linearity and respon- effective antithrombotic treatment even in those siveness . Standardization might be an issue patients in whom an adequate anticoagulation is because of the variety of phospholipids and purity of currently denied due to the limitations of VKAs. How- snake venom used for testing. (vi) Others. Other tests, ever, some aspects of the use of these new drugs re- for example using portable coagulometers with test quire careful consideration before they are intro- strips calibrated in terms of INR valid for patients on duced for broad clinical use. Some of these drugs VKAs and thrombin generation tests, have been pro- (although to different degrees) are excreted via the posed  but their advantages compared with previ- kidney (see Table 1). This indicates a potential risk of ous tests have not been investigated. drug accumulation in patients with renal failure, sug- gesting the need for dosage reduction and ⁄ or thera- peutic monitoring. This is important especially in the Recommended tests elderly people in whom close clinical control and fre- Based on the aforementioned considerations and the quent renal function assessment is highly advisable. present limited experience, we recommend ECT or di- Although lower with the new drugs than with VKAs, lute TT for dabigatran and the PT for rivaroxaban and there is still some risk of drug–drug interactions. Fur- apixaban. The ECT and dilute TT are simple, readily thermore, previous experience with a different agent available tests that can be run using a standard coag- (ximelagatran) suggests that it is advisable to collect ulometer; the PT is also a simple and readily available long-term data on the safety of these drugs, For in- test. Furthermore, we recommend that results for stance, clearance of apixaban is mainly hepatic (@ these tests should be expressed as the ratio 75%) and caution should be used in patients with li- (patient ⁄ normal) of the clotting time. In particular, PT ver dysfunction. results should be expressed as the clotting time ratio (patient ⁄ normal), or INR valid for this drug. Expres- Whichever drug is used, good patient compliance is of sion of the results as an INR valid for patients on VKAs paramount importance for an effective and safe anti- is strongly discouraged as it was shown that this dra- coagulant treatment. Poor compliance, however, can matically increases the between-reagent variability be even more serious with the new drugs than with . VKAs. Indeed, whilst the anticoagulant effect of VKAs lasts for several days, the relatively rapid offset of ac- tion of the new drugs may result in a rapid and com- Future directions plete lack of anticoagulant protection in patients who Until now, VKAs have been the only oral agents avail- forget their medication for only 1 or 2 consecutive able for long-term anticoagulation and are therefore days. Furthermore, in contrast to VKA therapy, where widely used for prevention of thromboembolic com- an erratic compliance to the treatment can be de- plications and for treatment in many clinical condi- tected in a timely manner due to repeated clinic visits tions that involve a large and increasing number of and INR determinations, the absence of periodic vis- patients. its with the new drugs may reduce the attention to ad- 10 ª 2012 The Association for the Publication of the Journal of Internal Medicine Journal of Internal Medicine A. Tripodi & G. Palareti | Review: New anticoagulants for venous thromboembolism and atrial ﬁbrillation here to therapy. The use of the new drugs requires and the recent approval by the US Food and Drug maximum effort to ensure adherence to treatment Administration as well as other regulatory agen- and to develop interventions to maintain good com- cies, they can be prescribed for treatment and pliance during long-term therapy. prophylaxis of thrombotic events, thus opening a new era in antithrombotic treatment for millions The new drugs do not need routine coagulation moni- of patients worldwide. However, because results toring, which is an important and favourable charac- from clinical trials cannot be easily generalized to teristic. However, there are instances when the drug the real-life situation, these new drugs must un- effect may need to be evaluated (see Table 4). dergo close scrutiny in phase IV clinical trials to establish their efﬁcacy and safety. Potential differences between the results obtained in clinical trials and the effects of the new drugs in the real-life clinical situation should be carefully as- Conﬂict of interest statement sessed with speciﬁcally designed postmarketing No conﬂict of interest was declared. studies. The phase III clinical studies of these drugs, in general, included only highly selected patients and important groups were excluded, particularly the el- References derly people or those with chronic co-morbid condi- 1 Cushman M, Tsai AW, White RH et al. Deep vein thrombosis and tions (especially cancer and renal or cardiac failure) pulmonary embolism in two cohorts: the longitudinal investiga- or intercurrent illnesses. In the real-life situation, one tion of thromboembolism etiology. Am J Med 2004; 117: 19–25. must address the needs of these patients, and an 2 Prandoni P, Lensing AWA, Cogo A et al. The long-term clinical assessment of the beneﬁt ⁄ risk ratio associated with course of acute deep venous thrombosis. Ann Intern Med 1996; the new drugs is essential. Speciﬁcally, designed 125: 1–7. 3 Heit JA, Silverstein MD, Mohr DN, Petterson TM, Ofallon WM, postmarketing, long-term studies are therefore Melton LJ. Predictors of survival after deep vein thrombosis and needed. pulmonary embolism - A population-based, cohort study. Arch Intern Med 1999; 159: 445–53. Finally, we believe there is a need for a change in 4 Piovella F, Wang CJ, Lu H et al. Deep-vein thrombosis rates after the role and activity of anticoagulation clinics that major orthopedic surgery in Asia. An epidemiological study manage patients requiring long-term VKA therapy. based on postoperative screening with centrally adjudicated bilateral venography. J Thromb Haemost 2005; 3: 2664–70. In the future their activity should be more focused 5 Kearon C, Kahn SR, Agnelli G, Goldhaber S, Raskob GE, Comero- on clinical aspects, including education of or infor- ta AJ. Antithrombotic therapy for venous thromboembolic dis- mation for patients for the optimal use of the new ease: American College of Chest Physicians Evidence-Based drugs, the best possible compliance to treatment Clinical Practice Guidelines (8th Edition). Chest 2008; 133: and periodic clinical surveillance of patients and 454S–545S. their management in case of complications during 6 Singer DE, Albers GW, Dalen JE et al. Antithrombotic therapy in treatment. Moreover, anticoagulation clinics are atrial ﬁbrillation: American College of Chest Physicians Evi- dence-Based Clinical Practice Guidelines (8th Edition). Chest ideal for performing collaborative studies to assess 2008; 133: 546S–92S. the effectiveness and safety of the new drugs in 7 Sellers MB, Newby LK. Atrial ﬁbrillation, anticoagulation, fall comparison with VKAs. risk, and outcomes in elderly patients. Am Heart J 2011; 161: 241–6. 8 Salem DN, O’Gara PT, Madias C, Pauker SG. Valvular and struc- Conclusions tural heart disease: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest New direct thrombin or factor Xa inhibitors are 2008; 133: 593S–629S. now becoming available. These drugs have the po- 9 vanWalraven C, Jennings A, Oake N, Fergusson D, Forster AJ. tential to overcome some of the problems associ- Effect of study setting on anticoagulation control - A systematic ated with the older antithrombotic agents. They review and metaregression. Chest 2006; 129: 1155–66. can be administered orally, have a relatively short 10 Oake N, Fergusson DA, Forster AJ, vanWalraven C. Frequency of half-life, acceptable therapeutic window, predict- adverse events in patients with poor anticoagulation: a meta- analysis. Can Med Assoc J 2007; 176: 1589–94. able dose–response relationship and no need for 11 Schulman S, Kearon C, Kakkar AK et al. Dabigatran versus war- regular laboratory monitoring and dose-adjust- farin in the treatment of acute venous thromboembolism. N Engl ment. As a result of all these characteristics, they J Med 2009; 361: 2342–52. are much more manageable and appealing than 12 Schulman S, Kearon C. Deﬁnition of major bleeding in clinical heparins or VKAs for patients and physicians. investigations of antihemostatic medicinal products in non- Following the completion of phase III clinical trials surgical patients. J Thromb Haemost 2005; 3: 692–4. ª 2012 The Association for the Publication of the Journal of Internal Medicine 11 Journal of Internal Medicine A. Tripodi & G. Palareti | Review: New anticoagulants for venous thromboembolism and atrial ﬁbrillation 13 Schulman S, Eriksson H, Goldhaber S et al. Dabigatran or warfa- 24 Hart RG, Pearce LA, Aguilar MI. Meta-analysis: antithrombotic rin for extended maintenance therapy of venous thromboembo- therapy to prevent stroke in patients who have nonvalvular atrial lism. J Thromb Haemost 2011; 9: O-TH-033. ﬁbrillation. Ann Intern Med 2007; 146: 857–67. 14 Schulman S, Baanstra D, Eriksson H, Group eaftR-SS. Dabiga- 25 Connolly SJ, Eikelboom J, Joyner C et al. Apixaban in patients tran versus placebo for extended maintenance therapy of venous with atrial ﬁbrillation. N Engl J Med 2011; 364: 806–17. thromboembolism. J Thromb Haemost 2011; 9: O-MO-037. 26 Stangier J, Rathgen K, Staehle H, Gansser D, Roth W. The phar- 15 Connolly SJ, Ezekowitz MD, Yusuf S et al. Dabigatran versus macokinetics, pharmacodynamics and tolerability of dabigatran warfarin in patients with atrial ﬁbrillation. N Engl J Med 2009; etexilate, a new oral direct thrombin inhibitor, in healthy male 361: 1139–51. subjects. Br J Clin Pharmacol 2007; 64: 292–303. 16 Connolly SJ, Ezekowitz MD, Yusuf S, Reilly PA, Wallentin L. New- 27 Samama MM, Amiral J, Guinet C, Perzborn E, Depasse F. An op- ly identiﬁed events in the RE-LY trial. N Engl J Med 2010; 363: timised, rapid chromogenic assay, speciﬁc for measuring direct 1875–6. factor Xa inhibitors (rivaroxaban) in plasma. Thromb Haemost 17 Hohnloser SH, Oldgren J, Yang S et al. Myocardial Ischemic 2010; 104: 1078–9. Events in Patients with Atrial Fibrillation Treated with Dabiga- 28 Samama MM, Martinoli JL, LeFlem L et al. Assessment of labora- tran or Warfarin in the RE-LY Trial. Circulation 2012; 125: tory assays to measure rivaroxaban–an oral, direct factor Xa 669–76. inhibitor. Thromb Haemost 2010; 103: 815–25. 18 Uchino K, Hernandez AV. Dabigatran association with higher 29 Tripodi A, Chantarangkul V, Guinet C, Samama MM. The Inter- risk of acute coronary events: meta-analysis of noninferiority national Normalized Ratio calibrated for rivaroxaban has the po- randomized controlled trials. Arch Intern Med 2012; 172: tential to normalize prothrombin time results for rivaroxaban- 397–402. treated patients: results of an in vitro study. J Thromb Haemost 19 Bauersachs R, Berkowitz SD, Brenner B et al. Oral rivaroxaban 2011; 9: 226–8. for symptomatic venous thromboembolism. N Engl J Med 2010; 30 Kubitza D, Becka M, Voith B, Zuehlsdorf M, Wensing G. Safety, 363: 2499–510. pharmacodynamics, and pharmacokinetics of single doses of 20 Patel MR, Mahaffey KW, Garg J et al. Rivaroxaban versus warfa- BAY 59-7939, an oral, direct factor Xa inhibitor. Clin Pharmacol rin in nonvalvular atrial ﬁbrillation. N Engl J Med 2011; 365: Ther 2005; 78: 412–21. 883–91. 31 Raghavan N, Frost CE, Yu Z et al. Apixaban metabolism and 21 Granger CB, Alexander JH, McMurray JJ et al. Apixaban versus pharmacokinetics after oral administration to humans. Drug warfarin in patients with atrial ﬁbrillation. N Engl J Med 2011; Metab Dispos 2009; 37: 74–81. 365: 981–92. 22 Glader EL, Sjolander M, Eriksson M, Lundberg M. Persistent use Correspondence: A. Tripodi, Angelo Bianchi Bonomi Hemophilia of secondary preventive drugs declines rapidly during the ﬁrst and Thrombosis Center, Department of Internal Medicine, IRCCS 2 years after stroke. Stroke 2010; 41: 397–401. ` ` Ca Granda Maggiore Hospital Foundation and Universita degli 23 Olesen JB, Lip GY, Hansen ML et al. Validation of risk stratiﬁca- Studi di Milano, Via Pace 9, 20122-Milano, Italy. tion schemes for predicting stroke and thromboembolism in pa- (fax: +39 02 50320723; email: email@example.com) tients with atrial ﬁbrillation: nationwide cohort study. BMJ 2011; 342: d124. 12 ª 2012 The Association for the Publication of the Journal of Internal Medicine Journal of Internal Medicine
"New anticoagulant drugs for treatment of venous thromboembolism "