NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Rivaroxaban (Xarelto): Treatment of Venous Thromboembolic Events (Deep Vein Thrombosis [DVT], Pulmonary Embolism [PE]) and Prevention of Recurrent DVT and PE [Internet]. Ottawa (ON): Canadian Agency for Drugs and Technologies in Health; 2015 Aug.

Cover of Rivaroxaban (Xarelto)

Rivaroxaban (Xarelto): Treatment of Venous Thromboembolic Events (Deep Vein Thrombosis [DVT], Pulmonary Embolism [PE]) and Prevention of Recurrent DVT and PE [Internet].

Show details


3.1. Findings From the Literature

A total of two studies were identified from the literature for inclusion in the systematic review (Figure 1). The included studies are summarized in Table 2 and described in Section 3.2. A list of excluded studies is presented in Appendix 3.

Figure 1. QUOROM Flow Diagram for Inclusion and Exclusion of Studies.

Figure 1

QUOROM Flow Diagram for Inclusion and Exclusion of Studies.

3.2. Included Studies

3.2.1. Description of studies

EINSTEIN DVT and EINSTEIN PE trials were two multi-centre, randomized, open-label, parallel-group, active controlled, event-driven non-inferiority studies with a treatment duration of three, six, or 12 months.

The primary objective of the two EINSTEIN trials was to evaluate whether rivaroxaban is at least as effective as enoxaparin/VKA in the treatment of patients with acute symptomatic VTE and in the prevention of its recurrence.

The main difference between the two trials was that the EINSTEIN DVT trial evaluated rivaroxaban in patients who had symptomatic DVT without symptomatic PE, while the EINSTEIN PE trial was concerned with patients who had PE with or without symptomatic DVT.

In both trials, allocation was stratified by country and by the intended treatment duration.

3.2.2. Populations

a. Inclusion and exclusion criteria

Patients were eligible for inclusion in the EINSTEIN PE trial if they presented with symptomatic PE and/or DVT. The PE diagnosis criteria included the following:

  • an intraluminal filling defect in segmental or more proximal branches on spiral computed tomography (sCT) scan
  • an intraluminal filling defect or an extension of an existing defect or a new sudden cut-off of vessels more than 2.5 mm in diameter on pulmonary angiogram
  • a perfusion defect of at least 75% of a segment with a local normal ventilation result (high probability) on ventilation/perfusion lung scintigraphy
  • inconclusive sCT, pulmonary angiography, or lung scintigraphy results, with demonstration of DVT in the lower extremities by compression ultrasound (CUS) or venography.

On the other hand, the EINSTEIN DVT trial included patients with symptomatic proximal DVT only, and no symptomatic PE. In both trials, the diagnosis of DVT was based on the following criteria:

  • a non-compressible proximal vein on CUS
  • an intraluminal filling defect in the proximal veins on venography.

In addition to the exclusion criteria in Table 3, patients were excluded if they used VKA for indications other than PE or DVT, had significant liver disease (e.g,. acute hepatitis, chronic active hepatitis, cirrhosis) or an alanine transaminase level more than three times the upper limit of normal, suffered from bacterial endocarditis, or used strong CYP3A4 inhibitors (e.g., HIV protease inhibitors, systemic ketoconazole) or strong CYP3A4 inducers such as rifampicin. Patients were also excluded if they had a life expectancy of less than three months or severe hypertension; women of child-bearing age without proper contraceptive measures, and women who were pregnant or breast feeding were also excluded.

Table 3. Details of Included Studies.

Table 3

Details of Included Studies.

b. Baseline characteristics

A summary of the baseline characteristics for both trials is presented in Table 4 and Table 5. Baseline characteristics were balanced between treatment groups in the included trials. The mean age was 57 years; the EINSTEIN PE trial included more patients older than 75 years (17%) than the EINSTEIN DVT trial (13%). The mean weight was around 82 kg, while approximately 30% of patients had a body mass index ≥ 30 kg/m2. Caucasian was the major ethnic group.

Table 4. Summary of Baseline Characteristics.

Table 4

Summary of Baseline Characteristics.

Table 5. Baseline Medical History and Diagnosis.

Table 5

Baseline Medical History and Diagnosis.

The majority of the included patients had normal renal function (i.e., creatinine clearance ≥ 80 mL/min) in both trials; about 7% of patients had creatinine clearance < 50 mL/min. Spontaneous VTE was reported in > 60% of the DVT trial population and in > 64% of the PE trial. The main reasons reported for provoked VTE were recent surgery/trauma, immobilization, and use of estrogen. Previous VTE events were reported in almost 19% of included patients in both trials.

The perfusion score was recorded in the PE trial as an indicator for severity; a score of 0 indicates perfusion defect in all lung lobes, and 1 means no perfusion defect. Almost 71% of the trial patients had a perfusion score above 0.75.

3.2.3. Interventions

In both trials, patients were treated for a duration of three, six, or 12 months. The decision concerning the duration of anticoagulation was based on the risk profile of the patient and on local treatment guidelines, and was made by the investigator at the time of randomization. Risk factors that were taken into consideration included active cancer, previous episodes of DVT/PE, and known thrombophilic conditions.5 Patients’ assignment to one of the three treatment periods varied between the two trials. In the DVT trial, 12% of patients were assigned to the three-month, 60% to the six-month, and 25% to the 12-month group. In the PE trials, the distribution was 5%, 57%, and 37%, respectively (Table 6).

Table 6. Pre-randomization Treatment.

Table 6

Pre-randomization Treatment.

Pre-randomization treatment with any anticoagulant was allowed for a maximum duration of 48 hours. However, only a single pre-randomization dose of VKA was allowed. Table 6 summarizes the pre-randomization treatments. As a result, 72% of patients in the DVT trial and 92% in the PE trial received pre-randomization treatment with LMWH, unfractionated heparin, or fondaparinux. The length of pre-randomization treatment varied between the two trials. In the DVT trial, about 67% of patients received the pre-medication for one day and 4% received it for two days, whereas in the PE trial, 72% of the patients received it for one day and 19% received for two days. There were no differences between treatment groups within each study.

The time elapsed from the VTE event to randomization was higher in the DVT trial (10.5 days) compared with the PE trial (8.7 days) (Table 6). After randomization, patients allocated to the rivaroxaban group received rivaroxaban 15 mg twice daily for three weeks, followed by rivaroxaban 20 mg once daily. Patients allocated to the comparator group received enoxaparin twice daily for at least five days in combination with VKA (overlap four to five days) and continued with VKA only after the INR had been ≥ 2.0 for two consecutive measurements at least 24 hours apart. Warfarin and acenocoumarol were allowed as VKAs. The actual treatment durations are presented in Table 11 in Appendix 4 and are summarized in the Patient Disposition section.

3.2.4. Outcomes

a. Efficacy

The primary efficacy outcome in both trials was symptomatic recurrent VTE, defined as the composite of recurrent DVT or non-fatal or fatal PE. In addition to the fixed-interval contacts with the two trial sites, patients were informed about the sign and symptoms of efficacy and safety events, and they were instructed to immediately contact the study site if these occurred. The events for primary efficacy and safety outcomes were evaluated by a central, blinded, independent adjudication committee (CIAC). Definitions used by the CIAC for the primary outcome events are reported in Appendix 6. All confirmed events were considered up to one month after the intended duration of treatment, irrespective of the actual treatment duration. Secondary outcomes in both trials included the composite of recurrent DVT, non-fatal PE, and all-cause mortality. Net clinical benefit was another secondary outcome that was defined as the composite of recurrent DVT or non-fatal or fatal PE (the primary efficacy outcome) and major bleeding events. A post hoc analysis evaluated a modified net clinical benefit, defined as the composite of recurrent DVT or non-fatal or fatal PE (the primary efficacy outcome), major bleeding events, cardiovascular deaths, myocardial infarctions, strokes, and non-central nervous system systemic embolisms.

b. Safety

Clinically relevant bleeding was the primary safety outcome and was defined as the composite of major or clinically relevant non-major bleeding. If a patient had more than one event within the time window for a composite outcome, only the first event was counted and displayed among the components.

A bleeding event was considered major if it was clinically overt and accompanied by at least one of the following:

  • a fall in hemoglobin level of 20 g/L (2 g/dL) or more
  • transfusion of two or more units of packed red blood cells or whole blood
  • bleeding that was retroperitoneal, intracranial, occurred in a critical site or contributed to death.

Clinically relevant non-major bleeding was defined as overt bleeding not meeting the criteria for major bleeding but associated with at least one of the following:

  • medical intervention or unscheduled contact with a physician
  • interruption or discontinuation of study treatment
  • bleeding associated with any other discomfort (e.g., pain or impairment of activities of daily life).

An adverse event (AE) was defined as any untoward medical occurrence in a patient who had been administered a pharmaceutical product.

3.2.5. Statistical analysis

a. Efficacy criteria

The primary efficacy analysis was based on the time to the first event of the composite primary efficacy outcome during the intended treatment period or the one-month observation period. A Cox proportional hazards model was used for the analysis, with intended treatment duration as stratum, adjusted for the baseline presence of cancer. To demonstrate non-inferiority, the manufacturer intended to maintain at least 66% of the difference between the proven efficacy of standard therapy versus placebo, no treatment, or “less effective” treatments.8 Prins and Lensing conducted a meta-analysis in order to support the estimation of the NIM,8 and they identified 14 studies. Based on this meta-analysis, four different values of the NIM were reported; these values were provided in terms of hazard ratios that ranged from 1.54 to 2.00, depending on the calculation method. The manufacturer used the most liberal value estimated from the meta-analysis and did not specify any clear justification why this NIM was used.

The trial size was based on the assumption of equal efficacy and on the requirement for a total of 88 events to give a power of 90% to demonstrate the non-inferiority of rivaroxaban over a standard therapy consisting of enoxaparin and a VKA at a two-sided alpha level of 0.05.5,6 With an estimated 3% incidence for the primary outcome, the trial needed to enrol a total of approximately 3,000 patients.5,6 If the planned number of events were reached before all patients had completed the intended treatment duration, the study would be stopped and the remaining patients would be treated to reach the intended treatment duration, but only up to a maximum of six months.

b. Analysis populations

Analyses were performed using the intention-to-treat (ITT) population, with an additional supportive analysis using the per protocol (PP) population. In non-inferiority trials, ITT analysis tends to bias the results toward non-inferiority and, therefore, is not considered a conservative approach. The PP analysis, however, is composed of a selected group of patients, which may provide a partial perspective and result in bias in either direction.30 Both PP and ITT analyses should be reported and should have equal importance in drawing a conclusion.31 In the EINSTEIN trials, patients who discontinued study drug prematurely were to be followed until the end of the intended treatment period and included in the ITT analysis.

The ITT population consisted of all randomized patients; patients were analyzed according to the assigned treatment group. On the other hand, the PP population consisted of all randomized patients without any major deviation from the protocol. Reasons for exclusion from the PP population included baseline VTE not confirmed by the CIAC, not receiving the allocated treatment, and inadequate compliance with the study treatment.

The safety population consisted of all randomized patients who received at least one dose of study drug and were analyzed according to the treatment actually received.

3.3. Patient Disposition

Patient disposition is summarized in Table 7. A total of 3,449 patients in the EINSTEIN DVT and 4,832 patients in the EINSTEIN PE trial were randomized. Premature discontinuation was higher in the DVT trial than in the PE trial; discontinuation on rivaroxaban was 17.2% and 10.7% and on enoxaparin/VKA was 19.7% and 12.3% in DVT and PE trials, respectively. These were similar between treatment groups, and also similar in terms of reason for discontinuation. However, twice as many patients withdrew consent in the enoxaparin/VKA treatment groups than in the rivaroxaban groups (4.5% versus 2.1% and 4.9% versus 2.7% in the DVT and PE trials, respectively).

Table 7. Summary of Patient Disposition.

Table 7

Summary of Patient Disposition.

3.4. Exposure to Study Treatments

Table 11, Appendix 4 summarizes the actual study treatments. The majority of patients in both EINSTEIN trials were allocated to the six-month anticoagulation regimen (approximately 63% in the DVT trial and 57% in the PE trial). However, the two trials differed in the percentages of patients assigned to the three-month and 12-month durations (12% and 25% in the DVT trial, respectively, compared with 5% and 37% in the PE trial, respectively). In general, the percentage of patients who had ≥ 80% compliance with the study treatment ▬▬▬▬

Table 12, Appendix 4 summarizes the INR compliance in the EINSTEIN trials. Overall, the majority of patients treated with enoxaparin/VKA in both trials had mean INR within the therapeutic range (62.4% in the DVT trial and 63.4% in the PE trial). The percentage of time spent within the therapeutic range was 57.7% in the DVT trial and 62.7% in the PE trial.4,7

Table 13, Appendix 4 provides a summary of the concomitant antithrombotic medications in the EINSTEIN trials. ▬▬▬▬ The reason and impact of this additional anticoagulation was not reported.

3.5. Critical Appraisal

3.5.1. Internal validity

a. Validity of the non-inferiority margin

CDR reviewers retrieved the full published texts of studies included in the Prins and Lensing study8 and verified the quality of the extracted data and the validity of the meta-analysis calculation methods; Appendix 5 provides a summary of the non-inferiority validation. This review revealed three main issues with the estimated NIM. First, the meta-analysis included five trials in which patients with distal DVT comprised all or a proportion of the trial patients.3236 Another trial included patients with unstable angina or arterial ischemia along with VTE patients.37 The management and outcomes of these patients might not reflect the proximal DVT patients included in the EINSTEIN trials. Second, some discrepancies between the reported recurrence rates and the original articles were found. Finally, the assessment of the meta-analysis calculation method showed that the use of the fixed treatment effects model was not justified, and the random effects model would have been more appropriate given the clinical heterogeneity in the included studies. The use of the corrected data set that excluded patients with non-proximal DVTs and of the random effects model produced NIM estimates that ranged from 1.49 to 1.60 on the geometric scale (on the arithmetic scale, the random effect model produced margins that ranged from 1.75 to 2.28). These margins are considerably lower than the NIM of 2.0 used in the EINSTEIN trials, highlighting uncertainty concerning the margin chosen for the trials. However, in considering such evaluation, one should keep in mind that the EINSTEIN trials were powered based on a NIM of 2.0, and using a lower margin a posteriori may raise further uncertainties concerning the conclusions.

b. Selection, allocation, and disposition of patients

The EINSTEIN trials were open-label non-inferiority trials. No justification was provided as to why a double-blind design was not selected and a double-dummy technique not used to maintain blinding. However, events were evaluated by a CIAC.4,7 Therefore, the open-label design is unlikely to have influenced outcome measurement.38

c. Intervention and comparator

Patients treated with enoxaparin/VKA in the trials spent 58% to 63% of the time within the target INR values of 2.0 to 3.0. Non-optimal INR results obviously affect treatment efficacy and safety in the comparator group, which may bias the results in favour of rivaroxaban. However, this also improves generalizability, since INR values are expected to vary naturally. In discussion with the clinical expert involved in the review, the time spent within the target INR range was consistent with clinical practice.

▬▬▬▬ The clinical relevance of this finding and impact on the trials’ results are uncertain.

3.5.2. External validity

a. Patient characteristics

Although inclusion and exclusion criteria were generally relevant and reasonable, important sub-populations were excluded from the trial, including patients with a high risk of bleeding or significant renal or hepatic disease. In addition, the two trials included only a small proportion of patients with active cancer, which is a common cause of DVT and an important risk factor for recurrent VTE. For these patients, the treatment of choice is long-term LMWH.39 Therefore, the findings from EINSTEIN trials are unlikely to be generalizable to these categories of patients.

In EINSTEIN DVT, 25% of patients had a planned treatment duration of 12 months; however, premature discontinuation of study drug lowered the average to 10 months. More than 60% of these patients were treated between nine and 12 months, while only 10% actually reached 12 months of treatment with the study drug. This is an issue, considering that more than 60% of patients in the trial had an unprovoked DVT, for which treatment duration in clinical practice may often be indefinite. The most frequent reason for premature discontinuation was “study terminated by sponsor,” which limited treatment duration to six months rather than 12 months for the last included patients. No data are available to compare the benefits and risks of rivaroxaban with those of a standard treatment for more than 12 months, although several patients with irreversible risk factors may require chronic treatment.

3.6. Efficacy

Only those efficacy outcomes identified in the review protocol are reported below (see Section 2.2, Table 2). See Appendix 4: Detailed Outcome Data.

3.6.1. Survival

Mortality rates are summarized in Table 14, Appendix 4 for the ITT population and in Table 15, Appendix 4 for the safety population; Figure 2 presents the relative risks of mortality in each EINSTEIN trial and the trials’ pooled results. All-cause mortality and death by cause were reported as part of the composite outcomes in the trials, and individually as secondary outcomes. The EINSTEIN DVT trial recorded numerically fewer all-cause death events in the rivaroxaban group (2.2%) than in the enoxaparin/VKA group (2.9%). In contrast, the EINSTEIN PE trial showed numerically higher incidence of all-cause death in the rivaroxaban group (2.4%) compared with the comparator group (2.1%).

Figure 2. Plot Presentation of the Relative Risk of Mortality in the Einstein Trials.

Figure 2


Subgroup analysis by the intended treatment duration is summarized in Table 16, Appendix 4. ▬▬▬▬ Figure 3 shows the change in mortality incidence based on the treatment durations.

Figure Icon

Figure 3

Trend of the Relative Risk of Mortality by the Intended Treatment Duration Contained. CONFIDENTIAL DATA AND WAS REMOVED AT THE MANUFACTURER’S REQUEST.

Morality in the subgroup of patients with active cancer at baseline was reported in the clinical study reports for the trials in terms of hazard ratios only. ▬▬▬▬ (Table 17, Appendix 4). These results should be interpreted with caution because of the small number of cancer patients included in both studies (Table 5).

Survival rates based on age subgroups or renal function were not reported in either EINSTEIN trial.

3.6.2. Recurrent pulmonary embolism and/or deep vein thrombosis

Recurrence of VTEs was recoded as primary and secondary outcomes. The primary outcome, in both EINSTEIN trials, was the composite of symptomatic recurrent VTE, defined as the composite of recurrent DVT or non-fatal or fatal PE. The secondary recurrent VTE outcome included the individual end points included in the primary outcome, plus all-cause mortality. All confirmed events were considered up to one month after the end of the intended treatment period, irrespective of the actual treatment duration. Results for the primary and secondary recurrence outcomes are summarized in Table 18, Appendix 4 and Table 19, Appendix 4 for the ITT and PP datasets.

a. Recurrent venous thromboembolism: composite of recurrent deep vein thrombosis or non-fatal or fatal pulmonary embolism

Figure 4 provides a plot summary of the VTE recurrence in the EINSTEIN trials.

Figure 4. Plot Presentation of the Risk of VTE Recurrence in the EINSTEIN Trials.

Figure 4


Results for the EINSTEIN DVT trial showed that rivaroxaban met the pre-specified NIM of 2.0 for the upper limit of the two-sided 95% CI associated with the observed hazard ratio. This result was consistent in the PP and ITT data; the hazard ratio in the PP data was ▬▬▬▬ and in the ITT data 0.68 (95% CI, 0.44 to 1.04). In the DVT trial, rivaroxaban was associated with numerically fewer events than enoxaparin/VKA (PP data: 32 [2.1%] versus 46 [2.9%], respectively). In EINSTEIN PE, however, rivaroxaban was associated with numerically more events (38 [1.7%]) than the comparator group (36 [1.6%]); the associated hazard ratio was ▬▬▬▬ in the PP data and 1.12 (95% CI, 0.75 to 1.68) in the ITT data. The upper limit of the confidence interval was below the pre-specified NIM for the trial.


b. Individual outcomes included in the composite

Death events in both trials are summarized in the Survival section. Events of symptomatic PE only (without symptomatic DVT) occurred at similar rates in the treatment groups in both trials. ▬▬▬▬ On the other hand, the incidence of symptomatic DVT alone varied in the two trials. The EINSTEIN DVT trial had numerically lower events in the rivaroxaban group (PP dataset: ▬▬▬▬) than in the enoxaparin/VKA group ▬▬▬▬. In the EINSTEIN PE trial, the two treatment groups showed similar event rates: 0.7% in the rivaroxaban group and 0.6% in the comparator group. ▬▬▬▬

c. Subgroup analysis

Table 20, Appendix 4 summarizes the subgroup analyses for the primary composite outcome, recurrent VTE. Only results in the ITT population were reported and presented in the table. Of note, none of the reported subgroups had a statistically significant interaction with the recurrence of VTE.

By intended treatment duration

In both trials, the sample size for patients with a three-month intended treatment duration was small, and a low number of events was observed for the primary efficacy outcome, which limits the conclusion that can be drawn. The majority of patients in the trial had an intended treatment duration of six months. The recurrence incidence among patients in the six-month duration subgroup of EINSTEIN DVT was numerically lower in the rivaroxaban group (2.3%) than in the comparator group (2.7%); the hazard ratio did not reach statistical significance. In the EINSTEIN PE trial, the rivaroxaban group had a slightly higher incidence rate (1.9%) than the comparator (1.7%); the difference did not reach statistical significance. Results for the 12-month treatment duration showed that rivaroxaban was associated with a statistically significantly lower incidence of recurrence (1.4%) than enoxaparin/VKA (4.4%); the hazard ratio was 0.3 (95% CI, 0.12 to 0.75). The EINSTEIN PE trial provided similar incidence rates for the two treatment groups; the difference did not reach statistical significance. ▬▬▬▬

Malignancy at baseline

A small number of patients with cancer were included in the two studies, and the recurrence rate in this population was very low. ▬▬▬▬ Results for non-cancer patients were similar to the overall results of both trials.

Patients’ age


Renal function

Both EINSTEIN trials excluded patients with severe renal dysfunction; nevertheless, subgroups based on baseline renal function could be helpful in observing rivaroxaban performance in patients with borderline renal function (i.e., creatinine clearance < 50 mL/min). Both EINSTEIN studies conducted subgroup analyses by baseline renal function (subgroups: < 50 mL/min, 50 to 80 mL/min, and ≥ 80 mL/min). For this subgroup, the differences in recurrence rates between rivaroxaban and the control group did not reach statistical significance.

d. Recurrent venous thromboembolism/all-cause mortality: composite of recurrent deep vein thrombosis or non-fatal or fatal pulmonary embolism

For both trials, results for this composite variable were consistent with those for the primary outcome. ▬▬▬▬

3.6.3. Quality of life

Quality of life was not evaluated in the EINSTEIN trials.

3.6.4. Length of hospitalization

Length of hospitalization was reported in the EINSTEIN trials as part of the health care utilization assessment. Results for health care utilization were reported for the ITT population only and are presented in Table 21, Appendix 4.



3.6.5. Post-thrombotic syndrome

Post-thrombotic syndrome was not evaluated in the EINSTEIN PE trial.

3.6.6. Degree of pulmonary reperfusion

Pulmonary reperfusion was not evaluated in the EINSTEIN PE trial.

3.6.7. Pulmonary hypertension

▬▬▬▬ Table 8 ▬▬▬▬

Table 8. Incidence of Pulmonary Hypertension Events (Safety Population).

Table 8

Incidence of Pulmonary Hypertension Events (Safety Population).

Table 9. Key Efficacy Outcomes.

Table 9

Key Efficacy Outcomes.

3.7. Harms

Only those harms identified in the review protocol are reported below (see 2.2. Methods). See Appendix 4: Detailed Outcome Data.

3.7.1. Adverse events

The overall incidence of AEs was similar between treatment groups in the two trials. A total of 62.7% and 80.3% of patients in the rivaroxaban groups experienced at least one treatment-emergent AE compared with 63.1% and 79.0% of patients in the enoxaparin/VKA group in EINSTEIN DVT and EINSTEIN PE, respectively. The most common AEs, in both treatment groups, included nasopharyngitis, headache, epistaxis, pain in extremity, and cough; chest pain was common in both groups in the EINSTEIN PE trial (Table 22, Appendix 4).

3.7.2. Serious adverse events

The incidence of serious adverse events (SAEs) was similar between rivaroxaban and enoxaparin/VKA in EINSTEIN DVT (12.0% versus 13.6%, respectively) and in EINSTEIN PE (20.9% versus 20.7%, respectively). In general, the incidence of individual SAEs was small (< 0.1% of patients). For both treatment groups, the most common SAEs (Medical Dictionary for Regulatory Activities [MedDRA] terms) were cardiac disorders, gastrointestinal disorders, infections and infestations, respiratory disorders, and neoplasms (Table 23, Appendix 4). Of note, these SAEs occurred more frequently in the EINSTEIN PE trial than in the EINSTEIN DVT trial.

3.7.3. Withdrawals due to adverse events

Table 24, Appendix 4 summarizes withdrawals due to adverse events (WDAEs).

Overall, the rate of WDAEs was low and similar between rivaroxaban and enoxaparin/VKA groups (4.9% versus 4.7% in EINSTEIN DVT and 5.1% versus 4.1% in EINSTEIN PE, respectively).

3.7.4. Bleeding and hospitalization due to bleeding

Table 25, Appendix 4 presents the rates of confirmed bleeding events.

Treatment-emergent clinically relevant bleeding was the primary safety outcome in both trials, and it was defined as the composite of major or clinically relevant non-major bleeding. Only the first event for each patient was counted. For this outcome, results were similar between treatment groups in EINSTEIN DVT (8.1% in both groups). For EINSTEIN PE, rivaroxaban had a numerically lower incidence rate (10.3%) than enoxaparin/VKA (11.4%).

A total of five fatal bleeding events were reported for rivaroxaban in both trials combined, as compared with eight events in the enoxaparin/VKA group.

Table 10. Harms.

Table 10


Copyright © CADTH 2015.

You are permitted to make copies of this document for non-commercial purposes, provided it is not modified when reproduced and appropriate credit is given to CADTH. You may not otherwise copy, modify, translate, post on a website, store electronically, republish, or redistribute any material from this document in any form or by any means without the prior written permission of CADTH.

Please contact CADTH’s Vice-President of Corporate Services at ac.htdac@secivresetaroproc with any inquiries about this notice or other legal matters relating to CADTH’s services.

Except where otherwise noted, this work is distributed under the terms of a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International licence (CC BY-NC-ND), a copy of which is available at

Bookshelf ID: NBK344311


  • PubReader
  • Print View
  • Cite this Page

Other titles in this collection

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...