EINSTEIN-PE

Published

Büller HR, et al. "Oral rivaroxaban for the treatment of symptomatic pulmonary embolism". The New England Journal of Medicine. 2012. 366(14):1287-1297.
PubMed • Full text • PDF

Clinical Question

Among patients with acute PE, is rivaroxaban noninferior to warfarin in preventing recurrent VTE or bleeding?

Bottom Line

Among patients with acute PE, rivaroxaban is noninferior to warfarin in preventing recurrent VTE, and is associated with similar bleeding rates.

Major Points

The oral factor Xa inhibitor rivaroxaban demonstrated effectiveness in the prevention of VTE following orthopedic surgery in RECORD1 (2009), in DVT treatment in EINSTEIN-DVT (2010), and in AF in ROCKET AF (2011). EINSTEIN-PE was an extension of the EINSTEIN program and was the first large, randomized study of rivaroxaban in the treatment of acute PE.

Published in 2012, EINSTEIN-PE randomized 4,832 patients with acute PE to rivaroxaban or standard therapy with enoxaparin and a VKA. At a mean follow-up of 7 months, rivaroxaban was noninferior to standard therapy in terms of the rate of recurrent symptomatic VTE (2.1% vs. 1.8%) and had a similar risk of clinically significant bleeding (10.3% vs. 11.4%).

EINSTEIN-PE is similar to other studies of non-VKA anticoagulants in treating VTE. Like the others, it employed a noninferiority (rather than a superiority) design, and enrolled a relatively heterogeneous patient population. It differed from these studies in several notable ways, however. EINSTEIN-PE was one of the first of these studies to use a single oral agent without a LMWH bridge. To compensate for this, the study used a higher dose during the first 3 weeks of therapy (15mg BID) followed by a lower maintenance dose (20mg daily). It was also one of the first to employ an open-label design lacking matching placebos between groups.

Some of these characteristics contribute to the study's limitations. For example, the study's noninferiority design may have rendered it unable to detect small differences in relative efficacy between treatment arms. In addition, its open-label design may have biased both patients and investigators. (For comparison, the 2013 trial AMPLIFY which studied apixaban in acute VTE, used matching placebos to achieve a double-blind design.) The trial's generalizability is limited for several reasons, including the fact that 1) patients were younger (mean age 58 years) than the general acute PE population and 2) the trial excluded patients with cancer. Despite these limitations, there remains a reasonably strong evidence base for rivaroxaban in acute VTE, which led to the FDA approval of rivaroxaban for these indications in November 2012.

Guidelines

CHEST Antithrombotic Therapy for VTE Disease (2016, adapted):

• In patients with proximal DVT or pulmonary embolism (PE), recommended long-term (3 months) anticoagulant therapy over no such therapy (Grade 1B).
  • For patients with DVT of the leg or PE and no cancer who are not treated with dabigatran, rivaroxaban, apixaban, or edoxaban, VKA therapy is recommended over low-molecular weight heparin (LMWH) (Grade 2C)
• In patients with DVT of the leg or PE and no cancer, as long-term (first 3 months) anticoagulant therapy, dabigatran, rivaroxaban, apixaban, or edoxaban are recommended over vitamin K antagonist (VKA) therapy (all Grade 2B)
• In patients with DVT of the leg or PE and cancer (“cancer-associated thrombosis”), as long-term (first 3 months) anticoagulant therapy, LMWH is recommended over other agents (Grade 2C)
• In patients with DVT of the leg or PE who receive extended therapy, recommend no need to change the choice of anticoagulant after the first 3 months (Grade 2C)
• In patients who have recurrent VTE on VKA therapy (in the therapeutic range) or on dabigatran, rivaroxaban, apixaban, or edoxaban (and are believed to be compliant), recommend switching to treatment with LMWH at least temporarily (Grade 2C).

ESC Guidelines on the diagnosis and management of acute pulmonary embolism (2014, adapted):

• LMWH or fondaparinux is the recommended form of acute phase parenteral anticoagulation for most patients (class I, level A).
• In parallel to parenteral anticoagulation, treatment with a VKA is recommended, targeting an INR of 2-3 (class I, level B).
• As an alternative to the combination of parenteral anticoagulation with a VKA, one of the following methods of anticoagulation are recommended (class I, level B):
  • Rivaroxaban 15 mg BID for 3 weeks, then 20 mg daily
  • Apixaban 10 mg BID for 7 days, then 5 mg twice daily
  • Dabigatran 150 mg BID (with dose adjustments for some patients)
  • Edoxaban (after acute phase parenteral anticoagulation)
• New oral anticoagulants are not recommended in patients with severe renal impairment (class III, level A).

Design

• Multicenter, randomized, open-label, noninferiority trial
• N=4,832
  • Rivaroxaban (n=2,419)
  • Standard therapy (n=2,413)
• Setting: 263 sites in 38 countries
• Enrollment: 2007-2011
• Mean follow-up: ~7 months
• Analysis: Intention-to-treat
• Primary outcomes:
  • Symptomatic recurrent VTE
  • Major or clinically relevant non-major bleeding while on treatment

Population

Inclusion Criteria

• Acute symptomatic PE confirmed on imaging

Exclusion Criteria

• Treatment with LMWH, fondaparinux, or UFH for >48 hours before randomization
• >1 dose of a VKA before randomization
• Treatment of the event with thrombectomy, IVC filter placement, or a fibrinolytic
• Clear indication or contraindication to LMWH or VKA
• CrCl <30 mL/min
• Liver disease or ALT >3x ULN
• Bacterial endocarditis
• High risk for bleed or active bleeding
• SBP >180 mmHg or DBP >110 mmHg
• Childbearing not on contraception
• Pregnancy or breastfeeding
• Use of a strong CYP3A4 inhibitor or inducer
• Experimental pharmacologic therapy in prior 30 days
• Life expectancy <3 months

Baseline Characteristics

From the rivaroxaban group.

• Demographics: Age 57.9 years, male 54.1%,
• PMH: thrombophilia 5.7%, VTE 18.8%
• Health data:
  • Weight:
    • ≤50 kg: 1.6%
    • >50 - 100 kg: 84.1%
    • >100 kg: 14.3%
  • CrCl:
    • <30 mL/min: 0.2%
    • 30 - <50 mL/min: 8.6%
    • 50 - <80 mL/min: 26.3%
    • ≥80 mL/min: 64.3%
• Diagnostics performed: CT 87.4%, V/Q scan 11.7%, angiography 0.8%
• PE extent:
  • ≤25% of vasculature in 1 lobe: 12.8%
  • Intermediate: 57.5%
  • Multiple lobes and >25% of vasculature: 24.7%
• Symptomatic DVT concurrently: 25.1%
• Hospitalized: 89.1%
• ICU admission: 12.9%
• PE etiology:
  • Unprovoked 64.7%
  • Surgery/trauma: 17.2%
  • Immobilization: 15.9%
  • Estrogen therapy: 8.6%
  • Cancer: 4.7%
• Duration of treatment, intended (actual):
  • 3 months: 5.3% (93 days)
  • 6 months: 57.3% (182 days)
  • 1 year: 37.4% (355 days)
• Treatment duration, mean: 216 days

Interventions

• Randomization to either rivaroxaban or to standard treatment
  • Rivaroxaban: 15mg BID for 3 weeks then 20mg daily
  • Warfarin: Enoxaparin 1mg/kg BID and a VKA (warfarin or acenocoumarol) with discontinuation of enoxaparin when INR was ≥2 (goal 2-3) for 2 consecutive days and after ≥5 days of enoxaparin therapy
• NSAIDS and other antiplatelet agents were discouraged in both groups, with the exception of ASA <100mg daily and clopidogrel 75mg daily
• Followed for duration of treatment at regular intervals with evaluation of recurrent VTE if any concerning symptoms

Outcomes

Comparisons are rivaroxaban vs. warfarin. P values are for noninferiority unless otherwise specified.

Primary Outcomes

Symptomatic recurrent VTE

2.1% vs. 1.8% (HR 1.12; 95% CI 0.75-1.68; P=0.003)

Major or clinically relevant non-major bleeding

10.3% vs. 11.4% (HR 0.90; 95% CI 0.76-1.07; P=0.23)

Secondary Outcomes

Recurrent VTE types

Fatal PE: 2 vs. 1 event

Death, PE not ruled out: 8 vs. 5 events

Non-fatal PE: 22 vs. 19 events

DVT with PE: 0 vs. 2 events

DVT without PE: 18 vs. 17 events

Bleeding

Major: 1.1% vs. 2.2% (HR 0.49; 95% CI 0.31-0.79; P=0.003 for superiority)

Fatal: 0 vs. 1 event

Non-fatal in critical site: 0.3% vs. 1.1%

Drop in Hgb ≥2 g/dL or requiring ≥2 units PRBC transfusion: 0.7% vs. 1.1%

Clinically relevant non-major bleeding: 9.5% vs. 9.8%

VTE plus major bleeding

3.4% vs. 4.0% (HR 0.85; 95% CI 0.63-1.14; P=0.28)

All-cause mortality during treatment

2.4% vs. 2.1% (HR 1.13; 95% CI 0.77-1.65; P=0.53)

INR therapeutic (in warfarin group)

62.7%

Adverse Events

Adverse event during treatment

80.5% vs. 79.1% (P=0.24)

Serious: 19.7% vs. 19.5% (P=0.86)

Resulting in permanent discontinuation: 5.1% vs. 4.1% (P=0.10)

Leading to prolonged hospitalization: 19.7% vs. 17.9% (P=0.82)

Treatment discontinuation

10.7% vs. 12.3% (P=0.07)

Adverse event: 4.6% vs. 3.8%

Withdrawal of consent: 2.7% vs. 4.9%

Loss of follow-up: 0.3% vs. 0.4%

Criticisms

• INR only in therapeutic range ~60% of the time (although this is standard for clinical trials of warfarin antagonists)
• Open-label design introduces potential bias
• Small subgroup of patients with cancer so efficacy in this group is unclear and should probably remain LMWH per the CLOT trial (2003)
• No stratification based upon PE-related mortality risk factors
• The average participant was relatively young (58 years)

Funding

• Bayer HealthCare
• Janssen Pharmaceuticals, the makers of Xarelto (the brand name of rivaroxaban)