MAGELLAN
PubMed • Full text • PDF
Clinical Question
In patients admitted to the hospital with an acute medical illness, is short-course rivaroxaban non-inferior to short-course enoxaparin in preventing VTE or its complications without increasing bleeding? Also, is extended-course rivaroxaban superior to short-course enoxaparin in preventing VTE or its complications without increasing bleeding?
Bottom Line
Short-course rivaroxaban is non-inferior to short-course enoxaparin in preventing VTE and its complications, although it increases the risk of bleeding. Extended-course rivaroxaban is superior to short-course enoxaparin for prevention of VTE and its complications, but is also associated with more bleeding.
Major Points
Guidelines recommend LMWH, unfractionated heparin, or fondaparinux for the prevention of VTE in hospitalized, non-surgical patients at increased risk for clot development.[1] The role of rivaroxaban, a direct factor Xa inhibitor, is unknown in prevention of VTE and its complications in this population. RECORD1[2] and RECORD3[3] (both published 2008) demonstrated efficacy of short-course rivaroxaban over short-course enoxaparin in prevention of VTE after hip and knee replacements, respectively. These interventions were not associated with increased rates of bleeding. Interestingly, RECORD2[4] (2008) demonstrated that extended-course rivaroxaban was more effective in preventing VTE than short-course enoxaparin without increasing bleeding rates following hip replacement. The efficacy and safety of rivaroxaban as a VTE prophylactic in acute illness was unknown.
The 2013 Multicenter, Randomized, Parallel Group Efficacy and Safety Study for the Prevention of Venous Thromboembolism in Hospitalized Acutely Ill Medical Patients Comparing Rivaroxaban with Enoxaparin (MAGELLAN) trial randomized 8,101 patients to rivaroxaban for 35 days or enoxaparin for 10 days followed by 25 days of placebo. Rivaroxaban was non-inferior to enoxaparin in the first 10 days in preventing VTE or its complications, though it was associated with an increased risk of bleeding with a NNH of 63 for major bleeding. Extended-course rivaroxaban was associated with significantly fewer events of VTE and its complications than short-course enoxaparin with a NNT of 35. This was offset by a NNH of 77 for bleeding.
Guidelines
As of April 2015, no guidelines have been published that reflect the results of this trial.
Design
- Multicenter, randomized, prospective, double-blind, controlled trial
- N=8,101
- Rivaroxaban (n=4,050)
- Enoxaparin and placebo (n=4,051)
- Setting: 556 centers in 52 countries
- Enrollment: 2007-2010
- Follow-up: Not specified, at least 35 +/- 4 days
- Analysis: Modified intention-to-treat and per-protocol anaysis
- Primary outcomes:
- Asymptomatic proximal DVT, symptomatic proximal or distal DVT, symptomatic PE (nonfatal), or VTE mortality at first 10 days (non-inferiority)
- Asymptomatic proximal DVT, symptomatic proximal or distal DVT, symptomatic PE (nonfatal), or VTE mortality at first 35 days (superiority)
- Clinically-relevant bleeding
Population
Inclusion Criteria
- Age ≥40 years
- Decreased mobility
- Hospitalization within 72 hours prior for one of the following:
- NYHA class III or IV HF
- Active cancer
- Acute ischemic CVA
- Acute infectious and inflammatory diseases
- Acute respiratory insufficiency
- Plus ≥1 risk factor (unless admitted for HF NYHA class III or IV symptoms, active cancer, or acute ischemic CVA with lower extremity weakness or paralysis):
- Severe varicosities
- Chronic venous insufficiency
- Cancer history
- DVT/PE history
- HF history with NYHA class III or IV
- Thrombophilia
- Age ≥75 years
- BMI ≥35 kg/m2
- Acute infectious disease
Exclusion Criteria
- Contraindications for LMWH or rivaroxaban
- Elevated bleeding risk
- Severe renal insufficiency
- Liver abnormalities
- HIV
- Uncontrolled HTN
- Drug or alcohol abuse
- Pregnancy or breastfeeding
- >2 days of prophylactic anticoagulation
- CYP450 3A4 inhibitor use
- Indication for fibrinolysis
- Indication for >14 days of anticoagulation
- Use of sequential compression devices
Baseline Characteristics
From the rivaroxaban group.
- Demographics: Age 71 years, male 55.6%
- Race: White 68.7%, Asian 19.9%, other 6.9%
- Baseline health data: Weight 77.5 kg, BMI 28.2 kg/m2
- Hospital duration: 11 days
- Creatinine clearance:
- <30 mL/min: 2.0%
- 30-50 mL/min: 19.5%
- 50-80 mL/min: 37.0%
- >80 mL/min: 39.1%
- D-dimer: 0.94 ug/mL
- Medical disease: Infection 45.8%, HF 32.3%, respiratory insufficiency 27.3%, ischemic CVA 17.3%, active cancer 7.3%, inflammatory or rheumatic condition 3.8%, other 0.8%, ≥2 diseases 30.6%
- VTE risk:
- Age ≥75: 38.3%
- HF, NYHA class III or IV: 34.8%
- Cancer: 17.3%
- Acute ischemic CVA with leg paresis: 16.3%
- Chronic venous insufficiency: 15.2%
- BMI ≥15.1%
- Acute infection: 14.0%
- Varicosis, severe: 12.4%
- Previous DVT or PE: 5.0%
- HRT: 1.2%
- Recent major surgery: 0.7%
- Inherited or acquired thrombophilia: 0.4%
- Recent "serious trauma": 0.2%
Interventions
- Randomization to one of two groups:
- Rivaroxaban 10 mg daily for 35 +/- 4 days
- Enoxaparin 40 mg daily for 10 +/- 4 days
- Placebo version of opposite group's therapy was given.
- All patients had ultrasonographic studies to evaluate for DVT after the last day of the medication or placebo or whenever clinically indicated
- 1,760 patients did not have this follow-up and were excluded from analysis.
- CT, V/Q, or angiographic modalities were used to evaluate for PE when clinically relevant
Outcomes
Presented as rivaroxaban vs. enoxaparin and placebo. Analyses are modified intention-to-treat unless otherwise specified.
Primary Outcomes
- Asymptomatic proximal DVT, symptomatic proximal or distal DVT, symptomatic PE (nonfatal), or VTE mortality at first 10 days
- 2.7% vs. 2.7% (RR 0.97; 95% CI 0.71-1.31; P=0.003 for non inferiority, per-protocol)
- Asymptomatic proximal thrombosis or DVT, symptomatic PE (nonfatal), or VTE mortality at first 35 days
- 4.4% vs. 5.7% (RR 0.77; 95% CI 0.62-0.96; P=0.02 for superiority; NNT=77)
- Major bleeding or clinically relevant non-major bleeding
- Within two days of last dose of study medication.
- 10 days: 2.8% vs. 1.2% (RR 2.3; 95% CI 1.63-3.17; P<0.001; NNH=63)
- 35 days: 4.1% vs. 1.7% (RR 2.5; 95% CI 1.85-3.25; P<0.001; NNH=42)
Secondary Outcomes
- Asymptomatic proximal DVT
- 10 days: 2.4% vs. 2.4% (no P given)
- 35 days: 3.5% vs. 4.4% (no P given)
- Symptomatic proximal or distal DVT
- 10 days: 0.2% vs. 0.2% (no P given)
- 35 days: 0.4% vs. 0.5% (no P given)
- Symptomatic non-fatal PE
- 10 days: 0.2% vs. <0.1% (no P given)
- 35 days: 0.3% vs.0.5% (no P given)
- VTE mortality
- 10 days: 0.1% vs. 0.2% (no P given)
- 35 days: 0.6% vs. 1.0% (no P given)
- Asymptomatic proximal DVT, symptomatic DVT, symptomatic PE (non-fatal), or all-cause mortality
- Mortality was through day 41
- 35 days: 8.6% vs. 9.2% (RR 0.93; 95% CI 0.80-1.09; P=0.38)
- Asymptomatic proximal DVT, symptomatic DVT, symptomatic PE (non-fatal), or VTE mortality
- Analysis of patients who had "adequate assessment of venous thromboembolism at day 10"
- 10 days: 3.0% vs. 3.1% (RR 0.99; 95% CI 0.75-1.30; P=0.95)
- Symptomatic VTE (non-fatal)
- 10 days: 0.5% vs. 0.3% (RR 1.50; 95% CI 0.72-3.11; P=0.28)
- 35 days: 0.6% vs. 0.7% (RR 0.82; 95% CI 0.47-1.43; P=0.48)
- All-cause mortality
- Through day 15 for 10 day analysis and day 41 for the 35 day analysis.
- 10 days: 2.2% vs. 2.0% (statistics not given)
- 35 days: 5.1% vs. 4.8% (statistics not given)
- CV mortality, MI, or acute ischemic CVA
- 10 days: 1.0% vs 1.0% (RR 1.02; 95% CI 0.66-1.58; P=0.91)
- 35 days: 1.8% vs. 1.6% (RR 1.11; 95% CI 0.79-1.55; P=0.55)
- Net clinical benefit or harm
- Composite of primary non-bleeding outcome, major bleeding,or clinically relevant non-major bleeding during treatment (days 15 and 41 respectively).
- 10 days: 6.6% vs. 4.6% (RR 1.44; 95% CI 1.18-1.77; P<0.001)
- 35 days: 9.4% vs. 7.8% (RR 1.21; 95% CI 1.03-1.43; P=0.02)
Adverse Events
- Fatal bleeding
- 10 days: 5 patients vs. 1 patient (statistics not given)
- 35 days: 7 patients vs. 1 patient (statistics not given)
- 7 fatal bleeds include pulmonary hemorrhage, ICH, retroperitoneal bleeding, and GI bleeding.
Criticisms
- Routine ultrasonographic screening for DVT is not the standard of care
- Exclusion of many patients because of no screening ultrasound
- Safety of rivaroxaban in acutely ill patients is unknown, especially as there is no known antidote[5]
- Did not look for below-knee thrombus[5]
- Uneven treatment duration; claims of rivaroxaban superiority by the authors is misleading[5]
- No report of VTE after day 35, concerning because of rebound hypercoaguability after cessation of Xa inhibitors[5]
Funding
- Bayer HealthCare Pharmaceuticals
- Janssen Research and Development
- Authors with multiple financial disclosures
Further Reading
- ↑ Guyatt GH, et al. "Executive summary: Antithrombotic therapy and prevention of thrombosis, 9th ed: American College of Chest Physicians evidence-based clinical practice guidelines." Chest. 2012;141(2 suppl):7S-47S.
- ↑ Eriksson BI, et al. "Rivaroxaban versus enoxaparin for thromboprophylaxis after hip arthroplasty." The New England Journal of Medicine. 2008;358:2765-2775.
- ↑ Lassen MR et al. "Rivaroxaban versus enoxaparin for thromboprophylaxis after total knee arthroplasty." The New England Journal of Medicine. 2008;358:2776-2786.
- ↑ Kakkar AK, et al. "Extended duration of rivaroxaban versus short-term enoxaparin for prevention of venous thromboembolism after total hip arthroplasty: A double-blind, randomised controlled trial." The Lancet. 2008;372(9632)31-39.
- ↑ 5.0 5.1 5.2 5.3 Multiple authors. "Correspondence: Rivaroxaban for thromboprophylaxis in acutely ill medical patients." The New England Journal of Medicine. 2013;368(20):1944-1946.