Interpretation of Rivaroxaban Noninferiority Trial Results for Pulmonary Embolism Treatment
Rivaroxaban is noninferior in efficacy to enoxaparin or a vitamin K antagonist (VKA) for treating pulmonary embolism. 1, 2
Understanding the Noninferiority Results
The noninferiority trial comparing rivaroxaban to standard therapy (enoxaparin/VKA) showed a hazard ratio of 1.12 (95% CI, 0.75-1.68) for the primary efficacy outcome of recurrent symptomatic venous thromboembolism (VTE). 1, 2
To correctly interpret these results, we need to understand the key components:
- Prespecified noninferiority margin: The trial used a margin of 2.0 2
- Hazard ratio and confidence interval: HR 1.12 (95% CI, 0.75-1.68) 2
- Statistical interpretation: For noninferiority to be established, the upper bound of the 95% CI must be below the prespecified margin
Since the upper bound of the confidence interval (1.68) is less than the prespecified noninferiority margin (2.0), rivaroxaban meets the criteria for noninferiority. 2
Clinical Evidence Supporting Noninferiority
The EINSTEIN-PE study, which specifically evaluated rivaroxaban for pulmonary embolism treatment, demonstrated:
- Primary efficacy outcome (recurrent VTE) occurred in 50 patients (2.1%) in the rivaroxaban group versus 44 patients (1.8%) in the standard-therapy group 2
- The hazard ratio of 1.12 with 95% CI of 0.75-1.68 established noninferiority (p=0.003 for noninferiority) 2
Safety Profile Comparison
Notably, rivaroxaban demonstrated a favorable safety profile compared to standard therapy:
- Principal safety outcome (major or clinically relevant non-major bleeding) occurred in 10.3% of rivaroxaban patients versus 11.4% in the standard-therapy group 2
- Major bleeding was significantly lower with rivaroxaban: 1.1% versus 2.2% with standard therapy (HR 0.49; 95% CI, 0.31-0.79; p=0.003) 2
Practical Advantages of Rivaroxaban
Rivaroxaban offers several practical advantages over standard therapy:
- Single-drug approach without the need for initial parenteral anticoagulation
- Fixed dosing without routine coagulation monitoring
- Potential for reduced hospital length of stay (mean reduction of 1.6 days) 1
Important Caveats
- Patients with severe renal impairment (CrCl <30 mL/min) were excluded from the trials 1
- Cancer patients were underrepresented in the clinical trials 1
- The European Society of Cardiology guidelines confirm rivaroxaban's noninferiority to standard therapy for VTE treatment 3
Common Pitfalls in Interpreting Noninferiority Trials
Misunderstanding the noninferiority margin: The margin of 2.0 means rivaroxaban could have up to twice the risk and still be considered noninferior. The actual results were well within this margin.
Confusing noninferiority with equivalence: Noninferiority trials are designed to show the new treatment is not unacceptably worse than the standard treatment, not that they are exactly equivalent.
Overlooking the confidence interval: The point estimate (HR 1.12) suggests slightly higher risk with rivaroxaban, but the confidence interval (0.75-1.68) includes 1.0, meaning the difference is not statistically significant.
In conclusion, based on the trial results with a hazard ratio of 1.12 (95% CI, 0.75-1.68) and a prespecified noninferiority margin of 2.0, rivaroxaban is noninferior in efficacy to standard therapy with enoxaparin/VKA for treating pulmonary embolism.