A Pooled Analysis of the Xalia and Xalia-Lea Non-Interventional Studies of Rivaroxaban Versus Standard Anticoagulation in Venous Thromboembolism

Introduction

XALIA was a prospective, non-interventional study of rivaroxaban versus standard anticoagulation for the treatment of deep vein thrombosis (DVT) with or without pulmonary embolism (PE). The companion XALIA-LEA study included regions not involved in XALIA (Latin America, Eastern Europe, the Middle East, Africa, and Asia-Pacific). Unlike XALIA, however, patients with isolated PE were also eligible for inclusion. This report is a pre-specified pooled analysis of the XALIA and XALIA-LEA study populations.

Methods

XALIA and XALIA-LEA are international, prospective, non-interventional studies of rivaroxaban for the treatment of acute venous thromboembolism (VTE) with the same basic design. Patients aged ≥18 years scheduled to receive ≥3 months of anticoagulation with rivaroxaban or standard anticoagulation (heparin/fondaparinux alone or overlapping with/followed by a vitamin K antagonist [VKA]) were eligible; follow up ended 12 months after final patient enrollment. Patients who received >2-14 days heparin/fondaparinux or 1-14 days VKA before rivaroxaban were called 'early switchers' and analyzed in a separate sensitivity analysis. The primary outcomes were major bleeding, VTE recurrence, and all-cause mortality.

Baseline characteristics, hospitalization, and primary outcome rates for the XALIA and XALIA-LEA pooled analysis set are described. Propensity score stratification (with 10 subclasses) was used to adjust for imbalances in baseline characteristics between treatment groups. Hazard ratios (HRs) for major bleeding and recurrent VTE were adjusted for active cancer at baseline and stratified by study, subclass, and type of VTE. The HR for all-cause mortality was stratified by active cancer at baseline, study, subclass, and type of VTE.

Results

7129 patients were enrolled; 3904 received rivaroxaban and 2551 received standard anticoagulation, comprising the safety analysis set. Twenty-one patients did not receive study medication; 653 patients were early switchers and not included in the main analysis. For the propensity score-stratified analysis, 2 patients from the rivaroxaban group and 8 standard anticoagulation-treated patients were excluded from the safety analysis set.

Patients given rivaroxaban were younger, and had lower rates of renal impairment, cancer, and provoked VTE than those given standard anticoagulation (Table 1).

In the propensity score-stratified analysis set, the least-squares mean duration of initial hospital stay was 6.5 days with rivaroxaban versus 9.7 days with standard anticoagulation (geometric means ratio 0.67; 95% confidence interval [CI] 0.63-0.73).

Annualized major bleeding rates were 1.7 % (n=39) and 3.9% (n=63) (propensity-score stratified HR 0.65; 95% CI 0.39-1.08), recurrent VTE rates were 2.5% (n=55) and 4.5% (n=71) (propensity score-stratified HR=0.85; 95% CI 0.54-1.32) and all-cause mortality was 1.8% (n=41) and 7.3% (n=117) (propensity score-stratified HR=0.55; 95% CI 0.33-0.91) with rivaroxaban and standard anticoagulation, respectively.

Summary and conclusions

This pooled analysis of the XALIA and XALIA-LEA studies provides a prospective global dataset for rivaroxaban treatment in routine clinical practice. Rivaroxaban-treated patients had lower rates of risk factors than patients given standard anticoagulation. In the propensity score-stratified analysis, adjusting for baseline characteristic imbalances, hospital stay duration was lower in the rivaroxaban group than the standard anticoagulation group. The propensity score-stratified analysis showed lower risks of major bleeding, recurrent VTE and all-cause mortality in patients treated with rivaroxaban versus those treated with standard anticoagulation. These results show that the single-drug approach with rivaroxaban is a safe and effective treatment for VTE in routine clinical practice. Rivaroxaban may also reduce healthcare resource utilization in VTE. However, although propensity score stratification can balance baseline covariates between the treatment groups, potential effects of unmeasured characteristics and confounders cannot be excluded.

View largeDownload slide

View largeDownload slide

Close modal