Oral rivaroxaban versus standard therapy for the treatment of symptomatic venous thromboembolism: a pooled analysis of the EINSTEIN-DVT and PE randomized studies

Martin H Prins; Anthonie WA Lensing; Rupert Bauersachs; Bonno van Bellen; Henri Bounameaux; Timothy A Brighton; Alexander T Cohen; Bruce L Davidson; Hervé Decousus; Gary E Raskob; Scott D Berkowitz; Philip S Wells

doi:10.1186/1477-9560-11-21

Thrombosis Journal

Table 2 Efficacy and safety outcomes and net clinical benefit in all patients and selected patient subgroups

From: Oral rivaroxaban versus standard therapy for the treatment of symptomatic venous thromboembolism: a pooled analysis of the EINSTEIN-DVT and PE randomized studies

	Mean duration of treatment		Recurrent venous thromboembolism			Nonmajor clinically relevant and major bleeding			Major bleeding			Net clinical benefit
Patient group	Rivaroxaban	Enoxaparin/VKA	Rivaroxaban n/N (%)	Enoxaparin/VKA n/N (%)	HR (95% CI)	Rivaroxaban n/N (%)	Enoxaparin/VKA n/N (%)	HR (95% CI)	Rivaroxaban n/N (%)	Enoxaparin/VKA n/N (%)	HR (95% CI)	Rivaroxaban n/N (%)	Enoxaparin/VKA n/N (%)	HR (95% CI)
	Days ± SD	Days ± SD
All	207.6 ± 95.9	204.0 ± 97.2	86/4150 (2.1)	95/4131 (2.3)	0.89 (0.66–1.19)	388/4130 (9.4)	412/4116 (10.0)	0.93 (0.81–1.06)	40/4130 (1.0)	72/4116 (1.7)	0.54 (0.37–0.79)	134/4150 (3.2)	169/4131 (4.1)	0.77 (0.61–0.97)
Fragile	196.8 ± 97.5	187.2 ± 102.5	21/791 (2.7)	30/782 (3.8)	0.68 (0.39–1.18)	97/788 (12.3)	109/779 (14.0)	0.85 (0.64–1.11)	10/788 (1.3)	35/779 (4.5)	0.27 (0.13–0.54)*	36/791 (4.6)	66/782 (8.4)	0.51 (0.34–0.77)^†
Nonfragile	210.1 ± 95.4	208.0 ± 95.5	65/3359 (1.9)	65/3349 (1.9)	0.98 (0.69–1.38)	291/3342 (8.7)	303/3337 (9.1)	0.95 (0.81–1.12)	30/3342 (0.9)	37/3337 (1.1)	0.80 (0.49–1.29)	98/3359 (2.9)	103/3349 (3.1)	0.93 (0.71–1.23)
Cancer	179.0 ± 96.9	167.7 ± 103.1	16/316 (5.1)	20/281 (7.1)	0.69 (0.36–1.33)	48/316 (15.2)	44/278 (15.8)	0.93 (0.62–1.41)	9/316 (2.8)	14/278 (5.0)	0.53 (0.23–1.23)	26/316 (8.2)	37/281 (13.2)	0.60 (0.36–0.99)
No cancer	209.3 ± 95.6	205.8 ± 96.6	70/3834 (1.8)	75/3850 (1.9)	0.93 (0.67–1.29)	340/3820 (8.9)^‡	368/3832 (9.6)^‡	0.92 (0.79–1.06)	31/3820 (0.8)^‡	58/3832 (1.5)^‡	0.53 (0.34–0.82)	108/3834 (2.8)	132/3850 (3.4)	0.81 (0.63–1.05)
Previous DVT/PE	259.0 ± 102.0	252.6 ± 107.1	11/791 (1.4)	25/819 (3.1)	0.45 (0.22–0.91)^§	70/788 (8.9)	91/813 (11.2)	0.77 (0.57–1.06)	7/788 (0.9)	14/813 (1.7)	0.51 (0.21–1.27)	20/791 (2.5)	39/819 (4.8)	0.52 (0.31–0.90)
No previous DVT/PE	195.4 ± 90.3	192.0 ± 90.8	75/3359 (2.2)	70/3312 (2.1)	1.04 (0.75–1.45)	318/3342 (9.5)	321/3303 (9.7)	0.96 (0.82–1.12)	33/3342 (1.0)	58/3303 (1.8)	0.54 (0.35–0.83)	114/3359 (3.4)	130/3312 (3.9)	0.84 (0.66–1.08)
Clot burden (limited)^¶	197.6 ± 93.2	197.1 ± 94.1	10/799 (1.3)	19/815 (2.3)	0.51 (0.24–1.10)	73/796 (9.2)	76/813 (9.3)	0.97 (0.70–1.34)	8/796 (1.0)	11/813 (1.4)	0.75 (0.30–1.87)	20/799 (2.5)	30/815 (3.7)	0.65 (0.37–1.16)
Clot burden (intermediate)^¶	215.0 ± 94.5	206.1 ± 96.4	41/1873 (2.2)	49/1881 (2.6)	0.82 (0.54–1.24)	181/1864 (9.7)	189/1876 (10.1)	0.95 (0.78–1.17)	20/1864 (1.1)	32/1876 (1.7)	0.62 (0.36–1.09)	65/1873 (3.5)	81/1881 (4.3)	0.79 (0.57–1.10)
Clot burden (extensive)^¶	205.8 ± 97.8	205.2 ± 100.0	35/1364 (2.6)	26/1327 (2.0)	1.29 (0.78–2.15)	126/1359 (9.3)	134/1326 (10.1)	0.90 (0.71–1.15)	11/1359 (0.8)	28/1326 (2.1)	0.36 (0.18–0.73)	48/1364 (3.5)	56/1327 (4.2)	0.81 (0.55–1.19)

*For major bleeding, the p-value for the treatment group × fragility interaction was 0.011, suggesting heterogeneity. ^†For net clinical benefit, the p-value for the treatment group × fragility interaction was 0.017. ^‡In this analysis, six DVT patients were randomized to rivaroxaban but received comparator. ^§For recurrent VTE, the p-value for the treatment group × previous DVT/PE was 0.032. For all other outcomes and analyses the interactions between treatment group and the subgroups of fragility [yes/no], baseline cancer [yes/no], previous DVT/PE [yes/no], clot burden for outcomes [limited/intermediate/extensive]), the p_interactions were larger than 0.05 (nonsignificant) calculated by the corresponding Cox proportional hazards models. ^¶In the clot burden analysis, a total of 71 patients in the rivaroxaban group and 68 patients in the enoxaparin/VKA group had missing data. A further 43 and 40 patients in the rivaroxaban and enoxaparin/VKA groups, respectively, did not have confirmed DVT and PE.
CI Confidence interval, DVT Deep vein thrombosis, HR Hazard ratio, PE Pulmonary embolism, VKA Vitamin K antagonist.

Back to article page