King's College London

Research portal

Choosing wisely: The impact of patient selection on efficacy and safety outcomes in the EINSTEIN-DVT/PE and AMPLIFY trials

Research output: Contribution to journalArticlepeer-review

Jan Beyer-Westendorf, Anthonie W.A. Lensing, Roopen Arya, Henri Bounameaux, Alexander T. Cohen, Philip S. Wells, Saskia Middeldorp, Peter Verhamme, Rodney Hughes, Nils Kucher, Akos F. Pap, Mila Trajanovic, Martin H. Prins, Paolo Prandoni, Jeffrey I. Weitz

Original languageEnglish
JournalThrombosis Research
Early online date21 Nov 2016
Accepted/In press13 Nov 2016
E-pub ahead of print21 Nov 2016


King's Authors


Background The results of the EINSTEIN-DVT/PE and AMPLIFY trials, which compared rivaroxaban and apixaban with conventional anticoagulation therapy for acute venous thromboembolism (VTE), respectively, are often compared. However, the trials differed in duration of therapy (3–12 and 6 months, respectively) and in patient selection (few exclusion criteria and more stringent exclusion criteria, respectively). Methods To determine the effect of these methodological differences on outcomes, the patients enrolled in EINSTEIN-DVT/PE were divided into 2 cohorts; the 5253 patients that matched the exclusion criteria for AMPLIFY and were treated for at least 6 months (cohort 1) and the 2368 patients who would have been ineligible for AMPLIFY (cohort 2). Results Compared with patients in cohort 2, those in cohort 1 were older and more often male and there were more with unprovoked VTE, prior VTE, cancer and known thrombophilia. In cohort 1, rivaroxaban would have significantly reduced recurrent VTE (relative risk [RR], 0.64; 95% confidence interval [CI], 0.43–0.95) and major bleeding (RR, 0.50; 95% CI, 0.30–0.82) compared with conventional therapy, whereas the two treatments would have had similar effects on recurrent VTE (RR, 1.08; 95% CI, 0.65–1.79) and major bleeding (RR, 1.03; 95% CI, 0.48–2.18) in cohort 2. Conclusions This analysis illustrates the influence of patient selection and treatments duration on outcome results and highlights the limitations of cross-trial comparisons.

Download statistics

No data available

View graph of relations

© 2020 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454