The Lancet, ISSN: 0140-6736, Vol: 374, Issue: 9691, Page: 682

Rivaroxaban versus enoxaparin after total knee arthroplasty

Gómez-Outes, Antonio; Suárez-Gea, M Luisa; Blázquez-Pérez, Antonio; Pozo-Hernández, Carmen; Vargas-Castrillón, Emilio

Alexander Turpie and colleagues

report the fourth comparative study of rivaroxaban and enoxaparin in major orthopaedic surgery (RECORD4), and conclude that there was no significant difference between treatment groups in the risk of major bleeding.

However, major bleeding rates in the RECORD4 study were three to six times lower than expected from the ODIXa dose-finding studies with rivaroxaban,

and about five times lower than those from contemporary studies in orthopaedic surgery.

This fact could be explained by the exclusion of most wound bleedings from major bleeding assessment in the RECORD studies. With the low rate of major bleeding events seen in RECORD4 (0·66% vs 0·27%), the probability of a type II error is 73% (probability of concluding that no difference between treatment groups exists when, in fact, there is a difference).

We have done sensitivity analyses to test what would have occurred in the RECORD4 study if overall major bleeding rates were similar to those in the ODIXa dose-finding studies

or other contemporary studies.

Such analyses suggest that an increase in major bleeding of 147% (relative risk 2·47) with rivaroxaban would have been significant in all the scenarios tested (table). Ancillary analyses by the US Food and Drug Administration

⁴

have also shown a consistent increased bleeding risk with rivaroxaban compared with enoxaparin in the RECORD programme.

Differences in the definition of major bleeding have important implications when assessing the relative safety of anticoagulants, making between-study comparisons difficult. Further experience is needed to assess whether the very low major bleeding rates reported in the RECORD programme are representative of what we would expect in unselected populations in clinical practice.

We declare that we have no conflicts of interest. The opinions expressed in this letter are of the authors alone and do not necessarily reflect the position or opinion of their institution or any other party.

Link to article