| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
Department of Anesthesiology and Critical Care; Centre Hospitalier Universitaire Pitié Salpêtrière; Assistance Publique des hôpitaux de Paris; Université Pierre et Marie-Paris 6., 47 bld de l'hôpital, 75013; Paris, France; yannick.le-manach{at}psl.ap-hop-paris.fr
In Response:
We agree with Dr. Pace1 that the nonrandomized design is probably the main weakness of our study2; nevertheless, the bias reduction has been our main concern. Indeed, three different statistical methods were used, but we assume that the unknown (or not recorded) variables could not have been taken into account. This limitation is observed in each nonrandomized study. Furthermore, the comment about a potential bias related to the before-after design has been assessed by a pooled analysis, which is presented and discussed in the original publication.2
Our study does not demonstrate that the postoperative withdrawal of a chronic statin therapy induces postoperative cardiac events, because this causality could only be proven by randomized studies. However, this causality link has never been asserted, and the well-recognized limitations of a retrospective study have been acknowledged. Nevertheless, the observed association remains important and should be considered in clinical practice, especially since these results had been reported by other groups in the same3 or similar4 conditions.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
|
