Author + information
Boudriot E, Thiele H, Walther T, et al. Randomized Comparison of Percutaneous Coronary Intervention With Sirolimus-Eluting Stents Versus Coronary Artery Bypass Grafting in Unprotected Left Main Stem Stenosis. J Am Coll Cardiol 2011;57:538–45.
The above paper recently published in the Journal was not the final version of the submitted manuscript. It was an earlier version that prompted comments from reviewers regarding statistical power and the interpretation of nonsignificant findings in a noninferiority test. The authors revised the manuscript to note that statistical power was an issue and to make it clear that their findings did not suggest that PCI was inferior to CABG. The changes are as follows:
1. On page 538, in the abstract, the “Conclusions” section should be revised as follows:
In patients with ULM stenosis, PCI with sirolimus-eluting stents did not show noninferiority to CABG at 12-month follow-up with respect to freedom from major adverse cardiac events, which is mainly influenced by repeated revascularization, whereas for hard endpoints, PCI results are favorable. A longer follow-up is warranted.
2. On page 544, right column, the following paragraph should be added to the “Study limitations” section:
Furthermore, our assumption in the sample size calculation of lower event rates in PCI (12.5%) than CABG (15.0%) resulted in a relatively small sample size estimate. If equal event rates had been assumed (15%), then the given sample size would have yielded only 62% power. Likewise, 80% power for the equal rate assumption would have required 160 subjects per group. Technically, the present paper might therefore be considered underpowered. From a statistical point of view, the failure to demonstrate noninferiority for PCI should not be misinterpreted as a demonstration of inferiority of PCI (in fact, a post-hoc head-to-head comparison yielded nonsignificant results arguing against an interpretation of inferiority).
3. On page 544, right column, the first 2 sentences of the “Conclusions” section should be revised to state:
PCI with sirolimus-eluting stents did not show noninferiority to CABG at 12-month follow-up with respect to freedom from major adverse cardiac events in patients with ULM stenosis. Failure to show noninferiority was mainly driven by the higher repeat revascularization rate, whereas death and myocardial infarction rates seem to be noninferior in PCI patients at lower perioperative morbidity.
Unfortunately, these changes addressing the reviewers' concerns did not appear in the published paper. We apologize for this error.
- American College of Cardiology Foundation