Author + information
- Simon Maltais, MD, MSc,
- Denis Bouchard, MD, FRCSC and
- Louis P. Perrault, MD, PhD, FRCSC, FACS⁎ ()
- ↵⁎Montreal Heart Institute, 5000 Belanger Street East Montreal, Quebec H1T 1C8 Canada
We read with great interest the study by Byrne et al. (1). This study claims percutaneous intervention before valve surgery is a safe approach compared to standard surgical revascularization concomitantly with valve surgery. We believe the investigators’ conclusions are extrapolated and incorrect based on nonsignificant scientific facts along with a study design that has insufficient power to support their findings.
First, the primary objectives were not clearly defined, as the researchers want to show the relative advantage of angioplasty over standard combined surgery. To demonstrate a difference in mortality, risk of bleeding, or number of tranfusions, one has to make a clear hypothesis and harbor a strict scientific methodology and study design. A retrospective analysis of a single-center heterogeneous group of patients (12 men and 14 women between 53 and 91 years of age) cannot scientifically conclude due to an unclear hypothesis.
Second, the mean time to angioplasty does not respect the recommendations accepted by the general scientific community, which should be 4 to 6 h. Only three patients had their intervention done limit the critical edge of 6 h (3 on 26; 11%).
The control group that was used to demonstrate a difference in mortality is inadequate. Comparison of 26 heterogeneous patients with a historical cohort of 100,000 patients introduces a fundamental bias. The calculated predicted mortality of 22% represents the median produced from a complex mathematical algorithm of a stratified risk ranging from 3.5% to 63.5%. From our standpoint, the comparison of the study group based on this extrapolated risk is questionable. Despite a mortality of 3.8% for the presented population (1 of 26 patients), the difference is not statistically significant (p < 0.05) and therefore the general conclusions cannot be supported. Finally, the survival curve at 36 months included fewer than 10 patients. Because this study seemed to be initially designed to look at the 30-day mortality, this portion of the curve is not interpretable.
The discussion addresses the risks of transfusions, reoperations for bleeding, and bleeding during surgery following antiplatelet therapy, but this study does not have the power or the design to conclude so. This type of secondary analysis represents, from a statistical standpoint, a “fishing expedition” for significant results.
In summary, although the concept of hybrid treatment may have some merit, the major biases and methodological flaws forgo the possibility of establishing valuable conclusions. Until proven otherwise with properly designed prospective controlled studies, surgical myocardial revascularization remains the gold standard for patients with combined coronary and valvular heart disease. The data presented in the study by Byrne et al. (1) do not support modification of our current treatment strategy in favor of the hybrid approach.
- American College of Cardiology Foundation