doi: 10.1161/CIRCULATIONAHA.106.663047
(Circulation. 2006;114:e649.)
© 2006 American Heart Association, Inc.
Correspondence |
Response to Letter Regarding Article, "Impact of Internal Mammary Artery Conduit on Long-Term Outcomes After Percutaneous Intervention of Saphenous Vein Graft"
Division of Cardiology, Department of Internal Medicine, Duke University Medical Center and the Duke Clinical Research Institute, Durham, NC
We thank Drs Brilakis and Banerjee for their insightful comments concerning our recently published study.1 Our goal was to evaluate whether percutaneous interventions of saphenous vein graft (SVG) had any influence on long-term outcomes. In accordance with the retrospective and observational nature of our study, we used currently accepted methodology that included adjustments using a Cox proportional hazard model to account for differences in baseline confounders in the 4 groups. In addition, we adjusted for the severity of coronary artery disease, left ventricular ejection fraction, and graft indices using previously published models.2 The graft indices account for the differences in the target vessels to which the grafts are attached and also for the presence or absence of stenosis of the native coronary arteries and graft. Despite these rigorous adjustments, which included accounting for differences in coronary and vein graft disease, mortality was not influenced by SVG intervention (SVG intervention adjusted hazard ratio, 0.94; 95% confidence interval, 0.81 to 1.10).
For myocardial infarction, we used similar adjustment strategies. We found that adjusted event-free rates from nonfatal myocardial infarction at follow-up were significantly lower for both SVG intervention groups compared with the non-SVG intervention group (SVG intervention adjusted hazard ratio, 3.19; 95% confidence interval, 2.18 to 4.66). Thus, within the known constraints and limitations of an observational analysis, we believe that our conclusions are reasonable regarding a lack of influence of SVG intervention on survival and adverse association with myocardial infarction.
The observational nature of our analysis precludes any inference regarding causation. We agree that vigorous statistical adjustment may not completely account for the pathophysiological behavior of SVG and that, despite these adjustments, diseased SVG grafts per se may still have higher long-term incidence of myocardial infarction compared with nondiseased grafts. This hypothesis, however, remains to be proven. Finally, the clinical dilemma of the influence of SVG interventions on long-term outcomes would be best resolved by a rigorous randomized clinical trial that assigned patients with SVG disease randomly to intervention or medical management. The data from our study and other similar investigations should be regarded as hypothesis-generating and should form the foundation for future prospective randomized trial addressing this issue.
 |
Acknowledgments |
|---|
Disclosures
None.
|
References |
|---|
 Top References |
|---|
- Mehta RH, Honeycutt EF, Peterson ED, Granger CB, Halabi AR, Honeycutt EF, Shaw LK, Califf RM, Harrington RA, Sketch MH. Impact of internal mammary artery conduit on long-term outcomes after percutaneous intervention of saphenous vein graft. Circulation. 2006; 114 (suppl I): I-396–I-401.[CrossRef][Medline]
[Order article via Infotrieve]
- Liao L, Kong DF, Shaw LK, Sketch MH Jr, Milano CA, Lee KL, Mark DB. A new anatomic score for prognosis after cardiac catheterization in patients with previous bypass surgery. J Am Coll Cardiol. 2005; 46: 1684–1692.
[Abstract/Free Full Text]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||