| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
University of Virginia, PO Box 800679, Charlottesville, VA 22908
(Email: gorav{at}virginia.edu).
We thank Dr Savage [1] for his interest in our article demonstrating improved early outcomes after mitral repair vs replacement in patients aged older than 75 years [2]. He raises some important points that, in his opinion, do not reflect our conclusion and warrant a response.
The study was addressed to compare the results of mitral repair versus replacement in an elderly population. There appears to be a bias towards replacement in this group of patients according to the Society of Thoracic Surgeons (STS) database that is not present in younger patients. Because this study was retrospective, we do not have the ability to determine what factors were involved in the decision to repair or replace the valve. There are no randomized data on mitral repair versus replacement in any age population or etiology. As such, we are currently involved in studies randomizing patients to mitral repair or replacement through the National Institutes of Health-funded Cardiothoracic Surgery Network.
The letter to the editor raises concerns about the omission of cross-clamp and cardiopulmonary bypass time, from the multivariate analysis. In the present study, cross-clamp time was not statistically different between mitral repair and replacement both in the young and elderly patients by univariate analysis. Cardiopulmonary bypass times were longer in elderly patients undergoing replacement compared with repair (p = 0.04) but not in young patients. Model selection for multivariate analysis in relatively small studies such as this can be the subject of much debate because there is no consensus on the best model selection paradigm. With a relatively small sample size and number of outcomes, we are forced to select from among many possible variables only a limited few to include in the model to avoid over-fitting. We used the common technique of selecting those factors that we thought were most clinically relevant. When cardiopulmonary bypass time is included in the model, statistical aberrations consistent with colinearity appear, creating an unreliable model in which the mitral valve replacement does loose statistical significance but retains a strong trend. Clearly, in our experience, mitral repair is often a simpler and shorter operation than replacement.
The long-term survival was significantly improved with mitral repair (p = 0.04) in elderly patients. Dr Savage is correct in noting that once postoperative deaths are accounted for, survival is equivalent with repair versus replacement. Although this is an important point to note, it would not be judicious to ignore the improved perioperative mortality with mitral repair in the present study.
We believe that there exists a bias towards mitral replacement in elderly patients due to concerns of failed repair. We demonstrate a low reoperation rate and improved survival in elderly patients undergoing repair and believe this is an important contribution to the growing literature on our aging population. We should emphasize that our conclusions note that mitral repair can be performed when feasible, even in elderly patients. We also concur with the growing literature suggesting that mitral procedures should be performed by experienced surgeons and believe this is especially true in elderly patients.
On behalf of all the coauthors, we would like to thank the editor for the opportunity to respond to these comments.
 |
References |
|---|
 Top References |
|---|
Related Article
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
