|
|
||||||||
|
Interact CardioVasc Thorac Surg 2009;8:522. doi:10.1510/icvts.2008.192757A1 © 2009 European Association of Cardio-Thoracic Surgery
eResponse: Individualization of blood cardioplegia administration modeDresden Heart Centre, Dresden University of Technology, Fetscherstrasse 76, 01307 Dresden, Germany Is repeated administration of blood-cardioplegia really necessary? First of all, we would like to thank Professor Apostolakis for his rich eComment. 1. More than one study brought unquestionable proofs of the biochemical changes taking place during ischemia/reperfusion. Nevertheless, our clinical study [1] has brought very acceptable clinical results, and that is why we published our work, to raise the question of the clinical impact of the biochemical changes recorded. Therefore, we recommend performing more prospective studies investigating that. 2. The non-randomization resulted because some of our surgeons operate using single shot cardioplegia, while others use the classic method. Statistically, this is not a reliable randomization, therefore, the study was analyzed as a non-randomized one. The presumption that the non-randomization was built upon the preoperative surgical risk evaluation is simply not correct. 3. We wonder that Professor Apostolakis concluded that the positive effect on myocardial infarction (with P-value of 0.002) is of no value because the study is not randomized, but at the same time accepted and even highlighted the negative effect on intraoperative need of inotropics (with P-value of 0.038) resulting from the same study. However, we stand by both results. 4. Concerning the comparison between the two groups, we think that there is a fundamental misunderstanding concerning statistical methodology. The statistical methodology was built entirely on a multivariate analysis testing the independent effect of single-shot cardioplegia on each outcome. The regression models were built for all patients, and the variable to be tested was introduced in the logistic regression model built for each outcome to test its significance. So there are no groups, there is only one pool of patients. 5. Myocardial damage was expressed in our study [1] as postoperative myocardial infarction, defined as elevated myocardial enzymes and/or pathological ECG changes. 6. Concerning the independent effect of CPB time, the answer is actually quite simple and is related to the above-mentioned statistical methodology. As the operative time and the CPB time are normally highly correlated, introducing the more significant of them into the logistic regression model will automatically decrease the statistical significance of the other. In case of intraoperative need of inotropics, the correlation was high enough to decrease the independent effect of CPB time to the level of insignificance. That was not the case in intraoperative need for intra-aortic balloon pump (IABP), indicating that other factors prolonged the operative time other than CPB time (e.g. TEE before or after CPB, dissection time in redo, etc.). 7. Concerning postoperative dialysis, Professor Apostolakis repeated the question that we had already admitted in our article that we do not have an answer to. Our comment is already published in the discussion section of our article [1]. 8. We agree with Professor Apostolakis with his opinion about the individual threshold of every patient to the CPB time. As this topic is beyond the scope of our study, we hope that future studies will address this issue.
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 |