| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 1996;111:900
© 1996 Mosby, Inc.
LETTERS TO THE EDITOR |
Oxford Heart Centre
John Radcliffe Hospital
Headington
Oxford OX3 9DU, United Kingdom
To the Editor:
In a "double-blind," randomized study of 316 patients, which compared two methods of perfusion techniques (nonpulsatile vs pulsatile) and pH management (alpha-stat versus pH-stat), Murkin and colleagues
1 report that pulsatile perfusion resulted in a highly significant reduction in myocardial infarction, death, the need for an intraaortic balloon pump, and major complications. These dramatic conclusions are significantly flawed, however, both by fundamental limitations in the study design and by the particularly high mortality and morbidity in the nonpulsatile group:
The operations were performed by four surgeons, two of whom used crystalloid cardioplegia and two of whom used blood cardioplegia (we are not told of the route of administration of the latter). The fact that the four surgeons performed 94, 99, 21, and 102 operations introduces a further major difference between the groups, because simple calculation reveals that between 36% and 39% of patients must have received one type of cardioplegia and 64% to 61% of patients the other. To complicate matters even further, the surgeons using crystalloid cardioplegia performed significantly more grafts with significantly shorter crossclamp times (by approximately 20 minutes) than those using blood cardioplegia. Despite the highly significant differences in cardioplegic techniques, number of grafts, and duration of ischemia in the two groups, which are all major independent determinants of cardiac morbidity, the authors merely state, without providing any objective data, that "cardioplegia type did not correlate with adverse outcome."
The use of the term "double-blind" in the abstract is totally misleading. The four operating surgeons must have been aware both of the mode of perfusion (by observation of analog displays of arterial pulse pressure) and the type of cardioplegic solution administered. If so, it is conceivable that individual surgeons were influenced, albeit subconsciously, in the number of grafts they performed (as indeed was the case for surgeons using crystalloid cardioplegia) or the need for an intraaortic balloon pump. The fact that the research nurse and technician were blinded to the mode of perfusion is irrelevant with regard to the collection of objective data such as the number of patients who died, who had electrocardiographically confirmed infarcts, or who required intraaortic balloon pumps.
A further major concern in this study is the particularly high mortality and morbidity in the nonpulsatile group considering their favorable demographic features (their mean age was 61 years, and 95% had a left ventricular ejection fraction >35%). A mortality of 6% and a requirement for an intraaortic balloon pump in 7% of such good-risk elective patients appears particularly excessive and is unlikely to be simply explained by the use of nonpulsatile perfusion.
Although the authors are to be congratulated on undertaking a potentially important clinical trial, major reservations regarding their conclusions must remain until objective data are presented to eliminate the effects of different surgeons, cardioplegic techniques, number of grafts, and ischemic times on postoperative cardiac mortality and morbidity. Furthermore, the possibility that the apparent benefit of pulsatile perfusion was simply due to an excessively high mortality in the nonpulsatile group for some other reason must remain a genuine concern.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
