HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Permission Requests Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Utley, J. R. Search for Related Content PubMed PubMed Citation Articles by Utley, J. R.

Ann Thorac Surg 1996;61:1042-1043
© 1996 The Society of Thoracic Surgeons

---

Correspondence

Coronary Bypass in Women

Cardiothoracic Research & Education Foundation, 100 E Wood St, Suite 300, Spartanburg, Sc 29303

To the Editor:

The invited commentary by Dr Karl E. Hammermeister has raised important questions about our article, ``Intraoperative Blood Transfusion Is a Major Risk Factor for Coronary Artery Bypass Grafting in Women'' [1]. I write to describe how the analysis was developed and to address some of the issues.

Many times we did multiple regression or stepwise logistic regression analysis to determine patient and operative factors predictive of mortality and morbidity. Transfusion almost always came out as the strongest predictor of mortality and morbidity. The prediction was even stronger when transfusion during operation was the factor analyzed.

Because of our large database, which included 189 pieces of data on each patient, we ultimately had to eliminate patient and operative factors. We could not use all factors because of the number of events (eg, 74 deaths). We used the following criteria to select factors for analysis: (1) factors that were significantly different in men and women, (2) factors that were related to poor results in other studies (eg, severity of disease or vessel size), (3) pathophysiologic reason relating the factor to increased risk (eg, there is reason for the relation of low ejection fraction to risk but not for small body size), and (4) relative independence of factors.

One of the important points is that three of the predictors of mortality are more common or abnormal in men, and one is not different between men and women, yet women suffer greater mortality. Therefore there must be something about women that causes them to have greater risk other than ejection fraction, preoperative creatinine level, and atherosclerosis of the ascending aorta, which are more common or abnormal in men, or emergency operation, which is not different between men and women. If this strongly predictive factor is not transfusion during operation, then what is the factor or ``root cause''? Predictors of transfusion during operation in our database are age, female sex, hypertension, unstable angina, claudication, low ejection fraction, elevated left ventricular end-diastolic pressure, prebypass hematocrit, preoperative creatinine level, and weight. Except for preoperative hematocrit, none of these predictors are correctable.

We do not believe that small body weight, per se, should carry greater risk for coronary artery operations, especially because vessel size index is not predictive of mortality or morbidity. When body weight, preoperative hematocrit, and intraoperative transfusion are the three factors subjected to stepwise logistic regression analysis to predict mortality and morbidity, we find that intraoperative transfusion is always the strongest predictor.

With cardiotomy suction and cell saving, virtually all the patient's red blood cells are preserved during the operation. The indication for intraoperative transfusion, in our hands, is low hematocrit. In response to Dr Hammermeister's comments, we use identical operative technique and indication for transfusion in men and women. Intraoperative bleeding is not an indication for intraoperative transfusion with the use of cell-saving techniques. In our discussion we mentioned that ``our protocols call for transfusion during operation to increase oxygen carrying capacity rather than for volume replacement.'' The comment by Dr Hammermeister that there might be a difference in coagulability of women on cardiopulmonary bypass seems rather far-fetched because all patients are anticoagulated with heparin to an activated clotting time of greater than 400 seconds. Hemostasis is not dependent on coagulation during complete heparinization and cardiopulmonary bypass. The cardiopulmonary bypass apparatus and the volume of hemodilution were identical in men and women; therefore, the suggestion that there might be a difference in hemodilution was not the case.

We thoroughly understand the comments by Dr Hammermeister about patient-related risk factors and factors related to process of care. We must totally disagree with Dr Hammermeister that patient-related risk factors must be determined before the episode of care or entrance into the operating room occurs. Many risk factors that are totally related to the patient are often not determined until observed in the operating room. Such factors include atherosclerosis of the aorta, recent myocardial infarction, and vessel size.

Recognition that intraoperative transfusion is associated with greater mortality and morbidity offers the opportunity to explore strategies that might reduce risk. The benefit would be greatest in women.

Reference

1. Utley JR, Wilde EF, Leyland SA, Morgan MS, Johnson HD. Intraoperative blood transfusion is a major risk factor for coronary artery bypass grafting in women. Ann Thorac Surg 1995;60:570–5.[Abstract/Free Full Text]

This Article Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Permission Requests Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Utley, J. R. Search for Related Content PubMed PubMed Citation Articles by Utley, J. R.