Ann Thorac Surg 2001;71:530-531
© 2001 The Society of Thoracic Surgeons
Invited commentary
Timothy J. Gardner, MDa
a Division of Cardiothoracic Surgery, Hospital of the University of Pennsylvania, 4 Silverstein, 3400 Spruce St, Philadelphia, PA 19104, USA
e-mail: gardner{at}mail.med.upenn.edu
Although one might assume that the influence of patient size and body habitus on outcome after major surgery such as CABG is already well understood, the present report, in fact, adds some important core information to the issue. The authors’ use of body mass index as opposed to body surface area results in greater physiological specificity when defining patient size. In addition, the handling of body size as a continuous variable rather than erroneously treating obesity versus normal body size as a dichotomous variable allows for more-reliable statistical analysis. Furthermore, the present study provides some interesting collateral insights about the influence of body mass, such as the apparent relationship between small body mass and the more likely development of hemodilution after exposure to cardiopulmonary bypass.
This report confirms several widely held assumptions regarding the influence of body size on outcome after coronary artery bypass grafting. It is widely believed that severely obese patients have more complications in the immediate postoperative period, including the need for prolonged ventilatory support and a higher incidence of sternal wound infections. The authors’ finding that operative death rates among these severely obese patients were not significantly increased is somewhat unexpected. It has been fairly widely accepted that women have a higher operative mortality than do men. The fact that there is a higher proportion of women in the low body mass index group adds a possible explanation for the higher operative risk for women than for men. What perhaps is the most unexpected finding from this report is the significantly higher operative mortality in those patients with the lowest body mass index values. It was in this group of patients that the incidence of comorbid conditions was highest. The increased early operative mortality and poorer two- and five-year survival in these patients can be explained by the existence of such extensive comorbid conditions.
What is not clarified in the present study, however, is the extent to which body mass index is an independent predictor of outcome. While very high body mass index may be correlated with such postoperative events as wound breakdown and time on the ventilator, surely it is the presence of comorbid conditions that accounts for the increased operative mortality and poor long-term results in patients at the extremes of body size. Because multivariate analysis was not done, this question remains unanswered here. The specific clinical relevance of deviations from median body mass index might be more clearly defined. Size, per se, appears only rarely to be a major determinant of early outcome, except perhaps in very obese patients.
