| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL ARTICLE |
Heart Institute Lahr/Baden, Lahr, Germany
Juergen Ennker, MD, Tel: +49 7821 9251000, Fax: +49 7821 9251 391000, Email: juergen.ennker{at}mediclin.de, Medical Director, Surgeon in Chief, Mediclin Heart Institute Lahr/Baden, Hohbergweg 2, 77933 Lahr, Germany.
ABSTRACT
Following recent studies concerning the increased risk of coronary artery bypass surgery for women, the impact of sex is still a controversial issue. Between 1996 and 2006, 9,527 men and 3,079 women underwent isolated coronary artery bypass in our institute. To adjust for dissimilarities in preoperative risk profiles, propensity score-based matching was applied. Before adjustment, clinical outcomes in terms of operative mortality, arrhythmias, intensive care unit stay, and maximum creatine kinase-MB levels were significantly different for men and women. After balancing the preoperative characteristics, including height, no significant differences in clinical outcomes were observed. However, there was decreased use of internal mammary artery, less total arterial revascularization, and increasing creatine kinase-MB levels with decreasing height. This study supports the theory that female sex per se does not increase operative risk, but shorter height, which is more common in women, affects the outcome, probably due to technical difficulties in shorter patients with smaller internal mammary arteries and coronary vessels. Thus women may especially benefit from sequential arterial grafting.
Key Words: Coronary Artery Bypass • Mortality • Sex Factors
INTRODUCTION
Coronary artery bypass grafting (CABG) has evolved in the last 50 years from being a rarely performed high-risk operation into a routine procedure. This is mainly due to improved surgical techniques, better medications, superior patient care, and advanced education of physicians.1,2 The prediction of surgical outcome based on preoperative data is both very beneficial and in some instances, controversial. Therefore, the mortality risk of CABG has been the focus of numerous studies in the last few years, which have differed in respect of the time period examined, data compared, and the inclusion of patients with various concomitant procedures. The impact of sex on mortality and morbidity after CABG is an ongoing issue. Reports dating back to the early 1970s implicated female sex as a risk factor for increased morbidity and mortality.4,5–10 Although these data are controversial, there is evidence that the sex difference in patients undergoing CABG is mainly due to different preoperative risk profiles between men and women.5–7 We compared clinical outcomes after CABG in men and women, considering the different preoperative patient characteristics using propensity score (PS)-based matching.
PATIENTS AND METHODS
Institutional review board approval was obtained from the Heart Institute Lahr, Germany, to perform research analysis on the institution’s database. The study population consisted of 12,606 patients undergoing isolated CABG between 1996 and 2006; 3,079 were female and 9,527 were male. Patient characteristics are given in Table 1
.
|
, irrespective of their significance level. Female and male patients were then matched one-to-one by the nearest propensity score (maximum difference <0.01 and by time of surgery. In a first step, all presented variables excluding height were included in the propensity score 1 (PS1) to differentiate between the effect of disease status and height on outcome after CABG. In a second step, all variables from the PS1 model as well as height were included in the propensity score model 2 (PS2). The specified outcomes of interest were 30-day mortality, intensive care unit (ICU) stay, arrhythmia, and maximum creatine kinase-MB (CK-MB) level. RESULTS
Comparison of the patients’ characteristics showed significant differences between men and women (Table 1). The women were older, of shorter stature, and they had higher incidences of diabetes (OR = 1.5; 95%CI: 1.41–1.68), arterial hypertension (OR = 1.8; 95%CI: 1.62–2.02), hyperlipidemia (OR = 1.2; 95%CI: 1.07–1.28), angina pectoris (OR = 1.2; 95%CI: 1.11–1.32), congestive heart failure (OR = 1.4; 95%CI: 1.19–1.61), carotid stenosis (OR = 1.2; 95%CI: 1.07–1.37), neurological disorders (OR = 1.2; 95%CI: 1.04–1.41), and more advanced New York Heart Association class than men. Men had greater incidences of myocardial infarction (OR for female to male = 0.83; 95%CI: 0.77–0.91), liver disease (OR = 0.84; 95%CI: 0.70–1.002), smoking (OR = 0.39; 95%CI: 0.35–0.43), previous cardiac surgery (OR = 0.7; 95%CI: 0.56–0.88), and more diseased vessels. The ejection fraction, determined by echocardiography, was significant higher in women than men in the original study population (Table 1). After PS1 matching, the preoperative patient characteristics (except height and body surface area) were balanced in the PS1 population of 2,747 pairs, accounting for 89% of all women undergoing CABG in the selected time period (Table 1). In the PS2 population consisting of 1,461 pairs, (47% of female patients) and including height, no significant sex differences were observed (Table 1). There was no significant difference in the percentage of off-pump procedures between men and women (Table 2). IMA was used less frequently in women (original population). However, after adjustment for height, no significant difference was observed (PS2 population). As smaller men are included in the PS2 population, due to matching by height, this may indicate less frequent use of IMA in smaller patients. This effect is supported by Figure 1 where IMA use according to height is shown for men and women. The same effect may be expected for total arterial revascularization: adjusting for height, fewer men received total arterial revascularization. Figure 2 shows that the percentage who received total arterial revascularization also depended on height; men with equivalent height to women received total arterial revascularization less often. Analyzing the main outcome data (Table 3), we found that in the original population, women had higher 30-day mortality (OR = 1.7; 95%CI: 1.23–2.25), a higher rate of arrhythmias (OR = 1.2; 95%CI: 1.06–1.25), and more prolonged ICU stay (OR = 1.3; 95%CI: 1.20–1.43). In addition, the maximum CK-MB in women was higher than in men. In the PS1 population, after adjustment by patient characteristics excluding height, all OR decreased and were not significantly different for 30-day mortality (OR = 1.2; 95%CI: 0.81–1.78) and arrhythmias (OR = 0.96; 95%CI: 0.86–1.06). ICU stay>3 days (OR = 1.2; 95%CI: 1.04–1.3) and maximum CK-MB remained significantly different between males and females. In the PS2 population (adjusted for height), 30-day mortality (OR = 0.93; 95%CI: 0.55–1.57), arrhythmias (OR = 1.2; 95%CI: 0.99–1.34), prolonged ICU stay (OR = 1.1; 95%CI: 0.96–1.31), and maximum CK-MB were not significantly different for men and women after isolated CABG. After adjustment for height, no significantly different values in CK-MB maximum were observed, again indicating a dependency on height, which is supported by Figure 3. The incidence of deep mediastinal wound infection was not significantly different for men and women, either before or after adjustment.
|
|
|
|
|
Women are generally considered to have a higher risk of operative mortality after CABG, but this issue is still controversial. Recent studies show that after adjusting for sex dissimilarities in preoperative risk profiles, women have no increased operative risk for CABG.15–17 In contrast, other investigations found an increased risk for women, despite the application of risk-adjustment methods.8–10 Conflicting results may be caused by differing approaches to the model selection process.11 As a consequence, the final models may differ in the confounding factors included. Applying propensity score methods summarizing baseline characteristics into a single score, a balanced preoperative risk profile between men and women can be achieved (PS2). A necessary requirement is the application of a saturated propensity score containing as many confounding variables as possible. In contrast, the number of adjusting variables in conventional multivariate regression analysis is limited due to overestimation. We want to emphasize that it is essential to present the final models so that the reader knows what adjustments were made relating to the confounders.
In our study of 12,606 patients undergoing CABG, we found no sex differences in operative mortality, postoperative arrhythmias, ICU stay, and maximal CK-MB levels after full adjustment for different preoperative risk profiles between men and women; whereas these differences were prevalent in the unadjusted study cohort. The main effect contributing to a balanced outcome was observed by including height in the propensity score in addition to age, body mass index, and several disease characteristics. We found that quality of surgery, indicated by IMA use and total arterial revascularization, depended on height (Figures 1 and 2). Smaller patients are less likely to receive IMA or total arterial revascularization. Because women are generally smaller than men (12 cm less in our cohort), their lower height may contribute to the adverse outcomes found in women. In our completely matched study cohort (PS2), only 47% of women were included, and these were the tallest women undergoing CABG at our hospital, with a mean height of 165 cm vs. 157 cm in the women left out of the matched cohort.
The most common explanations for adverse outcomes in women are their smaller coronary vessels and less frequent use of IMA grafting.13,14 It is well known that use of at least one IMA reduces operative and long-term mortality.16 Surgery in patients with small coronary vessels is more complex and difficult, requiring higher skills of the surgeon. Moreover, vessels with smaller luminal areas will close up faster by thrombosis, especially near the suture line, causing more myocardial injury. The increasing maximal CK-MB levels with decreasing height may be an indicator of this process. This is in agreement with previous studies reporting an association of female sex with more perioperative myocardial infarction and indirect signs of myocardial injury after coronary revascularization, which may be a consequence of the generally shorter height of women.17,18
Distinctive operative strategies are proposed to manage the issue of small target vessels. The use of arterial grafts (IMA or radial artery) seems to protect against graft failure predominantly in patients presenting with small coronary vessels; a recent randomized study demonstrated a strong effect of radial artery vs. saphenous vein conduits on the patency of grafts on small coronaries and in female patients.19 The physiological basis of this finding may be the preserved endothelial function when grafted. Another operative strategy is the use of sequential grafting; in an angiographic study of 408 patients undergoing off-pump coronary revascularization, the graft patency and stenosis-free rates were better with side-to-side anastomosis than with end-to-side anastomosis, especially in small target vessels (<1.5 mm).20
Our finding that women received fewer arterial grafts despite the potential benefit, may be best explained by their smaller IMA and target vessels and consequently more complex and difficult surgery. Therefore, it is mandatory to train the surgeons in sequential arterial grafting to optimize the results of CABG in women who often present with smaller vessels.
REFERENCES
Asian Cardiovasc Thorac Ann 2009;
17:253-258
© 2009 by SAGE Publications
DOI: 10.1177/0218492309104746
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
