Interact CardioVasc Thorac Surg 2009;8:408-411. doi:10.1510/icvts.2008.189340
© 2009 European Association of Cardio-Thoracic Surgery
Institutional report - Coronary |
Adverse influence of female gender on outcomes after coronary bypass surgery: a propensity matched analysis
Vigneshwar Kasirajan*,
Luke G. Wolfe and
Angel Medina
Division of Cardiothoracic Surgery, Department of Surgery, Virginia Commonwealth University Health System, Richmond, VA, USA
Received 4 August 2008;
received in revised form 9 November 2008;
accepted 13 November 2008
 Presented at Poster Session at Society of Thoracic Surgery Annual Meeting, Fort Lauderdale, Florida, January 28–29, 2008.
*Corresponding author. MCV Campus, West Hospital, 7th Floor, South Wing, 1200 East Broad Street, PO Box 980068, Richmond, VA 23298-0068, USA. Tel.: +1 804 828 2774; fax:+1 804 827 0527.
E-mail address: vkasirajan{at}mcvh-vcu.edu (V. Kasirajan).
 |
Abstract
|
|---|
Female patients have been shown to have a different risk profile and increased adverse events after coronary bypass grafting (CAB). The objective was to create a risk profile based on gender and look at outcomes in propensity matched groups with similar factors differing by gender. From May 2001 to December 2006, 976 patients underwent isolated CAB. Univariable analysis created a risk profile for female and male patients. Multivariable logistic regression was used to develop independent predictors of mortality. Propensity matching for the most predictive variables of adverse was used to create matched sets of 269 male and female patients to analyze outcomes independent of these variables. Of the 976 patients 31.7% were female. Of the 19 preoperative risk factors analyzed, diabetes, hypertension, older age, higher body mass index (BMI), African-American race were more predictive of female gender. Males had more smokers and acute myocardial infarction within seven days before surgery. Independent predictors of mortality were female gender (P=0.01), diabetes (P=0.02), increased age (P=0.02), acute MI <7 days (P=0.003). Propensity matching (for smoking, diabetes, hypertension, MI <7 days, age, BMI and race) still showed increased in-hospital complications and mortality for female patients. Female patients undergoing CAB have a different risk profile and have a higher incidence of adverse outcomes including death, which are not mitigated by careful matching with male patients.
Key Words: Coronary bypass surgery; Gender; Propensity matching
|
1. Background
|
|---|
Multiple studies have shown that female patients have an increased incidence of complications and mortality after coronary artery bypass (CAB) surgery [1–3]. However, after matching for a similar population, it is believed that most of these outcomes are equivalent [1]. The intent of this study is to examine outcomes by gender after CAB and create matched population for the most significant risk factors. This would allow us to examine outcomes then impacted by gender.
|
2. Methods
|
|---|
Using the STS database, 976 patients undergoing CAB from May 2001 to December 2006 at a single institution were identified. Univariable analysis (t-test and Fisher's Exact Test) was used to identify risk factors between male and female patients. Multivariable logistic regression was then used to further develop independent predictors for mortality. The most predictive variables between male and female patients were then used to create matched sets of patients to look at outcomes. PROC LOGISTIC was used to perform multivariate logistic regression which calculated the propensity score for each observation that was used in the matching process. All statistical analysis was performed using SAS.
|
3. Results
|
|---|
Nine hundred and seventy-six patients at a single urban tertiary care medical center were studied. All these patients had isolated coronary bypass surgery. Patients having concomitant procedures such as valve surgery were excluded.
Three hundred and nine (31.66%) of patients were female. Females were older, had a higher body mass index, a higher incidence of diabetes and hypertension. African-Americans accounted for 55% of females compared to 36% of males (Table 1). The mean BMI of female patients (31.0) makes this group obese (the normal BMI being 17 to 25).
Male patients had a higher proportion of smokers and those with acute myocardial infarctions within seven days of surgery (preoperative). Variables not different between the groups were family history of coronary artery disease, dyslipidemia, renal failure, history of cerebrovascular events, peripheral vascular disease, chronic lung disease and use of immunosuppressive agents (including steroids) preoperatively (Table 2). The overall incidence of renal failure preoperatively (creatinine >2.0) was 15.5% in women and 13.3% in men.
By stepwise logistic regression on mortality using all data, female gender (P=0.0107, OR 2.691, 95% CI 1.26–5.75), Diabetes (P=0.015, OR 2.67, 95% CI 1.2–5.89), increasing age (P=0.016, OR 1.04, 95% CI 1.01–1.08), MI within seven days (P=0.002, OR 3.26, 95% CI 1.52–7.02) were independent predictors of mortality post CAB.
Taking the most statistically significant variables that characterized each gender (smoking, diabetes, hypertension, MI <7 days, age, BMI and Race), 269 male and female patient matched groups were created (Table 3). All in-hospital complications, prolonged ventilation (over 24 h), postoperative renal failure and overall mortality was higher in female patients (Fig. 1). A much higher percentage of female patients in the matched groups also had blood transfusions (Fig. 2). However, the transfusion data were available only in 423 of the 538 patients.
View larger version (20K):
[in this window]
[in a new window]
|
Fig. 1. Female patients have significantly more complications compared to males even after propensity matching.
|
|
The use of arterial conduits was not different between female and male patients (86.57% vs. 88.06%). However, overall female patients had less number of distal anastomosis (arterial and venous) than male patients (2.6 vs. 2.8, P=0.0048). But in the matched set, there was no difference between male and females in the number of distal anastomosis either for arterial or venous conduits (Fig. 3).
Looking at the actual patients who died in the propensity matched groups; there were 15 female and 5 males. All these patients had urgent or emergent surgery due to worsening symptoms or acute MI. There were some important differences, men were older than women (69.6 years, range 42–81 vs. 64.1 years, range 62–84), and men had lower ejection fractions, but overall had better controlled diabetes, hypertension and no active smokers.
The women though younger had more active smokers, poorly controlled diabetes, hypertension, end stage renal failure requiring dialysis and peripheral vascular disease.
|
4. Discussion
|
|---|
This study has a number of important differences from other reported experience in the literature. The overall proportion of women is higher. The number of African-American women is high, constituting a majority of women (55%). Women tended to a higher BMI (31), indicating that most were obese. The overall incidence of renal failure was high. As reported in other studies, the incidence of hypertension and diabetes was higher in women, who also tended to be older. This is reflected in a greater proportion of women being medicare beneficiaries. However, there were more men who also had commercial insurance (Table 1; P<0.0001). Whether these reflect socioeconomic discrepancies are not known from this study. More men were smokers. The incidence of acute myocardial infarction within seven days before surgery was higher in men, who then required more urgent surgery [4, 9].
In the propensity matched groups, the effects of female gender could not be eliminated in increased risk of complications and death.
We then looked at two other variables in the matched groups that may have a bearing on outcomes. First, we looked at the total number of distal anastomosis by arterial and venous conduits. There was no difference in the matched cohorts, but there was a difference in the overall groups indicating perhaps this may be due to smaller numbers (Type II error). The quality of venous conduits between the groups could not be ascertained, but we could not show any increase in acute MI post-op indicating early graft closure. Anecdotally, we believe that older women have thinner and often varicose veins. Unlike others, we had a similar use of LIMA to the LAD in these groups [5, 6].
The second variable of transfusion during or after surgery showed an increased use of blood products in women. The exact reasons for this are unclear, but women have been shown to have a lower circulating blood volume. The hemoglobin levels preoperatively and triggers for transfusion are not known in this study.
Clearly, propensity matching could not account for the differences. This could simply be due to the fact that retrospective matching is not perfect. However, other factors such as effectiveness of antihypertensive therapy, control of diabetes could not be ascertained. Historically, women tend to be undertreated for these conditions. The quality of venous conduits will need to be studied prospectively. Also the extent of distal CAD is not known. The role of transfusions and resulting inflammatory responses in contributing to outcomes is not known. Though there is evidence to suggest that transfusion increases risk of postoperative low output states and infection [10–13].
The use of postoperative medications to control hyperglycemia, for antiplatelet therapy, beta blocker, effective postoperative antihypertensive and antilipid medication use were not studied. However, the effects of these medications are most important for long-term outcomes than immediate postoperative events.
In spite of this and many other papers showing that women tend to have higher risk of complications after surgery, no prospective studies comparing men vs. women in the CABG population exist. Prospective follow-up of a sufficiently large cohort of patients with a significant number of women (unlike past studies, who have predominantly men) with data on effectiveness of preoperative medical therapy, control of diabetes, HTN, anemia are needed [8]. The data on the role of transfusions in affecting outcomes after CAB are growing. Sometimes, the data seem to be conflicting showing worse outcomes with anemia and some showing transfusions increasing risk of complications. Clearly with women having lower blood volumes, the risk of transfusion is high. Whether this can be reduced by increasing thresholds for transfusion (example, do not transfuse unless hemoglobin is <6.5) without affecting outcomes for the worse is not known. The is also evidence from retrospective case series that off-pump CAB may improve outcomes in women, which may simply be due to perhaps less blood dilution and transfusions in this population [7]. More data on conduit selection are needed. We have found that contrary to popular belief, women tend to have excellent IMAs and radial arteries, where as veins tend to be inferior in quality compared to men. Long-term studies looking at effective postoperative medical management, the role of depression, participation in cardiac rehab are all needed to further optimize long-term survival in women.
The ultimate aim of studies such as this and others is not to deny surgical therapy for women with CAD, but to emphasize that after 40 years of surgical therapy for CAD, we have not really understood why these differences exist and how to eliminate them.
|
References
|
|---|
- Koch CG, Khandwala F, Nussmeier N, Blackstone EH. Gender profiling in coronary bypass grafting. J Thoracic Cardiovasc Surg 2003;126:2044–2051.[Abstract/Free Full Text]
- Liao Y, Cooper RS, Mensah GA, McGee DL. Left ventricular hypertrophy has a greater impact on survival in women than in men. Circulation 1995;92:805–810.[Abstract/Free Full Text]
- Bolooki H, Green R, Vargas A, Kaiser GA, Ghahramani A. Coronary artery surgery in women. Cardiovascular Disease, Bulletin of the Texas Heart Institute 1974;1:215–222.
- Christakis GT, Weisel RD, Buth KJ, Panagiotopoulos KP, Ivanov J, Goldman B, David TE. Is body size the cause for poor outcomes of coronary bypass operations in women? J Thoracic Cardiovasc Surg 1995;110:1344–1358.[Abstract/Free Full Text]
- Kurlansky PA, Traad EA, Galbut DL, Singer S, Zucker M, Ebra G. Coronary artery bypass surgery in women: a long-term comparative study of quality of life following bilateral internal mammary artery grafting in men and women. Ann Thorac Surg 2002;74:1517–1525.[Abstract/Free Full Text]
- Rahimtoola SH. Editorial. Bilateral internal mammary artery grafting for coronary artery bypass grafting: why men vs. women? Ann Thorac Surg 2002;74:1435–1437.[Free Full Text]
- Mack MJ, Brown P, Houser F, Katz M, Kugelmass A, Simon A, Battaglia S, Tarkington L, Culler S, Becker E. On pump vs. off pump coronary artery bypass surgery in a matched sample of women: a comparison of outcomes. Circulation 2004;110(Suppl_II):II-1–II-6.[Medline]
- Mosca L, Ferris A, Fabunmi R, Robertson RM. Tracking women's awareness of heart disease: an American Heart Association National Study. Circulation 2004;109:573–579.[Abstract/Free Full Text]
- Vaccarino V, Abramson JL, Veledar E, Weintraub WS. Sex differences in hospital after coronary artery bypass surgery. Evidence for higher mortality in younger women. Circulation 2002;105:1176–1181.[Abstract/Free Full Text]
- Habib RH, Zacharias A, Schwann TA, Riordan CJ, Durham SJ, Shah A. Adverse effects of low hematocrit during cardiopulmonary bypass in the adult: should current practice be changed? J Thorac Cardiovasc Surg 2003;125:1438–1450.[Abstract/Free Full Text]
- Ranucci M, Biagiollli B, Scolletta S, Grillone G, Cazzaniga A, Cattabriga I, Isgro G, Giomarelli P. Lowest hematocrit on cardiopulmonary bypass impairs the outcome in coronary surgery. Tex Heart Inst J 2006;33:300–305.[Medline]
- Surgenor SD, DeFoe GR, Fillinger MP, Likosky DS, Groom RC, Clark C, Helm RE, Kramer RS, Leavitt BJ, Klemperer JD, Krumholz CF, Westbrook BM, Galatis DJ, Frumiento C, Ross CS, Olmstead EM, O'Connor GT. Intraoperative red blood cell transfusion during coronary artery bypass graft surgery increases the risk of postoperative low output heart failure. Circulation 2006;114(Suppl I):I-43–I-48.[Medline]
- Rogers MAM, Blumberg N, Heal JM, Hicks GL. Increased risk of infection and mortality in women after cardiac surgery related to allogenic blood transfusion. J Women's Health 2007;16:1412–1420.[CrossRef]
|
|
Top
Abstract
1. Background
