Abstract
Objectives
Coronary artery bypass grafting shows impaired outcomes in women compared to men. Investigation of sex-related outcome differences mostly comprises different operative strategies, potentially hampering interpretation of data. We herein aimed to investigate sex-related outcome differences in on-pump procedures only.
Methods
All consecutive patients undergoing isolated on-pump bypass grafting between 2012 and 2021 at our centre were included in this study. Propensity score-matching was applied to compare outcomes of men and women. Kaplan-Meier analysis was used for investigating differences in survival probabilities. Identification of risk factors for long-term mortality was conducted by multivariable Cox regression.
Results
In the matched cohort (326 pairs), women presented with a higher EuroSCORE II (European System for Cardiac Operative Risk Evaluation II) and less prior surgery. Distribution of arterial, venous, and radial bypass grafts was similar between sexes. Women presented more often with superficial wound healing disorders (P = .004), a higher amount of red blood cell transfusions (P < .001), and a longer hospital stay. Mortality at 30 days, 1 year, and 10 years revealed no difference between sexes (10-year mortality: men 27.7% vs women 22.6%; P = .228). Female sex was not significantly associated with long-term mortality.
Conclusions
On-pump CABG in women is not associated with a higher prevalence of early or long-term all-cause mortality. Female sex is not a risk factor for long-term mortality. Women need more red blood cell transfusions, present with more superficial wound-healing disorders, and have a longer hospital stay. These findings underscore the overall safety of on-pump CABG in women.
Keywords: coronary artery bypass grafting, myocardial infarction, cardiopulmonary bypass
According to recent guidelines for myocardial revascularization, coronary artery bypass grafting (CABG) is recommended for patients with complex coronary artery disease (CAD), particularly those with intermediate to high SYNTAX scores or type 2 diabetes.
Graphical abstract
INTRODUCTION
According to recent guidelines for myocardial revascularization, coronary artery bypass grafting (CABG) is recommended for patients with complex coronary artery disease (CAD), particularly those with intermediate to high SYNTAX scores or type 2 diabetes.1,2 Despite being the most frequently performed cardiac surgery worldwide, only 20%-30% of CABG patients are women.3,4 Multiple studies report worse outcomes in women compared to men, including higher rates of early mortality, stroke, and postoperative myocardial infarction.5,6 Contributing factors may include delayed CAD diagnosis due to atypical symptoms,7 higher comorbidity burden and urgency at presentation,8 a greater prevalence of non-obstructive CAD,9 smaller coronary arteries prone to spasm,10 and less frequent complete revascularization with greater use of unfavourable grafts.11,12
Most previous studies included mixed surgical strategies (on-pump, off-pump, and beating heart), potentially hampering data interpretation. We therefore specifically investigate on-pump CABG to assess sex-related outcome differences and to identify risk factors for long-term mortality in men and women.
METHODS
Ethical statement
Due to the retrospective and anonymized nature of the study and in accordance with German law (§12 HmbKHG), patient inclusion in Hamburg (Germany) did not require ethics committee approval, and informed patient consent was waived. Patient data were collected, stored, and processed in accordance with institutional data protection standards and consistent with the requirements of the WMA Declaration of Taipei. The analysis was conducted according to the Declaration of Helsinki.
Patients and definitions
All consecutive patients undergoing isolated on-pump CABG at our centre between 2012 and 2021 were included. Exclusion criteria were off-pump/beating-heart CABG, concomitant procedures including valve and aortic surgery, and patients with post-infarction left ventricular aneurysm.
For definitions of complete revascularization, deep sternal wound infection (DSWI), superficial wound-healing disorders (WHD), perioperative myocardial infarction, and urgent procedures, see Table S1.
Surgical strategy and perioperative management
All patients were discussed in a multidisciplinary heart team. Indication for surgery followed current guideline recommendations, coronary anatomy, and patient-specific characteristics. Surgical procedures followed standardized institutional protocols. Bilateral internal mammary artery (BIMA) grafting with a predominantly arterial revascularization strategy was preferred whenever feasible, with IMA harvesting in a skeletonized technique. Contraindications to BIMA included clinical conditions or haemodynamic instability. Myocardial protection and cardiopulmonary bypass management followed uniform institutional standards and were not sex specific. The choice between on-pump and off-pump CABG was based on anatomy, patient factors, and surgeon expertise; sex was not a decision criterion. Incomplete revascularization was mainly related to chronic total occlusions or small target vessel diameter. Perioperative medical therapy followed contemporary guideline-based recommendations without sex-specific differences.
Statistical analysis
Primary end-points were short- and long-term all-cause mortality; secondary end-points included a composite of myocardial infarction, cerebrovascular events, and renal failure. Mortality data were obtained from national registries, with follow-up until June 2025.
All variables included in the predefined analyses were complete, as they are mandatory fields within the institutional database; therefore, no missing data were present.
Propensity-score matching was performed using 1:1 nearest neighbour matching (calliper width 0.1). Matching parameters were ‘urgency’, ‘three-vessel CAD’, ‘STEMI’, ‘age’, ‘EuroSCORE II’, ‘creatinine’, ‘NYHA ≥III’, and ‘extracardiac arteriopathy’. Matching parameters and standardized mean differences before and after matching are shown in Table S2 and Figure S1, with propensity score overlap illustrated in Figure S2.
Continuous variables were tested for normality (Shapiro-Wilk test), reported as medians with IQR, and compared using the Mann-Whitney U test. Categorical variables were reported as counts and percentages and compared using the χ2 test.
Median follow-up time for mortality was estimated using the reverse Kaplan-Meier method. Cumulative incidence of the composite end-point was assessed using Kaplan-Meier method and compared using the log-rank test. Restricted mean survival time (RMST) for all-cause mortality was calculated at 30 days, 1 year, and 10 years; expressed in days with 95% CIs and compared using Wald’s test.
Risk factors for 10-year all-cause mortality were identified using multivariable Cox regression, including clinically relevant covariates.
To assess age-dependent differences in the association between established risk factors and long-term mortality, two complementary approaches were applied: age-stratified Cox regression analyses (cutoff 65 years) and formal testing of age-group interaction terms within the overall Cox model. Covariates included BIMA use, insulin-dependent diabetes mellitus, body mass index, ST-elevation myocardial infarction, non–ST-elevation myocardial infarction, left ventricular ejection fraction (LVEF), and hypercholesterolemia.
A P-value of <.05 was considered statistically significant. All analyses were performed with R statistical software version 4.0.3 (R Foundation for Statistical Computing, Vienna, Austria).
Preliminary results of this study were presented as a conference abstract at the 55th Annual Meeting of the German Society for Thoracic and Cardiovascular Surgery (DGTHG).13
RESULTS
Baseline demographics
From 2012 to 2021, 2545 patients underwent on-pump CABG at our centre. Of these, 2149 (84.4%) patients were men and 396 (15.6%) were women. A total of 326 matched pairs were analysed after propensity-score matching.
In the unmatched cohort, women showed higher age (66.0 [IQR 59.0-72.0] vs 69.0 [IQR 64.0-75.0] years, P < .001) and a higher comorbidity and symptom burden according to NYHA class ≥III (40.4% vs 51.0%, P < .001). Additionally, women presented with lower baseline creatinine levels (1.00 [IQR 0.88-1.15] vs 0.84 [IQR 0.70-1.00] mg/dL, P < .001) and a lower amount of 3-vessel CAD (79.1% vs 71.7%, P = .002).
In the matched cohort, median age in both sexes was similar with 70.0 years (IQR 63.0-75.0) in men and 69.0 years (IQR 62.0-74.0) in women. Women had a lower prevalence of prior cardiac surgery (3.4% vs 0.0%; P = .002) and showed a higher EuroSCORE II (European System for Cardiac Operative Risk Evaluation II) (1.30% [IQR 0.94-2.02] vs 1.58% [IQR 1.15-2.40])). Other baseline characteristics of the matched cohort did not show significant differences between sexes.
Detailed patient demographics are summarized in Table 1 for the matched cohort and Table S3 for the unmatched cohort.
Table 1.
Baseline Characteristics
| Men matched (n = 326) | Women matched (n = 326) | P-value | |
|---|---|---|---|
| Age (years) | 70.0 (63.0-75.0) | 69.0 (62.0-74.0) | .126 |
| BMI (kg/m²) | 27.2 (24.8-30.1) | 27.9 (24.8-31.2) | .143 |
| Prior cardiac surgery | 11 (3.4) | 0 (0.0) | .002 |
| Prior PCI | 73 (22.4) | 62 (19.0) | .334 |
| Myocardial infarctiona | |||
| STEMI | 13 (4.0) | 16 (4.9) | .704 |
| NSTEMI | 87 (26.7) | 67 (20.6) | .080 |
| Hypertension | 270 (82.8) | 281 (86.2) | .279 |
| COPD | 24 (7.4) | 18 (5.5) | .425 |
| Prior stroke | 30 (9.2) | 26 (8.0) | .675 |
| Extracardiac arteriopathy | 65 (19.9) | 60 (18.4) | .691 |
| LVEF ≤ 35% | 23 (7.1) | 12 (3.7) | .082 |
| IDDM | 18 (5.5) | 20 (6.1) | .875 |
| Hypercholesterolemia | 204 (62.6) | 227 (69.6) | .077 |
| Smoking | |||
| Active | 60 (18.4) | 69 (21.2) | .466 |
| Former | 65 (19.9) | 45 (13.8) | .041 |
| Dialysis | 2 (0.6) | 0 (0.0) | .479 |
| Creatinine (mg/dL) | 0.89 (0.79-0.99) | 0.87 (0.74-1.00) | .663 |
| Haemoglobin (g/dL) | 14.3 (13.4-15.1) | 13.2 (12.2-14.1) | <.001 |
| Three-vessel CAD | 231 (70.9) | 235 (72.1) | .795 |
| EuroSCORE II (%) | 1.30 (0.94-2.02) | 1.58 (1.15-2.40) | <.001 |
| NYHA ≥III | 156 (47.9) | 154 (47.2) | .937 |
Continuous variables are shown as median (IQR), categorical variables are shown as n (%).
Abbreviations: BMI, body mass index; CABG, coronary artery bypass grafting; CAD, coronary artery disease; COPD, chronic obstructive pulmonary disease; EuroSCORE II, European System for Cardiac Operative Risk Evaluation II; IDDM insulin-dependent diabetes mellitus; LVEF, left ventricular ejection fraction; NSTEMI, non–ST-elevation myocardial infarction; NYHA, New York Heart Association; PCI, percutaneous coronary intervention; STEMI, ST-elevation myocardial infarction.
During 30 days prior to CABG.
Periprocedural data
In the unmatched cohort, the number of performed distal bypass anastomoses was lower in women. Accordingly, procedure time, extracorporeal circulation time (ECC) and aortic cross-clamp time (ACC) were shorter in women. Utilization of BIMA was less frequently conducted in women (60.3% vs 49.5%, P < .001), whereas a combination of 1 IMA and saphenous vein grafts (SVG) or SVG alone was more often applied in women. Periprocedurally, women had a higher need for prolonged catecholamine therapy (2.3% vs 5.1%, P = .003) and mechanical circulatory support (MCS) (1.7% vs 4.0%, P = .004). Moreover, in women, a higher number of administered red blood cell (RBC) units was documented (0.0 [0.0] vs 0.0 [IQR 0.0-2.0], P < .001).
In the matched cohort, procedure time, ECC and ACC were similar in both sexes. Conducted bypass strategy, grafts, number of performed distal bypass anastomoses, and rate of complete revascularization were similar between sexes. Periprocedural prolonged catecholamine therapy or MCS use, and prolonged ventilation time were without significant differences. In women, a higher number of administered RBC was documented (0.0 [IQR 0.0-1.0] vs 0.0 [IQR 0.0-2.0], P < .001).
Detailed periprocedural data are summarized in Table 2 for the matched cohort and Table S4 for the unmatched cohort.
Table 2.
Periprocedural Data
| Men matched (n = 326) | Women matched (n = 326) | P-value | |
|---|---|---|---|
| Urgency | 34 (10.4) | 33 (10.1) | 1.000 |
| Procedure time (min) | 270.0 (225.0-310.0) | 255.0 (215.0-307.0) | .087 |
| ECC (min) | 104.0 (84.0-127.0) | 100.0 (83.0-124.0) | .299 |
| ACC (min) | 71.0 (56.0-89.0) | 68.0 (55.0-84.0) | .154 |
| Number of bypasses (n) | 3.0 (2.0-3.0) | 3.0 (2.0-3.0) | .475 |
| Bypass graft strategy | |||
| Composite | 133 (40.8) | 135 (41.4) | .937 |
| Sequential | 186 (57.1) | 175 (53.7) | .431 |
| CX territory | 97 (29.8) | 103 (31.6) | .671 |
| RCA territory | 164 (50.3) | 177 (54.3) | .347 |
| Bypass graft | |||
| BIMA only | 168 (51.5) | 173 (53.1) | .754 |
| BIMA + SVG | 33 (10.1) | 21 (6.4) | .118 |
| BIMA + radialis | 4 (1.2) | 1 (0.3) | .369 |
| IMA only | 6 (1.8) | 7 (2.1) | 1.000 |
| IMA + SVG | 111 (34.0) | 121 (37.1) | .462 |
| IMA + radialis | 3 (0.9) | 0 (0.0) | .247 |
| SVG | 1 (0.3) | 3 (0.9) | .616 |
| Complete revascularization | 246 (75.2) | 264 (81.0) | .136 |
| Catecholamines ≥24 h | 7 (2.1) | 16 (4.9) | .089 |
| Ventilation time ≥24 h | 12 (3.7) | 9 (2.8) | .657 |
| Mechanical circulatory support | 9 (2.8) | 12 (3.7) | .800 |
| Numbers of RBC | 0.0 (0.0-1.0) | 0.0 (0.0-2.0) | <.001 |
Continuous variables are shown as median (IQR), categorical variables are shown as n (%).
Abbreviations: ACC, aortic cross-clamp time; BIMA, bilateral internal mammary artery; CABG, coronary artery bypass grafting; CX, circumflex artery; ECC, extracorporeal circulation time; IMA, single internal mammary artery; RBC, red blood cells; RCA, right coronary artery; SVG, saphenous vein graft.
Thirty-day outcome after on-pump CABG
The unmatched cohort showed no difference in secondary end-points during 30 days after CABG. Women showed a higher rate of surgical revision and longer intensive care unit stay (2.0 [IQR 1.0-3.0] vs 2.0 [IQR 2.0-4.0] days, P = .010) and hospital stay (7.0 [6.0-8.0] vs 7.0 [6.0-9.0] days, P = .001). Rates of WHD (3.0% vs 10.1%, P < .001) and DSWI (1.6% vs 3.3%, P = .035) were significantly higher in women.
In the matched cohort, women showed more WHD (4.3% vs 10.4%; P = .004) but not DSWI. Moreover, hospital stay was prolonged in women (7.0 [IQR 6.0-8.0], mean 8.175 days vs 7.0 [IQR 6.0-8.0], mean 8.377 days; P = .043).
Detailed 30-day outcome parameters are summarized in Table 3 for the matched cohort and Table S5 for the unmatched cohort.
Table 3.
Thirty-Day Outcome Parameters of Men and Women Undergoing On-Pump CABG
| Men matched (n = 326) | Women matched (n = 326) | P-value | |
|---|---|---|---|
| WHDa | 14 (4.3) | 34 (10.4) | .004 |
| DSWI | 9 (2.8) | 12 (3.5) | .657 |
| Surgical revision | 13 (4.0) | 14 (4.3) | 1.000 |
| Coronary angiography | 4 (1.2) | 5 (1.5) | .547 |
| ICU stay (days) | 2.0 (1.0-3.0) | 2.0 (1.0-3.0) | .517 |
| Hospital stay (days) | 7.0 (6.0-8.0) (mean 8.175) | 7.0 (6.0-8.0) (mean 8.377) | .043 |
| Myocardial infarction | 2 (0.6) | 3 (0.9) | 1.000 |
| Disabling stroke | 5 (1.5) | 6 (1.8) | 1.000 |
| Renal failure (KDIGO ≥ III) | 3 (0.9) | 2 (0.6) | 1.000 |
Continuous variables are shown as median (IQR), categorical variables are shown as n (%).
Abbreviations: CABG, coronary artery bypass grafting; DSWI, deep sternal wound infection; ICU, intensive care unit; KDIGO, Kidney Disease: Improving Global Outcomes; WHD, wound-healing disorders.
Including superficial wound-healing disorders.
Cumulative incidence of the composite end-point during the early period after CABG was similar in both sexes (P = .919). For the cumulative incidence curve of the composite end-point, see Figure 1A. The rate of 30-day mortality was similar (men 1.2% [95% CI, 0.0-2.4] vs women 1.8% [95% CI, 0.4-3.3]; P = .317). RMST for end-point all-cause mortality at 30 days was 29.8 (95% CI, 29.6-30.0) days in men and 29.6 (95% CI, 29.3-29.9) days in women (P = .369). For Kaplan-Meier survival curve of 30-day mortality stratified by sex, see Figure 1B.
Figure 1.
Sex-specific short-term outcomes after on-pump coronary artery bypass grafting (CABG). (A) Sex-specific cumulative incidence of the composite endpoint after on-pump CABG. Cumulative incidence curves showing no significant difference in rate of composite endpoint including myocardial infarction, disabling stroke, and renal failure (KDIGO ≥III) (p=0.919) after on-pump CABG between sexes. (B) 30-day mortality stratified by sex after on-pump CABG. Short-term mortality was similar in both sexes (1.2% [95% CI, 0.0-2.4] vs 1.8% [95% CI, 0.4-3.3]; P = .317). RMST for endpoint all-cause mortality at 30 days was 29.8 (95% CI, 29.6-30.0) days in men and 29.6 (95% CI, 29.3-29.9) days in women (P = .369)
Long-term mortality after on-pump CABG
Median follow-up time of matched patients was 8.5 years (men 8.6 [95% CI, 8.1-9.4] years vs women 8.5 [95% CI, 7.9-9.2] years).
Rate of mortality was similar in both sexes in a 1-year follow-up (men 3.2% [95% CI, 1.2-5.1] vs women 2.8% [95% CI, 1.0-4.6]; P = .428). RMST for end-point all-cause mortality at 1 year was 356.2 (95% CI, 350.6-361.7) days in men and 357.0 (95% CI, 351.5-362.5) days in women (P = .842).
Rate of mortality was similar in both sexes in a 10-year follow-up (men 27.7% [95% CI, 21.5-33.4) vs women 22.6% [95% CI, 16.8-28.0]; P = .228). RMST for end-point all-cause mortality at 10 years was 3153.7 (95% CI, 3042.4-3265.0) days in men and 3259.9 (95% CI, 3159.3-3360.6) days in women (P = .166).
For Kaplan-Meier survival curve of 1-year mortality and 10-year mortality stratified by sex see Figure 2A and B.
Figure 2.
Sex-specific mid- and long-term mortality after on-pump CABG. (A) 1-year mortality stratified by sex after on-pump CABG. Rate of mortality was similar in both sexes in a 1-year follow-up (3.2% [95% CI, 1.2-5.1] vs 2.8% [95% CI, 1.0-4.6]; P = 0.428). RMST for endpoint all-cause mortality at 1 year was 356.2 (95% CI, 350.6-361.7) days in men and 357.0 (95% CI, 351.5-362.5) days in women (P = .842). (B) 10-year mortality stratified by sex after on-pump CABG. Long-term mortality was similar in both sexes (27.7% [95% CI, 21.5-33.4] vs. 22.6% [95% CI, 16.8-28.0]; P = .228). RMST for endpoint all-cause mortality at 10 years was 3153.7 (95% CI, 3042.4-3265.0) days in men and 3259.9 (95% CI, 3159.3-3360.6) days in women (P = .166)
Risk factors for long-term mortality after on-pump CABG
For the entire patient cohort, age (hazard ratio [HR], 1.045; 95% CI, 1.022-1.069; P < .001) and LVEF ≤35% (HR, 3.035; 95% CI, 1.725-5.340; P < .001) were associated with long-term mortality after on-pump CABG. Sex was not significantly associated with all-cause mortality.
Sex-specific analyses of risk factors for long-term mortality after on-pump CABG identified age (HR, 1.076; 95% CI, 1.035-1.119; P < .001) in women and LVEF ≤35% (HR, 3.693; 95% CI 1.815—7.513; P < .001) in men.
Detailed results of multivariable Cox regression analysis of the matched cohort are shown in Figure 3.
Figure 3.
Risk Factors for Long-Term Mortality After On-Pump CABG. In multivariable Cox regression risk factors for long-term mortality were identified. For the entire patient cohort age (HR, 1.045; 95% CI, 1.022-1.069; P < .001) and LVEF ≤ 35% (HR, 3.035; 95% CI, 1.725-5.340; P < .001) were significantly associated with long-term mortality after on-pump CABG
Age-stratified Cox regression analyses (<65 vs ≥65 years) identified preoperative LVEF as the only consistent independent predictor of long-term mortality in both subgroups. No other covariate reached statistical significance in either group (Tables S6 and S7). Formal testing of interaction terms between each covariate and age group confirmed that none of the investigated variables showed a significantly different association with long-term mortality across age groups (Table S8).
DISCUSSION
Main findings of the herein-conducted study on sex-related differences in on-pump CABG are:
in the unmatched cohort, women presented with higher comorbidity, greater symptom burden, and higher surgical risk;
in the unmatched cohort, women received fewer bypass grafts, less often BIMA grafting, and more frequently required prolonged catecholamine therapy and RBC administration;
in the matched cohort, graft strategy and early outcomes, including myocardial infarction, disabling stroke, and renal failure, were similar between sexes;
women required more RBC transfusions, showed more WHD, and had longer hospital stays;
mortality at 30 days, 1 year, and 10 years after CABG did not differ between matched groups; and
female sex was not independently associated with all-cause mortality.
The herein-seen conspicuousness of women in the unmatched cohort undergoing CABG at a higher age, with a higher comorbidity burden and more frequent significant symptoms, is in accordance with other studies investigating outcome differences between men and women in myocardial revascularization. Gaudino and colleagues showed in a large meta-analysis involving 13,193 patients that women are older at the time of presentation to CABG, present more often with significant comorbidities including diabetes and previous stroke and are more likely to suffer from NYHA functional class III or IV.5 The higher prevalence of NYHA class ≥III in women likely reflects a pattern of later and more symptomatic presentation, potentially related to sex-specific pathophysiology including microvascular dysfunction and a higher prevalence of heart failure with preserved ejection fraction. The same applies to patients undergoing percutaneous coronary intervention, as seen in previous studies,14 suggesting an overall delayed referral of women to surgical or percutaneous revascularization in the instance of CAD. This assumption is endorsed by a work of Fink and colleagues in which male sex was associated with a higher likelihood of referral to CABG (OR: 2.27; P < .001).7 In the herein-presented work, after propensity score matching, these differences were no longer detectable, suggesting that the NYHA imbalance was largely captured by the overall comorbidity burden and that recognition of (atypical) CAD symptoms in women and a timely referral may be of paramount importance to improve outcomes in women after CABG.
Furthermore, women in the unmatched cohort received less BIMA graftings and, overall, less bypass grafts when undergoing on-pump CABG in this study. Attia and colleagues described this sex disparity in 57 943 patients undergoing CABG from 1972 to 2011 with less completeness of revascularization, fewer BIMA grafting and overall fewer arterial grafts in women.15 In the matched patients, no differences in bypass grafts or completeness of revascularization between men and women were found. Considering short- (10%-25%) and long-term failure rates (40%-50%) of SVG,16 a BIMA-first approach seems to be reasonable in women undergoing CABG to improve outcomes, especially with regard to smaller coronary artery anatomy, although it may still not be clear which patient subgroups benefit most from multiple arterial grafting.17
A recent analysis of 3229 patients undergoing CABG with either single IMA or BIMA grafts found male sex to be a protective factor against superficial WHD,18 consistent with our findings of a higher rate of superficial WHD in women. Notably, no differences were observed regarding DSWI, which is known to significantly affect long-term outcome and mortality. The higher rate of superficial WHD in women is likely explained by the well-documented convergence of female sex, elevated BMI, and diabetes as independent risk factors for sternal wound infections after CABG.19 The higher prevalence of both risk factors in our female cohort suggests a synergistic effect, highlighting optimized perioperative glycaemic control and weight management as modifiable targets.
Several multicentre studies have demonstrated that women undergoing CABG are more likely to receive red blood cell transfusions. Female sex emerged as an independent risk factor for transfusion, largely attributed to lower preoperative haematocrit levels and smaller circulating blood volume.20–22 This is reflected in our cohort by significantly lower preoperative haemoglobin levels in women compared to men (13.2 vs 14.3 g/dL, P < .001), underscoring the need for targeted preoperative blood optimization strategies in female patients before CABG.
The mean hospital stay was slightly longer in women compared to men, with a statistically significant difference. Despite this, the clinical relevance of the difference may be limited due to the small absolute difference in means and overlapping standard deviations.
EuroSCORE II differed significantly between men and women, with both sexes remaining within a low predicted risk range. Importantly, this difference did not translate into a disparity in long-term all-cause mortality. Despite its inclusion as a matching variable, a difference in EuroSCORE II persisted after matching, likely reflecting an intrinsic sex-specific risk profile rather than a methodological limitation (SMD after matching 0.054).
CABG as a redo procedure is associated with higher long-term but not short-term or 1-year mortality.23 Although men more often had prior cardiac surgery, this subgroup comprised only 3.4% of men and did not translate into a significant difference in long-term survival. This is consistent with evidence that the male sex independently favours surgical over percutaneous repeat revascularization, reflecting a sex-specific difference in long-term coronary revascularization strategies.24
While most studies investigating outcome differences between men and women undergoing CABG describe higher incidences of major adverse cardiovascular events in women including stroke and myocardial infarction, higher mortality rates in women are also reported.5 This difference may be driven by the herein-performed inclusion of on-pump CABG-only patients since a large meta-analysis showed that off-pump CABG may be associated with higher rates of mid- and long-term mortality and higher rates of repeat revascularization.25 Furthermore, the absolute number of composite end-point events in this study may limit the statistical power to detect differences. In the herein-conducted Cox-regression, female sex was not significantly associated with long-term mortality.
In the matched cohort, women were of similar age, exhibited comparable comorbidities, and underwent the same grafting strategies as their male counterparts, resulting in equivalent short- and long-term outcomes. However, in the unmatched population, women presented at an older age, with a greater symptomatic burden, more comorbidities, and consequently a higher operative risk. These findings highlight the need for earlier detection and treatment of women with CAD.
Limitations
Limitations include the retrospective, single-centre design. A limited number of composite end-point events potentially reduces statistical power. Patients were not randomized to on- or off-pump CABG. Despite limited overlap in propensity scores, covariate balance after matching was adequate, making the matched sample suitable for comparison, though results mainly apply to patients within the region of common support.
Long-term follow-up was confined to all-cause mortality, with missing data on cause-specific mortality and repeat interventions limiting conclusions on cardiovascular-specific outcomes.
CONCLUSIONS
In the unmatched cohort, women undergoing on-pump CABG presented with more comorbidities, a higher symptomatic burden, and a higher risk profile. Women received fewer bypass grafts and more SVGs than men. After matching, graft number, strategy, and early composite outcomes were similar. Women required more RBC transfusions, experienced more WHD, and had longer hospital stays. No differences in short-term or long-term all-cause mortality were observed. Female sex was not significantly associated with long-term mortality. These findings underscore the overall safety of on-pump CABG in women. Improving outcomes in women may require a multimodal approach, including strategies to minimize anaemia and risk factors for WHDs.
Supplementary Material
Glossary
Abbreviations
- BIMA
bilateral internal mammary artery
- CABG
coronary artery bypass grafting
- CAD
coronary artery disease
- DSWI
deep sternal wound infection
- ICU
intensive care unit
- KDIGO
Kidney Disease: Improving Global Outcomes
- NYHA
New York Heart Association
- RBC
red blood cells
- SVG
saphenous vein graft
- WHD
wound healing disorder
Contributor Information
Tim Knochenhauer, Department of Cardiac Surgery, University Medical Center Hamburg-Eppendorf, Hamburg 20246, Germany; DZHK (German Centre for Cardiovascular Research), Partner Site North, Hamburg 20246, Germany.
Friedrich Sobik, Department of Cardiac Surgery, University Medical Center Hamburg-Eppendorf, Hamburg 20246, Germany.
Jonas Pausch, Department of Cardiac Surgery, University Medical Center Hamburg-Eppendorf, Hamburg 20246, Germany.
Inken Detlef, Department of Cardiac Surgery, University Medical Center Hamburg-Eppendorf, Hamburg 20246, Germany.
Jens Brickwedel, Department of Cardiac Surgery, University Medical Center Hamburg-Eppendorf, Hamburg 20246, Germany.
Beate Reiter, Department of Cardiac Surgery, University Medical Center Hamburg-Eppendorf, Hamburg 20246, Germany.
Svante Zipfel, Department of Cardiac Surgery, University Medical Center Hamburg-Eppendorf, Hamburg 20246, Germany.
Yvonne Schneeberger, Department of Cardiovascular Surgery, Asklepios Klinik St. Georg, Hamburg 20099, Germany.
Hermann Reichenspurner, Department of Cardiac Surgery, University Medical Center Hamburg-Eppendorf, Hamburg 20246, Germany; DZHK (German Centre for Cardiovascular Research), Partner Site North, Hamburg 20246, Germany.
Evaldas Girdauskas, Department of Cardiac Surgery, University Medical Center Hamburg-Eppendorf, Hamburg 20246, Germany; DZHK (German Centre for Cardiovascular Research), Partner Site North, Hamburg 20246, Germany.
Bjoern Sill, Department of Cardiac Surgery, University Medical Center Hamburg-Eppendorf, Hamburg 20246, Germany.
Andreas Schaefer, Department of Cardiac Surgery, University Medical Center Hamburg-Eppendorf, Hamburg 20246, Germany.
AUTHOR CONTRIBUTIONS
Tim Knochenhauer (Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Visualization, Writing—original draft, Writing—review & editing), Friedrich Sobik (Data curation, Methodology, Writing—review & editing), Jonas Pausch (Data curation, Methodology, Writing—review & editing), Inken Detlef (Data curation, Methodology, Writing—review & editing), Jens Brickwedel (Data curation, Methodology, Writing—review & editing), Beate Reiter (Data curation, Methodology, Writing—review & editing), Svante Zipfel (Data curation, Methodology, Writing—review & editing), Yvonne Schneeberger (Conceptualization, Data curation, Investigation, Methodology, Writing—review & editing), Hermann Reichenspurner (Project administration, Supervision, Writing—review & editing), Evaldas Girdauskas (Project administration, Supervision, Writing—review & editing), Bjoern Sill (Conceptualization, Data curation, Methodology, Supervision, Writing—review & editing), and Andreas Schaefer (Conceptualization, Data curation, Methodology, Project administration, Supervision, Writing—original draft, Writing—review & editing)
SUPPLEMENTARY MATERIAL
Supplementary material is available at ICVTS online.
FUNDING
None declared.
CONFLICTS OF INTEREST
None declared.
DATA AVAILABILITY
The data underlying this article will be shared on reasonable request to the corresponding author.
REFERENCES
- 1. Neumann FJ, Sousa-Uva M, Ahlsson A, et al. ; ESC Scientific Document Group. 2018 ESC/EACTS guidelines on myocardial revascularization. Eur Heart J. 2019;40:87-165. [DOI] [PubMed] [Google Scholar]
- 2. Lawton JS, Tamis-Holland JE, Bangalore S, et al. 2021 ACC/AHA/SCAI guideline for coronary artery revascularization: a report of the American college of cardiology/American heart association joint committee on clinical practice guidelines. Circulation. 2022;145:e18-e114. [DOI] [PubMed] [Google Scholar]
- 3. Benjamin EJ, Muntner P, Alonso A et al. Heart disease and stroke statistics-2019 update: a report from the American heart association. Circulation, 2019;139:e56-e528. Correction appears in Circulation 2020;141:e33. [DOI] [PubMed] [Google Scholar]
- 4. Gaudino M, Di Mauro M, Fremes SE et al. Representation of women in randomized trials in cardiac surgery: a meta-aAnalysis. J Am Heart Assoc, 2021;10:e020513. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Gaudino M, Di Franco A, Alexander JH, et al. Sex differences in outcomes after coronary artery bypass grafting: a pooled analysis of individual patient data. Eur Heart J. 2021;43:18-28. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Bryce Robinson N, Naik A, Rahouma M, et al. Sex differences in outcomes following coronary artery bypass grafting: a meta-analysis. Interact CardioVasc Thorac Surg. 2021;33:841-847. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Fink N, Nikolsky E, Assali A, et al. Revascularization strategies and survival in patients with multivessel coronary artery disease. Ann Thorac Surg. 2019;107:106-111. [DOI] [PubMed] [Google Scholar]
- 8. Enumah ZO, Canner JK, Alejo D, et al. Persistent racial and sex disparities in outcomes after coronary artery bypass surgery: a retrospective clinical registry review in the drug-eluting stent era. Ann Surg. 2020;272:660-667. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Reynolds HR, Picard MH, Spertus JA, et al. Natural history of patients with ischemia and No obstructive coronary artery disease: the CIAO-ISCHEMIA study. Circulation. 2021;144:1008-1023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Hessian R, Jabagi H, Ngu JMC, et al. Coronary surgery in women and the challenges we face. Can J Cardiol. 2018;34:413-421. [DOI] [PubMed] [Google Scholar]
- 11. Jawitz OK, Lawton JS, Thibault D, et al. Sex differences in coronary artery bypass grafting techniques: a society of thoracic surgeons database analysis. Ann Thorac Surg. 2022;113:1979-1988. [DOI] [PubMed] [Google Scholar]
- 12. Schwann TA, Habib RH, Wallace A, et al. Operative outcomes of multiple-arterial versus single-arterial coronary bypass grafting. Ann Thorac Surg. 2018;105:1109-1119. [DOI] [PubMed] [Google Scholar]
- 13. Knochenhauer T, Sobik F, Pausch J, et al. Sex related long-term outcomes in on-pump coronary artery bypass grafting: a propensity score-matched analysis. Thorac Cardiovasc Surg. 2026;74:S1-S70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Sambola A, Del Blanco BG, Kunadian V et al. Sex-based differences in percutaneous coronary intervention outcomes in patients with ischaemic heart disease. Eur Cardiol. 2023;18:e06. 10.15420/ecr.2022.24 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Attia T, Koch CG, Houghtaling PL, et al. Does a similar procedure result in similar survival for women and men undergoing isolated coronary artery bypass grafting? J Thorac Cardiovasc Surg. 2017;153:571-579.e9. [DOI] [PubMed] [Google Scholar]
- 16. Goldman S, Zadina K, Moritz T, et al. ; VA Cooperative Study Group #207/297/364. Long-term patency of saphenous vein and left internal mammary artery grafts after coronary artery bypass surgery: results from a Department of Veterans Affairs Cooperative Study. J Am Coll Cardiol. 2004;44:2149-2156. [DOI] [PubMed] [Google Scholar]
- 17. Gaudino M, Samadashvili Z, Hameed I, et al. Differences in long-term outcomes after coronary artery bypass grafting using single vs multiple arterial grafts and the association with sex. JAMA Cardiol. 2021;6:401-409. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Knochenhauer T, Schaefer A, Brickwedel J, et al. Association of HbA1c and utilization of internal mammary arteries with wound infections in CABG. Front Cardiovasc Med. 2024;11:1345726. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Crabtree TD, Codd JE, Fraser VJ, et al. Multivariate analysis of risk factors for deep and superficial sternal infection after coronary artery bypass grafting at a tertiary care medical center. Semin Thorac Cardiovasc Surg. 2004;16:53-61. [DOI] [PubMed] [Google Scholar]
- 20. Stammers AH, Tesdahl EA, Mongero LB, et al. Gender and intraoperative blood transfusion: analysis of 54,122 non-reoperative coronary revascularization procedures. Perfusion. 2019;34:236-245. [DOI] [PubMed] [Google Scholar]
- 21. Wester ML, Sampon F, Olsthoorn JR, et al. ; Cardiothoracic Surgery Registration Committee of the Netherlands Heart Registration. Gender is independently associated with red blood cell and platelet transfusion in patients undergoing coronary artery bypass grafting: data from The Netherlands heart registration. J Cardiothorac Vasc Anesth. 2024;38:924-930. [DOI] [PubMed] [Google Scholar]
- 22. Räsänen J, Ellam S, Hartikainen J, et al. Sex differences in red blood cell transfusions and 30-Day mortality in cardiac surgery: a single center observational study. J Clin Med. 2023;12:7674. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Bianco V, Kilic A, Gleason TG, et al. Long-term outcomes after reoperative coronary artery bypass grafting. Ann Thorac Surg. 2021;111:150-158. [DOI] [PubMed] [Google Scholar]
- 24. Mohamed MO, Shoaib A, Gogas B, et al. Trends of repeat revascularization choice in patients with prior coronary artery bypass surgery. Catheter Cardiovasc Interv. 2021;98:470-480. [DOI] [PubMed] [Google Scholar]
- 25. Thakur U, Nerlekar N, Muthalaly RG, et al. Off- vs. on-pump coronary artery bypass grafting long-term survival is driven by incompleteness of revascularisation. Heart Lung Circ. 2020;29:149-155. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data underlying this article will be shared on reasonable request to the corresponding author.