Impact of planned concomitant coronary artery bypass grafting on risk of major adverse cardiovascular events in elective aortic hemiarch surgery

Abstract

Background

Hemiarch replacement of the ascending aorta has become routine in many aortic centers. While the addition of coronary bypass does not add a lot of time to the procedure, it carries with more significant comorbidities. We hypothesize that the addition of CABG carries a higher risk of complication than hemiarch alone.

Methods

This is a single-center, retrospective cohort study of 419 patients undergoing elective hemiarch surgery between February 2010 and May 2023. Patients were categorized into concomitant CABG (n = 42) and non-CABG (n = 379) groups. Perioperative variables and outcomes were analyzed. Both univariate and multivariate logistic regressions were used to identify predictors for MACE.

Results

Of 419 patients, 42 (10%) patients received adjunctive CABG. This group was older (68.1 vs. 60.4 years, p < 0.001) with more comorbidities associated with coronary artery disease (CAD), such as hypertension (92.9% vs. 59.2%, p < 0.001), type 2 diabetes (33.3% vs. 8.8%, p < 0.001), and atrial fibrillation (19% vs. 5.8%, p = 0.006). CABG patients had longer cardiopulmonary bypass (158 vs. 131 min, p < 0.001) and aortic cross-clamp (115.5 vs. 95 min, p < 0.001) times and required more intraoperative blood products, FFP (4 vs. 2 units, p = 0.010) and platelets (2 vs. 1 units, p < 0.001). Postoperative complications, including arrhythmia (40.5% vs. 21.8%, p = 0.012), mechanical circulatory support (11.9%, 1.9%, p = 0.004), acute kidney injury (16.7% vs. 0.5%, p < 0.001), infection (11.9% vs. 3.7%, p = 0.032), mortality (9.5% vs. 0.5%, p = 0.001), stroke (9.5% vs. 2.1%, p = 0.024), and the composite outcome– MACE (21.4% vs. 2.9%, p < 0.001) were higher in the CABG group. Multivariate analysis identified the number of bypassed vessels (OR: 2.23, CI 1.33–3.69, p = 0.002), age (OR: 1.07, CI: 1.02–1.13, p = 0.006), and female gender (OR: 3.53, CI: 1.31–9.64, p = 0.012) as significant risk factors for MACE.

Conclusions

Concomitant CABG may increase the risk of MACE compared to other patients undergoing hemiarch. These data argue that the risk may be higher for concomitant CABG but should still undergo revascularization. Future research should focus on preoperative optimization, operative strategies, and sex-specific risk factors to improve elective hemiarch replacement outcomes.

Introduction

With advancements in surgical techniques and perioperative care, elective aortic hemiarch replacements have a low rate of operative and postoperative mortality in high-volume centers [1,2,3,4]. In this institution’s experience, more patients who require surgery for aortic pathologies present with higher age and more comorbidities. Older age is a risk factor for higher mortality following arch surgery, and patients with advanced age typically present with more comorbidities [3]. About 15–25% of patients with aortic aneurysms also have concomitant coronary artery disease (CAD) [5, 6] and they may benefit from revascularization. Needing concomitant coronary artery bypass grafting (CABG) can be seen as a proxy for impaired ventricular function, which can potentially increase perioperative risks. For instance, the STS aortic and mitral valve replacement risk calculators indicate that adding CABG increases mortality risk by 1.5 to 2 times (https://stsmultivalvecalculator.research.sts.org/).

On the other hand, the effect of CABG on the perioperative risks during aortic surgery has been reported with mixed results. Some studies demonstrated that concomitant CABG did not increase the risk of major adverse events in ascending or total arch replacement [5, 6]. Nevertheless, another study suggested that concomitant CABG might increase the operative risk in total arch replacement when three or more coronary vessels were bypassed [7]. Although elective hemiarch arch replacement is generally safe to perform in experienced hands, one study showed that adding concomitant procedures, including CABG, increased the risk of adverse outcomes (stroke and death) in patients receiving ascending/hemiarch surgery [8]. In these cases, left ventricular dysfunction might have played a role in the increased operative risk. Given these mixed findings, this study aims to further explore and clarify the impact of concomitant CABG on the risk of major adverse cardiovascular events (MACE) in patients undergoing elective hemiarch replacement in the current era.

Methods

Study design and data collection

This is a retrospective cohort study analyzing a prospectively maintained database of a high-volume aortic center. This study has been reviewed and approved by the Colorado Multiple Institutional Review Board (COMIRB #17–0198, February 6th, 2017). All adult patients (n = 419) presenting for elective hemiarch replacement between February 2010 and May 2023 were reviewed and included in the cohort. The data collection was finalized in December 2024. Unplanned CABGs were excluded from this study. Seven surgeons including two high-volume aortic surgeons contributed cases to this study. Case volume breakdown of elective hemiarch replacement for individual surgeons is shown Supplemental Table S1.

Patient categorization

All patients underwent coronary artery evaluation with cardiac catheterization, stress test, or coronary artery cross-sectional imaging, prior to the hemiarch replacement to determine whether a concomitant CABG was indicated. The patients were categorized into two primary groups: those who underwent concomitant, planned CABG and those who did not. Additionally, patients were stratified based on the year of surgery into Early Cohort (EC, 2011–2018) and Later Cohort (LC, 2019–2023) to assess potential chronological differences in outcomes. The cutoff denotes the first implementation of the Shaggy Aorta protocol for cerebral perfusion, which involves a brief period of retrograde cerebral perfusion (RCP) followed by antegrade cerebral perfusion (ACP) for the remainder of the circulatory arrest time [9]. The results for the comparison between EC and LC are shown in Supplementary Materials.

Surgical techniques

Chest exposure

A full median sternotomy was performed in all cases. The soft tissues were dissected down to the sternum, which was then divided with a reciprocating saw. A hemostatic agent, such as Gelfoam, was applied to the bone marrow to achieve hemostasis. After further retraction, the pericardium was opened, and the pericardial well was secured. Carbon dioxide gas was insufflated into the pericardial well during the surgery to reduce the risk of air embolism.

Cardiopulmonary bypass

After systemic heparinization, the cardiopulmonary bypass (CPB) setup began with arterial cannulation. Prior to the implementation of the Shaggy Aorta protocol for cerebral perfusion at this institution in December 2018, arterial cannulation was typically performed via the distal ascending aorta, innominate artery, or right axillary artery. Cannulation of the innominate or right axillary artery was performed either over a synthetic graft or a cannula directly. After adoption of the Shaggy Aorta protocol, the primary site for arterial cannulation was the distal ascending aorta using a soft flow cannula. In rare cases of unfavorable anatomy, alternative sites, such as the right common carotid artery, left common carotid artery, left subclavian artery, or peripheral cannulation, were utilized. After completing the distal anastomosis, distal perfusion was transitioned to the sidearm of the hemiarch graft.

Venous cannulation was predominantly performed at the right atrium using a dual-stage cannula. In only a few cases, bicaval cannulation was utilized. A left ventricular vent was routinely placed. CPB and cooling were initiated once the activated clotting time exceeded 480 seconds. The target cooling temperature for hypothermic circulatory arrest (HCA) was 24 to 28 °C, depending on surgeon’s preference.

Hypothermic circulatory arrest and cerebral perfusion

Once CPB cannulation was completed, cerebral perfusion during HCA was setup. Before 2019, cerebral perfusion was typically achieved with unilateral, selective ACP using the innominate or right axillary graft/cannula. There were only two patients who had bilateral ACP. In cases where the Shaggy Aorta protocol was used, an SVC cannula was placed and connected to the arterial line for RCP. For anticipated HCA time longer than 10 min, the innominate artery was accessed if not already done during the CPB setup. Once target hypothermia and cardioplegia were achieved, cerebral perfusion could be initiated.

Cardioplegia

The type of cardioplegia used was based on the patient’s coronary artery anatomy and the degree of stenosis or occlusion. Different cardioplegia solutions were used throughout the study period: cold blood and Plegisol (before 2011), Custodiol (2012–2015), and del Nido (since 2014). Antegrade cardioplegia was delivered via the root vent cannula after cross-clamping the ascending aorta. Retrograde cardioplegia was given via a coronary sinus cannula. Selective cardioplegia was used when standard antegrade or retrograde methods were insufficient. Depending on the patient’s coronary artery conditions and surgeon’s preferences, antegrade cardioplegia, retrograde cardioplegia, or both were utilized.

Concomitant coronary artery bypass grafting

The coronary arteries and their branches were inspected to identify suitable segments for bypass, avoiding areas with extensive calcification. Graft selection was determined by the location of the target vessels. Arterial grafts, such as left internal mammary artery (LIMA), were predominantly used for left anterior descending artery (LAD) and the saphenous vein graft (SVG) was primarily used for bypassing right coronary artery (RCA), posterior descending artery, obtuse marginal, or other branches (e.g., posterior lateral, diagonal, and ramus).

SVG was harvested using an endoscopic technique, during which the vein was identified and dissected out under direct visualization. The vein was prepared, flushed, and tested for patency. The leg incision was then closed.

CABG was begun following cardioplegia administration and before hemiarch replacement. The distal anastomosis of the SVG to targeted vessels was performed first. The proximal anastomosis of the SVG was completed onto the ascending interposition graft following hemiarch replacement or onto the root graft in cases where root replacement was performed. When LIMA was required, it was mobilized from its origin to the bifurcation and subsequently anastomosed to the LAD. Details of CABG configurations for each patient are provided in Supplemental Table S2.

Hemiarch replacement

The ascending aorta and proximal arch were dissected free, followed by establishing HCA and cerebral perfusion. Rummel tourniquets were placed for both the innominate and left common carotid arteries to control the bypass flow. After the aortic cross-clamp was removed, the aorta was fashioned up to the innominate artery and excised from just above the sinotubular junction into the transverse aortic arch. The hemiarch graft was beveled and sewn to the distal aorta. After deairing, the hemiarch graft was clamped, and the perfusion inflow was transferred to its side arm. The whole body was then reperfused for 5 minutes before rewarming started. Once the distal anastomosis was confirmed to be hemostatic, the graft was trimmed, and the proximal anastomosis to the sinotubular junction was completed.

Weaning from cardiopulmonary bypass and completion

Following the hemiarch replacement and proximal anastomosis of the SVG, the heart was deaired via the root vent. The heart was reperfused for 15 minutes before calcium administration and weaning from CPB. The venous cannula, LV vent, and root vent were then removed. The arterial cannula was also removed (the graft was ligated and divided), and the hemiarch graft side arm was ligated and divided. All cannulation sites were tied down and inspected for bleeding to ensure hemostasis. The patient was rewarmed to 36 °C, and protamine was administered to reverse the effects of heparin. Epicardial pacing wires were inserted, and two mediastinal drains were placed. The sternum was closed with wires, followed by soft tissue and skin closure in multiple layers.

The steps for adjunctive aortic valve intervention, root replacement, atrial fibrillation procedures, and mitral valve intervention were not the primary focus of this study and are therefore not listed.

Data and variables

Preoperative variables included patient demographics (e.g., age, sex, BMI, baseline hemoglobin A1c level, and baseline blood pressure), medical comorbidities (e.g., hypertension, hyperlipidemia, smoking history, diabetes, chronic kidney disease, stroke, pulmonary diseases including obstructive sleep apnea, atrial fibrillation, and peripheral artery disease), left ventricular ejection fraction (LVEF), the presence of carotid stenosis > 70% if preoperative carotid imaging was performed, and history of prior cardiac and aortic surgery. Intraoperative variables included adjunctive procedures (CABG, root replacement, aortic valve intervention, mitral valve intervention, and atrial fibrillation procedures), strategies of circulatory arrest and cardioplegia, nadir bladder temperature, cardiopulmonary bypass time, aortic cross-clamp time, hypothermic circulatory arrest time, antegrade and retrograde cerebral perfusion times, and the number of blood transfusions. Postoperative outcomes including length of stay, reoperation, arrhythmia, new hemodialysis, stroke, myocardial infarction (MI), acute kidney injury (AKI), deep venous thrombosis (DVT), delirium, prolonged ventilation (> 48 h), infection, need for mechanical circulatory support, and death were assessed. The primary endpoint was MACE, a composite outcome of stroke, MI, and death within 30 days.

Statistical analysis

All statistical analyses were performed by using R software (version 4.4.1) with custom-built R scripts and open-source libraries. Continuous variables were initially assessed for data distribution using the Shapiro-Wilk test. Nonparametric data were analyzed using the Mann-Whitney U test for comparisons between two groups and the Kruskal-Wallis test for comparisons among more than two groups. Parametric data were compared using the t-test for two groups and ANOVA for more than two groups. Non-parametric results were reported as median [interquartile range] and parametric results were reported as mean ± standard deviation. For categorical variables, Chi-Square and Fisher’s exact tests were used, depending on the expected frequencies in the data. To identify risk factors for MACE in patients undergoing elective hemiarch replacement, first, univariate logistic regression analysis was performed on all clinically relevant variables. Variables that were statistically significant and those considered borderline significant in univariate analysis were included in the first multivariate logistic regression model (Supplemental Table S4). To construct the final multivariate model, a stepwise approach was employed to reduce the number of covariates and maintain an appropriate event-to-variable ratio, minimizing the risk of overfitting while preserving the predictive power of the model. The models were validated using multiple statistical metrics. Model performance was assessed with the area under the receiver operating characteristic curve and the Brier score. Multicollinearity among predictors was evaluated using the variance inflation factor. Cross-validation was performed by dividing the cohort into training and testing sets in a 4:1 ratio. A p-value < 0.05 was considered statistically significant.

Results

Patient demographics and comorbidities (Table 1)

Table 1 Comparison of preoperative variables between the concomitant CABG and non-CABG groups among hemiarch replacement patients

Of 419 patients in the cohort, 42 (10%) patients underwent concomitant CABG. The concomitant CABG group was significantly older (68.1 [65.5–72.8] vs. 60.4 [49.3–69] years, p < 0.001), along with more comorbidities including hyperlipidemia (p < 0.001), hypertension (p < 0.001), type 2 diabetes (p < 0.001), pulmonary disease (p = 0.020), peripheral artery disease (p = 0.025), and atrial fibrillation (p = 0.006). There were no significant differences in measured baseline systolic and diastolic blood pressures, baseline hemoglobin A1c level, preoperative LVEF, history of smoking, chronic kidney disease, previous stroke, or prior cardiac and aortic surgery. Approximately 25% of patients received preoperative carotid imaging using ultrasound or CT scan. Among these, three patients in the non-CABG group had carotid stenosis greater than 70%, with no significant difference in the carotid stenosis rates observed between the two groups.

Operative techniques (Table 2)

Table 2 Comparison of operative details between the concomitant CABG and non-CABG groups among hemiarch replacement patients

The median number of bypassed vessels was 1 [1–2] in the CABG group. Seven patients (16.7%) had 3-vessel CABG. The CABG group had marginally more adjunctive atrial fibrillation procedures performed than the non-CABG group [5 (11.9%) vs. 18 (4.8%), p = 0.068]. However, the two groups had no significant differences in other adjunctive procedures, such as aortic and mitral valve interventions and root replacement. Regarding hypothermia, no difference was observed in target temperature (p = 0.175) and nadir bladder temperature (p = 0.274). The times during cardiopulmonary bypass (158 [135.2–220.5] vs. 131 [111–165] minutes, p < 0.001) and aortic cross-clamp (115.5 [95–149] vs. 95 [72–121.2] minutes, p < 0.001) were longer in the concomitant CABG group. The strategies and setups of cardioplegia and circulatory arrest did not differ between the two groups. del Nido was the most commonly used cardioplegia solution with Custodiol and cold blood being the second most commonly used solutions for non-CABG and CABG groups, respectively.

Operative outcomes (Table 3)

Table 3 Comparison of patient outcomes between the concomitant CABG and non-CABG groups among hemiarch replacement patients

In those who received intraoperative blood products, the CABG group required more fresh frozen plasma (4 [2–4] vs. 2 [0–4] units, p = 0.010) and platelets (2 [1–2.8] vs. 1 [0–2] units, p < 0.001), compared to the non-CABG group. Additionally, there was a trend for more transfusion quantity of packed red blood cells in the CABG group (p = 0.087).

The CABG group experienced a longer hospital stay (8 [7–10] vs. 7 [6–9] days, p = 0.002). Patients undergoing concomitant CABG had higher rates of postoperative morbidities such as infection (p = 0.032), need for MCS (p = 0.004), arrhythmia (p = 0.012), AKI (p < 0.001), mortality (p = 0.001), stroke (p = 0.024), and MACE (p < 0.001), compared to the non-CABG group. The infection types included pneumonia (7 vs. 3), urinary tract infections (4 vs. 1), bacteremia (2 vs. 2), sternal osteomyelitis/mediastinitis (1 vs. 0), Clostridium difficile colitis (1 vs. 0), and cholecystitis (1 vs. 0) in the non-CABG and CABG groups, respectively. Operative mortality was attributed to cardiogenic shock (2 vs. 3) and septic shock (1 vs. 2), while stroke was not a cause of death. There was a tendency for higher rates of prolonged ventilation (p = 0.079), and new hemodialysis (p = 0.052). Late mortality rate was similar between the two groups (p = 0.492), and the death of two patients was associated with stroke.

Risk factors for MACE (Fig. 1)

Fig. 1

Forest plot showing the adjusted odds ratios (OR) and 95% confidence intervals (CI) for mutivariate logistic regression model in predicting major adverse cardiovascular events (MACE) in patients undergoing hemiarch replacement

Concomitant CABG (OR: 9.07, CI: 3.44–23.51, p < 0.001), the number of bypassed vessels (OR: 2.54, CI: 1.55–4.02, p < 0.001), age (OR: 1.09, CI: 1.04–1.16, p < 0.001), and female gender (OR: 3.43, CI: 1.37–8.62, p = 0.008) were significant predictors in the unadjusted model (Table 4). Cardiovascular (CV) comorbidities (p = 0.058) and mitral valve intervention (p = 0.055) demonstrated borderline significance.

Table 4 Univariate and multivariate logistic regression models in predicting MACE for patients undergoing elective hemiarch replacement

In the final multivariate model, the number of bypassed vessels (OR: 2.23, CI 1.33–3.69, p = 0.002), age (OR: 1.07, CI: 1.02–1.13, p = 0.006), and female gender (OR: 3.53, CI: 1.31–9.64, p = 0.012) were identified as independent predictors of MACE (Table 4). Concomitant CABG was excluded from this model due to overlapping with another variable. CV comorbidity and mitral valve intervention were not included in the final model as they did not maintain significance in adjusted analyses.

Discussions

As more patients with advanced age and more comorbidities present for aortic surgery, in addition to aortic pathologies, some of them may have concomitant coronary artery disease that may benefit from revascularization during the same operating room visit. How adjunctive or concomitant CABG affects the risks of postoperative major adverse cardiovascular events, including stroke, myocardial infarction, and death, remains controversial in elective aortic arch surgery, particularly hemiarch replacement. This study aimed to address this clinical gap. Unlike a previous study that combined only stroke and death, the present study grouped stroke, MI, and death into a single composite outcome for analysis. This reflects the clinical significance of MI, a complication of coronary artery disease that is strongly associated with increased mortality. Additionally, stroke is often perceived by patients as being as severe as death and is similarly linked to early mortality [8].

In this study, the concomitant CABG group was older and had more baseline comorbidities typically associated with CAD, including hypertension, hyperlipidemia, diabetes, peripheral artery disease, and atrial fibrillation. The similar preoperative LVEF suggests that the burden of CAD had not significantly affected the left ventricular contraction. Most of the patients receiving concomitant CABG presented with significant stenosis in one or two coronary arteries, and when CAD involved RCA or smaller branches, it did not necessarily cause a decrease in baseline LV function in the study cohort.

Preoperative carotid imaging was not a routine study performed prior to elective hemiarch replacement by some surgeons at this institution, except in patients with known carotid artery pathology, history of aortic dissection, or clinical signs suggesting carotid pathology, such as a carotid bruit. Given that only 3 non-CABG patients had baseline high-grade carotid stenosis, further analysis was not performed to look into the association between carotid stenosis and stroke.

Operative details

The setups of CPB, cardioplegia, HCA, and cerebral perfusion were similar between the CABG and non-CABG groups, although some techniques for cerebral perfusion and the choice of cardioplegia solution were modified over time. Compared to the non-CABG patients, those undergoing concomitant CABG had longer CPB and aortic cross-clamp times, which could be explained by the addition of CABG procedures. However, adding CABG did not significantly prolong the HCA, ACP, and RCP times.

Notably, the Shaggy Aorta protocol was utilized in 81.5% of patients in the LC, with 71.9% receiving RCP only during HCA (Supplemental Table S3). Compared to the EC, the LC had significantly shorter CPB, HCA, ACP (if used) times. When ACP was used, it was performed through direct cannulation rather than over a graft in the LC. The improvements in operative times suggest the success of implementation of the Shaggy Aorta Protocol and the standardized CPB setup.

Concomitant CABG was chronologically evenly distributed over the entire study period. Additional adjunctive procedures, including aortic valve intervention, mitral valve intervention, root replacement, and atrial fibrillation were occasionally combined with hemiarch replacement. There was no significant difference in the rates of these additional procedures between the CABG and non-CABG groups.

Patient outcomes

Although the CABG patients seemed to need more blood products, the surgery team at this center managed to avoid extensive transfusions when possible, to reduce the risk of graft thrombosis. Postoperatively, the concomitant CABG group experienced higher rates of stroke and mortality, resulting in a higher rate of MACE, but a similar rate of MI. The perioperative mortality was not associated with stroke. The CABG patients also had more AKI, need for MCS, infection, arrhythmia, compared to the non-CABG patients. Several surgeons contributed to this study with no differences in MACE among them, and there was no chronological difference in the rates of MACE between EC and LC.

To further explore the potential risk factors for MACE, univariate and multivariate logistic regression models were constructed. Given the total of 20 MACE events, three strong covariates were included in the final model, which demonstrated excellent model performance and cross-validation. Consistent with the findings of Ehrlich et al., where concomitant procedure was a risk factor for adverse outcomes [8], concomitant CABG was associated with increased likelihood of MACE in the present study. Each additional revascularized coronary artery increased the odds of MACE 2.23-fold, whereas CV comorbidity was not a significant predictor. This suggests that while patients requiring CABG are generally sicker at baseline, the elevated risk of MACE is primarily attributable to the procedure itself rather than their baseline comorbidities.

The goal of concomitant CABG is to reverse any potential ventricular dysfunction and facilitate postoperative cardiac recovery. It may prolong an already complex hemiarch procedure with longer CPB and aortic cross clamp times, which can exacerbate systemic inflammation, coagulopathy, and subsequent end-organ damage, thereby increasing the risk of mortality. Additionally, reperfusion injury associated with CABG may also contribute to the increased mortality. Patients requiring CABG typically have both coronary and systemic atherosclerosis, and manipulation of calcified coronary arteries, aorta, and its branches may increase the risk of embolization, leading to stroke.

Age was also found to be a significant, independent predictor for MACE, with each additional year increasing the odds of MACE by 7%. When stratified by age, older patients (age ≥ 75) had a higher stroke rate of 7.8% compared to 2.2% in younger patients (age < 75). The higher stroke rate in older patients is likely due to a greater burden of systemic atherosclerosis. However, operative mortality was not significantly different between older (7.8%) and younger (4.3%) patients.

Female gender was associated with a 3.53-fold increased likelihood of MACE compared to male patients. Further sub-analysis showed that female patients were healthier and had shorter CPB and aortic cross-clamp times but experienced a higher rate of stroke (Supplemental Table S5). There were no significant differences found in MI and operative mortality between male and female patients. These findings suggest that female gender is an independent risk factor for MACE in hemiarch replacement, specifically stroke, and is not confounded by other factors. Anatomical differences (e.g., smaller vessel size) and hormonal influences may play a role in this disparity, though further research is needed to confirm the hypotheses. Other additional procedures (e.g., aortic and mitral valve interventions, root replacement, and atrial fibrillation procedures) were not significant risk factors for MACE.

Impact of surgical practice

Based on the results of this study, concomitant CABG may increase the risk of MACE and should be performed selectively. It may be most appropriate for patients with single-vessel disease, as the risk of MACE rises further with an increasing number of bypassed vessels. Surgeons should carefully evaluate the number of bypassed vessels when planning the procedure. For patients with extensive CAD requiring multiple bypasses, a staged approach (e.g., separating CABG and hemiarch replacement) should be considered to potentially reduce the risk of MACE. Further prospective cohort studies are needed to determine whether the staged approach results in better outcomes.

Regarding non-modifiable risk factors, stroke is the primary driving factor of MACE in older and female patients. However, advanced age alone should not be considered an absolute contraindication for hemiarch surgery as the potential long-term benefits, such as preventing aortic rupture and sudden death, may outweigh perioperative risks. The mechanism by which female gender increases the risk of MACE, particularly stroke, remains unclear, and further investigation into sex-specific biological factors is warranted to improve outcomes in female patients. Future research should also focus on preoperative optimization and operative strategies to enhance outcomes for patients undergoing elective hemiarch surgery.

Limitations

There are several limitations of this study. First, this is a retrospective single center study. Secondly, the study includes surgical results from eight surgeons, but the outcomes of individual surgeon were not evaluated, which may introduce variability in the results. Thirdly, the sample size of this study was relatively small, with a low rate of concomitant CABG (10%) and most of the patients receiving CABG had 1–2 vessels revascularized (Supplemental Table S2). Additionally, preoperative risk scoring was not assessed, and the details of coronary artery anatomy were not included in the risk prediction models.

Conclusions

In this study cohort, concomitant CABG was associated with an increased risk of MACE, with the risk increased with each additional bypassed vessel. As such, it should be performed selectively, particularly in patients with single-vessel disease, where the benefits may outweigh the risks. For patients with extensive CAD requiring multiple bypasses, a staged approach separating CABG and hemiarch should be considered as a potential strategy to lower the risk of MACE, though this approach warrants further investigation. Future research should focus on preoperative optimization, operative strategies, and sex-specific risk factors to improve outcomes in patients undergoing elective hemiarch replacement with and without concomitant CABG.