Short-Term Outcomes of Total Arterial Revascularization Compared to Conventional Coronary Artery Bypass Graft in Patients with Multivessel Disease and Left Ventricular Dysfunction

,


Introduction
Coronary artery disease (CAD) is a leading cause of ischemic heart disease, which is associated with increased morbidity and mortality [1].Multivessel CAD reduces the blood and oxygen supply to the myocardium, resulting in ischemia, angina, and heart failure [2].One of the most efective treatments for patients with multivessel CAD is coronary artery bypass grafting (CABG) [3], which restores the blood fow to the ischemic areas of the heart by using grafts to bypass the obstructed coronary arteries.
However, not all patients who undergo CABG have the same prognosis.A major factor that infuences the outcomes of CABG is the presence of left ventricular dysfunction (LVD) [4].LVD is a common complication of multivessel CAD, which impairs the cardiac output and increases the risk of adverse cardiac events [5].LVD can be caused by previous myocardial infarction, chronic ischemia, or other factors that afect the structure and function of the left ventricle.
Terefore, it is important to optimize the surgical strategy for this high-risk population to improve their longterm outcomes.One specifc technique of CABG that has gained interest in recent years is total arterial revascularization (TAR).TAR involves the use of only arterial grafts to perform CABG, without using any venous grafts.
Recent evidence has suggested that TAR may ofer superior long-term patency rates and improved clinical outcomes in comparison to conventional revascularization (CR) [6,7].However, TAR is technically challenging and may pose higher operative risks [8], especially in patients with LVD.Moreover, few studies have examined the outcomes of TAR in patients with multivessel CAD and LVD.
In this paper, we aim to compare the early outcome of TAR and CR in patients with multivessel CAD and LVD who underwent isolated CABG at our institution between January 2014 and December 2022, ultimately providing valuable insights to guide optimal surgical decision-making in this challenging clinical context.We hypothesize that TAR would be associated with lower rates of postoperative complications, better graft patency, and improved cardiac function than conventional CABG in this high-risk population.

Study Design and Population.
Tis was a retrospective cohort study based on a single-center registry of patients who underwent isolated CABG for multivessel CAD and LVD between January 2014 and December 2022 at our institution.Multivessel CAD was defned as the presence of at least two coronary arteries with more than 50% stenosis on coronary angiography.LVD was defned as a left ventricular ejection fraction (LVEF) of less than 40% on preoperative echocardiography.Te exclusion criteria were as follows: (1) patients with combined ventricular wall aneurysm, ventricular septal perforation, chordae tendineae rupture, etc.; (2) patients with single-vessel disease, emergency or salvage CABG; (3) patients with prior cardiac surgical procedure; (4) concomitant valve surgery or other cardiac procedures.Perioperative insertion of an intra-aortic balloon pump (IABP) was indicated for patients with one or more of the following conditions: cardiogenic shock or ventricular failure that did not respond to medical therapy, unstable hemodynamics, persistent angina, ventricular arrhythmia, or severe left main coronary artery stenosis (>70%).
Patients were divided into two groups according to the type of revascularization: TAR (using total arterial revascularization strategy) or CR (using venous or mixed arterial and venous conduits).Total arterial revascularization was defned as the use of only arterial conduits (left internal thoracic, right internal thoracic, and/or radial arteries) for bypass grafts.Conventional CABG was defned as the use of ≥1 saphenous vein grafts in addition to arterial conduits [9].Isolated single left internal thoracic artery (LITA) to left anterior descending (LAD) grafting alone did not meet criteria for TAR in this study.Te study protocol was approved by the institutional review board, and the requirement for informed consent was waived.

Data Collection and Defnitions.
Data on baseline characteristics, operative details, postoperative complications, and follow-up outcomes were collected from electronic medical records, surgical databases, and telephone interviews.Te primary outcome was in-hospital mortality.Te secondary outcomes were early complications, freedom from angina rate within 6 months, graft patency rate at 3 months, and readmission rate within 6 months.Postoperative complications included respiratory complication, prolonged ventilation, low cardiac output syndrome (LCOS) requiring extracorporeal membrane oxygenation (ECMO), cardiocerebral events (myocardial infarction, and stroke), bleeding requiring reoperation, sternal wound infection, mediastinitis, acute kidney injury, and atrial fbrillation.Inhospital mortality was defned as death for any reason occurring within 30 days after the operation.Myocardial infarction was diagnosed based on the presence of typical symptoms, electrocardiographic changes, and elevated cardiac enzymes.Stroke was defned as a new focal neurological defcit lasting more than 24 hours and confrmed by imaging studies.Postoperative complications were defned according to the Society of Toracic Surgeons criteria.Graft patency was assessed by computed tomography angiography (CTA) at 3 months after surgery.Tis was a routine practice in our institution for all patients who underwent CABG, regardless of the type of revascularization strategy.We defned graft occlusion as the absence of contrast opacifcation within the graft lumen, and graft stenosis as a reduction of more than 50% in the graft diameter.Follow-up data were obtained from outpatient clinic visits, telephone interviews, or linkage with the national death registry.Te last follow-up date was June 30, 2023.Telephone interviews were conducted for patients who did not attend the clinic visits or who had missing data.Linkage with the national death registry was performed to verify the vital status of the patients and to identify any deaths that occurred outside the hospital.By using these methods, we were able to obtain complete follow-up data for all 112 patients within 6 months.Te clinical indication for CABG was categorized as silent ischemia, stable angina, unstable angina, non-ST-segment elevation myocardial infarction (NSTEMI), or ST-segment elevation myocardial infarction (STEMI), based on the diagnosis at admission.Prior MI was defned as a history of myocardial infarction before the current admission.

Surgical Procedures and Postsurgical
Treatment.We performed CABG under general anesthesia and median sternotomy in all patients.Te choice of revascularization strategy was based on the surgeon's preference and the patient's characteristics: age, comorbidities, anatomical features (coronary-artery blockage location and extent, coronary-artery size, and availability of arteries and veins), and preference.We performed CABG using three diferent techniques: of-pump coronary artery bypass (OPCAB), on-pump coronary artery bypass (ONCAB), and ONCAB + cardioplegic arrest.Te conduits of choice were left and right ITA, radial artery, and saphenous vein.Te LITA was preferentially used to bypass the LAD artery when possible.If the LITA was unsuitable for use, the next choice was the radial artery.For other coronary territories, the right internal thoracic artery (RITA) and the radial artery were used preferentially when they were of 2 Journal of Cardiac Surgery adequate quality and diameter.In some cases, we used composite grafts, such as T grafts or snake grafts, to achieve maximal revascularization with fewer anastomoses or more arterial conduits.We decided the choice of conduits and target vessels based on our discretion and the quality and size of the vessels.Te LITA was harvested in a pedicled fashion in all cases.Te RITA was harvested as a short skeletonized conduit to preserve sternal blood fow and reduce the risk of spasm.Te radial artery and saphenous vein graft were harvested using an open technique.Te completeness of revascularization was assessed by comparing the number of diseased vessels (>70% stenosis) with the number of grafts performed.Te number and location of grafts were determined by the extent and severity of CAD, the quality of target vessels, and the hemodynamic status of the patient.
In the OPCAB technique, we stabilized and positioned the heart using a tissue stabilizer device (Octopus, Medtronic) and a heart positioner device (Starfsh or Urchin, Medtronic) and performed anastomoses on a beating heart.In the ONCAB technique, we established cardiopulmonary bypass using ascending aortic and a single right atrial cannula and performed anastomoses on a beating heart.In the ONCAB + cardioplegic arrest technique, we used the same method as in ONCAB, but we also induced cardiac arrest by delivering cold blood cardioplegia antegrade or retrograde.
All patients received standard postoperative care in the intensive care unit and the ward according to institutional protocols.Antiplatelet medicines, such as aspirin, were started within 6 hours after surgery and continued indefnitely.Other antiplatelet agents, such as clopidogrel, were added according to the discretion of the treating physician.Blood thinners, such as warfarin or rivaroxaban, were prescribed for some patients with concomitant atrial fbrillation, according to their CHA2DS2-VASc score and bleeding risk.Other medications, such as beta blockers, angiotensin-converting enzyme inhibitors, statins, and nitrates, were prescribed according to current guidelines.

Statistical Analysis.
We expressed continuous variables as mean ± standard deviation or median (interquartile range [IQR]), depending on their distribution, and compared them using Student's t-test or Wilcoxon signed-rank test, respectively.We presented categorical variables as counts or percentages and compared them using chi-squared test or Fisher's exact test, as appropriate.We considered a twosided p-value of <0.05 as statistically signifcant.We performed all analyses using SPSS v.26.0 (IBM SPSS Inc., Armonk, NY) and R 4.3.0.

Baseline Characteristics.
A total of 112 cases were included in this study, 52 patients for TAR and 60 patients for CR.Te patients' baseline characteristics and comorbidities are shown in Table 1.Te median age of the total sample was 56 years (IQR 49-73), and 68 patients (60.7%) were male.Te most common clinical indication was unstable angina (42 patients, 37.5%), followed by stable angina (25 patients, 22.3%).Te median LVEF was 35.4% (IQR 33.1-36.9),and the median EuroSCORE II was 2.965% (IQR 2.2-4.4175).Te majority of the patients had three-vessel disease (88 patients, 78.6%) and were in NYHA class III (60 patients, 53.6%).Tere were no signifcant diferences between the two groups in terms of age, sex, BMI, smoking history, diabetes, hypertension, clinical indication, prior myocardial infarction, chronic obstructive pulmonary disease, liver dysfunction, dialysis, peripheral vascular disease, cancer history, stroke history, arrhythmias, NYHA class, left main disease, three-vessel disease, two-vessel disease, left ventricular end-diastolic diameter, LVEF, and EuroSCORE II.Te two groups were well matched and had similar risk profles.

Operative Data and Early Outcomes.
Table 2 shows the operative data and early outcomes of the study participants.Te TAR group had a signifcantly longer operation duration than the CR group (TAR: 268.5 vs. CR: 220 minutes, p � 0.004).Te characteristics of the CABG procedure (OPCAB, ONCAB, or ONCAB + cardioplegic arrest) were not signifcantly different between the two groups (p � 0.180).Of the 43 cases of ONCAB, 37 were elective and six were conversions from OPCAB due to hemodynamic instability, bleeding, or poor exposure.Of the 22 cases of ONCAB + cardioplegic arrest, 17 were elective and fve were conversions from OPCAB for the same reasons.In the TAR group, 21 patients (40.4%) underwent OPCAB, 21 patients (32.7%) underwent ONCAB, and 17 patients (26.9%) underwent ONCAB + cardioplegic arrest.In the CR group, 26 patients (43.3%) underwent OPCAB, 26 patients (43.3%) underwent ONCAB, and 8 patients (13.3%) underwent ONCAB + cardioplegic arrest.Te number of grafts and the rate of complete revascularization were also comparable between the two groups (p � 0.837 and p � 0.807, respectively).Complete revascularization was achieved in 37 patients (71.2%) in the TAR group and in 45 patients (75.0%) in the CR group.Te reasons for incomplete revascularization were technical difculties, poor distal targets, and lack of suitable conduits in some cases.
Intraoperative intra-aortic balloon pump (IABP) placement was required for two patients in the TAR group and one patient in the CR group.No patients in the TAR group and two patients in the CR group required postoperative IABP placement.Te overall incidence of IABP usage did not difer signifcantly between the two groups (TAR: 3.85% vs. CR: 5.00%, p � 0.869).Te TAR group had a signifcantly shorter intensive care unit (ICU) stay (TAR: 3.5 vs. CR: 5 days, p � 0.016) and hospital stay (TAR: 10.5 vs. CR: 12 days, p � 0.007) than the CR group.No cases of reexploration for bleeding or tamponade were observed in either group.Additionally, no sternal wound infections occurred after CABG surgery in either group.Te hospital mortality rate was not signifcantly diferent between the two groups (TAR: 2 deaths, 3.85% vs. CR: 4 deaths, 6.67%, p � 0.810).To determine whether the mortality rates were clinically equivalent, an equivalence test was conducted using a 10% margin.Te 90% confdence interval for the Journal of Cardiac Surgery mortality rate diference between groups was −7.3% to 4.6%, falling entirely within the equivalence region.Te rates of early complications, such as respiratory complication, prolonged ventilation, LCOS requiring ECMO, and cardiocerebral events, were also similar between the two groups (p > 0.05 for all).

Discussion
CABG remains one of the most efective treatments for patients with multivessel CAD and LVD [5].However, the optimal revascularization strategy in this high-risk patient population is still a matter of debate [10,11].In this study, we compared the early outcomes of TAR versus CR in patients with multivessel CAD and LVD.Te main fndings 4 Journal of Cardiac Surgery of this study are as follows: (1) TAR was feasible and safe in patients with multivessel CAD and LVD, without increasing the operative risk or hospital mortality; (2) TAR was associated with shorter ICU and hospital stay, higher early graft patency rate, and lower readmission rate within the frst 6 months postprocedure than conventional CABG.Te benefts of TAR over conventional CABG have been well established in previous studies.TAR has been shown to improve long-term survival, reduce the need for repeat revascularization, and lower the incidence of cardiac events in patients with multivessel CAD [7,12].Te superior outcomes of TAR are mainly attributed to the higher patency rates of arterial grafts compared to venous grafts [13].Arterial grafts have better resistance to atherosclerosis, thrombosis, and spasm than venous grafts, and they also have better endothelial function and vasoreactivity.Moreover, arterial grafts can provide more physiological fow patterns and adapt to changes in coronary fow demand [14].Our study supports these fndings, as we observed a significantly higher graft patency rate at 3 months (91.7% vs. 83.7%,p � 0.034) and a similar freedom from angina rate within 6 months (86.0% vs. 75.0%,p � 0.240) in the TAR group compared to the CR group.
However, several concerns have limited the application of TAR in patients with LVD.TAR is more technically demanding and time-consuming than CR, especially when using multiple arterial conduits or performing of-pump surgery.Moreover, TAR may increase the risk of sternal wound infection or mediastinitis, particularly in patients with diabetes, obesity, or chronic lung disease [15].Terefore, it is important to evaluate the feasibility, safety, and efcacy of TAR in patients with multivessel CAD and LVD.To our knowledge, this is the frst study to compare the early outcomes of TAR and CR in this high-risk population.We found that TAR was feasible and safe in patients with multivessel CAD and LVD, without increasing the operative risk or hospital mortality.Te TAR group had a longer operation duration than the CR group.Te use of arterial grafts in TAR requires meticulous and time-consuming anastomosis techniques, which might contribute to the longer operation time.However, this did not translate into higher rates of postoperative complications or mortality.On the contrary, we found that TAR was associated with shorter ICU and hospital stay.Tis fnding may be attributed to the superior hemodynamic stability, lower incidence of postoperative complications, and potentially fewer woundrelated issues with the absence of venous graft harvesting sites, as suggested by previous studies [16,17].Shorter hospital stays not only reduce the economic burden on patients and healthcare systems but also have implications for patient recovery and overall satisfaction.
A notable fnding in our study was the higher graft patency rate observed in the TAR group at 3 months postsurgery than the CR group.Te use of arterial grafts in TAR might contribute to the improved graft patency rates, as arterial conduits have been associated with better long-term patency compared to venous grafts [18].Although it is challenging to draw defnitive conclusions from early graft patency data, these results are promising and warrant further investigation in long-term follow-up studies to assess the impact of graft patency on clinical outcomes.While the superior graft patency rate of TAR observed in our study is encouraging, it is important to acknowledge the lack of a signifcant diference in freedom from angina within 6 months between the two groups.Tis may be due to several reasons, such as the small sample size, the short follow-up period, and the multifactorial nature of angina symptoms in patients with LVD.Long-term studies are therefore necessary to assess the full impact of graft patency on symptom relief.
Another important observation in our study was the lower readmission rate within 6 months in the TAR group than the CR group (4.0% vs. 17.9%,p � 0.024).Reduced readmission rates could be attributed to the superior graft patency observed in the TAR group, leading to fewer ischemic events such as angina, myocardial infarction, or heart failure requiring hospitalization [19].However, it is essential to consider that readmission rates can be infuenced by various factors, such as patient compliance, lifestyle modifcations, and medical management, and thus, the true impact of TAR on readmission rates requires careful evaluation in larger, prospective studies [20,21].
However, one third of the patients in the TAR group and 25% in the CR group had incomplete revascularization.Tis rate was higher than expected, but we would like to emphasize that our decision to perform incomplete revascularization was not arbitrary, but rather based on several factors, such as technical difculties, poor distal targets, or lack of suitable conduits.Moreover, TAR was associated with a higher graft patency rate than CR (91.7% vs. 83.7%,p � 0.034), which might have mitigated some of the adverse efects of incomplete revascularization.None of our patients required repeat percutaneous or surgical intervention because most of them were asymptomatic or had mild angina at the 6-month follow-up, which might be attributed to improved medical therapy, collateral circulation, or myocardial hibernation.
Besides graft patency and angina relief, another potential beneft of TAR is the improvement of left ventricular function, which may afect the prognosis and quality of life of patients with multivessel CAD and LVD [22].We did not measure the postoperative LVEF or other parameters of left ventricular function because this study was mainly focused on some clinical outcomes, such as mortality, complications, graft patency, and angina relief.We considered these outcomes to be more relevant and important for patients with multivessel CAD and LVD who underwent CABG.However, we recognize that left ventricular function is also a crucial outcome that may afect the prognosis and quality of life of these patients [23].Terefore, we plan to investigate the efect of TAR on left ventricular recovery in our future studies.
Te mortality rates in our study were higher than those predicted by EuroSCORE II (TAR: 3.85% vs. 2.85%, CR: 6.67% vs. 3.04%).Tis discrepancy could be attributed to some factors that the EuroSCORE model did not adequately account for in this population, such as small coronary artery size, difuse coronary artery disease, and incomplete revascularization.

Limitations
It is important to note that our study has several limitations.First, this was a retrospective cohort study from a single center, which may limit the generalizability of the fndings.Te small sample size may have reduced the statistical power to detect a signifcant diference in in-hospital mortality between TAR and CR.Terefore, our study was underpowered and the lack of diference could be a result of a type II error.Multicenter, randomized controlled trials with larger sample sizes and longer follow-up durations are needed to confrm our results and assess potential long-term benefts of TAR.Second, the choice of revascularization strategy was not randomized but rather based on the surgeon's preference and patient characteristics, introducing potential selection bias.Randomized controlled trials could address this limitation and provide more robust evidence on the benefts of TAR in this patient population.Tird, we did not have data on the subtype of prior myocardial infarction (STEMI vs. NSTEMI).Tis may have provided further insights into the patients' ischemic burden and risk profles.Fourth, we did not examine if incomplete revascularization contributed to the cases of readmission for angina and heart failure.Future studies should investigate the association between completeness of revascularization and adverse cardiac events requiring hospitalization.Additionally, while the early outcomes of TAR are promising, long-term data regarding left ventricular recovery, survival, and quality of life are needed to fully understand the potential benefts of this approach in patients with multivessel CAD and LVD.

Conclusion
In conclusion, our study provides valuable insights into the early outcomes of total arterial revascularization compared to conventional CABG in patients with multivessel CAD and LVD.Both strategies demonstrated similar mortality rates and early complications, indicating the safety and efcacy of both approaches.TAR was associated with shorter ICU and hospital stay.Notably, TAR showed potential advantages in terms of graft patency and reduced readmission rates within 6 months, suggesting possible improved long-term outcomes.Further larger-scale, prospective studies with longer follow-up are warranted to validate these fndings and determine the true impact of TAR on long-term clinical outcomes in this high-risk patient population.

Table 1 :
Comparison of patients' baseline demographic and clinical characteristics.

Table 2 :
Operative data and postoperative in-hospital outcomes., total arterial revascularization; CR, conventional revascularization; CABG, coronary artery bypass grafting; OPCAB, of-pump coronary artery bypass; ONCAB, on-pump coronary artery bypass; LCOS: low cardiac output syndrome; ECMO: extracorporeal membrane oxygenation.a Non-normally distributed variables are presented as the median (interquartile range (IQR)) and categorical data as number.b Of the 4 patients in the CR group who received a single graft, 2 had a LIMA to LAD graft, and 2 had a single vein graft. TAR