Mid-term angiographic evaluation of LIMA-LAD anastomoses following LAD endarterectomy in coronary artery bypass grafting

Abstract

Objective

Diffuse coronary artery disease remains a critical issue that heart surgeons continue to research in terms of treatment options. An alternative method applied during coronary bypass surgery to achieve complete revascularization is coronary artery endarterectomy. Since the reliability of this technique and its effects on mortality and morbidity are still debated in the literature. Our objective in conducting this study is to determine the mid-term patency rates in patients who underwent LAD endarterectomy and to explore its future applicability.

Materials and Methods

This study reviewed 20 patients who underwent coronary endarterectomy during CABG in our clinic between January 2014 and December 2021. The data, including contrast imaging to check graft patency, were evaluated retrospectively by reviewing hospital archives and patient files. Patients without LAD endarterectomy were excluded from the study.

Results

LAD endarterectomy and patch-plasty were performed on all patients in the study. In 17 patients, the LIMA graft was anastomosed onto the patch applied to the LAD. It was observed that 12 of the grafts anastomosed to the LAD (60%) were patent during a mean follow-up period of 32.8 ± 25.9 months. The average 4-year survival rate was found to be 95%. Perioperative myocardial infarction occurred in 5 patients (25%). There was no need for high-dose inotropic support, IABP, or ECMO in the postoperative period.

Conclusion

Coronary artery endarterectomy should be considered a viable option for surgeons to achieve satisfactory revascularization in cases where suitable anastomosis sites on coronary arteries to ensure adequate outflow are not available. Our findings align closely with literature reports indicating that endarterectomy performed on the LAD, with LIMA used as the conduit, leads to promising outcomes.

Introduction

The primary goal of surgical interventions in coronary artery disease is the revascularization of coronary arteries with critical lesions that ensure the nourishment of viable myocardial tissue. The presence of widespread and complex coronary artery lesions poses challenges for achieving complete revascularization using both percutaneous coronary intervention (PCI) and standard coronary artery bypass grafting (CABG) techniques. Therefore, during CABG, reconstructive procedures such as coronary endarterectomy (CE) and patch-plasty are used as alternative methods to ensure adequate distal runoff. CE is a technique in which atheromatous plaque is removed from the inner wall of the coronary arteries. The arteries are then reconstructed with a patch over a wide arteriotomy to allow free blood flow.

As a surgical technique, CE was introduced before CABG surgery in the treatment of coronary artery disease [1]. However, there is no consensus on the preoperative indications for this technique. Several authors have reported using this technique in occluded vessels with multiple stenoses and distal involvement [2].

Uncertainties regarding morbidity and mortality have led to hesitation among surgeons today regarding the role of CE. Residual plaque in the Left Anterior Descending (LAD) artery, in particular, is considered the most important parameter affecting perioperative and late-term mortality [3].

The aim of our study is to determine the patency rates of left internal mammary artery (LIMA)-LAD anastomoses performed after LAD endarterectomy using angiographic methods (conventional angiography or coronary computed tomography (CT) angiography) and to explore their usability.

Materials and methods

In this study, 20 patients who underwent coronary endarterectomy and were imaged to check graft patency at a University Hospital, Department of Cardiovascular Surgery between 01.01.2014 and 31.12.2021 were examined. Patients who specifically underwent LAD endarterectomy were included in our study, and their data, including contrast imaging to check graft patency, were retrospectively reviewed using the hospital database, archive records, and patient files.

Ethics approval and consent to participate

Our study was conducted after the approval of the Local Clinical Research Ethics and Approval Committee (March 10,2023; reference number E-83045809–604.01.01–638942). Every human participant provided their informed consents in terms of involving the study. Our study was conducted in accordance with the Declaration of Helsinki.

Statistical analysis

The preoperative, perioperative, and postoperative data of the patients included in the study were obtained retrospectively. Descriptive statistics, such as mean and standard deviation values, were provided for the analyzed data, and categorical variables were presented as numbers and percentages. Survival analysis was conducted using the Kaplan–Meier survival curve.

Surgical technique

After being brought into the operating room, the patients were placed supine on the table. Five electrodes were placed on the back for electrocardiographic monitoring. An intravenous line was established with an 18 G catheter through peripheral veins. Blood pressure monitoring was performed using an 18 G or 20 G catheter inserted into the radial or brachial artery with 2 mg of midazolam (0.03 mg/kg) administered to the patient and 20 mg of lidocaine applied locally. For anesthesia induction, 0.05 mg/kg midazolam, 2 mg/kg fentanyl, 1 mg/kg propofol, and 0.6 mg/kg rocuronium were injected. Patients were ventilated with an Ambu bag during the effective muscle relaxation period, followed by orotracheal intubation. Sevoflurane was inhaled during ventilation. A 7 F central venous catheter was routinely placed in the right internal jugular vein. Patients were fitted with a bladder catheter with a temperature probe. After the draping and sterile covering procedures, a median sternotomy was performed. Based on the number of vessels to be operated on, the left internal mammary artery and the great saphenous vein grafts were prepared. Heparin was administered at 300–400 U/kg to achieve an activated coagulation time (ACT) of over 460 s. The pericardium was opened and suspended. Cardiopulmonary bypass was initiated with arterial cannulation from the ascending aorta and two-stage venous cannulation from the right atrium. Antegrade cardioplegia cannula from the aortic root and retrograde cardioplegia cannula from the coronary sinus were placed. All patients were given cold blood cardioplegia enriched with potassium initially to achieve diastolic arrest. Myocardial protection was maintained with intermittent antegrade and continuous retrograde cardioplegia throughout the surgery. Patients generally underwent mild hypothermia (30 °C – 32 °C). Distal anastomoses were performed under cross-clamp, while proximal anastomoses were performed under side clamp. After the distal anastomoses were completed, the cross-clamp was removed, and patients were gradually warmed until the bladder temperature reached 36 °C. After myocardial contraction resumed, the proximal part of the LIMA was occluded to check the LIMA-LAD anastomosis, and the pulse was palpated distally. Then, the distal part of the LIMA was occluded, and the pulse was palpated proximally. Electrolyte levels and patient temperature were normalized, and the response of the myocardium was observed during gradual weaning from the bypass. In patients who did not achieve adequate myocardial circulation and required high-dose inotropic support, cardiopulmonary bypass was resumed after re-heparinization to recheck graft patency and anastomosis quality. In patients who remained hemodynamically stable, with no increase in central venous pressure, adequate urine output, and requiring low-dose inotropic support, cardiopulmonary bypass was terminated after confirming that the myocardium had taken over circulation. The heart was decannulated, and heparin was neutralized with protamine infusion. After hemostasis, the sternum was closed with wire sutures, and the skin and subcutaneous tissue were closed anatomically, concluding the surgery.

Endarterectomy procedure

In the patients who underwent endarterectomy, the coronary arteries were of small diameter, with diffuse and critically calcified plaques, where suitable anastomosis sites to ensure adequate outflow on the coronary arteries were not available. During open endarterectomy, the arteriotomy was extended proximally and distally to reach the healthy arterial wall. This allowed for the complete removal of the plaque under direct vision, including the parts of the plaque involving the septal branches (Fig. 1). For arteriotomies longer than 1 cm, patch-plasty was applied with a vein (Fig. 2). The LIMA graft, prepared with small leaf-like incisions on the patch, was then anastomosed (Fig. 3). In all patients who underwent endarterectomy, unless postoperative drainage volume and ACT levels were excessive, protamine administration was avoided. Low-molecular-weight heparin was administered to patients six hours postoperatively as drainage decreased. Dual antiplatelet therapy (100 mg aspirin + 75 mg clopidogrel) was started from the first postoperative day, and low-molecular-weight heparin therapy was discontinued on the third postoperative day.

Fig. 1.

Coronary Artery Endarterectomy

Fig. 2.

Vena Saphena Magna (VSM) Patch on LAD

Fig. 3.

LIMA Anastomosis on VSM Patch after LAD Endarterectomy. A: LIMA; B: VSM Patch

Follow-up plan and antiplatelet therapy for patients after CE

In the early postoperative period, including intensive care unit (ICU) follow-up, patients received low-molecular-weight heparin (LMWH) at a dose of 100 IU/kg, continued through the first 3 postoperative days. Dual antiplatelet therapy [acetylsalicylic acid (ASA) + clopidogrel] was initiated as early as possible—unless contraindicated—and was prescribed at discharge for patients who underwent CE. If the patient had an indication for warfarin use, we opted for an ASA + warfarin combination instead of dual antiplatelet therapy.

During the follow-up period, conventional angiography was preferred in symptomatic cases to enable potential interventional procedures, while coronary CT angiography was used for graft evaluation in asymptomatic cases (Fig. 4).

Fig. 4.

Flow diagram of the study. Ao: Aorta; CE: Coronary endarterectomy; CT: Computed tomography; GSV: Great saphenous vein; LAD: Left anterior descending artery; LIMA: Left internal mammary artery

Results

Of the 20 patients whose data were analyzed, 3 (15%) were female, and 17 (85%) were male. The average age of the patients was 64.4 ± 9.3 years, ranging from 50 to 82 years. Eighty-five percent of the patients were over 55 years old. The body mass index (BMI) distribution of the patients ranged from a minimum of 22.4 to a maximum of 35.3, with an average of 27.6 ± 3.1. The comorbid diseases and risk factors accompanying coronary artery disease in the patients included in the study are shown in Table 1.

Table 1.

Preoperative patient characteristics

Characteristics	All Patients (N = 20)
Age (years), mean ± SD	64.4 ± 9.3
Male, n (%)	17 (85)
BMI (kg/m2), mean ± SD	27.6 ± 3.1
Hypertension, n (%)	13 (65)
Hyperlipidemia, n (%)	1 (5)
Diabetes Mellitus, n (%)	3 (15)
Nicotine Abuse, n (%)	16 (80)
Chronic Obstructive Pulmonary Disease, n (%)	2 (10)
Peripheral Vascular Disease, n (%)	3 (15)
Carotid Artery Disease, n (%)	4 (20)
Acute Kidney Failure, n (%)	2 (10)
Chronic Kidney Failure, n (%)	2 (10)
Preoperative Myocardial Infarction, n (%)	9 (45)
LV-Function, n (%)
EF ≥ 50%	16 (80)
EF = 30–50%	4 (20)

BMI body mass index, EF ejection fraction, SD standard deviation

When preoperative risk factors were examined, the most common risk factor was smoking (80%), followed by hypertension (65%) and a history of preoperative myocardial infarction (45%). Three (15%) patients were diagnosed with diabetes mellitus, and one (5%) patient had a history of hyperlipidemia. Regarding non-cardiac comorbidities, it was found that 3 (15%) patients had peripheral artery disease (PAD), 4 (20%) patients had carotid artery disease, 2 (10%) patients had chronic obstructive pulmonary disease (COPD), and 2 (10%) patients had chronic kidney failure. No patients had a history of cerebrovascular events. Preoperative echocardiographic evaluation revealed that 16 (80%) patients had a left ventricular ejection fraction (LVEF) ≥ 50%, while 4 (20%) had an LVEF between 30–50%. Only 3 (15%) patients had moderate mitral valve regurgitation (MR), and 1 (5%) had moderate aortic valve regurgitation (AR), with no severe valve pathologies detected (Table 1).

Coronary angiography reports before surgery were available for only 17 of the 20 patients included in the study (Table 2). As seen in the table, 12 (70.6%) patients did not have left main coronary artery (LMCA) stenosis, and only 3 (17.6%) patients had LMCA stenosis of less than 50%. All patients had LAD stenosis above 70%, with 4 (23.5%) having stenosis greater than 90%. Based on the target coronary vessels, 3 (15%) patients underwent double CABG, 7 (35%) underwent triple CABG, 9 (45%) underwent quadruple CABG, and 1 (5%) underwent quintuple CABG (Table 3). A total of 68 coronary artery anastomoses were performed. The anastomoses included 20 (100% of patients) LAD, 15 (75% of patients) diagonal artery, 19 (95% of patients) circumflex artery (CX), and 14 (70% of patients) right coronary artery (RCA) anastomoses (Table 3).

Table 2.

Preoperative angiography results

Stenosis	n (%)
LMCA
Normal	12 (70.6)
> 50%	2 (11.8)
< 50%	3 (17.6)
LAD
0–70%	-
71–90%	13 (76.5)
> 90%	4 (23.5)
CX
0–70%	5 (29.4)
71–90%	7 (41.2)
> 90%	5 (29.4)
RCA
0–70%	2 (11.8)
71–90%	5 (29.4)
> 90%	10 (58.8)

LMCA left main coronary artery, LAD left anterior descending artery, CX circumflex artery, RCA right coronary artery

Table 3.

Intraoperative results

Characteristics	N = 20
Total Number of Anastomotic Arteries, n
LAD, n (% of patients)	20 (100)
Diagonal Artery, n (% of patients)	15 (75)
CX, n (% of patients)	19 (95)
RCA, n (% of patients)	14 (70)
CABGx2, n (%)	3 (15)
CABGx3, n (%)	7 (35)
CABGx4, n (%)	9 (45)
CABGx5, n (%)	1 (5)
Type of Grafts, n
LIMA, n (%)	17 (25)
VSM, n (%)	51 (75)
CE Territory
LAD, n	20
RCA, n	2
Cross-clamp Time, mean ± SD	102.4 ± 29.3 min
Cardiopulmonary Bypass Time, mean ± SD	141.79 ± 44.42 min

LAD left anterior descending artery, CE coronary endarterectomy, CX circumflex artery, RCA right coronary artery, CABG coronary artery bypass grafting, LIMA left internal mammary artery, VSM vena saphena magna, SD standard deviation

Endarterectomy and patch plasty were performed on the LAD artery in all 20 patients. Only 2 (10%) patients underwent endarterectomy on the RCA artery (Table 3). A total of 68 grafts were used, 17 (25%) of which were LIMA grafts, and 51 (75%) were Vena Saphena Magna (VSM) grafts (Table 3). All LIMA grafts were anastomosed onto the LAD artery over the patch. In only 3 patients, due to the small caliber of the LIMA graft, a VSM graft was anastomosed onto the LAD artery as a long segment patch. The early postoperative data of the patients are shown in Table 4. Sixteen (80%) patients were extubated within 6–8 h postoperatively, while 4 (20%) were extubated within 12–16 h postoperatively. None of the patients required prolonged mechanical ventilator support. The total drainage amount within the first 24 h in the intensive care unit ranged from a minimum of 200 ml to a maximum of 1400 ml, with an average drainage amount of 572 ml ± 249 ml. Two (10%) patients underwent revision surgery due to bleeding in the early postoperative period.

Table 4.

Postoperative data

Outcome	n (%)
Extubation Time
6–8 h	16 (80)
12–16 h	4 (20)
Re-exploration for Bleeding	2 (10)
Peroperative Myocardial Infarction	5 (25)
IABP	0 (0)
ECMO	0 (0)
Postoperative Hemodialysis	0 (0)
Tracheostomy	0 (0)
Wound Infection	2 (10)
Sepsis	0 (0)
Respiratory Insufficiency	1 (5)
Inotropic Support
Medium Dose	2 (10)
Low Dose	18 (90)
Mean ± SD
Intensive Care Unit Stay (days)	2.6 ± 1.32
Amount of Bleeding (ml)	572.5 ± 249.24
Postoperative Hospital Stay (days)	9.75 ± 4.3
CK-MB Level (IU/L)	532.6 ± 320.7
Troponin Level (ng/ml)	0.75 ± 0.58

IABP intraaortic balloon pump, ECMO extracorporeal membrane oxygenation, CK-MB creatine kinase-myocardial band, SD standard deviation

In terms of early postoperative complications, perioperative myocardial infarction was observed in 5 (25%) patients. Two (10%) patients required medium-dose inotropic support, while 18 (90%) required low-dose inotropic support, with no patients requiring high-dose inotropic support. Additionally, none of the patients required intra-aortic balloon pump (IABP) or extracorporeal membrane oxygenation (ECMO). During the first postoperative week, 2 (10%) patients developed a wound infection at the sternum incision, which responded well to antibiotic therapy, and 1 (5%) patient developed pneumonia. None of the patients required hemodialysis, tracheostomy, or developed sepsis. The average length of stay in the ICU was 2.6 ± 1.32 days, with a minimum of 1 day and a maximum of 6 days. The total hospital stay averaged 9.75 ± 4.3 days, with the earliest discharge on the 5th postoperative day and the latest discharge on the 23rd day (Table 4).

Seventeen (85%) patients were discharged with aspirin 100 mg + clopidogrel 75 mg, and three (15%) patients were discharged with aspirin 100 mg + warfarin 2.5 mg, after receiving low-molecular-weight heparin (100 IU/kg) therapy for the first 3 days, following confirmation that there were no contraindications.

Of the patients who underwent endarterectomy, 17 (85%) underwent contrast imaging (coronary CT angiography and conventional angiography) after 12 months postoperatively, and 3 (15%) underwent contrast imaging within 12 months postoperatively to evaluate graft patency. The average imaging time was 32.8 ± 25.9 months. The earliest imaging was performed at 1 month postoperatively, and the latest at 8 years postoperatively. The imaging results showed that 12 (60%) LAD anastomoses, 15 (79%) CX anastomoses, and 13 (93%) RCA anastomoses were patent (Table 5). Endarterectomy had been performed on the LAD artery in all 20 patients. The longest patent LAD anastomosis graft was observed at 96 months (8 years) postoperatively, while the earliest occluded graft was observed at 1 month postoperatively. One of the 2 patients who underwent RCA endarterectomy was found to have an occluded anastomosis.

Table 5.

Follow-up outcomes

Outcome	n (%)
Timing of Imaging (months), mean ± SD	32.8 ± 25.9
Anastomosis
LAD, n (%)	20 (29.4)
CX, n (%)	19 (27.9)
RCA, n (%)	14 (20.6)
Graft Patency
LAD, n (%)	12 (60)
CX, n (%)	15 (79)
RCA, n (%)	13 (93)
Symptomatic Patients
Stable Angina Pectoris, n (%)	4 (20)
Acute Coronary Syndrome, n (%)	4 (20)
PCI/Stenting, n (%)	3 (15)
Re-CABG, n (%)	0 (0)
Overall Follow-up Time (months), mean ± SD	47.7 ± 23.9
Overall Survival, n (%)	19 (95)

LAD left anterior descending artery, CX circumflex artery, RCA right coronary artery, PCI percutaneous coronary intervention, CABG coronary artery bypass grafting, SD standard deviation

In terms of postoperative symptoms and findings, 4 (20%) patients developed stable angina pectoris, 4 (20%) developed acute coronary syndrome, 3 (15%) underwent percutaneous transluminal coronary angioplasty (PTCA), and none of the patients required repeat bypass surgery (Table 5). Only 1 (5%) patient included in the study died in the 4th postoperative month, while 19 (95%) patients were alive. The average survival time was 47.7 ± 23.9 months (Fig. 5).

Fig. 5.

Kaplan–Meier Survival Curve of the patients

A total of 17 cases (85%) had a LIMA anastomosis to the vein patch over the LAD CE area, while 3 cases (15%) used the GSV. Among the 17 patients with LIMA, 12 (70.6%) remained patent at a mean follow-up time of 35.65 ± 26.39 months. For the 3 patients with GSV, 1 (33.3%) remained patent at a mean follow-up time of 16.33 ± 17.90 months. These two groups showed no statistically significant difference in terms of occlusion (p = 0.270). When compared using the Kaplan–Meier event-free survival plot and the Log-rank test, there was also no statistically significant difference between the two groups (p = 0.084) (Fig. 6).

Fig. 6.

Kaplan Meier Event-Free Survival Curve of Patients. LIMA: left internal mammary artery; VSM: Vena saphena magna (great saphenous vein). *: Survival without graft occlusion in the coronary endarterectomy area

A multivariate analysis was conducted to evaluate possible variables that could affect one-year graft occlusion rates. This analysis showed no significant variables, including perioperative myocardial infarction, comorbidities, and intraoperative factors (Table 6).

Table 6.

Multivariate analysis of one-year graft patency rates

Variable	OR	(95% CI)	P value
Age ≥ 65 years old	0.917	0.068–12.322	0.948
BMI	0.737	0.438–1.239	0.250
Cross-clamp Time	1.035	0.987–1.085	0.155
Cardiopulmonary Bypass Time	1.026	0.990–1.064	0.157
Preoperative-CK-MB	0.977	0.892–1.069	0.611

BMI body mass index, CI confidence interval, CK-MB creatine kinase-myocardial band, OR odds ratio

Discussion

Diffuse coronary artery disease is one of the biggest challenges faced by both interventional cardiologists and cardiac surgeons. Treatment options for these patients are limited, and there are no specific criteria to guide the selection of an appropriate treatment. In patients indicated for coronary artery bypass surgery, coronary artery endarterectomy is applied as an alternative method to achieve complete revascularization. Although studies on this topic yield mixed results regarding the procedure’s safety and its impact on morbidity and mortality, endarterectomy can be inevitable for cardiac surgeons when performing bypass on coronary arteries with critical and multiple calcified stenoses. Several factors lead to the need for endarterectomy, including the poor quality of anastomoses performed over plaques, the risk of distal embolization when suturing through plaque, blockage of coronary artery side branches by the plaque that may render the anastomosis ineffective, and the presence of residual clinical symptoms and findings in these patients [4].

The most important questions that intrigue surgeons after coronary endarterectomy are the effects of the procedure on mortality and morbidity in these patients, the improvement in clinical symptoms during the perioperative period, the rate of myocardial infarction, graft patency rates, the impact of the chosen graft structure on patency, and the appropriate application of antiplatelet therapy in the postoperative period.

Studies comparing patients who underwent coronary endarterectomy with those who underwent standard CABG have yielded mixed results, leading to varying interpretations in the literature regarding the use of this method. In a meta-analysis conducted by Wang et al., which included 30 studies published between 1970 and 2015, it was found that patients who underwent CABG + CE had higher 30-day mortality and postoperative complication rates, and they recommended performing coronary endarterectomy only when necessary [5]. Additionally, according to the results of this meta-analysis, patients who underwent coronary endarterectomy had a higher incidence of preoperative comorbidities such as hypertension (HT), previous myocardial infarction, diabetes mellitus (DM), and peripheral vascular disease. In our study, DM was found in 15% of patients, HT in 65%, smoking in 80%, peripheral artery disease in 15%, and preoperative MI in 45%. In a study by Tiruvoipati et al., it was found that the group of patients who underwent CE had a higher mortality rate compared to those who did not undergo endarterectomy (8.6% vs. 4.6%) [6]. According to the results of a review by Soylu et al. of 30 studies, CE was associated with a significant increase in 30-day mortality and the rates of pre- and postoperative MI [7]. On the other hand, some studies with positive outcomes and low mortality rates have also been reported. For example, a study by Shapira et al., which compared 151 patients who underwent CE with a control group, reported a low mortality rate of 2% [8]. In studies conducted by Okur et al. and Nemati et al., low mortality rates of 1.5% and 1.25%, respectively, were reported for patients who underwent CE [9, 10]. As studies showing lower mortality rates and satisfactory patency rates have increased in the literature, surgeons have become more encouraged to perform CE [10–13]. In our study, the average 4-year mortality rate was 5%, which is similar to the results of studies reported in the literature. No late-term mortality was observed among the patients included in the study. In the patient who died, postoperative angiography performed in the second month after surgery revealed that the LIMA graft was patent, but there was partial stenosis at the LIMA-LAD anastomosis and occlusion of the vein graft used for the CX artery.

Another controversial issue is graft patency rates in patients who have undergone endarterectomy. Ferraris et al. reported graft patency rates of 40% in patients who underwent endarterectomy and 58% in those who did not [14]. They suggested that performing endarterectomy on a completely occluded artery might be beneficial. In a study by Nemati et al., a lower patency rate of 66.7% was reported in the endarterectomy group compared to 95% in the control group [10]. In a study by Okur et al. that compared a control group, high patency rates of 92.3% in the early period and 82.6% in the late period were reported in the CE group [9]. Takanashi et al., in their study of patients who underwent long arteriotomy and endarterectomy with patch plasty on the LAD, reported an early patency rate of 91.2% [15]. Several factors are suggested to influence graft patency in patients who have undergone CE. For example, the characteristics of the artery undergoing endarterectomy, the use of IMA, the technique of endarterectomy, the diameters of the coronary arteries, and the choice of postoperative antiplatelet therapy can all affect the outcome [11, 12, 16, 17]. Shapira et al. reported that performing endarterectomy on coronary arteries with an external diameter of less than 1.5 mm may lead to early occlusion [16]. There are several studies showing the relationship between high patency rates and endarterectomies performed on the LAD. According to some studies, the use of IMA after CE is considered an important factor, and there is a relationship between the anastomosis of IMA to the LAD after endarterectomy and high patency rates [12, 18]. This is attributed to the contribution of anti-atherosclerotic and anti-vasospastic factors, such as nitric oxide and prostacyclin, released from the endothelium when using IMA, to high patency rates, even if distal blood flow is weak [9, 17]. There are also various studies in the literature on the impact of the endarterectomy technique on graft patency. As previously mentioned, endarterectomy can be performed using open or closed methods. In a study by Nishi et al. that compared 127 patients who underwent open and long endarterectomy with IMA anastomosis with those who underwent closed endarterectomy, the graft patency rate was 89.1% in the open endarterectomy group and 81% in the closed endarterectomy group after 21 months of follow-up [19]. In a study evaluating mortality, Soylu et al. found more favorable early mortality outcomes in the open endarterectomy group compared to the closed endarterectomy group [7]. In a study published by Efem et al. in 2022, the graft patency rate after coronary endarterectomy was reported as 73.4% [20]. In our study, open endarterectomy was performed in all 20 patients, and the graft patency rate for the LAD was found to be 60%, which is in line with the average rate reported in recent studies and meta-analyses. Endarterectomy was performed on the RCA in only 2 patients, and one of them was found to have an occluded graft. The grafts anastomosed to 12 RCA arteries that did not undergo endarterectomy were found to be patent. A previous study found no statistically significant difference between the outcomes of coronary endarterectomy performed on the LAD and those on other vessels [21]. However, in contrast to these findings, our study primarily focused on cases involving LAD endarterectomy, so a direct comparison with other vessels was not possible. Hypothetically, however, LAD endarterectomy offers the advantage of providing a larger run-off area through opened septal regions, and the additional benefit of using a LIMA graft in this area should also be considered. Therefore, better outcomes are generally expected following LAD endarterectomies. Supporting this hypothesis, our study found graft occlusion in half of the RCA endarterectomies performed in addition to LAD endarterectomy. Although the total number of cases was limited for drawing definitive conclusions, this finding aligns with the primary expectation. A recent study also supports our findings that LAD endarterectomy had better outcomes when compared with other locations [22].

In evaluations of the average follow-up period similar to those in studies in the literature, it was observed that our study also includes a comparable follow-up duration [21]. Previous studies have identified an operative and perioperative mortality risk of approximately 1.5% [21]. In this study, there was no significant difference in event-free survival between on-pump and off-pump cases following coronary endarterectomy. In our study, all cases were performed on-pump. Contrary to some studies in the literature claiming that off-pump procedures yield better outcomes [21], our findings align more closely with studies showing no difference [22]. Furthermore, in our study, the average follow-up period for patients who underwent coronary endarterectomy was 32.75 ± 25.89 months. This relatively limited follow-up period may not capture long-term graft dysfunction at later stages. However, while there is a clear advantage in increasing run-off in the artery bed where the graft is anastomosed and enabling perfusion of all septal areas, the removal of the entire endothelial surface may hypothetically increase the risk of early thrombosis and graft failure—at least until re-epithelialization occurs. We believe that if the disadvantages of the early period can be managed without mortality or morbidity, the long-term risk of graft failure may not necessarily be higher than with other bypass grafts. Nonetheless, further studies with extended follow-up data are needed to reach a definitive conclusion on this matter.

An additional point that warrants discussion is the type of graft used in the CE area and over the patch. In our study, 17 cases (85%) had a LIMA anastomosis to the vein patch over the LAD CE area, while 3 cases (15%) used the GSV. Among the 17 patients with LIMA, 12 (70.6%) maintained patency at a mean follow-up period of 35.65 ± 26.39 months. In contrast, only 1 of the 3 patients with GSV (33.3%) remained patent, with a mean follow-up period of 16.33 ± 17.90 months. The difference in occlusion rates between these two groups was not statistically significant (p = 0.270). Similarly, Kaplan–Meier event-free survival analysis and the Log-rank test showed no significant difference between the groups (p = 0.084). Although patency rates appeared to vary by follow-up duration, favoring the use of LIMA, this difference did not reach statistical significance due to the limited number of cases in our study. We believe this is an important point to highlight. While the difference was not statistically significant, our study showed a trend favoring LIMA over GSV in graft patency when used in the endarterectomy area. These results further reinforce the expectation that LIMA anastomosis to the LAD endarterectomy area could yield more favorable outcomes. Qui Z et al. also claimed that LIMA graft had advantages over GSV use in the CE location in terms of event-free survival rates and outcomes [22].

When our patients were evaluated in terms of perioperative MI and the need for high mechanical and inotropic support, the results of the meta-analysis by Soylu et al. indicated that endarterectomy significantly increased the risk of perioperative and postoperative MI, as well as the need for mechanical support and inotropes [7]. The incidence of postoperative MI reported in other studies ranges from 5 to 25% [23]. In a study by LaPar et al., which compared patients who underwent CABG + CE with those who underwent isolated CABG between 2003 and 2008, the incidence of prolonged mechanical ventilator support and extended ICU stay was found to be higher in the endarterectomy group [24]. In our study, as noted in the literature, postoperative clinical symptoms and findings (stable angina pectoris 20%, acute coronary syndrome 20%, PTCA 15%) were observed at a high rate in patients who underwent endarterectomy. However, we obtained more favorable results than those reported in studies regarding the need for high-dose inotropic support, IABP or ECMO, prolonged mechanical ventilator support, extended ICU stay, and hospitalization duration.

Mortality, graft patency, and postoperative complications in patients who have undergone coronary endarterectomy should not be solely attributed to the endarterectomy itself. Multivariate analyses in studies reported in the literature have shown that being over 65 years of age significantly increases the risk for event-free survival during follow-up after coronary endarterectomy, while other parameters have no effect. In our study, however, this factor did not present a significant risk increase in parallel with some literature findings [22].

Currently, there is no unified guideline on the use of antiplatelet or anticoagulation therapy in patients undergoing CE. Although most authors have followed different anticoagulation regimens based on institutional protocols, the reported differences in bleeding and mortality are generally clinically insignificant. The anticoagulation regimens used by different authors vary. LaPar DJ et al. reported using dual antiplatelet therapy for at least 3 months in endarterectomy patients, observing a postoperative myocardial infarction rate of 1% and an operative mortality rate of 4% [24]. Schmitto JD et al. initiated heparin infusion 4 h after surgery if bleeding was ≤ 50 mL/hour and started patients on 100 mg aspirin daily from the first postoperative day [25]. Marzban M et al. fully reversed heparin at the end of surgery and began heparin six hours after ICU admission if bleeding was ≤ 100 mL/hour, followed by warfarin for 2–3 months [International normalised ratio (INR): 2.5–3.5], with a reported hospital mortality rate of 5% [26]. Kumar S et al. administered 300 mg aspirin rectally six hours after the procedure, followed by 75 mg aspirin and clopidogrel daily from the first postoperative day. Clopidogrel was continued for six months, and aspirin for life. Patients with atrial fibrillation and/or a mechanical valve received warfarin and aspirin, with a reported mortality rate of 9.8% [27]. Today, the combination of clopidogrel and aspirin is generally considered sufficient. It has been reported that continuing dual antiplatelet therapy for 3 months to a year, followed by single antiplatelet therapy, is effective in maintaining graft patency. Therefore, there is no standard treatment protocol for thromboprophylaxis in patients undergoing CE. However, some studies have found that dual antiplatelet therapy with aspirin and clopidogrel is sufficient [28]. It has been reported that continuing dual antiplatelet therapy for 3 months to 1 year, followed by single antiplatelet therapy, is effective in maintaining graft patency [11]. In our clinic, we prescribed dual antiplatelet therapy to patients who underwent endarterectomy at the time of discharge. Aspirin and clopidogrel were recommended for 17 patients, while aspirin and warfarin were recommended for 3 patients. Some patients were transitioned to single antiplatelet therapy after 6 months, while others transitioned after 1 year.

In cardiovascular surgery practice, CE is generally not a first-choice procedure during CABG unless absolutely necessary, as it is previously reported as to be associated with risks of early thrombosis and myocardial infarction. However, in cases where it is indicated, it can be a life-saving intervention with promising outcomes. Contrary to early studies, more recent studies indicate that it can yield favorable results [22]. The data obtained in our study align with these findings.

Limitations of the study

The main limitations of our study were its single-center, retrospective design, the limited number of cases, and the absence of a control group of patients who underwent CABG without endarterectomy. Additionally, the relatively short follow-up period may not capture long-term graft dysfunction in later stages. Therefore, it would be more appropriate to design a study that includes both patients who underwent endarterectomy and a control group of patients who underwent standard CABG to obtain more reliable results.

Conclusion

The approach to diffuse coronary artery disease remains a topic of ongoing research as there is no clear treatment protocol that cardiac surgeons must follow. Since the primary goal of coronary artery bypass surgery is to fully revascularize the blocked coronary arteries, different approaches have been proposed to determine the most appropriate treatment options for these patients. Coronary endarterectomy has emerged in recent years as a surgical option that has piqued the interest of cardiac surgeons, and research on this technique continues from various perspectives.

In our study, conducted to evaluate the outcomes of our clinic, although the number of cases was limited and we achieved an average graft patency rate, the results were found to be satisfactory in terms of both mortality and postoperative complications. Based on the findings of recent studies on this subject, we believe that coronary endarterectomy is an acceptable option for patients who require it to ensure adequate blood circulation to the myocardium.

In conclusion, in suitable patient groups, especially in cases where suitable anastomosis sites on coronary arteries to ensure adequate outflow are not available, particularly when faced with blocked coronary arteries supplying a large myocardial area, CE should be considered a viable option for surgeons with promising outcomes. Our findings align closely with literature reports [22] indicating that CE performed on the LAD, with LIMA used as the conduit, leads to better outcomes.

Abbreviations

ACT

Activated coagulation time

Ao

Aorta

AR

Aortic valve regurgitation

ASA

Acetylsalicylic acid

BMI

Body mass index

CABG

Coronary artery bypass grafting

CE

Coronary endarterectomy

CI

Confidence interval

COPD

Chronic obstructive pulmonary disease

CK-MB

Creatine kinase-myocardial band

CT

Computed tomography

CX

Circumflex artery

DM

Diabetes mellitus

ECMO

Extracorporeal membrane oxygenation

EF

Ejection fraction

HT

Hypertension

IABP

Intra-aortic balloon pump

ICU

Intensive care unit

INR

International normalised ratio

GSV

Great saphenous vein

LAD

Left anterior descending

LIMA

Left internal mammary artery

LMCA

Left main coronary artery

LMWH

Low-molecular-weight heparin

LVEF

Left ventricular ejection fraction

MR

Mitral valve regurgitation

OR

Odds ratio

PAD

Peripheral artery disease

PCI

Percutaneous coronary intervention

PTCA

Percutaneous transluminal coronary angioplasty

RCA

Right coronary artery

SD

Standard deviation

VSM

Vena saphena magna

Authors’ contributions

F.Z. primarily contributed in all of the steps of the study and writing processes of the manuscript including designing the study, analysing the data, conducting all of the processes. O.O.B. and contributed in analysing the data, conducting all of the processes of the study and writing the manuscript, preparing tables and figures, revised the critical points in terms of content of the manuscript. All of the authors contributed all steps of the study and writing the final manuscript at varying rates and all of them approved final version of the manuscript. OOB is the guarantor as the conductor of the study.

Funding

None.

Data availability

The corresponding author can be contacted to obtain the anonymous dataset used and/or analyzed during the current study upon reasonable request.

Declarations

Ethics approval and consent to participate

Our study was conducted after the approval of the Local Clinical Research Ethics and Approval Committee (March 10,2023; reference number E-83045809-604.01.01-638942). Every human participant provided their informed consents in terms of involving the study. Our study was conducted in accordance with the Declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.