검색
검색 팝업 닫기

Advanced search

Article

Split Viewer

J Chest Surg 2025; 58(1): 21-30

Published online January 5, 2025 https://doi.org/10.5090/jcs.24.080

Copyright © Journal of Chest Surgery.

Endoscopic-Assisted Multivessel Off-Pump Coronary Artery Bypass Grafting: Experience of the First 100 Procedures

De Qing Görtzen, B.Sc., Fleur Sampon , M.D., Naomi Timmermans , M.D., Joost Ter Woorst , M.D., Ph.D., Ferdi Akca , M.D., Ph.D.

Department of Cardiothoracic Surgery, Catharina Hospital, Eindhoven, The Netherlands

Correspondence to:Ferdi Akca
Tel 31-40-239-8680
E-mail ferdi.akca@catharinaziekenhuis.nl
ORCID
https://orcid.org/0000-0002-1748-3235

Received: August 1, 2024; Revised: September 5, 2024; Accepted: September 26, 2024

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Commentary: J Chest Surg. 2025;58(1):31-33 https://doi.org/10.5090/jcs.24.110

Background: This study presents an overview of our technique and the perioperative outcomes for the first 100 patients who underwent minimally invasive endoscopic-assisted off-pump multivessel bypass grafting (endoscopic coronary artery bypass [endo-CAB]) at the Catharina Hospital in Eindhoven.
Methods: The first 100 patients undergoing multivessel endo-CAB from May 2022 to March 2024 were included in this retrospective, single-center, observational study (N=100). The study encompassed both elective and urgent surgical revascularization. In all cases, endoscopic-assisted harvesting of the internal mammary artery, radial artery, or saphenous vein was performed, followed by beating-heart anastomoses through a mini-thoracotomy.
Results: A total of 226 distal anastomoses were performed, utilizing 102 left internal mammary arteries, 80 radial arteries, 30 right internal mammary arteries, and 14 saphenous veins. On average, each patient had 2.3 anastomoses. A Y graft configuration was employed in 78 patients, in-situ bilateral internal mammary artery inflow in 19 patients, and a proximal aortic graft in 3 patients. Four patients underwent concurrent arrhythmia surgery. Eleven patients received hybrid revascularization. There was 1 conversion to sternotomy (1%) and 3 instances where cardiopulmonary bypass was required (3%). The median operation time was 3.3 hours (interquartile range, 3.0–3.7 hours), and the median hospital stay was 4.0 days (interquartile range, 3–4 days). The in-hospital mortality rate was 1%.
Conclusion: Multivessel off-pump endo-CAB surgery can be safely performed with endoscopic-assisted conduit harvesting. Combining the benefits of a minimally invasive and anaortic approach may improve perioperative outcomes for patients requiring surgical revascularization. Further studies are necessary to establish the role of this technique in routine coronary surgery.

Keywords: Endoscopic coronary artery bypass grafting, Off-pump coronary bypass grafting, Minimally invasive cardiac surgery, Minimally invasive direct coronary artery bypass

The standard of care for managing advanced coronary artery disease is coronary artery bypass grafting (CABG). As the global burden of cardiovascular disease continues to rise, CABG remains the most commonly performed procedure in cardiac surgery [1]. CABG is conventionally performed through a full sternotomy on a non-beating heart, utilizing cardiopulmonary bypass (CPB). However, off-pump coronary artery bypass grafting (OPCAB) has become increasingly popular due to its lower risk of cerebrovascular accidents. This benefit arises because the technique avoids manipulation of the aorta, a method referred to as the “anaortic approach” [2-4].

Minimally invasive coronary surgery is a rapidly evolving and technically demanding field, especially for multivessel revascularization. Using minimally invasive techniques, a median sternotomy is avoided in favor of a mini- thoracotomy or total endoscopic techniques [5]. Various methods have been developed for both conduit harvesting—including direct vision, thoracoscopic-assisted, or robotic-assisted mammary artery harvest—and graft deployment, which can be performed off-pump, on a beating heart, or on an arrested heart [6-12]. The ideal revascularization strategy would avoid sternotomy entirely and achieve complete revascularization using an anaortic approach, thus providing a truly minimally invasive long-term solution for the patient.

We aimed to share our experience with minimally invasive, multivessel, endoscopic-assisted, off-pump coronary surgery (endoscopic coronary artery bypass [endo-CAB]) and provide a comprehensive overview of the perioperative data and outcomes for the first 100 patients undergoing this procedure at the Catharina Hospital in Eindhoven.

Ethical statement

The institutional review board and the ethical committee of Medical Research Ethics Committees United approved the study and waived the requirement for informed consent (registration no., W23.029).

Patient inclusion

From May 2022 to March 2024, 100 patients undergoing multivessel endo-CAB were included in this retrospective, single-center, observational study at Catharina Hospital in Eindhoven. All procedures were performed by a dedicated minimally invasive off-pump coronary surgeon. Patient baseline characteristics were obtained from the Department of Cardiothoracic Surgery’s database. Our aim was to provide a comprehensive overview of our initial experience and procedural parameters, focusing on the first 100 patients treated with multivessel endo-CAB. The patients had either double (n=87) or triple (n=13) vessel coronary artery disease. For 11 patients, a hybrid revascularization strategy was determined by the heart team, which included a percutaneous intervention alongside the multivessel endo-CAB procedure.

Patient selection and preoperative screening

All patients were evaluated by a heart team, which included a cardiac surgeon and an interventional cardiologist, to assess the need for surgical revascularization. Patients diagnosed with multivessel disease and deemed suitable for either elective or urgent surgical revascularization were considered eligible for endo-CAB. Initially, during the implementation phase of endo-CAB, patients with a lower body mass index (BMI) and double vessel disease were specifically chosen for this procedure. Morbid obesity and a salvage situation involving hemodynamic instability were established as definitive exclusion criteria.

All patients underwent the standard preoperative workup required for any coronary surgery procedure, which included coronary angiography, chest radiography, electrocardiogram, transthoracic echocardiogram, bilateral blood pressure measurements, and blood panel laboratory tests. Following the preoperative workup, we determined whether a patient was eligible for standalone multiple-vessel endo- CAB, hybrid revascularization, or concomitant arrhythmia surgery. Routine preoperative pulmonary function tests or computed tomography scans were not included in our preoperative workup.

Patient positioning and preparation

The procedure is performed under general anesthesia with the patient in a supine position. To improve surgical access, a pillow is placed under the left scapula, elevating the left hemithorax. Standard arterial and central venous lines, along with an endotracheal tube, are positioned before the procedure begins. A selective endobronchial blocker (Teleflex, Wayne, PA, USA) is inserted through the endotracheal tube to enable unilateral ventilation during the grafting process. Additionally, the groin is marked and draped in preparation for potential femoral cannulation if hemodynamic instability occurs. If the radial artery is planned to be harvested endoscopically at the same time, the right arm is positioned at a 90° angle from the thorax. Furthermore, transesophageal echocardiography is utilized to evaluate left ventricular and valve function.

Conduit harvesting

Three small incisions are made on the left side of the chest: 1 at the third intercostal space at the anterior axillary line, and 2 more at the second and fourth intercostal spaces, positioned more anteriorly. Through these incisions, 5-mm endoscopic ports are inserted, and capnothorax is established at a pressure of 8 to 12 mm Hg (Fig. 1A). A 5-mm, 0°, 2-dimensional camera (Karl Storz GmbH, Tuttlingen, Germany) is then introduced. Using the other ports, standard long-shafted video-assisted thoracic surgery instruments and a Ligasure Maryland device (Medtronic, Dublin, Ireland) are also introduced. This setup enables harvesting of both the left internal mammary artery (LIMA) and the right internal mammary artery (RIMA).

Figure 1.(A) The setup for thoracoscopic left internal mammary artery (LIMA) harvesting. (B) The needle guided marking for mini-thoracotomy by Ohtsuka et al. [24]. (C) Ligasure-aided LIMA harvest. (D) The ramus descendens posterior visualized on the video monitor endoscopically.

Before harvesting the mammary artery, the pericardium is initially opened anterior to the left phrenic nerve to identify and assess coronary targets. A clip is placed near the grafting site on the left anterior descending artery (LAD) to facilitate easy identification through the small thoracotomy. When harvesting the bilateral mammary arteries (BIMAs), both the anterior mediastinum and the right pleural space are opened. The RIMA is harvested first, following the opening of the endothoracic fascia. Utilizing long-shafted video-assisted thoracic surgery instruments and the Ligasure device, the RIMA is harvested along with its adjacent veins in a semi-skeletonized fashion. The RIMA can be harvested proximally beyond the subclavian vein and distally up to its bifurcation. Chest wall deformities, such as pectus excavatum, can complicate harvesting the RIMA from the left side. A right thoracoscopic approach or opting for a radial artery can mitigate this issue. Once the RIMA is harvested to the desired length, the LIMA is then harvested.

Thoracoscopically, the phrenic nerve can be easily identified. The LIMA is harvested from the proximal area up to the level of the phrenic nerve and extends distally to its bifurcation. Any side branches of the LIMA are ligated using the Ligasure device (Fig. 1C). Once the LIMA has been fully dissected, heparin is administered intravenously to achieve an activated clotting time of over 300 seconds. The distal portion of the LIMA is secured with titanium clips (Endo-Clip; Medtronic). The LIMA is then sutured to the pericardium using a 5-0 polypropylene stitch. After dividing the LIMA, the location for the mini-thoracotomy is marked. Intrathoracic pressure is relieved to accurately determine the location of the coronary target without the influence of artificial capnothorax pressure. A transthoracic needle is used to puncture the intercostal space above the primary coronary target, ensuring an optimal placement for the mini-thoracotomy (Fig. 1B). The precise location may vary depending on the patient’s specific anatomy.

When a radial artery is used as a second arterial conduit, the LIMA and radial artery are harvested simultaneously to achieve complete arterial revascularization. Endoscopic radial artery harvesting was performed following the methods described by Reyes et al. [13] and Connolly et al. [14], utilizing an endoscopic vessel harvesting system (Vasoview Hemopro; Getinge, Gothenburg, Sweden). Similarly, endoscopic techniques were employed for saphenous vein harvesting.

Mini-thoracotomy

Once all conduits have been harvested and the mini-thoracotomy site has been marked, the endoscopic ports are removed, and unilateral ventilation is initiated. The 5–6 cm incision is made at the previously marked location, between the third and fifth intercostal spaces. For grafting the LAD and right descending posterior, we generally adopt a more medial approach. If grafting of lateral branches is required, a more lateral access must be established. In female patients, the incision is preferably made in the inframammary area for aesthetic reasons. When the mini-thoracotomy is correctly positioned, the LIMA will be attached to the pericardium directly beneath the incision. Fig. 2A illustrates the LIMA as visible through the thoracotomy.

Figure 2.(A) The harvested left internal mammary artery (LIMA) in situ through a mini-thoracotomy. (B) A Y-graft through a mini-thoracotomy. (C) The LIMA-radialis Y-graft on the Octopus cardiac stabilizer. (D) Cardiac positioning with the wet sponges.

Graft configurations

During multivessel procedures, various graft configurations may be employed. These include both BIMA in situ grafts and Y- or T-graft configurations using a free-RIMA or radial artery. Fig. 2A–C illustrate a T-graft comprising the LIMA and radial artery performed through a mini- thoracotomy. If a saphenous vein is utilized, the proximal anastomosis is created on the ascending aorta using techniques previously described by Issa and Ruel [10] and McGinn et al. [11]. Traction sutures are applied to the right lateral pericardium, and a sponge is positioned adjacent to the aorta. The Octopus Nuvo (Octopus Nuvo Tissue Stabiliser; Medtronic) is employed to depress the pulmonary artery, increasing maneuverability.

Cardiac positioning for distal anastomosis

For cardiac positioning, we use the epicardial stabilizer and the heart positioning device (Starfish Evo; Medtronic), as previously described by Davierwala et al. [9], Issa and Ruel [10], and Kikuchi and Endo [15]. Depending on the coronary targets, we either use both devices or only the epicardial stabilizer. The heart is positioned with wet sponges, and the stabilizers are placed as shown in Fig. 2D. For the LAD and diagonal branches, only the epicardial stabilizer is used, as depicted in Fig. 3A. For the anterolateral, lateral, inferolateral, and inferior anastomoses, the armless Starfish Evo is also utilized, as illustrated in Fig. 3B and C. The armless Starfish Evo is employed to elevate the apex, facilitating access to the coronary targets, and a silk 1-0 suture is tied around the Starfish to assist in positioning. For the inferior wall, we elevate the apex of the heart with wet sponges and position the armless Starfish towards the apex. We often loosen our pericardial sutures, as they may limit visibility and maneuverability of the heart. We pull the Starfish cranially to stabilize the position for exposing the right posterior descending coronary artery. Subsequently, we position the epicardial stabilizer to secure a stable field for the coronary anastomosis. Once the heart is positioned, we ensure cardiac stability before continuing to manipulate the coronary artery. In cases of hemodynamic instability, we first elevate the legs to increase venous inflow and reposition the heart to stabilize it. If hemodynamic instability or ventricular arrhythmias persist, femoral CPB support is initiated. When using CPB, anastomoses are performed on the beating heart using similar techniques.

Figure 3.(A) The left internal mammary artery-left anterior descending anastomosis process between the Octopus cardiac stabilizer. (B) The Octopus and Starfish cardiac stabilizers for the margo obtusus grafting. (C) The ramus descendens posterior anastomosis. (D) The esthetic result.

Closing-up and flow measurements

Once all coronary targets had been grafted, protamine was administered. The flow through the graft was measured using transit-time flow measurement (TTFM) to ascertain graft function (Medistim Vascular TTFM; Medistim, Oslo, Norway). After confirming that the graft function was sufficient, a 28F chest tube was inserted into the left pleural cavity through 1 of the 3 incisions previously made for the LIMA harvest. Ropivacaine was administered at 3 intercostal levels, and the skin was then closed (Fig. 3D).

Concomitant arrhythmia surgery

Patients were considered for concomitant arrhythmia surgery if they presented with symptomatic paroxysmal or persistent atrial fibrillation (AF) and had a left atrial volume index of less than 70 mL/m2. A detailed account of our criteria for patient selection and the techniques used in concomitant thoracoscopic arrhythmia surgery has been previously published [16]. For patients with a high BMI, the primary surgical objective was to achieve minimally invasive coronary revascularization. If limited intrathoracic space increased the risk of complications during concomitant pulmonary vein isolation, only the left atrial appendage (LAA) clip (AtriClip; Atricure, Mason, OH, USA) would be applied to mitigate the risk of stroke.

Patient postoperative care

All patients were transferred to the intensive care unit (ICU) after the procedure. Patients with radial artery grafts were administered calcium antagonists postoperatively to prevent graft spasms. Paracetamol, oral opioids, and intravenous nonsteroidal anti-inflammatory drugs were utilized for postoperative pain management. Following discharge, patients attended a follow-up appointment at the outpatient clinic 14 days later, where they underwent chest radiography and laboratory blood tests.

Data collection and statistics

The data was analyzed using JASP ver. 0.18.1.0 (Universiteit van Amsterdam, Amsterdam, Netherlands) to obtain descriptives, boxplots, and Q-Q plots, identifying outliers for each continuous variable. Categorical variables are presented as frequency and percentage, while continuous variables are shown as mean and standard deviation if they are normally distributed, and as median and interquartile range (IQR) if they are not, based on the Q-Q plot analysis. RStudio ver. 4.4.0 (R Core Team, Boston, MA, USA) was utilized to visualize data and compute the Pearson correlation coefficient.

Demographics, procedural parameters, and perioperative characteristics

The patient demographics, procedural parameters, and perioperative characteristics are displayed in Table 1. Most of the patients were men (90.0%) and the mean age was 65.5±9.1 years. The mean BMI was 26.7±3.3 kg/m2, and the median body surface area was 2.00 m2 (IQR, 1.91–2.10 m2). Seventeen percent of the patients (17.0%) had diabetes, 7 patients had peripheral artery disease (7%), and 5 patients had AF (5%). The median Euroscore II for the population was 1.00 (IQR, 0.80–1.61).

Table 1. Demographics and perioperative characteristics of the endoscopic coronary artery bypass group (N=100)

CharacteristicValue
Sex (men)90 (90.0)
Age (yr)65.5±9.1
Height (cm)175 (171–180)
Weight (kg)82.3±12.8
Body mass index (kg/m2)26.7±3.3
Body surface area (m2)2.00 (1.91–2.10)
Diabetes17 (17.0)
Peripheral vascular disease7 (7.0)
Atrial fibrillation5 (5.0)
Previous cardiac surgery0
Recent myocardial infarction28 (28.0)
Prior percutaneous coronary intervention33 (33.0)
Left ventricular ejection fraction
Good (<50%)78 (78.0)
Moderate (31%–50%)18 (18.0)
Poor (21%–30%)3 (3.0)
Very poor (<20%)1 (1.0)
Chronic obstructive pulmonary disease4 (4.0)
Pulmonary hypertension0
Cerebrovascular accident6 (6.0)
Renal impairment5 (5.0)
Dialysis0
Level of urgency
Elective55 (55.0)
Urgent44 (44.0)
Emergency1 (1.0)
Euroscore II1.00 (0.80–1.61)
Preoperative hemoglobin (mmol/L)9.1 (8.3–9.6)
Double vessel disease87 (87.0)
Triple vessel disease13 (13.0)
Concomitant arrhythmia surgery4 (4.0)
Isolated LAA clip3 (3.0)
Thoracoscopic AF ablation including LAA clip1 (1.0)
No. of distal anastomoses
275 (75.0)
324 (24.0)
41 (1.0)
No. of anastomoses with left IMA
196 (96.0)
23 (3.0)
No. of anastomoses with right IMA
126 (26.0)
22 (2.0)
No. of anastomoses with radial artery
142 (42.0)
219 (19.0)
No. of anastomoses with SVG
16 (6.0)
24 (4.0)
Y-graft configuration78 (78.0)
Bilateral internal mammary artery in situ19 (19.0)
Proximal aorta3 (3.0)
Hybrid revascularization11 (11.0)
PCI, circumflex artery5 (5.0)
PCI, right coronary artery6 (6.0)
Conversion to sternotomy1 (1.0)
CPB-assisted3 (3.0)
Operation time (hr)3.3 (3.0–3.7)

Values are presented as frequency (%) for categorical variables; for continuous variables, they are shown as mean±standard deviation if normally distributed, or as median (interquartile range) if non-normally distributed.

LAA, left atrial appendage; AF, atrial fibrillation; IMA, internal mammary artery; SVG, saphenous vein graft; PCI, percutaneous coronary intervention; CPB cardiopulmonary bypass.



Four patients underwent concomitant arrhythmia surgery. Three underwent isolated LAA clip placement and 1 patient underwent thoracoscopic AF ablation, including an LAA clip. The patients received an average of 2.3 distal anastomoses. The total number of grafts per conduit is plotted in Fig. 4A. The first choice of conduit was the LIMA with 102 anastomoses, while the radial artery was used for 80 anastomoses, the RIMA for 30 anastomoses, and the saphenous vein for 14 anastomoses. A Y-graft configuration was used in 78 patients (78%), the BIMAs in situ were utilized in 19 patients (19%), and a proximal aorta graft was placed in 3 cases (3%) (Fig. 4B). Coronary anastomoses were performed on the anterior (n=100), anterolateral (n=42), lateral (n=65), inferolateral (n=5), and inferior wall (n=14) (Fig. 4C, Table 2). In our population, 11 patients underwent hybrid revascularization (11%). In 1 patient, a conversion to sternotomy was performed due to limited visibility (1%), and 3 cases needed CPB support (3%). In 1 patient, this was planned preoperatively, while in 2 patients, CPB was used due to hemodynamic instability. The median operation time was 3.3 hours (IQR, 3.0–3.7 hours). The Pearson correlation coefficient was calculated to illustrate a possible learning curve in Fig. 5. The Pearson correlation was calculated to be r(98)=-0.109 (p=0.279).

Figure 4.(A) The number of distal anastomoses for each conduit. (B) The used graft configurations. (C) The number of anastomoses for each myocardial territory. LIMA, left internal mammary artery; RIMA, right internal mammary artery; BIMA, bilateral internal mammary artery.

Figure 5.The operative time in minutes per procedure since implantation. The Pearson correlation coefficient has been calculated and visualized for the operative time since implementation.

Table 2. Graft locations in the endoscopic coronary artery bypass group

LocationNo.
Anterior100
Anterolateral42
Lateral65
Inferolateral5
Inferior14


Postoperative characteristics and complications

Table 3 describes the postoperative characteristics and complications. The total ventilation time was 7.6 hours (IQR, 6.6–9.3 hours), and the total postoperative blood loss was 350 mL (IQR, 200–550 mL). The median stay in the ICU was 1.0 days (IQR, 0.5–1 days), and the total postoperative hospital stay was 4 days (IQR, 3–4 days). In-hospital mortality occurred in 1 patient (1%) due to shock caused by a postoperative stress response due to previously undiagnosed cancer metastases. One patient required postoperative re-exploration for bleeding (1%), and 1 patient had postoperative ischemia requiring revision of the bypass graft (1%). One patient had a pulmonary embolism that required oral anticoagulation. There was no occurrence of stroke. During hospital admission, 7 patients developed new-onset AF that required oral anticoagulation (7%).

Table 3. Postoperative characteristics of the endoscopic coronary artery bypass group (N=100)

VariableValue
Total ventilation time (hr)7.6 (6.6–9.3)
Postoperative blood loss (mL)350 (200–550)
Intensive care unit admission time (day)1.0 (0.5–1)
Total hospital stay (day)4 (3–4)
In hospital mortality1 (1.0)
Intra-aortic balloon pump0
Extracorporeal membrane oxygenation0
CK-MB max (ng/mL)25.5 (19.0–35.3)
Discharge hemoglobin (mmol/L)7.7 (6.8–8.3)
Transfusion6 (6.0)
Re-exploration for bleeding1 (1.0)
Ischemia postoperative1 (1.0)
Graft revision1 (1.0)
Acute percutaneous coronary intervention0
Cerebrovascular accident0
Tamponade during initial admission0
Tamponade after discharge1 (1.0)
Pulmonary embolism1 (1.0)
Wound infection0
New-onset atrial fibrillation7 (7.0)
Hospital-acquired pneumonia3 (3.0)
Thoracentesis3 (3.0)
30-day readmission2 (2.0)
Target vessel reintervention3 (3.0)

Values are presented as number (%) for categorical variables and median (interquartile range) for continuous variables if non-normally distributed.

CK-MB, creatine kinase-myoglobin binding.



During follow-up, 1 patient developed late tamponade (1%) that required pericardiocentesis and 3 patients had left-sided pleural effusion that required thoracentesis (3%). Two patients were readmitted within 30 days after discharge for symptomatic postoperative anemia (n=1) and dyspnea due to pleural effusion requiring thoracentesis (n=1). Three patients received unexpected target vessel reintervention (3%) during follow-up.

In this study, we described the outcomes of our first 100 endoscopic-assisted, off-pump, multivessel coronary bypass grafting procedures. All patients underwent endoscopic-assisted graft harvesting and received coronary bypasses via a mini-thoracotomy. On average, each patient had 2.3 distal anastomoses. Eleven patients underwent hybrid revascularization. There was 1 conversion to sternotomy and 3 instances where CPB was required. The median operation time was 3.3 hours, and the in-hospital mortality rate was 1%. Our findings indicate that multivessel off-pump endo-CAB can be performed safely with favorable postoperative outcomes.

Learning curve and technical challenges

Multivessel endo-CAB is a technically demanding procedure with a steep learning curve. It utilizes conventional off-pump coronary techniques. This approach differs technically from bypass grafting performed on an arrested heart [17]. Patients with advanced atherosclerosis of the aorta, chronic lung disease, or kidney problems may benefit from this approach. Other advantages include reduced risks of postoperative inflammation, infection, and arrhythmia [18]. Safely performing OPCAB requires specific surgical training and dexterity to achieve complete familiarity [19,20]. As Shaefi et al. [2] previously described, maintaining hemodynamic stability and minimizing coronary ischemia are critical considerations. We believe that to safely perform multivessel endo-CAB as described above, a surgeon must be comfortable with sternotomy off-pump techniques, ensuring adequate exposure and stabilization for the precise creation of anastomoses on a beating heart.

We further discuss the challenges associated with the thoracoscopic harvest of the mammary artery. Our group has previously detailed an extensive examination of the learning curve associated with thoracoscopic harvesting. In our initial 80 cases, we observed a significant reduction in LIMA harvesting time, from an average of over 100 minutes to approximately 60 minutes [7,21]. Vassiliades was the first to demonstrate the feasibility of thoracoscopic BIMA harvest with favorable outcomes [22]. Starting the procedure with endoscopic assistance offers several advantages. First, it allows for the opening of the pericardium to identify coronary targets and potentially alter the graft configuration. Second, it facilitates the precise identification of the optimal location for the mini-thoracotomy. Finally, it enables the performance of concurrent thoracoscopic arrhythmia surgery, thus maintaining the principles of minimally invasive surgery while effectively addressing multiple comorbidities [22].

Our second most frequently used conduit is the radial artery, which can be harvested through minimally invasive techniques [14]. Although the RIMA is an excellent graft conduit option, it is less often utilized due to the slightly more challenging harvesting process from the left side of the thorax, leading us to prefer the endoscopic radial artery approach [23]. In this study, we resorted to venous grafts only when arterial conduits were not available, or when technical difficulties or patient comorbidity were present. In almost all cases, we preferred the Y configuration or BIMA in situ as the graft configuration, to provide an all- arterial, anaortic approach. The benefits of this approach have been described by Puskas et al. [3].

This procedure involves both surgical and anesthesiological learning curves. Close collaboration with the anesthesiologist is crucial during heart manipulation, similar to what is required in sternotomy off-pump revascularization. Factors such as table positioning, the administration of inotropes or vasopressors, and the sequencing of grafts are critical for ensuring the safety of the procedure. Occasionally, the heart may need to be positioned slightly differently to maintain stable hemodynamics during the coronary anastomosis, which necessitates effective communication with the anesthesia team.

Cardiopulmonary bypass support

Ideally, the endo-CAB procedure is performed off-pump and without manipulating the aorta, termed “anaortic.” If the surgeon lacks experience with off-pump beating heart surgery, they can decide preoperatively to use a pump. Therefore, pump-assisted endo-CAB may be an attractive option for those with limited experience, as it allows the surgeon to become acquainted with the minimally invasive aspects of the procedure.

Limitations

The study is a single-center retrospective analysis. Additionally, our follow-up period was brief, and the angiographic evidence is confined to cases of hybrid revascularization and target vessel reinterventions. Nevertheless, we can tentatively conclude that for selected patients, multivessel endo-CAB offers favorable perioperative outcomes, with target vessel reintervention rates comparable to those of sternotomy techniques.

Conclusion

Multivessel off-pump endo-CAB can be safely performed using thoracoscopic IMA harvesting techniques. We believe that combining the benefits of minimally invasive surgery with an anaortic approach could enhance perioperative outcomes for patients undergoing coronary surgery. Mastery of conventional off-pump techniques is crucial to successfully perform this procedure. In selected patients, multivessel endo-CAB yields favorable procedural outcomes.

Authors contributions

Conceptualization: DQG, FA. Data curation: DQG, FA. Formal analysis: DQG, FA. Methodology: DQG, FA. Software: DQG, FA. Writing original draft: DQG, FA. Writing review & editing: DQG, FS, NT, JTW, FA. Approval of final manuscript: all authors.

Conflict of interest

Joost Ter Woorst is a proctor for OPCAB at Medtronic. Ferdi Akca is a proctor for Endo-CAB and OPCAB at Medtronic. Except for that, no potential conflict of interest relevant to this article was reported.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Acknowledgments

Data will be made available by the corresponding author upon reasonable request.

  1. Bachar BJ, Manna B. Coronary artery bypass graft. StatPearls Publishing; 2024.
  2. Shaefi S, Mittel A, Loberman D, Ramakrishna H. Off-pump versus on-pump coronary artery bypass grafting: a systematic review and analysis of clinical outcomes. J Cardiothorac Vasc Anesth 2019;33:232-44. https://doi.org/10.1053/j.jvca.2018.04.012.
    Pubmed CrossRef
  3. Puskas JD, Yanagawa B, Taggart DP. Off-pump, multiple arterial grafting with minimal aortic manipulation: is it for everyone?. J Thorac Cardiovasc Surg 2016;151:4-6. https://doi.org/10.1016/j.jtcvs.2015.09.116.
    Pubmed CrossRef
  4. Vallely MP, Potger K, McMillan D, et al. Anaortic techniques reduce neurological morbidity after off-pump coronary artery bypass surgery. Heart Lung Circ 2008;17:299-304. https://doi.org/10.1016/j.hlc.2007.11.138.
    Pubmed CrossRef
  5. Marin-Cuartas M, Sa MP, Torregrossa G, Davierwala PM. Minimally invasive coronary artery surgery: robotic and nonrobotic minimally invasive direct coronary artery bypass techniques. JTCVS Tech 2021;10:170-7. https://doi.org/10.1016/j.xjtc.2021.10.008.
    Pubmed KoreaMed CrossRef
  6. Mavioglu I, Vallely MP. Minimally invasive off-pump anaortic coronary artery bypass (MACAB). J Card Surg 2022;37:4944-51. https://doi.org/10.1111/jocs.17180.
    Pubmed CrossRef
  7. Akca F, Ter Woorst J. Learning curve of thoracoscopic nonrobotic harvest of the left internal mammary artery in minimally invasive coronary artery bypass grafting. Innovations (Phila) 2023;18:262-5. https://doi.org/10.1177/15569845231178012.
    Pubmed CrossRef
  8. Yilmaz A, Robic B, Starinieri P, Polus F, Stinkens R, Stessel B. A new viewpoint on endoscopic CABG: technique description and clinical experience. J Cardiol 2020;75:614-20. https://doi.org/10.1016/j.jjcc.2019.11.007.
    Pubmed CrossRef
  9. Davierwala PM, Verevkin A, Sgouropoulou S, et al. Minimally invasive coronary bypass surgery with bilateral internal thoracic arteries: early outcomes and angiographic patency. J Thorac Cardiovasc Surg 2021;162:1109-19. https://doi.org/10.1016/j.jtcvs.2019.12.136.
    Pubmed CrossRef
  10. Issa HM, Ruel M. Beating heart multi-vessel minimally invasive direct coronary artery bypass grafting: techniques and pitfalls. J Vis Surg 2023;9:5. https://doi.org/10.21037/jovs-22-5.
    CrossRef
  11. McGinn JT Jr, Usman S, Lapierre H, Pothula VR, Mesana TG, Ruel M. Minimally invasive coronary artery bypass grafting: dual-center experience in 450 consecutive patients. Circulation 2009;120(11 Suppl):S78-84. https://doi.org/10.1161/CIRCULATIONAHA.108.840041.
    Pubmed CrossRef
  12. Kikuchi K, Une D, Suzuki K, et al. Off-pump minimally invasive coronary artery bypass grafting with a heart positioner: direct retraction for a better exposure. Innovations (Phila) 2015;10:183-7. https://doi.org/10.1097/IMI.0000000000000167.
    Pubmed CrossRef
  13. Reyes AT, Frame R, Brodman RF. Technique for harvesting the radial artery as a coronary artery bypass graft. Ann Thorac Surg 1995;59:118-26. https://doi.org/10.1016/0003-4975(94)00673-U.
    Pubmed CrossRef
  14. Connolly MW, Torrillo LD, Stauder MJ, et al. Endoscopic radial artery harvesting: results of first 300 patients. Ann Thorac Surg 2002;74:502-6. https://doi.org/10.1016/s0003-4975(02)03717-7.
    Pubmed CrossRef
  15. Kikuchi K, Endo Y. Assistive techniques for proximal anastomosis in minimally invasive coronary artery bypass grafting. Innovations (Phila) 2017;12:224-6. https://doi.org/10.1097/IMI.0000000000000366.
    Pubmed CrossRef
  16. Tjon Joek Tjien A, Akca F, Lam K, et al. Concomitant atrial fibrillation ablation in minimally invasive cardiac surgery. Ann Cardiothorac Surg 2024;13:91-8. https://doi.org/10.21037/acs-2023-afm-21.
    Pubmed KoreaMed CrossRef
  17. Ofoegbu CK, Manganyi RM. Off-pump coronary artery bypass grafting; is it still relevant?. Curr Cardiol Rev 2022;18:e271021197431. https://doi.org/10.2174/1573403X17666211027141043.
    Pubmed KoreaMed CrossRef
  18. Livesay JJ. The benefits of off-pump coronary bypass: a reality or an illusion? Tex Heart Inst J 2003;30:258-60.
  19. Song HK, Petersen RJ, Sharoni E, Guyton RA, Puskas JD. Safe evolution towards routine off-pump coronary artery bypass: negotiating the learning curve. Eur J Cardiothorac Surg 2003;24:947-52. https://doi.org/10.1016/s1010-7940(03)00616-x.
    Pubmed CrossRef
  20. Murzi M, Caputo M, Aresu G, Duggan S, Angelini GD. Training residents in off-pump coronary artery bypass surgery: a 14-year experience. J Thorac Cardiovasc Surg 2012;143:1247-53. https://doi.org/10.1016/j.jtcvs.2011.09.049.
    Pubmed CrossRef
  21. Akca F. Thoracoscopic (non-robotic) harvesting of bilateral internal mammary artery grafts. Multimed Man Cardiothorac Surg 2023 Nov 15. [Epub]. https://doi.org/10.1510/mmcts.2023.069.
    CrossRef
  22. Vassiliades TA Jr, Reddy VS, Puskas JD, Guyton RA. Long-term results of the endoscopic atraumatic coronary artery bypass. Ann Thorac Surg 2007;83:979-85. https://doi.org/10.1016/j.athoracsur.2006.10.031.
    Pubmed CrossRef
  23. Tatoulis J, Buxton BF, Fuller JA. The right internal thoracic artery: the forgotten conduit: 5,766 patients and 991 angiograms. Ann Thorac Surg 2011;92:9-17. https://doi.org/10.1016/j.athoracsur.2011.03.099.
    Pubmed CrossRef
  24. Ohtsuka T, Ninomiya M, Maemura T, Takamoto S. Needle-guided mini- entry in video-assisted coronary artery bypass. Eur J Cardiothorac Surg 2003;24:644-6. https://doi.org/10.1016/s1010-7940(03)00380-4.
    Pubmed CrossRef

Article

Clinical Research

J Chest Surg 2025; 58(1): 21-30

Published online January 5, 2025 https://doi.org/10.5090/jcs.24.080

Copyright © Journal of Chest Surgery.

Endoscopic-Assisted Multivessel Off-Pump Coronary Artery Bypass Grafting: Experience of the First 100 Procedures

De Qing Görtzen, B.Sc., Fleur Sampon , M.D., Naomi Timmermans , M.D., Joost Ter Woorst , M.D., Ph.D., Ferdi Akca , M.D., Ph.D.

Department of Cardiothoracic Surgery, Catharina Hospital, Eindhoven, The Netherlands

Correspondence to:Ferdi Akca
Tel 31-40-239-8680
E-mail ferdi.akca@catharinaziekenhuis.nl
ORCID
https://orcid.org/0000-0002-1748-3235

Received: August 1, 2024; Revised: September 5, 2024; Accepted: September 26, 2024

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Commentary: J Chest Surg. 2025;58(1):31-33 https://doi.org/10.5090/jcs.24.110

Abstract

Background: This study presents an overview of our technique and the perioperative outcomes for the first 100 patients who underwent minimally invasive endoscopic-assisted off-pump multivessel bypass grafting (endoscopic coronary artery bypass [endo-CAB]) at the Catharina Hospital in Eindhoven.
Methods: The first 100 patients undergoing multivessel endo-CAB from May 2022 to March 2024 were included in this retrospective, single-center, observational study (N=100). The study encompassed both elective and urgent surgical revascularization. In all cases, endoscopic-assisted harvesting of the internal mammary artery, radial artery, or saphenous vein was performed, followed by beating-heart anastomoses through a mini-thoracotomy.
Results: A total of 226 distal anastomoses were performed, utilizing 102 left internal mammary arteries, 80 radial arteries, 30 right internal mammary arteries, and 14 saphenous veins. On average, each patient had 2.3 anastomoses. A Y graft configuration was employed in 78 patients, in-situ bilateral internal mammary artery inflow in 19 patients, and a proximal aortic graft in 3 patients. Four patients underwent concurrent arrhythmia surgery. Eleven patients received hybrid revascularization. There was 1 conversion to sternotomy (1%) and 3 instances where cardiopulmonary bypass was required (3%). The median operation time was 3.3 hours (interquartile range, 3.0–3.7 hours), and the median hospital stay was 4.0 days (interquartile range, 3–4 days). The in-hospital mortality rate was 1%.
Conclusion: Multivessel off-pump endo-CAB surgery can be safely performed with endoscopic-assisted conduit harvesting. Combining the benefits of a minimally invasive and anaortic approach may improve perioperative outcomes for patients requiring surgical revascularization. Further studies are necessary to establish the role of this technique in routine coronary surgery.

Keywords: Endoscopic coronary artery bypass grafting, Off-pump coronary bypass grafting, Minimally invasive cardiac surgery, Minimally invasive direct coronary artery bypass

Introduction

The standard of care for managing advanced coronary artery disease is coronary artery bypass grafting (CABG). As the global burden of cardiovascular disease continues to rise, CABG remains the most commonly performed procedure in cardiac surgery [1]. CABG is conventionally performed through a full sternotomy on a non-beating heart, utilizing cardiopulmonary bypass (CPB). However, off-pump coronary artery bypass grafting (OPCAB) has become increasingly popular due to its lower risk of cerebrovascular accidents. This benefit arises because the technique avoids manipulation of the aorta, a method referred to as the “anaortic approach” [2-4].

Minimally invasive coronary surgery is a rapidly evolving and technically demanding field, especially for multivessel revascularization. Using minimally invasive techniques, a median sternotomy is avoided in favor of a mini- thoracotomy or total endoscopic techniques [5]. Various methods have been developed for both conduit harvesting—including direct vision, thoracoscopic-assisted, or robotic-assisted mammary artery harvest—and graft deployment, which can be performed off-pump, on a beating heart, or on an arrested heart [6-12]. The ideal revascularization strategy would avoid sternotomy entirely and achieve complete revascularization using an anaortic approach, thus providing a truly minimally invasive long-term solution for the patient.

We aimed to share our experience with minimally invasive, multivessel, endoscopic-assisted, off-pump coronary surgery (endoscopic coronary artery bypass [endo-CAB]) and provide a comprehensive overview of the perioperative data and outcomes for the first 100 patients undergoing this procedure at the Catharina Hospital in Eindhoven.

Methods

Ethical statement

The institutional review board and the ethical committee of Medical Research Ethics Committees United approved the study and waived the requirement for informed consent (registration no., W23.029).

Patient inclusion

From May 2022 to March 2024, 100 patients undergoing multivessel endo-CAB were included in this retrospective, single-center, observational study at Catharina Hospital in Eindhoven. All procedures were performed by a dedicated minimally invasive off-pump coronary surgeon. Patient baseline characteristics were obtained from the Department of Cardiothoracic Surgery’s database. Our aim was to provide a comprehensive overview of our initial experience and procedural parameters, focusing on the first 100 patients treated with multivessel endo-CAB. The patients had either double (n=87) or triple (n=13) vessel coronary artery disease. For 11 patients, a hybrid revascularization strategy was determined by the heart team, which included a percutaneous intervention alongside the multivessel endo-CAB procedure.

Patient selection and preoperative screening

All patients were evaluated by a heart team, which included a cardiac surgeon and an interventional cardiologist, to assess the need for surgical revascularization. Patients diagnosed with multivessel disease and deemed suitable for either elective or urgent surgical revascularization were considered eligible for endo-CAB. Initially, during the implementation phase of endo-CAB, patients with a lower body mass index (BMI) and double vessel disease were specifically chosen for this procedure. Morbid obesity and a salvage situation involving hemodynamic instability were established as definitive exclusion criteria.

All patients underwent the standard preoperative workup required for any coronary surgery procedure, which included coronary angiography, chest radiography, electrocardiogram, transthoracic echocardiogram, bilateral blood pressure measurements, and blood panel laboratory tests. Following the preoperative workup, we determined whether a patient was eligible for standalone multiple-vessel endo- CAB, hybrid revascularization, or concomitant arrhythmia surgery. Routine preoperative pulmonary function tests or computed tomography scans were not included in our preoperative workup.

Patient positioning and preparation

The procedure is performed under general anesthesia with the patient in a supine position. To improve surgical access, a pillow is placed under the left scapula, elevating the left hemithorax. Standard arterial and central venous lines, along with an endotracheal tube, are positioned before the procedure begins. A selective endobronchial blocker (Teleflex, Wayne, PA, USA) is inserted through the endotracheal tube to enable unilateral ventilation during the grafting process. Additionally, the groin is marked and draped in preparation for potential femoral cannulation if hemodynamic instability occurs. If the radial artery is planned to be harvested endoscopically at the same time, the right arm is positioned at a 90° angle from the thorax. Furthermore, transesophageal echocardiography is utilized to evaluate left ventricular and valve function.

Conduit harvesting

Three small incisions are made on the left side of the chest: 1 at the third intercostal space at the anterior axillary line, and 2 more at the second and fourth intercostal spaces, positioned more anteriorly. Through these incisions, 5-mm endoscopic ports are inserted, and capnothorax is established at a pressure of 8 to 12 mm Hg (Fig. 1A). A 5-mm, 0°, 2-dimensional camera (Karl Storz GmbH, Tuttlingen, Germany) is then introduced. Using the other ports, standard long-shafted video-assisted thoracic surgery instruments and a Ligasure Maryland device (Medtronic, Dublin, Ireland) are also introduced. This setup enables harvesting of both the left internal mammary artery (LIMA) and the right internal mammary artery (RIMA).

Figure 1. (A) The setup for thoracoscopic left internal mammary artery (LIMA) harvesting. (B) The needle guided marking for mini-thoracotomy by Ohtsuka et al. [24]. (C) Ligasure-aided LIMA harvest. (D) The ramus descendens posterior visualized on the video monitor endoscopically.

Before harvesting the mammary artery, the pericardium is initially opened anterior to the left phrenic nerve to identify and assess coronary targets. A clip is placed near the grafting site on the left anterior descending artery (LAD) to facilitate easy identification through the small thoracotomy. When harvesting the bilateral mammary arteries (BIMAs), both the anterior mediastinum and the right pleural space are opened. The RIMA is harvested first, following the opening of the endothoracic fascia. Utilizing long-shafted video-assisted thoracic surgery instruments and the Ligasure device, the RIMA is harvested along with its adjacent veins in a semi-skeletonized fashion. The RIMA can be harvested proximally beyond the subclavian vein and distally up to its bifurcation. Chest wall deformities, such as pectus excavatum, can complicate harvesting the RIMA from the left side. A right thoracoscopic approach or opting for a radial artery can mitigate this issue. Once the RIMA is harvested to the desired length, the LIMA is then harvested.

Thoracoscopically, the phrenic nerve can be easily identified. The LIMA is harvested from the proximal area up to the level of the phrenic nerve and extends distally to its bifurcation. Any side branches of the LIMA are ligated using the Ligasure device (Fig. 1C). Once the LIMA has been fully dissected, heparin is administered intravenously to achieve an activated clotting time of over 300 seconds. The distal portion of the LIMA is secured with titanium clips (Endo-Clip; Medtronic). The LIMA is then sutured to the pericardium using a 5-0 polypropylene stitch. After dividing the LIMA, the location for the mini-thoracotomy is marked. Intrathoracic pressure is relieved to accurately determine the location of the coronary target without the influence of artificial capnothorax pressure. A transthoracic needle is used to puncture the intercostal space above the primary coronary target, ensuring an optimal placement for the mini-thoracotomy (Fig. 1B). The precise location may vary depending on the patient’s specific anatomy.

When a radial artery is used as a second arterial conduit, the LIMA and radial artery are harvested simultaneously to achieve complete arterial revascularization. Endoscopic radial artery harvesting was performed following the methods described by Reyes et al. [13] and Connolly et al. [14], utilizing an endoscopic vessel harvesting system (Vasoview Hemopro; Getinge, Gothenburg, Sweden). Similarly, endoscopic techniques were employed for saphenous vein harvesting.

Mini-thoracotomy

Once all conduits have been harvested and the mini-thoracotomy site has been marked, the endoscopic ports are removed, and unilateral ventilation is initiated. The 5–6 cm incision is made at the previously marked location, between the third and fifth intercostal spaces. For grafting the LAD and right descending posterior, we generally adopt a more medial approach. If grafting of lateral branches is required, a more lateral access must be established. In female patients, the incision is preferably made in the inframammary area for aesthetic reasons. When the mini-thoracotomy is correctly positioned, the LIMA will be attached to the pericardium directly beneath the incision. Fig. 2A illustrates the LIMA as visible through the thoracotomy.

Figure 2. (A) The harvested left internal mammary artery (LIMA) in situ through a mini-thoracotomy. (B) A Y-graft through a mini-thoracotomy. (C) The LIMA-radialis Y-graft on the Octopus cardiac stabilizer. (D) Cardiac positioning with the wet sponges.

Graft configurations

During multivessel procedures, various graft configurations may be employed. These include both BIMA in situ grafts and Y- or T-graft configurations using a free-RIMA or radial artery. Fig. 2A–C illustrate a T-graft comprising the LIMA and radial artery performed through a mini- thoracotomy. If a saphenous vein is utilized, the proximal anastomosis is created on the ascending aorta using techniques previously described by Issa and Ruel [10] and McGinn et al. [11]. Traction sutures are applied to the right lateral pericardium, and a sponge is positioned adjacent to the aorta. The Octopus Nuvo (Octopus Nuvo Tissue Stabiliser; Medtronic) is employed to depress the pulmonary artery, increasing maneuverability.

Cardiac positioning for distal anastomosis

For cardiac positioning, we use the epicardial stabilizer and the heart positioning device (Starfish Evo; Medtronic), as previously described by Davierwala et al. [9], Issa and Ruel [10], and Kikuchi and Endo [15]. Depending on the coronary targets, we either use both devices or only the epicardial stabilizer. The heart is positioned with wet sponges, and the stabilizers are placed as shown in Fig. 2D. For the LAD and diagonal branches, only the epicardial stabilizer is used, as depicted in Fig. 3A. For the anterolateral, lateral, inferolateral, and inferior anastomoses, the armless Starfish Evo is also utilized, as illustrated in Fig. 3B and C. The armless Starfish Evo is employed to elevate the apex, facilitating access to the coronary targets, and a silk 1-0 suture is tied around the Starfish to assist in positioning. For the inferior wall, we elevate the apex of the heart with wet sponges and position the armless Starfish towards the apex. We often loosen our pericardial sutures, as they may limit visibility and maneuverability of the heart. We pull the Starfish cranially to stabilize the position for exposing the right posterior descending coronary artery. Subsequently, we position the epicardial stabilizer to secure a stable field for the coronary anastomosis. Once the heart is positioned, we ensure cardiac stability before continuing to manipulate the coronary artery. In cases of hemodynamic instability, we first elevate the legs to increase venous inflow and reposition the heart to stabilize it. If hemodynamic instability or ventricular arrhythmias persist, femoral CPB support is initiated. When using CPB, anastomoses are performed on the beating heart using similar techniques.

Figure 3. (A) The left internal mammary artery-left anterior descending anastomosis process between the Octopus cardiac stabilizer. (B) The Octopus and Starfish cardiac stabilizers for the margo obtusus grafting. (C) The ramus descendens posterior anastomosis. (D) The esthetic result.

Closing-up and flow measurements

Once all coronary targets had been grafted, protamine was administered. The flow through the graft was measured using transit-time flow measurement (TTFM) to ascertain graft function (Medistim Vascular TTFM; Medistim, Oslo, Norway). After confirming that the graft function was sufficient, a 28F chest tube was inserted into the left pleural cavity through 1 of the 3 incisions previously made for the LIMA harvest. Ropivacaine was administered at 3 intercostal levels, and the skin was then closed (Fig. 3D).

Concomitant arrhythmia surgery

Patients were considered for concomitant arrhythmia surgery if they presented with symptomatic paroxysmal or persistent atrial fibrillation (AF) and had a left atrial volume index of less than 70 mL/m2. A detailed account of our criteria for patient selection and the techniques used in concomitant thoracoscopic arrhythmia surgery has been previously published [16]. For patients with a high BMI, the primary surgical objective was to achieve minimally invasive coronary revascularization. If limited intrathoracic space increased the risk of complications during concomitant pulmonary vein isolation, only the left atrial appendage (LAA) clip (AtriClip; Atricure, Mason, OH, USA) would be applied to mitigate the risk of stroke.

Patient postoperative care

All patients were transferred to the intensive care unit (ICU) after the procedure. Patients with radial artery grafts were administered calcium antagonists postoperatively to prevent graft spasms. Paracetamol, oral opioids, and intravenous nonsteroidal anti-inflammatory drugs were utilized for postoperative pain management. Following discharge, patients attended a follow-up appointment at the outpatient clinic 14 days later, where they underwent chest radiography and laboratory blood tests.

Data collection and statistics

The data was analyzed using JASP ver. 0.18.1.0 (Universiteit van Amsterdam, Amsterdam, Netherlands) to obtain descriptives, boxplots, and Q-Q plots, identifying outliers for each continuous variable. Categorical variables are presented as frequency and percentage, while continuous variables are shown as mean and standard deviation if they are normally distributed, and as median and interquartile range (IQR) if they are not, based on the Q-Q plot analysis. RStudio ver. 4.4.0 (R Core Team, Boston, MA, USA) was utilized to visualize data and compute the Pearson correlation coefficient.

Results

Demographics, procedural parameters, and perioperative characteristics

The patient demographics, procedural parameters, and perioperative characteristics are displayed in Table 1. Most of the patients were men (90.0%) and the mean age was 65.5±9.1 years. The mean BMI was 26.7±3.3 kg/m2, and the median body surface area was 2.00 m2 (IQR, 1.91–2.10 m2). Seventeen percent of the patients (17.0%) had diabetes, 7 patients had peripheral artery disease (7%), and 5 patients had AF (5%). The median Euroscore II for the population was 1.00 (IQR, 0.80–1.61).

Table 1 . Demographics and perioperative characteristics of the endoscopic coronary artery bypass group (N=100).

CharacteristicValue
Sex (men)90 (90.0)
Age (yr)65.5±9.1
Height (cm)175 (171–180)
Weight (kg)82.3±12.8
Body mass index (kg/m2)26.7±3.3
Body surface area (m2)2.00 (1.91–2.10)
Diabetes17 (17.0)
Peripheral vascular disease7 (7.0)
Atrial fibrillation5 (5.0)
Previous cardiac surgery0
Recent myocardial infarction28 (28.0)
Prior percutaneous coronary intervention33 (33.0)
Left ventricular ejection fraction
Good (<50%)78 (78.0)
Moderate (31%–50%)18 (18.0)
Poor (21%–30%)3 (3.0)
Very poor (<20%)1 (1.0)
Chronic obstructive pulmonary disease4 (4.0)
Pulmonary hypertension0
Cerebrovascular accident6 (6.0)
Renal impairment5 (5.0)
Dialysis0
Level of urgency
Elective55 (55.0)
Urgent44 (44.0)
Emergency1 (1.0)
Euroscore II1.00 (0.80–1.61)
Preoperative hemoglobin (mmol/L)9.1 (8.3–9.6)
Double vessel disease87 (87.0)
Triple vessel disease13 (13.0)
Concomitant arrhythmia surgery4 (4.0)
Isolated LAA clip3 (3.0)
Thoracoscopic AF ablation including LAA clip1 (1.0)
No. of distal anastomoses
275 (75.0)
324 (24.0)
41 (1.0)
No. of anastomoses with left IMA
196 (96.0)
23 (3.0)
No. of anastomoses with right IMA
126 (26.0)
22 (2.0)
No. of anastomoses with radial artery
142 (42.0)
219 (19.0)
No. of anastomoses with SVG
16 (6.0)
24 (4.0)
Y-graft configuration78 (78.0)
Bilateral internal mammary artery in situ19 (19.0)
Proximal aorta3 (3.0)
Hybrid revascularization11 (11.0)
PCI, circumflex artery5 (5.0)
PCI, right coronary artery6 (6.0)
Conversion to sternotomy1 (1.0)
CPB-assisted3 (3.0)
Operation time (hr)3.3 (3.0–3.7)

Values are presented as frequency (%) for categorical variables; for continuous variables, they are shown as mean±standard deviation if normally distributed, or as median (interquartile range) if non-normally distributed..

LAA, left atrial appendage; AF, atrial fibrillation; IMA, internal mammary artery; SVG, saphenous vein graft; PCI, percutaneous coronary intervention; CPB cardiopulmonary bypass..



Four patients underwent concomitant arrhythmia surgery. Three underwent isolated LAA clip placement and 1 patient underwent thoracoscopic AF ablation, including an LAA clip. The patients received an average of 2.3 distal anastomoses. The total number of grafts per conduit is plotted in Fig. 4A. The first choice of conduit was the LIMA with 102 anastomoses, while the radial artery was used for 80 anastomoses, the RIMA for 30 anastomoses, and the saphenous vein for 14 anastomoses. A Y-graft configuration was used in 78 patients (78%), the BIMAs in situ were utilized in 19 patients (19%), and a proximal aorta graft was placed in 3 cases (3%) (Fig. 4B). Coronary anastomoses were performed on the anterior (n=100), anterolateral (n=42), lateral (n=65), inferolateral (n=5), and inferior wall (n=14) (Fig. 4C, Table 2). In our population, 11 patients underwent hybrid revascularization (11%). In 1 patient, a conversion to sternotomy was performed due to limited visibility (1%), and 3 cases needed CPB support (3%). In 1 patient, this was planned preoperatively, while in 2 patients, CPB was used due to hemodynamic instability. The median operation time was 3.3 hours (IQR, 3.0–3.7 hours). The Pearson correlation coefficient was calculated to illustrate a possible learning curve in Fig. 5. The Pearson correlation was calculated to be r(98)=-0.109 (p=0.279).

Figure 4. (A) The number of distal anastomoses for each conduit. (B) The used graft configurations. (C) The number of anastomoses for each myocardial territory. LIMA, left internal mammary artery; RIMA, right internal mammary artery; BIMA, bilateral internal mammary artery.

Figure 5. The operative time in minutes per procedure since implantation. The Pearson correlation coefficient has been calculated and visualized for the operative time since implementation.

Table 2 . Graft locations in the endoscopic coronary artery bypass group.

LocationNo.
Anterior100
Anterolateral42
Lateral65
Inferolateral5
Inferior14


Postoperative characteristics and complications

Table 3 describes the postoperative characteristics and complications. The total ventilation time was 7.6 hours (IQR, 6.6–9.3 hours), and the total postoperative blood loss was 350 mL (IQR, 200–550 mL). The median stay in the ICU was 1.0 days (IQR, 0.5–1 days), and the total postoperative hospital stay was 4 days (IQR, 3–4 days). In-hospital mortality occurred in 1 patient (1%) due to shock caused by a postoperative stress response due to previously undiagnosed cancer metastases. One patient required postoperative re-exploration for bleeding (1%), and 1 patient had postoperative ischemia requiring revision of the bypass graft (1%). One patient had a pulmonary embolism that required oral anticoagulation. There was no occurrence of stroke. During hospital admission, 7 patients developed new-onset AF that required oral anticoagulation (7%).

Table 3 . Postoperative characteristics of the endoscopic coronary artery bypass group (N=100).

VariableValue
Total ventilation time (hr)7.6 (6.6–9.3)
Postoperative blood loss (mL)350 (200–550)
Intensive care unit admission time (day)1.0 (0.5–1)
Total hospital stay (day)4 (3–4)
In hospital mortality1 (1.0)
Intra-aortic balloon pump0
Extracorporeal membrane oxygenation0
CK-MB max (ng/mL)25.5 (19.0–35.3)
Discharge hemoglobin (mmol/L)7.7 (6.8–8.3)
Transfusion6 (6.0)
Re-exploration for bleeding1 (1.0)
Ischemia postoperative1 (1.0)
Graft revision1 (1.0)
Acute percutaneous coronary intervention0
Cerebrovascular accident0
Tamponade during initial admission0
Tamponade after discharge1 (1.0)
Pulmonary embolism1 (1.0)
Wound infection0
New-onset atrial fibrillation7 (7.0)
Hospital-acquired pneumonia3 (3.0)
Thoracentesis3 (3.0)
30-day readmission2 (2.0)
Target vessel reintervention3 (3.0)

Values are presented as number (%) for categorical variables and median (interquartile range) for continuous variables if non-normally distributed..

CK-MB, creatine kinase-myoglobin binding..



During follow-up, 1 patient developed late tamponade (1%) that required pericardiocentesis and 3 patients had left-sided pleural effusion that required thoracentesis (3%). Two patients were readmitted within 30 days after discharge for symptomatic postoperative anemia (n=1) and dyspnea due to pleural effusion requiring thoracentesis (n=1). Three patients received unexpected target vessel reintervention (3%) during follow-up.

Discussion

In this study, we described the outcomes of our first 100 endoscopic-assisted, off-pump, multivessel coronary bypass grafting procedures. All patients underwent endoscopic-assisted graft harvesting and received coronary bypasses via a mini-thoracotomy. On average, each patient had 2.3 distal anastomoses. Eleven patients underwent hybrid revascularization. There was 1 conversion to sternotomy and 3 instances where CPB was required. The median operation time was 3.3 hours, and the in-hospital mortality rate was 1%. Our findings indicate that multivessel off-pump endo-CAB can be performed safely with favorable postoperative outcomes.

Learning curve and technical challenges

Multivessel endo-CAB is a technically demanding procedure with a steep learning curve. It utilizes conventional off-pump coronary techniques. This approach differs technically from bypass grafting performed on an arrested heart [17]. Patients with advanced atherosclerosis of the aorta, chronic lung disease, or kidney problems may benefit from this approach. Other advantages include reduced risks of postoperative inflammation, infection, and arrhythmia [18]. Safely performing OPCAB requires specific surgical training and dexterity to achieve complete familiarity [19,20]. As Shaefi et al. [2] previously described, maintaining hemodynamic stability and minimizing coronary ischemia are critical considerations. We believe that to safely perform multivessel endo-CAB as described above, a surgeon must be comfortable with sternotomy off-pump techniques, ensuring adequate exposure and stabilization for the precise creation of anastomoses on a beating heart.

We further discuss the challenges associated with the thoracoscopic harvest of the mammary artery. Our group has previously detailed an extensive examination of the learning curve associated with thoracoscopic harvesting. In our initial 80 cases, we observed a significant reduction in LIMA harvesting time, from an average of over 100 minutes to approximately 60 minutes [7,21]. Vassiliades was the first to demonstrate the feasibility of thoracoscopic BIMA harvest with favorable outcomes [22]. Starting the procedure with endoscopic assistance offers several advantages. First, it allows for the opening of the pericardium to identify coronary targets and potentially alter the graft configuration. Second, it facilitates the precise identification of the optimal location for the mini-thoracotomy. Finally, it enables the performance of concurrent thoracoscopic arrhythmia surgery, thus maintaining the principles of minimally invasive surgery while effectively addressing multiple comorbidities [22].

Our second most frequently used conduit is the radial artery, which can be harvested through minimally invasive techniques [14]. Although the RIMA is an excellent graft conduit option, it is less often utilized due to the slightly more challenging harvesting process from the left side of the thorax, leading us to prefer the endoscopic radial artery approach [23]. In this study, we resorted to venous grafts only when arterial conduits were not available, or when technical difficulties or patient comorbidity were present. In almost all cases, we preferred the Y configuration or BIMA in situ as the graft configuration, to provide an all- arterial, anaortic approach. The benefits of this approach have been described by Puskas et al. [3].

This procedure involves both surgical and anesthesiological learning curves. Close collaboration with the anesthesiologist is crucial during heart manipulation, similar to what is required in sternotomy off-pump revascularization. Factors such as table positioning, the administration of inotropes or vasopressors, and the sequencing of grafts are critical for ensuring the safety of the procedure. Occasionally, the heart may need to be positioned slightly differently to maintain stable hemodynamics during the coronary anastomosis, which necessitates effective communication with the anesthesia team.

Cardiopulmonary bypass support

Ideally, the endo-CAB procedure is performed off-pump and without manipulating the aorta, termed “anaortic.” If the surgeon lacks experience with off-pump beating heart surgery, they can decide preoperatively to use a pump. Therefore, pump-assisted endo-CAB may be an attractive option for those with limited experience, as it allows the surgeon to become acquainted with the minimally invasive aspects of the procedure.

Limitations

The study is a single-center retrospective analysis. Additionally, our follow-up period was brief, and the angiographic evidence is confined to cases of hybrid revascularization and target vessel reinterventions. Nevertheless, we can tentatively conclude that for selected patients, multivessel endo-CAB offers favorable perioperative outcomes, with target vessel reintervention rates comparable to those of sternotomy techniques.

Conclusion

Multivessel off-pump endo-CAB can be safely performed using thoracoscopic IMA harvesting techniques. We believe that combining the benefits of minimally invasive surgery with an anaortic approach could enhance perioperative outcomes for patients undergoing coronary surgery. Mastery of conventional off-pump techniques is crucial to successfully perform this procedure. In selected patients, multivessel endo-CAB yields favorable procedural outcomes.

Article information

Authors contributions

Conceptualization: DQG, FA. Data curation: DQG, FA. Formal analysis: DQG, FA. Methodology: DQG, FA. Software: DQG, FA. Writing original draft: DQG, FA. Writing review & editing: DQG, FS, NT, JTW, FA. Approval of final manuscript: all authors.

Conflict of interest

Joost Ter Woorst is a proctor for OPCAB at Medtronic. Ferdi Akca is a proctor for Endo-CAB and OPCAB at Medtronic. Except for that, no potential conflict of interest relevant to this article was reported.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Acknowledgments

Data will be made available by the corresponding author upon reasonable request.

Fig 1.

Figure 1.(A) The setup for thoracoscopic left internal mammary artery (LIMA) harvesting. (B) The needle guided marking for mini-thoracotomy by Ohtsuka et al. [24]. (C) Ligasure-aided LIMA harvest. (D) The ramus descendens posterior visualized on the video monitor endoscopically.
Journal of Chest Surgery 2025; 58: 21-30https://doi.org/10.5090/jcs.24.080

Fig 2.

Figure 2.(A) The harvested left internal mammary artery (LIMA) in situ through a mini-thoracotomy. (B) A Y-graft through a mini-thoracotomy. (C) The LIMA-radialis Y-graft on the Octopus cardiac stabilizer. (D) Cardiac positioning with the wet sponges.
Journal of Chest Surgery 2025; 58: 21-30https://doi.org/10.5090/jcs.24.080

Fig 3.

Figure 3.(A) The left internal mammary artery-left anterior descending anastomosis process between the Octopus cardiac stabilizer. (B) The Octopus and Starfish cardiac stabilizers for the margo obtusus grafting. (C) The ramus descendens posterior anastomosis. (D) The esthetic result.
Journal of Chest Surgery 2025; 58: 21-30https://doi.org/10.5090/jcs.24.080

Fig 4.

Figure 4.(A) The number of distal anastomoses for each conduit. (B) The used graft configurations. (C) The number of anastomoses for each myocardial territory. LIMA, left internal mammary artery; RIMA, right internal mammary artery; BIMA, bilateral internal mammary artery.
Journal of Chest Surgery 2025; 58: 21-30https://doi.org/10.5090/jcs.24.080

Fig 5.

Figure 5.The operative time in minutes per procedure since implantation. The Pearson correlation coefficient has been calculated and visualized for the operative time since implementation.
Journal of Chest Surgery 2025; 58: 21-30https://doi.org/10.5090/jcs.24.080

Table 1 . Demographics and perioperative characteristics of the endoscopic coronary artery bypass group (N=100).

CharacteristicValue
Sex (men)90 (90.0)
Age (yr)65.5±9.1
Height (cm)175 (171–180)
Weight (kg)82.3±12.8
Body mass index (kg/m2)26.7±3.3
Body surface area (m2)2.00 (1.91–2.10)
Diabetes17 (17.0)
Peripheral vascular disease7 (7.0)
Atrial fibrillation5 (5.0)
Previous cardiac surgery0
Recent myocardial infarction28 (28.0)
Prior percutaneous coronary intervention33 (33.0)
Left ventricular ejection fraction
Good (<50%)78 (78.0)
Moderate (31%–50%)18 (18.0)
Poor (21%–30%)3 (3.0)
Very poor (<20%)1 (1.0)
Chronic obstructive pulmonary disease4 (4.0)
Pulmonary hypertension0
Cerebrovascular accident6 (6.0)
Renal impairment5 (5.0)
Dialysis0
Level of urgency
Elective55 (55.0)
Urgent44 (44.0)
Emergency1 (1.0)
Euroscore II1.00 (0.80–1.61)
Preoperative hemoglobin (mmol/L)9.1 (8.3–9.6)
Double vessel disease87 (87.0)
Triple vessel disease13 (13.0)
Concomitant arrhythmia surgery4 (4.0)
Isolated LAA clip3 (3.0)
Thoracoscopic AF ablation including LAA clip1 (1.0)
No. of distal anastomoses
275 (75.0)
324 (24.0)
41 (1.0)
No. of anastomoses with left IMA
196 (96.0)
23 (3.0)
No. of anastomoses with right IMA
126 (26.0)
22 (2.0)
No. of anastomoses with radial artery
142 (42.0)
219 (19.0)
No. of anastomoses with SVG
16 (6.0)
24 (4.0)
Y-graft configuration78 (78.0)
Bilateral internal mammary artery in situ19 (19.0)
Proximal aorta3 (3.0)
Hybrid revascularization11 (11.0)
PCI, circumflex artery5 (5.0)
PCI, right coronary artery6 (6.0)
Conversion to sternotomy1 (1.0)
CPB-assisted3 (3.0)
Operation time (hr)3.3 (3.0–3.7)

Values are presented as frequency (%) for categorical variables; for continuous variables, they are shown as mean±standard deviation if normally distributed, or as median (interquartile range) if non-normally distributed..

LAA, left atrial appendage; AF, atrial fibrillation; IMA, internal mammary artery; SVG, saphenous vein graft; PCI, percutaneous coronary intervention; CPB cardiopulmonary bypass..


Table 2 . Graft locations in the endoscopic coronary artery bypass group.

LocationNo.
Anterior100
Anterolateral42
Lateral65
Inferolateral5
Inferior14

Table 3 . Postoperative characteristics of the endoscopic coronary artery bypass group (N=100).

VariableValue
Total ventilation time (hr)7.6 (6.6–9.3)
Postoperative blood loss (mL)350 (200–550)
Intensive care unit admission time (day)1.0 (0.5–1)
Total hospital stay (day)4 (3–4)
In hospital mortality1 (1.0)
Intra-aortic balloon pump0
Extracorporeal membrane oxygenation0
CK-MB max (ng/mL)25.5 (19.0–35.3)
Discharge hemoglobin (mmol/L)7.7 (6.8–8.3)
Transfusion6 (6.0)
Re-exploration for bleeding1 (1.0)
Ischemia postoperative1 (1.0)
Graft revision1 (1.0)
Acute percutaneous coronary intervention0
Cerebrovascular accident0
Tamponade during initial admission0
Tamponade after discharge1 (1.0)
Pulmonary embolism1 (1.0)
Wound infection0
New-onset atrial fibrillation7 (7.0)
Hospital-acquired pneumonia3 (3.0)
Thoracentesis3 (3.0)
30-day readmission2 (2.0)
Target vessel reintervention3 (3.0)

Values are presented as number (%) for categorical variables and median (interquartile range) for continuous variables if non-normally distributed..

CK-MB, creatine kinase-myoglobin binding..


References

  1. Bachar BJ, Manna B. Coronary artery bypass graft. StatPearls Publishing; 2024.
  2. Shaefi S, Mittel A, Loberman D, Ramakrishna H. Off-pump versus on-pump coronary artery bypass grafting: a systematic review and analysis of clinical outcomes. J Cardiothorac Vasc Anesth 2019;33:232-44. https://doi.org/10.1053/j.jvca.2018.04.012.
    Pubmed CrossRef
  3. Puskas JD, Yanagawa B, Taggart DP. Off-pump, multiple arterial grafting with minimal aortic manipulation: is it for everyone?. J Thorac Cardiovasc Surg 2016;151:4-6. https://doi.org/10.1016/j.jtcvs.2015.09.116.
    Pubmed CrossRef
  4. Vallely MP, Potger K, McMillan D, et al. Anaortic techniques reduce neurological morbidity after off-pump coronary artery bypass surgery. Heart Lung Circ 2008;17:299-304. https://doi.org/10.1016/j.hlc.2007.11.138.
    Pubmed CrossRef
  5. Marin-Cuartas M, Sa MP, Torregrossa G, Davierwala PM. Minimally invasive coronary artery surgery: robotic and nonrobotic minimally invasive direct coronary artery bypass techniques. JTCVS Tech 2021;10:170-7. https://doi.org/10.1016/j.xjtc.2021.10.008.
    Pubmed KoreaMed CrossRef
  6. Mavioglu I, Vallely MP. Minimally invasive off-pump anaortic coronary artery bypass (MACAB). J Card Surg 2022;37:4944-51. https://doi.org/10.1111/jocs.17180.
    Pubmed CrossRef
  7. Akca F, Ter Woorst J. Learning curve of thoracoscopic nonrobotic harvest of the left internal mammary artery in minimally invasive coronary artery bypass grafting. Innovations (Phila) 2023;18:262-5. https://doi.org/10.1177/15569845231178012.
    Pubmed CrossRef
  8. Yilmaz A, Robic B, Starinieri P, Polus F, Stinkens R, Stessel B. A new viewpoint on endoscopic CABG: technique description and clinical experience. J Cardiol 2020;75:614-20. https://doi.org/10.1016/j.jjcc.2019.11.007.
    Pubmed CrossRef
  9. Davierwala PM, Verevkin A, Sgouropoulou S, et al. Minimally invasive coronary bypass surgery with bilateral internal thoracic arteries: early outcomes and angiographic patency. J Thorac Cardiovasc Surg 2021;162:1109-19. https://doi.org/10.1016/j.jtcvs.2019.12.136.
    Pubmed CrossRef
  10. Issa HM, Ruel M. Beating heart multi-vessel minimally invasive direct coronary artery bypass grafting: techniques and pitfalls. J Vis Surg 2023;9:5. https://doi.org/10.21037/jovs-22-5.
    CrossRef
  11. McGinn JT Jr, Usman S, Lapierre H, Pothula VR, Mesana TG, Ruel M. Minimally invasive coronary artery bypass grafting: dual-center experience in 450 consecutive patients. Circulation 2009;120(11 Suppl):S78-84. https://doi.org/10.1161/CIRCULATIONAHA.108.840041.
    Pubmed CrossRef
  12. Kikuchi K, Une D, Suzuki K, et al. Off-pump minimally invasive coronary artery bypass grafting with a heart positioner: direct retraction for a better exposure. Innovations (Phila) 2015;10:183-7. https://doi.org/10.1097/IMI.0000000000000167.
    Pubmed CrossRef
  13. Reyes AT, Frame R, Brodman RF. Technique for harvesting the radial artery as a coronary artery bypass graft. Ann Thorac Surg 1995;59:118-26. https://doi.org/10.1016/0003-4975(94)00673-U.
    Pubmed CrossRef
  14. Connolly MW, Torrillo LD, Stauder MJ, et al. Endoscopic radial artery harvesting: results of first 300 patients. Ann Thorac Surg 2002;74:502-6. https://doi.org/10.1016/s0003-4975(02)03717-7.
    Pubmed CrossRef
  15. Kikuchi K, Endo Y. Assistive techniques for proximal anastomosis in minimally invasive coronary artery bypass grafting. Innovations (Phila) 2017;12:224-6. https://doi.org/10.1097/IMI.0000000000000366.
    Pubmed CrossRef
  16. Tjon Joek Tjien A, Akca F, Lam K, et al. Concomitant atrial fibrillation ablation in minimally invasive cardiac surgery. Ann Cardiothorac Surg 2024;13:91-8. https://doi.org/10.21037/acs-2023-afm-21.
    Pubmed KoreaMed CrossRef
  17. Ofoegbu CK, Manganyi RM. Off-pump coronary artery bypass grafting; is it still relevant?. Curr Cardiol Rev 2022;18:e271021197431. https://doi.org/10.2174/1573403X17666211027141043.
    Pubmed KoreaMed CrossRef
  18. Livesay JJ. The benefits of off-pump coronary bypass: a reality or an illusion? Tex Heart Inst J 2003;30:258-60.
  19. Song HK, Petersen RJ, Sharoni E, Guyton RA, Puskas JD. Safe evolution towards routine off-pump coronary artery bypass: negotiating the learning curve. Eur J Cardiothorac Surg 2003;24:947-52. https://doi.org/10.1016/s1010-7940(03)00616-x.
    Pubmed CrossRef
  20. Murzi M, Caputo M, Aresu G, Duggan S, Angelini GD. Training residents in off-pump coronary artery bypass surgery: a 14-year experience. J Thorac Cardiovasc Surg 2012;143:1247-53. https://doi.org/10.1016/j.jtcvs.2011.09.049.
    Pubmed CrossRef
  21. Akca F. Thoracoscopic (non-robotic) harvesting of bilateral internal mammary artery grafts. Multimed Man Cardiothorac Surg 2023 Nov 15. [Epub]. https://doi.org/10.1510/mmcts.2023.069.
    CrossRef
  22. Vassiliades TA Jr, Reddy VS, Puskas JD, Guyton RA. Long-term results of the endoscopic atraumatic coronary artery bypass. Ann Thorac Surg 2007;83:979-85. https://doi.org/10.1016/j.athoracsur.2006.10.031.
    Pubmed CrossRef
  23. Tatoulis J, Buxton BF, Fuller JA. The right internal thoracic artery: the forgotten conduit: 5,766 patients and 991 angiograms. Ann Thorac Surg 2011;92:9-17. https://doi.org/10.1016/j.athoracsur.2011.03.099.
    Pubmed CrossRef
  24. Ohtsuka T, Ninomiya M, Maemura T, Takamoto S. Needle-guided mini- entry in video-assisted coronary artery bypass. Eur J Cardiothorac Surg 2003;24:644-6. https://doi.org/10.1016/s1010-7940(03)00380-4.
    Pubmed CrossRef

Stats or Metrics

Share this article on :

  • line