Minimally invasive coronary artery bypass grafting: a systematic review

Introduction

Coronary artery bypass grafting (CABG) is the standard of care for patients with coronary artery disease. With converging outcomes between percutaneous coronary intervention and CABG in selected patient populations, there has been increasing attention on minimizing surgical morbidity. In this regard, minimally invasive cardiac surgery has gained popularity in recent years. Currently, the most dominant forms of minimally invasive surgical coronary arterial revascularization can be classified as off-pump CABG via a median sternotomy (OPCAB), minimally invasive direct coronary artery bypass (MIDCAB), and minimally invasive coronary surgery-CABG (MICS CABG). OPCAB was introduced in the late 1990s and popularized in the early 2000s for its potential benefit in reducing the morbidity of cardiopulmonary bypass (CPB) and its ability to achieve complete revascularization.^1,2 Conventionally, it is conducted via a median sternotomy and allows a multivessel operation, including utilization of bilateral internal thoracic arteries (BITA). MIDCAB without CPB was introduced by Calafiore and colleagues ³ and is characterized by a left anterior small thoracotomy that allows harvesting and grafting of the left internal mammary artery (LIMA) to the left anterior descending artery (LAD) without CPB. Open, robotic, and endoscopic forms of LIMA harvesting have been described for this approach, ⁴ but a major limitation is its ability to revascularize only the LAD. Hence, in the MIDCAB domain, the current paradigm is shifting toward a hybrid approach where the LIMA graft is anastomosed to the LAD and the remaining lesions are treated via stenting when deemed appropriate. ⁵ MICS CABG was popularized in the late 2000s and similarly utilizes a small left anterior thoracotomy, ⁶ but is differentiated from MIDCAB by its ability to conduct multivessel revascularization without the use of CPB.^7,8 Our group recently reported techniques to harvest BITA under direct vision via a small thoracotomy incision to achieve BITA revascularization in MICS CABG,^9,10 and some centers have successfully undertaken this robotically and endoscopically. ¹¹ In an extreme form, CABG under epidural anesthesia (awake CABG) has been described. ¹² The access to BITA for a more complete arterial revascularization is a major advantage of MICS CABG, because the use of BITA grafts provides superior outcomes compared to single internal thoracic artery (SITA) revascularization.^13–15 Following general discussions on OPCAB and total arterial revascularization, this review focuses on the proposal of indications for minimally invasive total arterial CABG, a summary of the surgical techniques, and current outcomes of total arterial minimally invasive CABG.

Method

We systematically searched the OVID version of Medline (1946 through June 2016) for relevant studies using search terms for coronary artery bypass surgery, and utilized filters to identify randomized controlled trials, retrospective studies, and guidelines on this topic. The search was restricted to publications in the English language. From the generated lists, relevant studies were identified for inclusion in the review.

Cardiopulmonary bypass and off-pump CABG

Regardless of the incisional approach and conduit choices, the cornerstone of minimally invasive coronary artery bypass is off-pump surgery without the use of CPB and with minimal aortic manipulation. The morbidity of CPB is well established, ¹⁶ and the heightened risk of stroke associated with retrograde arterial perfusion has been highlighted in experiences of minimally invasive mitral valve procedures. ¹⁷ A meta-analysis conducted by the International Society of Minimally Invasive Cardiac Surgery (ISMICS) concluded that compared to conventional CABG, OPCAB reduces the risk of stroke (class IIa, level of evidence [LOE] A), renal dysfunction/failure (class IIa, LOE A), blood transfusion (class I, LOE A), respiratory failure (class I, LOE A), atrial fibrillation (class I, LOE A), wound infection (class I, LOE A), ventilation time, and intensive care unit and hospital length of stay (class I, LOE A). OPCAB is associated with a reduced number of grafts placed (class I, LOE A) and diminished graft patency (class IIa, LOE A), with increased coronary reintervention at 1 year and beyond (class IIa, LOE A). ¹⁸ It should be noted that most of the evidence used to generate the above meta-analysis originated from earlier eras of OPCAB, with significant heterogeneity among the studies included.

Numerous trials have demonstrated that off-pump CABG reduces operative time and perioperative morbidity with equivalent survival and risk of myocardial infarction at 30 days and 1 year following surgery, despite higher incidences of graft occlusion at 30 days. ¹⁸ Initial concern regarding the short-term outcomes of OPCAB was raised by the Randomized On/Off Bypass (ROOBY) trial, a randomized controlled trial enrolling 2203 patients from the Veterans Affairs system, ¹⁹ which demonstrated a significantly higher rate of the composite event at 1 year with OPCAB than with on-pump CABG (9.9% vs. 7.4%; relative risk 1.33; p = 0.04) with lower rates of graft patency in the OPCAB group. The main concerns regarding the trial design were that the setting consisted exclusively of the Veterans Affairs system and a significant proportion of the primary surgeons were trainees. A subsequent randomized controlled trial, CABG Off or On Pump Revascularization Study (CORONARY), a multinational trial enrolling 4752 patients operated on by surgeons with more than 2 years of experience and who had completed more than 100 cases with the assigned technique, demonstrated no significant difference in the rate of the primary composite outcome between the two techniques at 1-year follow-up. ²⁰ The study design and sample size are perhaps the most robust of the existing trials on this topic. Of note, in CORONARY, there were weak trends toward lower rates of the primary event with OPCAB compared to on-pump CABG in the mid- to high-risk population (EuroSCORE 3–5 and >5), while there was a trend toward lower rates of the primary event in on-pump CABG in the low-risk population (EuroSCORE ≤ 2).

The long-term outcome beyond 1 year has been under intense scrutiny and several randomized controlled trials have shown that the graft patency and survival is inferior in OPCAB compared to on-pump CABG, although the majority of trials with long-term follow-up are from the earlier era of OPCAB. ¹⁸ The Best Bypass Surgery Trial, a single-center randomized controlled trial enrolling 341 patients with 3-vessel disease and high-risk (EuroSCORE ≥ 5), demonstrated increased all-cause mortality at the 3-year follow-up in the OPCAB cohort compared to on-pump CABG (24% vs. 15%; hazard ratio 1.66, 95% confidence interval 1.02–2.73; p = 0.04) but no significant difference in cardiac-related death. ²¹ The MASS III trial, a single-center randomized controlled trial enrolling 308 patients, demonstrated that at 5-year follow-up, there was no difference in the primary composite endpoint between OPCAB and on-pump CABG. ²² Both the Best Bypass Surgery Trial and the MASS III trial are limited by being single-center studies with relatively small sample sizes. Trials consisting exclusively of experienced surgeons and routine use of standard safety measures (epiaortic scanning, humidified CO₂ blower, selective use of intracoronary shunts, and intraoperative assessment of graft patency) is still awaited to provide further insights into the long-term outcome of OPCAB. Importantly, the use of OPCAB is disproportionate globally; its use in the USA has plummeted in recent years, but 65% of all CABG surgery is performed off-pump in Japan. ²³ With regards to the use of CPB in MICS CABG, we agree with the assessment by Ruel and colleagues ²⁴ that it may be useful as a safety margin in the initial phase of overcoming the learning curve of the MICS CABG technique. However, care must be taken to rigorously assess the state of the descending aorta and iliofemoral system.

Importance of multi-arterial BITA revascularization

Total arterial revascularization has been shown to improve long-term outcomes in patients undergoing CABG, with the cornerstone being the use of BITA when available. However, an analysis of the Society of Thoracic Surgeons Adult Cardiac Surgery Database demonstrated that the current incidence of a second arterial graft use is less than 7%. ²⁵ In addition, only 4% of patients undergoing CABG in the USA received BITA grafts. ²⁶ The incidence is much lower even compared to the USA participating site in the SYNTAX trial (17% second arterial conduit use) or the overall trial cohort (35.3%). Further analysis of SYNTAX registries showed that compared to patients undergoing CABG in European centers, where the use of multi-arterial grafts was more prevalent, patients in the USA experienced significantly higher rates of repeat revascularization and graft occlusion, although there was no difference in the rates of major adverse cardiac and cerebrovascular events. ²⁷

Current guidelines, although based mostly on studies of on-pump CABG, which include recommendations regarding conduit use in CABG are: the 2014 European Society of Cardiology/European Association for Cardio-Thoracic Surgery Guidelines on myocardial revascularization, ²⁸ the 2011 American College of Cardiology Foundation/American Heart Association guideline for coronary artery bypass graft surgery, ²⁹ and the 2015 Society of Thoracic Surgeons guidelines on arterial conduits for coronary artery bypass grafting. ²⁵ In the European guidelines, use of BITA is a class IIa recommendation in patients <70 years of age (or life-expectancy >5 years), and total arterial revascularization is a class IIa recommendation for patients with reasonable life-expectancy. The 2015 Society of Thoracic Surgeons guideline recommends consideration of the use of a second arterial graft (right internal mammary artery or radial artery) for appropriate patients with reasonable life-expectancy and appropriate targets (class IIa), and to consider use of BITA in patients without excessive risk of sternal complications (class IIa). While robust evidence of the long-term advantages of BITA use compared to SITA is awaiting the completion of the Arterial Revascularisation Trial (ART), a recent meta-analysis demonstrated a significant survival benefit of BITA compared to SITA revascularization in the long-term (hazard ratio 0.79, 95% confidence interval 0.75–0.84). ³⁰ In addition, BITA use has been shown to confer a survival advantage in spite of diabetic status. ³¹ Routine measures to minimize the risk of sternal infection with BITA use (smoking cessation, graft skeletonization, glycemic control) are also encouraged.

Indications for minimally invasive multi-arterial BITA revascularization

As discussed above, in revascularization of patients with multivessel disease and a low risk of sternal wound complications, current evidence supports the use of multiple arterial grafts, with the second conduit ideally being the right internal mammary artery. Conventionally, this was accomplished via CABG with a median sternotomy, either off- or on-pump. Current guidelines offer minimal guidance on the indications for minimally invasive coronary revascularization, owing to the paucity of data on long-term outcomes. Recent reports have demonstrated reduced hospital length of stay, reduced need for blood transfusion, and expedited postoperative recovery in patients who underwent MICS CABG compared to conventional off-pump CABG via a median sternotomy.^32,33 In addition, MICS CABG may confer a pronounced benefit in patients aged 75 years and older. ³⁴

Perceived advantages of MICS CABG are: while offering minimal invasiveness, it still allows harvesting and utilization of BITA; MICS CABG allows total arterial revascularization with the first and second conduits being BITA and with additional arterial grafts; because the incision avoids a median sternotomy, this approach likely reduces the risk of sternal wound complications significantly, supporting this approach in patients who were previously denied BITA revascularization due to the risk of sternal wound infection. Thus MICS CABG may be considered in patients with a high risk of mediastinitis, multivessel disease, and a strong desire for early societal return or improved cosmesis. MICS CABG may also be considered in high-risk older patients in view of the benefits conferred by the minimally invasive approach. ³⁴ As reported by McGinn and colleagues ⁶ in their initial series of MICS CABG, absolute contraindications include emergency surgery with hemodynamic compromise, severe chest wall deformity such as pectus excavatum, and severe lung disease precluding single-lung ventilation. In addition, MICS CABG should be avoided in patients with severely low ejection fractions (with possible need for emergency CPB), cardiomegaly (difficult conduit anastomosis), diffuse multivessel disease (difficult target identification), and history of chest radiation that raises the risk of significant adhesions (complicating BITA harvesting).

Surgical technique

Preoperative examinations, positioning, and techniques of BITA harvesting have been published previously.^10,35 Preoperatively, 3-dimensional computed tomography is routinely obtained to examine the ascending aorta, the lengths and courses of the BITA, and the anatomical relationship between the sternum and the right internal mammary artery (Figure 1). The imaging also aids in identifying the intercostal space above the apex to be opened. The patient is positioned in a 40-degree right lateral decubitus position with an infusion bag placed under the left scapula (Figure 2) to optimize BITA exposure. Techniques of harvesting BITA in MICS CABG have been published previously.^7,9,10,35 In brief, the use of retractors, stabilizers to depress the right lung, and patient positioning are noted to be important factors for creating the optimal anatomic configuration to conduct multivessel bypass surgery and proximal anastomoses (Figure 3). ⁷ The use of a long-tipped cautery device via a surgical port also aids in minimizing obscuring of the surgical field while providing enough reach to the proximal area of the right internal mammary artery. After harvesting BITA, the pericardium is opened completely from the distal side of the ascending aorta to the left apex and to the inferior vena cava along the diaphragm, followed by proximal anastomosis. For the proximal anastomosis, we utilize the method described in our previous report. ⁷ A flexible side-biting clump is placed on the ascending aorta (Figure 4), and the proximal anastomosis is hand sewn to the ascending aorta and tied with a knot pusher. The extensive retrosternal dissection conducted for right internal mammary artery harvesting aids in increasing the mobility of the heart and improving access for the distal anastomosis. Two sutures are placed in the left lateral side of the pericardium to retract the pericardium to the lateral chest wall. Two to three deep pericardial sutures are placed as in OPCAB to assist displacement of the heart (Figure 5). ⁹ For anastomosis of the obtuse marginal artery, the heart is displaced rightward, and for the posterior lateral artery, the heart is displaced in the right and cranial direction. For the posterior distal artery, the heart is displaced in the cranial direction and the left ventricle apex is elevated vertically. Distal anastomoses are completed as in off-pump CABG through the thoracotomy with this technique. Postoperative computed tomography-angiography is shown in Figure 6. Meticulous attention to the intraoperative net fluid balance is important to minimize distension of the heart, which would make the positioning and anastomosis challenging.

OPEN IN VIEWER

Figure 1.

Preoperative 3-dimensional computed tomography showing the relationship between the intercostal spaces (ICS) and the left ventricular apex.

OPEN IN VIEWER

Figure 2.

The patient is positioned in a 40-degree right lateral decubitus position with an infusion bag placed under the left scapula.

OPEN IN VIEWER

Figure 3.

A 32-cm dissecting-hook-type Harmonic Scalpel (Ethicon Endo-Surgery, Inc., NJ, USA) is inserted through the surgical port to harvest the internal thoracic arteries.

OPEN IN VIEWER

Figure 4.

The main pulmonary artery is retracted to the caudal direction with octopus NUVO stabilizer (Medtronic, Inc.). And proximal anastomosis is hand sewn to the ascending aorta with a Cygnet (Vitalitec, Inc., Plymouth, MA, USA) flexible side-biting clamp.

OPEN IN VIEWER

Figure 5.

Distal anastomosis is performed with 2 or 3 deep pericardial sutures placed as in off-pump coronary artery bypass, to assist displacement of the heart.

OPEN IN VIEWER

Figure 6.

Postoperative computed tomography showing well-patent bilateral internal mammary arteries: right internal mammary artery to the left anterior descending artery; left internal mammary artery to the obtuse marginal and posterior lateral artery.

Outcomes

Outcomes currently available on MICS CABG are limited to single- or dual-center experience with mostly early to mid-term follow-up, and all consist of SITA revascularization. The first large series of MICS CABG using SITA was reported in 2009 by McGinn and colleagues. ⁶ Excellent perioperative outcomes in this series of 450 patients include perioperative mortality of 1.3%, use of CPB in 8%, and a 4% rate of conversion to sternotomy. The average number of grafts was 2.1, with first being LIMA and the second being a venous or arterial conduit. A case-matched study between MICS CABG and OPCAB via sternotomy by Lapierre and colleagues ³² involving a subgroup of the above study showed 0% mortality in the entire cohort, 7% conversion to sternotomy in the MICS CABG group, and 2% on-pump conversion in the OPCAB group. A significantly lower rate of wound infection (0% vs. 4%, p = 0.03) and earlier return to full physical activity (12 days vs. >5 weeks, p < 0.001) were demonstrated in the MICS CABG cohort compared to the OPCAB cohort. In addition, excellent follow-up results of graft patency have been demonstrated: Ruel and colleagues ⁸ reported 100% patency of LIMA grafts and 92% patency of all grafts at 6 months following MICS CABG with SITA in 91 patients prospectively enrolled after their initial experience with the MICS CABG technique. The advantage of minimal invasiveness in MICS CABG may have a pronounced benefit in older patients. A subgroup analysis limited to patients older than 75 years of age, comparing MICS CABG to CABG with sternotomy, demonstrated significantly lower all-cause mortality in the MICS CABG group (19.7% vs. 47.7%, p < 0.001) with a multivariate model showing a hazard ratio of 0.51 (p = 0.04) associated with MICS CABG compared to CABG with sternotomy. This result is certainly consistent with the findings that minimally invasive techniques have a tendency to demonstrate greater benefits in high-risk groups. A study involving a general high-risk population not limited to the age cut-off is also anticipated. In our center, 27 cases of BITA MICS CABG have been performed. Multivessel revascularization was performed in 82% with the average number of distal anastomoses being 2.7 ± 1.2. By the design of the operation, the right internal mammary artery was commonly grafted to the LAD territory as an in-situ graft in the majority of cases, and the rest had a composite right internal mammary artery graft, with a Y-graft being the most common design. No perioperative mortality occurred in this series. Surgical site infection was observed in 3%, and all of these patients were female, presumably due to breast tissue covering the incision site.

Hybrid coronary revascularization

This technique that allows the use of BITA in MICS CABG may expand the existing armamentarium for coronary arterial revascularization, providing surgeons, cardiologists, and patients with an additional option. With the recent advancement in hybrid coronary artery revascularization with excellent outcomes,^5,36–38 MICS CABG using BITA provides a potential for further advancing the range of patients who would benefit from this by combining hybrid coronary artery revascularization for a more complete revascularization while maintaining minimal invasiveness. Indeed, there are patients who present with diffuse disease and complex vascular anatomy in the LAD that strongly argues for surgical revascularization while having short localized disease with clean distal vessels in the right coronary artery or circumflex system that are amenable to percutaneous coronary intervention. With the evolving paradigm of the heart team, discussions should take place on revascularization strategies for each coronary artery. ³⁹ With MICS CABG as a new alternative surgical revascularization strategy, patients may benefit from the optimal combinations of minimal invasiveness and the long-term benefit of BITA revascularization by advancing the collaborative approach in hybrid coronary revascularization.

Conclusions

Off-pump MICS CABG is a safe and feasible alternative to existing strategies of surgical coronary revascularization in selected patient populations. Existing reports, although limited in the quality of evidence, have demonstrated that BITA can be safely harvested in a reproducible fashion under direct vision via a small left thoracotomy. Extreme care during the internal thoracic artery harvesting and adequate retrosternal dissection are critical in achieving an optimal surgical field and ensuring excellent graft quality. In addition to access to BITA, the eliminated risk of sternal wound infection is a major advantage of this technique.