Single-Center Initial Experience with Robotic Assisted Minimally Invasive Coronary Artery Bypass Surgery

Abstract

Background

Robotic assisted minimally invasive direct coronary artery bypass grafting (RA-MIDCAB) represents an effective alternative to conventional coronary artery bypass grafting for revascularization of the left anterior descending artery (LAD) with the left internal thoracic artery (LITA). The first LITA-LAD bypass using a robot was performed in 1998. Since then, progress has been made, bringing the benefits of reduced surgical trauma and faster patient recovery. We report on our single-center initial experience of RA-MIDCAB surgery.

Methods

This retrospective study included the first 27 patients who underwent RA-MIDCAB between July 2024 and September 2025. Baseline characteristics, perioperative and postoperative outcomes were collected from institutional databases and patient records.

Results

All RA-MIDCAB procedures were successfully performed by 2 expert surgeons. Intraoperative graft patency was measured. No mortality was registered, neither on 30-day nor long-term. Conversion to sternotomy never occurred. The mean surgery time was 245.7 minutes. Early outcomes included postoperative new onset atrial fibrillation (n = 2; 7.4%) and respiratory failure (n = 1; 3.7%). Mean length of intensive care unit stay and hospital stay were, respectively, 2.4 and 3.9 days. No patients underwent repeat revascularization or suffered major adverse cardiovascular events.

Conclusions

RA-MIDCAB surgery can be performed safely with excellent results. Despite an initial investment, starting a RA-MIDCAB program improves quality of care.

Visual Abstract

In Short.

• ▪RA-MIDCAB is an effective alternative to conventional CABG for LAD revascularization.
• ▪At our center, the first 27 patients who underwent RA-MIDCAB showed no mortality and minimal complications rate.
• ▪Considering our initial outcomes, starting a RA-MIDCAB program should be encouraged in other centers globally.

Robotic assisted minimally invasive direct coronary artery bypass (RA-MIDCAB) integrates robotic left internal thoracic artery (LITA) harvesting with a left minithoracotomy for the left anterior descending artery (LAD) anastomosis, offering a less invasive alternative to conventional coronary artery bypass graft (CABG).

MIDCAB was developed to avoid sternotomy in isolated LAD disease through left anterior 4-6 cm minithoracotomy and hand-sewn anastomosis. MIDCAB has been combined with robotic internal thoracic artery harvest, creating RA-MIDCAB. This procedure is suited for cases involving isolated LAD disease and favorable coronary anatomy. Hybrid coronary revascularization (HCR) is increasingly utilized, combining robotic LAD grafting with percutaneous coronary intervention (PCI) to non-LAD vessels.

In 1998, the advent of the da Vinci led to the first robotic CABG.¹ In 2005, Subramanian and associates² reported excellent clinical outcomes with no mortality and robotically harvested 1 or both internal thoracic arteries. In the same year, Kiaii and colleagues³ performed and reported the first simultaneous robotically assisted HCR. In 2018, Patel and coworkers⁴ published a propensity-matched comparison showing RA-MIDCAB had notably lower blood transfusion rates and shorter hospital stays compared with conventional CABG. Nowadays, minimally invasive CABG enables multivessel, robotic, and/or hybrid revascularization with excellent outcomes.

In this context, we report our initial single-center RA-MIDCAB experience, highlighting perioperative outcomes and recovery time.

Patients and Methods

Study Design and Patient Population

This is a retrospective single-center study of the first patients who underwent off-pump RA-MIDCAB at South Shore University Hospital between July 2024 and September 2025. A total of 27 patients were included. Primary inclusion criteria were (1) patients undergoing RA-MIDCAB for revascularization and (2) availability of comprehensive perioperative and follow-up data. Procedures were performed by 2 expert surgeons.

Surgical Technique

The patient is positioned with the left chest elevated 30 degrees with the left arm tucked and the left shoulder abducted. A 1-cm incision is made in the third intercostal space, 2 cm anterior to the axillary line. The left lung is deflated and the pleural space entered bluntly with an endoscopic port trocar. The robotic endoscopic camera is inserted, and entry into the pleural space confirmed. The chest is insufflated with CO² to a maximum pressure of 8 mm Hg with flow of 6. The LITA is identified. The previously draped robotic surgical cart is transported to the operating room table and fixed in position. The central camera arm is affixed to the camera port, and the camera inserted. Next, two more 1-cm incisions are made, in the fifth and seventh intercostal spaces, slightly anterior to the camera port site, through which the right and left arms of the robotic system are inserted sequentially, under direct videoscope guidance. The arms are fitted with a scalpel cautery on the right and DeBakey micro forceps on the left. The LITA is dissected from the chest wall (Figure 1), with cautery and clipping of all branches. The artery is freed as far proximally. Distally, the LITA is dissected free to the seventh intercostal space, just proximal to its bifurcation. Heparin is administered and the LITA is clipped and transected and clipped to the pericardium.

Figure 1.

Left internal thoracic artery isolation from chest wall with micro forceps and cautery spatula.

The pericardium is opened. The anterolateral surface of the heart is visualized, allowing identification of the LAD. At this point, the robotic system is removed from the chest and the robotic surgical cart moved away from the operating room table.

A 4-cm anterolateral minithoracotomy is made in the inframammary line, and the fourth interspace is entered. Soft tissue retractors are placed and an Estech thoracic retractor is positioned. The distal LITA is identified. The LAD is identified. The vessel is stabilized with a Medtronic stabilizer, and an arteriotomy is made. The LITA-LAD anastomosis is performed. Blood pressure is well maintained throughout the anastomosis. Protamine is given. Medistim graft flow evaluation and pulsatility index measurements are made. An angled chest tube is inserted into the left chest via the lowest port. A Blake drain is placed in the upper port. The left lung is inflated, and the Blake placed above the lung and below the rib space. The deep fascia, subcutaneous tissues, and skin are closed in layers. Once stable, the patient is transferred to the intensive care unit (ICU).

In cases involving HCR, additional PCI to non-LAD lesions is performed either during the same hospitalization or in a staged fashion, 6 weeks before or after surgery, based on heart team assessment (Figure 2).

Figure 2.

Post robotic assisted minimally invasive direct coronary artery bypass left internal mammary artery–left anterior descending artery angiogram at staged percutaneous coronary intervention.

Data Collection and Outcomes

Baseline characteristics (Supplemental Table 1) and perioperative and postoperative data were retrospectively extracted from our institutional database. Perioperative variables included operative time, conversion to full sternotomy, graft patency, and intraoperative blood loss. Early postoperative outcomes, ICU and hospital stay, postoperative complications, and 30-day mortality (Table 1, Supplemental Table 2). Long-term outcomes, including long-term mortality (defined as >30-day), freedom from myocardial infarction, repeat revascularization, and major adverse cardiovascular events (MACE), were recorded (Supplemental Table 3). Mean follow-up for the first postoperative visit was calculated. Continuous variables are presented as mean and standard deviation (SD). Categorical variables are reported as number and percentage.

Table 1.

Perioperative Patients Outcomes

Surgery Time, min	Medistim Flow, mL/min	Medistim Pulsatility Index	ICU Stay, d	Hospital Stay, d	30-d Mortality	No-AF	Respiratory Failure
287	42	2.7	2	3	N	N	N
171	61	2.2	2	4	N	N	N
236	41	3.5	5	5	N	N	N
249	41	3.5	2	4	N	N	N
247	35	2.4	2	4	N	N	N
173	53	2.2	4	4	N	N	N
236	9	3.8	1	3	N	N	N
206	53	2.4	2	3	N	N	N
298	42	3.0	3	4	N	N	N
206	41	2.5	2	3	N	N	N
268	44	3.0	2	3	N	N	N
235	34	3.6	3	3	N	Y	Y
297	40	3.0	2	3	N	N	N
205	34	2.8	2	3	N	N	N
283	44	2.4	2	3	N	N	N
195	43	2.6	2	3	N	N	N
258	11	1.2	2	3	N	N	N
289	33	3.1	2	3	N	N	N
310	47	2.1	2	5	N	N	N
259	31	2.1	2	3	N	N	N
329	29	3.1	3	3	N	N	N
196	15	2.7	2	3	N	N	N
274	19	3.1	2	4	N	N	N
246	14	3.1	2	4	N	N	N
211	22	2.7	2	4	N	N	N
240	40	3.0	2	7	N	N	N
230	40	3.0	6	11	N	N	N

ICU, intensive care unit; No-AF, new-onset atrial fibrillation.

Results

Demographics

Between July 2024 and September 2025, a total of 27 patients (22 male, 81.5%; 5 female, 18.5%) underwent RA-MIDCAB at our institution. The mean age was 66.9 (SD, 9.9) years. The mean body mass index was 28.9 (SD, 3.9) kg/m². A total of 18 patients (66.7%) had history of smoking. The most frequent comorbidities included hypertension (n = 25, 92.6%), dyslipidemia (n = 24, 88.9%), diabetes mellitus (n = 12, 44.4%), atrial fibrillation (n = 5, 18.5%), and prior cerebrovascular accident (n = 5, 18.5%). A total of 6 patients underwent prior PCI (22.2%). The mean left ventricular ejection fraction was 59% (SD, 7.3%). A total of 9 patients (33.3%) underwent HCR.

Intraoperative Outcomes

In all cases, robotic assistance was used for LITA harvesting and a minithoracotomy was performed for LITA-LAD anastomosis. The mean operative time was 245.7 (SD, 41.9) minutes. Conversion to full sternotomy was never required. Mean intraoperative blood loss was 48.2 (SD, 39.1) mL. Mean Medistim graft flow and pulsatility index were 35.48 (SD, 13.2) mL/min and 2.8 (SD, 0.6), respectively.

Postoperative Outcomes and Follow-up

The mean ICU stay was 2.4 (SD, 1.0) days, and mean hospital stay 3.9 (SD, 1.7) days. Postoperative new onset atrial fibrillation occurred in 1 patient (3.7%), and 1 patient (3.7%) required chest tube placement for left pleural effusion. Myocardial infarction, stroke, renal failure and 30-day mortality did not occur. The first clinical follow-up was completed in all of patients at a mean of 13.1 (SD, 4.4) days, with all patients returning to regular daily activities and tolerable pain levels through the postoperative period. Further follow-up, at a mean of 7.2 (SD, 5.0) months, demonstrates 100% overall survival with complete freedom from myocardial infarction, repeat revascularization, and MACE.

Comment

RA-MIDCAB is a major advancement in coronary revascularization. In our single-center experience, program initiation was associated with excellent clinical outcomes, including no 30-day mortality and no long-term mortality. Mean ICU and hospital stay were 2.4 and 3.9 days, respectively. New onset atrial fibrillation occurred in 1 patient, and 1 patient required chest tube placement for left pleural effusion. At the first follow-up, patients reported prompt return to normal activities and high satisfaction with the minimally invasive approach.

Compared with conventional MIDCAB, RA-MIDCAB enables longer grafts and facilitates bilateral mammary revascularization. The technique is associated with less postoperative pain, fewer pulmonary complications, and shorter ICU and hospital stay.⁵ Robot utilization enhances MIDCAB by enabling precise 3-dimensional LITA harvesting with minimal incision and rib trauma, preserving graft integrity. This leads to significantly reduced postoperative bleeding, as noted in propensity-matched comparisons, even in the early phases of the learning curve.⁶

A meta-analysis of 29,523 patients by Hwang and associates⁷ assessed the outcomes of robotic assisted CABG. The analysis revealed a low conversion rate to full sternotomy of 3.2% (95% CI, 2.1%-5.2%; I², 39%). At a mean follow-up of 5.2 years, overall graft patency was 96%. Freedom from MACE was 83.2% (95% CI, 72.0%-90.4%; I², 90%) for RA-MIDCAB. These findings suggest that robotic assisted CABG offers acceptable perioperative and mid-to-long-term outcomes with promising graft patency.

RA-MIDCAB programs require an initial financial investment and learning curve. Gianoli and colleagues⁸ found that RA-MIDCAB was 2.8%-10.3% more expensive than conventional CABG, but higher procedural costs were largely offset by postoperative savings, resulting in minimal impact on hospital resources. Patrick and coworkers⁹ analyzed 1195 RA-CABG cases by 114 novice surgeons, revealing a steep early learning curve, where complications rates declined and procedural success improved by the tenth case.

In addition, starting a RA-MIDCAB program allows hybrid strategies in multivessel coronary disease. Although rates of major complications such as stroke, myocardial infarction, and mortality remain comparable, HCR offers perioperative advantages over conventional CABG, including reduced blood transfusions and shorter ICU and hospital stay.¹⁰ Nevertheless, to define the optimal approach in multivessel disease, adequately powered randomized trials comparing CABG, HCR, and PCI are necessary.

RA-MIDCAB innovated coronary revascularization by reducing trauma, implementing recovery, and improving quality of life. We encourage other centers to invest in developing and expanding robotic coronary programs, ideally following structured training supported by an experienced robotic surgeon. Such a pathway can accelerate the learning curve and ensure safe, reproducible, and high-quality outcomes.