Abstract
Background: Automated distal connecting devices have been recently introduced to facilitate coronary anastomosis. This could have a large impact on the capacity of robotic systems to perform completely endoscopic off-pump bypass, where the quality of anastomosis and the prolonged operative time for the performance of the anastomosis have until now been cause for concern. Our group tried to determine the feasibility and efficacy of the JoMed distal graft connector using the ZEUS robotic system.
Methods: Six swine, with a mean weight of 25.8 (standard deviation [SD] 2.2) kg, underwent endoscopic off-pump internal thoracic artery–left anterior descending (ITA-LAD) anastomosis with a special stabilizing system using ZEUS robotic assistance. The anastomosis was performed with the JoMed distal connector.
Results: The connector was employed successfully in 4 of 6 cases using a special delivery instrument. Two animals fibrillated within 2 minutes after the application of proximal occluding snares and were excluded from the analysis. The total device deployment time was 2 minutes 4 seconds (SD 50 s) in 4 of 6 survivors, which remained hemodynamically stable and in sinus rhythm until euthanasia. Coronary angiography and transonic flow measurements were used to verify patency.
Conclusion: The JoMed distal graft connector may facilitate the use of robot-assisted endoscopic bypass on a beating heart. Long-term patency issues will need to be assessed.
Several attempts have been made recently to introduce automated anastomotic devices to facilitate coronary anastomosis.1–4 Successful, reproducible, rapid and safe deployment of these devices could have a huge impact on the capacity of robotic systems to perform completely endoscopic beating-heart multivessel bypass. The quality of anastomosis constructed on a beating heart and the prolonged operative times needed for the performance of endoscopic coronary anastomosis have been cause for concern.5 Our group tried to determine the feasibility and efficacy of using a recently developed distal graft connector (JoMed Inc., Helsingborg, Sweden) that allows expeditious sutureless anastomosis between blood vessels using the ZEUS (Computer Motion, Goleta, Calif.) robotic system.
Methods
The graft connector is a dedicated coronary artery nitinol stent, covered by a 10-μm thick layer of polytetrafluoroethylene (PTFE), with a side branch or tower of 4-mm PTFE graft into which the internal thoracic artery (ITA) is inserted and fixated by a ligature. The covered stent is kept in a crimped loaded configuration (1-mm internal diameter) and is inserted into the recipient coronary artery through a 4–6-mm arteriotomy. Once in position, the release mechanism is activated and the covered stented graft expands and is held fixed in the coronary artery. The shunt comes in varying sizes to accommodate vessels ranging from 1.5 to 3.5 mm.6
Six swine with a mean weight of 25.8 (standard deviation [SD] 2.2) kg were selected, and general anesthesia was induced with ketamine, 10 mg/kg, and maintained with isoflorane. Electrocardiographic and O2 saturation monitoring was done continuously. At the end of the experiment, euthanasia was performed. All animals received humane care in compliance with the Guide for the Care and Use of Laboratory Animals (www.nap.edu/catalog/5140.html), and the experiment was approved by the animal ethics committee of Case Western Reserve University, Cleveland, Ohio.
Each animal was placed in a supine position and a partial lower sternotomy performed. The ITA was harvested under direct vision. Then, 100 IU/kg of heparin was injected and 1 mg/kg of lidocaine bolus given, and the sternotomy was closed.
The ITA pedicle was brought out through a stab wound and prepared and mounted onto the graft connector manually outside the chest. Three ZEUS robotic arms were then fixed on the operating table to hold an endoscope and 2 endoscopic instruments. This has been described previously.4 The mean set-up time was 4.2 (SD 0.6) minutes. A 10-mm endoscope and two 5-mm endoscopic instruments were introduced into the left hemithorax to make an isosceles triangle. The mean port placement time was 11 (SD 1.7) minutes. A fourth port via the second intercostal space was used to insert a special endostabilizer, which was positioned to straddle the left anterior descending coronary artery (LAD). A proximal occluding snare of 4′0′ polypropylene (Ethicon, Somerville, NJ) was passed around the LAD and brought out through a small stab incision and the snare tightened. The arteriotomy was then carried out using a special articulating arteriotomy knife. The ITA pedicle with the graft connector in place was re-introduced into the thoracic cavity and, using a special delivery device attached to the robotic arm via a 5-mm port (Fig. 1), was manipulated into the arteriotomy and the device deployed (Fig. 2). The occluding clip from the ITA pedicle was released, and the proximal snare was left in place while flow measurements and coronary angiography were carried out.
Results
The connector was deployed successfully in 4 of 6 cases. Two animals fibrillated within 2 minutes after the application of the proximal occluding snares and were excluded from the analysis. The mean device implantation time was 2 minutes 4 seconds (SD 50 s) in the 4 survivors, and they remained hemodynamically stable after heparin reversal and in sinus rhythm until euthanasia 1 hour later. There were no anastomotic bleeding complications, and the patency of the shunt was confirmed at autopsy in all 4 swine using probes and saline injection through the ITA graft. In 1 animal, transonic flow measurements (Transonic Systems, Ithaca, NY) demonstrated a flow of 40 mL/min (Fig. 3), and this was confirmed by intraoperative angiography using a digital fluoroscopy system (Fig. 4) (Series 9800, Mobile C-Arm OEC Medical Systems, Salt Lake City, UT).
Discussion
Robot-assisted microsurgery systems have helped to increase the precision and scope of endoscopic cardiac surgery by improving ergonomics, eliminating tremor, providing motion scaling and intrathoracic articulation, and allowing the surgeon to have a magnified, 3-dimensional, high-resolution image.7,8
Recent reports of totally endoscopic coronary artery bypass (TECAB) with robotic assistance have been encouraging.9,10 However, with the present techniques, it is still not possible to achieve the goal of completely endoscopic multivessel bypass on a beating heart in a safe and reproducible manner. The conversion rate for totally endoscopic single-vessel bypass on a beating heart was 64% in Boyd’s experience.9
In a recent FDA-approved multicentre prospective randomized trial of robot-assisted versus conventional bypass, the time needed to perform anastomosis using robotic assistance was 62.7 (SD 25) minutes versus 6.5 (SD unavailable) minutes in the control arm (beating heart group) (p = 0.03) and 28.5 (SD 5) minutes and 10.2 (SD 3) minutes, respectively (stopped heart group) (p = 0.003).11 These times are clearly unacceptable.
The introduction of single-shot anastomotic devices may enhance the success of TECAB, if they allow a high-quality anastomosis to be carried out in a less technically demanding manner, expeditiously with specially designed robotic instruments.
The shunts come in sizes that accommodate vessels ranging from 1.5 to 3.5 mm, making them potentially useful clinically, because the large size of previous systems has been a limitation. In the current experiments, all the shunts were 2.5 mm in size.
Tozzi and colleagues12 demonstrated earlier that off-pump coronary artery bypass grafting could be carried out successfully with excellent results at 6-month angiographic follow-up in the sheep model. These experiments were conducted on cardiopulmonary bypass through a sternotomy, and the shunt was manually deployed. Histological examination showed 100% angiographic patency with some increase in the thickness of myointimal hyperplasia between the graft connector and the control group, but because of the overall oversizing and distension of the anastomotic site in the graft connector group, there was no difference in luminal width at histology. (1.7 [SD 0.2] mm v. 1.6 [SD 0.1] mm).
The present study demonstrated that endoscopic beating-heart single-vessel ITA–LAD anastomosis could be successfully and effectively done with robotic assistance in under 3 minutes, though long-term patency issues in humans will need to be addressed by clinical trials in animals and then in humans.
Footnotes
Competing interests: None declared.
- Accepted October 5, 2004.