Reoperative cardiac surgery, performed via redo sternotomy, poses technical challenges not experienced when doing primary sternotomy cardiac operations. Redo median sternotomy is associated with increased risk of graft injury and inadvertent cardiotomy, potentially causing serious bleeding and other complications.1
In an attempt to reduce the procedural risks of repeat sternotomy, robotic and other minimally invasive approaches have been used for patients needing primary mitral valve repair or replacement in the setting of previous sternal incisions and previous bypass operations.2 Although the complexity of repeat valve procedures has thus far precluded a robotic approach at many centers, increased experience with minimally invasive techniques and instrumentation, along with advances in perfusion and myocardial protection strategies, has allowed us and other groups to expand the use of robotics to these more challenging patients,3,4 as described in the following case.
In July 2013, a 53-year-old man (height = 188 cm; body mass index = 33) presented with heart failure (New York Heart Association class III). For reasons that are unclear, the patient received an Edwards 29-mm pericardial mitral prosthesis at another institution 10 years ago, via sternal incision, for mitral insufficiency secondary to fibroelastic deficiency disease (Carpentier type II). Left and right heart catheterization revealed normal coronary anatomy, an estimated left ventricular ejection fraction of 45%, pulmonary arterial pressure of 100/35 mm Hg, and pulmonary capillary wedge pressure of 34 mm Hg. The mitral valve gradient was calculated at 16 mm Hg and valve area at 0.63 cm2. Transesophageal echocardiography showed severe central mitral insufficiency without paravalvular leak. Preoperative computed tomography angiography revealed no aortic or iliofemoral pathology. Comorbidities included hypertension, diabetes mellitus, and chronic renal insufficiency with a baseline creatinine of 1.8 mg/dL.
Under general double-lumen endotracheal anesthesia, bilateral brachial arterial lines were placed for monitoring endoaortic balloon deployment and clamping. A right internal jugular cannula was placed for retrograde cardioplegia, as well as a right internal jugular pulmonary artery vent. The patient's right shoulder was elevated to 15 degrees, and the right chest and both groins were prepared into the operative field.
All cannulation was performed using a modified Seldinger technique through polypropylene purse-string sutures (Prolene; Ethicon, Somerville, NJ USA). The right femoral vein was cannulated with a 25F venous cannula (Edwards Lifesciences Corp, Irvine, CA USA) and advanced into the superior vena cava. The right femoral artery was cannulated with a 23F cannula (Edwards Lifesciences Corp) with a side arm, through which an endoaortic occlusion balloon (IntraClude, Edwards Lifesciences Corp) was advanced into the ascending aorta and prepositioned for later use.
Before port placement, the right lung was dropped and single, left-lung ventilation was initiated. Simultaneous to exposure of the femoral vessels, a 5-mm camera port was created in the right fourth intercostal space to allow inspection of the pleural space for adhesions and for planning the placement of the remaining ports. This was later enlarged to accommodate a 12-mm port for the robotic camera, and the remaining ports were placed under direct vision. Three 8-mm ports were created in the second, sixth, and fifth intercostal spaces for the left, right, and third (retractor) robotic arms, respectively. A 2-cm working port placed 2 cm lateral to the camera port, also in the fourth intercostal space, allowed for a totally endoscopic, port-access approach (Fig. 1A).
After placement of the ports and cannulation of the femoral vessels, the robot (DaVinci S; Intuitive Surgical, Sunnyvale, CA USA) was docked. Pleural adhesions were dissected robotically using a Bovie paddle, aided by deflation of the right lung, with carbon dioxide inflation under 15 mm Hg of pressure. Thymic and pericardial fat were excised. Full cardiopulmonary bypass was initiated and the temperature allowed to drift with a final core temperature of 32°C or greater. The pericardium was entered approximately 3 to 4 cm anterior to the right phrenic nerve and dissected free from the diaphragm to the ascending aorta, creating a flap to hold the lung back and to facilitate instrument exchange into the left atrium.
The balloon was inflated to occlude the aorta. The heart was arrested with 1 L of antegrade blood cardioplegia. Additional doses of cardioplegia were delivered every 12 to 15 minutes through the retrograde cannula. The working port and camera port were united and a soft-tissue retractor (Alexis, Applied Medical, Rancho Santa Margarita, CA USA) placed to allow for adequate working space (Fig. 1B). A small rib spreader was used to separate the ribs enough to allow the old prosthesis to be extracted and to deliver the new one through the chest wall, but the remainder of the operation was accomplished without the rib spreader.
A vertical left atriotomy was created beginning at the right superior pulmonary vein and carried around behind the inferior vena cava. The dynamic left atrial robotic retractor was advanced into the left atrium to expose the degenerated mitral valve prosthesis (Fig. 2). The prosthesis was resected using a robotic Bovie paddle set at 70 coag, beginning at the surgeon's 2 o'clock position near the right fibrous trigone to create atrial-ventricular communication. Then, hugging the prosthesis, the valve was excised using a robotic needle holder in the left arm and robotic scissors in the right arm. Residual debris, including old pledgets, was mechanically extracted using the robotic instruments and passed to the tableside assistant. The most difficult and challenging portion of the removal was scissor excision of the old prosthesis. Using the Bovie to create an initial plane was critical. A 27-mm Mosaic bioprosthesis (Medtronic, Inc, Minneapolis, MN USA) was sutured to the annulus with 2–0 polyester pledgeted mattress sutures (Ethibond; Ethicon Inc, Somerville, NJ USA). The prosthesis was seated and secured with a knot-tying device (COR-KNOT; LSI SOLUTIONS, Victor, NY USA). The left atrial appendage was oversewn and the left atriotomy closed with double-layer 2–0 expanded polytetrafluoroethylene suture (Gore-tex; W. L. Gore & Associates, Inc, Flagstaff, AZ USA) over a vent. Two chest tubes were placed through the right- and third-arm ports. Cross-clamp and cardiopulmonary bypass times were 91 and 145 minutes, respectively. Total operative time was 192 minutes.
Recovery was uneventful and the patient was discharged on postoperative day 6. Follow-up echocardiography in January 2017 showed a stable and well-functioning prosthesis with no leak and no stenosis. The patient remains alive and well with no symptoms of heart failure (New York Heart Association class I) as of March 30, 2017.
In centers experienced with the technology, robotic-assisted cardiac surgery has been shown to shorten hospital length of stay and allow for an earlier return to full activity than traditional cardiac surgery.5
Because the first robotic-assisted, primary mitral valve repair was reported by Mohr et al6 in 1999, such procedures have represented most robotic cardiac operations, along with a growing number of primary coronary-artery revascularizations.7,8 As reflected in the literature, however, development of robotic-assisted approaches for more complex procedures or those in patients with history of cardiac surgery has generally been approached slowly and with understandable caution.
With experience and using techniques for remote perfusion and myocardial protection, we and other groups have begun to expand the role of robotic approaches into complex operative and clinical situations.3,4 A detailed discussion of the equipment, training, patient selection, techniques, and potential complications associated with robotic mitral valve surgery is beyond the limited scope of this case report, but readers interested in learning more about these aspects of robotic-assisted surgery are directed to Lehr et al.9
Although they did not occur in this case, complications of potential concern would be atrio-ventricular separation or injury to the circumflex coronary artery. In the event of circumflex injury, our strategy would be to convert to a sternotomy and perform distal bypass. For an acute atrio-ventricular dissociation, we would attempt an intracardiac repair via robot and, if unsuccessful, convert to a sternotomy. Our group has successfully repaired one chronic atrio-ventricular dissociation and left ventricular pseudoaneurysm with the robot, using a Dacron graft.
A 2013 literature review by Botta et al2 reported good results for minimally invasive but nonrobotic repeat mitral valve procedures by a limited right thoracotomy in patients with a previous open-heart surgery. The mean ± SD early mortality among the 10 papers analyzed was 5.7 ± 2.3%, and other clinical outcomes were comparable with or better than traditional repeat sternotomy. However, we were unable to identify any reports of redo mitral valve replacement by a robotic-assisted, totally endoscopic, port-access approach.
Removal of the previous prosthesis was challenging in the setting of dense postoperative adhesions and a limited operative field, but we found that visual detail was excellent and that our robotic approach was suitable for excision and replacement of the prosthesis. We therefore conclude that reoperative mitral valve surgery can be performed safely with the robotic approach, in selected patients, in experienced centers. Additional experience with this approach will be needed to objectively assess its advantages and disadvantages relative to a sternotomy or nonrobotic thoracotomy approach. In addition, the potential for transcatheter mitral approaches may further complicate the choice for redo mitral valve replacement.
The authors thank Jeanne McAdara, PhD, for professional assistance with article preparation and Jill Rhead, MA, CMI, FAMI, for medical illustrations.
1. Kim DC, Chee HK, Song MG, et al. Comparative analysis of thoracotomy and sternotomy approaches in cardiac reoperation
. Korean J Thorac Cardiovasc Surg
2. Botta L, Cannata A, Bruschi G, et al. Minimally invasive approach for redo mitral valve surgery. J Thorac Dis
. 2013;5(Suppl 6):S686–S693.
3. Lewis CT, Stephens RL, Tyndal CM, Cline JL. Concomitant robotic mitral and tricuspid valve repair: technique and early experience. Ann Thorac Surg
4. Nifong LW, Rodriguez E, Chitwood WR Jr. 540 consecutive robotic mitral valve repairs including concomitant atrial fibrillation cryoablation. Ann Thorac Surg
5. Lehr EJ, Rodriguez E, Chitwood WR. Robotic cardiac surgery. Curr Opin Anaesthesiol
6. Mohr FW, Onnasch JF, Falk V, et al. The evolution of minimally invasive valve surgery—2 year experience. Eur J Cardiothorac Surg
7. Canale LS, Mick S, Mihaljevic T, Nair R, Bonatti J. Robotically assisted totally endoscopic coronary artery bypass surgery. J Thorac Dis
. 2013;5(Suppl 6):S641–S649.
8. Kayatta MO, Halkos ME. Reviewing hybrid coronary revascularization: challenges, controversies and opportunities. Expert Rev Cardiovasc Ther
9. Lehr EJ, Guy TS, Smith RL, et al. Minimally invasive mitral valve surgery III: training and robotic-assisted approaches. Innovations