Minimally Invasive and Robot-Assisted Mitral Valve Surgery

W. Randolph Chitwood Jr, MD, FACS, FRCS


In the last 25 years the treatment of mitral valve disease has advanced markedly. No longer can we rely on a valve replacement or simple repair to sustain us as comprehensive cardiac valve surgeons. Recent advances in catheter-based mitral repair and replacement techniques, albeit not perfect, have foisted the necessity to be able to perform minimally invasive operations. Moreover, this combined with patients being referred earlier for mitral surgery requires us to have similar or better clinical results than conventional operations.

Herein, the most up to date information regarding minimally invasive mitral valve surgery (MIMVS) is conveyed and should be a guide for residents in training as well as established surgeons. Learning pathways for both videoscopic and robot- assisted MIMVS are included. We also focus on the preoperative selection of patients, imaging technology, anesthesia methods, surgical set-up, cardiopulmonary perfusion, complications, and current outcomes. Direct, videoscopic (endoscopic), and robot-assisted visualization discussions should help surgeons select his/her planned minimally invasive approach. Most surgeons have abandoned hemi-sternotomy MIMVS, and thus, this chapter will focus mainly on mini-thoracotomy approaches. Concomitant procedures, including the Cox-MAZE operation to treat atrial fibrillation and our approach for minimally invasive tricuspid surgery are discussed briefly. Other chapters in this book cover detailed mitral and tricuspid valve repair techniques; however, specific repair and replacement methods that we use to facilitate MIMVS are included.

Descriptive text is not available for this image


Very few things in cardiac surgery are completely new or un-discovered. T. Lauder Brunton first suggested that mitral valve stenosis could be treated surgically.[1] The thoracotomy for mitral surgery has been used since Sir Henry Souttar’s time (1925) when he performed the first “finger” fracture commisurotomy. [2]Smithy,Bailey, Harken, Lillehei, Morrow and many others used either a right or left thoracotomy for mitral commissurotomies and early valve replacements. [3],[4],[5],[6],[7] Many of us learned to do complex mitral reoperations through a right thoracotomy and found that we were looking directly at the mitral valve.[8] Despite these facts, the sternotomy has prevailed as the “standard approach” for mitral and tricuspid valve surgery.

The first coronary angioplasty by Gruntzig (1979) heralded the dawn of minimally invasive methods for treating cardiovascular disease. [9] In 1991 catheter-based aortic occlusion was conceived and developed by an innovative group of surgeons at Stanford University, and the Heartport company was founded subsequently by them. By 1995 they were able to apply these concepts during coronary and mitral valve surgery. With the evolving “Heartport Platform”, the heart now could be arrested using an occlusive balloon aortic “Endoclamp” and catheter-based coronary sinus cardioplegia. Mitral valves could be repaired and replaced through small port incisions, using newly developed long-shafted instruments. [10],[11] Soon thereafter, Cosgrove and Cohn independently developed hemi-sternotomy approaches to MIMVS, using conventional surgical instruments and modified perfusion methods. [12],[13] During this period, antagonistic views were published that cautioned surgeons regarding minimally invasive cardiac surgery.[14],[15]Nevertheless,subsequent large series proved that minimally invasive mitral valve surgery could be done safely and with comparative results to conventional techniques. [16],[17],[18],[19],[20]Later, both propensity matched studies and meta-analyses confirmed that there were similar operative risks and outcomes with both sternotomy-based and minimally invasive methods. [21],[22]

In 1923 Duff Allen and Evarts Graham originated the concept of operating inside the heart using videoscopic vision.[23] Harken experimented with intra-cardiac visualization techniques in 1943.[24],[24]In 1958 Sakakibara predicted that valve operations could be done using videoscopic secondary vision. [25] Kaneko (1995) used video-assistance through a sternotomy to aid in mitral repairs and commissurotomies.[26] In the spring of 1996 Carpentier performed the first videoscopic MIMVS in which a simple ring repair was done through a mini-thoracotomy under hypothermic ventricular fibrillation. [27] Later that year, our group developed a specialized trans-thoracic aortic clamp. Using this clamp, we first replaced a mitral valve under videoscopic vision using antegrade cardioplegia.[28],[29] Several subsequent series showed that videoscopic vision could be a facilitating adjunct to MIMVS. [17],[19],[30] In these series mitral repair results were excellent, and the method was proven to be safe. Recent comparisons between the balloon EndoClamp (now Intraclude™ Edwards Lifesciences, Inc, Irvine, Calif.) and trans-thoracic aortic clamp occlusion have shown both methods to be equally safe and effective. [31],[32],[33]Nevertheless, the clamp technique does provide an economic advantage with simpler application and less need for monitoring.

In 1995 the Intuitive Surgical Company was founded after acquiring patents for early robot prototypes from the Stanford Research Institute. In late 1997 the author first tested the early daVinci™ robot prototype with the notion that it could be applied in MIMVS. In early 1998 Carpentier (Paris) and Mohr (Leipzig), did the first robotic mitral repairs using the same prototype. [34],[35] The author was present in Leipzig for many of these seminal operations. In 1999 our group purchased the first commercial daVinci™ surgical system in the United States and began laboratory development of our approach to robotic mitral valve repairs. The first of two FDA clinical trials began with our first operation, which was done in May of 2000. [36],[37] Both safety and efficacy as well as a multicenter investigational device exemption trial proved that mitral repairs could be done safely using the first daVinci™ surgical robot, and the device was FDA approved for intra-cardiac use in 2002. To date, the daVinci™ Surgical System is the only surgical robot FDA approved for intra-cardiac surgery. Our inaugural series showed that robotic mitral surgery was safe and could render equivalent outcomes to conventional operations. [38],[39]Since then many centers in the United States have proven safety, efficacy, and excellent outcomes in large groups of patients, establishing robot-assistance as a standard of care for many experienced mitral surgeons.[40],[41],[42],[43] Dedicated robotic referral programs have shown outcomes similar to operations done either through a hemi-sternotomy or a minimally invasive thoracic incision. The economic costs have been challenged; however, surgeons at the Mayo and Cleveland Clinics optimized robotic care pathways, which rendered economic parity with other incisional approaches. [44],[45]

Some might think that the historical development of surgical techniques, such as for MIMVS, should be relegated to minor footnotes. Nonetheless, this over 25-year evolution has placed surgeons in a better position to embrace the future requirement for the least invasive cardiac therapies. This aforementioned progress has brought us port-like incisions, new perfusion and aortic occlusion techniques, catheter-based operator Seldinger wire skills, long-shafted instruments, smaller retractors, 2-D and 3-D videoscopic vision, and 3-D robotic vision with ergonomic robotic wristed instruments. I believe that 2-D or 3-D videoscopic vision has become the new major "touchstone" for facilitating all types of minimally invasive mitral valve surgery.

Learning Pathways

The basic operative pathway, or sequence, is the same for either minimally invasive or robot-assisted mitral valve surgery. Although the operative setup and instruments are different, the sequence in performing these operations is very similar. Mastery of the operative sequence for MIMVS outlined in Figure 1 will place learning surgeons at a great advantage before proceeding to robot-assisted surgery (Figure 2).

Figure 1
Descriptive text is not available for this image
Learning Pathway for Minimally Invasive Mitral Valve Surgery.
Figure 2
Descriptive text is not available for this image
Learning Pathway for Robot-assisted Mitral Valve Surgery.

Direct Vision and Videoscopic

Surgeons beginning to learn the nuances of MIMVS should have prior experience with replacement and repair techniques that include echocardiographic imaging and planning. Moreover, it is most advantageous to learn from an experienced minimally invasive surgeon. For the first operations one should enlist an experienced proctor, no matter the level of prior surgical experience. There are several important MIMVS technique differences from mitral valve operations done through a full sternotomy. These include: 1) patient positioning. 2) anesthesia preparation 3) the size and location of the incision 4) instruments 5) the method of visualization 6) cannulation for cardiopulmonary perfusion 7) the perfusion circuit 8) cardiac exposure 9) aortic occlusion and cardioplegia administration 10) mitral valve exposure and 11) repair and replacement techniques. Each of these steps should be mastered through a stepwise learning pathway. (Figure 1) Later in this chapter, a more detailed procedural workflow is described for both videoscopic and robot-assisted MIMVS. As mentioned, most cardiac surgeons have abandoned the hemi-sternotomy when performing MIMVS and now prefer a right-sided mini-thoracotomy, which is sometimes is called a “working port”, depending upon the incision size. It is best to learn to do these minimally invasive mitral valve operations first using direct vision, through a slightly larger incision, before attempting either videoscopic or robot-assisted methods.


The learning sequence for robotic mitral surgery should begin as listed previously for MIMVS. However, additional maneuvers also include: 1) instrument and camera port placement 2) surgical cart to operating table docking 3) instrument arm engagement, deployment, and exchange 4) surgeon console operation and tele-manipulation 5) overcoming ergonomic pitfalls and 6) troubleshooting robot defaults. After the basics of MIMVS have been mastered and applied both safely and effectively in a number of patients, surgeons should be ready to begin the robotic training sequence outlined in Figure 2.

System-based training with the daVinci™ device should be done first and under the direction of a qualified trainer. Generally, this is done at a dedicated training center. This part should include working with simulation platforms to develop surgeon efficiency and ergonomics. Thereafter, these evolving skills should be translated to the surgeon’s operating console. At the same time, the tableside assistant and nursing staff should have hands-on experience in operating the surgical and vision carts cart as well as all aspects of instrument engagement, insertion, and activation.

The next part of training is procedure-based and should be done using animal and cadaver platforms. During this part of training, learners specifically focus on mitral valve repair. Generally, this involves the entire dedicated team, which should include the surgeon leader, a tableside assistant, a “scrub” technician or nurse, a circulating nurse, an anesthesiologist, and a perfusionist. Procedure training promotes team “synchrony”, which later generally translates into efficiency with decreased operative times. This part of training should be combined with actual operative case observations, either at the same time or shortly thereafter. Before applying robotic methods, the team should master needed peripheral perfusion techniques as well as operating through minimally invasive incisions along the previously described learning pathway. After team comfort has been achieved, progression to robotic mitral valve operations is reasonable and should be gratifying. An experienced surgeon proctor should be present for several early operations. It is important to remember that robotic instruments are only access tools and not the operation. To date, no absolute guidelines for robotic program development have been published: however, two recent publications relate optimal pathways for establishing one. [46],[47]The foundations of both the Society for Thoracic Surgeons and American Association for Thoracic Surgery now are sponsoring Cardiac Surgery Robotic Fellowships, which are funded by educational grants from the Intuitive Surgical company.


For years airline pilots have been required to undergo airplane specific simulation training before they are allowed to enter the flight deck. As their tolerance for pilot error is 0%, we must have the same mission as the lives of our patients are in our surgical hands. In 2008 representatives from six professional cardiothoracic surgical societies met at Harvard University along with individuals from industry and the NHLBI to evaluate simulation as part of the training paradigm for young surgeons. Their publication entitled Envisioning Simulation in the Future of Thoracic Surgical Education was the springboard for technologic, procedural, and situational simulation training pathways. [48] The American Board of Thoracic Surgery now requires that residents undergo 20 hours of simulation training in cardiothoracic training. A CTS Net editorial (2009) by M. Blair Marshall, MD entitled Simulation in Cardiothoracic Surgery stated, “Simulation in cardiothoracic surgery will become a staple in cardiothoracic education of the future. Residents will one day be able to practice repeatedly outside of the operating room prior to being expected to produce a virtuoso performance during an operation.” [49]

Videoscopic Simulation

Recently, surgeons at the University of Maastricht developed an advanced simulator for training in videoscopic MIMVS. [50] (Figure 3 A and B) The European Society for Cardiothoracic Surgery now is sponsoring simulation programs there for surgeons desiring to learn videoscopic MIMVS. This simulator consists of a polymer thoracic shell that has access ports for long shafted instruments to be passed to a mitral valve model. Echocardiographic reconstructions of actual degenerative mitral pathology are 3-D printed, and these silastic models, along with subvalvular structures, are placed in the simulator. As surgeons begin to learn their first attempts at a repair, their hand instrument maneuvers are computer registered. The simulator program registers needle angles and penetration depth, ergonomic trajectories, speed, and accuracy. Then, by reviewing an output report card, mid-course technical corrections can be made. After multiple videoscopic directed sessions, most learning surgeons become quite adroit at repairing leaflet pathology and implanting annuloplasty rings as well as placing ePFTE neochords between model papillary muscles and leaflets.

Figure 3A
Descriptive text is not available for this image
Minimally Invasive Mitral Valve Repair Simulator: This minimally invasive mitral valve repair simulator was developed at the University of Maastricht. Three-dimensional printed silicone copies of actual mitral valve pathology emulate different repair challenges.
Photograph by the author (WRC).
Figure 3B
Descriptive text is not available for this image
A Surgeon Practicing with the Simulator: Computer metrics are displayed for different tasks, such as suturing and cutting. These include measurements of speed and accuracy as well as needle trajectory and tissue bite depth characteristics.
Photograph by the author (WRC).

Robot-Assisted Simulation

MIMIC Technology, Inc. (Seattle, WA) has developed two simulation platforms for the daVinci™ Surgical System. The Skills Simulator™ and newer SimNow™ “backpacks” can be mounted on the surgeons’ console for both the daVinci™ SI and XI robots. (Figure 4 A and B) The MIMIC dv-Trainer™ has a dedicated operative console that directly emulates hand motions of the surgeon operator and table-side assistant. (Figure 5) These simulators are designed to train surgeons in overall system settings and camera control as well instrument arm clutching, and end effector manipulation. Moreover, fine instrument maneuvers focus on needle passage, tissue dissection and deformation, cautery application, and positioning materials within the surgical field. Each of these tasks are computer assessed, and a skills progress report is provided. Not only are the quality of specific tasks assessed, but also accuracy, speed, and ergonomics are evaluated. Experienced robotic surgeons often find that their accuracy and speed are very good but at the cost of inferior ergonomics of both external hand motions and instrument tips. With these simulators one soon learns that the daVinci™ clutching mechanism should be engaged often to restore an ideal hand and instrument arm attitude within the operative field. Both the Intuitive Skills Simulator™ and the MIMIC dv-Trainer™ have been shown to provide equivalent robotic training advantages. [51],[52]

Figure 4A
Descriptive text is not available for this image
Intuitive Skills™ Backpack Simulator.
Reproduced with permission from Intuitive Surgical Inc., Sunnyvale, Calif.
Figure 4B
Descriptive text is not available for this image
Intuitive SimNow™ Backpack Simulator.
Photograph by the author (WRC).
Figure 5
Descriptive text is not available for this image
MIMIC dV-Trainer – MIS and daVinci™ Robot Simulator.
Reproduced with permisison from Mimic Technologies, Inc., Seattle, Washington.


With the advent of new catheter-based mitral repair techniques, the need for bio-simulation has arisen. Trans-apical and trans-catheter simulated mitral repairs can now be done on a beating animal heart. LifeTec, Inc. (Eindhoven, Netherlands) has developed an integrated bio-simulator that is comprised of an explanted beating animal heart, a videoscope, an external echo probe, and hemodynamic monitoring system. [53] The heart is pulsated from the ventricular apex by a complex closed-loop pump. Intra-cardiac vision is provided by a 3-D echocardiographic probe placed directly on the heart. (Figure 6 A, B, C) Moreover, a 2-D videoscope is placed across the left atrial wall to provide real time en face views of the mitral valve. The hemodynamic monitoring system measures contemporaneous systemic and left atrial pressures as well as cardiac output. Thus, when a flail mitral leaflet, causing severe regurgitation, is being repaired, the operator can monitor the adequacy of the repair by both echo and videoscopic vision as well as through hemodynamic corrections.

Figure 6A
Descriptive text is not available for this image
This photograph shows the biosimulator with an incorporated animal heart. The left ventricle is pressurized cyclically with saline from a pulse duplicator (PD) inserted in the apex of the heart. A three-dimensional echocardiographic probe (EP) is in place along the posterior left atrium. A chord replacement device (D) is being inserted over a guidewire.
Photograph by the author (WRC).
Figure 6B
Descriptive text is not available for this image
LifeTEC Biosimulator (LifeTEC Group, Inc., Eindhoven Netherlands)
An endoscope is placed through the left atrial wall to visualize the dynamic mitral valve.
Photograph by the author (WRC).
Figure 6C
Descriptive text is not available for this image
LifeTEC Biosimulator (LifeTEC Group, Inc., Eindhoven, Netherlands): 6C - Simultaneously, cardiac output, aortic blood pressure (AoP), and left atrial pressure (LAP) are displayed. Note, that following a flail mitral leaflet repair the AoP rises and the LAP decreases concomitantly as the regurgitation is reduced.
Photograph by the author (WRC).

Surgical Anatomy

Surgeons beginning MIMVS via a right mini-thoracotomy should become familiar with the view shown in Figure 7A of the pericardium, phrenic nerve, and diaphragm. In some patients the right hemi-diaphragm may be elevated and obstruct this view of the pericardium. In this instance to render exposure, a retraction suture should be placed in the central diaphragmatic tendon and exteriorized through the chest wall and tightened. Figure 7B shows an important opened pericardial surgical view of the transverse sinus that shows the right pulmonary artery, left atrial appendage, as well as left atrial roof. After the left atriotomy has been made and the mitral valve exposed, one has an orthogonal en face view of the mitral valve that is more difficult to attain through a traditional sternotomy. This is one of the major advantages of this approach, which is equally amenable to repairing and replacing mitral valves.

Figure 7A
Descriptive text is not available for this image
Working Incision: Pericardial Anatomic View – After the patient is on full cardiopulmonary bypass and the right lung has been deflated, surgeons should have this right 4th interspace view of the pericardium, phrenic nerve, and diaphgram. The collapsed pericardium is shown here draped over the aorta. After longitudinally opening the pericardium in this region, the important cardiac anatomy for a minimally invasive mitral operation is exposed – see Figure 7B.
Photograph by the author (WRC).
Figure 7B
Descriptive text is not available for this image
Cardiac and Transverse Sinus Anatomy: After opening the pericardium and distracting it laterally, important heart and transverse sinus anatomy are exposed. The posterior aorta, right pulmonary artery, left atrial appendage, and atrial roof are shown. Often the left main coronary artery also can be seen. Before applying the trans-thoracic aortic cross clamp (described later in the chapter), all of these structures should be visualized to avoid complicated injuries.

Patient Selection:

Patients should have the indications for mitral valve repair surgery as outlined in either the United States 2020 ACC/AHA or European 2021 ESC/EACTS Guidelines. [54],[55] Patient selection for a minimally invasive mitral operation should be determined by skeletal anatomy, pre-existing medical conditions, cardiac function, complexity of mitral pathology, previous thoracic procedures, overall fragility, and the patient’s willingness to undergo a possible conversion sternotomy. The mini-thoracotomy incision approach, with or without assisted vision or robotics, is indicated for both primary and reoperative mitral valve surgery in patients with no or few comorbidities. Moreover, combined mitral and tricuspid valve operations can be performed safely through this incision.

Patients requiring either concomitant coronary revascularization, an aortic valve replacement, or have a significantly dilated ascending aorta should have a traditional sternotomy operation. Table 1 lists the inclusion and exclusion criteria that our group has used successfully for many years. Some of these are absolute contraindications but some can be managed, providing that a number of lesser co-morbidities are not compounded. Regarding patients with central and/or peripheral vascular disease, perfusion and aortic occlusion methods differ and must be considered during both patient and operative technique selection. Institutional patient selection algorithms can be developed for MIMVS and are helpful to provide the best safety when beginning a new program.

Table 1: Contraindications: Minimally Invasive Mitral Valve Surgery

  • Previous right thoracotomy with a lung resection, empyema, or pleurodesis
  • Poor pulmonary function
  • Pulmonary hypertension (Fixed PAS > 70 torr)*
  • Pulmonary hypertension with a dilated poorly contracting right ventricle
  • Severe liver dysfunction
  • Bleeding disorders*
  • Significant aortic valve disease
  • Coronary artery disease requiring revascularization*
  • Recent myocardial ischemia (< 30 days)
  • Recent stroke (< 30 days)
  • Severely calcified mitral valve annulus
  • Severe aorto-iliac atherosclerosis or aneurysm*
  • Severe left ventricular failure – EF< 25%

* Relative contraindications: Some may be managed either by a detailed preoperative evaluation or intervention and/or the use of alternate techniques.

Informed Consent

Informed consent for a minimally invasive or robotic mitral repair should include the choice of a traditional sternotomy operation, as well as the possibility of conversion to a sternotomy for safety or to achieve the operative goal. Although conversions at most large centers are very low, the patient must be informed of the occasional necessity.

Diagnostic Screening Tests

Patients should be studied with 2-D and 3-D echocardiography and when appropriate, cerebrovascular ultrasound and coronary angiography as well as chest, abdomen, and peripheral vascular computed tomography.

Echocardiography: Diagnosis and Repair Planning

Echocardiography is the mainstay to define mitral valve pathology and ventricular function. Careful preoperative echo planning helps during patient discussions as well should provide surgeons a pre-operative “blueprint” for consideration during operative reconstructions. Moreover, mobile aortic arch atheromas can be defined optimally at the time of surgery by transesophageal echocardiography (TEE), which should always be used when passing intra-vascular guide-wires or cannulas.

Many cardiologists refer patients having had only a 2-D trans-thoracic echocardiogram. However, in complex degenerative disease a pre-operative 3-D trans-esophageal study is very helpful. A comprehensive 2-D and 3-D TEE study always should be done in the operating room to render the necessary planning information listed in Table 2. At our center all minimally invasive, robotic, and traditional mitral repairs are planned and directed from intra-operative TEE studies. Figure 8 shows the echo data that our anesthesiologists collect in the operating room. To determine an appropriate annuloplasty ring or band size, we have developed a TEE-based nomogram, which has been validated with actual leaflet measurements and commercial ring sizers. [56](Table 3) Annuloplasty prosthetics differ in sizing measurements depending upon the manufacturer. Thus, a similar nomogram can be developed for each ring or band by comparing static sizers to linear measurements (obtained at surgery with an “on-valve” paper mm ruler), and then to TEE measurements. Because of accentuated leaflet curvilinearity associated with extreme billowing in severe Barlow’s pathology, the TEE can underestimate actual leaflet lengths. In this circumstance either direct linear or sizer measurements should be made in the arrested heart. Some manufacturers make flexible silastic sizers to ease trans-thoracic passage through small working incisions.

Table 2: Trans-esophageal Echocardiographic Planning Data

  • A2 and P1-P3 leaflet segment lengths
  • Specific mitral leaflet prolapse or restricted regions
  • Direction and numbers of regurgitant jets
  • Posterior leaflet segment indentations/clefts
  • Annular posterior-anterior diameter and shape
  • Aorto-mitral valve annular plane angle (< 120 degrees)
  • Inter-ventricular septal thickness
  • Coaptation-septal distance
  • Right and left ventricular function
  • Left atrial diameter
  • A topographic valve reconstruction from 3-D images

Figure 8
Descriptive text is not available for this image
Operating Room Echocardiographic Data Sheet: This data sheet shows the 3-D echocardiographic information that is collected in the operating room by our anesthesiologists. Defining data include the length, mobility, and pathologic characteristics of each mitral leaflet segment as well as the site and direction of any leaks. By using this information, surgeons can formulate a “blueprint“ plan for the repair, while the heart still is functioning physiologically.

Table 3: Echo-Based Annuloplasty Band* Sizing

Anterior Leaflet

(A2) Length

Band Size

20-mm or <










30-mm or >


*Cosgrove-Edwards™ Annuloplasty System (Edwards Lifesciences, Irvine, CA)

*Note: As these sizes differ between manufacturers, a similar nomogram can be derived

for the different annuloplasty bands or prostheses.

The combination of intra-operative 2-D and 3-D TEE studies best define: 1) leaflet lengths and segment (scallop) pathologies (flail, prolapsing, or tethered) 2) commissure pathology, 3) annular size and geometry (septo-lateral distance), 3) ventricular septal thickness, 4) aortic outflow diameter (coaptation-septal distance), 5) the aorto-mitral plane angle, and 6) annular/leaflet calcium. Moreover, echo topographic models (Figure 9A and B) of the mitral pathology are useful for planning reconstructions. [56]

Figure 9A
Descriptive text is not available for this image
Echocardiographic Topographic Model: All of P1, P2, and P3 Posterior Leaflet Prolapsing. In the operating room topographic models are created from trans-esophageal echocardiographic data. These are quite helpful in planning any mitral valve repair. Papillary muscles are shown in bright red below the annulus. In this image the entirety of a large P2 is shown to prolapse. (A) anterior leaflet; (P) posterior leaflet; (AL) antero-lateral commissure; (PM) posterio-medial commissure.
Figure 9B
Descriptive text is not available for this image
Echocardiographic Topographic Model: All of P1 and Part of P2 Posterior Leaflet Prolapsing. (A) anterior leaflet; (P) posterior leaflet; (AL) antero- lateral commissure; (PM) posterio-medial commissure.

Long anterior leaflets (> 30mm A2), especially in the presence a high (long) posterior leaflet (>20-mm), a narrow aorto-mitral plane angle (< 120°), or a thickened outflow inter-ventricular septum, may portend the development of post-repair systolic anterior leaflet motion (SAM).This can create a major hemodynamic problem when weaning from cardiopulmonary bypass. Preventative structural repair measures should include: 1) implantation of a large, true-sized ring/band, 2) reduction of the posterior leaflet height to 15-mm or less, and 3) achievement of an optimal linear leaflet coaptation surface (8 to10-mm). In the arrested heart the saline ventricular pressure test, done after the final repair, is helpful to determine degree and symmetry of leaflet coaptation.

Computed Tomography

Computed tomography (CT) of the peripheral arteries and complete aorta should be done in patients having any risk of atherosclerosis. CT studies are especially important when using peripheral retrograde cardiopulmonary bypass perfusion and/or intra-aortic endoballoon occlusion. Some groups recommend performing CT scans on all patients undergoing minimally invasive and robot-assisted cardiac surgical operations. A recent study showed clearly that CT studies were key in operative planning to decrease the risk perioperative strokes.[43]Ileo-femoral arterial measurements are also important to determine inflow cannula size and especially the need for bi-femoral cannulation when using an endoballoon occluder, as it can compromise perfusion inflow when placed through a small arterial cannula. Diffuse atherosclerosis or a significantly dilated aorta usually precludes intra-aortic occlusion. Moreover, a calcified or dilated ascending aorta may prevent safe direct clamping. Patients with significant thoracic skeletal anomalies (pectus excavatum, kyphosis, scoliosis) should have a detailed CT done before selecting any minimally invasive approach versus a sternotomy.

Coronary Angiography

The patients with degenerative mitral valve disease often are younger and usually do not have significant coronary disease. However, it is essential to define any coronary pathology. Our policy has been to screen younger patients (< 45 years old), who have no family history or major risk factors for coronary disease, using CT angiography. We still have a low threshold for requesting catheter-based angiography in older patients and those having any possibility of significant coronary disease. In patients with single and double vessel disease, we still consider them for a minimally invasive or robotic mitral operation if the lesions are not flow limiting, ischemia inducing, or have been percutaneously stented several weeks before surgery.

Carotid Ultrasound

Patients having any prior neurological event, symptoms, or significant peripheral atherosclerosis should have a flow determining carotid ultrasound done prior to surgery. In patients with significant carotid disease, we have deferred MIMVS until after either an endarterectomy or stenting.

Surgical Access and Techniques

Hemi-Sternotomy - Direct Vision

Today, most MIMVS operations are done either through a right mini-thoracotomy or small ports. However, in the past many minimally invasive mitral repairs/replacements were done through a hemi-sternotomy. [12],[21](Figure 10) With this technique the skin incision should begin below the sternal notch and carried for 5-8 cm to access the fourth or fifth intercostal space. Here, the hemi-sternotomy is directed rightward into the interspace, being careful to preserve the internal thoracic artery. After the pericardium is opened, it should be suspended to the sternal fascia and a small retractor inserted. By opening the retractor, the pericardial sling will elevate the aorta and heart toward the incision. Standard sternotomy-based cannulation and aortic occlusion techniques can be performed through this incision. However, some prefer peripheral perfusion to decrease cluttering the surgical field. Surgeons using this approach generally have used an extended right atrial/trans-septal approach to access the mitral valve. [12],[21]

Figure 10
Descriptive text is not available for this image
Hemi-sternotomy: Through an upper hemi-sternotomy, an extended trans-septal approach to the mitral valve is shown. The vena cavae (SVC and IVC) have been isolated for venous drainage. Through a right atrial incision (RA) with extension into the roof, the intra-atrial septum (IAS) has been divided and retracted. The left atrium (LA) and mitral valve are shown to be exposed widely.

Mini-Thoracotomy - Videoscopic and Robot-Assisted

As mentioned, preoperative patient positioning, anesthesia management, echocardiographic evaluation including repair planning, the operative set-up, and perfusion techniques are similar for both robotic and non-robotic MIMVS. This approach has become a most popular way to perform minimally invasive mitral operations as the surgeon sees the valve en face. The Atlas of Robotic Cardiac Surgery details the conduct of robotic and MIMVS at several well-known centers. [56]

Patient Positioning

For both approaches, the patient should be positioned with the right chest elevated by 30° on a commercial lift or a cloth roll. (Figure 11) This position widens the intercostal interspaces, decreasing the need for mechanical rib spreading. The right arm should be distracted laterally and supported safely on a sling inferior to the posterior axillary line. The right axillary area remains exposed for insertion of the trans-thoracic aortic cross clamp. During positioning, anterior and posterior Zoll defibrillator patches should be placed to subtend the cardiac axis. Standard skin preparation and draping should provide wide exposure to the right hemi-thorax, mid-sternal line, and both groins.

Figure 11
Descriptive text is not available for this image
Patient Positioning: The patient should be positioned with the right hemithorax elevated 30° using either a cloth roll or commercial lift device. The right arm should be distracted laterally and then positioned safely just below the posterior axillary line using a sling. Before sterile preparation, it is helpful to determine and mark the 2nd, 3nd, 4th and 5th intercostal spaces as well as the midsternal line. The position for working an incision in a non-robotic videoscopic mitral operation is shown. This is placed farther anteriorly than for a robot-assisted mitral operation.

Anesthesia Management

For most MIMVS and robotic operations our anesthesiologists prepare patients as shown in Figure 12. Either a double-lumen endotracheal tube or a right endo-bronchial blocker can be placed to enable single lung ventilation. Some centers prefer to us only a single endotracheal tube during MIMVS. However, the ability to deflate the right lung is helpful when inspecting for bleeding sites. In most cases a single right radial arterial catheter is placed for systemic blood pressure monitoring. When selecting endoballoon aortic occlusion, bilateral radial arterial monitoring is mandatory to assure that the inflated balloon is not obstructing innominate artery blood flow. A 3-D TEE probe is inserted at this time. Thereafter, a right internal jugular vein introducer is placed for both drug infusions and Swan-Ganz pulmonary artery catheter insertion. Using the “double-puncture” guide-wire technique, a thin-walled (15-Fr or 17-Fr) Bio-Medicus (Medtronic, Inc., St. Paul, MN) right internal jugular venous drainage cannula is placed. Some surgeons prefer to insert this cannula from the surgical field. In selected patients a single femoral to superior vena cava cannula can provide sufficient venous drainage. However, we have preferred dual caval venous drainage to assure constant mitral valve exposure and ideal systemic cooling of the right ventricle and atrium. If desired, a guidewire directed retrograde coronary sinus cardioplegia catheter can be placed via the right internal jugular vein and positioned either under echocardiography or fluoroscopy. If a jugular venous return cannula is to be placed on the same side as the retrograde cardioplegia cannula, the Swan Ganz catheter should be inserted through the left internal jugular or subclavian vein.

Figure 12
Descriptive text is not available for this image
Anesthesia Preparation: After general anesthesia has been administered, either a double lumen endotracheal tube or bronchial blocker is placed to isolate the right lung. Thereafter, the transesophageal echo probe is positioned. A right internal jugular vein introducer is inserted to accommodate placement of a Swan-Ganz pulmonary artery catheter as well as medication infusions. Lastly, using the jugular vein “double puncture” guide-wire technique, the superior vena caval venous drainage cannula is inserted just inferior to the existing introduce.

Working Incision/ Access Port

Today, most minimally invasive and robotic mitral valve operations are performed through a mini-thoracotomy. The choices for visualization include either a direct view through the incision, [2-D or 3-D] videoscopy, or 3-D robotic. The size of the working incisionor port usually relates to both the visualization and operative method chosen by the surgeon. For both direct vision and videoscopic MIMVS, a 4-5 cm mini-thoracotomy is made in the 4th intercostal space placed near the anterior axillary line. (Figure 11) Smaller working port-access incisions (2-3 cm) are made more posterior for robot-assisted operations. We place a flexible Alexis™ soft tissue wound protector (Applied Medical, Inc., Rancho Santa Margarita, Calif) in preference to using a rib-spreading retractor. This device provides excellent exposure, minimizes postoperative pain, and provides good working incision exposure for either direct vision, videoscopic, or robotic MIMVS.

Surgical Instruments

Standard forceps, needle holders, and scissors can only be rotated approximately 160 to 180-degrees by the human wrist. When using these instruments through a sternotomy, surgeon body re-orientation generally compensates for natural ergonomic limitations. However, this is not possible with small incisions. Using long-shafted instruments, finger rotation can be combined wrist motion, enabling near 360-degree tip rotation. Despite added benefits of these instruments, they lack the full articulation abilities of robotic endo-wrist instruments. To become familiar with using these long-shafted instruments, we suggest using them first in sternotomy procedures. Figures 13 (A-D) show several of these long-shafted minimally invasive instruments. Similarly, hand knot-tying and suture cutting are difficult to impossible through very small access incisions. Figure 14 (A,B,C) show the knot pusher and suture cutter that we use to facilitate these operations. Several trans-thoracic aortic clamps have been developed that either are passed through the chest wall or through the small working incision. Figure 15 (A - B) shows the atraumatic trans-thoracic aortic cross clamp that we use for both minimally invasive and robot-assisted mitral valve surgery.

Figure 13A
Descriptive text is not available for this image
Long-shafted Instruments for minimally invasive mitral valve surgery. Note, in all of these instruments the round handles provide additional range of motion via finger-tip manipulation. (A) Curved jaw needle holder
Reproduced with permission from Scanlan International, Inc.
Figure 13B
Descriptive text is not available for this image
Long-shafted Resano mitral valve tissue forceps.
Reproduced with permisison from Scanlan International, Inc.
Figure 13C
Descriptive text is not available for this image
Long-shafted 45° angled (Potts) scissors.
Reproduced with permission from Scanlan International, Inc.
Figure 13D
Descriptive text is not available for this image
Long-shafted Curved scissors. Note, the round handles provide additional range of motion via finger-tip manipulation.
Reproduced with permission from Scanlan International, Inc.
Figure 14A
Descriptive text is not available for this image
Minimally Invasive Knot Pusher.
Reproduced with Permission of Scanlan International, Inc.
Figure 14B
Descriptive text is not available for this image
Long suture cutter for minimally invasive mitral valve surgery.
Reproduced with permission from Scanlan International, Inc.
Figure 14C
Descriptive text is not available for this image
Long knot pusher and suture cutter tips.
Reproduced with permission from Scanlan International, Inc.
Figure 15A
Descriptive text is not available for this image
Long trans-thoracic aortic cross clamp.
Reproduced with permission from Scanlan International, Inc.
Figure 15B
Descriptive text is not available for this image
Atraumatic cross clamp jaws of the trans-thoracic aortic cross clamp.
Reproduced with permission from Scanlan International,Inc.

Cannulation for Cardiopulmonary Perfusion

Methods for cardiac cannulation and cardiopulmonary perfusion (CPB) during MIMVS and robotic surgery are detailed in Cardiopulmonary Bypass and Mechanical Support: Principles and Practice. [57]All perfusion cannulas are placed under echocardiographic guidance using the Seldinger guide-wire technique. The right femoral artery is cannulated with either a 17 or 19-Fr Bio-Medicus cannula placed through a 2-cm oblique groin incision. (Figure 16) For both arterial perfusion and Intraclude™ endoballoon introduction, the Thruport EndoReturn cannula (Edwards Lifesciences, Inc., Irvine, Calif) (Figure 17) is available as a 19-Fr single sheath or as a 21-23 Fr dual lumen Y catheter. If vascular disease precludes retrograde peripheral arterial perfusion, it may be feasible to cannulate the ascending aorta directly through the incision or an additional port. Alternatively, we use the right axillary artery for antegrade perfusion. In this instance, the artery is exposed through a right infra-clavicular incision. Generally, we sew an 8-mm GelSoft™ knitted graft (Vascutek, Terumo, Ann Arbor, MI) end-to-side to the axillary artery with 5-0 polypropylene suture. This is then tied over a 3/8th inch perfusion tube-connector with heavy silk sutures. Terumo (Terumo Medical Products, Somerset, NJ) also makes a cannula that is annealed to an 8-mm PTFE graft that can be hooked directly to the perfusion circuit tubing.

Figure 16
Descriptive text is not available for this image
Femoral Cannulation for Cardiopulmonary Bypass: To expose the femoral vessels, a small oblique incision is made in the right groin, just below the iliac ligament. Both vessels should be exposed but not isolated completely. (A) Small purse-string sutures are placed superficially along the anterior surface of the vein and artery. Using the Seldinger technique and under echocardiographic guidance, guide-wires are passed sequentially into both the artery and vein. Thereafter, appropriately sized vessel dilators and cannulas are passed over each guide-wire and positioned using echo vision. (B and C) Small flexible Rummel tourniquet tubes are passed over the purse-string suture tails and secured with clamps.
Figure 17
Descriptive text is not available for this image
Thruport EndoReturn™ Cannula: For (A) arterial perfusion and (B) insertion of the Intraclude™ Aortic Balloon catheter. During cardiopulmonary bypass and aortic occlusion the arterial cannula lumen is occupied by the balloon catheter.
Reproduced with permission from Edwards Life Sciences ,Inc., Irvine California.

For inferior vena caval drainage either a 22-Fr (single stage) or a 23/25-Fr (dual stage) RAP (Livanova Inc., Houston, Texas) or a Quick Draw 22–25 Fr (Edwards Lifesciences, Irvine, Calif) femoral (single stage) venous cannula is passed into the right atrium. As mentioned, we prefer to use bi-caval drainage by placing a 15 to 17 Fr internal jugular venous catheterVacuum-assisted venous drainage is used in all of these operations. Intra-thoracic carbon dioxide is insufflated throughout each procedure to minimize intra-cardiac air entrapment. We monitor oxygen saturation levels continuously in each leg, using the Invos System (Somanetics Inc., Troy, MI), to assure adequate limb perfusion during cardiopulmonary bypass (CPB). If the arterial saturation decreases significantly in the cannulated leg during CPB, we place a 5-Fr catheter in the superficial femoral artery distally and connect it to the arterial perfusion circuit. To date, we have had no problems with residual leg ischemia. Ilio-femoral CT scanning can help guide arterial cannula selection size, possibly obviating the need for monitoring limb perfusion.

Cardiopulmonary Perfusion Circuit

In median sternotomy-based cardiac surgery, gravity venous drainage with large diameter cannulas use to be the mainstay of cardiopulmonary perfusion. However, for mini-thoracotomy minimally invasive and robotic cardiac surgery, direct access for inserting large cannulas is not feasible. As described above, newer ultra-thin wall catheter designs with multiple side holes optimize flow rates and provide maximum drainage. With MIMVS gravity-based venous drainage systems usually are inadequate because of circuit tubing length and small internal cannula diameters. Most minimally invasive operations require ideal venous return to provide total cardiopulmonary support, optimal cardiac decompression, and the best cardiac protection. Most commonly either kinetically assisted (KVAD) or vacuum-assisted (VAVD) venous drainage is used for MIMVS.

KVAD utilizes a centrifugal pump, placed between the venous cannula and reservoir. This method improves venous drainage by 20 - 40% compared with gravity circuits. With KVAD, venous blood is pumped actively into either an open or closed venous reservoir. Centrifugal pump inertia regulates siphoning of venous blood, and as the pump speed is increased, more central suction is generated. Pressure in the venous conduit is monitored proximal to the pump inlet, allowing the perfusionist to regulate the amount of blood returning to the pump. Excessive negative pressure can cause hemolysis and/or right atrial collapse around the venous cannula, thus impeding venous return. Major drawbacks surrounding KVAD include the additional cost of the centrifugal pump head, and the potential of de-priming the pump if enough gross air is introduced from the venous cannulation site.

Our group prefers the VAVD circuit shown in Figure 18. VAVD utilizes regulated vacuum suction connected to the venous circuitry. VAVD was accomplished originally by connecting the wall vacuum directly into an open system hard shell venous reservoir. Current VAVD practice involves equal application of negative pressure exerted on the venous circuit, creating augmented venous drainage. VAVD requires close regulation of the vacuum source to avoid blood component trauma and either cracking or imploding of the venous reservoir. Positive and negative pressure relief valves must be incorporated into the reservoir to prevent both over and under pressurization and to ensure consistent extracorporeal perfusion.

Figure 18
Descriptive text is not available for this image
Cardiopulmonary Perfusion Circuit: (VAVD) Vacuum Assisted Venus Drainage; (ART) pump - Arterial perfusion pump; (del Nido) crystalloid cardioplegia solution.

Cardiac Exposure

After going on CPB, the pericardium is opened longitudinally 3–cm anterior to the phrenic nerve, using either long shafted endoscopic or robotic instruments. Two to three well-spaced pericardial retraction sutures are placed along the inferior (dorsal) pericardial edge. (Figure 19A) At the end of each retraction suture we tie a loop. By passing the needle back through the end-loop, only one suture is withdrawn through the chest wall using a crochet hook instrument. Care should be taken not to stretch or injure the phrenic nerve. The anterior (ventral) pericardial edge then is suspended with two sutures that are brought through the working incision. This helps exposure when placing a bi-polar pacing wire on the inferior surface of the right ventricle, where there is less epicardial fat. Using the right superior pulmonary vein as a guide the inter-atrial groove should be only minimally dissected to expose the left atrium. (Figure 19B).

Figure 19A
Descriptive text is not available for this image
Cardiac Exposure: The pericardium should be opened 3-cm anterior to the phrenic nerve, and then the posterior edge should be retracted laterally using two or three trans-thoracic traction sutures. This should expose the right superior pulmonary vein (RSPV) and inter-atrial groove. (SVC) Superior vena cava.
Figure 19B
Descriptive text is not available for this image
Cardiac Exposure: The inter-atrial groove should be dissected minimally, using the right superior superior pulmonary vein (RSPV) as guide. Opening the oblique sinus (OS) helps prevent entering the right atrium or injuring the right lower pulmonary vein inadvertently during the left atriotomy (Lt Atrium).
Photograph by the author (WRC).

Aortic Occlusion

For aortic occlusion, we prefer a trans-thoracic cross clamp (Scanlan International, St. Paul, MN) as it has been proven to be safe, reliable, economic, and simple to apply. (Figure 15A and B) The posterior clamp arm should be passed through the transverse sinus either under direct or videoscopic vision. The anterior arm is positioned across the aorta until it reaches the main pulmonary artery. (Figure 20) Care must be taken not to injure the right pulmonary artery, left atrial appendage, left main coronary artery, or aorta. When the trans-thoracic clamp is placed near the posterior axillary line (as in robot-assisted mitral operations), it can compress the superior vena cava. Bicaval (jugular and femoral) venous drainage cannulas should avoid cerebral congestion in this circumstance. Some surgeons prefer to use the catheter-based Intraclude™ (Edwards Lifesciences, Irvine, CA) device for balloon aortic occlusion. (Figure 21) This technique has a steeper learning curve than using the clamp. We have published details and nuances of using the Intraclude™ technique. [56],[57]) The balloon position must be precise and remain stable in the ascending aorta just above the sino-tubular junction and proximal to the innominate artery. To avoid the potential of either innominate artery occlusion or intra-ventricular displacement, TEE and bilateral radial arterial blood pressure monitoring are mandatory. Also, balloon catheter introduction through the femoral arterial cannula can limit perfusion flow. In this circumstance the endoballoon catheter must be inserted separately through the contralateral femoral artery. Despite these concerns, this method combines the benefits of effective aortic occlusion, an antegrade cardioplegia delivery route, and a ventricular air vent.

Figure 20
Descriptive text is not available for this image
Trans-thoracic Aortic Cross Clamp Placement: This illustrates how to position the trans-thoracic aortic cross-clamp. It should be inserted through the third intercostal space along the posterior axillary line. The clamp should pass just in front of the superior vena cava (SVC) at the pericardial reflection. (Inset) Under either direct or video-assisted vision, the posterior arm of the clamp should be passed carefully into the transverse sinus before application (see Figure 7) Care must be taken to avoid injury to transverse sinus structures or impinge upon the cardioplegia catheter.
Figure 21
Descriptive text is not available for this image
Intraclude ™ Aortic Perfusion and Occlusion System: Using the side-arm of the Thruport EndoReturn™ perfusion cannula (A) that was shown in Figure 20, the Intraclude ™ balloon (B) is guide-wire positioned in the ascending aorta. Through a right internal jugular vein introducer, a pulmonary artery vent catheter (C) is placed for venting the right side of the heart, and if desired a retrograde cardioplegia catheter is inserted here as well.

Myocardial Protection

As mentioned, we have always used the trans-thoracic aortic clamp method for both minimally invasive and robot-assisted mitral valve surgery. Either using direct, videoscopic, or robot vision, a pledgeted 4-0 ePTFE purse-string suture is placed in the ascending aorta just proximal to the fatty fold of Rindfleisch. (Figure 22) For antegrade cardioplegia administration, a long cardioplegia-vent catheter (Medtronic, Inc., St Paul, MN) is inserted into the aorta through the purse-string, secured, and passed eternally either through a chest wall trocar or the working incision.

Figure 22
Descriptive text is not available for this image
Cardioplegia/Vent Cannula Placement: (A) Using either direct, videoscopic, or robot-assisted vision, a 4 - 0 ePTFE pledgeted “square” purse string suture is placed in the adventitia of the ascending aorta.; (B) Then, the cardioplegia cannula is grasped with an instrument and needle-tip is inserted into the aorta.; (C) Thereafter, opposite side purse string pledgets are placed over the anchoring platforms of the cannula. A Rummel tourniquet is used to secure the cannula in place.

Several years ago, our group switched from intermittent cold blood cardioplegia to Custodiol - Bretschneider’s HTK solution (Franz Köhler Chemie Bensheim GMBH, Germany) as it provides much longer myocardial protection without requiring frequent reinfusions. [58] Del Nido cardioplegia solution also has similar characteristics and has been shown to be safe and very effective during these operations. For maximal myocardial protection, we combine cold antegrade cardioplegia with a systemic blood inflow temperature of 28° C. If a surgeon selects cold blood cardioplegia, repeat administrations should be delivered at 15 to 30-minute intervals. However, multiple cardioplegia infusions, requiring atrial retractor relaxation and repositioning, can entrain air into the aortic root. Additionally, it is very important to have excellent venous drainage to minimize systemic cardiac warming and provide the best intra-cardiac visualization. For patients having had a previous sternotomy, we cool the patient systemically to 26° C and induce ventricular fibrillation by rapid pacing using a custom Swan-Ganz catheter. As long as the aortic valve remains competent, excellent visualization can be attained. However, in patients with mild aortic insufficiency, additional intra-cardiac suction can be combined with dropping perfusion flow briefly for difficult suture placement.

Mitral Valve Exposure

The left atriotomy should be made just medial to the right superior and inferior pulmonary veins. (Figure 19B) Generally, we have used the electrocautery to dissect the intra-atrial groove. Less intra-atrial groove dissection is needed than for mitral valve operations done through a sternotomy. One has to be careful not to injure the inferior pulmonary vein or enter the right ventricle when extending the lower atriotomy. We have found that prior opening tissue leading to the oblique sinus helps prevent these two problems.

For MIMVS operations, a trans-thoracic atrial retractor arm (Atrial Lift System™, Livanova, Houston, Texas) is passed carefully through the chest wall, just lateral to the sternum in the fourth interspace, and coupled to an appropriate size blade. (Figure 23) This arm then is attached to a table-mounted support that facilitates lifting of the inter-atrial septum. Inside the thorax an appropriate size retractor blade is attached. After adjusting the retractor for optimal exposure, the mitral valve repair or replacement is done using long shafted instruments. An intra-atrial left superior pulmonary vein suction sump should be passed through the chest wall to aid in visualization. When sutures are externalized for annuloplasty band/ring or prosthetic valve prosthesis deployment, we place Gabby-Frater™ guides (Teleflex Medical, Research Triangle, NC) externally around the working incision for suture organization.

Figure 23
Descriptive text is not available for this image
2-D or 3-D Videoscopic MIMVS: This illustration shows the general set-up for a non-robotic endoscopic mitral valve operation. The 3-D or 2-D endoscopic camera is inserted via a trocar in either the second or third interspace. The trans-thoracic aortic cross clamp is passed through the chest wall generally in the third interspace along the posterior axillary line. Here, the working incision is shown in the fourth interspace. A soft tissue retractor is used to avoid rib spreading. The trans-thoracic atrial retractor arm is shown in the same interspace as the working incision.

Atrial Closure and De-airing

After the mitral repair or replacement is complete and the valve tested (using injected saline for the repairs), we prefer using 4-0 Gore-Tex™ ePTFE suture (Gore Medical Inc., Flagstaff, Arizona) to close the atriotomy. The suture line should be started at both ends of the atriotomy and run toward the center of the incision. Care must be taken as not to “skew” or close the suture line unevenly as pulmonary vein anatomy can be altered. Leaving an intraventricular vent prior to tying these sutures helps with deairing. Echocardiography monitoring is essential to be sure that air is removed from the heart. Deairing is done by first stopping suction on the ventricular vent is briefly. Then, the heart should be filled by reducing venous pump return while both lungs are ventilated gently to remove air from the pulmonary veins. At this time, we loosen the atrial closure suture line to expel residual air and turn on the ventricular vent suction simultaneously. This is done at the same time that suction is applied to the aortic root vent. Thus, air is removed simultaneously through the aortic and ventricular vents as well as the atriotomy. After the echo studies confirm that de-airing has been successful, the ventricular vent should be removed, and the sutures tied. Thereafter, the aortic clamp is released. When using the Intraclude™ device, deairing is done in a similar fashion except that aortic venting is done through the balloon catheter tip. As mentioned, several times earlier, all of these operative techniques are nearly the same when performing either direct vision, video-assisted or robotic mitral operations.

Videoscopic Vision: 2 and 3-D Vision

Video-assistance is advantageous when very small incisions are used. After completing the above maneuvers and establishing cardiac arrest, we pass a 5-mm high definition 2-D endoscope (KARL STORZ GmbH & Co. KG, Tuttlingen, Germany) through a third intercostal space trocar and attach it to a hand-positioned holder, which is affixed to the operating table. Recently, 10-mm 3-D endoscopes (0 and 30 degrees) have been developed and provide excellent imaging during mini-thoracotomy non-robotic operations (Image1 S-3D® by Karl Storz, Inc and Aesculap® 3-D Einstein Vision®system by Braun, Inc.) [59]3-D endoscopes enabled surgeons to have increased procedural accuracy and are quite facilitating for non-robotic MIMVS. Figure 24 shows the general set-up for a non-robotic videoscopic mitral valve operation. In Figure 24 a 3-D endoscope is being used at the University of Maastricht during a mitral valve repair.

Figure 24
Descriptive text is not available for this image
3-D Endoscopic Vision: This photograph at the University of Maastricht shows the set-up for a minimally invasive 3-D vision endoscopic mitral valve repair. An Image1 S-3D® Storz endoscope (KARL STORZ GmbH & Co. KG, Tuttlingen, Germany) has been inserted in the third intercostal space at the anterior axillary line. The surgeon is wearing special eye-wear to visualize the 3-D image and is doing the operation completely under endoscopic vision.
Photograph by the author (WRC).

Robot-Assisted: 3-D Vision

Robot-assisted cardiac surgery should be called surgical tele-manipulation as the term” robot” means autonomous. The daVinci™ surgical system (Intuitive Surgical, Inc., Sunnyvale, CA) merely provides access with superb three-dimensional visualization and enhanced ergonomics, which allows surgeons to perform complex operations through small port-like incisions. When using long hand-held instruments, the working incision becomes a fulcrum that can limit accurate suture placement and tissue manipulation. The accuracy achieved with da Vinci™ instruments relates to wide freedom of motion achieved at the operative plane. The surgeon always has control of instruments by tele-manipulation, as if they were activated directly by his/her hands.

The daVinci™Surgical System

The da Vinci™ surgical systems consists of tele-manipulator instruments with micro-instrument (end-effectors) that are controlled remotely from the surgeon’s operating console. Currently, these robotic systems are the only ones FDA approved to perform intra-cardiac surgical procedures. Until recently, we used the da Vinci™ SI HD (high-definition) dual console surgical system, which first was commercialized in 2009.The daVinci™ XI System (2014) is now being used by most surgeons for robot-assisted mitral valve surgery. (Figure 25A-C) Both SI and XI systems are comprised a mobile surgical instrument cart, an electronic vision cart, and a surgeon operating console. The da Vinci XI™ has a laser targeting positioning system that facilitates bedside instrument cart docking. Moreover, an additional robotic joint decreases the risk of external instrument arm collisions. The 3-D vision system is improved over the da Vinci SI™ device. For non-cardiac thoracic operations, the da Vinci XI™ automated stapler has been a major advance. Both of the SI and XI systems have dual surgeon operating console capabilities, which greatly facilitates teaching robotic mitral repair and other operations. (Figure 26)

Figure 25
Descriptive text is not available for this image
daVinci™ Xi Surgical System: (A) Table-side instrument cart; (B) Electronic vison cart; (C) Surgeon operating console.
Reproduced with Permission of Intuitive Surgical, Inc. Sunnyvale, California.
Figure 26
Descriptive text is not available for this image
daVinci Si™ Dual Operating Consoles: Here, the author is proctoring a surgeon who is learning to perform robot-assisted mitral valve repairs. The proctor can guide the learning surgeon, using on-screen telestration, and then transfer instrument control to him/her to perform suggested individual repair steps.

For mitral valve surgery the instrument cart should be positioned along the patient’s left side with the instrument arm activators arching over to the right chest. (Figure 27) Individual instrument trocars are inserted through specific intercostal spaces. After the two instrument arms, the 3-D camera, and dynamic retractor arm have been inserted, two hand-driven sensors transmit instructions from the surgeon to instrument end effectors. A clutching mechanism enables frequent hand-position readjustments to maintain an optimal ergonomic attitude with respect to the visual field. The dynamic atrial retractor arms can be repositioned easily from the surgeon console.

Figure 27
Descriptive text is not available for this image
daVinci™ Robotic Operating Room Set-up: The da Vinci™ surgical cart is docked along the left side of the operating table. Instrument arm activators arch over the table to reach the right side of the patient. A separate vision cart sends images of the operative field to monitors placed throughout the room. The surgical assistant works with the scrub nurse or technician to position the robot arms and to exchange instruments. The surgeon works from the operating console.

Figure 28 shows the daVinci XI™ operating room is arranged for a mitral valve repair being done at the Heart Hospital-Plano/Baylor. Both the daVinci SI™ and da Vinci XI™ have dual-console capability, which enables surgeon collaboration during complex cases and can facilitates training. The EndoWrist™ instruments have seven degrees of ergonomic freedom and allow tremor free dexterity with both dominant and non-dominant hands. (Figure 29A-B) Thus, the wrist-like micro-instrument articulations improve dexterity in tight spaces.

Figure 28
Descriptive text is not available for this image
daVinci XI™ operating room: (S) The surgeon is at the operating console performing a mitral valve repair; (A) Tableside assistant; (N) Scrub nurse; (IC) daVinci XI™ instrument cart. The vision cart is just to the side of the instrument cart.
Photograph by the author WRC.
Figure 29A
Descriptive text is not available for this image
daVinci ™ Instrument: Range of Motion:This illustration shows the full range of motion capabilities of da Vinci™ Endowrist™ instruments. A full seven degrees of freedom of motion in space are provided, which closely recapitulates the ergonomics of the human wrist.
Figure 29B
Descriptive text is not available for this image
daVinci ™ Instrument: Endowrist™ Resano tissue forceps.

Trocar Placement and Instruments

Correct instrument arm trocar-port placement is paramount to achieve optimal robot-assisted ergonomics. Thus, we always mark a “topographic thorax surface map” prior to patient positioning and skin preparation. Many surgeons prefer to insert the camera first to define both the intra-thoracic anatomy and/or determine the presence of adhesions prior to placing other ports or creating the working incision. For complex thoracic skeletal anatomy, CT scan reconstructions may be helpful in guiding trocar placement. The goal is to provide the best converging instrument trajectory at the mitral valve annular plane. The most important cardiac landmark for guiding the camera port placement and working port incision locations is the right superior pulmonary vein.

Figure 30 shows the ideal trocar entry sites for instrument insertions. Most often left and right instrument trocars are inserted in the 3d and 5th intercostal spaces, respectively. The left trocar is positioned near the anterior axillary line and the right one at the mid-axillary line. The 3-D robot camera is inserted either through the 4th interspace working incision, or via a trocar placed anterior to it in the same interspace. The dynamic retractor is inserted through a mid-clavicular line trocar placed in the 5th interspace. Figure 31 illustrates how instruments should be deployed during a da Vinci XI™ mitral valve repair.

Figure 30
Descriptive text is not available for this image
Instrument Trocar Placement: The working incision is shown in the fourth intercostal space along the anterior axillary line. Trocars for the left and right robot arms are inserted in the third and fifth intercostal spaces, respectively. Note, that the trocar for the right instrument arm is positioned posterior to the left arm trocar and near the mid-axillary line. The camera port should be placed just anterior to the working incision and in the same (5th) interspace. The trocar for the dynamic atrial retractor is shown along the midclavicular line in the fifth interspace.
Figure 31
Descriptive text is not available for this image
Deployed daVinci XI™ Instruments: (L ARM) Left instrument arm; (R ARM) Right instrument arm; (C)camera; (LAR) Left atrial retractor; (WP) Working port.
Photograph by the author (WRC).

The most common wristed da Vinci™ instruments used for robot-assisted mitral valve surgery are curved scissors, electrocautery scissors, suture-cut needle holders, large needle holders, and Resano forceps. Specialized “ring” forceps are available positioning round cryoprobes that are used for atrial fibrillation ablation operations.

Operative and Postoperative Management

Operative Workflow

As mentioned earlier, the basic operative workflow is similar for either direct vision, videoscopic, robot-assisted MIMVS and is shown in Figure 32. After the “working port” incision has been made, the patient is heparinized and cannulated for CPB. For robotic operations the da Vinci™ system instruments are “trocar-docked” at the operating table. Thereafter, CPB is begun with systemic cooling to 28 degrees C. To expose the heart, the pericardium is opened linearly 3-cm anterior to the phrenic nerve with the robotic cautery scissors or a hand-held electrocautery. Pericardial retraction sutures are placed and then exteriorized laterally. As described earlier, we place the aortic purse-string cardioplegia suture and cannula using either shafted or robotic instruments, depending on the MIMVS method selected. Those using the Intraclude™ aortic occlusion device should position it at this time. For atrial fibrillation patients, right-sided cryo-MAZE lesions are made before cardioplegic arrest. The trans-thoracic cross clamp is then applied under secondary vision and cardioplegia is infused. For Intraclude balloon users, the device is inflated and cardioplegia given. After the left atriotomy is made, either a static or dynamic robotic retractor is deployed and manipulated to expose for the valve repair or replacement as well any sites for left atrial cryo-MAZE lesions. Thereafter, we close the left atrial appendage using a two-layer 4-0 ePTFE suture line. After specific leaflet repairs are completed, an annuloplasty prosthesis is anchored in place. We place a bipolar pacing wire on the diaphragmatic surface of the arrested right ventricle. After the left atriotomy has been closed and the heart deaired, we release the aortic clamp and hopefully reestablish normal sinus rhythm. When indicated, tricuspid repairs then are completed on the beating heart, and thereafter the patient is withdrawn from CPB.

Figure 32
Descriptive text is not available for this image
Basic Operative Workflow for Direct Vision, Videoscopic, or Robotic MIMVS

Post-Operative Management

The post-operative management for all of our minimally invasive mitral valve patients is similar to those having a sternotomy-based operation. Generally, the chest tubes and/or silastic drains are removed on the second postoperative day. Pacing wires are discontinued on the third post-operative day. In the absence of prior or post-operative atrial fibrillation, we have not anti-coagulated most mitral repair patients. For patients having had a cryo-MAZE ablation for atrial fibrillation and atrial appendage closure, we reinitiate anti-coagulation on the third postoperative day. Most often patients having a lone mitral repair or replacement are discharged on the 4th postoperative day. Nevertheless, patients receiving an adjunctive cryo-MAZE usually require one to two days longer hospitalization. Amiodarone has been our first choice for treating patients who develop post-operative atrial fibrillation. We rarely cardiovert these patients while in the hospital, especially if they are rate controlled and stable hemodynamically. If the immediate postoperative TEE was satisfactory, we have not performed echocardiographic studies routinely prior to discharge

Repair Techniques

In this chapter, we use the Carpentier nomenclature for topographic descriptions of mitral valve anatomy, but to facilitate MIMVS operations, we have simplified some his classic mitral repair techniques. [60]Table 4 shows our “Technique Toolbox” that we use for all direct vision, videoscopic, and robotic mitral valve repairs. Most mitral repairs can be accomplished using one or more of these simplified techniques.

Table 4: “Technique Toolbox”
Used for all Minimally Invasive and Robot-assisted Mitral Valve Repairs

Posterior Leaflet Prolapse

Small Segment

Posterior Leaflet Prolapse

Large Segment

Anterior Leaflet Prolapse

Bileaflet Prolapse (Barlow)

Commissure Prolapse

Triangular Resection*

Triangular* or Trapezoid Resection

Triangular Resection

(Small Isolated Segment)

AL = ePTFE Neochords*

PL = Multiple Triangular Resections

Commissure Closure

Alfieri Stitch or “Magic Stitch”

ePTFE Neochords*

ePTFE Neochords*

ePTFE Neochords*

(Large Segment)

AL = ePTFE Neochords*

PL = Multiple Folding-plasties

ePTFE Neochords*

Native Chord Transfer

"Haircut” P2 Edge Resection*

+ Native Chord Re-attachment or ePTFE Chords

Papillary Folding-plasty*

AL = ePTFE Neochords*

PL = Leaflet Sliding-plasty*

PL = Sliding-plasty + ePTFE Neochords

Leaflet Folding-plasty*

Leaflet Folding-plasty*

Translocation of A2 Strip*

along with attached chords

Combined Techniques

Papillary Folding-plasty*

(Elongated or multi papillary with PL and AL Chords)

Inter-scallop Cleft Closure

Inter-scallop Cleft Closure

Combined Techniques



* a chapter illustration

(AL) Anterior Leaflet

(PL) Posterior Leaflet

Posterior Leaflet Repairs

We prefer to maintain leaflet-annular junction integrity during posterior leaflet repairs when possible. We generally ascribe to the “respect” rather than “resect” leaflet repair philosophy. Nevertheless, for isolated leaflet scallop prolapse or chordal ruptures, we often implement small triangular/trapezoidal resections that do not extend to the annulus. (Figure 33) Resection defects are closed with interrupted figure of eight sutures using a 5-0 or 4-0 polypropylene or polyamide monofilament Cardionyl suture material (Peters Surgical, Inc., Paris, France). This suture material is strong and has very little material memory. Many surgeons close leaflet resections with either two-layer running monofilament or ePTFE sutures. In the presence of multiple prolapsing posterior leaflet scallops (Barlow’s pathology), several small triangular resections can produce an effective repair.

Figure 33
Descriptive text is not available for this image
Leaflet Triangular Resection: Curved scissors are used for this procedure: (a) posterior mid-P2 leaflet flail; (b and c) small triangular leaflet resection – Note: the leaflet incision does not go completely to annulus: (d) P2 edges on each side of the resection are reapproximated with either interrupted or running monofilament sutures. Thereafter, either in annuloplasty ring or band is placed. With the robotic technique, we place the first annuloplasty suture at the right fibrous trigone and proceed clockwise. At the midportion of the band, we switch the needle holder to the left instrument arm and continue suture placement clockwise to left fibrous trigone.

Folding-plasty and leaflet imbrication techniques can be used for repairing some prolapsing posterior leaflets. (Figure 34) For folding-plasties, we pass a 4-0 ePTFE suture from the annulus, through the leaflet tip, and back through the annulus. After the annuloplasty band/ring insertion, we adjust individual folded scallop lengths with the ePTFE suture to create a symmetric coaptation line. We still use the sliding technique occasionally for a complex posterior leaflet repair. (Figure 35).

Figure 34
Descriptive text is not available for this image
Leaflet Folding-Plasty: When either part or all of the posterior leaflet is prolapsing or flail, either a single or multiple folding-plasties can be quite effective. (a) A 4-0 ePTFE suture is passed from the annulus, around the tip of the prolapsed leaflet and then back to the annulus. (b) Guided by the sailing pressure test, both limbs of the suture are withdrawn to reduce the prolapsing segment to the appropriate height. (c) After folding adjustments of all prolapsing areas have been made, sutures are tied. (d) Thereafter, either an annuloplasty ring or band is implanted.
Figure 35
Descriptive text is not available for this image
Leaflet Sliding-plasty: We generally reserve sliding plasties for degenerative mitral valves that have severe multi-scallop leaflet prolapse with massive redundant tissue. Although many surgeons use multiple ePTFE neochords to reduce multiple leaflet segments, we believe that the sliding-plasty still is a very good alternative. (a) A large quadrangular resection of P2 with native chords is carried to the annulus, and then radial incisions are made to detach both P1 and P3 from the annulus. Secondary chords then are cut to enable leaflet “sliding”. (b) Compression sutures may be needed to reduce the annular length. Then, using 4-0 ePFTE sutures, both sides (P1 and P2) are advanced serially with each “bite” advancing toward the mid-line. (c) After the two leaflet edges meet, they are suture reapproximated. (d) Thereafter, an annuloplasty ring or band completes the repair.

In the presence of a very large P2 with diminutive P1 and P3 segments, one of three posterior leaflet-plasty or “haircut” techniques is very effective and helps maintain leaflet motion and annular continuity. After resection of the long scallop tip, native chords either are re-attached (Figure 36) or folded over a (Figure 37) long the coapting edge to render “new” P2 support. [56] Another way to provide support for the “new” P2 is to implant several ePTFE neochords. (Figure 38) Today, most surgeons are using ePTFE neochords (Gore-Tex™ - Gore, Inc. Phoenix, AZ) for all or some part of both anterior and posterior leaflet repairs. (Figure 38) Chord-X™ (Cryolife, Inc., Kennesaw, GA) premeasured multiple ePTFE chords are now available and attach at a single point on the selected papillary muscle.

Figure 36
Descriptive text is not available for this image
“Haircut” Posterior Leaflet-plasty #1: Techniques #1 to #3 can be used in the presence of an extremely large P2 posterior leaflet flail or prolapse. (a and b) The large prolapsing or flail P2 segment tip is resected to resemble the length of the other posterior leaflet segments (P1 and P3), while preserving native chords. (c) These chords are reattached symmetrically along the “new” coapting edge. Remaining large clefts between P2 and the other segments should be closed. (d) Finally, an annuloplasty ring or band should be inserted.
Figure 37
Descriptive text is not available for this image
“Haircut” Posterior Leaflet-plasty #2: (a and b) Here a rectangular shaped segment of the large P2 is removed with preservation of “good” chords on the remaining tissue “sidearms”. (c) Each of these “sidearms” then is folded onto the coapting edge of the “new” P2 and sutured in place. Thus, any residual prolapse will be reduced as attached chords on these “sidearms” are moved centrally. (d) Thereafter, an annuloplasty ring or band is implanted.
Figure 38
Descriptive text is not available for this image
“Haircut” Posterior Leaflet-plasty #3: (a) As in the figure 30, the prolapsing end of P2 is resected to the height of the other posterior leaflet scallops. (b) ePTFE neochords are passed through the papillary muscle tip and then the corresponding region of the resected P2 edge. Artifical chord lengths are adjusted using the saline test and then tied. (d) Shows how ePTFE neochords should emanate from the papillary muscle to corresponding leaflet edge, never crossing the annulus vertical midline.

Anterior Leaflet Repairs

The most common methods used include tiny triangular leaflet edge resections, chordal transfers, prosthetic ePTFE chord replacements, or leaflet edge translocation.Theda Vinci™ 3-D high magnification camera and micro-scissors also allow us to easily detach anterior leaflet secondary chords and transfer them directly to a flail or prolapsed edge. Segmental and global anterior or posterior leaflet prolapse can be reduced effectively by inserting multiple ePTFE neochords. (Figure 39) Again, the Chord-X™ solution is very helpful as only one papillary muscle attachment is required. Neochords should never cross the annular midline. That is from one papillary muscle to a nonadjacent anterior leaflet region (e.g. anterior papillary to A3 or posterior papillary to A1) Proper neochord lengths can be estimated from measurements of adjacent normal chords. To achieve symmetric leaflet coaptation, we implant the annuloplasty band/ring before final neochord adjustment. When the entirety of A1 through A3 is prolapsed symmetrically, a wide A1, A2, and A3 chord-bearing leaflet strip can be trans-located (shifted) along the anterior leaflet and toward the aorto-mitral hinge. (Figure 40) This will provide a symmetric corrective reduction of the anterior leafletand should restore natural coaptation.

Figure 39
Descriptive text is not available for this image
ePTFE Neochord Anterior Leaflet Repair: As mentioned in the text, insertion of multiple ePTFE neochords has become very popular to reduce either flail or prolapsing mitral valve leaflet segments. This technique can be used for either anterior or posterior leaflet pathology. (a) Chordae tendenae to the anterior leaflet are ruptured. (b) First, a single long double-armed ePTFE suture is anchored to the fibrous tip of the corresponding papillary muscle, using a “crisscross” technique that is similar to that used in a tendon repair. This provides an excellent anchor without causing papillary ischemia or the necessity of a pledget. Next, each suture arm is passed through the prolapsing leaflet 2 to 3-mm from the edge. Thereafter, the suture is looped around the edge and passed back through the leaflet. (c) Guided by the saline pressure test, the ePTFE neochords are adjusted for ideal leaflet coaptation and ‘locked” onto the leaflet tissue before tying. (d) This shows the completed repair with reduction of the anterior leaflet prolapse.
Figure 40
Descriptive text is not available for this image
Leaflet & Chord Translocation: (a) This type of repair works very well with a large symmetric anterior leaflet prolapse of A2. In this circumstance the native cords are elongated equally on each side of A2. In this illustration the coapting edge of A2 has been marked for translocation. (b) This marked segment of A2 is then pared, leaving the dedicated native chords intact. (c) Then this segment is translocated with intact chords to the anterior surface of remaining A2. (d) Guided by the saline pressure test, the translocated segment is adjusted along the A2 surface to create a uniform valve coaptation line. Thereafter, several sutures are used to affix this to the remainder of A2. This technique allows uniform reduction of a symmetrically prolapsed large mid-anterior leaflet.

Para-Commissural Repairs

For limited regional prolapse, commissures can be closed using an imbrication suture (Alfieri or ‘magic stitch”). Severe bi-leaflet (A3 and P3)posterior commissure prolapse can result from elongated or mobile papillary muscle heads, where multiple originating chords support both leaflets. In this instance shortening papillary muscle heads can reduce both prolapsing commissural leaflets simultaneously. (Figure 41) Anterior commissure prolapse is less common as most anterior papillary muscles have a single head with effective commissural support.

Figure 41
Descriptive text is not available for this image
Papillary Muscle Shortening for Para-commissural Prolapse: (a) In the presence of a large A3 and P3 commissural prolapse associated chords to both generally are elongated by the same amount and most often are attached to single papillary muscle. (a) By reducing these cords by the same amount during papillary muscle shortening, commissural competence can be re-established. After removing a small wedge from the associated papillary muscle and suture approximating the defect, associated chords are “de facto’ shortened. (c) This shows the completed papillary muscle reduction, reducing the para-commissural prolapse.

Annuloplasty Band/ring Deployment

Currently, we perform an adjunctive prosthetic annuloplasty in all repairs, to prevent further dilatation and reinforce the repair. Reducing the antero-septal annular distance increases the leaflet coaptation surface significantly. As discussed before, annuloplasty band sizes are selected from an echocardiographic nomogram shown in Table 3. In most of our robotic and minimally invasive mitral repairs Cosgrove™ annuloplasty bands (Edwards Lifesciences, Irvine Calif.) have been used. Generally, with myxomatous mitral valves, a “trigone to trigone” posterior band provides optimal coaptation while preserving a “saddle-shaped” systolic configuration. For patients having mitral insufficiency from annual dilatation secondary to ischemia or a cardiomyopathy, we implant a complete annuloplasty ring.

To secure annuloplasty bands, we place the initial interrupted 2-0 Ticron™ (Covidien, Mansfield, MA) suture in the right fibrous trigone and proceed in a clockwise direction. Others use effectively a continuous annuloplasty band suturing technique. [61]The automated Cor-Knot™ (LSI Solutions, Victor, NY) device is ideal to secure annuloplasty rings/bands sutures firmly. [62](Figures 42 and 43) This device has reduced our cross-clamp times dramatically, when compared with former instrument tying.

Figure 42
Descriptive text is not available for this image
Annuloplasty Band Implantation: Cor-knot™ Device (LSI Solutions, Inc., Victor New York): (a) After all annuloplasty band/ring sutures have been placed and exteriorize through the working incision, individual ones are passed through a wire-loop extending from the side-hole in the Cor-knot™ applicator. (b) An individual suture is then withdrawn through the applicator side-hole. (b) While maintaining firm suture countertraction, the tubular end of the applicator should be pressed firmly against the annuloplasty prosthesis surface. On activating the device mechanism, a titanium ‘shim-like’ fastener anchors the suture tightly and then cuts it. (c) This illustration shows the implanted annuloplasty band after all Cor-Knot™ all suture fasteners have been deployed.
Figure 43
Descriptive text is not available for this image
Cor-knot™ applicator device: Exiting annuloplasty band sutures have been organized serially in guides. Thereafter, each one is “tied” using the Cor-knot ™ Device (CKD). (WP) working port/incision; (L ARM) Left robot instrument arm; (R ARM) Right robot instrument arm; (C) Camera; (LAR) Left atrial retractor.
Photograph by the author (WRC).

Replacement Techniques

Direct Vision/videoscopic Mitral Valve Replacements

Generally, a direct vision or videoscopic mitral valve replacement is done in a similar fashion to a sternotomy operation. However, when using long instruments through a small incision, ergonomic limitations often prevent traditional suturing methods. Thus, surgeons must learn different needle angle patterns to compensate for these difficulties. For these minimally invasive mitral replacements, we use a larger working incision to accommodate the valve prosthesis. A chord sparing operation is done either by excising the central portion of the anterior leaflet tissue with chordal remnants sutured to the annulus. Alternatively, if the anterior leaflet native chords must be excused, replacement ePTFE ones should be implanted to help preserve ventricular function. When using long instruments through a small incision, pledgeted annular sutures are placed more easily from the ventricular side. The posterior sutures should “incorporate and fold ” the posterior leaflet against the annulus, leaving native chords intact for support. Valve sutures then are exteriorized, organized in the suture holders, and then passed through the prosthetic valve sewing cuff. After the prosthesis has been seated in the annulus, Cor-Knot™ suture fixation completes the replacement.

Robot-Assisted Mitral Valve Replacements

Robot-assisted mitral valve replacements are similar to repairs with several exceptions.To deliver and position a prosthetic mitral valve, a larger working incision is necessary. Plastic holders are placed around the incision for organizing exteriorized annular sutures. (Figure 44) In patients with degenerated myxomatous valves, tissue excision is easy using the curved robot scissors. However, with either thickened rheumatic tissue or a failed bioprosthesis, the robotic instruments often are not strong enough for excision. In this circumstance, we suggest enlarging the working incision to enable access of standard stronger scissors. Large robotic needle holders are used to place either sub or supra annular pledgeted sutures. (Figure 45A) After placing suture needles externally through the prosthesis sewing ring, the holder is grasped to lower and stabilized the valve into the annular position. (Figure 45 B) Thereafter, we use the Cor-Knot™ device toaffix the sutures firmly against the sewing ring. (Figure 45 C) As mentioned, when a typical valve sparing operation cannot be done, because of adherent rheumatic chords and/or thickened papillary muscles, excised native chords should be replaced with ePTFE neochords.

Figure 44
Descriptive text is not available for this image
Robotic Mitral Valve Replacement: To replace a native thickened or calcified rheumatic mitral valve or to explant a degenerated prosthesis, a larger working incision may be required. In these circumstances robotic scissors often are too delicate. Conversely, with myxomatous leaflet tissue the robotic instruments generally work very well. Instrument ports are the same as described earlier. Gabbay-Frater holders are placed around the incision to organize valve sutures as they exit the incision.
Figure 45
Descriptive text is not available for this image
Robotic Mitral Valve Replacement: These illustrations show the dynamic left atrial retractor in place. (a) The posterior leaflet has been folded into the ventricular side of the annulus with robot placed pledgeted valve sutures. The anterior leaflet has been excised and native chords have been replaced with ePTFE ones. (b) Exteriorized suture needles have been passed through the valve prosthesis sewing cuff by the patient-side assistant. The holder is left in place to facilitate grasping with a robotic needle holder when lowering and positioning the prosthetic valve. (c) We generally anchor prosthetic valves in place by Cor-Knot™ titanium suture clips. However, robot-assistance facilitates tying sutures using traditional methods.


Patients having had prior coronary artery bypass surgery with and intact internal thoracic artery, who have developed mitral insufficiency, can benefit from MIMVS. The morbidity and mortality have been shown to be less than with traditional sternotomy reoperations. [63],[64]The potential benefits include less bleeding and lower risk of ventricular dysfunction in patients with reduced ejection fractions. Because of either regional or global ischemia, patients may either have Carpentier Type 3b ischemic insufficiency and or a Type 1 leak from a dilated annulus. In the past a reduction ring annuloplasty was the standard used for these patients. The recent Cardiac Surgery Network NIH studies showed that mitral insufficiency can recur in up to 50% in patients after a lone ring repair.[65]In this circumstance, replacement may be the best option and can be accomplished using MIMVS techniques. Thus, in patients with severely tethered chords from a dilated ventricle, we recommend valve replacement.

The right thorax reoperative set-up is exactly the same as described earlier for all three minimally invasive approaches. For prosthetic mitral repair or replacement reoperations, we systemically cool patients to 26° C and induce ventricular fibrillation either by rapid pacing or electrically via pacing wires. Generally, old vein grafts that traverse the right atrial wall can be identified in the collapsed heart. The adherent pericardium is entered anterior to the phrenic nerve using the superior pulmonary vein as a landmark. Very little dissection is needed to make the left atriotomy. In patients having Type 3b ischemic insufficiency without a severely dilated ventricle, we use a complete annuloplasty ring to re-establish leaflet coaptation. Using either direct vision, endoscopic, or daVinci™ visualization, valve suture placement is begun along the anterior annulus proceeding clockwise. When replacing a failed mitral bioprosthesis, it is best to make a larger working incision and excise the old prosthesis via direct vision with long instruments. It very important to keep sutures well organized in Gabby-Frater suture guides as they exit from the working incision. If the aortic root has to be displaced for annular exposure, suture placement at the left fibrous trigone can be difficult because of resulting valve insufficiency. By systemic cooling to 26° C, perfusion flow can be reduced safely for a few minutes until several difficult sutures are placed here. The advantage of the robotic method in these reoperations resides in the ability to use minimal atrial retraction, which helps maintain a competent aortic valve.

Tricuspid Repair: Adjunctive

Minimally invasive techniques for tricuspid valve repairs and replacements are very similar for the direct vision, video-assisted, or robot-assisted approaches. Cardiac and valve exposure are exactly the same. Most tricuspid repairs are done for annular dilatation secondary to mitral insufficiency. The mitral valve repair should be done first. It has been our practice to complete the tricuspid annuloplasty the beating heart while the patient is being rewarmed. Alternatively, this can be done during cardiac arrest. As we use jugular and femoral drainage cannulas, both vena cava can be isolated by intra-thoracic tape occlusion. Alternatively, repositioning the inferior vena caval cannula just below the liver (without occlusion) can provide adequate drainage, provided a hard-shell pump reservoir is used to manage any entrained venous air. When the beating heart technique is used to place a tricuspid annuloplasty band/ring, visualization of the posterior right atrium can be impaired by blood. Thus, we place the first sutures at the 2 o’clock position along the annulus and proceed first counterclockwise from this point and then clockwise. If the sutures are Cor-Knot™ secured serially, then the band can be instrument held and distracted to expose the area around Koch’s A-V node bearing triangle.

Cox - Maze Cryoablation: Adjunctive

Other chapters will detail surgery to treat atrial fibrillation, however, the minimally invasive approach is ideal to perform a complete adjunctive Cox-MAZE cryoablation. [56],[66] A specialized daVinci™ instrument is designed to fit the form of currently manufactured round cryoprobes. To be sure that there are no lesion gaps, both robotic as well as 2-D and 3-D video-assistance aid greatly in probe placement. We prefer to apply the Atricure CryoForm™ cryoablation probe for two minutes for each lesion as it provides rapid freeze to -60 C with quick tissue release.

Figures 46, 47 and 48 detail the cryo-lesion set recommended for a complete biatrial ablation.[56],[66]Generally, we perform right atrial cryo-MAZE lesions first, while cooling the patient. (Figure 46A-C) These include: (A) intra-caval (B) trans-atrial - from the oblique sinus crossing the intra-caval line and across to the right atrial appendage (C) tricuspid isthmus – by passing the probe, incorporating the end of the trans-atrial ablation line, into the right atrium via a tiny incision. Then, the probe should subtend the endocardium of the anterior atrial wall and onto the tricuspid annulus. When making this cryolesion the probe is internal to the right coronary artery. We have never had a coronary injury using this endocardial approach. Extra care must be taken to protect the phrenic nerve during this part of the procedure.

Figure 46
Descriptive text is not available for this image
Right Atrial Lesions: The following cryo-lesions are created on the empty beating heart while on cardiopulmonary bypass. (a) The inter-caval freeze line should be placed along the right atrial wall, carefully avoiding the region of the SA node as it includes the superior vena cava. Moreover, care must be taken to keep the probe away from the pericardial region near the phrenic nerve. (b) This lesion should begin at the oblique sinus, crossing the inter-caval line, and extended as far along the right atrium to the appendage. (c) The trans-atrial lesion now should be joined with one that extends to the tricuspid annulus. To perform lesion, a tiny incision is made near the right atrial appendage at the end of the trans-atrial lesion. The cryo-probe then is passed inside the right atrium to the tricuspid annulus to complete the right atrial lesion set.
Figure 47
Descriptive text is not available for this image
Left Atrial Lesions: After cardioplegic cardiac arrest, a generous left atriotomy is made and then the dynamic retractor is used to provide exposure for each cryo-lesion. (a) This first lesion should begin and incorporate the lower end of the atriotomy. It should extend along t he endocardium to surround the left inferior pulmonary vein. (b) The next line should extend along the endocardium between the posterior mitral annulus and left pulmonary veins. It should overlap the previous cryo-lesion. The curved end of the probe should be molded to surround the left superior pulmonary vein. (c) Next the freeze line should continue from the left superior pulmonary vein to the upper end of the left atriotomy. Care must be taken to avoid “skipped” endocardium at each end of the atriotomy.
Figure 48
Descriptive text is not available for this image
Left Atrial Lesions (Continued): (d) The next cryo-lesion should incorporate the mitral valve isthmus by crossing previous lesion A with extension to the mitral valve annulus near the P3 region. (e) The last lesion should begin on the epicardium at the origin of the oblique sinus and extend parallel with endocardial lesion D to cross the coronary sinus. Thereafter, the left atrial appendage is oversewn in two layers with a 4-0 ePTFE suture.

The left atrial cryo-MAZE lesions should be completed before a mitral repair is begun. Again, video-assistance (robotic or videoscopic) facilitates accurate probe placement, especially around the left pulmonary vein and across the coronary sinus. The left atriotomy should span across the origins of the right pulmonary veins with cryo lesions completing a “box lesion” around the left pulmonary veins. The left atrial sequence is as follows (Figure 47A-C): (A) inferiorly from the atriotomy edge and around the right inferior pulmonary vein (B) across the tip of the previous line and then between the mitral annulus and left pulmonary veins to behind the left superior pulmonary vein (C) crossing the last lesion and around the left superior pulmonary vein to connect with the superior (cephalad) left atriotomy edge. Figure 48 shows the remaining left atrial cryo-lesion set. (D) across the mitral isthmus by connecting lesion (A) to the P3 leaflet annulus. (E) through the oblique sinus and across the epicardium and coronary sinus with this line parallel to the previous (D) endocardial isthmus one. Thereafter, the left atrial appendage should be closed in two layers using a 4-0 PTFE suture. Alternatively, the appendage can be occluded by placing an occluding clip using one of the Articulating AtriClip PRO (AtriCURE Inc. Mason, Ohio) devices placed through the transverse sinus.

Complications and Avoiding Them

The most common complications are related to retrograde cardiopulmonary bypass perfusion (neurological or aortic dissection), vascular or cardiac injury (aorta, pulmonary artery, iliac artery, left atrial appendage), pulmonary injury, and/or poor ventricular function. Tables 5,6, and 7 detail these potential complications as well as ways toavoid them. Today, detailed patient screening algorithms can help avoid many of these complications.

Table 5: Perfusion and Aortic Clamping: Pitfalls and Prevention



Retrograde Perfusion

  • Aortic Dissection
  • Leg Ischemia
  • Cerebral Athero-emboli

  • Pre-operative vascular assessment
  • Preoperative CT scan with contrast
  • Echo-guided Seldinger cannulation
  • Monitor leg O2 saturations - distal femoral artery shunt
  • Rt. Axillary artery cannulation

Trans-thoracic Aortic Clamping

  • Aortic injury
  • PA Injury
  • Left atrial appendage tear
  • Left main coronary injury

  • Pre-operative CT scan
  • Direct or endoscopic transverse sinus visualization during clamping
  • Decrease pump flow when clamping

Endoballoon Aortic Occlusion

  • Femoral arterial injury
  • Aortic dissection
  • Atheroma embolization
  • Aortic endoballoon migration
  • Innominate artery occlusion
  • Into left ventricle

  • Pre-operative vascular assessment
  • Preoperative CT scan with contrast
  • Echo-guided Seldinger cannulation
  • Intra-operative TEE

Table 6: Pulmonary: Pitfalls and Prevention



  • Phrenic nerve injury
  • Cross clamp lung injury
  • Unilateral pulmonary edema
  • Residual air leak

  • Visualize the phrenic nerve – limited stretch - care with cryoablation
  • Visualize clamp tip and lung when inserting
  • No barotrauma – PEEP during CPB – frequent lung recruitment – limited lung manipulation – limit transfusions and CPB time
  • Care when placing trans-thoracic sutures and trocars

Table 7: Ventricular Dysfunction: Pitfalls and Prevention



  • Inadequate myocardial protection
  • Ventricular distention
  • Coronary disease
  • Aortic insufficiency
  • Pulmonary artery hypertension
  • Air embolism – coronary
  • Circumflex coronary injury

  • Frequent cardioplegia infusions
  • Systemic cooling
  • Good venous drainage
  • Ventricular venting - no LV distension
  • Milrinone & nitric oxide therapy
  • Meticulous de-airing
  • No air in cardioplegia catheter
  • Meticulous annular sutures – not off annulus

Comparative Effectiveness

Videoscopic Operations

Both a number of observational and propensity-matched series have shown that direct-vision and video-assisted MIMVS is safe with results similar to operations done through a sternotomy (ST). [21],[22],[30],[67],[68],[69],[70],[71],[72],[73] Moreover, short and long-term mitral repair results and survival are the same. Five recent patient series that had over 1000 robot-assisted or endoscopic mitral operations confirmed these tenets.[30],[42],[43],[70],[71] The majority of these patients were operated through a right mini-thoracotomy using retrograde CPB perfusion. The collective group operative mortality was 0.75% and stroke rate was 1.2%. Of these studies Mohr et al reported the largest series with 2829 minimally invasive video-assisted mitral repairs.[71]At that center the freedom from reoperation was 97 and 93% at 5 years and 10 years, respectively, and the long-term survival was 87 and 74% at 5 and 10 years, respectively. These data are comparable to large series of patients undergoing a mitral valve repair through a sternotomy.

A meta-analysis from the International Society of Minimally Invasive Cardiac Surgery showed that versus a conventional sternotomy, MIMVS patients had less postoperative atrial fibrillation, chest tube drainage, transfusions, and infections.[22]There was no difference in the repair quality or freedom from reoperation, and the operative mortality was not statistically different (1.5% ST vs 1.2% MIMVS). Other benefits included reduced intensive care times and length of hospitalization as well as faster return to normal activity. These advantages were shown despite significantly longer perfusion and aortic cross clamp times. Conversely, they reported comparatively more strokes (1.2% vs 2.0%), aortic dissections (0.0% ST vs 0.2%), and phrenic nerve injuries (0% ST vs 3.0%) when comparing the sternotomy group to the minimally invasive cohort. The independent meta-analyses of Modi, Cao, and Sundermann showed similar benefits and clinical outcomes as with traditional sternotomy operations, but with similar incidences of strokes to sternotomy patients. [67],[68],[69] Additionally, the recent Virginia Cardiac Services Quality Initiative report of propensity matched robotic, minimally invasive, and conventional mitral valve surgery showed equivalence in patient safety and clinical outcomes. [72] Their earlier multi-instutional analysis showed that minimally invasive mitral surgery could be done without adverse economic consequences. [73] When using retrograde femoral CPB perfusion or mode of aortic clamping, stroke risks have been a major concern since the inception of MIMVS. [22],[31],[32],[33],[74],[75],[76],[77],[78]In most of these studies no CT angiographic screening was not done prior to using retrograde perfusion or aortic clamping. The use of endoaortic balloon occlusion in early series showed a higher rate of strokes and aortic dissections, owing to retrograde endo-balloon catheter passage in unscreened patients. Some series showed that trans-thoracic aortic clamping was safer than using the endoballoon catheter.[74]However, Grossi found that there were no differences in perioperative strokes between antegrade and retrograde perfusion when high-risk vascular patients were CT screened.[78]However, several series, in which CT patient screening was done, both endoballoon and direct aortic clamping had similar neurological events.[31],[32],[33]A metanalysis by Modi and Chitwood determined retrograde perfusion to be safe when patients were screened for aortic and peripheral vascular disease.[77]A recent study from the Cleveland Clinic showed that by prior CT screening patients, who underwent retrograde femoral perfusion for robot-assisted mitral surgery, reduced the stroke rate from 2.0 to 0.8%. [43] Thus, it is incumbent, if not mandatory, that patients at risk for peripheral or central atherosclerosis should have a careful physical examination and undergo CT analysis before considering using retrograde perfusion. In the presence of peripheral arterial atherosclerosis, alternate antegrade perfusion routes should be considered and include either direct trans-thoracic aortic cannulation or using the axillary artery. We suggest developing a center screening algorithm to ensure the safest clinical results.

In our 167 reoperative minimally invasive mitral valve series, the overall mortality was 3.0% and in the last 50% of patients was 0.0%. The overall neurological complication rate was 2.4%. [63]Most of these patients had a prior coronary artery bypass operation. Sardari found in a meta-analysis of his reoperative MIMVS patients, similar mortality and morbidity with improved lengths of stay and less bleeding. [64]Losennoalso found better outcomes in reoperative patients using the MIMVS approach.[79]

Robot-Assisted Operations

Our early series of robot-assisted mitral repairs (RMVP), done between years 2000 and 2010, consisted of 454 patients who underwent a lone procedure with 86 having a concomitant atrial fibrillation ablation (M-RMVP).[39]In the lone repair cohort, cross clamp and CPB times averaged 116 and 153 minutes, respectively, and the operative mortality was 0.2%. Of those treated for preoperative atrial fibrillation, 96.5% were atrial fibrillation and drug free at 351 days (15–946 days). During patient follow-up, 2.9% of all required a re-operation for a failed repair. This series included the original FDA clinical trial patients and was the inaugural United States robotic mitral valve repair series.

To date, we have performed well over 1000 operations at our institution. A single surgeon’s (WRC) experience of 944 RMVPs (years 2000–2015) included the inaugural FDA trial series. [80] For these operations, the original daVinci, the daVinci S™, and the daVinci SI™ devices were used serially. Trento showed improving results with every robotic device iteration, as did our group.[41] Of course, in this series surgeon and team experience increased with time as did increased pathology complexity.

In our later series 677 (68%) had a lone robot assisted mitral repair (RMVP), 323 (32%) had a concurrent cryo-MAZE procedure (M-RMVP) and 38 (3.8%) were reoperations in patients, who had a prior sternotomy (Re-RMVP). For the entire series, the in-hospital mortality was 1.4%; however, for an isolated RMVP was 0.15%. Of the entire series, 2.5% had reoperations for a failed repair, which was 1.7% for a lone RMVP. For each of these cohorts the mean CPB times were 148, 187, and 176 minutes, respectively. For RMVP and M-RMVP groups, mean aortic cross-clamp times were 108 and 128 minutes, respectively. Re-RMVP operations were done under hypothermic ventricular fibrillation (113 minutes, mean). In the RMVP patients, major complications included myocardial infarctions (0.9%) and strokes (0.9%). Re-explorations for bleeding were required in 2.7% of patients. There were two incidences of a residual phrenic nerve palsy. Packed red cell transfusions were required in 28% of patients and 38% with any blood product. Of the RMVP patients, 56.9% were discharged within 4 days with 75.0% within 5 days. When leaving the operating room, 97.1% of patients had no or trivial mitral regurgitation. Recently, the Cedars-Sinai, Emory University and Cleveland Clinic groups reported series with over 1000 RMVP operations. [42][81],[43] Other large series of robotic mitral repairs showed have shown similar clinical outcomes.[40],[41] Each large series has shown conclusively that robot-assisted mitral surgery is safe and can render excellent high-quality results. However, these are experienced mitral repair surgeons who have focused on the heart team approach and have learned the best ways to provide safety, increase operative efficiency, and carefully select patients for the robot-assisted method. Adjunctive newer techniques and devices, such as less complicated repair techniques and annuloplasty suturing devices, along with team experience, have reduced cross clamp times. These factors have allowed surgeons to expanded robot-assistance to combination operations, including concomitant tricuspid repairs and cryo-MAZE procedures to treat atrial fibrillation.

The Minimally Invasive Philosophy

This chapter outlines the best ways to learn and adopt the minimally invasive philosophy. First, surgeons and their teams must truly believe that these approaches benefit patients and improve clinical outcomes. Next, appropriate training is mandatory even for experienced surgeons and their teams. There will be surgeon and team frustration at first because of slower operative times, which also incurs more economic costs. The hospital administration also must be willing to support the program by understanding these initial cost increases per procedure compared with traditional sternotomy operations. However, by both gaining operative experience and tailoring patient intensive care workflow, economic benefits can be achieved, especially in hospitals where the cost of the robotic device is amortized over several surgical services. If these tenets are adopted early, then minimally invasive and/or robot-assisted heart valve surgery can benefit patients, the hospital, and surgeons. We believe that this philosophy is the pathway to the future of cardiac surgery.

The Future

The future is minimally invasive heart valve surgery. As the singer Bob Dylan (Nobel laureate 2017) said in 1964 “The Times They Are a Changin” and indeed they are changing ߪ. and fast. Bruce Springsteen also once wrote “the door is open but the ride ain’t free”. We believe that they both could have predicted the future of heart valve surgery as these statements can be applied today. Long-term clinical data have established the expectation that mitral repair is the “gold” standard. Cardiology guidelines now suggest that asymptomatic patients should be referred for surgery earlier, before atrial fibrillation, ventricular dysfunction, annular dilatation, and/or advanced valve pathology develops. Echocardiographic imaging continues to advance every year providing improved operative guidance. In experienced hands, minimally invasive and robot-assisted repair results now parallel those of sternotomy operations. Although minimally invasive mitral surgery in the United States has lagged behind European centers, our younger surgeons now know that the future favors even less invasive operations. Finally, the Internet and social media have given patients many more clinical choices and empowerment than in the past. They are more knowledgeable about their health care and demand to participate. Collectively, these facts seem like the “Sword of Damocles” for surgeons who scoff at the evolving mitral technology. The advent and application of the Mitra-Clip™ (Abbott Vascular, Santa Clara, CA) and completion of the Everest Trials proved that in high-risk patients, mitral regurgitation could be reduced to the point where symptoms abated. Even when some residual regurgitation remained, patients were better. This set the stage for off pump ePTFE neochord chord replacements to reduce leaflet prolapse in patients with a nondilated annulus. Also, transcatheter annuloplasty prostheses are being developed to reduce dilatation. The co-application of transcatheter annular reduction and leaflet/chordal devices may be an answer for some patients. The story regarding treatment of ischemic mitral regurgitation is not complete, but we do know that transcatheter replacements could stabilize some patients more than a surgical annuloplasty repair or a Mitra-Clip™. Heretofore, less attention was paid to the leaking “lonely” tricuspid valve. The natural history of this malady is better understood and transcatheter remedies are evolving.

Where does that leave surgeons? We are in the enviable position of being able to surgically repair mitral valves while simultaneously coopting trans-catheter therapy into our armamentarium. Surgeons will have to adopt the minimally invasive and transcatheter philosophy to remain as a major cardiovascular structuralist in the future. The heart team approach to patient referrals, evaluations, and treatment has matured in most institutions, and cardiac surgeons who embrace this philosophy participant will be winners as will their patients. Owing to proven long-lasting results, surgical repair in younger patients should continue for a number of years. To be sure, if we can provide the least invasive complete repairs, based on sound engineering principles, we will prevail. Hopefully, principles set forth in this chapter comprise one entry portal to the future of mitral valve therapy.


  1. Brunton TL. Preliminary note on the possibility of treating mitral stenosis by surgical methods. Lancet 1902;1:352.
  2. Souttar HS. THE SURGICAL TREATMENT OF MITRAL STENOSIS. Br Med J. 1925;2(3379):603-6.  [PMID:20772184]
  3. Smithey HG, Boone JA, Stallworth JM. Surgical treatment of constrictive valvular disease of the heart. Surg Gynecol Obstetr 1950;90:175-192.
  4. BAILEY CP. The surgical treatment of mitral stenosis (mitral commissurotomy). Dis Chest. 1949;15(4):377-97.  [PMID:18114648]
  5. Harken DE, Ellis LB, Ware PF, Norman LR. The surgical treatment of mitral stenosis. I. Valvuloplasty. NEJM 1948:239;801-809.
  6. LILLEHEI CW, GOTT VL, DEWALL RA, et al. Surgical correction of pure mitral insufficiency by annuloplasty under direct vision. J Lancet. 1957;77(11):446-9.  [PMID:13476052]
  7. BRAUNWALD NS, COOPER T, MORROW AG. Complete replacement of the mitral valve. Successful clinical application of a flexible polyurethane prosthesis. J Thorac Cardiovasc Surg. 1960;40:1-11.  [PMID:13804040]
  8. Tribble CG, Killinger WA, Harman PK, et al. Anterolateral thoracotomy as an alternative to repeat median sternotomy for replacement of the mitral valve. Ann Thorac Surg. 1987;43(4):380-2.  [PMID:3566384]
  9. Grüntzig AR, Senning A, Siegenthaler WE. Nonoperative dilatation of coronary-artery stenosis: percutaneous transluminal coronary angioplasty. N Engl J Med. 1979;301(2):61-8.  [PMID:449946]
  10. Pompili MF, Stevens JH, Burdon TA, et al. Port-access mitral valve replacement in dogs. J Thorac Cardiovasc Surg. 1996;112(5):1268-74.  [PMID:8911323]
  11. Fann JI, Pompili MF, Burdon TA, et al. Minimally invasive mitral valve surgery. Semin Thorac Cardiovasc Surg. 1997;9(4):320-30.  [PMID:9352947]
  12. Navia JL, Cosgrove DM. Minimally invasive mitral valve operations. Ann Thorac Surg. 1996;62(5):1542-4.  [PMID:8893611]
  13. Aklog L, Adams DH, Couper GS, et al. Techniques and results of direct-access minimally invasive mitral valve surgery: a paradigm for the future. J Thorac Cardiovasc Surg. 1998;116(5):705-15.  [PMID:9806377]
  14. Baldwin JC. Editorial (con) re minimally invasive port-access mitral valve surgery. J Thorac Cardiovasc Surg. 1998;115(3):563-4.  [PMID:9535442]
  15. Cooley DA. Antagonist's view of minimally invasive heart valve surgery. J Card Surg. 2000;15(1):3-5.  [PMID:11204385]
  16. Mihaljevic T, Cohn LH, Unic D, et al. One thousand minimally invasive valve operations: early and late results. Ann Surg. 2004;240(3):529-34; discussion 534.  [PMID:15319724]
  17. Mohr FW, Onnasch JF, Falk V, et al. The evolution of minimally invasive valve surgery--2 year experience. Eur J Cardiothorac Surg. 1999;15(3):233-8; discussion 238-9.  [PMID:10333015]
  18. Reichenspurner H, Boehm D, Reichart B. Minimally invasive mitral valve surgery using three-dimensional video and robotic assistance. Semin Thorac Cardiovasc Surg. 1999;11(3):235-43.  [PMID:10451254]
  19. Vanermen H, Farhat F, Wellens F, et al. Minimally invasive video-assisted mitral valve surgery: from Port-Access towards a totally endoscopic procedure. J Card Surg. 2000;15(1):51-60.  [PMID:11204388]
  20. McClure RS, Cohn LH, Wiegerinck E, et al. Early and late outcomes in minimally invasive mitral valve repair: an eleven-year experience in 707 patients. J Thorac Cardiovasc Surg. 2009;137(1):70-5.  [PMID:19154906]
  21. Svensson LG, Atik FA, Cosgrove DM, et al. Minimally invasive versus conventional mitral valve surgery: a propensity-matched comparison. J Thorac Cardiovasc Surg. 2010;139(4):926-32.e1-2.  [PMID:19945121]
  22. Cheng DC, Martin J, Lal A, et al. Minimally invasive versus conventional open mitral valve surgery: a meta-analysis and systematic review. Innovations (Phila). 2011;6(2):84-103.  [PMID:22437892]
  23. Allen DS, Graham EA. Intracardiac surgery – a new method: preliminary report. JAMA 1922;79:1028-1030
  24. Harken DE, Glidden EM. Experiments in intracardiac surgery. II. Intracardiac Visualization. J Thorac Surg 1943;12:566-572.
  25. Sakakibara H, Ichikawa T, Hattori J. An intraoperative method for observation of cardiac septal defect using a cardioscope. Operation 1956;10:285-290.
  26. Kaneko Y, Kohno T, Ohtsuka T, et al. Video-assisted observation in mitral valve surgery. J Thorac Cardiovasc Surg. 1996;111(1):279-80.  [PMID:8551780]
  27. Carpentier A, Loulmet D, Carpentier A, et al. [Open heart operation under videosurgery and minithoracotomy. First case (mitral valvuloplasty) operated with success]. C R Acad Sci III. 1996;319(3):219-23.  [PMID:8761668]
  28. Chitwood WR, Elbeery JR, Moran JF. Minimally invasive mitral valve repair using transthoracic aortic occlusion. Ann Thorac Surg. 1997;63(5):1477-9.  [PMID:9146354]
  29. Felger JE, Chitwood WR, Nifong LW, et al. Evolution of mitral valve surgery: toward a totally endoscopic approach. Ann Thorac Surg. 2001;72(4):1203-8; discussion 1208-9.  [PMID:11603437]
  30. Modi P, Rodriguez E, Hargrove WC, et al. Minimally invasive video-assisted mitral valve surgery: a 12-year, 2-center experience in 1178 patients. J Thorac Cardiovasc Surg. 2009;137(6):1481-7.  [PMID:19464468]
  31. Atluri P, Goldstone AB, Fox J, et al. Port access cardiac operations can be safely performed with either endoaortic balloon or Chitwood clamp. Ann Thorac Surg. 2014;98(5):1579-83; discussion 1583-4.  [PMID:25266263]
  32. Bentala M, Heuts S, Vos R, et al. Comparing the endo-aortic balloon and the external aortic clamp in minimally invasive mitral valve surgery. Interact Cardiovasc Thorac Surg. 2015;21(3):359-65.  [PMID:26093955]
  33. Barbero C, Krakor R, Bentala M, et al. Comparison of Endoaortic and Transthoracic Aortic Clamping in Less-Invasive Mitral Valve Surgery. Ann Thorac Surg. 2018;105(3):794-798.  [PMID:29289366]
  34. Carpentier A, Loulmet D, Aupècle B, et al. [Computer assisted open heart surgery. First case operated on with success]. C R Acad Sci III. 1998;321(5):437-42.  [PMID:9766192]
  35. Autschbach R, Onnasch JF, Falk V, et al. The Leipzig experience with robotic valve surgery. J Card Surg. 2000;15(1):82-7.  [PMID:11204392]
  36. Chitwood WR, Nifong LW, Elbeery JE, et al. Robotic mitral valve repair: trapezoidal resection and prosthetic annuloplasty with the da vinci surgical system. J Thorac Cardiovasc Surg. 2000;120(6):1171-2.  [PMID:11088044]
  37. Nifong LW, Chitwood WR, Pappas PS, et al. Robotic mitral valve surgery: a United States multicenter trial. J Thorac Cardiovasc Surg. 2005;129(6):1395-404.  [PMID:15942584]
  38. Chitwood WR, Rodriguez E, Chu MW, et al. Robotic mitral valve repairs in 300 patients: a single-center experience. J Thorac Cardiovasc Surg. 2008;136(2):436-41.  [PMID:18692654]
  39. Nifong LW, Rodriguez E, Chitwood WR. 540 consecutive robotic mitral valve repairs including concomitant atrial fibrillation cryoablation. Ann Thorac Surg. 2012;94(1):38-42; discussion 43.  [PMID:22560967]
  40. Suri RM, Burkhart HM, Rehfeldt KH, et al. Robotic mitral valve repair for all categories of leaflet prolapse: improving patient appeal and advancing standard of care. Mayo Clin Proc. 2011;86(9):838-44.  [PMID:21757782]
  41. Ramzy D, Trento A, Cheng W, et al. Three hundred robotic-assisted mitral valve repairs: the Cedars-Sinai experience. J Thorac Cardiovasc Surg. 2014;147(1):228-35.  [PMID:24199760]
  42. Murphy DA, Moss E, Binongo J, et al. The Expanding Role of Endoscopic Robotics in Mitral Valve Surgery: 1,257 Consecutive Procedures. Ann Thorac Surg. 2015;100(5):1675-81; discussion 1681-2.  [PMID:26296272]
  43. Gillinov MA, Mihaljevic T, Javadikasgari H, et al. Safety and effectiveness of robotically assisted mitral valve surgery: analysis of 1000 consecutive cases. J Thorac Cardiovasc Surg. 2018 Jan;155(1):82-91
  44. Mihaljevic T, Koprivanac M, Kelava M, et al. Value of robotically assisted surgery for mitral valve disease. JAMA Surg. 2014;149(7):679-86.  [PMID:24848944]
  45. Suri RM, Thompson JE, Burkhart HM, et al. Improving affordability through innovation in the surgical treatment of mitral valve disease. Mayo Clin Proc. 2013;88(10):1075-84.  [PMID:24079678]
  46. Rodriguez E, Nifong LW, Bonatti J, et al. Pathway for Surgeons and Programs to Establish and Maintain a Successful Robot-Assisted Adult Cardiac Surgery Program. Ann Thorac Surg. 2016;102(1):340-4.  [PMID:27343505]
  47. Suri RM, Dearani JA, Mihaljevic T, et al. Mitral valve repair using robotic technology: Safe, effective, and durable. J Thorac Cardiovasc Surg. 2016;151(6):1450-4.  [PMID:27012453]
  48. Carpenter AJ, Yang SC, Uhlig PN, et al. Envisioning simulation in the future of thoracic surgical education. J Thorac Cardiovasc Surg. 2008;135(3):477-84.  [PMID:18329455]
  49. Marshall, MB. Simulation in cardiothoracic surgery. CTSNet 2009
  50. Sardari Nia P, Heuts S, Daemen J, et al. Preoperative planning with three-dimensional reconstruction of patient's anatomy, rapid prototyping and simulation for endoscopic mitral valve repair. Interact Cardiovasc Thorac Surg. 2017;24(2):163-168.  [PMID:27677879]
  51. Tanaka A, Graddy C, Simpson K, et al. Robotic surgery simulation validity and usability comparative analysis. Surg Endosc. 2016;30(9):3720-9.  [PMID:26581619]
  52. Whitehurst SV, Lockrow EG, Lendvay TS, et al. Comparison of two simulation systems to support robotic-assisted surgical training: a pilot study (Swine model). J Minim Invasive Gynecol. 2015;22(3):483-8.  [PMID:25543068]
  53. Leopaldi AM, Wrobel K, Speziali G, van Tuijl, Drasutiene A. Chitwood WR Jr. The dynamic cardiac biosimulator: A method for training physicians in beating-heart mitral valve repair procedures. Jour Thorac Cardiovasc Surg 2018:135;147-155
  54. Otto CM, Nishimura RA, Bonow RO, Carabello BA, Erwin JP 3rd, Gentile F, Jneid H, Krieger EV, Mack M, McLeod C, O’Gara PT, Rigolin VH, Sundt TM 3rd, Thompson A, Toly C. 2020 ACC/AHA Guideline for the Management of Patients With Valvular Heart Disease: Executive Summary: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation. 2021 Feb 2;143(5):e35-e71.
  55. Vahanian A, Beyersdorf F, Praz F, Milojevic M, Baldus S, Bauersachs J, Capodanno D, Conradi L, De Bonis M, De Paulis R, Delgado V, Freemantle N, Gilard M, Haugaa KH, Jeppsson A, Jüni P, Pierard L, Prendergast BD, Sádaba JR, Tribouilloy C, Wojakowski W; ESC/EACTS Scientific Document Group. 2021 ESC/EACTS Guidelines for the management of valvular heart disease. Eur Heart J. 2022 Feb 12;43(7):561-632.
  56. Chitwood WR, Jr. Atlas of Robotic Cardiac Surgery. New York: Springer, 2014.
  57. Kypson AP, Sanderson DA, Nifong LW, et al. Cardiopulmonary bypass and myocardial protection for minimally invasive cardiac surgery. In: Gravlee GP, Davis RF, Hammon JW, et al. editors. Cardiopulmonary Bypass and Mechanical Support: Principles & Practice. Philadelphia: Wolters Kluwer, 2016.
  58. Hummel BW, Buss RW, DiGiorgi PL, et al. Myocardial Protection and Financial Considerations of Custodiol Cardioplegia in Minimally Invasive and Open Valve Surgery. Innovations (Phila). 2016;11(6):420-424.  [PMID:27879532]
  59. Marie-Elisabeth S, Laufer G, Wisser W. The Endoscopic Aesculap Einstein Vision 3D System: can it compete with the daVinci System? (abstract). AATS Mitral Valve Conclave, 2015.
  60. Carpentier A, Adams DH, Filsoufi F. Carpentier’s Reconstructive valve Surgery: From Valve Analysis to Valve Reconstruction. 2010;Saunders (Elsevier), New York.
  61. Mihaljevic T, Jarrett CM, Gillinov AM, et al. A novel running annuloplasty suture technique for robotically assisted mitral valve repair. J Thorac Cardiovasc Surg. 2010;139(5):1343-4.  [PMID:20189192]
  62. Etiwy M, Javadikasgari H, Houghtaling P, et al. Automated Titanium Fasteners Versus Hand-Tied Knots: A Randomized Controlled Trial. Ann Thorac Surg. 2018;106(4):1160-1163.  [PMID:29883653]
  63. Arcidi JM, Rodriguez E, Elbeery JR, et al. Fifteen-year experience with minimally invasive approach for reoperations involving the mitral valve. J Thorac Cardiovasc Surg. 2012;143(5):1062-8.  [PMID:21821266]
  64. Gaemen JHT, Heuts S,, Olsthoon JR, Maessen JG, Sardari P. Right minithoracotomy versus median sternotomy for reoperative mitral valve surgery: a systematic review and meta-analysis of observational studies. Eur J cardiothoracic Surg 2018:Apr 23 (Epub ahead of print)
  65. Michler RE, Smith PK, Parides MK, et al. Two-Year Outcomes of Surgical Treatment of Moderate Ischemic Mitral Regurgitation. N Engl J Med. 2016;374(20):1932-41.  [PMID:27040451]
  66. Rodriguez E, Cook RC, Chu MWA, Chitwood Jr. Minimally invasive bi-atrial cryoMaze operation for atrial fibrillation . Op Tech in Thorac Cardiovasc Surg2008:14;208–223
  67. Modi P, Hassan A, Chitwood WR. Minimally invasive mitral valve surgery: a systematic review and meta-analysis. Eur J Cardiothorac Surg. 2008;34(5):943-52.  [PMID:18829343]
  68. Cao C, Gupta S, Chandrakumar D, et al. A meta-analysis of minimally invasive versus conventional mitral valve repair for patients with degenerative mitral disease. Ann Cardiothorac Surg. 2013;2(6):693-703.  [PMID:24349970]
  69. Sündermann SH, Sromicki J, Rodriguez Cetina Biefer H, et al. Mitral valve surgery: right lateral minithoracotomy or sternotomy? A systematic review and meta-analysis. J Thorac Cardiovasc Surg. 2014;148(5):1989-1995.e4.  [PMID:24589199]
  70. Glauber M, Miceli A, Canarutto D, et al. Early and long-term outcomes of minimally invasive mitral valve surgery through right minithoracotomy: a 10-year experience in 1604 patients. J Cardiothorac Surg. 2015;10:181.  [PMID:26643038]
  71. Davierwala PM, Seeburger J, Pfannmueller B, et al. Minimally invasive mitral valve surgery: "The Leipzig experience". Ann Cardiothorac Surg. 2013;2(6):744-50.  [PMID:24349976]
  72. Hawkins RB, Mehaffey JH, Mullen MG, et al. A propensity matched analysis of robotic, minimally invasive, and conventional mitral valve surgery. Heart. 2018;104(23):1970-1975.  [PMID:29915143]
  73. Downs EA, Johnston LE, LaPar DJ, et al. Minimally Invasive Mitral Valve Surgery Provides Excellent Outcomes Without Increased Cost: A Multi-Institutional Analysis. Ann Thorac Surg. 2016;102(1):14-21.  [PMID:27041453]
  74. Reichenspurner H, Detter C, Deuse T, et al. Video and robotic-assisted minimally invasive mitral valve surgery: a comparison of the Port-Access and transthoracic clamp techniques. Ann Thorac Surg. 2005;79(2):485-90; discussion 490-1.  [PMID:15680820]
  75. Murzi M, Cerillo AG, Gasbarri T, et al. Antegrade and retrograde perfusion in minimally invasive mitral valve surgery with transthoracic aortic clamping: a single-institution experience with 1632 patients over 12 years. Interact Cardiovasc Thorac Surg. 2017;24(3):363-368.  [PMID:28040754]
  76. Khan H, Hadjittofi C, Uzzaman M, et al. External aortic clamping versus endoaortic balloon occlusion in minimally invasive cardiac surgery: a systematic review and meta-analysis. Interact Cardiovasc Thorac Surg. 2018;27(2):208-214.  [PMID:29506260]
  77. Modi p, Chitwood WR, Jr, retrograde femoral arterial perfusion and stroke risk during minimally invasive mitral valve surgery: is there cause for concern? Ann Cardiothorac Surg 2013;2:1-5.
  78. Bedeir K, Reardon M, Ramchandani M, et al. Elevated Stroke Risk Associated With Femoral Artery Cannulation During Mitral Valve Surgery. Semin Thorac Cardiovasc Surg. 2015;27(2):97-103.  [PMID:26686431]
  79. Losenno KL, Jones PM, Valdis M, et al. Higher-risk mitral valve operations after previous sternotomy: endoscopic, minimally invasive approach improves patient outcomes. Can J Surg. 2016;59(6):399-406.  [PMID:28234615]
  80. Chitwood WR. Robotic mitral valve surgery: overview, methodology, results, and perspective. Ann Cardiothorac Surg. 2016;5(6):544-555.  [PMID:27942486]
  81. Roach A, Trento A, Emerson D, et al. Durable Robotic Mitral Repair of Degenerative Primary Regurgitation With Long-Term Follow-Up. Ann Thorac Surg. 2022;114(1):84-90.  [PMID:34437853]
  82. Grossi EA, Loulmet DF, Schwartz CF, et al. Evolution of operative techniques and perfusion strategies for minimally invasive mitral valve repair. J Thorac Cardiovasc Surg. 2012;143(4 Suppl):S68-70.  [PMID:22285326]
Last updated: April 14, 2023