Minimally invasive mitral valve surgery is commonly performed through a right anterior minithoracotomy. Although this approach has been proven to be safe and effective, achieving the repair even of complex mitral lesions,1–3 a number of disadvantages come with this approach, including the need of extra incisions in the groin and in the neck to perform arterial and venous cannulations and the impossibility to perform concomitant procedures on the aortic valve and on the aortic root. Moreover, a recent meta-analysis comparing the outcomes of minimally invasive toward conventional mitral surgery shows that the minithoracotomy approach could be associated to an increased risk of myocardial infarction, aortic dissection, groin infections, and phrenic nerve palsy when compared with the conventional sternotomy approach, although the level of this evidence is low because of the nonrandomized design of most of the included studies.4 Last but not least is the necessity to have special and expensive surgical instruments (cannulae, needle holders, rib retractors, and aortic clamps) to perform mitral repairs through a minithoracotomy.
Surgical access to the mitral valve through the left atrial roof (LAR) has been initially described by Meyer in 1965 as a variation of the vertical biatrial approach to the mitral valve.5,6 In 2007, given the increased interest in minimally invasive surgery, the LAR approach was chosen at our institution as the preferential approach to repair or replace mitral valve.
From October 2007 to October 2011, a total of 512 consecutive patients underwent mitral valve surgery by a single surgeon using the LAR approach for isolated mitral surgery or combined procedures including the mitral valve and the aortic valve, the aortic root, or the ascending aorta. Only patients requiring concomitant tricuspid valve procedures or specifically requesting a right minithoracotomy for aesthetical reasons underwent a surgical approach different from the LAR. In case of concomitant tricuspid valve repair, patients underwent full sternotomy, bicaval cannulation, and a transatrial approach to the mitral valve, allowing to expose the tricuspid valve as well.
The indications to choose a J-shaped upper ministernotomy were the presence of diabetes, chronic obstructive pulmonary disease, obesity, age older than 75 years, and osteoporosis, to reduce the chances of wound complications or, in younger female patients, for aesthetical reasons when a combined procedure on the aortic and mitral valve was planned and a submammary minithoracotomy was therefore not possible.
After general anesthesia was induced, the heart was accessed via a midsternotomy or via a J-shaped upper ministernotomy and 300 mg/kg of heparin was given intravenously. The ascending aorta and right atrium were cannulated, the latter with a single two-stage cannula. CO2 was insufflated into the pericardium. The cardiopulmonary bypass was instituted with a target temperature of 32°C. In cases of persistent AF with a diameter of the left atrium of less than 5 cm or of paroxysmal AF, pulmonary vein isolation was performed using a radiofrequency clamp (Cardioablate LP Medtronic Inc, Minneapolis, MN USA, and Cobra Adhere XL Estech, San Ramon, CA USA), and subsequently, a vent was positioned into the right superior pulmonary vein. The aorta was then cross-clamped and antegrade cold blood cardioplegia was delivered into the ascending aorta if the procedure involved only the mitral valve.
In cases of a planned concomitant procedure on the ascending aorta, aortic root, or aortic valve, the aorta was opened and resected first to further enhance the exposure through the LAR and the cardioplegia was delivered directly into the coronary ostia.
Once the cardioplegia was given, the ascending aorta was completely deflated by the suction on the root vent and was gently displaced by the assistant, exposing the LAR. A vertical incision of the LAR (Fig. 1) was performed parallel to the superior vena cava and the ascending aorta between the lower edge of the right pulmonary artery and the right atrium, carefully avoiding the sinoatrial node artery normally running below the lower end of this incision. A Ross retractor was used to expose the mitral valve. A video illustrating the LAR approach can be found online at http://links.lww.com/INNOV/A23. A mitral repair or replacement was performed as planned. In all the mitral repairs, a complete semirigid ring (Memo 3D, Sorin Group, Saluggia, Italy) was implanted.
In patients with mitral valve prolapse, the most commonly performed procedure on the posterior leaflet was the quadrangular or triangular resection of P2 associated with a folding plasty at the base of the leaflet with a 4/0 prolene to reduce the extent of the resection and prevent systolic anterior motion. Prolapses of P1 and P3 were commonly addressed with the implant of 4/0 Gore-Tex chordae. Prolapses of the anterior leaflet were repaired by secondary chordae transposition or implantation of additional 4/0 Gore-Tex chordae.
In patients with chronic ischemic mitral regurgitation, according to the degree of leaflet tethering demonstrated at the TOE, a repositioning of the posterior papillary muscle and a cut and transfer of the secondary chordae in the primary position were performed together with a complete semirigid annuloplasty ring.
The left atrium was then closed with a 3/0 and 4/0 prolene running suture. Once the deairing was completed, the cross-clamp was removed, reperfusing the heart. The following steps, including weaning from cardiopulmonary bypass and chest closure, were performed with conventional techniques.
The follow-up was complete for 481 of 500 discharged patents (96.2%). Follow-up length ranged between 6 and 48 months, whereas the mean was 22.9 ± 4 months. All the patients had an outpatient appointment and a transthoracic echo 6 weeks and 1 year after the operation and, after that, once a year or earlier if it was clinically indicated.
Categorical variables were presented as absolute numbers and percentage. Continuous variables were presented as mean ± standard deviation. The Kaplan-Meier product limit method was used to calculate survival. The statistical software used was Stat-View 5.0 statistical software package (SAS Institute Inc, Cary, NC USA).
Preoperative characteristics are depicted in Table 1. Although all the common etiologies of mitral valve disease are represented in this sample, degenerative and ischemic mitral regurgitation was by far the most frequent, accounting for more than 80% of the patient population included in this study.
Intraoperatively, almost two thirds of the patients underwent a full sternotomy, whereas the other third underwent a J-shaped ministernotomy (Table 2). The overall incidence of combined procedure was 36.5% (39.1% through a full sternotomy and 32.2% through a ministernotomy; P = 0.12).
Of 460 patients operated for mitral regurgitation, 398 underwent mitral valve repair (86.5%) and 62 patients had mitral valve replacement. Among the patients who underwent mitral valve repair, the most frequently repaired leaflet was the posterior (62%), whereas 29% required a repair on both the mitral leaflets and only 9% required a repair of the anterior leaflet only. Combined procedures on the aortic valve or the aortic root were performed in 154 patients (30%). Pulmonary vein isolation and left atrial appendage closure was performed in 20% of patients.
In-hospital mortality was 2.3% (12 patients). A laceration of the roof of the left atrium requiring a repair with a pericardial patch occurred in 10 patients (1.9%), whereas reexploration for bleeding was necessary in 25 patients (4.8%). Only 3.1% of patients required a permanent pacemaker insertion for the persistence of atrioventricular block. The incidence of myocardial infarction was 1.7% (3 patients). The overall incidence of sternal wound infection was 1.95% (10 patients), but none of the patients who underwent a ministernotomy had any complication to the wound. Mean in-hospital stay was 7.1 ± 3.9 days.
Overall 4-year survival was 91% ± 4.2%. For patients who underwent mitral valve repair, the freedom from recurrence of mitral regurgitation greater than 2 was 97.4% at 1 year.
Interest toward minimally invasive cardiac surgery has been growing during the last two decades, aiming to make cardiac surgery less traumatic, shorten postoperative recovery, and increase patient satisfaction.
A right anterior minithoracotomy is currently the most commonly used minimally invasive approach to the mitral valve, but this incision limits the possibility to associate procedures to the aortic valve; carries a slightly higher risk of myocardial infarction, aortic dissection, groin wound infections, and phrenic nerve palsy4; and requires special surgical instruments.
Ministernotomy is commonly used for aortic valve and aortic root procedures: compared with a full sternotomy, the use of ministernotomy is associated to less postoperative pain,7,8 lower incidence of sternal wound infections,8 improved early respiratory function and earlier extubation,8,9 faster return to normal life, and less need for posthospital rehabilitation.10 Nevertheless, performing mitral surgery via a J-shaped upper ministernotomy and using the conventional paraseptal or transeptal incisions is extremely difficult because of the limited space to perform a bicaval cannulation and very limited visualization of the mitral valve.
Initially described in 1965 by Meyer et al,5 the LAR incision has the potential to combine adequate mitral valve exposure with limited tissue trauma. Nevertheless, in the modern era, the only report on the use of LAR is a case series of 95 patients who underwent mitral surgery via a lower ministernotomy by Ngaage and coworkers.11,12
In our opinion, the first advantage of using the LAR approach is the extreme simplicity and rapidity of cardiopulmonary bypass setup that eliminates the technical complexity of the minithoracotomy approach. The use of bicaval cannulation remains mandatory when using a paraseptal approach because the significant retraction of the right atrium could impair the venous return and the inadvertent incision of the right atrium could cause a venous air lock. In our series, the use of a single cannula into the right atrium provided an adequate venous drainage and significantly improved access to the operating field when through a ministernotomy. In addition, deairing maneuvers are facilitated with the incision of the LAR because most of the air into the left atrium tends to accumulate at the level of the roof.
The exposure of the mitral valve and especially of the anterior leaflet through the LAR was, in our experience, excellent, as confirmed by the high percentage of bileaflet mitral valve repairs and the low rate of repair failure during the follow-up. The access and visualization of the subvalvular apparatus were also very good, with the LAR approach allowing to perform complex repair on the chordae and papillary muscles in patients with ischemic mitral regurgitation. These benefits were particularly enhanced in the presence of small left atria when the visualization of the mitral valve through paraseptal approach becomes very difficult.
The LAR approach is also particularly valuable for training purposes, allowing the assistant to easily visualize every single step of the mitral repair during supervised training cases.
In this series, the percentage of combined procedures was 36.5% for the entire cohort and 32.2% for the ministernotomy subgroup. Most of these procedures involved the aortic valve, the aortic root, and the ascending aorta, confirming the suitability of this approach for mitroaortic procedures.
The two main caveats of the LAR approach are the risk of laceration of the LAR and the injury of the sinoatrial node artery. In this series, a tear of the LAR requiring a repair with pericardial patch was reported in 10 patients. Although thickness of the LAR is variable and lacerations are unpredictable we believe that a particularly careful retraction of the LAR edges with a Ross retractor by the assistant especially in old patients can prevent most of the time this complication.
The second possible complication of this procedure is the damage of the sinoatrial node artery causing the dysfunction of the sinoatrial node.
The sinoatrial node artery most frequently arises from the proximal segment of the right coronary artery and reaches the sinoatrial node running close to the superior junction between the left and the right atrium.13 Few authors have evaluated the effects of the systematic sacrifice of the sinoatrial node artery during a full vertical approach to the mitral valve without identifying any significant correlation with the onset sinoatrial node dysfunction and postoperative arrhythmias.14–16 In our experience with the LAR incision, preserving at least 5 mm of left atrial tissue from the interatrial septum can prevent the accidental transection of this artery and the overall incidence of permanent pacemaker insertion was 3.1%.
At our institution, this approach has been extensively used for every patient undergoing mitral surgery, except patients undergoing concomitant tricuspid surgery or specifically requesting a right thoracotomy for aesthetical reasons. Nevertheless, in redo procedures, where the dissection necessary to identify the LAR can lead to the damage of the right pulmonary artery, the superior vena cava, or the aortic root, this approach should be performed only by surgeons already comfortable with this technique.
The main limitations of this study are the retrospective collection of data and the lack of a control, possibly randomized group to benchmark the performance of this approach.
In conclusion, the LAR incision has been, in our experience, a safe and effective alternative to the conventional approaches, allowing to visualize clearly the mitral valve and to perform even complex repairs on the mitral leaflets and the subvalvular apparatus in settings traditionally considered challenging, such as minimally invasive incisions, small left atria, or combined procedures on the aortic valve and root.
1. Seeburger J, Borger M, Falk V, et al.. Minimal invasive mitral valve repair for mitral regurgitation: results of 1339 consecutive patients. Eur J Cardiothorac Surg
. 2008; 34: 760–765.
2. Seeburger J, Borger MA, Doll N, et al.. Comparison of outcomes of minimally invasive mitral valve surgery for posterior, anterior and bileaflet prolapse. Eur J Cardiothorac Surg
. 2009; 36: 532–538.
3. Vollroth M, Seeburger J, Garbade J, et al.. Minimally invasive mitral valve surgery is a very safe procedure with very low rates of conversion to full sternotom. Eur J Cardiothorac Surg.
2012; 42: e13–15.
4. Cheng DC, Martin J, Lal A, et al.. Minimally invasive versus conventional open mitral valve surgery: a meta-analysis and systematic review. Innovations
. 2011; 6: 84–103.
5. Meyer BW, Verska JJ, Lindesmith GG, et al.. Open repair of mitral valve lesions-the superior approach. Ann Thorac Surg
. 1965; 1: 453–457.
6. Saksena DS, Tucker BL, Lindesmith GG, et al.. The superior approach to the mitral valve: a review of clinical experience. Ann Thorac Surg
. 1971; 12: 146–153.
7. Sun L, Zheng J, Chang Q, et al.. Aortic root replacement by ministernotomy: technique and potential benefit. Ann Thorac Surg
. 2000; 70: 1958–1961.
8. Bonacchi M, Prifti E, Giunti G, et al.. Does ministernotomy improve postoperative outcome in aortic valve operation? A prospective randomized study. Ann Thorac Surg
. 2002; 73: 460–465.
9. Tabata M, Khalpey Z, Aranki SF, et al.. Minimal access surgery of ascending and proximal arch of the aorta: a 9-year experience. Ann Thorac Surg
. 2007; 84: 67–72.
10. Cohn LH, Adams DH, Couper GS, et al.. Minimally invasive cardiac valve surgery improves patient satisfaction while reducing costs of cardiac valve replacement and repair. Ann Surg
. 1997; 226: 421–426.
11. Ngaage DL, Nair UR. The left atrial roof incision: an asset for minimally invasive mitral valve surgery. Heart Surg Forum
. 2002; 5 (suppl 4): S421–S430.
12. Ngaage DL, Mulpur A, Nair RU. Mitral valve surgery using a modified superior left atrial approach. Surg Technol Int
. 2000; 9: 224–226.
13. Ballesteros LE, Ramirez LM, Quintero ID. Right coronary artery anatomy: anatomical and morphometric analysis. Rev Bras Cir Cardiovasc
. 2011; 26: 230–237.
14. Shin H, Yozu R, Higashi S, Kawada S. Sinus node function after mitral valve surgery using the superior septal approach. Ann Thorac Surg
. 2001; 71: 587–590.
15. Misawa Y, Fuse K, Kawahito K, Saito T, Konishi H. Conduction disturbances after superior septal approach for mitral valve repair. Ann Thorac Surg
. 1999; 68: 1262–1264.
16. Gaudino M, Nasso G, Minati A, et al.. Early and late arrhythmias in patients in preoperative sinus rhythm submitted to mitral valve surgery through the superior septal approach. Ann Thorac Surg
. 2003; 75: 1181–1184.
This large case series from Dr. Esposito and his colleagues in Bergamo, Italy, reports impressive results in over 500 patients in whom the mitral valve was exposed using a left atrial roof incision. This incision was first described in 1965 but is not commonly employed in many centers. This report shows that this approach can be used to effectively perform mitral valve surgery with low mortality and excellent results. Their repair rate in patients with mitral regurgitation was 87%, and 97% were free from mitral regurgitation >2+ at 4 years. While the majority of patients were done through a sternotomy, 189 were performed in a mini-sternotomy. The advantages of this approach include the excellent exposure and the possibility of using a single venous cannula. Moreover, it is attractive for combined mitral and aortic procedures through a median-sternotomy. The disadvantages are the potential for tearing of the left atriotomy and the possibility for injuring the sino-atrial node. In this series, a tear was seen in 10 patients and was repaired with a pericardial patch. The overall incidence of permanent pacemaker insertion was an acceptable 3%. The main limitation with this study was the lack of a control group to benchmark their performance. However, this nicely written report demonstrates the feasibility and potential advantages of the left atrial roof incision for mitral valve surgery.