Minimally Invasive Approach for Complex Cardiac Surgery Procedures

Minimally Invasive Approach for Complex Cardiac Surgery Procedures Pasquale Totaro, MD, Simone Carlini, MD, Matteo Pozzi, MD, Francesco Pagani, MD, Giuseppe Zattera, MD, Andrea Maria D Armini, MD, and Mario Vigano, MD Division of Cardiac Surgery, Istituto Di Ricovero e Cura a Carattere Scientifico University Hospital Foundation San Matteo, Pavia, Italy Background. A minimally invasive approach through an upper ministernotomy (UMS) has been used in our Division since 1997. On the basis of favorable outcome we have gradually extended this approach from isolated aortic valve replacement (AVR) to more complex cardiac surgery procedures and it is currently our first choice for a variety of procedures. Here we report our 11 years experience. Methods. From 1997 to December 2007, 1,126 procedures were performed at our department, using UMS. Isolated procedures on the aortic valve were performed in 695 patients (61%). Isolated procedures on the aortic valve as redo operation were performed in 77 patients (7%). Complex cardiac surgery procedures (including double valve replacement-repair, ascending aorta-aortic arch replacement, aortic root replacement, aortic dissection, AVR combined with coronary surgery, and complex redo procedures) were performed in 354 patients (32%). Early postoperative outcome was evaluated considering three different groups according to the surgical procedure (first time AVR, redo AVR, and complex procedure). Results. Overall conversion to full sternotomy was required in 16 patients (1.4%) with no significant differences between isolated AVR (9 patients, 1.3%) and complex or redo procedures (1 patient [1.2%] and 6 patients [1.6%], respectively). Forty-seven patients died in hospital (cumulative in-hospital mortality of 4.1 %). Mortality according to the procedure was 6.7, 3.8, and 2.8% for complex, redo AVR, or isolated AVR procedures, respectively, with a significant difference only for the complex procedures. Similarly, early postoperative outcome in terms of incidence of prolonged mechanical ventilation and ICU stay was significantly different only in the complex procedure group. Incidence of surgical revision (5.1, 2.9, and 2.7% for complex, redo, or isolated AVR procedures, respectively) showed no statistically significant differences regardless the type of procedures. Conclusions. Our experience clearly shows that a minimally invasive approach through upper ministernotomy is feasible and safe not only for isolated AVR but that it can also be utilized for a variety of complex surgical procedures. Minimizing surgical access may be helpful in patients undergoing complex surgical procedures, especially redo procedures, without compromising the surgical result. (Ann Thorac Surg 2009;88:462 7) 2009 by The Society of Thoracic Surgeons In order to avoid full median sternotomy during valve surgery procedures, minimally invasive approaches were introduced in clinical practice in the late 1990s [1 4]. Among these approaches, the so-called ministernotomy has been clearly correlated to relevant clinical advantages [5 9] but it has never gained complete worldwide recognition as many authors continue to consider such an approach only for its cosmetic advantages, and therefore just as a minimal incision instead of a minimally invasive procedure. Furthermore, although favorable preliminary results have been reported using ministernotomy also for redo [10] and for complex procedures including aortic root replacement [11, 12], experiences have been limited to few centers. Interest in ministernotomy as a first choice approach for a variety of cardiac Accepted for publication April 16, 2009. Presented at the Fifty-fifth Annual Meeting of the Southern Thoracic Surgical Association, Austin, TX, Nov 5 8, 2008. Address corresponce to Dr Totaro, Via San Vincenzo, 38, Milan, 20123, Italy; e-mail: ptotaro@yahoo.com. surgery procedures has been proposed by Doty and colleagues since 1998 [13] and has recently been reviewed by Bakir and colleagues [14, 15]. In our Department we introduced the upper ministernotomy (UMS) approach in 1997 for isolated first-time aortic valve repair (AVR) and, on the basis of favorable outcomes, we have gradually extended this approach to more complex cardiac surgery procedures. Since 1999 it has become our first choice for a variety of procedures apart from coronary artery bypass grafting (CABG) and heart transplant. Here we report our 11 year single center experience. Material and Methods From February 1997 to December 2007 we used the UMS approach in 1,126 patients undergoing a variety of surgical procedures (type of procedures and patient population characteristics are summarized in Tables 1 and 2). All patients signed the informed consent form and the study was approved by the IRCCS Foundation San Matteo Review Board and Ethics Committee. Since 1999 2009 by The Society of Thoracic Surgeons 0003-4975/09/$36.00 Published by Elsevier Inc doi:10.1016/j.athoracsur.2009.04.060

Ann Thorac Surg TOTARO ET AL 2009;88:462 7 COMPLEX CARDIAC SURGERY PROCEDURES THROUGH UMS 463 Table 1. Cardiac Surgery Procedures Performed Through UMS Approach Upper Ministernotomy (n 1,126) No. (%) First-time surgery (n 1,001; 89%) AVR: 695 (61%) Ascending aorta procedures: Bentall ( / associate procedures) 67 (5.9%) Ascending aorta repair-replacement 78 (6.9%) AVR associate procedures: AVR MVS 50 (4.4%) AVR AAS 96 (8.5%) Other procedure 15 (1.3%) Redo surgery (n 125, 11%) AVR 77 (6.9%) Other procedures 48 (4.3%) Type of surgical procedures performed using upper ministernotomy approach. AVR aortic valve replacement; AAS ascending aorta surgery; MVS mitral valve surgery; UMS upper ministernotomy. (Fig 1), only a minority of procedures involving the aortic valve and (or) thoracic aorta have been performed through midline sternotomy, usually when the patient has had a previous procedure performed by full sternotomy or needed multiple coronary bypass graft. The conversion rate in midline sternotomy and early postoperative outcome were evaluated and analyzed according to the type of surgical procedure performed. Details of Surgical Technique Over the study period different techniques of ministernotomy were used. However, the majority of patients underwent a reverse T ministernotomy while a minority of patients received an inverted L ministernotomy. A ministernotomy was prolonged as far as the third or fourth intercostal spaces depending on the surgical procedure, the patient s characteristics, and the surgeon s confidence. According to the surgeon s preference either a conventional vertical saw or an oscillating saw was used in first time procedures. An oscillating saw was routinely used in redo procedures. As far as the cardiopulmonary bypass (CPB) setup was concerned, all primary AVR procedures were performed using direct central aortic and right atrium cannulation, while peripheral aortic cannulation (through the femoral artery or axillary artery) was used for redo and complex procedures. In 40 patients (3.5%) undergoing redo complex procedures, venous return was also achieved by means of peripheral cannulation (femoral vein, jugular vein, or axillary vein) and CPB was started before approaching the redo sternotomy in order to decompress the heart. Surgical exposure and strategies are summarized in Table 3. In the case of redo surgery, once the sternum was reentered dissection from adhesions was limited to the structure directly involved in the procedure (ascending aorta, aortic arch) or necessary for venous cannula insertion (right atrium). Out of 125 redo procedures 24 patients (19%) had had a previous CABG. When a previous left internal mammary artery graft was patent it was dissected free only when it strongly adheres to the ascending aorta. Accurate transesophageal echocardiography (TEE) monitoring was used in order to minimize the risk of air embolism in all patients. Carbon dioxide insufflation was used only at the beginning of our experience and is not used any more as routine procedure. Similarly, the left ventricle venting through the right superior pulmonary vein was not used as routine procedure but only according to the surgeon preference. Myocardial protection was routinely obtained using an antegrade infusion of cold crystalloid cardioplegia (through aortic root selective cannulation of coronary ostium). However, selective cannulation of the coronary sinus for a retrograde infusion of cardioplegia is technically feasible and has been used occasionally in the present series. Cerebral protection during aortic arch aneurysm or acute aortic dissection repair was obtained by hypothermic circulatory arrest and either monolateral (through axillary artery cannula) or bilateral (using also selective left carotid artery cannulation) selective cerebral perfusion. External defibrillation pads were regularly used when needed but occasionally internal pediatric pads were also used. The sternum was usually closed using 3-4 sternal wires fixing the inferior part of the inverted T (or L) after the insertion of one or two small chest drains. Two epicardial pacing wires were positioned and tested before aortic declamping as it would be very difficult to do so when the right ventricle is no longer decompressed. Statistical Analysis Data are expressed in mean standard deviation. Comparison of hemodynamic and clinical parameters were made using analysis of variance for continuous variables and the 2 (or the Fisher exact test when appropriate) test for categoric variables. Results Surgical parameters are summarized in Table 3. Conversion to full sternotomy was required in 16 patients (cumulative 1.4%) with no statistical differences regarding the type of procedure (1.3, 1.2, and 1.7% for AVR, redo AVR, and Table 2. Preoperative Patient Characteristics Characteristic No. (%) Age (years) 62 13 Elderly ( 70 years) 338 (30%) Sex Male 687 (61%) Female 439 (39%) Previous cardiac procedures 125 (11%) Timing of surgery: Elective 1,063 (94%) Emergency 63 (6%) LVEF 0.53 0.12 LVEF left ventricular ejection fraction.

464 TOTARO ET AL Ann Thorac Surg COMPLEX CARDIAC SURGERY PROCEDURES THROUGH UMS 2009;88:462 7 Fig 1. Number of procedures involving aortic valve and (or) thoracic aorta performed at IRCCS San Matteo, Pavia, Italy using upper ministernotomy (UMS; black column) compared with conventional full sternotomy (FS; grey column). complex procedures, respectively). Suboptimal surgical exposition was the most frequent cause for conversion (11 patients, 69%) followed by accidental damage of a previous coronary graft and (or) native mammary artery (3 patients, 19%). Cumulative in-hospital mortality was 4.1% (47 patients). Complex procedure was a significant risk factor for higher postoperative mortality (24 patients, 6.7%) while no significant difference was shown comparing primary AVR to redo AVR (20 patients, 2.8% and 3 patients, 3.8%, respectively). Multiple organ failure (MOF) was the most frequent cause of death (11 patients, 23%) followed by low cardiac output syndrome (10 patients, 21%). In the subgroup of complex procedures, however, MOF (9 patients, 37%) was followed by gout, ischemia, and sepsis (5 patients each, 20%). Dividing our experience in three periods, overall mortality did not change significantly, although the rate of redo and complex procedure increased (Fig 2). Early postoperative outcome was similarly influenced by the type of procedure. Cumulative incidence of prolonged mechanical ventilation ( 24 hours) and intensive care unit (ICU) stay ( 3 days) were 9% and 10.5%, respectively, with no differences for patients receiving primary AVR or redo AVR, but with a significant increased risk in patients Table 3. Cumulative Surgical Parameters Parameter No. (%) Skin incision (cm) 7.2 1.4 Intercostal space: Third 867 (77%) Fourth 259 (23%) Arterial cannulation: Aortic 990 (88%) Femoral artery 136 (12%) Venous cannulation: Central 988 (87%) Peripheral 138 (13%) CPB time (minutes) 100 45 Ischemic time (minutes) 70 26 Conversion to full sternotomy 16 (1.4%) CPB cardiopulmonary bypass. receiving a complex procedure. Postoperative bleeding was significantly different in the complex procedure group but the incidence of surgical revision needed was not significantly different, regardless the type of procedure. Early postoperative outcomes are summarized in Table 4. Finally, we matched postoperative outcomes of 24 patients undergoing complex surgical procedures through UMS to a group of 24 patients undergoing complex procedures through full sternotomy and, although two groups were obviously not randomized, a clear trend in reduced postoperative bleeding and total in-hospital stay was shown in patients receiving UMS (Table 5). Comment In this paper we report our 11-years experience using a UMS approach for cardiac surgery procedures. Following preliminary experiences reported in the late 1990s [1 4], we started to use UMS in 1997 for isolated AVR. Based on the fact that patients were clearly satisfied with the procedures performed using a minimally invasive approach [7], we soon decided to extend the indication for this approach; which, since 1999, has become our first choice also for redo AVR and for a variety of complex procedures addressing the aortic valve and the thoracic aorta (including ascending aorta and arch). Clinical advantages of UMS in terms of reduced postoperative mechanical ventilation, bleeding, Fig 2. Cumulative mortality (black column) compared with the percentage of redo (dark grey area) and complex (light gray area) procedures during the study period.

Ann Thorac Surg TOTARO ET AL 2009;88:462 7 COMPLEX CARDIAC SURGERY PROCEDURES THROUGH UMS 465 Table 4. Early Postoperative Outcome Variable AVR (n 695) Redo AVR (n 77) Complex (n 354) p Value Mechanical ventilation: Mean (hours) 14 36 31 46 35 66 p 0.05 Prolonged ( 24 hours) 46 (6.6) 7 (8.6) 72 (20.4) a a p 0.05 ICU stay: Mean (days) 1.5 2.7 2.1 3.7 3.6 6.7 a a p 0.05 Prolonged ( 3 days) 47 (6.8) 8 (10.4) 88 (25) a a p 0.05 Postoperative bleeding (ml) 356 393 326 298 460 457 a a p 0.05 Surgical revision 18 (2.7) 2 (2.8) 18 (5.1) p 0.116 Total hospital stay (days) 10 7 11 6 12 8 b b p 0.05 a Complex vs AVR and Complex vs REDO AVR. b Complex vs AVR. AVR aortic valve replacement; ICU intensive care unit. and hospital stay, when compared to standard full sternotomy, have been extensively reported [8, 9]. However, such an approach has never gained complete recognition because the majority of surgeons continue to believe that the only advantages are cosmetic. A few centers, however, have continued to extend UMS application to more complex procedures. In 2000, Byrne and colleagues [10] reported the preliminary experience using UMS for redo cardiac surgery and for aortic root and complex ascending aortic surgery [11, 12]. Extended application of the ministernotomy approach (but using a lower ministernotomy) was also proposed by Doty and colleagues [13]. These studies, however, have not changed the criticism by some authors who consider such an approach only as part of minimal access surgery rather than minimally invasive surgery [16]. Despite such criticism, extensive use of the UMS approach has been carried out mainly by the groups of Mihaljevic and colleagues [17] and Bakir and colleagues [18]. In both these extended experiences, however, a comparable number of patients were still operated on full sternotomy. An interesting and peculiar aspect of our experience is that, since 1999, we have not had a real control group as UMS was used as a first choice approach and midline sternotomy (Fig 1) was conversely used only in selected cases (combined multiple CABG, procedure performed by junior resident, etc). Our results, however, seem to confirm data reported by Mihajlevic and colleagues [17] who, replying to previous criticism about the prolonged surgical time [16], clearly showed that, after appropriate training, AVR using UMS can be performed with reduced CPB and ischemic time compared with full sternotomy. In our hands, cumulative surgical time for isolated AVR decreased from 260 minutes (2001) to 190 minutes (2007) using UMS, and this approach is currently used by all staff surgeons. Bakir colleagues [18] also showed a reduced hospital stay and a significantly reduced blood loss when using UMS. We could not confirm such findings as the number of patients undergoing first time AVR using midline sternotomy was very low. However, in our experience using UMS, early postoperative outcomes (in terms of mortality, prolonged ICU stay and mechanical ventilation, postoperative bleeding, and total hospital stay) after redo AVR were comparable with those after first time AVR. These findings, therefore, compare favorably with previously reported experiences in redo AVR [19-21]. As far as UMS in redo AVR is concerned, we could therefore conclude that it is not only feasible and safe but that it allows outcomes comparable to first time surgery and should be strongly encouraged for redo AVR. The UMS application in complex procedures deserves a further and accurate discussion. Our low conversion rate to full sternotomy during such procedures seems to confirm technical feasibility of such an approach. Furthermore, although the patient population is quite inhomogeneous, our experience also seems to confirm the safety of this approach in such high-risk procedures. Our results in terms of mortality compare favorably with previous experiences using UMS for complex procedures that did not include patients with complex aortic surgery [11, 14]. The safety of the UMS approach also seems to be confirmed by the comparison of our results with those achieved in complex procedures on the ascending aorta [22] and on redo aortic root replacement [23] by an experienced team using median full sternotomy. Our limited matched comparison between two homogeneous (but not randomized) groups of patients undergoing complex procedures using midline sternotomy or UMS shows, furthermore, an interesting trend toward a reduced postoperative morbidity using UMS also in complex procedures. In conclusion, even if a clear advantage of using UMS in complex procedures is difficult to demonstrate without a prospective randomized trial, we would like to share the optimistic view of previous authors [24, 25] in considering the minimally invasive approach as an expanding field with potential true benefit, especially in high-risk patients. In our minds, a further extension of UMS to more complex procedures than isolated AVR should therefore be encouraged. Our study carries some clear limitations. First, although all data were prospectively collected in a general database, it is a retrospective study. Second, patients included in the complex procedures group are quite inhomogeneous including either first time or redo procedures and scheduled or emergency (such as aortic dissection) procedures. Third, we do not have a proper control group because, since 1999, minimally invasive approaches (either upper ministernotomy or minithoracotomy) have been the approach of choice for all non-cabg and nontransplant procedures performed at our

466 TOTARO ET AL Ann Thorac Surg COMPLEX CARDIAC SURGERY PROCEDURES THROUGH UMS 2009;88:462 7 Table 5. Early Postoperative Outcome UMS Versus Midline Sternotomy (for Complex Surgical Procedures) Group Division. We enclosed, however, a small matched comparison of postoperative outcomes for complex procedures performed through midline sternotomy or UMS. The authors thank Roger Lakelin for his precious support. References UMS (n 24) Midline Sternotomy (n 24) p Value Preoperative characteristics: Age 69 14 68 13 p 0.799 Elderly ( 70 years) 10 (42) 9 (37) p 1 Sex: Male 11 (45) 12 (50) p 1 Female 13 (54) 12 (50) Procedure performed: Aortic dissection 5 (20) 3 (13) Redo AVR 8 (33) 6 (22) Ascending aorta 8 (33) 9 (39) p 0.72 aneurysm AVR MVR 1 (4) 3 (13) AVR MVR (redo post- 2 (8) 3 (13) AVR) Redo 10 (42) 9 (37) p 1 Atrial fibrillation 9 (37) 9 (37) p 1 LVEF 57 6 59 3 p 0.151 Timing of surgery: Elective 18 (75) 21 (87) p 0.46 Emergency 6 (25) 3 (13) Postoperative outcomes Cumulative mortality 2 (8.3) 3 (12.5) p 1 Postoperative bleeding 421 301 737 532 p 0.001 Revision for bleeding 0 1 (4) p 1 Patients transfused 14 (58) 16 (66) p 0.12 Units of RBC transfused 2.5 2.1 4.1 3.0 p 0.03 Mechanical ventilation Mean 23 20 15 10 p 0.08 Prolonged 4 (16) 3 (12) p 0.7 ICU stay: Mean 3.3 3.2 2.5 3.1 p 0.38 Prolonged 6 (25) 6 (25) p 1 Total hospital stay 11 8 16 9 p 0.04 AVR aortic valve replacement; ICU intensive care unit; LVEF left ventricular ejection fraction; MVR mitral valve replacement; RBC red blood cells; UMS upper ministernotomy. 1. Gundry SR, Shattuck OH, Razzouk AJ, del Rio MJ, Sardari FF, Bailey LL. Facile minimally invasive cardiac surgery via ministernotomy. Ann Thorac Surg 1998;65:1100 4. 2. Cosgrove DM III, Sabik JF. Minimally invasive approach for aortic valve operations. Ann Thorac Surg 1996;62:596 7. 3. Svensson LG. Minimal-access J or j sternotomy for valvular, aortic, and coronary operations or reoperations. Ann Thorac Surg 1997;64:1501 3. 4. Izzat MB, Yim AP, El-Zufari MH, Khaw KS. Upper T ministernotomy for aortic valve operations. Chest 1998;114:291 4. 5. Bonacchi M, Prifti E, Giunti G, Frati G, Sani G. Does ministernotomy improve postoperative outcome in aortic valve operation? A prospective randomized study. Ann Thorac Surg 2002;73:460 6. 6. Masiello P, Coscioni E, Panza A, Triumbari F, Preziosi G, Di Benedetto G. Surgical results of aortic valve replacement via partial upper sternotomy: comparison with median sternotomy. Cardiovasc Surg 2002;10:333 8. 7. 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 8. 8. Liu J, Sidiropoulos A, Konertz W. Minimally invasive aortic valve replacement (AVR) compared to standard AVR. Eur J Cardiothorac Surg 1999;16(suppl 2):S80 3. 9. Mächler HE, Bergmann P, Anelli-Monti M, et al. Minimally invasive versus conventional aortic valve operations: a prospective study in 120 patients. Ann Thorac Surg 1999;67:1001 5. 10. Byrne JG, Karavas AN, Adams DH, et al. Partial upper re-sternotomy for aortic valve replacement or rereplacement after previous cardiac surgery. Eur J Cardiothorac Surg 2000;18:282 6. 11. Byrne JG, Adams DH, Couper GS, Rizzo RJ, Cohn LH, Aranki SF. Minimally-invasive aortic root replacement. Heart Surg Forum 1999;2:326 9. 12. Byrne JG, Karavas AN, Cohn LH, Adams DH. Minimal access aortic root, valve, and complex ascending aortic surgery. Curr Cardiol Rep 2000;2:549 57. 13. Doty DB, DiRusso GB, Doty JR. Full-spectrum cardiac surgery through a minimal incision: mini-sternotomy (lower half) technique. Ann Thorac Surg 1998;65:573 7. 14. Bakir I, Casselman F, De Geest R, et al. Minimally invasive aortic root replacement: a bridge too far? J Cardiovasc Surg (Torino) 2007;48:85 91. 15. Bakir I, Casselman FP, De Geest R, et al. Should minimally invasive aortic valve replacement be restricted to primary interventions? Thorac Cardiovasc Surg 2007;55:304 9. 16. Gulbins H, Pritisanac A, Hannekum A. Minimally invasive heart valve surgery: already established in clinical routine? Expert Rev Cardiovasc Ther 2004;2:837 43. 17. Mihaljevic T, Cohn LH, Unic D, Aranki SF, Couper GS, Byrne JG. One thousand minimally invasive valve operations: early and late results. Ann Surg 2004;240:529 34. 18. Bakir I, Casselman FP, Wellens F, et al. Minimally invasive versus standard approach aortic valve replacement: a study in 506 patients. Ann Thorac Surg 2006;81:1599 604. 19. Kumar P, Athanasiou T, Ali A, et al. Re-do aortic valve replacement: does a previous homograft influence the operative outcome? J Heart Valve Dis 2004;13:904 13. 20. Potter DD, Sundt TM 3rd, Zehr KJ, et al. Operative risk of reoperative aortic valve replacement. J Thorac Cardiovasc Surg 2005;129:94 103. 21. Davierwala PM, Borger MA, David TE, Rao V, Maganti M, Yau TM. Reoperation is not an independent predictor of mortality during aortic valve surgery. J Thorac Cardiovasc Surg 2006;131:329 35. 22. Krähenbühl ES, Immer FF, Stalder M, et al. Technical advances improved outcome in patients undergoing surgery of the ascending aorta and/or aortic arch: ten years experience. Eur J Cardiothorac Surg 2008;34:595 9. 23. David TE, Feindel CM, Ivanov J, Armstrong S. Aortic root replacement in patients with previous heart surgery. J Card Surg 2004;19:325 8. 24. Soltesz EG, Cohn LW. Minimally invasive valve surgery. Cardiol Rev 2007;15:109 15. 25. Caffarelli AD, Robbins RC. Will minimally invasive valve replacement ever really be important? Curr Opin Cardiol 2004;19:123 7.

Ann Thorac Surg TOTARO ET AL 2009;88:462 7 COMPLEX CARDIAC SURGERY PROCEDURES THROUGH UMS 467 DISCUSSION DR MICHAEL MACK (Dallas, TX): Dr Totaro, congratulations on an excellent presentation and on the work that your team is performing in Pavia; also I thank you for forwarding me the manuscript prior to the meeting. I would like to express our appreciation for choosing our meeting in which to present your work and hope you have experienced our Southern hospitality and hope it has been as enjoyable as Northern Italian hospitality has been to Americans. I have four questions for you. You reported the results on 1,126 patients. What is the denominator? It wasn t clear to me from the manuscript. In other words, what proportion of all patients undergoing aortic valve procedures, simple or complex, were done minimally invasively? How were they chosen; was this surgeon preference? Did all surgeons perform limited sternotomy and were there specific patients that were avoided? DR TOTARO: Thank you. First of all, it is an honor for me to have as a primary discussant the President-Elect of the Society. In our department there is an extensive minimally invasive program now, and as a first choice option, we go for a ministernotomy, upper ministernotomy in all procedures but coronary artery bypass graft and transplant, and if this is mitral surgery, we go for a right minithoracotomy. So the percentage of minimally invasive in all our activities is around 50%. For this specific group of patients (ie, aortic surgery, AVR, redo AVR) the percentage of procedures performed in ministernotomy is even higher. There was just a minority of patients operated on full sternotomy, even for AVR or for redo AVR, but these were only the cases performed by residents or with a very selected indication, such as combined multiple coronary artery bypass graft or combined redo mitral procedure. DR MACK: You said early in your series that you performed upper ministernotomy by a reverse L and then switched to an inverted T. Did you find that just the right upper hemisternotomy was insufficient and that you obtained better exposure by the T incision? DR TOTARO: Well I have to say that of course we started with the reverse L only for isolated AVR procedure, then when we started to perform more complex procedures we found that the reverse T could offer a better exposure. But nowadays I think that it depends mainly upon surgeon choice. DR MACK: You mention in the manuscript that a left ventricular sump was used occasionally or variably. What is your method of deairing the heart; was CO 2 used across the operative field? Why were sumps only used occasionally and has inadequate deairing and stroke been an issue at all? DR TOTARO: Well, I have to, first of all, stress that I joined this team only in the last two years. So I am not 100% aware of the initial experience. But what I can say is that left ventricle deairing using a left ventricle vent through the right superior pulmonary vein has been used in a minority of cases. What we do use routinely is the transesophageal echo, to be sure before we take the clamp off that there is no air. CO 2 has been used occasionally in the first three or four years and it is not used routinely anymore. We have never had, anyway, a major problem with air embolism. DR MACK: My last question is the one that I struggle with most. We perform these procedures the same way as your team. You stated in your presentation that there are clear clinical advantages to this. Surgeons come to meetings to learn what they should be doing differently than what they have been doing before, and although you have proved that this technique is feasible in complex patients, I am not sure it is clear what those clinical advantages are, especially since there was no control and there was no randomization. Is it cosmesis or less pain or less respiratory difficulty? What would you say to a surgeon in the audience as to why they should go home and adopt this technique? What are those clear clinical advantages? DR TOTARO: Thank you very much for this question, because this is the most important key point. And I would like to stress, first of all, that when I joined this group I was as suspicious as many of the surgeons here on using this approach for a complex procedure or even for a redo procedure. But I have changed my mind very quickly looking at the advantage of such approach in peculiar situations. Anyway, I think that we have passed the step of thinking about this operation only for cosmetic reasons. For redo AVR, I think that there are clear advantages, which have been shown also by previous papers, in terms of bleeding, length of stay, and also on early postoperative outcome. So for redo AVR I just invite all of you to start using it and I bet that you will change your mind in six months, as it happened to me. In complex procedures this is surely a bit more difficult to demonstrate because we need a randomized study, which is not easy to perform. But I am sure that with the more complex patients we are going to operate on, the more we reduce the invasiveness of surgery, the more we could get a clinical advantage. Of course, once more this is difficult to prove. For complex procedures, furthermore, I do believe that better results could depend upon an appropriate learning curve and on the ability of the surgeon. Once more I would like to thank the society for the privilege of presenting our experience. DR R. BRENT NEW (Austin, TX): Very nice presentation, Dr Totaro. Could you describe a little bit your techniques for myocardial protection with this approach and how it may differ with whether it is an aortic valve procedure or a mitral valve, that sort of thing? DR TOTARO: Thank you very much for your interesting question. The policy of our division is not to use retrograde cardioplegia, so regardless the type of approach we use we go for antegrade crystalloid cold cardioplegia. However, I tried in my previous experience to use retrograde cardioplegia, even with a minimal approach, and I have to say that it is technically feasible. It is, of course, more demanding but it is feasible. Furthermore, one thing I want to stress is that for the complex procedures involving aortic arch we could use all type of cerebral protection. So, even if it is a very limited approach, we used deep hypothermia and selected antegrade cerebral perfusion.