Portal de Periódicos da CAPES

Panel Discussion: Session I—Ascending Aorta

2007; Elsevier BV; Volume: 83; Issue: 2 Linguagem: Inglês

10.1016/j.athoracsur.2006.11.088

1552-6259

D. Craig Miller, Jean Bachet, Duke E. Cameron, Oved Cohen, Tirone E. David, Gébrine El Khoury, Armin W. Erasmi, Matthias Karck, Hans‐Joachim Schäfers, Fabrizio Settepani,

Otitis Media and Relapsing Polychondritis

DR D. CRAIG MILLER (Stanford, CA): I would prefer you to direct questions to a specific individual. If you want different opinions, direct it to all 10 of us, and we will each probably say a different thing. Let’s not be shy: you heard a lot of provocative stuff this morning, and there is not a dearth of controversial areas to be discussed. DR THORALF SUNDT (Rochester, MN): Could I ask Tirone a question about the application of the technique of the valve-sparing root? It seems we are seeing a lot of patients with giant cell arteritis and root aneurysms, and it raises the issue of in which subpopulations valve-sparing is appropriate. I have been concerned that if a patient has an aneurysm on the basis of an inflammatory process, that the leaflets themselves will develop retraction on the basis of that process, but I don’t have any data on that. Do you have any experience in giant cell arteritis or other inflammatory conditions? DR TIRONE E. DAVID (Toronto, Ontario, Canada): We do, but it is limited. I think, in eight of the 120, the pathological diagnosis was some sort of arteritis, an inflammatory aneurysm. The cusps are spared. Actually, even the sinuses in these patients are often spared. So if the cusps are reasonable, we do valve-sparing operations in these cases. But eight cases doesn’t make much of an experience. DR MILLER: Did you have a specific disease, Tirone, polymyalgia rheumatica, gonococcal arthritis, or was there no diagnosis? DR DAVID: Not really. I am afraid of other diseases, however, like lupus. DR MILLER: I think that is a very wise word of caution, especially with the Ross II procedure: avoid the polymyalgia rheumaticas; avoid lupus. DR DAVID: I have seen a couple with rheumatic arthritis with aortic insufficiency and the cusps appear to be normal, but I was reluctant to preserve them. Connective tissue disorders, as a rule, except for Marfan’s, I am reluctant to repair. Ehlers-Danlos I have not repaired; I have replaced. DR MILLER: Which type? There are 10 types. DR FRANCISCO NISTAL (Santander, Spain): This is a question to Dr Joachim Schäfers concerning the alternative modality of treatment for patients with bicuspid aortic valves. Aside from the dimensions of the root or the features of the leaflets, oftentimes one encounters in these patients sinuses of Valsalva which have very thin and delicate walls. Do you take this particularity into account, Joachim? DR HANS-JOACHIM SCHÄFERS (Homburg/Saar, Germany): Yes, sometimes. We all have subjective judgment; we call it surgical judgment in some details of our operations. What I have tried is to be as disciplined as possible. So actually only maybe in one or two patients in the presence of reasonable root dimensions have I gone to replacement of the root just because I was uncertain about wall quality. In effect, while I am not 100% certain that the cutoffs I use are correct, my feeling is that if you go by root dimensions, you are relatively safe and less dependent on a very subjective gut feeling. DR MILLER: Let’s ask a question of everybody, because I noticed in the Hannover experience you had several patients over 70. Tirone, I doubt you are doing valve-sparing root replacements in patients over 70; Duke Cameron certainly isn’t, and we certainly aren’t. What are we doing in these old people? Where do we cut it off and do something simple and dependable? DR DAVID: I don’t do reimplantation of the aortic valve in patients over 70, but the 70-year-olds, more often than not, do not have aortic root aneurysms. They have an ascending aortic aneurysm that causes secondary dilatation of the sinotubular junction. In those patients, the cusps are often normal. In others, they require adjusting the sinotubular junction and replacing the arch and whatever else. We just reviewed our experience with this group again. It is not a bad operation. It doesn’t matter if it is a bicuspid or a tricuspid valve. Actually we have no reoperations in that group, and one patient of 109 has moderate AI. That is a very good operation for the 70-, 80- or 90-year-old who comes in primarily because of an ascending aortic arch aneurysm with AI. DR MILLER: But that is a different disease, aneurysm of the tubular segment of the ascending aorta with relatively normal sinuses. And as you have taught us, we can just remodel the sinotubular junction. DR DAVID: Absolutely: those cases do well. If you have a 75-year-old with a root aneurysm, I will do a Bentall and put a porcine valve in them. DR MILLER: And as Dr Schäfers reiterated today, shorten the free margins because you are going to create a prolapse. How about in Hannover, Matthias, are you still doing David operations in people over 70? DR MATTHIAS KARCK (Hannover, Germany): Well, if the ventricle is good, if the patient is going for a swim every other day and there is AI of perhaps grade 2. I probably wouldn’t correct AI grade 1 and an ascending aortic aneurysm in a 75-year-old patient. But if there is somewhat more than AI grade 1, and an aneurysm of 5 cm, and the general condition is good and the ventricle is good, we would try for a David reconstruction in these patients. DR MILLER: Would you agree with what Tirone just said, that isolated sinus aneurysms, root aneurysms, are rare in the elderly? DR KARCK: Definitely, because they probably would have ruptured years before they come to this age. DR NISTAL: I have another question for Professor Jean Bachet. When you recommend replacing the arch for Marfan patients with aortic dissection, you are speaking about hemiarch or total arch? DR JEAN BACHET (Paris, France): A difficult question. I suppose I mean a total arch, because if we stay in the middle of the river, then we will have exactly the same kind of reoperation in the future as I described here in this experience. So if we go to replace the aortic arch in Marfan patients during acute type A dissection initial surgery, we should replace the total arch as they do in some places in Japan. DR MILLER: A very good question. Let’s ask Dr Kazui to come to the microphone. Dr Kazui has been one of the leaders in concomitant total arch replacement at the time of acute type A aortic dissection. I think 30% or 35% of your acute type A patients underwent concomitant total arch replacement with your little short elephant trunk. They were selected patients, but your reoperation rate thereafter, as I recall, was not any lower than your historical reoperation rate. So the corollary to Dr Nistal’s question and to Jean Bachet is what are we accomplishing? DR TERUHISA KAZUI (Hamamatsu, Japan): Well, I highly recommend replacing the whole aortic arch in the case of acute type A dissection, particularly in the Marfan patient, using our separate graft technique with the aid of antegrade cerebral perfusion. The mortality is now below 5%, even in emergency cases, and a previous operation on the ascending aorta or the arch is a risk factor for reoperation on the aortic arch. I think we should not leave the pathological portion of the aortic arch when we do the initial operation. DR MILLER: Let me reiterate that and make sure I got it straight. For all Marfan patients with acute type A aortic dissection, you do a complete total arch multibranch graft? DR KAZUI: Yes, with the elephant trunk technique. DR MILLER: But we all must remember that doing so on a 40- or 50-kg Japanese subject is different than doing it on the 110- or 130-kg gentleman or lady that we see. DR KAZUI: The average patient weight in my Marfan patient is about 60 kg. DR MILLER: Is Professor Loisance here? He has shown from Paris that the reoperation rate over time is actually going up, not going down, as we get more aggressive at the initial acute type A operation. DR DAVID: Can I make a comment? I think it is wonderful that Dr Kazui can do what he does with a mortality of 5%: take an acute type A dissection in a Marfan patient, replace the arch, reimplant both of the vessels, do a valve-sparing operation, and whatever else has to be done at the root, and only 5% die. That is remarkable. I can’t do that. I don’t do this operation. My partners cannot do an aortic arch resection in the dissection patients they treat: 90% would be dead in 2 days. My service has a 30% mortality for acute type A dissection. I don’t know about the referrals in different places, but most of us should be able to do an operation in acute dissection to save the patient’s life. Replacing the entire aortic arch and putting three branches in a brain: this, my friend, is a lot of surgery. And unless you do it every day, I do not believe that this is the standard treatment, and will never be in North America, because most heart surgeons in the United States do not do 10 dissections a year. DR MILLER: Let’s ask Dr Erasmi, because in Luebeck, one third of your 164 valve-sparing root replacements were done for acute type A aortic dissection. I think that’s the highest frequency of anybody. You only had 17% to 19% Marfan patients, but how many of those are you also doing the total arch on? That is a remarkable experience with a very low mortality rate. DR ARMIN W. ERASMI (Luebeck, Germany): In Marfan’s we try to replace the arch, but in non-Marfan patients, we just do a partial arch or open distal anastomosis and reconstruct the layers inside and outside with Teflon felt. DR MILLER: So would you or would you not do a total arch multibranch graft in a young Marfan with acute dissection? DR ERASMI: We try to do it, but we are several surgeons and it depends on the surgeon’s preference. DR MILLER: You told us today that you have 13% operative mortality rate for valve-sparing aortic root repairs in the face of acute type A aortic dissection. So, Tirone, it can be done relatively safely elsewhere. I congratulate you. DR PETER SKARSGARD (Vancouver, British Columbia, Canada): A question perhaps for Drs. Cameron and David. For aortic root aneurysms, sometimes by the time we have the patient in the operating room, the cusp pathology is actually quite bad. So my question is what types and degrees of cusp pathology do you think preclude repair? DR DAVID: If one cusp is bad—and by bad I mean it is overly stretched and/or there is a large fenestration—in other words, when prolapse is due to a tear and detachment at the commissural area, I will repair the cusp. If two are bad, the patient gets a root replacement, a Bentall. Have I done many with two? No, never. It seems too much to me to take two of the three cups that are totally flailed because they have no commissural area. With other problems, we repair the cusps. If the cusps are intact, I don’t care how bad the degree of prolapse is: you can shorten enough the nodule of Aranti, and reinforce it with Gore-Tex. And, as Craig said, it is not that difficult to weave a 7-0 or 6-0 Gore-Tex along the free margin. Unlike Dr El Khoury, who weaves over the free margin, I do it beneath the free margin: I go in and out a millimeter or so below the free margin. We had just one patient that died, with a ruptured type B dissection, who had Gore-Tex in the free margin. You don’t see the Gore-Tex 5 years later. It becomes a fibrous sheath along where the fenestration was and works very well. So the answer is if one cusp is bad, we repair, but with more than one, we replace the valve. DR MILLER: And that is on the basis of having 30% Marfan’s in your experience? DR DAVID: 38% Marfan’s. DR MILLER: Duke, you probably have 80 or 90% Marfan’s. What is your feeling? DR DUKE E. CAMERON (Baltimore, MD): As I alluded to in my talk, we have a very low threshold for moving to a Bentall if there really is any significant leaflet pathology. If there are significant fenestrations, we will go ahead and replace the valve. I think we are lucky that the Bentall is such a good operation. If that is your fallback, you are in great shape. We won’t work with leaflet asymmetries or marked fenestrations. And I think probably in our total of 120-some valve-sparing root and valve repairs, we have only done perhaps two or three leaflet repairs, and it has always been a fold in the mid portion of the leaflet, never anything along the free edge. With the Marfan leaflets, at least the ones that I operate on, it would be very difficult to run a 7-0 suture along the free edge: the leaflet is very thin. DR MILLER: But you cautioned us that your patients are very carefully selected; almost none have severe AR and I am sure zero have eccentric AR due to cusp prolapse. So our patient populations vary from center to center. DR CAMERON: That’s right. DR MILLER: Any other comments of wisdom from the panelists about what leaflets are too shredded, too far gone to try and save? DR GEBRINE EL KHOURY (Bruxelles, Belgium): I have to make one comment. I think we have to assess the quality of the leaflets just after the preparation of the root. If there is prolapse or fenestration, or the quality is not acceptable, we have to decide at this time if we should preserve or repair the valve. But if you go for repair, sometimes the three leaflets are found to be prolapsing after the completion of the procedure. And so, at that time again you have to decide whether to replace the valve or repair it. In my opinion, it is always possible to repair the three leaflets with Gore-Tex after the completion of the procedure. DR MILLER: In our experience, the most common reason to abort is good judgment. In old people and big people, we say we will look at it, but then we don’t follow through. The second reason to hesitate is aberrant coronary artery origin with an intramural coronary artery, usually the right sinus or the circumflex. It is better to know about that before you get there. A repair can be done, but the aberrant coronary complicates things. DR PHILIPPE DELEUZE (Le Chesnay, France): After having done several years of what I would call the Yacoub procedure, I was very impressed by Dr David, whom I visited, and then have been even more impressed by the use of the Valsalva tube. I know that some surgeons on this panel are using the Valsalva graft, but other surgeons don’t use it. I would like to know if it is only because it is European and a good idea, or do you have scientific reasons not to use it? DR MILLER: We have learned that it is not just used in Italy anymore. Fabrizio, do you have comments about the Vascutek prosthesis? DR FABRIZIO SETTEPANI (Rozzano, Italy): There are scientific reasons to use the Valsalva graft. Ruggero De Paulis did a lot of studies on the graft. He published two papers in the Annals, in 2001 and 2002, in which he made the comparison between healthy people and patients with Valsalva grafts and found that the distensibility at the site of the Valsalva sinus is really similar to the distensibility of the normal aortic root. On top of that, the valve opens and closes really smoothly in a good way. DR MILLER: In Luebeck, do you use the sinus graft, the Valsalva graft? DR ERASMI: We do both the David and the Yacoub procedure. DR MILLER: But what sort of graft do you use? Do you use the commercially made Vascutek? DR ERASMI: Yes. DR MILLER: On every case? DR ERASMI: Yes. DR SCHÄFERS: I find the Valsalva graft a little impractical because of the measuring of commissural height you have to do. So I have always used straight grafts. They allow me to work essentially from the valve base up, and extend the incisions in the graft to any length I find I need. So for reasons of practicality, I have not used it. DR BACHET: I use it, but in different patterns. I use it systematically now when I do a Bentall procedure. My experience with valve-sparing procedures in about 60 patients is mostly with the Yacoub technique, and when I do this remodeling procedure I do an annuloplasty with a small band of Teflon around the aortic annulus, and then I use a straight tube. But when you look at the CT scans during follow-up, you see that straight tubes used with the Yacoub procedure do exactly what the De Paulis valve prosthesis does: it remodels exactly the natural root. So I am not sure that for a valve-sparing procedure the sinus graft is very useful. DR CAMERON: We use it 100% of the time, and the attraction is just the simplicity. If this operation is going to be widely adopted, it has to be safe, and it has to be simple, and in our hands this has been a very good graft that you can adapt to a number of different situations. The height of the commissures, as you know, can be just below the sinotubular ridge, or just above it. The graft itself goes up and down a little bit. I think, frankly, a few millimeters often doesn’t make much difference. I was interested by the discussion this morning about the similarities with mitral valve repair and the same principles of getting good leaflet coaptation and making sure that the coaptation is all either below in the case of the mitral valve, or above the annulus in the case of the aortic valve. It is a bit similar to the arguments over ring sizes, in which people will wring their hands over whether they should put a 34- or a 32-mm graft. And then you see other units that use only a 28-mm graft for everybody, and they seem to get pretty good results. That is a long-winded way of saying that I still think that the graft can be adapted to a very broad range of patients. It is simpler to sew in than customizing a graft. The argument that a customizable graft can be used for giant roots doesn’t hold much attraction for me, because I am still ambivalent about whether we should be retaining the valve in those giant roots. DR EL KHOURY: I use the Valsalva 50% of the time, and in the other 50% I use the tube. The Valsalva is very, very helpful, but I have some principles. First, the commissures must or should reach the sinotubular junction or be even above the sinotubular junction. I never use the collar; I think it is not a good idea to have the collar. Because I think a good position between the prosthesis and the valve is important, I don’t hesitate to cut. I never respect the integrity of the Valsalva. I like to cut it in order to make the commissure at the level of the sinotubular junction or even higher. DR MILLER: So you cut the collar off; you like to notch it. Your technique looks like a Cochran operation from 1995 first described by Karen Kunzelman and Pat Cochran. I don’t understand why you are doing that. Tirone, do you use the Valsalva graft? Never? DR DAVID: Never. It has the wrong shape, the wrong symmetry. Everything is wrong about that graft. For those who don’t know anatomy, and I guess the designers don’t know anatomy, the aortic annulus evolves to a single horizontal plane. This is the aortic annulus, and you cannot put an aortic annulus in a hemispherical graft. Secondly, it is an illusion to think you can take a piece of Dacron graft, implant it in a patient, and expect that 6 months later it is going to be distensible. The fibrotic reaction against the collagen-impregnated graft is so intense that the stuff is as stiff as a piece of bone. DR MILLER: To turn to my cynical side, I think the reason it is so popular for valve-sparing, and also for regular mechanical composite valve grafts, is because the surgeons out there doing root replacements don’t know how to do root replacements and they don’t know how to mobilize the coronaries adequately. So diameter plus eight, giving you 4 millimeters on each side, may save lives, but if you mobilize the buttons enough, you don’t need it. DR JOSEPH E. BAVARIA (Philadelphia, PA): That was a great editorial comment, with which I disagree totally. This is for Drs Cameron, Miller, David, Karck, El Khoury, and my colleague, Settepani. Dr Lansman just told us we have 1100 people here in this room and they really want to know one thing about the valve-sparing operation and that is how do you size your grafts? And not a single person really said anything about it; they just said, well, you know, it is a little arty and you take a look at it and you put some stuff through your brain and it comes out that you pick a 28 or a 30 or a 32. I am challenging here. Let’s give everybody something hard, since we have gotten to the point where we have a real number that we come out with. How do you size these grafts? Is there really a way that we can size these grafts for the 1100 people in this room? DR MILLER: A good question. You have already heard several people say they don’t have a formula. It is in their head. I will go first. I still measure the Feindel-David formula, and then I discard it and ignore it, and then I go up to a 6, 8, 10, and bigger if it is a Marfan’s with a lot of bulk, because I hate crowding all that stuff, as I did in the first 25 Tirone David-I. With a big Marfan’s, a lot of tissue and a little tiny chimney, it isn’t any fun. And then I am going to make the annulus whatever the annulus needs to be. That is the beauty of having this flexibility. The sinotubular junction has to be the same as what you have made the annulus. So that is when you pick the little graft. Pretty simple, but not reproducible and not a formula. DR DAVID: I take the height and multiply it by two and that is the size of the graft to do the reimplantation. With remodeling, suspend the three commissures, and—to make sure the cusps coapt centrally—take one size up, because you can always reduce; you cannot make it larger if you take too small a size. DR MILLER: Height plus two? DR DAVID: For the reimplantation, the sinotubular junction diameter of the hypothetical triangle that allows the three cusps to coapt. DR MILLER: That is not far off from Dr Aybek in Frankfurt and Dr Gleason at Northwestern, two H plus two, but they get these 40, 42, 44. DR DAVID: I haven’t used larger than 34 yet. DR MILLER: Do you remember when Dr Griepp said years and years ago, when you first introduced the Feindel-David formula, that it is the height, and you average them twice, and multiply by 0.6, then divide by the square root of zero, and then add 26, because it always came out to be 26. Do you remember that? That is a pretty small graft. DR DAVID: It was an overstatement. The average is 32. DR MILLER: Thirty-two nowadays. I think earlier, and even in Hannover, it is probably a smaller size. Is there a formula or not, and what is your average size? DR EL KHOURY: After preparation of the aortic root, I move in the three commissures, make the valve competent, get a good configuration of the valve, size the insides of the sinotubular junction, and add 4 mm. DR CAMERON: I do exactly the same: transect it above the sinotubular ridge, look at the valve, and determine what sinotubular junction diameter provides best apposition of the leaflets. You have to add the 4 mm because the graft is sitting outside the root complex. DR BACHET: When I do a remodeling procedure, I measure the annulus with a Hager dilator, and I take a prosthesis of the same size. If I do a Tirone technique, I do exactly what Duke has said. DR SCHÄFERS: For remodeling, I take the size of the aortoventricular junction of the annulus minus one or two. For reimplantation, I take a 30-mm graft in men and 28 in women. DR ERASMI: For remodeling we lift up the commissures and measure the distance like Dr El Khoury, and for the reimplantation we orient on the annulus diameter, because we do the David-I. DR SETTEPANI: We use the same method as Dr El Khoury. We size the ideal circle that includes the three commissures when we lift them up, and then we add 5 mm to pick the graft. DR KARCK: We use the conventional valve sizer put on the commissures and put the valve sizer near the sinotubular junction and see if we get a nice coaptation area, and then we select the graft according to the sizer. DR MILLER: Matthias, has your graft size, on average, increased over the last 10 years? Have you looked at that? DR KARCK: It is still 28 mm in the non-Marfan male, and in the Marfan subcohort it is 2 mm more. This applies for male and female. DR MILLER: And at UCLA I guess you don’t have to pick a graft because you never replace the aorta, you just wrap it. DR OVED COHEN (Los Angeles, CA): We certainly replace, but when it is possible, we wrap it, and we believe that simpler is better. DR RANDALL B. GRIEPP (New York, NY): Can I make one comment? I would just like to remind the panelists and the room that if you do a CT scan a week after the operation, the average graft increases in size between 16% and 20%. So all of your grafts go up in size by a factor of .2 over the first week. DR MILLER: The UCLA wrap (and you made Francis Robicsek very, very proud) I am not convinced is a durable fix. I noticed in your echo example and in the angiographic example that the dilatation was only in the tubular segment of the ascending aorta. The sinuses were normal, suggesting strongly that those patients had a bicuspid valve. How many had a bicuspid valve? DR COHEN: About 20% of the patients. If it is possible, we correct the valve. Usually we correct the valve. DR MILLER: But how many had a bicuspid valve? You replaced two thirds of them. Most had bicuspid valves? DR COHEN: Yes. DR MILLER: So your operation might be okay where the dilatation is above the sinotubular junction? DR COHEN: Usually, yes. DR MILLER: But not okay for dilatation of the sinuses? DR COHEN: I showed in the presentation that there is a technique to reinforce from the outside the dilated sinuses, but if the dilatation of the aorta is above 6.5 cm, we usually replace it with a graft, and do not wrap it. DR MILLER: I was shocked to see your aortoplasty picture. I thought that went out 50 years ago: that is an obsolete operation. Now, you are wrapping it, so maybe it is going to work, but that seems rather silly and perhaps dangerous. DR NICHOLAS T. KOUCHOUKOS (St Louis, MO): I would like to go back to the issue of the tubular graft versus the sinus graft. Dr David’s series, which is one of the longest and the largest, has a not insignificant number of patients who had only a straight graft. Did you see any difference in the long-term outcomes between the straight graft and the T. David-I, T. David-II, T. David-III, T. David-IV, T. David-V, the ST. David-V, (and maybe there is an S2 T. David-V or VI)? Tell us the differences in the outcomes between the patients with a straight graft and a modified graft. DR DAVID: My partner, Dr Feindel, hasn’t changed. He believes that we should not change things that work. So we have a good comparison between two surgeons who never change technique. There is a learning curve. The first 5 patients that I showed you were operated on in the first 2 years of the series. Since 1989, not a single patient has developed more than mild AI, or less than mild or trace AI. The answer is that it makes no difference after 14 years. The longest follow-up—in the 3 or 4 patients who are now at 15 years and had a straight tube—shows a normally functioning aortic valve. But 15 years in a 20-year-old Marfan patient is perhaps inadequate. We need 30 years of follow-up. But the reality is that the cusps have not changed after 13 or 14 years of follow-up. The straight tube is a good operation. DR KOUCHOUKOS: I would agree, and I think the echocardiographic studies that you look at in the patients, even with Marfan’s syndrome, at 10 and 12 years with a straight graft show excellent coaptation and excellent function of the leaflets. So there are a lot of theoretical reasons why we should use a sinus graft, but the long-term data don’t really support it. DR DAVID: The only thing that changes, if you create a sinus of Valsalva, is that the velocity of closure of the cusps decreases. It is not normal, but it decreases. In a straight tube, the average closure is 35 to 45 meters per second. If you create sinuses, it drops to 25 to 35 meters per second. It is a wide range, and I don’t know why there is so much variability among the patients. DR KOUCHOUKOS: But it doesn’t translate to any difference in long-term outcomes, at least from information we currently have? DR DAVID: Absolutely not, up to 14, 15 years. DR MILLER: It is a theoretical question that will probably remain unsolved forever. DR THIERRY CAUS (Amiens, France): Preservation of the bicuspid aortic valve is a controversial issue, and I have two questions. The first is addressed to Dr Schaefers, and the second to the panel. Dr Schaefers, you described a technique of plicating the valve in case of prolapse in order to increase the height of coaptation. I feel that the risk of doing that if you take too many stitches might induce some kind of stenosis. Can you provide us with the mean gradient postoperatively for the 57 patients operated on with this technique? DR SCHÄFERS: I am always concerned when I plicate, not that I will create stenosis, but that cusp motion will become restrictive, which can also result in regurgitation. This is one of the reasons why I have started this measurement of effective height, because once I have achieved 8 mm, I can be 98% certain that cusp configuration will be okay on the echo. If the height looks good but I am a little uncertain, I simply take one stitch out. There was one assumption in your question, which I would like to address. If you place so many stitches, is that perhaps bad? I usually go for interrupted stitches, and I have placed as many as eight or nine single interrupted sutures in the free margin. If there is a lot of tissue redundancy in the cusps, a centimeter or more, I then go for triangular resection in order to do the rest with plication sutures. DR CAUS: And therefore you never observe an increased gradient after this procedure? DR SCHÄFERS: No. Mean gradients in bicuspid valves are in the range of 5- to 6-mm, and thus essentially physiologic. Even under exercise conditions, these gradients are only slightly higher than for normal tricuspid valves. DR MILLER: If you pick your bicuspids correctly. As you have always told us, you are ultra-selective in your bicuspids, and that is why your durability is better in bicuspid valve disease than trileaflet valve disease. DR CAUS: The second question is for the panel. When operating for a bicuspid valve, one should be concerned about the potential late aortic complications. What cutoff would the panel recommend for replacing the sinuses of Valsalva? What is the diameter of the aortic root in cases of bicuspid valve above which you should replace the root? DR DAVID: I think he is asking, in patients with an ascending aortic operation, when do you replace the root when you do something for these sinuses? What is the cutoff for tolerance to leave these sinuses in or to excise them during aortic surgery? DR MILLER: When do you just replace the supracoronary tubular ascending aorta and when do you replace the sinuses? Is that the question? DR CAUS: The question is in the case of a bicuspid valve, whether you preserve the valve or you replace the valve. DR MILLER: The sinuses tend to be small and only rarely do those sinuses, unless they are very thin, need what you do in a Marfan’s, because most of the dilatation is in the transverse arch and the distal tubular ascending aorta. You can get big sinuses in bicuspids, but it is a black and white situation compared to the root aneurysms in the Marfan’s. I would be interested in what our friend from Luebeck says, the naturally perfect bicuspid valve, two sinuses, two commissures, two symmetrical cusps. They now call it the Sievers-Schmidtke type 9D or something like that. They are rare but they exist, and I think they are wonderful valves to spare, and the sinuses in the handful I have done are thick, normal, and not dilated. I leave the whole sinuses. What is going on in Luebeck about leaving the sinuses in bicuspid valve disease? DR ERASMI: In the adult, if the root diameter is 5 cm or more, we don’t leave the sinuses. And if there is an ascending aortic aneurysm and a 4.5 cm root, we will replace the root too. I wouldn’t leave a root with more than 4 cm in when I do a supracommissural graft replacement. DR MILLER: I think the gentleman from Amiens is trying to tell us you can leave the sinuses more often than we think? DR CAUS: No. What I am trying to say is that we should maybe replace the sinuses more often, because those patients are more likely to present with acute type B dissection during follow-up. DR MILLER: But that is bicuspid valve disease. It is a different disease. The sinuses are abnormal but less so than the rest of the ascending aorta. DR MARTIN MISFELD (Luebeck, Germany): I would like to make a general comment about which graft design is superior. I share the opinion of Dr David that there is no difference, at least in the outcome, after 15 years. And so far with the David-V Stanford modification, also the De Paulis graft, we have failed to imitate the anatomy of the normal aortic root because we create one sort of sinus but not three sinuses. And I think we are far away from imitating the physiology of the complex changes of the aortic root dimensions during one cardiac cycle. So I think we should focus in the future on coming away from the Dacron graft and on developing new materials that have flexibility and imitate the windkessel function of the aorta. DR MILLER: We need new materials instead of rigid Dacron: a good point. DR EMMANUEL LANSAC (Paris, France): As has been seen this morning with a beautiful presentation, and as we can see from the present discussion, there is still room for a standardized approach for conservative aortic valve surgery. Furthermore, as has been clearly explained, the main concern of a valve-sparing procedure is durability and long-term benefit for the patients. My question is at the time of evidence-based medicine, how can we define proper indications of conservative aortic valve surgery without a randomized study comparing the aortic valve–sparing procedure to the gold standard composite graft? DR MILLER: A good question. Let me speak for the National Marfan’s Foundation. There are now 22 centers in Europe and North America where we are not randomizing, but at least prospectively analyzing what is done in patients with the Marfan syndrome, whether it be a Yacoub valve-sparing, a David valve-sparing or a composite valve graft. Why isn’t it randomized? Because to enter enough patients and to power it adequately to detect a difference in anything less than 10 years is just prohibitively expensive. The events that are going to occur are rare, they are late, and you have to have adequate statistical power. That means hundreds and hundreds of patients in each arm and long follow-up with annual imaging, and the costs are just astronomical. But we will have registry data within 5 years from 22 centers in North America and Europe, and the question will be addressed, nonrandomized. Does anyone else have a reason why not to do a randomized study? Why can’t the German Ministry of Health force a randomized study? I think that is what Emmanuel has gotten the French Health Ministry to do. DR KARCK: For a single institution it is not possible, and it would be a multi-institutional task to design and launch such a study. We have not started that, and I don’t know whether or not we will embark on this endeavor in the future. DR LANSAC: The French study, the CAVIAAR study on conservative aortic valve surgery, does not involve a huge number of patients. The calculation is 120 conservative aortic valve operations compared with 120 mechanical composite grafts over 3 years, so that we can have an answer. The main thing is that we as cardiovascular surgeons agreed to a standardized approach. As we have seen this morning, this is pretty difficult.