Dr G. Dreyfus
(Monte Carlo, Monaco
). Dr David, as always, an excellent presentation; however, you told us that tricuspid regurgitation (TR) does not increase with time and unless severe, it is not an issue when performing mitral valve repair, if I did understand well.
(Toronto, Ontario, Canada
). Yes, when we started practicing, only severe TR or in patients with right-sided failure was the tricuspid valve was repaired.
Dr Dreyfus. Well, you would not be surprised if I did say right at the beginning that I disagree with the message that the title of your presentation conveys. But I would like to come back to your presentation.
You are dealing with a series of 1171 patients operated by one single surgeon, and a great surgeon, from 1985 until 2005. So first, your series can clearly be divided into 2 time periods, 1985 until 2005 and another one from 2005 until 2010. Roughly you have about 20 years of 25 where you have not really valid data on the preoperative tricuspid valve and its regurgitation grade. I quote you in your paper:
“In reviewing the echo reports from studies done in the 1980s and the 1990s we found that the TV was frequently overlooked by echocardiographers because their written reports contained no information on functional status of the valve or, simply stated, that the right-side valves were normal without quantification. With time we realized that a few patients developed transient right-side failure during the first couple of weeks after mitral repair and they often had evidence of moderate to severe TR before surgery.”
This is probably one of the most important issues of your presentation, as you are telling us that you had no preoperative data concerning the tricuspid valve, and let's hope that you will not tell us that these data were available at the time of surgery under anesthesia, because then you would agree that grading would then become completely irrelevant, even more so for the tricuspid valve than for the mitral valve. So how can you therefore conclude that there is no TR increase as you have no reference point? You did say you had 138 who had moderate severe TR preoperatively but only 55 received an annuloplasty. Why the others did not and what was their outcome?
But overall, can you seriously say from your data that TR is uncommon after mitral valve repair and do you think that your data supports such statements.
Dr David. The preoperative TR was based on intraoperative transesophageal echocardiography; if the intraoperative transesophageal echo showed no TR, that is how it was coded, but, if anything, we underestimated TR intraoperatively. Likely more patients have TR than reported, according to your argument.
Dr Dreyfus. Yes, of course, you underestimated it, but we do not know really what is the evolution of TR grading if you mix up, first of all, data from 25 years back where the tricuspid valve was not analyzed in the same way as it is today.
Dr David. But only 20% of the patients died; 80% survived over 30 years and they had echo afterwards.
Dr Dreyfus. Hold on. Death is another issue.
Dr David. And since 2003 echo was done according to guidelines. So even if they had TR or not before and if they had an echo after that time, what does it matter? The reality is 15 years later only 9% of patients had isolated TR.
Dr Dreyfus. You showed us a very important slide. You have a table, which is Table 1 in your manuscript, which assigns a grading to the 1171 patients subdivided into none, mild, moderate, and severe. But accordingly to American College of Cardiology/American Heart Association and European Society of Cardiology echo guidelines, only none/mild and severe TR can be identified. Moderate TR is not quantifiable. This is why in our paper from 2005 we stressed that annular dilatation was more relevant than TR grading, and, if I understand well, you have mixed up in this assignment, data from transesophageal echocardiography and data before surgery, which I think from a methodology standpoint is not fair.
It is funny enough to stress that you had about 13% of patients who showed moderate and/or severe TR, the same number as the Mayo Clinic did, who also denies that it is an ongoing process, the same figure as we had in our paper in 2005 to which you refer. They also had an increase up to 30% and they said this was not a problem, as you are saying that it is not a problem. It is not a question of data analysis, it is the interpretation of it which seems to be wrong.
Dr David. I don't quite understand your question, because these patients were followed prospectively. They had an echo every 2 or 3 years. Yes, in the 1980s and the 1990s the cardiologists didn't look at the tricuspid valve as they do now. So the reports frequently read normal right side valves. So I assumed that if it's normal there was no severe leakage.
Dr Dreyfus. Sorry, for an average European surgeon or French surgeon, it's difficult to understand. If you enter data and you go back to your data 25 years back you are telling me it is prospective? This is a retrospective analysis. You looked at the echo of the patients 25 years ago. Moreover, you relied on echos performed by referring cardiologists and not according to a core laboratory review. It does indeed weaken your analysis even if prospective.
Dr David. The patients were followed prospectively. I looked back 25 years, but our patients were followed prospectively, not retrospectively. The patients who had mitral valve repair, I saw them at 2 or 3 months, did an echo. The cardiologists continued following every year, had more echoes. They were followed prospectively.
Dr Dreyfus. Okay. Second point. Most likely for statistical purposes you defined as endpoints reoperation and occurrence of more than moderate TR. The main issue is that there is no way to know how many patients had TR and how many did not, especially if reoperations are mixed with moderate severe TR. Similarly, there are no clinical symptoms reported. It is impossible to know whether patients were symptomatic in relation with their TR other than those reoperated.
If I did report, for instance, 2000 coronary bypasses and did tell you that no patients had ischemia, occluded grafts, secondary stents because they did not come back for surgery, basically you would believe me? You are telling this exactly.
Dr David. But that isn't what I am saying. I said there are 5 possible outcomes and I summarized them.
Dr Dreyfus. But why did we mix up reoperation with an increase of TR grading? This is another issue.
Dr David. Wait a second. If you read the manuscript carefully, in the isolated TR, only 2 of them died from TR; they were considered inoperable.
Dr Dreyfus. Hold on. Dying is the ultimate issue. You can have TR; you can have no reoperation and not die. Your paper is that TR is uncommon after mitral valve repair.
Dr David. And it is uncommon.
Dr Dreyfus. You didn't stick to the title. I am not fighting your findings. I am fighting the message that you convey based on your data. You are mixing up reoperation, which is an entity, and not all patients with recurrent mitral regurgitation or TR are being reoperated, and basically as we do not know which are which, tell me how we can know how many patients have developed secondary TR? We can't.
Dr David. We did. I showed you there: 7% of the entire population developed isolated TR and 2% developed recurrent mitral and tricuspid regurgitation. In the slides it was listed as valve-related complications.
Dr Dreyfus. Seven percent is your incidence of annuloplasty.
Dr David. No, I am sorry, no. The annuloplasty incidence was 4.7%. Of those patients, by the way, 30% developed TR again at 15 years, indicating that tricuspid annuloplasty is not the answer either; it doesn't solve the TR permanently.
Dr Dreyfus. Well, I don't want to get into the tricuspid annuloplasty business, because basically you have used so many different technique over 25 years and did not mention or identify the tethering entity, which cannot be treated by annuloplasty alone.
Dr David. Yes, that's correct.
Dr Dreyfus. —that I don't think that it is relevant to talk about surgery, because you have been using De Vega, the Simplici-T, and the rigid or not rigid rings.
Dr David. That's correct.
Dr Dreyfus. So I don't think this is the point. And we are not talking here about surgery. We are talking about the development of TR and whether this disease is a self- perpetuated and ongoing process, and this is what you are negating and this is why I am not in agreement with your message.
Dr David. Gilles, what I am conveying is if you repair a mitral valve and the patient does not have TR before the operation, the chance that they develop TR within 15 years is only 5%, develop moderate severe. That is all I am saying. If you had TR before the operation, bad news.
Dr Moon. I am going to have to ask you guys to take this out back. I heard some criticism of the paper from Mount Sinai. Is there anyone here from Mount Sinai?
Dr Dreyfus. It's a pity. It's a very constructive discussion. I like the responses of Dr. David. I have not finished, however.
Dr D. Adams
(New York, NY
). Tirone, it's a pity we didn't have a microphone to record our dinner conversation together last night as we could have played it now. We have been discussing your paper for a while. I said last night I was going to tell everyone that I have always respected you as a master surgeon, but now I know you are a master magician as well. It is “magical” to suggest your analysis is an answer to our paper presented by Joanna Chikwe last year where we described a contemporary series of more than 600 patients treated exactly the same way in the operating room, with a preoperative echo grade of TR not just from the intraoperative transesophageal echocardiography but on preoperative studies, and describing long-term postoperative data on degree of tricuspid regurgitation, functional class, right atrial size, pulmonary hypertension, and the recovery of right ventricular dysfunction.
So I don't think that your retrospective analysis exactly “answers” the paper we presented last year; there are a lot of limitations … you agree?
Dr David. I agree entirely. Any retrospective analysis has limitations, although the patients were followed prospectively.
Dr Adams. They were followed, Tirone, but as Gilles stated, many cardiologists “lost” in Canada didn't read his paper in 2005 and probably have not read subsequent published papers on functional tricuspid valve disease. So one still wonders, were they really assessing tricuspid valve function postoperatively? We don't really know. The one thing we do know is your patients survived after surgery, right? Survival was not impacted, but you didn't present data on functional class, and we don't know if they will continue to survive. So you don't really know what the impact of moderate or severe TR is on your patients. Agree?
Dr David. Agree. I had 8 minutes to present. I went over a minute to present this paper. The functional class bears no relationship to tricuspid regurgitation. And we have an actuarial in the manuscript. We plotted cumulative functional classes 3 and 4 versus tricuspid regurgitation. Only 2 patients developed heart failure.
Dr Adams. Okay. Tirone, because the moderators are going to cut us off, I just have a few more questions I have to raise. One is, I am going to do your mitral valve repair for degenerative disease. In 4 years would you like an echo that says mild or no TR or one that finds moderate TR?
Dr David. No TR.
Dr Adams. So we agree that, even though you have not demonstrated a clinical impact of not addressing moderate TR in your patients, you want a more aggressive approach today because you don't want TR after your mitral valve repair. That's why in the conclusion of your manuscript, which you didn't present today, that you now recommend tricuspid repair if you have moderate TR.
Dr David. It was one of the items that moderate TR caused more TR afterwards and it affected survival, so we should repair it, moderate and severe, but not mild. In your paper three-fourths of them had no TR or mild TR.
Dr Adams. Okay, Tirone. So let me ask about a patient in our paper. If you see tomorrow a 65-year-old patient with severe MR, mild TR, right ventricular dilatation, moderate pulmonary hypertension, and ejection fraction of 55%, and their tricuspid diameter is 45, are they going to get an operation?
Dr David. Yes, they are.
Dr Adams. Are they going to get a tricuspid repair with mitral repair?
Dr David. Absolutely.
Dr Adams. But they only have mild TR.
Dr David. But the annulus is dilated.
Dr Adams. Exactly, they have other risk factors.
Dr David. They should have tricuspid annuloplasty.
Dr Adams. And this is the take-home message, because you and I have talked a lot about this since last year. Tirone, I just want to summarize that there are a lot of patients who benefit from concomitant tricuspid valve repair, including patients that don't have moderate or severe TR, and that is based on risk factors for progression of disease. So I think that we agree on. We also agree that tricuspid annular dilatation in a Barlow's patient, for example, may be less important than in a patient with a smaller valve and fibroelastic deficiency. Perhaps we have to be more proscriptive in certain subgroups we are treating without associated risk factors.
Dr David. Absolutely, and after this paper I will be even more aggressive. Likely it's going to be 12%, 15%, 20% of my cases, but never, David, 65.
Dr Adams. That is, again, the real take-home message. You may now repair 25% to 30% of functional TR; we are on the way back to 40% to 50%. We are actually not that far apart in our current thinking. Am I right?
Dr David. I agree.
Dr Moon. I am going to give everybody final words. Thirty seconds.
Dr Dreyfus. Finally, Tirone, I think we do not disagree as much as your title seems to be. We would agree that TR grading is unreliable, that tricuspid annular dilatation is more relevant than grading, and perhaps we need to tailor the indication more than just saying there is some regurgitation and there is an annulus at 40. You agree with this?
Dr David. I agree entirely and that is why a randomized trial is very timely, particularly regarding size of the annulus.
Dr Dreyfus. And you also agree that you should put some of kind of rigid or semirigid ring and not put bands or stitches or De Vega, because these are prone to fail rapidly?
Dr David. We published this. All our failures were De Vega or flexible bands.
Dr Dreyfus. When do you think we will have the data of your 2005-2010 patients?
Dr David. Next year at this meeting. The manuscript is being prepared.
Dr Dreyfus. Well, there will be no more debate then.
Dr Moon. We have 30 seconds from the United States.
Dr Adams. I enjoyed our discussion. Finally, I want to address your thoughts regarding a randomized trial being discussed in the United States. I think we agree that patients with isolated tricuspid dilatation without other important risk factors are an important population to study. Patients with moderate TR, pulmonary hypertension, right ventricular dilatation or dysfunction, low left ventricular ejection fraction, or atrial fibrillation less interesting to randomize. I don't think you would randomize many of these patient anymore. I am right?
Dr David. I agree. But, David, even in your paper, patients who had a tricuspid annuloplasty did not live as long. It is one single line in your manuscript. But Joanna will tell you this, she did the analysis I guess, that survival in those who had tricuspid annuloplasty was shorter, at seven years, significantly so.
Dr Adams. All the more reason to be aggressive because we were doing tricuspid repair in patients with more significant markers of advanced heart disease. Tirone, we are not saying we can cure them. Maybe their survival would have been even worse if we hadn't treated them.
Dr David. The tricuspid is a bystander. It's the patient.
Dr Adams. Tirone, I am going to see you in a few minutes at the coffee break. We will continue the discussion from dinner last night, to this morning, and I am sure for a long time to come.
Published online: March 03, 2017
© 2017 Published by Elsevier Inc. on behalf of The American Association for Thoracic Surgery