In this episode, Marc Gillinov, MD, Faisal Bakaeen, MD, and Siva Raja, MD, PhD, discuss Cleveland Clinic's highlights at the 2023 AATS Annual Meeting. Learn more about Tall Rounds online: http://2207941.mediaspace.kaltura.com/category/Tall+Rounds/75663251.
In this episode, Marc Gillinov, MD, Faisal Bakaeen, MD, and Siva Raja, MD, PhD, discuss Cleveland Clinic's highlights at the 2023 AATS Annual Meeting. Learn more about Tall Rounds online: http://2207941.mediaspace.kaltura.com/category/Tall+Rounds/75663251.
Announcer:
Welcome to the Talking Tall Rounds series, brought to you by the Sydell and Arnold Miller Family Heart, Vascular and Thoracic Institute at Cleveland Clinic.
Marc Gillinov, MD:
Welcome Everyone to this Tall Rounds. This rounds focuses on CCF highlights at the AATS where we had a tremendous and I would say dominant presence. And today we're going to hear about eight or nine of the key presentations from Cleveland Clinic. There was a lot of interest in robotic heart surgery at this AATS as there has been recently because more and more surgeons are getting into it. And most of the robotic talks were centered on visuals, the robotic video and strategy, not a whole lot of data.
Moby Rehman gave a talk in which he discussed our internal approach to posterior prolapse and how do you choose between respect and resect or cords versus resection? And here's what we do for our limited segmental posterior prolapse. We do a triangular resection, more extensive multi-segment prolapse. We do cords and if there's a high risk of SAM, we do a sliding repair. And this is what our setup looks like. We do not use an endo balloon, but there was a lot of controversy, endo balloon versus some kind of clamp. And the answer ultimately was, doesn't matter as long as it works and it's safe.
And then we showed this video in a session with Stephanie Mick, which was our most common approach, and I just wanted to show this here, which will be my whole talk really. This is someone with posterior leaflet prolapse, as you can see, a directed anteriorly directed jet rather. And there's the 3D echo that shows you clearly what is the problem with this valve, it's that piece sticking up. Here you see with the robot the attractive great view. If you said, why do robotic surgery from the surgeon perspective? It's because you can see really well. Of course you can also see well through a sternotomy, but this is particularly spectacular. And one of the keys to doing these resections is to leave some thick tissue on either side. You don't want to resect every bit of bad tissue because then you will have perhaps nothing to sew to or you'll have a lot of tension. So you'll get the idea here that that is actually triangular-shaped as the name implies. And when we put it together, we've left enough tissue that there's some laxity to it. And by the way, to give credit, this is Kevin Hodges operating. We all do this the same, Dan Burns, myself, Kevin Hodges. But you line that up with a stitch at the end and you would do something very similar with a sternotomy. Having a stitch to line up the two edges at the end is a good idea, and then run the next stitch down with the robot. Sometimes you can get a bit tangled as you're running to avoid that, the nurses cut the sutures to a particular length just to make it easier on us.
And here you see, you don't have anyone to follow, so you've got to make sure that you keep the sutures falling correctly.
But I think the main thing with robotic surgery is to recapitulate to the extent possible what we do with a sternotomy, meaning no cutting corners. You don't want to do a lesser operation just because you have a smaller incision. So we just run this back up and at the end of running this back up and tying it, you'll see that the free edge will look a little bit irregular, and that's something that you want to fix. So Kevin will tie this up and then he'll look at the free edge and say, that could cause some irregularity. And irregular line of coaptation can cause a bit of a leak, which is easily avoided. And there's what I was talking about, you can see at the very edge that would've caused a one plus leak. So it's worth looking at your repair of the leaflets when you're done just to make sure that you're happy that you've got a very smooth surface of coaptation.
And we did see a lot of more complicated repairs, but I think repairs become complicated two ways. Either the valve is actually complicated or the surgeon makes it so, and usually mitral valve repair is not that complicated. So I think keep it simple. A posterior leaflet resection is a very reliable way to do this, and every repair is going to include an angioplasty. We use always a flexible band. When you're doing mitral valve surgery for degenerative disease, it does not matter what kind of angioplasty you use. They all work. So use something that works for you. It's particularly easy to see the trigone. You can almost get the feel that's thick tissue, even though with the robot, there's no haptic feedback, meaning there's no tactile feedback. This suture is pre-knotted. You can see that knot has engaged the angioplasty, and that just prevents us from pulling it through. In theory, you don't have to tie this, but we always do just to keep it tight. And then we take this suture, this angioplasty suture, and we do a running horizontal mattress stitch, taking this down to the middle of P2 or about 6 o’clock and then begin a second stitch. The reason we begin a second stitch at 6 o’clock is the needle eventually gets dull. Every time the robot grasps the needle, it creates little ridges in the needle because the robot doesn't do delicate when it grasps things, it just grasps it really hard. You can actually bend a penny with the robot, which gives you an idea of be careful when you're handling tissue. So this is the running horizontal mattress suture, and there are multiple ways to do the angioplasty. This once your facile with it takes about 10 minutes. So it's actually quicker than using COR-KNOT and a whole lot less expensive. Many robotic surgeons perhaps most will use COR-KNOT, which is also A-OK. Simply not necessary.
One of the things you do lose with this technique is the ability to size precisely because we're not putting the band down with the template that maintains the size. So every time we cinch the ring, we're making it a little smaller than it's measured size, which I think is A-OK. We more or less just eyeball it and say just like shirts at some stores, large, medium, small, what do we need, a large or medium or small ring? And as with ring type, you don't have to be super precise because all of them work. This is the second stitch, also pre-knotted. We put it through the ring, then through the anulus and the anulus, you can see very well. Then it comes back up through the ring, and then we'll tie it and tie this stitch to the first stitch. And after this stitch is done, there'll be a third anchoring stitch at the left fibrous trigon to complete the repair. But first we got to make sure we're not tangled and complete this one coming back up.
And so there's the horizontal mattress suture coming up the left side structured to be wary of on the left side, circumflex coronary artery. You can see the beginning of the left atrial appendage off to the left. If you stray into that area, you may hit the circumflex. And here you can get a sense we're getting near the trigone, which we are. So there'll be one more stitch coming and that's going to complete the repair once we put this one through the trigone. And again, you get this idea with your eyes, that tissue up there is a little bit thicker, a little bit more stout, which is what you're looking for.
Once we get this tied, we can move the retractor so that we don't have a lot of tension on the valve and test the valve. One nice feature of the robot is that this retractor, the left atrial retractor drop above with the two arms is mobile in multiple different dimensions. So we can tie this stitch, release that retractor a little bit, and now we test it with a power injector coming from the side, run some antegrade cardioplegia at this point and make sure that we see no leak, good coaptation, and that the coaptation is posterior. And that's what you want it to look like. And that's what the echo looks like. So again, the robotic talks were a lot of show and tell. So I have remained in that vein. This is a good way to do a repair. Thank you.
Faisal Bakaeen, MD
First of all, I'd like to start with the disclosure that I have coronary disease in the family and my grandfather, my father and two of my uncles had CABG. So I feel safe at the Clinic because if ever need my CABG, I've got the greatest surgeons in the world here. Certainly if I have three vessel disease. Now, if I have one or two, I love our cardiologists. Many of them are in the room, and I trust them with my life too. At the Cleveland Clinic, we use a team-based approach to tailor the management as what's best for the patients.
But let's talk about the hot topic of the controversial guidelines. As you know, CABG was downgraded in those guidelines and the evidence provided simply did not support that downgrade. Let's look at the root causes. Well, if you look at the joint committee of the ACC and the AHA, who sits on that committee? 28 members, one of them is a surgeon. There are three pharmacists. So that's a structural problem with the guideline when the surgeons are so underrepresented, this has to change. When it comes to robust trustworthy guidelines, Lars and I represented the AATS and the STS and came up with this document outlining how the structural process and transparency and responsiveness of the guidelines can guarantee quality and robust guidelines that we put out from those associations. For example, when it comes to the process, we can't just pass a recommendation because we have a world-renowned surgeon who's a pusher, who decides he wants to make something Class 1. No, it's an anonymous voting process whereby 80% of the writing committee members should participate and there should be 75% agreement.
And we have a public comment period when the guidelines are written, they're not published right away. They're circulated amongst the membership and the public for two weeks for feedback, and we have to respond to that feedback. When you look at other associations, they follow simple majority vote. All they need is three people out of the writing committee to be present, and for two of them to decide it's a Class 1, Class 2 and Class 3, and this is passed. There's no Delphi process and there's no opportunity for a public comment. Now, they review it internally, etc. And they do the due diligent internally, but it's not externally validated or tested. That's why this guideline was controversial because the CABG downgrade didn't make any sense and there was a global outcry. Yeah, yeah, yeah. Those are surgeons being pissed off that the CABG that they do, which is more than 50% of their practice is CABG.
But we surgeons are very objective and very pragmatic, and we'll look at the actual evidence. So let's look at the actual evidence and see if we were emotional and irrational, or we actually were right in objecting to those guidelines. So if you look, I'm just going to pick one example, and that is the downgrade of CABG and three-vessel disease with normal EF from Class 1 to Class 2B. Four references were cited, reference number 10 and 13 through 15.
Let's go through them one by one and see if they actually support the downgrade. The first one, reference number 10, is the Salim Yusuf meta-analysis. And from the source document, it shows that CABG in three-vessel disease was associated with improved survival regardless of the ejection fraction. Reference number 13, the European study, again, CABG in three-vessel disease in two-vessel disease and proximal LAD was associated with improved survival as those graphs demonstrate. And also in the source document in patients with good left ventricular ejection fraction. So where did that downgrade come from?
Well, the writing committee rightly said, medical management has improved. Yes, it has. I love medical management. It should be the background therapy for every patient with coronary artery disease. And they said, "Well, that's why we should downgrade CABG. It's not as effective as before because medical therapy is more effective." Well, what about the increased safety and effectiveness of CABG? That was totally ignored. And just to give you an example, the FAME 3 Trial that was published in the New England Journal Medicine, the 30 mortality of CABG was 3 in 1000. 3 in 1000 identical to that of PCI. And that is 10 times less than the 30 mortality reported in the Salim Yusuf meta-analysis. So if the mortality, perioperative mortality of CABG improved by a factor of 10 since the Salim Yusuf meta-analysis, did medical therapy improve by a factor of 10?
Maybe, maybe not, but you should account for that, and this wasn't accounted for. And then we get presented with a new evidence because everything else that we presented is old, even though they cited it as evidence. ISCHEMIA trial, we love this trial. It's an excellent trial. I know the investigators, I respect them, but I think the whole study was stretched, manipulated, exaggerated. Really the scope of it wasn't intended to compare medical therapy versus CABG. It was intended to compare initial conservative management versus initial invasive management. And initial invasive management was diagnostic cath. It wasn't CABG, it wasn't PCI. So this reference 14, ISCHEMIA did not randomize a single patient to CABG, and the median survival was only 3.2 years. I mean, when we talk in cardiac surgery, we talk about lifetime therapy. We don't talk about three years.
And in fact, 24% crossed from the initial medical to initial invasive within that very short period of time. And that's not unique to ISCHEMIA. Over 35% crossed over in the Bari 2 Trial. And if you actually look at the kind of patients, the population that was studied, you will see that only 36% had a proximal LAD lesion of greater or equal 50%. Now remember, we use 70%, but even if you go down to 50%, just over a third had a proximal LAD lesion. I can't remember the last time I did CABG on a patient with no proximal LAD lesion. So this population is really not representative of modern-day CABG. And because the atherosclerotic burden and anatomically is low, this is reflected physiologically. If you compare it to the modern-day STS CABG patient, the STS patients are more likely to be diabetic, more likely to be hypertensive, and two to four times more likely to have peripheral vascular disease, cerebral vascular disease or heart failure. We're comparing apples to oranges. We're comparing rare to well-done. I mean, these are two different populations. How could that study inform CABG? It can't. Scientifically doesn't make any sense. And they say, well, it's not just ISCHEMIA. There's this meta-analysis. Well, in this meta-analysis, all the studies were small. The more predominant study, and it was again, ISCHEMIA and half the trials did not include a CABG arm. Half the trials, in fact, only 16% of patients had CABG in this meta-analysis, and the vast majority had low to moderate atherosclerotic burden. Again, not the STS population that we actually operate on. Most of those patients receive PCI. How could this inform CABG guidelines? But what did the guidelines and the evidence show, including ISCHEMIA and this meta-analysis, is that it is safe and reasonable to use initial medical management for patients with moderate or low atherosclerotic burden. And to closely follow them up, remember medium follow up 3.2 years, and to switch them over to intervention when that moment arrives. So that is really what the evidence and that what the guidelines should have focused on because that is what the data supports. It doesn't support the downgrade of CABG. Let's be objective, let's be scientific.
On the other hand, this is one of the relatives that I mentioned of mine who has total occlusion of the right, total occlusion of the left. He had three-vascular disease, normal EF 30 years ago plus. This patient is not living because of optimal medical therapy. This patient is living because of a patent LIMA to LAD that supplies the entire heart through collateral. And if you leave the OR with a patent LIMA to LAD, that patient will have a long life, not 3.2 years, 30 years plus.
So the conclusion that gained unanimous support at the AATS was that CABG improved survival and three-vascular coronary disease, and it should remain a Class 1 recommendation regardless of EF. Thank you very much.
Siva Raja, MD, PhD
Just like every meeting, there's a theme to the talks that are chosen and often it's involving a controversial topic. In thoracic surgery, the topic that has been controversial for the last few years and highly debated is something that started the debate almost 30 years ago. In 1995, Ginsburg, et al at Memorial Sloan Kettering published a paper that said that in early-stage lung cancer, T1 lung cancer, lobectomy, which is you take out the entire lobe of the lung that the cancer was in, was superior to limited resection. And that had been the paradigm for almost 25 years before studies started to challenge this. And so two of the plenary session talks there are both studies looking at a more granular analysis of early-stage lung cancer. As well as in the era of robotic surgery, minimally invasive surgery, where we have the ability to do an anatomic resection, almost like a lobectomy, but in a subsegmental fashion to see if the old paradigm where you had an open lobectomy with a thoracotomy or someone just did a quick wedge resection, a two-hour operation versus a twenty-minute operation, they found a difference. But in today where we have more sophisticated means, is it still true?
So the first one was the Japanese oncology group study that looked at segmentectomy versus lobectomy for two centimeter or smaller lesions. And their overall conclusion was that, in fact the opposite where the preservation of lung was actually beneficial in the overall survival of patients having small lung cancers. Now, just like everything else, if you slice the pie small enough, you're going to find a piece of pie that didn't quite meet that standard. So in patients under 70 years old, there were some areas where it didn't meet statistical significance, but the general consensus was in small tumors, there was a benefit.
Similarly, the CALGB, which is a western study that looked at T1A, which is lesions that are smaller than two centimeters looked at low bar versus sub low bar resection. And they found a similar outcome, which was that in this large study lobectomy, which is the removal of the entire lobe of the lung, segmentectomy, where you dissect out in an anatomic fashion, the artery, the vein, and the airway going to the segment along with all the lymph nodes and wedge resection, which is a more of a blunt instrument if you will, were associated with similar survival in patients with very early stage cancer. Numerically, there may have been some higher local regional recurrence, but statistically speaking, there were no meaningful differences between the groups. So this was the main theme of some of the plenary session talks at the AATS looking at early lung cancer.
There were three original manuscripts that were contributed from the Cleveland Clinic. The first one is once again looking at with this very trend where we are now going towards minimally invasive, lesser resections that have additional that have survival benefit. We looked at segments that are not benefiting from this. So the first one was presented by our research resident, Sadia Tasnim, where we looked at the preoperative predictors of something called STAS. I'm going to talk about that in a second to tell everyone what STAS is so that it makes sense.
So classically, lung cancer invasion is either through lymphovascular invasion where obviously as the name implies, that either the lymphatic channels or blood vessels are invaded, stromal invasion, which means that invades into the tissue around it or pleural invasion are the three classic types of lung cancer invasion. In recent years, we've identified a fourth type of invasion, which is called STAS or spread through the airways. This means that the tumor cells are found somewhere in the lobe that is not near the main tumor. If the tumor cells are not present right next to the tumor, you can imagine if you did a lesser resection, you're likely to miss those cancer cells.
We know from recent studies that it associated with the increased recurrence and obviously overall survival. The problem has always been that you only find this after the fact. You have to take out the tumor to find out if you had STAS. So our team looked to see if we could identify preoperative factors that would tell us who would have STAS and who would not, meaning that who would benefit from this cancer that potentially could have a worse survival and should not get a subsegmental resection versus a lobectomy. So we looked at radiographic clinical and even genetic factors that would tell us if that's the case. The only thing we couldn’t use was pathologic factors, because once the lung is in the bucket, then it doesn't really help you to make a decision.
You could say that you could do a lesser resection and then go back, but talking a patient into going back for a second operation is not as easy, especially for a statistical benefit. So ultimately, we identified six factors that all put together created a significant risk for patients, whether it was solid or subsolid, the size of the tumor and the SUV max on the PET scan, whether it was adenocarcinoma or squamous cell age of the patient, and did they have a KRAS mutation? So how did this clinically help anybody?
Well, currently if you had a two centimeter tumor, there are folks who would do a lobectomy, and there are some people who would decide that they were going to do a segmentectomy. We have two patient examples here who based on who was operating would've gotten the exact same operation today. A 45 year old patient with a solid nodule and a KRAS mutation and an SUV of five who had adenocarcinoma on the one side and patient two who's 80 years old, partially solid, who had a PD-L1 positive, which is a molecular marker for which we have an immunotherapy drug, max SUV of one and squamous cell cancer. So two patients who would get exactly the same operation, but their risk of recurrence or risk of having STAS is 60% in patient one and 20% in patient two. So if we can preoperatively identify who would benefit from a subsegmental resection, you could have saved that 80-year-old patient a lobectomy. But by doing a subsegmental resection on a 45 year old patient, you were not doing an oncologically sound operation. So that was the first one.
The second paper that we presented was from our second research resident, John Barron, looking at paraconduit hernias. So esophagectomy for esophageal cancer is the definitive treatment of choice. Anytime you do a big operation and the patient survives and they survive their cancer and their operation, you're going to have some post-operative complications. A paraconduit hernia is where some other organ, in this case, the colon that is depicted in this picture, herniates through the diaphragmatic hiatus. And now you have both the stomach and the colon, the chest.
The stomach is there because we left it there, the colon's there because even though we didn't want it there, it happens 2 to 10% of the time. And the controversy in the field is that if you were asymptomatic and they, on a CAT scan, found this colon in the chest, what should you do? Anybody who sees colon in the chest immediately thinks we should get it back down in the belly. It doesn't belong there. Well, it turns out that these are patients who have a high risk of cancer recurrence. Their life expectancy in the early years might be lower. Now, once you're cured of your cancer, that's a different story. And of course, if you're asymptomatic and you have this finding, is it the same as a garden variety hiatal hernia where the colon's in the chest?
And so the majority of the folks would actually recommend repairing it, but the recurrence rates after repair are very high. So how many times do you want to keep repairing this? The Cleveland Clinic approach for decades has been to watch folks and when they became symptomatic to repair them, and for the most part, we were alone. So we figured that if you're going to do something that others weren't doing, we should prove it. So we looked at all of our patients, at least in the last 10 years out of 1200 esophagectomies, it turns out we have a very low rate of recurrence, but within our patients, we had no emergency complications when we had patients who didn't have any symptoms, and all our patients are very closely followed. Yes, some patients did increase in size, and we did operate on some folks down the road when we identified it and they became symptomatic. So to quote Dr. Lytle, at some point was, that there are emergencies and then there are redos. There are no emergency redos. So asymptomatic patients were not rushed to the operating room just because of a CAT scan finding. Okay. So the other thing we identified was also that if you don't look, you don't find, or the converse is that if you do look, you're going to find them. Since we follow our patients more closely as opposed to patient coming, getting an operation and then going back to the community, we identified that over time this can occur. And if you follow them closely, you will find these patients and when they become symptomatic, you'll be there to take care of this. So in conclusion, the risk of paraconduit hernia does increase with time. Only a minority of patients will have symptoms. It is safe to watch, and the editors made us add the line. It was safe in our patient population.
The last, and I think one of the most significant papers that was presented there was our paper by Dr. Sudarshan, which looks at a universal esophageal patient-reported outcomes. In today's reimbursement window as well as in patient follow-up in places like us where we are a quaternary care center where patients have to come from far having a way to identify how patients are doing, who needs to be followed up, who needs an operation, how did we do. Having a way to measure this from a patient standpoint, it becomes very important. And the esophagus is this very small organ that's sort of shoved in between the aorta, the heart, the lungs, the mediastinum, everything. And so oftentimes it gets sort of forgotten about it. And so we have these patient-reported outcomes that talk about in general thoracic diseases and so on and so forth. So we developed internally an esophageal-specific patient-related outcome.
So our objective was to create one specifically for this and to be able to use this for multiple conditions. So if you have to give a patient a different questionnaire every time they have a different diagnosis, it's not going to be so easy for the nurses. So we have to incorporate two things: what were the symptoms? And just because you had a symptom doesn't mean it bothered you. So one of the unique aspects of this is to identify how much of this is bothering people as well. There were multiple domains that were looked at, and without going into too much detail, very thorough analysis and validation was done to see how acceptable was it for the patients? Does it measure what you want to measure? Can it be reproduced? And does it measure changes based on what treatment you've given them? And the answer to all of them was, yes. We are going to incorporate this now into our electronic medical records, so every patient coming in for any esophageal problem, we'll get this. And we plan to validate this across multiple specialties, multiple institutions, and hopefully in multiple countries. One of our former fellows is in Italy and is interested in translating this into Italian. So this is our brief update. Hopefully it's not more than eight minutes for a four day meeting. Thank you.
Announcer:
Thank you for listening. We hope you enjoyed the podcast. Like what you heard? Visit Tall Rounds online at clevelandclinic.org/tallrounds and subscribe for free access to more education on the go.