Aortic Valve Surgery: When to Perform Ross Procedure, Repair or Replace with Root Management

Podcast Transcript

Announcer:

Welcome to Cleveland Clinic Cardiac Consult, brought to you by Sydell and Arnold Miller Family Heart, Vascular & Thoracic Institute at Cleveland Clinic.

Lars Svensson, MD, PhD:

I'm Lars Svensson. I'm chairman of the Heart and Vascular and Thoracic Institute. We're going to talk about managing the aortic valve that's associated with enlarged aortic root or is leaking, and the options for repairing a three-leaflet valve. That may or may not be associated with enlarged aortic root, but usually, patients have enlarged aortic root and that means we do the reimplantation operation to preserve the valve and also replace the aorta. In the patients who don't have enlarged aorta, we use another type of operation, which involves working on the aortic valve leaflets and tailoring the sinotubular junction down to a normal size or tightening it in a sense.

I'll go over the operations, the options. And then, if you're not a candidate for having a reimplantation operation, I'll touch on also the operations we do here for enlarged aortic roots or patients who, say, have had a previous aortic valve replacement and now develop a large aneurysm of the root; or they've had a Ross procedure and it's failed and we have to redo that operation; or they've had a homograft put in. In other words, a human valve put in and those typically calcify and we have to replace them. I'll go over all the options and what the survival looks like with the various operations and what we call the freedom from reoperation, which we usually use as a standard of 10-year risk, and the pros and cons of the approaches and the results of the various operations we use.

This talk is entitled Aortic Valve Regurgitation, when to do the Ross versus repair versus replace with root management. Note, this is for aortic valve regurgitation in particular. Just to give you some idea of our numbers of aortic valve repairs during the decade of 2010 to 2020, we did 4,680 aortic valve repairs, of which just under third were bicuspid and unicuspid valve repairs. During this decade, our mortality rate for elective aortic valve replacements was 0.8%. In the last five years, three of those years, we've had no deaths related to isolated aortic valve replacements.

When it comes to root surgery, this is just a summary from 2021, we did 323 Bentalls, that means replacement of the root and the valve with a built-in valve. We did 76 homografts, which are using human valves usually now for only patients with infections. We did 19 Ross procedures. We used to do a lot more Ross procedures. We've backed off doing them because of the problems we've seen. And we did 125 root reimplantation operations. You see the division there into bicuspid versus tricuspid or three-leaflet valves.

Our results are considerably better than the rest of the country. In the database study from STS that we did or I was involved with, the risk of death was 2.7%. For us for elective surgery, it's around 0.25% if you take all patients who have composite valve grafts and reoperations. We just did a study on our reoperation for aortic root replacements and the risk of death was just over 2%. There's a new classification of bicuspid valves and they are phenotypic or describing the appearance of the valve, how the valve is fused, and then taking into account the symmetry of the valves and also the orientation and whether there's a raphe involved. This classification is much more practical than previous classifications. The phenotypic is, really, a description of the valve and the common findings we see. And then, the raphe, which is the leftover commissure of what would normally have been a three-leaflet valve and then the symmetry and then whether there's associated aortic disease or not. This then is used in the imaging studies to describe the findings.

Let me make some comments about the Ross procedure. Donald Ross was a South African who had learned about switching leaflets and valves in patients that was done on some research in Pretoria and then popularized the Ross procedure. Essentially, what the Ross procedure does is take out the pulmonary valve, puts it into the aortic position, and then attaches the coronary arteries to that. In most cases, some people have used a sub-coronary implantation technique and then the gap in the pulmonary valve and the pulmonary artery has to be replaced with a homograft, in other words, a human valve. That's usually the way the operation is done. You now have put two valves at risk and they have the risk of deteriorating, in other words, the aortic valve and the pulmonary valve, over time. These valves can break down over time.

I used to do a lot of them and I backed off when I found that, because I used to test the pulmonary valve before I used it to see if it was leaking by using intraoperative pressurization of it and looking at the echo. A lot of them were leaking, particularly in patients with bicuspid valves. When you look at the structure of it, deterioration over time, the freedom from structural valve deterioration in our hands at the Cleveland Clinic was about 85% at five years, which if you listen to my earlier talk, we talk about 10-year freedom from reoperation. For example, repair of a leaking bicuspid valve in our hands has about a 91% freedom from reoperation at 10 years, not five years. Here's just the risk of operation on the autograft, as it's called, so the pulmonary valve and aortic position over five years and then on the pulmonary valve.

Now, the pulmonary valve does calcify over time and we don't see that usually until about 8 to 10 years, then we start seeing the calcification of the pulmonary valve and then that also has to be addressed. We do a lot of revisions of Ross procedures, either reimplanting the pulmonary valve, a patient's own valve, into a new tube graft or reversing the Ross. In other words, we take the pulmonary autograft and put it back into the normal position and then replace the aortic valve. It's a complete reversal of the original operation. Here are some results with longer term data and the risk of reoperation over time.

Let me say, and I stress at the beginning, for aortic valve regurgitation, the general consensus is the Ross is not a good operation and we have much better operations for bicuspid valve repairs or, for that matter, three-leaflet valve repairs than doing the Ross procedure. That, to me, is the way to go. In patients with a ascending aortic aneurysm and needing potentially a Ross operation, the general consensus is probably not to do a Ross procedure, because over time that aortic root dilates and then we have to re-intervene and do a reoperation root procedure. When does the root need to be replaced when a patient has aortic valve regurgitation or incompetence? The way we looked at this many years ago is we took the cross-sectional area of the aorta and divided by the patient's height. If it was more than 10, we found that that was a good recommendation for surgery. In a subsequent study done here from the Cleveland Clinic using CAT scans and looking at the cross-sectional area, in patients who had a ratio more than 10, their long-term survival was considerably reduced by not having the aorta replaced. In that particular study, we found also that if the aortic root is enlarged, then that is also greater risk factor for death over time.

Several years ago, we looked at all our patients who had bicuspid valves and ascending aorta above 4.7 centimeters and looked what happened to them. 801 had surgery immediately and the other group of patients went under surveillance. The patients had surgery, the risk of death was 0.25%, so very low. In the patients who had surveillance, we then followed them over time. We found the hinge point is around about just over five centimeters for the aortic root and increasing the risk of a major aortic event, such as aortic dissection; for the ascending aorta may be slightly higher, probably around about 5.2, 5.3 centimeters, the cutoff to intervene.

When we looked at this from the various calculations, what was the best predictor in the cross-sectional area-to-height ratio was the best way to decide when patients should have surgery. When we looked at the aorta over time, the patients who started off with aorta at 4.7 centimeters, about 50% of those patients 10 years after surgery needed the aorta replaced. Obviously, the patients with big aortas had a greater need for having the aorta replaced. There's a recent study out of Yale that found essentially the same. In other words, the risk of death increases considerably when the aorta gets above 5 to 5.5 centimeters, here in the black curve. Here's just the data again for that particular group of patients and the risk of rupture or dissection or death related to the aorta.

Here's another way to look at them, particularly in that category of the 5 to 5.4 cent-... sorry, 5 to 5.5 centimeters and the risk of something happening. The hinge point is very similar here at about five, but more towards below five centimeters. They study the risk of a serious event increases tremendously. Also, in their study they found in older patients that size is a bigger predictor of an adverse aortic event. What do we do for bicuspid valve repairs and also for three-leaflet valve repairs? This just shows the figure-of-8 hitch up suture that I use in my patients to improve the chances of a successful aortic valve repair. This happens to be a bicuspid valve, but I also use this for three-leaflet valves. The way I teach my residents and trainees to look at the valve is to follow the schema, the class schema. So, you look at the commissures, the leaflets, the annulus, sinuses of Valsalva, and the sinotubular junction. This is a functioning unit, so everything has to be addressed and work together to result in a successful repair of a three-leaflet valve or a bicuspid valve.

Here's one study. I'm going to run through a couple of studies that we've done to try and clarify what the best approach is for patients with, in this case, bicuspid valves. The question was whether putting a tube graft above a bicuspid valve repair results some better results than just doing a bicuspid valve repair. This is the clinical question for the study, so we looked at 842 of our patients who had bicuspid valve repairs and divided them into those who had ascending aorta plus repair or just the bicuspid valve repair alone. Essentially, not much in the way of difference.

The patients who had the bicuspid valve repair alone had a slightly lower risk of survival out to 12 years but not statistically significant. Severe aortic valve regurgitation made no difference whether we replaced the aorta or not. As far as root dilatation also did not make a difference. Our cutoff for root replacement is 4.5 centimeters and freedom from reoperation is essentially the same in those two groups. Reoperation for aortic valve stenosis, so calcium buildup on the leaflets was essentially the same, and reoperation for a leaking valve was also very low. Remember, these were patients that had a leaking valve that we operated on and 91, 92% were free from needing an operation because the aortic valve was leaking. As far as root dilatation, very few of those patients needed a reoperation for that.

Our conclusions from that study were that the outcome from bicuspid valve repair is excellent with or without ascending aortic replacement. Stabilization of the sinotubular junction does not seem to influence the durability long-term of these repairs. Ascending aorta replacement, we believe, still a cutoff should be about 4.5 centimeters for replacing the aortic root or, for that matter, the aortic arch. We did another study looking from 1985 to 2011, all patients had bicuspid valve repairs that were just under 2,000. Then, we looked at the more modern cohort from 2001 to 2011 and, obviously, you then get long-term outcomes when you choose a population like this. 1,124 patients with bicuspid valve repairs. These are the various operation techniques that we use to repair the valve. I won't go into them and it all depends what the findings are at the time of surgery.

If you look at the risk of reoperation out to 10 years, it came in at 90, 91% freedom from reoperation at 10 years. If you look at our modern series versus the entire series, then the modern series had better outcomes. We've become better at repairing bicuspid valves over time. A couple of technical points. When we use the figure of eight suspension stitch that I showed you at the beginning of this talk, the risk of reoperation was much lower than the whole group of patients. When we also braced the aortic root, whether a reimplantation or some type of root procedure, then the long-term durability was also better in those patients. It's pretty good there at about 95% of 10 years.

I mentioned in one of the talks earlier that when we compared the patients who had the operation to age and sex and race match due as population. The outcomes as far as survival were the same as the general population. As far as the recommendations from that particular study, we think it's very important to follow the class schema, avoid repair in patients with stenosis and calcification, and patients with severe aortic valve regurgitation from fenestrations, and use the figure-of-eight suspension stitch and brace the annulus and root when needed, just as we do with mitral valve repairs.

What about comparing bicuspid valves with this three-leaflet valves and aortic root replacement using the reimplantation technique? In the reimplantation technique, we free up the valve as I described, and then we put pledgeted sutures through the left ventricular outflow tract below the aortic valve through a Dacron plastic tube graft, and then we hook up the coronary arteries to the side and the valve now sits within that tube.

We then looked and compared the outcomes of three leaflet and two leaflet valves during this period ending in 2017. 607 patients. What happened in the series showed that leakage of bicuspid valves occurred more quickly than the three-leaflet valves. When we look at the risk of reoperation, the bicuspid valves had a higher risk of reoperation. However, that was not statistically significant when compared to the patients who had the three-leaflet valves. The survival was also no different, but on the other hand, the bicuspid valves did particularly well in that series. As far as residual aortic valve regurgitation, bicuspid valve still had a bit more residual regurgitation over time.

Another study we did was asking the question in patients who don't have a big root: Would it be helpful to use the reimplantation operations to try and preserve the leaflets? The advantage of a reimplantation operation, you got much more control in preserving a three-leaflet valve that is leaking. This, obviously, is in valves where there are no major issues with the leaflets, and the results were pretty good longterm out to 10 years shown here.

Another study we looked at was, "Okay, so how does a reimplantation compare to the Bentall operation, where mechanical or biological valve is put in patients with three-leaflet roots?" We had 643 patients in this. The risk of death was 0.16, with reimplantation 0.22 for the Bentalls, and the freedom from reoperation for the series was 97% at 10 years. Really excellent risk of reoperation. In other words, in the patients we did the replacement of the aortic root, either with a Bentall or reimplantation operation, long-term results are really good and the risk of another operation is low. Here's the survival. No difference between the two groups. Here's the risk of developing severe aortic valve regurgitation with the reimplantation, with your own valve versus the Bentall. Really, no difference.

When it comes to the gradient, the pressure drop across the aortic valve, the reimplantation, because that's your own aortic valve. The results were better. If you look at the heart's recovery after the operation, slightly better with the reimplantation operation. Freedom from reoperation in the reimplantation patients was very good, over 97, about 98%, and none of the Bentalls needed another operation within 10 years in that category or time period that I mentioned.

Another operation we do for leaking three leaflet valves is we do a repair of the valve and we then tailor the sinotubular junction. This is a sinotubular junction and here's a smaller graft that we're using to narrow down this, because that's sometimes the reason the valve leaks, and putting tension on it to hitch it up. Here, we have put in what we call Cabrol stitches and you see a very nice apposition of the leaflets. How do these hold up? It turns out there's very little difference in our hands with a reimplantation, the remodeling operation, which I really didn't discuss in the tailoring operation as far as long-term outcomes. But bicuspid valves do have a higher risk of failure over time, and that's because there's a abnormal valve. Here, this just shows the risk of a bicuspid valve versus a three-leaflet valve. Also, we are very gratified that Marfan syndrome and... I'll talk in the other talk about connective tissue disorders of other types. The results are very good and also for patients with aortic dissection or generation. And then as I mentioned, the bicuspid valves aren't quite as good.

Here's another series. 957 patients who had a root procedure and we looked at four different operations to compare their outcomes over time in patients who need their aortic valve to be addressed and the aortic root to be addressed. Here are the operations. This is the remodeling or operation, sometimes known as the Yacoub operation, where you cut slits into the graft and then accommodate the commissures and sew the valve to the graft and you sew along the aortic valve annulus. Here's the reimplantation operation, as I showed you earlier. Here is a composite valve graft and, in this case, with a tube graft to the left main, an operation that I've done quite a lot of, in particular reoperations.

This patient had additional bypasses to the blood vessels coming off the aortic arch and an elephant trunk procedure. Then, there's the biological composite valve grafts. We often make those on the table ourselves. We take a biological valve, sew it into a new tube graft, and then we replace the aortic root. And then, there's the allograft or homografts, a human valve that is used. So how do they do? Well, we started off with a series that's more or less equal in proportion of patients in each category. Here's the survival over time. The valve preserving operations, so reimplantation or remodelings, and the patients who had Bentall mechanical valves had excellent survival out to 15 years, around about 85% 15-year survival. The Bentalls with biological valves did not do as well, and that reflects probably that usually the biological Bentalls are in the older population.

As far as gradients across the valve, in other words, resistant, the heart has to pump against the valve preserving operations or reimplantations. The transplants had the lowest gradients. As far as risk of reoperation over time for the valve preserving operations, it's usually leakage for composite valve grafts with mechanical or biological valves. It's usually because infection of the valves are endocarditis. In the patients who had the human valves, it's usually from deterioration of the valve and either calcification or leakage and the valve leaflets essentially fall apart over time.

Here's a freedom from first reoperation and here is what the freedom from reoperation looks like. Composite valve grafts hold up very well over the long-term out of 15 years with something like 97% free from another operation. Valve preserving operation hold up very well but then deteriorate over time, but remember, this is an older series and it's mainly remodeling operations, which we now avoid. The reimplantation operation holds up much better. And then, the allograft, as I mentioned, start deteriorating quite markedly after about seven years.

When it comes to the risk of operation and a reoperation for an aortic valve at the Cleveland Clinic after about 2006, maybe 2010, there's no difference in the risk of death at the Cleveland Clinic if you have a reoperation of your aortic valve versus a isolated aortic valve. Longterm survival is quite similar. This shows data out to 20 years. So why do we repair the valves? The reason is that in patients with mechanical valves, this is an older study from the turn century of 1999, and about 50% of patients with mechanical valves at 10 years will have had some problem with the mechanical valves. The survival is very good, but there is the risk of clots, bleeding and related to the valve. That's why we'd like to repair the valves.

The question that sometimes comes up: Should I have a cow valve, pericardial valve, or a human valve? There's no benefit. The cow valves over time deteriorate slowly, whereas the human valves can deteriorate quite suddenly, and it's much easier to reoperate on a patient who had a cow valve put in than a patient who had a human valve put in. That's the reason why here in powder blue, particularly in younger patients, we do a lot more repairs. Having said that, two weeks ago I operated on an 81-year-old lady who particularly wanted me to keep her valve leaflets for her, and so I did a reimplantation operation for her and she fortunately did well. I hope you find this talk useful and that you get some insights into what we do for these type of procedures.

Announcer:

Thank you for listening. We hope you enjoyed the podcast. We welcome your comments and feedback. Please contact us at heart@ccf.org. Like what you heard? Subscribe wherever you get your podcasts or listen at clevelandclinic.org/cardiacconsultpodcast.