The Heart Team Approach as a Model of Care

On February 21st of this year, Massachusetts General Hospital launched the Mass General Institute for Heart, Vascular and Stroke Care, one of the only institutes in the world to integrate cardiovascular and cerebrovascular care. We talk with Institute co-director Dr. Thoralf Sundt about the philosophy behind the Institute and his perspective as a cardiac surgeon on cutting-edge cardiovascular care.

What does multidisciplinary care really mean?

At our hospital, like many hospitals, care is more fragmented than we would like it to be. That’s actually the rule in healthcare today, in most places. Patients see physicians sequentially, often on different visits and days, often at different locations, and it makes it cumbersome for patients. It is also cumbersome for physicians as well. When you see a patient who has been seen by multiple other caregivers, you shuffle through the papers to try and find documents and office notes. It is not the same as actually having a conversation with the caregiver, seeing what their thoughts are and how they put the pieces together, and listening to their questions for you. 

The best way to deliver care is in a way that promotes communication amongst physicians and other caregivers around the patient in a timely manner. That is most appealing to the patients, it is most efficient for the caregivers, and it delivers the highest quality care — things don’t fall through the cracks and aren’t missed. So, recognizing the value in integrated multispecialty care, we have created the Institute to reorient all of our caregivers around this concept. It has been practiced for some time at other institutions, but it hasn’t really spread very much from places like the Mayo Clinic and the Cleveland Clinic. There are a handful of clinics in the country patterned on those institutions, but most care today remains fragmented. Transcatheter aortic valve replacement (TAVR), particularly, has emphasized the importance of this. CMS communications about payment for TAVR have emphasized the importance of a ‘heart team’ approach. In this specific instance, the need for that kind of integrated care has been emphasized, but in truth, for every patient and every condition, a format or structure that allows for that kind of integration is important. For example, a patient may come to a primary care physician or a cardiologist with mitral regurgitation secondary to degenerative mitral valve disease, have an echo, get referred to a surgeon and have a mitral valve repair. Right now, mitral valve repair does not require a heart team approach, and frankly, most of those people aren’t cared for in a heart team manner, but still go through the traditional channels. It is only because of the payment mechanism of TAVR that the payers have essentially forced this kind of integrated care, which is a good thing. The more it spreads to every kind of care, the better. The patient with aortic stenosis ought to be evaluated by a heart team, too. What do you do with the asymptomatic patient with mitral regurgitation or an asymptomatic patient with aortic stenosis? How do you make the decision about when they should or shouldn’t be referred for surgery? What are their options? Most of those patients still get cared for in a one-off, fragmented way. I think physicians make a great effort to communicate with each other, but it takes a great deal of effort. It is better to have a structure that facilitates that kind of care.

How is the Institute encouraging such care?

Some of it requires a change in mindset. The idea that you can really expect to see the primary care doctor, the cardiologist, and the surgeon all on the same day should be set as an expectation. It should be set as an expectation that people have communicated with each other. It’s not that every single patient has to be seen by every single caregiver, but the focus is on creating a system where it is easy to make those connections happen. We are still in the early stages. For example, if a question arises from the primary care doctor, it needs to be very easy to assemble a team and see the patient. When the phone rings and the question is asked as to when you can see this patient, the answer is at 2:30 or 3:30 this afternoon, rather than three weeks from now. Office clinics must restructure scheduling; changes in infrastructure must occur to make it work. Physicians can expect to have a more fluid schedule, where perhaps they don’t know who they are seeing this afternoon, but a consult just came in and now they are going to see that person. It is not the way most doctors have practiced their careers.  

The specialties need to get closer to one another and understand each other more. For example, in cardiac surgery training, we are spending more time in the cath lab with our resident trainees. That doesn’t mean that when they are done, these surgeons will go out and do angioplasties, but they need to have enough of an understanding of what interventionalists do in terms of percutaneous coronary intervention (PCI) to be more meaningful contributors in a discussion.

Can you talk more about the treatment of TAVR from your perspective as a cardiothoracic surgeon?

Valve replacement is a great example of the need for individuals with different sets of expertise to interact, whether it is for the aortic valve or for the mitral valve. These therapies, just like PCI and coronary artery bypass graft surgery or other interventions, need to be considered complementary and not competitive. During the time I was in training and early after training, it could get a little contentious between interventional cardiologists and surgeons. It was unfortunate for a lot of reasons — it was not as collegial or as much fun, and the patients got lost, because the real question got lost: what is the right therapy for this particular patient? Not who is going to bill. The SYNTAX trial made it clear that each procedure has a place. A recent editorial¹ written by David Holmes, Mike Mack, Jeff Rich and William Zoghbi talks about surgery, PCI and TAVR, emphasizing that each has a place, and that is the way we need to consider these procedures. We must keep the patient at the center. 

If an inoperable patient is going to have a procedure done, right now, TAVR is the way to go. Inoperable patients have no other option. For them, the issue is whether the procedure is futile or not. Being inoperable is a very bad prognostic sign. If you are inoperable, you are not going to live very long, no matter what happens. That has been shown very clearly with TEVAR (thoracic endovascular aortic repair) endografts. Whether it is thoracic endografts, abdominal endografts or aneurysmal disease, patients who are inoperable have a terrible survival. They tend not to die of their aneurysm, but of something else. It may play out in a similar fashion for inoperable patients receiving TAVR. 

The PARTNER trial showed that for high-risk patients, the surgical and percutaneous options are comparable, for both early and late outcomes. TAVR thus has become an appealing option for very high-risk patients. As we look to lower the criteria for TAVR to lower-risk patients, at some point, the lines are going to cross, meaning the risk of the open operation will become less than the risk of the TAVR. Why? TAVR, because of the balloon inflations and deployment of the valve, has a risk of stroke. There is also a risk of the valve dislodging. If a very low-risk surgical patient, let’s say a 40-year-old who is otherwise healthy, has straight-up aortic valve replacement, the operative risk is exceedingly low, far below 1%. In fact, in the Society of Thoracic Surgeons (STS) database, all-comers aortic valve replacement has an operative mortality rate of about 3%. That sets the bar. This is an average, meaning there are many patients well below that 3%. At some point, we are going to hurt people doing TAVR instead of help them. The question is, where is that point? That is what we have to find and titrate. 

TAVR also commits you to a tissue valve, not a mechanical valve. A mechanical valve can’t be done percutaneously. You may say, that’s fine, we don’t want to be putting in mechanical valves anyway, but today, about 15% of all heart valves placed in the aortic position in the U.S. are mechanical. Mechanical heart valves have superb durability and a very low reoperation rate. When people use home INR testing to control their warfarin, the risk of complications associated with the warfarin is very low. Just about every study that has compared the long-term outcome of people with tissue valves versus mechanical valves has actually shown superior survival for patients with mechanical valves. Many patients don’t want to take Coumadin, to which I am sympathetic, but let’s look at a 40-year-old man with a very low operative risk. He may be better off with a mechanical valve, with an extremely low risk of post-operative complications. Maybe he’s on Coumadin anyway for atrial fibrillation or some other indication. It would be a disservice to this patient to do a TAVR instead of an open aortic valve replacement. 

Depending on who the patient is and what the specific circumstances are, the two therapies are complementary. Over time, the operative risks of both TAVR and surgical valve placement will decline. Devices will be developed to prevent embolic phenomena reaching the brain during TAVR. But still, there will be some patients who are better served one way or the other. How that will all shake out, I don’t know. 

Valve repair is another procedure where having a heart team approach is useful. Perhaps a patient might be a candidate for repair rather than replacement. This will be even more of an issue when we consider interventions to deal with mitral regurgitation. A young person with mitral regurgitation secondary to prolapse of the middle segment of the posterior leaflet is repairable 99% of the time, with a 10-year freedom from reoperation of 95%. This patient is likely better off with an operation than they will be with a clip. To make that judgment about a specific patient in a specific circumstance, however, we need specialists with expertise in all related areas. There will still be a subset of people whose primary position is still in the operating room and a subset of people whose primary job, every day, is in the cath lab. In facilitating a fluid interaction between the two, you get the best of both worlds. 

How do your own research interests tie into multidisciplinary perspectives?

One of my interests has to do with bicuspid aortic valve disease and aortic aneurysmal disease. For a patient with a bicuspid valve, the question is, when do we need to intervene on their ascending aorta? There is a growing appreciation for the risk of aortic dissection and a growing interest in prophylactic surgery for aortic aneurysmal disease. In some ways, that’s a good thing. On the other hand, just as with TAVR, there is also a place where the lines cross, where we can be too aggressive about replacing the ascending aorta, and create more problems than we solve. The general trend right now for patients with bicuspid aortic valve disease is an intervention to replace the aorta if it is 5 cm or greater. There is somewhat more consensus around replacement of the aorta at 5.5 cm. One school of thought believes we should be even more aggressive, and intervene if the aorta is 4.5 cm. I can’t argue too much with that, if the patient is low risk. The risk of rupture or dissection between 4.5 and 5.5 is quite low, but not zero. However, if the patient is low operative risk, it is reasonable to intervene. You don’t need to hit the panic button, but it’s probably okay. So now we ask, if you are intervening on the aorta, what do you do with the valve — especially if the valve is only mildly dysfunctional? We see a good number of bicuspid aortic valves that are regurgitant with a moderate aortic enlargement. We will very often repair those valves. More and more, bicuspid valves are being repaired, rather than replaced. Patients certainly prefer it. Yet we don’t want to intervene too early on the valve, so there is some advantage to waiting. If the valve was completely normal and there were no issues, you could argue for intervention on the ascending aorta. That’s true if the valve is a tri-leaflet valve, but if it is a bicuspid valve, you pay a certain price by intervening too early. Some would argue that these patients are best served with a tissue prosthesis, and then planning on a valve-in-valve with a TAVR. We know it can be done, but we don’t know what the long-term durability will be. If someone is 40 years old, you can put in a tissue valve and say, well, you can have a valve-in-valve in ten years. Let’s say the patient has that done when they are 50, and then when they are 60, they are going to have a valve-in-valve-in-valve. You can see that it gets to a point where it doesn’t make sense. 

Are cath labs across the country going to start seeing more congenital and structural heart disease patients on an emergent basis?

That’s a good question. It’s hard to know what direction to go with that. On the one hand, cath labs will probably continue to proliferate for things like infarct PCI, so they will increase in number, but as they increase in number, it’s hard to expect every cath lab to be an expert in every one of these therapies. But there are going to be more and more people out there with a transcatheter valve in place that may have a complication. In addition, there are more adults with congenital heart disease now than there are children with congenital heart disease; this speaks to the importance of having care integrated amongst cath labs as well. 

What we are trying to do, along these lines, is facilitate remote communication and consults within the region. We want a situation where if a cath lab in Maine with a patient on the table wants to ask a surgeon what they think, that we can get a surgeon on the phone, quickly, who can look at the films electronically and discuss the matter. There isn’t a reason in the world that can’t happen. That doesn’t exist right now, because we don’t have the infrastructure built to do it, but that’s where we need to be. With an iPhone, I should be able to answer a call right here, right now, from anywhere, look at films on my iPhone, and say yes, I think we could or couldn’t repair that valve, or I think the patient is appropriate for coronary bypass, or no, you don’t have to do an angioplasty. 

I don’t think there will ever be a substitute for getting everyone together. It can be very efficient, if we structure it in the right way. I have been asked how we pay people for multidisciplinary conferences. It is a reasonable question, but again, in the modern electronic era, these conferences should be able to be accomplished very quickly. It’s truly not very much of a time investment. In my experience, it is more efficient to be in a setting where people are talking to each other about patients than be in a setting where a single physician has to sit down with a chart, flip through it, and find the office notes of some other physician from a previous visit. n

Reference

1. Holmes DR Jr, Rich JB, Zoghbi WA, Mack MJ. The heart team of cardiovascular care. J Am Coll Cardiol. 2013 Mar 5;61(9):903-7. doi: 10.1016/j.jacc.2012.08.1034.