Ƶ

Adult Complications of the Fontan Procedure

— This AP Cardiology podcast episode tackles cirrhosis, arrhythmias, and pregnancy risks

Ƶ MedicalToday

With of univentricular physiology treated with a Fontan palliative shunt, the long-term complications come to the fore. In this episode of the AP Cardiology podcast, Andrew Perry, MD, speaks with Eric Krieger, MD, of the University of Washington Adult Congenital Heart Disease group in Seattle.

They discuss the case of an infant born with tricuspid atresia, review the initial surgeries performed, and then talk about complications that develop in adulthood.

For further reading, Perry recommends the chapter on late complications following the Fontan operation in .

A transcript of the podcast follows:

Perry: Univentricular physiology has fascinated physicians for a very long time. This physiology occurs in specific congenital heart defects where one of the ventricles is underdeveloped. You may have heard of these patients referred to by the palliative surgery that they undergo, the Fontan procedure.

In this episode, I'm visiting with Dr. Eric Krieger, one of the faculty at the University of Washington, who specializes in adult congenital heart disease. We discuss the case of an infant born with tricuspid atresia and an underdeveloped right ventricle. We discuss the initial palliative surgeries that occur as an infant and then discuss what complications to look out for as that baby matures into an adult.

We review cirrhosis, atrial arrhythmias, and pregnancy. There is a lot of great information, and I hope you enjoy the episode. If you do, please go to iTunes or your podcast app to leave a review. Thanks.

This is AP Cardiology and this is your host, Andrew Perry. Hi! Thank you for meeting with me today, Dr. Krieger. Can I have you give us your name and your title for our audience?

Krieger: Sure. I'm Eric Krieger. I'm an associate professor of medicine here at the University of Washington, and I direct the adult congenital heart disease program here, as well as the adult congenital heart disease fellowship.

Perry: Perfect. Thank you. Today I'm visiting with you to talk about patients who have a Fontan procedure, Fontan physiology, which is fairly rare. To launch into that, let's start with a case. I want to start at birth, to talk about a baby that's born at 38 weeks with tricuspid atresia. How do we think about those patients when they're born with tricuspid atresia and a maldeveloped right ventricle? Starting from there, what are the initial problems that the baby is experiencing, and then subsequently, what are the initial surgical procedures to help this baby along?

Krieger: Sure. is one of several different univentricular anatomies, where what you end up with is one single functional ventricle. Now, actually, almost all of these babies formally have two ventricles, but one may just be so maldeveloped or so hypoplastic that it really can't achieve any functional goal.

If you have one single functional ventricle, you need to deal with the fact that all your oxygenated and deoxygenated blood are going to end up mixing and then being ejected to the aorta and the pulmonary artery, so it's necessarily going to start off life as a cyanotic congenital heart defect.

In someone with tricuspid atresia, what you typically have, of course, is an atretic tricuspid valve and a hypoplastic right ventricle. You need to get that blue blood out of the right atrium, usually across an ASD [atrial septal defect] or a PFO [patent foramen ovale] into the left atrium, and then you have purple blood in the left ventricle that's going to eject to the aorta and the pulmonary artery.

There are a lot of different flavors of tricuspid atresia, actually. You can have tricuspid atresia with normally-related great arteries, transposed great arteries. You can have pulmonic stenosis. How you manage each of these babies is going to depend on their initial anatomy, but your primary goal is to make sure you have unobstructed outflow out of the right atrium and then you need to have a balanced amount of flow to the aorta and the pulmonary artery. You neither want to overcirculate the pulmonary artery with excessive pulmonary blood flow, but you also need to make sure you have adequate pulmonary blood flow, for example, in the baby with pulmonic stenosis so that you can get oxygenated blood back to the heart.

Perry: Gotcha. In a baby with tricuspid atresia, one of our primary goals initially is that there is essentially minimal or low blood flow to the pulmonary arteries, and so one of the first goals is to augment the pulmonary circulation. How might that be performed or accomplished in the neonate?

Krieger: Sure. If you have a baby that has inadequate pulmonary blood flow, either from tricuspid atresia or really almost any other cyanotic congenital heart disease, generally we augment that with a systemic to pulmonary artery shunt. The one that you'll see now most frequently is called a Blalock-Taussig-Thomas shunt, or a BTT shunt, and that's a shunt that's placed in between the subclavian artery to the pulmonary artery. It diverts blood from the systemic circulation into the pulmonary circulation to increase pulmonary blood flow.

Perry: Perfect, and I might just give a plug-in. There is a great movie about the Blalock-Taussig-Thomas shunt,

Krieger: That's right.

Perry: It's like an HBO special from some years ago.

Krieger: Yeah.

Perry: After we've done that, that's one of the first procedures that's done within like the first couple weeks of life. Correct?

Krieger: Yeah, you can keep the ductus arteriosus open early on with prostaglandins to augment pulmonary blood flow, but you can't send babies home that way. You would do a BTT shunt before you sent a baby home if you need to augment their pulmonary blood flow.

Perry: OK. We do that for our infant. They go home with a BTT shunt. What time frame, I suppose, are they going to come back for something else? Or what might be the next step in their surgical procedures?

Krieger: Right. The next goal starts to be trying to separate out the pulmonary and systemic circulation. This, in a shunted infant with single ventricle anatomy, they are going to be profoundly cyanotic, or they're going to be quite cyanotic, at least. As we move forward through the Fontan pathway, the next steps we take are going to be aimed at trying to get the blue blood to the lungs and the red blood to the body.

Somewhere around maybe 6 months of age, a baby may return for what's called a Glenn shunt, which is a SVC [superior vena cava] to pulmonary artery shunt, where the SVC is diverted to the pulmonary artery [PA]. What this achieves is it diverts exclusively blue blood out to the pulmonary arteries, which will decrease the amount of cyanosis that the baby has.

Perry: OK. If I understand correctly, these shunts, the SVC to PA anastomoses, that's a newer surgery, and maybe some decades ago possibly that palliation, or that anastomoses, might have been different for like a right atrium to pulmonary anastomoses?

Krieger: Sure. Yeah. What you're bringing up is the fact that the Fontan operation, which we'll be coming to, has gone through a number of different variations, starting in the early 1970s with just the atrium to the pulmonary artery. But as we'll get to later, I'm sure, that surgery's been abandoned for over 25 years. Now, typically, a baby will have this Glenn procedure, the SVC to the right pulmonary artery, classically, as an inter stage before the completion of the Fontan operation.

Perry: OK. Perfect. Let's go on to what the completion of the Fontan procedure looks like. Now we have a baby with tricuspid atresia, so no functioning right ventricle. We have had a BTT shunt implanted, and then we've subsequently moved to having an SVC now connecting to the pulmonary artery, called a Glenn. Now, what's the final procedure?

Krieger: Sure. Just to clarify, at the time of that Glenn, now that you have a good source of pulmonary blood flow, you no longer need the BTT shunt. If a baby has had a BTT shunt to start with, you would ligate that or take that down at the time of the Glenn operation.

Perry: Got it. OK.

Krieger: After the Glenn shunt, all the blood from the arms and head are going to the pulmonary artery, but all the IVC [inferior vena cava] flow is still going back to the right atrium, left atrium, left ventricle, and out the aorta, so this baby's still going to be cyanotic. As these babies turn into kids that are running around, moving around, and using their legs more, they are going to at some point likely develop limitations related to their ongoing cyanosis.

It's very center-dependent, but somewhere between the ages of 18 months and 4 years most babies will move forward for their Fontan completion. The Fontan completion is bringing the flow from the IVC up to the pulmonary artery as well, so that now all the blue blood is going to the lungs and all the red blood is coming back to the left ventricle and pumping out to the aorta and turning it into non-cyanotic heart disease.

There is a number of different technical ways you can bring that IVC blood up to the pulmonary artery. You can either use an interposition graft, something like a Gore-Tex interposition graft -- that would be called an extracardiac Fontan -- or sometimes you'd tunnel it up through the right atrium itself using the lateral wall of the right atrium and some synthetic material to separate it out, and that would be called a lateral tunnel Fontan.

Both of those are reasonable approaches. What you referred to earlier, the atrial pulmonary Fontan -- where you let all the blue blood go to the right atrium and then directly anastomose something like the right atrial appendage onto the pulmonary artery -- has now been , because those patients had so many arrhythmic complications.

Perry: Now to summarize what we have achieved, somewhere between the ages of 18 months and 4 years of age is we have an infant or a child where the SVC drains directly into the pulmonary arteries. The IVC is then tunneled either through the right atrium with a Gore-Tex graft or using the lateral wall of the right atrium into the pulmonary artery. Thus, we have passive filling of the pulmonary arteries, which then drains into the left atrium, into the left ventricle, which then pumps it into the aorta, thereby bypassing the right ventricle entirely.

Krieger: Yeah. The nice thing about the Fontan completion is that you've dealt with cyanosis. The other thing that's important, although maybe less intuitive to a lot of people, is until you finish the Fontan you have a major volume load on that left ventricle. You're putting much more preload and much more return to that left ventricle, which causes volume loading and can predispose to ventricular dilation and potentially remodeling and dysfunction down the road. You also need to get rid of that excessive volume load in addition to dealing with the cyanosis.

You achieve all those things, but as you pointed out, the penalty is that you're supplying the pulmonary circulation via passive venous pressure. If you only have single ventricle anatomy and you're using your single ventricle to pump your red blood to the aorta, that leaves no ventricles left over to drive blue blood through the lungs, so the Fontan is defined by the absence of a subpulmonary pumping chamber.

Perry: Perfect. I think that essentially wraps up the portions for the child and the surgical procedures. Are there any other comments before we then progress? We're going to fast forward to when we're seeing this patient as they then grow up into a young adult.

Krieger: Yeah. I think we glossed over a couple things, which I think are fine. It can be sort of exciting, that interstage time for the baby with the Glenn operation before they get their Fontan. Pediatric cardiologists and pediatric cardiac surgeons really do have their hands full with these babies up until the point of Fontan completion. There are usually very intensive programs to monitor these kids very carefully.

That being said, following the Fontan completion, most kids -- not all of them, but most kids -- do really well for long periods of time. Most kids with a Fontan who are 10, 12 years old, who have simple anatomy with a systemic left ventricle -- which isn't everyone -- but kids like this with tricuspid atresia tend to do pretty well.

They play. They can do sports. They do reasonably well at school, and they can lead relatively normal childhood lives. It's a fairly quiescent period, those school-age years and middle school years.

Perry: Gotcha. Also remembering, I think we also glossed over about how common this is. So this is a rare disorder and affects about 2 in 10,000 births just in terms of a univentricular physiology. That can be manifested in many different types, one of which is the specific case in which we're describing, a tricuspid atresia, although I think the most common type is a hypoplastic left heart syndrome, unfortunately, which is a much more complicated issue.

Krieger: Yeah. Right. If we're talking about, for example, our congenital heart clinic, single ventricle anatomies, far less common than ASDs, VSDs, tetralogy, coarctation, and things like this. But that being said, these patients are among the most complex and the ones that take up the most resources. While we maybe don't follow as many Fontan patients as we do tetralogy patients, we know virtually all our Fontan patients quite intimately because, as we'll be getting to, in adulthood many Fontan patients end up with considerable complications related to it.

Perry: Maybe another question, or next, would be about what age do things start to not go as well as they were when they were 10 or 12, when they were like playing and doing things normally? About what age do some of the first manifestations or issues with this abnormal physiology start to develop?

Krieger: Sure. A lot of it depends on how complex their single ventricle anatomy is. We started with a kid with tricuspid atresia, relatively straightforward anatomy as far as single ventricle anatomy is concerned. The left ventricle is the systemic ventricle, so that's pumping to the aorta. Mitral valve problems and aortic valve problems are relatively uncommon. A kid like this you'd expect to do pretty well certainly into their 20s, maybe into their 30s.

On the other hand, conditions where you have a systemic right ventricle that's supporting the circulation, kids who have concomitant valve dysfunction or maybe prosthetic heart valves as part of their initial surgeries, kids who had complications from their early surgeries, those kids may never quite be well during childhood. But someone like this, most of our Fontans with straightforward anatomy, single left ventricle, tend to do well certainly into their 20s and up to their 30s oftentimes.

Perry: Which sounds really good from like the complicated issues that they were born with at birth. But going back to what are the first manifestations of any complications from this?

Krieger: It depends. There are a number of different issues that can come up with the adult patient with a Fontan operation. But as we think about that, let's back up to the physiology about what's going on with the Fontan so that we can start to understand maybe why it would deteriorate as a patient gets older.

With a Fontan circulation, we've talked about a couple times now that it's central venous pressure that's driving blood through the pulmonary vasculature and into the left atrium. What that means is that your central venous pressure [CVP] has to be higher than your left atrial pressure. How much higher does it have to be? It has to be higher enough that it can overcome the resistance of the lungs and still deliver blood.

Now, if you're a pliable, healthy 80-pound, 10-year-old kid who's running around, the pulmonary resistance may be low, the left atrial pressure may be low, so the central venous pressure can be low.

Just to throw some numbers out there, let's say a normal left atrial pressure is 5. You have a little bit of a pressure drop across the pulmonary vasculature. A kid like that could be walking around with a Fontan pressure or central venous pressure of 8 or 9, and that's awfully manageable. But as kids get older and as we turn into adults, a number of things happen.

One is our left atrial pressure just naturally goes up a bit as our ventricles become less compliant. We have some stiffness. Perhaps you have a little mitral or aortic valve dysfunction that's crept in over the years, so maybe your left atrial pressure goes up to 10, which isn't a bad left atrial pressure. You, as a cardiology fellow, would be delighted to see most of your patients have a left atrial pressure of 10.

Perry: Sure.

Krieger: But let's say the left atrial pressure is 10 or 11. Pulmonary vascular resistance maybe goes up as we get a little older. We get obese. We get sleep apnea. Maybe someone's taken up smoking. You put that on top of the fact that they have some restrictive lung disease from their thoracotomy that they got with their BTT shunt, their sternotomy they got at the time of their Glenn, and their Fontan operation, now the pulmonary resistance is higher and the pressure drop across the lungs is higher. So even a healthy 25-year-old Fontan can't have a CVP of 8 or 9 anymore. They are going to have a Fontan pressure, a CVP of 15, 16.

Not terrible, but that's not a normal central venous pressure, and that starts to have consequences as you get older. The consequences it has, we'll go through them, I think, one by one. But one major one is it puts passive congestive pressure on the liver, and we start to see liver dysfunction as these kids get older.

Two is it becomes more difficult to augment cardiac output without having prohibitively high Fontan or central venous pressures. Because the more you increase the flow, the more you increase the pressure drop across the lungs; so they may have cardiac output limitations.

As Fontan pressures go up and up and up, the body finds ways to deal with that. The body will form natural decompressing collateral vessels to decompress the Fontan, to try to find a way to get the blue blood from the systemic veins back to the heart while bypassing the resistance of the pulmonary circulation.

That might be through pulmonary AVMs [arteriovenous malformations] that would bypass the pulmonary circulation. That can be from decompressing systemic venous to pulmonary venous collateral vessels. We call those veno-venous collaterals.

Both of those will decompress the Fontan, but both of those will lead to cyanosis, so you might see these patients developing and dropping in their oxygen saturations from their baseline sat of 97%, 98% down to 92%, 93%.

Perry: Gotcha. To summarize or recap, as you're getting older, there can be issues with pulmonary resistance for many of the issues that you mentioned. It would be a sleep apnea, which I think is fairly common in patients with adult congenital heart disease, whether they become smoking, or they develop obesity, and they have some other issues with relaxing and just filling pressures in their left ventricle. Thereby they get this increase of pressures transmitted all the way back to their central venous pressure, which is the pressure needed to overcome their pulmonary pressure since there is no longer a pumping chamber below to overcome that.

Krieger: That's right. The only way you can maintain cardiac output in the setting of these problems is to increase your Fontan pressure or shunt and both of those things end up having serious consequences.

Perry: Now, to be clear, when we're talking about central venous pressure there is an issue when we talk about examining these patients because generally speaking I can look at someone's neck and estimate their central venous pressure. However, in a patient with a Fontan, estimating their central venous pressure based off of their internal jugular veins and their venous waveforms is no longer useful, correct?

Krieger: Yeah. It's actually extremely difficult to use things like neck veins and even IVC diameter by echo, the traditional tools we use to estimate CVP for Fontan patients. It just hasn't really proven to be reliable. Generally, you need to look at the patient as a whole.

Patients that are volume retaining and have ascites and edema, those patients have markedly elevated central venous pressure. But it can be difficult. All Fontan patients will have elevated central venous pressure by the time they get to adulthood. The question does become how high, though.

Perry: Sure. Let's put the situation where we're seeing this patient in their late 20s, early 30s, and we've noticed that on their vital signs that their resting oxygen saturations are now lower than they used to be. Some of the complications that you've mentioned that I should be aware of are these veno-venous collaterals?

Krieger: That's right. That's from the systemic veins like the SVC or the innominate vein draining straight down to the pulmonary vein, so that will put blue blood back into your left atrium.

Perry: And that's a development from these elevated pressures in the body figuring out a way to decompress?

Krieger: Yes, oftentimes. Sometimes you can have someone who's got great Fontan pressures, but they can re-cannulate these veno-venous collaterals that probably existed embryologically. In those cases, it can be reasonable to think about occluding the veno-venous collaterals in the cath lab. But if those are there because of very high Fontan pressures and they are the body's way of trying to decompress, then coiling them or occluding them ends up, first of all, being ineffective because the body will just form more, but also limits the patient's cardiac output because that blue blood returning to the left heart is helping to supplement the cardiac output.

Perry: Another thing I should be concerned about are pulmonary AVMs then?

Krieger: They are usually not large AVMs. These patients will sometimes have diffuse AVMs. They are not like the big vessels that you see in HHT [hereditary hemorrhagic telangiectasia], for example. They are these diffuse, very spidery, almost amorphous vessels that you could really only see on direct pulmonary angiography.

Perry: Gotcha. OK, let's circle back to these issues with elevated central venous pressure. Like in other patients with heart failure, they can then develop cirrhosis as a result from that because of this elevated pressure. How do you normally screen for those patients? Cirrhosis is something much more subtle that doesn't present as manifesting, "Oh, yes, now I'm hypoxemic." Or, "I'm having exercise intolerance." But it's just a slowly progressive issue.

Krieger: Yeah. For the last 15-plus years, there has been a real recognition that liver dysfunction is a major source of morbidity for adult Fontan patients. It's now universally understood throughout the adult congenital heart disease community. The prevalence of liver dysfunction after a Fontan operation is much higher than you see in acquired heart failure.

Our program, for example, now has a standardized surveillance protocol for all adult Fontan patients where we do screening for subclinical liver disease. What that means for us is, once a year, for all our adult Fontan patients, we get a full set of liver function labs. We do imaging, either with CT or ultrasound, of the liver. And we do surveillance for hepatocellular carcinoma [HCC], which we used to think was rare, but now it turns out there is case report after case report and now large-ish case series of HCC in patients with Fontan. For that, we screen using AFP [alpha-fetoprotein] and annual imaging.

Perry: Interesting. There is also additionally something else, while we're talking about GI complications, of protein losing enteropathy. These patients can get nutritional issues as well. Is that...?

Krieger: Yeah. Protein losing enteropathy is a problem for Fontan patients. Real pure protein losing enteropathy, or PLE, is probably seen a little bit more commonly in the pediatric population, but we have a handful of patients on the adult side.

You lose protein through your GI tract. It's characterized by hypoalbuminemia, ascites. It can really mimic right heart failure and cirrhosis, and you have to have a suspicion. If you have a hypoalbuminemic patient, then doing 24-hour fecal collections to look for protein is the right screening test for that.

Perry: Now, a couple of things when I was rotating on the adult congenital heart disease service that stood out in particular to me was, one, the risk of developing arrhythmias... and we can talk about what specific arrhythmias are most common. But also, I was struck by how devastating some of these more "benign" arrhythmias turn out to be, like patients with atrial fibrillation just don't tolerate that very well.

Perry: Yeah. Particularly with the old-style Fontans, these atrial-pulmonary Fontans, patients would get these massively dilated right atria in response to having a high pressure right atrium for years and years. They would develop atrial arrhythmias that we call interatrial reentrant tachycardia, or IART.

They are really a form of atrial flutter, but they look different from your typical flutter that you'd see in a patient with acquired heart disease. In a typical flutter, the flutter goes at 300 beats a minute. It often conducts 2 to 1, so these patients will classically present in 2 to 1 flutter at 150, and we're all used to recognizing the saw tooth pattern, et cetera.

Perry: Yeah.

Krieger: After a Fontan, it's a lot more subtle. Now, one of the things that happens is because the atria are bigger, the reentrant pathway is longer, so the flutter goes slower. The flutter rate might only be 200 beats a minute, and then when a patient goes to 2 to 1, their ventricular rate might only be 90 or 100.

That can be tricky, because we all know that detecting 2 to 1 flutter is hard because every other flutter wave is buried in the QRS. It can look a whole lot like sinus tachycardia at 90 or 100, and could sort of blow right by it, but you need to have a clinical suspicion for that. If their heart rate's higher than it typically is, if the P waves on their ECG look different than they have in the past, you ought to be suspicious.

We screen our patients with ECGs each year and patients with a Fontan in atrial tachyarrhythmias do not do well, so there's this large body of research in acquired heart disease that suggests rhythm control may be effective to rate control in patients with afib or a-flutter. That is just not true in patients with a Fontan.

Perry: Sure

Krieger: Almost universally we'd prefer rhythm control in that situation. A cardioversion may work. They often will need an ablation.

The other thing I'll quickly say is that any time you have a patient presenting with a new arrhythmia and congenital heart disease, you need to be very suspicious that might be your warning sign of some hemodynamic change and start to ask yourself, "Gee, I wonder why this patient who's been in sinus all these years all of a sudden went into some atrial tachyarrhythmia?" And use that to do a comprehensive evaluation with echo and consider catheterization to try to unmask what hemodynamic perturbation led to the patient developing their arrhythmia.

Perry: Gotcha. We had indicated earlier that there was an older style of Fontan, this atria to pulmonary anastomoses versus now the modern surgery is to do SVC to PA and IVC to PA. Are the risk of atrial arrhythmias any less with these more modern techniques -- when I say modern, like in the last couple of decades -- than they were previously?

Krieger: Yeah. Oh, without a doubt. With an atrial pulmonary Fontan -- and there's vanishingly few patients with atrial pulmonary Fontan left nowadays -- what's happened is they have either been converted to a more modern Fontan with a subsequent Fontan revision, or unfortunately, they have died. There are certainly some but there's fewer and fewer.

Those patients had a huge incidence of atrial arrhythmias, and it was a real problem. That's what led to the atrial pulmonary Fontan being abandoned and moving to this either a lateral tunnel or extracardiac Fontan. The extracardiac Fontan, where you resect a lot of the atrial tissue and bring up a Gore-Tex tube graft, has the lowest rate of atrial arrhythmias.

Lateral tunnels are still much better than the atrial pulmonary Fontan was. But with the extracardiac Fontan, if you do have an arrhythmia, there is not an easy path back from the systemic veins back into the heart if you ever did want to do an ablation. That's one of the challenges that you might deal with. It's not so easy to get from the veins back into the heart.

Perry: Yeah. OK. Another feared complication is pulmonary emboli. Pulmonary emboli are already to be feared within the general population, but you can kind of get used to it as a resident of seeing, as a trainee seeing, just like pulmonary emboli after pulmonary emboli, the little small ones. They are like, "Oh, we just need to anticoagulate them," and most people do just fine. Maybe they came in with a little sinus tachycardia sort of thing. Why are they so devastating in patients with Fontan?

Krieger: Sure. A real and big pulmonary embolism would be a major problem for a Fontan patient because, as we've pointed out, there is no ventricle to pump blood through the lungs. Anything that acutely raises the impedance of getting blood through your lungs can only be dealt with one of two ways, either by increasing central venous pressure or by dropping your cardiac output. Since it's awfully difficult to acutely raise your central venous pressure, generally what happens is they drop their cardiac output and it could be a devastating event.

Fortunately, real thromboembolic massive PE events are relatively uncommon in Fontan patients. They do happen. Fontan patients have chronically elevated central venous pressure. They have venous varicosities. That's been well documented, so it certainly can happen. But likely even more common than the acute thromboembolic event is in situ thrombus formation. There is some early pathology studies and early nuclear medicine studies that documented an alarmingly high rate of in situ thrombus in Fontan patients. I think no one quite knows exactly how high the prevalence really is.

I'd like to take a second just to talk about how unbelievably hard it is to diagnose a pulmonary embolism or pulmonary thrombus in these patients. Because this is an area where I often see mistakes being made in the emergency department, or mistakes being made when someone has a suspicion for a PE, they go down a typical pathway, so maybe...

Perry: Yeah. You're going to get a CT PE scan.

Krieger: They're going to get a CT PE scan.

Perry: Which relies a lot on your filling time and the timing sequence from your venous filling.

Krieger: Absolutely, and CT PE scans, if done according to a standard protocol that you might use for someone walking in the ER with a suspected PE, just don't work for a Fontan. They don't work for a variety of reasons.

But if you think about what happens to a patient with normal cardiac anatomy with a CT PE scan is you inject a bunch of contrast in the arm, it goes to the right atrium where it mixes with a bunch of unopacified blood that's coming from the IVC, and then it all homogenizes in the right atrium, homogenizes in the right ventricle, and then you get a first pass in the pulmonary arteries where they're nicely opacified because you get your timing just right.

That doesn't happen in a Fontan. In a Fontan you inject contrast -- let's say in a peripheral IV in the arm -- it comes down, and that contrast goes straight to the pulmonary artery. What happens is most of that contrast is going to stream right into the right pulmonary artery.

Simultaneously, you have a bunch of blood coming up the IVC that's got no contrast in it that's going to mix or collide with the blood coming from the SVC and preferentially stream to the left pulmonary artery, so you're going to have a huge lack of contrast in the left pulmonary artery. Time after time, we've seen that interpreted as a patient with a proximal left pulmonary artery embolism or you get streaming artifact into the right pulmonary artery from the unopacified blood. It really makes it incredibly difficult.

There are a couple approaches you can take. Probably the best way to deal with this, if you want to rule out a PE in a patient with a Fontan, is to get a couple acquisitions. You get an early acquisition and then a late acquisition at about 45 seconds after your injection, usually, maybe 25 or 30 seconds after your first acquisition.

Double the radiation. You don't have isolated contrast in the pulmonary arteries and no contrast in the pulmonary veins, which is what radiologists like, but you're getting that second pass. The contrast brightness isn't quite as good, so they're not beautiful images, but it's a lot better than just what you get on that first pass, which is full of streaming artifact.

Perry: Sure. OK.

Krieger: You have the same problem with the VQ scan, by the way. That's macroaggregated albumin, so you inject it in the arm. It mostly goes into the right pulmonary artery just because of streaming, and then it lodges there. That's what the macroaggregated album in a VQ scan does, so it never makes it out to the left lung. It looks like you end up with a big perfusion defect in the left lung, too.

If you really need to know for sure, you can get that CT with the late phase, but direct pulmonary angiography done by interventional cardiology with expertise in congenital heart disease, either a pediatric interventional cardiologist or an ACHD [adult congenital heart disease] interventional cardiologist, is what you need to get.

Perry: Gotcha. Interesting. Super interesting. When I was recently on service, we also had a patient who was admitted, Fontan physiology, who was pregnant and I think that's in part interesting because I think some centers just say, "No pregnancy." Or it's like such a high-risk event, or some centers, I think, are a little more lenient in their counseling of patients. What are your thoughts about pregnancy in patients with Fontan?

Krieger: It's certainly a hard one. I think saying, "No, no Fontan patient should or can get pregnant," is not the right approach. First of all, you can't stop patients from getting pregnant, even if you want to. You can counsel patients about risk; but patients won't always listen to you ,and you need to be aware of that. You need to counsel patients honestly about risk.

Fontan, no doubt, is a high-risk cardiac condition for pregnancy, but it's not the highest. It's not like pulmonary arterial hypertension, or severe symptomatic aortic stenosis, or Eisenmenger syndrome. It's not one where we say off the bat, "Absolutely not. Don't do it."

It is one where we would counsel them that, first of all, you need to counsel them on the risk of congenital heart disease in the offspring, which is, of course, going to be higher than the general population, and they need to know about that. Then you need to counsel them that there is real risk of some form of clinical deterioration during pregnancy, but most of those risks are manageable.

We've had dozens of women here with Fontan have successful pregnancies. Now, some have volume retention or heart failure, and some have had arrhythmia, but the majority have had reasonably successful and uneventful pregnancies.

I think that pregnancy while they're still young and healthy is preferable. I would be anxious about a pregnancy in a already decompensated or marginally decompensated Fontan patient. But we do not counsel specifically against pregnancy in well women with a Fontan. You just need to really tell them what you think the risk is, and we always talk about shared decision-making. But if there's ever a time for shared decision-making, it is really when it comes to counseling about pregnancy.

Perry: We talked a lot about these complications and, as we've mentioned, the load, then, on the remaining ventricle could be quite severe. I think almost universally at some stage these patients will develop symptoms of heart failure. Their EF will start to drop at a lower age. Maybe not necessarily true?

Krieger: Well, I agree. Fontan patients, as they enter their 30s and early 40s will, not invariably but often, develop symptoms of heart failure. Certainly not invariably. I have plenty of quite healthy, active Fontan patients in their 30s. But commonly, they'll end up with heart failure and we use heart failure or the term Fontan failure a little bit generically.

Sometimes it's due to ventricular systolic dysfunction, and that we understand pretty well, but oftentimes it's not and we don't always understand all the mechanisms of what leads to the physiology or the phenotype of heart failure in Fontan patients. Sometimes it's ventricular dysfunction, systolic or diastolic. Sometimes it's valvular. But sometimes as those left atrial pressures creep up and the pulmonary resistance creeps up and Fontan pressures start to approach 18, 19, 20, 21, those patients don't feel very well and they often don't do well. It accelerates their liver disease. They may get ascites that could be related to cirrhosis or just related to right heart failure. And so the mechanisms of heart failure in Fontan are variable and not always well understood, but they're almost all characterized by high Fontan filling pressures.

Perry: When do you start thinking about heart transplant for these patients?

Krieger: Boy, that's probably the hardest question in all of adult congenital heart disease. In general too late, and I don't mean that disparagingly. I just mean that we need to be thinking about heart failure and heart transplant in these patients from a really young age. We should be starting to think about it when they're kids and teenagers -- not starting to do it, but starting to think about it -- because we see time after time that adult patients with Fontans develop heart failure, frequently in their 30s, and transplanting a Fontan patient is not straightforward.

For starters, it needs to be done at a center that's really good at that. Even at really good centers with big ACHD programs and big transplant programs, there is no denying the fact that Fontan transplant is a high-risk transplant. It's high risk for a variety of reasons.

First of all, they have necessarily had multiple cardiac surgeries. They are often alloimmunized from multiple transfusions. They have multi-organ dysfunction with, often, clinical cirrhosis and the decision about whether to do a heart transplant or a combined heart/liver transplant is a really tough one. They have multiple comorbidities with renal dysfunction.

They have high venous pressure. We have talked about that, but that makes these patients bleed. They have collaterals all over the place, so they're bloody operations. It is not a surgery to be done casually. It's not a surgery to be done by somebody who dabbles in caring for Fontans, and it's not something to be undertaken lightly. Operating on really decompensated Fontan patients who have extensive collaterals and extensive cirrhosis is high risk. Even in good centers, 30-day mortality can be over 20%.

Perry: Wow.

Krieger: There are a few places around the country that have done really well with Fontan transplants, but it really needs to be highly regionalized.

Perry: OK. Wow. We've talked a lot about, maybe, where we stand with Fontans. What are the future directions or areas of active research that are ongoing that you know of?

Krieger: Sure. I think if you talk to adult congenital heart disease providers, we're all hopeful that there will be something "better" than the Fontan. And people have dabbled in advocating for leaving them shunted, maybe not even completing the Fontan. I think most pediatric cardiologists don't think that's a feasible long-term solution, that these cyanotic kids really don't thrive, but we currently don't have a great way of avoiding the ravages of the Fontan as kids get older.

There are some groups that are experimenting with bioengineering and tissue engineering to see if there's a way to create pulsatility in the Fontan conduit that would augment pulmonary blood flow. That would be really appealing, that you wouldn't be dependent just on central venous pressure, but that's nowhere near clinical trials quite yet. That's all preclinical.

A lot of the research is still going towards understanding the phenotype of Fontan failure, understanding what medical therapies might work to help improve outcomes in Fontan patients, understanding when to bring patients for transplant and how to select patients more appropriately for transplant so that we can get acceptable transplant outcomes.

We should probably touch a little bit on medical therapy for Fontan patients. We don't have the depth of information that we have for treating heart failure and acquired heart failure with things like ACE inhibitors and beta-blockers.

We don't have evidence to suggest that those treatments are specifically beneficial broadly for Fontan patients. They may be effective for those with ventricular dysfunction, but not broadly for Fontan patients.

The treatments that we know, or the treatments that we think are helpful for Fontan patients are anticoagulation for Fontan patients with risk factors for thrombosis and probably antiplatelet therapy for everyone else. You want to prevent thrombotic events.

Treating rhythm disorders, that's real important. You want to maintain activity, aerobic activity, exercise, maintain a healthy body weight, avoid smoking. All these things matter for everyone, but they probably matter more for Fontan patients. There are even clinical trials showing that breathing exercises and yoga are beneficial in improving Fontan outcomes.

Then there has been a couple of clinical trials on pulmonary vasodilators in Fontan patients. We've talked about the importance of needing to keep pulmonary resistance low, so a couple of studies have asked the question of what if we gave pulmonary vasodilators like endothelin receptor antagonists or PDE5 inhibitors, things like bosentan [Tracleer], sildenafil [Revatio], tadalafil [Adcirca], ambrisentan [Letairis], et cetera, to these patients? Even though they don't have pulmonary vascular disease, what if we could lower their pulmonary resistance?

Perry: Which is starting it on them while they're healthy?

Krieger: Starting on it while they're healthy.

Perry: Without symptoms to prevent... ?

Krieger: Yeah, and there has been a couple of clinical trials that seemed promising. They seemed promising. There was one randomized, double-blind, placebo-controlled trial of bosentan in Fontan patients, and it improved CPAP performance. It improved New York Heart Association functional class. This was given broadly to Fontan patients, not just those with pulmonary vascular disease.

That's in there, but we need to remember these medicines are awfully expensive. They are tough to get approved for this indication, because they don't have an FDA approval and insurance companies often fight tooth and nail with it.

Perry: In terms of anticoagulation, now that you they mention that, because a PE can be so devastating it's such an issue. You said patients with another risk factor for being hypercoagulable?

Krieger: Yeah. I'd say all Fontan patients should probably be on an antiplatelet agent or anticoagulation. The simplest way to think about it is your best Fontan patients should be on antiplatelet agents, and pretty much everyone else should be on anticoagulation.

That means Fontan patients with a history of arrhythmia, Fontan patients with a history of clot, Fontan patients with heart failure. Those patients. Fontan patients even with moderate ventricular dysfunction, I think you could make a case for using anticoagulation.

Perry: And including those with cirrhosis, probably, in that?

Krieger: Anticoagulating cirrhotic patients is always hard, higher risk, but yes. If they don't have varices and coagulopathy and actual bleeding events, then I'd strongly consider it.

Perry: Got it. OK. We've hit a lot of different things. I appreciate your time. Any last tidbits or pearls that we have missed or overlooked or glazed over?

Krieger: No, I think we've hit on a lot, other than that Fontan patients, all of them, should be cared for in a specialty ACHD center. I think that's a must. I think that the symptoms can be subtle, so the suspicion for deterioration needs to be high, which means that you need to watch them real carefully. You need to watch them with echoes, you need to get exercise testing on them, you need to do lab surveillance. You need to be pretty proactive in these patients.

Perry: Got it. OK. Again, I thank you for your time.

Krieger: Thanks so much. It was good to be here.

is a cardiology fellow at the University of Washington Medical Center in Seattle.