HAROLD J. FARBER: How do children who have pulmonary hypertension (PH) present?
GEORGE B. MALLORY: One of the great challenges of PH in children is the wide range of age groups, beginning with infants, who most commonly present with very dramatic symptoms, and most often in a fairly obvious clinical context, and then extend our care to older children and adolescents, who are probably the group that readers will be most interested in learning about and who are more like adults with idiopathic hypertension and can present with very subtle symptoms.
D. DUNBAR IVY: I think some of the symptoms of PH are very subtle in children, as George mentioned. But some are perhaps a little more striking. For example, children with idiopathic PH are more likely to present with syncope than adults. We also see patients who have shortness of breath, particularly in adolescents.
I think we also certainly need to think about the former premature infant who has bronchopulmonary dysplasia (BPD), in whom again the symptoms of PH can be subtle. For example, we and others are seeing children who have a long course of prematurity and BPD. They develop multiple viral infections, and can present with severe PH.
BRIAN D. HANNA: One of the things that we miss is that children, whether adolescents or premature infants, are not so much presenting with PH as with right ventricular (RV) failure. An adolescent can come in with exercise-induced syncope or presyncope and venous congestion in the legs, which is more like what one might see in a 60-year-old lady, but I have seen 4 kids in the last 6 months who have been referred by pediatricians to our cardiology service for such indications of RV failure.
GEORGE B. MALLORY: Would you not agree that many PH patients have been put on bronchodilators at the time of presentation, which suggests that pediatricians and pediatric pulmonologists who are seeing patients with breathlessness at a somewhat earlier stage may be jumping to a diagnosis of asthma as the most common cause of dyspnea and then going directly to the default treatment for asthma? I would be interested in what others might suggest as helpful signs and symptoms for differentiating PH in a child from the first manifestations of exercise-induced asthma.
FELIX SHARDONOFSKY: That is an interesting question. I think that in part, this confusion results from a superficial evaluation of the patient, and certainly exercise-induced asthma is a very common condition and associated with a postexertional cough and wheezing and airway obstruction. On the other hand, exercise-induced dyspnea associated with PH is a completely different situation. The patient typically has severe dyspnea during exertion, and chest pain, and sometimes syncope, and looks “clammy.” That should remind us that in many cases PH is also associated with air flow limitation, the mechanism of which is probably quite complex.
BRIAN D. HANNA: The idea of airflow limitation and exercise-induced asthma as a sign or symptom of PH, and that adolescent patients often come to us when they’re already taking bronchodilators, is very much of a problem. I have yet to find a way to tell the pediatricians in such cases what they missed. It is usually missed on a physical examination as opposed to an electrocardiogram (ECG), an echocardiogram, or a chest X-ray. But I think that even with the chest X-ray I report more positive findings for PH than do the radiologists. Their problem is that they answer the questions that they’re asked, and don’t do a differential diagnosis.
HAROLD J. FARBER: The initial presentations of PH, which is a rare disease, are very similar to common problems; we get lots of children coming in with chest pain and shortness of breath and very few with PH. What are the things that should make us start looking for PH?
FELIX SHARDONOFSKY: I think that on physical examination, the RV findings in PH include a loud, prominent pulmonic second sound, and a chest X-ray film shows a prominent pulmonary artery. Those are simple findings that should make one think about possible PH. Certainly PH is a rare condition, but on the other hand, the patient who presents with chest pain and dyspnea should be studied systematically, and an echocardiogram, and even an exercise test should be part of the evaluation of a patient who comes with symptoms of dyspnea or exercise limitation.
D. DUNBAR IVY: Let’s not forget the ECG. I think that along with its relatively lower cost than echocardiography, ECG is fairly sensitive, especially in combination with a good physical examination. However, the ECG is not 100% sensitive.
BRIAN D. HANNA: I must say that I am not as enamored of ECG as I would like to be, especially in the younger child with BPD or chronic lung disease or sickle-cell disease. I think that right axis deviation on the ECG is not an uncommon finding in the pediatric population. RV hypertrophy or an rSR prime pattern is relatively common, and right atrial enlargement is not as common in PH as I would like it to be. There is talk about relatively low sensitivity and specificity of ECG in adults, but I am not sure it is much better in the pediatric population. But on the other hand, I’m also really not certain that echocardiography is as sensitive and specific as we’d like it to be.
GEORGE B. MALLORY: I think that one of the reasons we have been spending so much time on diagnosis is that it is a crucial issue for pediatricians and pediatric pulmonologists, and for some pediatric cardiologists as well. Over the past few years, as I have gotten more involved with children with PH, I have considerably refined my own skills in physical diagnosis, and I’m concerned that most pediatric pulmonologists do not simply put their hand on a child’s anterior chest to feel for an RV heave, and their cardiac examination is also quite superficial.
I agree completely with Felix on this. A patient who comes in without dramatic wheezing and without classic asthma attacks but with a symptom complex of dyspnea and exercise intolerance should prompt the clinician to think that “this is a patient in whom I really have to do a careful physical examination and consider the possibility of PH.”
I also think that even though airflow limitation may occur in children with PH, their response to asthma therapy will be unsatisfactory and they will continue to have dyspnea and near syncope, and I therefore think that Felix’s recommendation for a systematic evaluation and early follow-up is critical in these patients.
D. DUNBAR IVY: I completely agree. I think we all agree that the presentation of PH in children is subtle. The physical examination findings are subtle, and our initial screening tests are not perfect, which makes this a difficult diagnosis. I agree that if you have a patient whose presentation doesn’t fit the classic pattern of asthma, you should consider an echocardiogram.
BRIAN D. HANNA: One of the questions we need to answer is whether in pediatrics a valid diagnosis of PH has to be based on a longer term of symptoms than in the adult population. The clinical literature points out that in adults, a long time can elapse from the beginning of symptoms to the time that a diagnosis of PH is made, and a further period can then pass from the time that the patient is told of the diagnosis to the beginning of therapy. My feeling is that in pediatrics we do a better job than that.
GEORGE B. MALLORY: It could also be argued that the child who becomes symptomatic with PH at age 8 often has a far more accelerated course than the lady whose PH diagnosis is made at the age of 35. At least once a year we get a call from a colleague about a child who is terminally ill with PH and in an intensive care unit asking whether our lung transplant center can accept the child for lung transplantation. In that situation it is usually too late to do anything, and unfortunately it is not rare for children to have very advanced disease by the time their clinical manifestations finally lead to a diagnosis of PH.
FELIX SHARDONOFSKY: I don’t think that we have sufficient data about pediatric patients to establish a time from the beginning of symptoms to the diagnosis of PH. From my own experience I think that depends on the age of the patient, the context of the hypertension, and the presenting symptoms. Often, the patient presents with syncope, which is a very dramatic presentation, is referred to a neurologist with a presumption of seizures, and we don’t get to treat that patient. Only eventually is the patient seen by a cardiologist or a pulmonologist.
On the other hand, there are patients who have dyspnea on exertion. I can remember instances of patients who were 6 or 7 years old and had a 3- or 4-year history of inability to participate in sports or poor physical performance, and who were ultimately diagnosed as having PH. So I think that a long time can pass between the presentation of symptoms of PH and its diagnosis.
HAROLD J. FARBER: There are atypical presentations of pediatric PH. With which ones should we be familiar?
GEORGE B. MALLORY: Our last 2 newly diagnosed patients presented to our emergency room with seizures and had a full neurologic workup, and in both of these cases it took more than 2 months before PH was considered as a possible underlying diagnosis. When I searched the literature about this, I found that it is not really well articulated and PH is a very uncommon cause of initial seizures. Therefore, once again, PH requires a careful history as well as physical examination for a prompt and accurate diagnosis. And I think that pediatric neurologists need to know that there is a differential diagnosis for children who are having seizures, especially when their evaluation does not show abnormalities on imaging or electroencephalography (EEG).
D. DUNBAR IVY: I agree. As you said, syncope, seizures, and failure to thrive in an infant can all be atypical presentations of PH.
HAROLD J. FARBER: How we should screen for pediatric PH? What are the pluses and minuses of electrocardiography and echocardiography? When should we turn to cardiac catheterization?
D. DUNBAR IVY: Continuing from our initial discussion, I don’t think that there is any perfect screening test with great sensitivity and specificity for pediatric PH, and that includes electrocardiography and echocardiography in the patient with BPD. That is why a high index of suspicion and a good physical examination are crucial to detecting pediatric PH. My own group and others have written articles pointing out that echocardiography may not always predict the presence or severity of PH in children with BPD, and that some patients may therefore require cardiac catheterization.1
BRIAN D. HANNA: The recommendations of the World Health Organization and of the 2008 Dana Point conference on PH with respect to the kinds of tests that should be done for its diagnosis and staging are as appropriate in pediatrics as elsewhere.2 What we don’t know are the sensitivity and specificity of echocardiography for making a diagnosis of pulmonary vascular disease in pediatric patients, which is different from saying that the diagnosis depends on a mean pulmonary artery pressure in excess of 25 mmHg at rest. Likewise, if cardiac catheterization showed a cardiac output of 3 L/min and a relatively low but abnormal pulmonary pressure or transpulmonary gradient, and you increased the cardiac output by pharmacologic means or by exercise, you might end up with supersystemic pressures.
I therefore think that in terms of answering these questions, we are truly neophytes. We don’t really have a clue as to the sensitivity or specificity of any of these tests.
FELIX SHARDONOFSKY: I read an article in which Dunbar compared the data from cardiac catheterization and echocardiography in infants.1
D. DUNBAR IVY: Yes, we found that we were not very good at predicting the level of PH by echocardiography in patients with BPD. The difficulty was that we did not do all of the echocardiograms and cardiac catheterizations at the same time in that study. Echocardiography in the absence of a reliable tricuspid regurgitation jet can underestimate the presence of pulmonary. There may be subtle signs, such as interventricular septal flattening, which I do not really believe reflect the absolute level of PH. It is in such patients that cardiac catheterization is valuable.
I would also point out that echocardiography is not very good in detecting some of the distal disease that we see in PH. It is not highly accurate in reflecting pulmonary vein disease, which we are discovering with increasing frequency, as has been shown in a recent manuscript3 and from our own experience.
GEORGE B. MALLORY: I would like to address that issue a bit differently. I think that in some clinical contexts, particularly in infancy, the diagnosis of PH is very straightforward, and I would suggest that in the young infant with congenital diaphragmatic hernia, and in some very small premature infants with BPD, the diagnosis, with clinical suspicion and the use of echocardiography, is relatively straightforward.
The problem is that infants with congenital diaphragmatic hernia and those with BPD have a tendency to show improvement in lung function and a decrease in PH as their lungs grow, with the result that most of these patients do not undergo cardiac catheterization in our institution.
An important issue is what to do with a child who is still on a ventilator after 6 months with BPD or CDH and who continues to show significant PH by echocardiography? I know from colleagues at smaller centers that their cardiologists are often very reluctant to consider right-heart catheterization in these patients. But I often quote Dunbar’s article1 to the effect that in a patient with ongoing morbidity, the echocardiogram may not be entirely satisfactory, especially as we try to make decisions about treatment. The implications of the PH in terms of the costs of treatment and the outcomes for the child and family are huge.
FELIX SHARDONOFSKY: Pulmonary hypertension in infants with BPD could result from diverse disease processes including developmental abnormalities of the lung circulation and pulmonary vascular remodeling as well as congenital heart or vascular lesions such as pulmonary veins stenosis, presence of systemic to pulmonary collateral vessels or left ventricular diastolic dysfunction. Therefore, a systematic diagnostic evaluation of these complex patients including invasive hemodynamic studies is warranted.
HAROLD J. FARBER: What do we think about the prognosis and treatment for infants and children with PH? Who should be treating these children?
BRIAN D. HANNA: The recommendations for adults from the 2008 Dana Point conference were that they go to a PH center. On the basis of what we have discussed I think that especially for small infants you need to have someone who has worked with this on a daily basis, and I therefore think that it is imperative for the pediatric treatment group to define the attributes of a PH center for the pediatric population and the kind of expertise that it needs to have. I think that we are lagging behind our adult-treatment colleagues on that.
The issue of how to treat and who to treat for pediatric PH is one that I think will be defined in the future. For babies with pulmonary hypoplasia, I think that we are going to find that pulmonary vasodilators may improve both alveolarization and the potential for lung growth, although the data supporting that are not yet concrete.4 For severe right heart failure in an adolescent, the mainstay of treatment has always been intravenous prostanoid therapy.
D. DUNBAR IVY: I agree with Brian, and I would add that the prognosis clearly depends on the underlying disease. As was pointed out, I think that the pulmonary vascular disease in most of our patients with diaphragmatic hernia or BPD tends to improve as they grow, if they don’t have any other anatomic abnormalities, such as pulmonary vein disease.
In the past, the prognosis for the child with idiopathic PH was dismal, and in the 1980s the survival for such children was probably around 10 months, according to some data from the National Institutes of Health, but the numbers were limited. I think that today we can say that the survival for children with idiopathic PH is probably 80% at 5 years according to some of the data in national registries such as that in the United Kingdom. Although we have seen an improvement in the overall prognosis for children with PH, this really depends on the underlying disease.
I would underscore the importance of treating any treatable cause. I would also like to add that Brian has done an excellent job of developing a lung hypoplasia clinic. The key to managing the problem of pediatric PH is multidisciplinarity. Almost all of these patients have more than one problem. They may have congenital heart disease; they may have pulmonary problems; they may have hematologic problems; and they may have gastrointestinal issues. We even need to involve rheumatologists for some of these patients, such as those with connective tissue disease.
GEORGE B. MALLORY: I agree completely, and that is one of the reasons why I have reached out to individuals like Brian and Dunbar and other members of the pediatric PH community, who come from different specialities and practice traditions, not only because of their rich experience, but because the cardiologist, to take but one example, clearly brings a different perspective, skill set, and approach that benefit the pulmonologist caring for such patients.
If you look at identified treatment centers for PH in adults, most are headed by pulmonologists. There are many fewer self-identified centers for treating pediatric PH, and most are directed by cardiologists. I think this is an important area for collaboration not only in clinical care but also in clinical research. Our patient population is more diverse in age, in presenting manifestations, in associated disease states and prognosis than the adult population with PH, which makes it more challenging and, I think, very interesting.
I would also like to add one thing about prognosis. As mentioned earlier, our center has a large lung transplant program, and two-thirds of our lung transplants are done in patients with cystic fibrosis. But we also do lung transplants in individuals with PH, and the mortality on the waiting list for those patients is far higher for those patients than it is for patients with cystic fibrosis. Brian discussed RV failure and mentioned that children with it are usually so resilient that they don’t complain. But once you get to RV failure, you have got a tough situation, and these patients are very fragile. But, on the other hand, while pediatric PH is a serious disease, it is also a disease for which we have a lot more treatments available today than were available even in the recent past. So we have to balance its seriousness of prognosis with a measured optimism about what we can do to help the children it affects to live longer and better lives.
BRIAN D. HANNA: The concept of what constitutes a good PH center is an interesting one for pediatrics. Dunbar is absolutely correct: You cannot run a pediatric PH center without major participation by everybody else in pediatric medicine. Our center began because we had patients with Eisenmenger complex who we needed to be able to treat. We had a lot of adult patients with what is now designated as class III PH, which is associated with lung disease,5 and I think that drew adult pulmonologists to our center, but we are having a difficult time in training pulmonologists and cardiologists in the area of pediatric PH.
HAROLD J. FARBER: Once upon a time we thought “Oh, PH is just a problem for which we need to work on lowering the blood pressure, and we can use vasodilators for that.” But now we are finding that endothelial dysfunction is a part of it. The different medications for treating PH have different mechanisms of action. What is effective, and how do we use those different mechanisms?
GEORGE B. MALLORY: Perhaps it has to do with a linguistic issue. Some people may call them pulmonary vasodilators, but, in fact, none of the medications that we use are pure pulmonary vasodilators. Mechanistically, all of them can and do have multiple effects on endothelial pathophysiology. We should therefore be talking about therapy for PH, not pulmonary vasodilator therapy.
That concept is important in a practical clinical context. Sometimes our cardiology colleagues misinterpret the significance of a poor response to inhaled nitric oxide in a newly diagnosed patient in the cardiac catheterization laboratory. I find all too often that I have to help the family understand that this does not mean that the medications available to us are likely to be ineffective.
D. DUNBAR IVY: I agree with George’s point about these drugs having multiple effects, including what in many cases is probably an antiproliferative effect on vascular smooth muscle cells.
I would also like second George’s comment about the failure of a child to respond acutely to a pulmonary vasodilator not meaning that such a drug should not be used in the long-term treatment of such a patient. I think this has been the paradox with intravenous epoprostenol, which many consider the “gold standard” agent for treating severe pulmonary arterial hypertension and right heart failure. Most children with severe PH and right heart failure will not respond to nitric oxide or intravenous epoprostenol in the short-term, but over the long-term the percentage of such patients who do respond with some reduction of blood pressure is probably 70% or 80%, which is why epoprostenol has a survival benefit.
In terms of the efficacy of such drugs, we are really at a disadvantage as compared with our adult-treatment colleagues. We have few if any randomized placebo-controlled trials on which to rely, and these drugs are not approved by the U.S. Food and Drug Administration (FDA) for use in young children. Some of them are approved for use in children older than 12 because that population was included in the trials of those drugs, but we don’t really have any FDA-approved drugs for children under 12.
Some data are just beginning to be published about one of the few randomized placebo-controlled trials. The European Respiratory Journal contains an abstract from the European Respiratory Society meeting held this past September on the use of sildenafil in children, suggesting that it has benefit in a subpopulation of children with PH.6
BRIAN D. HANNA: Dunbar is right. We need to know which drug to use for each patient. In the case of a drug like sildenafil, which has at least a theoretical benefit with respect to alveolarization on the basis of animal studies and models of BPD, we know that it helps substantially in weaning newborn infants and postoperative cardiac surgery patients from nitric oxide, and that it can significantly decrease symptoms and increase the 6-min walk time in adults.
But I am not sure that we know the answer to whether it changes the natural history of pediatric PH. However, if you were to say that children with idiopathic PH of New York Heart Association class I should probably be treated in the same way as adults are treated, I would concur in general, but I think that for children with BPD or pulmonary hypoplasia or chronic lung disease, we’re going to have to work that through and figure out where it is going to go in the future. We now have a huge number of pharmacologic agents at out disposal, and routes by which to administer them. Prostacyclin can now be given intravenously, subcutaneously, or by inhalation, and that does change a lot of the burden on the patient and the risks of such treatments.
HAROLD J. FARBER: Please briefly review the different classes of medications that we use and why we think each of them is of benefit in pediatric PH.
BRIAN D. HANNA: The 3 preeminent classes that we have are the phosphodiesterase type 5 inhibitors, endothelin receptor antagonists, and prostacyclins (Table 1 and see Ivy et al.7).
The etiology in some forms of PH appears to be associated with a deficiency or abnormality in nitric oxide production, and sildenafil seems to work very helpfully in weaning patients from nitric oxide, and may work beneficially in counteracting pulmonary hypoplasia. It is approved for treating PH of NYHA classes II and III, and I believe that bosentan was also recently approved for treating PH of class II, whereas the most widely used prostanoid, prostacyclin, is approved for treating classes III and IV disease.
The endothelin-1 receptor antagonists, both nonselective and selective, have been called a mainstay of treatment for treating PH in the adult population. Some people might question whether or not to use them in severely symptomatic patients, but I think they’re very beneficial in patients who have clear PH without substantial right heart failure.
And among the prostacyclins, we have in North America epoprostenol, treprostinil, and iloprost. The current recommendations call for their use in treating classes III and IV heart failure, which affect the sickest of our patients. Treprostinil is another prostacyclin analog for treating PH with NYHA class II to IV effects, but I don’t think it has received FDA approval for pediatric use.
FELIX SHARDONOFSKY: I would like to make a comment about the use of calcium-channel blockers. A small population of patients who have significant PH may benefit from treatment with calcium-channel blockers, a medication which is safe and inexpensive.
BRIAN D. HANNA: Yes, calcium-channel blockers were among the first agents to be used for treating PH. However, I think that it is currently believed that in the kind of pathophysiology that exists in Eisenmenger syndrome, with increased right-heart pressures, calcium-channel blockers don’t work well or for a long time. I have 250 kids being treated with one kind of agent or another for PH, and not one of them is being treated with a calcium-channel blocker, so clearly I have a bias.
HAROLD J. FARBER: What are some perils and pitfalls in therapy for pediatric PH?
GEORGE B. MALLORY: One of our most common therapeutic dilemmas occurs when the young infant with PH and diaphragmatic hernia or a pulmonary growth abnormality or BPD develops an acute illness associated with hypoxemia. The question surfaces as to whether the hypoxemia is directly attributable to worsening. My colleagues who are often less familiar with PH often think that we should prescribe more PH medication. But while there are exceptions, the problem in the vast majority of cases is the underlying lung disease, and the focus really has to be on treating that with antibiotics, alveolar recruitment strategies, or the like.
The converse situation is that some practitioners may get so comfortable with the use of sildenafil for children with PH, believing that they do not need specialized PH follow-up. These patients should be considered as very vulnerable, with the need to be followed either in a PH center or managed via good communication between the treating physician and a PH clinician with a great deal of experience.
BRIAN D. HANNA: I think the first major pitfall that we face is that of underdosing our patients. We’re now using the endothelin-1 receptor blockers and the phosphodiesterase inhibitors and some classes of prostacyclins that have very few side effects, and I think it’s very easy for us simply to use the FDA-approved dose of any of these and rest on the belief that we are doing what we should be doing. But our experience with epoprostenol suggests that we often need to escalate the dosage to get control of the disease.
The second significant issue for us is that we have to be very careful about side effects of treatment for other than the NYHA class I indications, especially in infants. For example, we don’t have any idea of the prospective incidence of visual or auditory difficulties with the long-term use of sildenafil or prostacyclins. And there are now reports of patients who have had interventricular hemorrhage with sildenafil.
And lastly, I only partly agree with George about the management of the child with PH who gets a respiratory infection or other acute illness. In this age in which influenza has been rampant and rhinovirus infection can be epidemic, many kids are presenting with profound right heart failure that is being missed in the emergency room and the general practitioner’s office and the pediatric intensive care unit, we have to be very careful not to miss RV failure in those patients.
FELIX SHARDONOFSKY: We may add that patients should be monitored very closely using an array of data such as symptoms, assessment of functional capacity, and exercise capacity using submaximal exercise tests such as 6-min walk test and incremental cardiopulmonary exercise test, biomarkers such as NT-proBNP, as well as invasive and noninvasive hemodynamic studies. The data obtained from periodic evaluations allows titration of therapy based on clinical severity and potential untoward effects of the available treatment options. Finally, it is important that any medical or surgical co-morbidity that a pediatric patient with PH may experience be managed in a medical center with expertise in the care of patients with pulmonary vascular disease.
D. DUNBAR IVY: My recommendation is to be certain that you have the correct diagnosis. Pulmonary hypertension can be associated with many diseases, and the underlying principle in its management is to treat the underlying disease if there is one. Examples include patients with PH associated with pulmonary capillaritis or follicular bronchiolitis, the treatment of which can markedly improve PH. This emphasizes the importance of a complete evaluation of any patient with significant PH.
HAROLD J. FARBER: I want to thank each of you for your participation and your time here today.