Plastic Bronchitis and Protein Losing Enteropathy in Children With Single Ventricle Physiology

Protein losing enteropathy (PLE) has emerged as an increasingly common complication of single ventricle palliation in children born with cyanotic congenital heart disease. This entity becomes manifest months or years after the modified Fontan operation and is thought to affect 5-15% of all patients with Fontan physiology. Patients with PLE often suffer from symptoms related to severe hypoproteinemia and intestinal malabsorption, including edema, ascites, pleural and pericardial effusions, chronic diarrhea and poor growth. Mortality is strikingly high, with only 50% of patients surviving for more than 5 years following diagnosis. The pathophysiology is poorly understood and several theories are presented as possible mechanisms which includes: 1) chronic low cardiac output and intestinal injury leading to loss of intestinal integrity, 2) chronic intestinal inflammation leading to loss of intestinal integrity, 3) chronic elevated portal venous pressure with intestinal injury leading to loss of intestinal integrity. Treatment is sporadically successful and has included 1) creating a fenestration between the Fontan pathway and pulmonary venous atrium which decreases Fontan pressure and increases cardiac output, 2) chronic use of unfractionated or low molecular weight heparin to reconstitute the intestinal basement membrane, 3) Use of oral steroids to reduce intestinal inflammation, 4) Use of pulmonary vasodilators to reduce Fontan pressure and increase cardiac output 5) implantation of pacemakers to treat sinus node dysfunction and improve cardiac output, 6) heart transplantation. Short of cardiac transplantation, no treatment of PLE, including efforts to change systemic and/or gastrointestinal hemodynamics, is routinely successful. Thus, it appears that the etiology of PLE is more complex than simply "high venous pressure in the gut" as has been previously believed.

A second entity seen in patients after modified Fontan procedure is "plastic bronchitis" (PB). This is a highly morbid clinical picture of persistent production of high mucoprotein-containing airway secretions that causes severe airway obstruction leading to increased work of breathing, V/Q mismatch and progressive respiratory failure leading to death. The etiology of plastic bronchitis is not understood at all and beyond supportive, palliative measures such as lung lavage, there is no current treatment for plastic bronchitis.

VIP and Substance P are both small peptides with significant autocrine and paracrine signaling ability. The role of VIP in controlling gut protein and solute secretion has been understood for approximately 20 years. The clinical entity where this has been best described are in rare "VIP-omas" and some other chromophore secreting gut tumors where massive amounts of VIP are secreted by the tumor and overwhelm the body's normal ability to clear VIP through cleavage by neutral endopeptidases in the pulmonary endothelium.

Until very recently, Substance P was known primarily for its role in peptidogenic signaling in sensory pathways. As described below, an important and previously not understood role for Substance P in the control of human airway mucus secretion has recently been described.

Scientific Basis for Proposed Research:

The basis for this proposal arises from three, independent streams of research.

1. First, extensive investigations in patients with increased portal venous pressures, usually from cirrhosis, has shown that increased portal venous pressures lead to significant increases in the excretion of a wide variety of small peptide signaling molecules. Included in these molecules are VIP, 5-HT and Substance P.As patients with Fontan physiology always have increased portal and gut venous pressures as a direct consequence of their anatomic palliation, it stands to reason that they should likewise have increased GI production of these signaling peptides.
2. Second, clearance of gut-derived signaling peptides such as VIP, Substance P, angiotensin, brandykinin and others occurs primarily in the lungs via peptide cleavage catalyzed by the neutral endopeptidases, a class of enzymes heavily resident in the pulmonary circulation. The role of the neutral endopeptidases is to cleave and thus inactivate small peptides and 'protect" the systemic circulation from the plethora of vasoactive peptides that come from the systemic circulation. With respect to the potential role of VIP and/or Substance P in the etiology of PLE or PB, it is important to note that in patients with Fontan physiology current practice is to fenestrate the Fontan shunt to allow some portion of the systemic venous blood flow to bypass the lungs. Thus, any vasoactive / signaling peptides returning in the lower body systemic venous circulation will bypass the pulmonary clearance mechanisms in proportion to the degree of fenestration shunt at the moment. This pulmonary clearance bypass physiology is relative unique in medicine to older patients with Fontan-palliated congenital heart disease and thus may be, in part, the basis that PLE and PB are seen almost uniquely in Fontan patient. Further implicating derangements in active peptide clearance physiology in the etiology of PLE and PB is the fact that the classic member of the neutral endopeptidase class is (the misnamed) Angiotensin Converting Enzyme (ACE) , which actually cleaves a number of these small peptide signaling molecules including VIP and Substance P. Virtually all patients with Fontan physiology are on large doses of ACE-inhibitors to lower systemic afterload. Captopril-induced inhibition of VIP and Substance P inactivation would significantly increase already elevated systemic levels of these signaling peptides and further potentiate the etiology of PLE and PB in Fontan-palliated patients. Variations of neutral endopeptidase inhibitor ("ACE inhibitor" ) drug dosage and/or genotype variation of ACE which causes dramatically altered activity can in part explain the variable incidence of PLE and PB in Fontan physiology patients. In this study, the use and dose of ACE inhibitor that each patient is on will be recorded to see if there is a correlation between ACE inhibitor dose and measured VIP and Substance P levels. There are wide variation of ACE gene genomics that have been described for decades. Intriguingly, approximately 10-15% of all persons have a newly described genomic variation of the ACE gene which leads to profoundly reduced circulating levels of ACE and increased susceptibility to ACE inhibition and 10-15% of patients with Fontans develop PB or PLE.
3. Lastly, very recent work has completely changed our understanding of the control of normal, basal mucus and solute secretion in the human airways and gut. Here-to-fore it was believed that cholinergic mechanisms controlled airway mucus secretion. This recent and still emerging work shows that in fact it is VIP and Substance P which act individually and synergistically in the control of both normal basal human airway mucus production and in regulating gut secretion. This very new work gives a unifying hypothesis to the seemingly disparate entities of PLE and PB: since Substance P and VIP control normal basal human airway secretion and stimulated human gut protein secretion, increases in pulmonary venous and/or gut arterial levels of these peptides can be the basis for PLE and PB. The GI, cardiac and pulmonary physiology and pharmacologic treatment of patients with fenestrated Fontans is a perfect setting for this interaction to occur.

Taken in totality, these three streams of independent research give a strong basis for the hypothesis that the palliative physiology and pharmacologic treatment of Fontan patients leads to increased levels of VIP, Substance P or both, and they may be causal in the pathophysiology of PB and PLE.

Rationale for Current Study Design and Subject Recruitment:

Consistent with our Aims and Hypotheses, four patient groups will be studied: i)Fontan Physiology Patients with PLE and PB Patients ii) Fontan Physiology Patients without PLE and PB iii) Patients with Glenn Physiology Congenital Heart Disease and iv) Patients with 2 ventricle physiology and an atrial septal defect who will serve as a Control group. Measurement of mixed venous and arterial levels of substance P and VIP in these 4 groups of subjects will allow us to determine the effects of elevated portal venous pressure on the production and clearance of substance P and VIP and well as the effects of altered hepatic-pulmonary blood flow on the clearance of substance P and VIP. This should allow us to determine if 1) systemic arterial levels of substance P and VIP levels are elevated in Fontan patients when compared to controls, and if elevated, 2) whether this is due to increased hepatic production, reduced pulmonary clearance or both. The rationale for the inclusion of each patient group is summarized below.

Subjects with Fontan physiology (Group I and II) have a unique physiology where both the superior vena cava (SVC) and inferior vena cava (IVC) are connected directly to the pulmonary arteries thereby bypassing the heart. This palliation scheme allows for separation of the pulmonary and systemic circulations in patients with univentricular hearts and essentially eliminates cyanosis. These patients have non-pulsitile pulmonary artery flow with elevated central venous and portal venous pressures. Therefore, it is suspected that these patients will have increased production of substance P and VIP. In addition, a fenestration (or hole), is created between the Fontan conduit and the atrium at the time of Fontan surgery in order to lower central venous pressure and increase cardiac output by creating a right to left intracardiac shunt. It is suspected that these patients will have elevated systemic venous and arterial levels of vasoactive peptides as a result of 1) increased hepatic/gut production, 2) decreased pulmonary vascular clearance, and/or 3) intracardiac right to left shunting. Studying these subjects will allow us to determine how systemic venous and systemic arterial levels of Substance P and VIP are altered by the Fontan circulation. By comparing systemic arterial and venous levels of vasoactive peptides among control, Glenn physiology and Fontan physiology groups we hope to measure differences in vasoactive peptide levels that could be responsible for the development of PLE and PB in Fontan physiology patients.

Subjects with Glenn physiology (Group III) have a unique physiology where hepatic venous and lower body venous blood returns to the heart and enters the systemic ventricle and is recirculated to the body after combining with the pulmonary venous blood. There is no direct connection thru the heart for hepatic venous blood to enter the pulmonary arteries. Therefore, it is suspected that these patients will likely have elevated systemic arterial circulating levels of Substance P and VIP due to intracardiac right to left shunting and reduced pulmonary clearance. However, since these patients typically have normal portal venous pressures, they are not thought to have increased production of these vasoactive peptides. In addition, these patients historically are not at risk for developing PLE and PB. Studying these subjects will allow us to determine how systemic venous and systemic arterial levels of Substance P and VIP are altered by the Glenn circulation. By comparing systemic arterial and venous levels of vasoactive peptides among control, Glenn physiology and Fontan physiology groups we hope to measure differences in vasoactive peptide levels that could be responsible for the development of PLE and PB in Fontan physiology patients.

Control group subjects (Group IV) will have normal hepatic-pulmonary blood flow and normal portal venous pressures. An atrial septal defect allows for left to right shunting of blood thru the heart but does not typically result in elevated pulmonary artery or right ventricular pressures. These subjects will have normal flow of hepatic venous blood into the pulmonary capillary bed. Therefore they should have normal hepatic production and pulmonary clearance of substance P and VIP.