Stout - 20
`
AHA Scientific Statement
Chronic Heart Failure in Congenital Heart Disease: A Scientific Statement From the American Heart Association
Karen Stout, Chair, Craig Broberg, Co-Chair, Wendy Book, Frank Cecchin, Jonathan Chen, Konstantinos Dimopoulos, Melanie Everitt, Michael Gatzoulis, Louise Harris, Daphne Hsu, Jeff Kuvin, Yuk Law, Cindy Martin, Anne Murphy, Heather Ross, Gautam Singh, Thomas Spray
Corresponding Author:
Karen Stout, MD
University of Washington
1959 NE Pacific, Box 356422
Seattle, WA 98195
PART I: GENERAL CONSIDERATIONS
INTRODUCTION
The last 50 years have brought remarkable advancements in the diagnosis and treatment of congenital heart disease (CHD). Early diagnosis and improvements in cardiac surgery and interventional cardiology have resulted in unprecedented survival of patients with CHD, even for those with the most complex lesions. Despite remarkable success in treatments, many interventions are palliative rather than curative, and often patients develop cardiac complications including heart failure (HF). HF management in the setting of CHD is challenged by the wide range of ages at which HF occurs, the heterogeneity of the underlying anatomy and surgical repairs, the wide spectrum of HF causes, the lack of validated biomarkers for disease progression, the lack of reliable risk predictors or surrogate endpoints and the paucity of evidence demonstrating treatment efficacy.
The purpose of this statement is to review the current literature pertaining to chronic HF in CHD, make recommendations where reasonable and elucidate important gaps in our knowledge, emphasizing the need for specific studies of mechanisms and of improving quality of life and outcome of those with HF. In this document, the definition of HF corresponds to that found in the multiple guidelines on diagnosis and management of HF. Though there are nuances and specific details that may be co1ntroversial, the broad definition from the Heart Failure Society of America guidelines states: “In physiologic terms, HF is a syndrome characterized by either or both pulmonary and systemic venous congestion and/or inadequate peripheral oxygen delivery, at rest or during stress, caused by cardiac dysfunction.” 2 The definition of “chronic HF” in this document concurs with that of the European Society of Cardiology guidelines, which emphasize chronic HF (whether stable, progressively worsening or decompensated) rather than acute. While specific definitions of acute vs. chronic HF are not universally accepted, herein we focus on chronic HF as a persistent syndrome that requires consideration for therapy to prevent progression, decompensation and/or death, similar to other guideline documents.1
The content of this document covers the age spectrum of pediatric to adult patients with CHD and HF with input from both pediatric and adult cardiologists. However, the bulk of available literature focuses on adult patients in whom there is a greater relative burden of HF, presumably reflecting the natural history of these diseases. Thus, the majority of the discussion herein is more applicable to adults with CHD and HF, though whenever able, specific issues in pediatric patients are discussed.
Some features of HF in CHD are common across diagnoses and are discussed in the general overview. However, special emphasis is given to topics with particularly unique anatomic and physiologic considerations, in particular for patients in whom the right ventricle (RV) is more vulnerable, whether in the normal subpulmonic position or as the systemic ventricle, and those patients with single ventricle (SV) physiology. In addition left ventricle (LV) pressure and volume loading lesions are not a, there are variations pressure or volume loading of the left ventricle (LV) that are unique to CHD, which re discussed separately as well.
This document focuses on mechanisms and treatment of myocardial dysfunction, while recognizing that HF symptoms may be due to underlying repairable hemodynamic abnormalities, such as valve dysfunction, outflow obstruction, coronary abnormalities or residual shunting. Therefore, all CHD patients with HF symptoms should undergo a careful hemodynamic assessment by CHD experienced cardiologists for any reversible hemodynamic abnormalities and receive appropriately targeted interventions if possible. Treatment recommendations for HF caused by valve dysfunction and/or ischemic heart disease are addressed elsewhere in the respective ACC/AHA guidelines, including the 2008 Guidelines on the Care of the Adult with Congenital Heart Disease. 3 Though this document focuses on HF treatment, palliative care should be considered a valuable and needed component of care in all patients with CHD and end-stage HF4.
OVERVIEW
Incidence of congenital heart disease
Structural heart disease is the most commonly occurring congenital disorder diagnosed in newborns, with birth prevalence reported as 10 per 1,000 live births.5,6 Registry studies have estimated an incidence between 3 and 20 per 1,000 live births.7 The incidence of CHD based on birth prevalence may be an underestimate, however, underestimates the true frequency as CHD is not necessarily apparent at birth but diagnosis may be delayed until childhood or adulthood. In fact, over one-quarter of CHD diagnoses are made after infancy.8
Survival in patients with congenital heart disease
Survival in children born with CHD has improved dramatically over the past several decades due in large part to surgical advances for children with complex CHD. Survival of newborns with complex CHD now approaches 90%, and 96% of newborns with CHD who survive the first year of life remain alive at 16 years of age.8 Infant survival in the present era is significantly better than for those born in prior decades but varies with CHD complexity; only 56% of newborns with heart defects of great complexity survive to 18 years of age. 9 Adults with CHD are also living longer, with the median age at death increasing from 37 years in 2002 to 57 years in 2007. 7 Even more striking is the change in mortality for patients with severe CHD in whom the median age at death has increased from 2 years prior to 1995 to almost 25 years currently. 10 [Figure 1]
Based on prevalence estimates of CHD as well as registry data, there are presently over 1 million adults with CHD in North America and 1.2 million in Europe. 5, 11, 12 While the majority of these survivors have simple forms of CHD such as atrial and ventricular septal defects, a significant proportion of them have more complex CHD; including 10% with defects of great complexity (such as SV) and 30% with moderate CHD (such as conotruncal defects and atrioventricular canal defects). 11
Importance of heart failure in CHD
The impact of cumulative survival means more patients are at risk for HF. Despite great success in the medical and surgical management of CHD, long-term survivors often have residual cardiac diseases and/or pulmonary or hepatic impairment due to sequelae of cardiac dysfunction. 13, 14 In a recent analysis, 76% of the deaths that occurred in patients with CHD who survived the first year of life occurred after the age of 18 years. 15 HF is an important problem for this expanding population of older children and adults, though he prevalence of HF in children and adults with CHD is unknown. However, HF has been reported to develop during childhood in approximately 5% of all patients with CHD and up to 10-20% of patients after Fontan procedure. 16-18 By adulthood, the prevalence of HF, variably defined, following the Fontan procedure is nearly 50%. 19, 20
Heart failure mortality and morbidity in CHD
In a recent population-based study, HF was the major cause of late death (>30 days) in children after pediatric cardiac surgery, contributing to 27% of the deaths and occurring at a median age of 5.2 years. 21 HF is the leading cause of death in adults with CHD as well, described in 26% of all deaths in a national registry of over 8000 adults with CHD, with similar findings in other reports. 18, 22, 23
In addition to decreased survival, adults with CHD face significant morbidity. The number of CHD hospitalizations has increased 101% from 1998 to 2005, with rates 2-3x higher than population norms. HF is a common reason for admission, though less common than arrhythmia. 16-18 Further highlighting the severity of the problem, CHD is the leading indication for heart transplant in the pediatric age group.24 In adulthood, because ischemic heart disease predominates, CHD is the indication for transplant in only 3% of cases. 25 This represents a small subset of the adults with end-stage HF due to CHD. One explanation may be that decisions regarding referral or listing are influenced by the high early mortality after transplant reported in the CHD population. 26
Heart failure classification in CHD
The clinical presentation of the HF patient with CHD may vary significantly by defect or age. CHD patients can have classic symptoms of fatigue, dyspnea and exercise intolerance, but may manifest more subtle signs of malnutrition, growth failure, or cachexia. 27, 28 One of the unique features of CHD patients is they often have adapted to their longstanding limitations, therefore they may not report symptoms despite significant objective exercise impairment. Thus, unless careful questioning is undertaken, application of general HF classifications such as the New York Heart Association (NYHA) categories or the Modified Ross Classification may underestimate the severity of disease, particularly in patients with complex or cyanotic CHD. 29 The Warnes-Somerville classification was developed to describe limitations in adults with CHD, though is not commonly used. None of the available grading scales [Table 1] have been validated in predicting outcomes.
The ACC/AHA guidelines for the diagnosis and management of HF, updated in 2013, specifically exclude children as well as CHD, valvular heart disease and infiltrative cardiomyopathies. 30 31, 32 The guidelines do, however, generalize recommendations to all other etiologies of HF. The staging system described in the guidelines recognizes risk factors for the development of HF including hypertension, diabetes, and coronary atherosclerosis. If the A-D staging in the HF guidelines were extrapolated to CHD, the vast majority of asymptomatic CHD patients would be categorized as Stage B. [Figure 2] However, there is little data to show that the medical or device therapies recommended for stages B-D are effective in patients with CHD of any age, thus applying all the recommendations may not optimally suit the CHD population, and there is inadequate evidence that categorizing CHD patients by this system enables management decisions or improves outcome. The guidelines are clear that HF is a clinical diagnosis, and that the presence of ventricular dysfunction or the result of any other single diagnostic test is not sufficient to make the diagnosis. This also applies to CHD patients. Also, recommendations regarding control of acquired heart disease risk factors, weight management and the need for routine health maintenance screening are broadly applicable to CHD patients.
POTENTIAL MECHANISMS OF HEART FAILURE IN CHD
General considerations
Clinical HF in CHD is multifactorial. An ineffective cardiovascular system in CHD, even after repair, can be the cumulative result of valvular abnormalities, shunts, flow obstruction, arrhythmia, or persistent anatomic defects such as a single ventricle, in addition to dysfunction of the myocardium itself. Likewise, myocardial dysfunction in CHD can be the end-result of any number of hemodynamic derangements such as abnormal pressure or volume loading, ventricular hypertrophy, myocardial ischemia, or even effects of prior cardiopulmonary bypass or ventriculotomy. Any of these may incite systolic and/or diastolic impairment [Table 2] together with their clinical manifestations such as arrhythmia or exercise intolerance. This section acknowledges these many causes but will focus on potential etiologies of myocardial dysfunction specifically, a final common pathway in CHD. 33 Much is unknown or speculative, based on extrapolation from other HF models, yet understanding specific mechanisms and pathways is vital to providing informed and effective treatment strategies.
Myocardial Architecture
The myocardial architecture in CHD can exhibit a certain disarray of ventricular myocardial fibers. 34,35 This is especially the case for the RV. Development of the RV is controlled by a different profile of transcriptional pathways than the LV. 36 The normal RV myocardium has only a superficial circumferential layer and deep longitudinal layer, but not a middle layer of circular fibers that normally makes up over half the wall thickness of a morphologic LV. 34 In an animal model of hypoplastic left heart syndrome (HLHS), abnormal RV and LV myocardial fiber orientation, reflected in abnormal patterns of anisotropic RV and LV deformation, was noted prenatally. 37 Different myofiber and connective tissue architecture has been observed in tricuspid atresia patients as well. 34 It is plausible, though still hypothetical, that these alterations impart a disadvantage to the myocardium and make it vulnerable to dysfunction, though how much of a role they play is unknown.
Abnormal Perfusion
Many CHD patients are cyanotic at birth, which can result in significant myocardial ischemia until repair or palliation. The early period of ischemia may not have a detectable impact on ventricular function acutely but may jeopardize or preprogram the myocardium to more serious dysfunction much later in life. In other cases there may be a coronary flow/demand mismatch, such as occurs in the systemic RV. Many studies have demonstrated perfusion abnormalities in patients with a systemic RV, where the typical coronary anatomy supplying the RV is insufficient for a hypertrophied, enlarged ventricle, though there is conflicting data regarding the frequency and clinical importance of these findings. 38-43 Some conditions such as transposition of the great arteries (TGA) are associated with coronary anomalies, which may subject the myocardium to prolonged ischemia and/or infarction either before or as a result of surgical repair. 44, 45 Myocardial perfusion assessed by positron emission tomography is often abnormal in Fontan patients and dTGA after an atrial switch procedure. 46 Even in the absence of coronary arterial abnormalities, tissue ischemia may be present. High wall stress from increased afterload in conjunction with decreased coronary flow reserve has been associated with myocardial hypoperfusion and supply/demand mismatch,47, 48 whose effects may only become manifest over decades.
Neurohormonal Activation
There is ample evidence from acquired heart disease that activation of cell signaling systems occurs in response to ischemia or to abnormal cardiac distension from deranged pressure or volume loading. 33 Activation of natriuretic peptide, sympathoadrenergic, endothelin, and renin–angiotensin-aldosterone systems (RAAS) can be driven by any of these adverse conditions, 49-53 which are ubiquitous in CHD. While less is known about specific activation pathways in CHD, there is certainly growing evidence, mainly in the form of elevated biomarkers, to support similar activation in CHD. Brain natriuretic peptide has been the most extensively documented biomarker, with elevated serum levels demonstrated in patients with poorer cardiovascular function or prognosis. 54-56 Data on RAAS and sympathoadrenergic axes in CHD are limited but also suggest activation, 33, 52 and argue in favor of similar HF pathways as those well studied in other models. Yet studies in CHD are small in size with limited follow up, and importantly do not show uptitration of biomarkers in all individuals. Therefore there is likely much more to understand regarding factors that govern neurohormonal activation than the biomarker evidence can provide.