In his insightful and witty educational offering, Dr. Seuss’s ABC , Theodor Seuss Geisel poses the intriguing rhetorical question “Big A, little a, what begins with A?” He might have answered, “An amalgamated assemblage of advancing ailments affecting the heart” (it works better if you consider the h to be silent.) Dr. Seuss contributed much to world literature but missed the opportunity to create a classification of cardiac conditions. Such a system was applied to stages of heart failure in the 2001 American College of Cardiology and American Heart Association guideline update for the diagnosis and management of chronic heart failure in adults, published 10 years after Dr. Seuss’s death. This document describes heart failure as a disease continuum, from stage A to stage D. Briefly, stage A patients have one or more risk factors for developing clinical heart failure, but they are asymptomatic and have normal cardiac structure and function. They may have any number of disease states that put them at risk, such as diabetes, coronary artery disease, and hypertension. Stage B patients also lack heart failure symptoms, but they have evidence of cardiac dysfunction, whether left ventricular (LV) hypertrophy, myocardial infarction with wall motion abnormalities, low LV ejection fraction (LVEF), diastolic dysfunction, or valvular disease. The label of stage C accompanies the onset of heart failure symptoms in the presence of structural heart disease. Stage D describes end-stage patients who have refractory symptoms.
The guidelines make it clear that stages A and B are not to be considered heart failure per se. If a patient lacks symptoms (dyspnea, fatigue) and abnormal signs on physical exam (lower extremity edema, pulmonary rales), even if he or she has obvious structural or functional abnormalities, heart failure is not present. In other words, an asymptomatic patient with an LVEF of 25% can be considered to have “pre–heart failure” but not heart failure. In addition, the degree of dysfunction that constitutes stage B is not defined. Thus, whether a patient with mild mitral regurgitation, stage I diastolic dysfunction (relaxation abnormality), and an LVEF of 51% has stage A or stage B heart failure is not settled. The guidelines also do not consider the gradation of structural and functional abnormalities within the stages. Thus, a patient with an LVEF of 10%, severe mitral regurgitation, and a restrictive filling pattern whose symptoms improve with diuresis and afterload reduction is categorized as stage C, not stage D.
The identification of stages A and B represent “an attempt to help healthcare providers with the early identification of patients who are at risk for developing heart failure.” The idea is that health care providers can intervene at the earlier stages to initiate therapies and lifestyle changes that will halt or slow progression of the disease. After patients have developed clinical heart failure, the designation of stage D allows tailored, advanced therapy (often involving referrals to specialists for advanced treatment strategies, such as ventricular assist devices, inotropic infusions, transplantation, or palliative or hospice care). Heart failure stages differ from the New York Heart Association (NYHA) classification scheme. NYHA class I encompasses stages A and B, while NYHA classes II to IV generally cover stages C and D. Patients can fluctuate in their NYHA classifications, going from IV to I and back again. Once patients develop symptoms, however, they are forever after considered to have at least stage C heart failure, even if they return to an asymptomatic state. Presumably, a patient in stage D heart failure who receives a temporary ventricular assist device, experiences recovery, has the device removed, and returns to an asymptomatic state can move from stage D to stage C.
“Big E, Little E, What Begins With E?”
Echocardiography, of course, is the big E in heart failure imaging. The guidelines label two-dimensional and Doppler echocardiography the “single most useful diagnostic test in the evaluation of patients with heart failure…to determine whether abnormalities of myocardium, heart valves, or pericardium are present and which chambers are involved.” Echocardiography can provide structural and functional explanations for heart failure symptoms that may be targets for therapies. It can also provide a baseline for comparison with future examinations to determine progression or regression of disease, which is noted to be of value, even if it has no impact on heart failure staging. The guidelines acknowledge that although specific abnormalities on echocardiography do not always correlate with symptoms, there is sometimes uncertainty as to whether symptoms and signs are due to cardiac dysfunction or other disorders. For example, a breathless patient with pulmonary rales and completely normal systolic and diastolic echocardiographic parameters may have noncardiogenic pulmonary edema. A recent study in this journal suggested that diastolic, as well as systolic, parameters may help differentiate cardiac from pulmonary causes of dyspnea.
Despite this enthusiastic and euphonious endorsement of echocardiography, all of the recommendations regarding the use of echocardiography in the guideline carry a level of evidence C, meaning that they are backed not by clinical trial or observational study data but only by expert opinion, case studies, or standard of care. And in the care of stage A patients, the guidelines recommend the assessment of LV size and function only in patients with strong family histories of cardiomyopathy and in patients receiving cardiotoxic medications (such as anthracycline chemotherapy).
These points highlight the importance of the study by Carerj et al. in this issue of the JASE . Their study essentially evidence for echocardiographic examination of patients with stage A heart failure. The authors screened an impressively large number (1,097) of asymptomatic patients at 19 different centers. Importantly, patients had no electrocardiographic or physical examination findings of cardiovascular disease but did have one or more cardiovascular risk factors. Thus, this population had already undergone one screening test (electrocardiography) beyond history and physical examination. To focus on myocardial dysfunction as a cause of heart failure, the authors excluded patients who demonstrated pericardial disease, pulmonary hypertension, aortopathy, and significant valvular dysfunction on echocardiography. They considered baseline LV systolic dysfunction (LVSD) to be an LVEF < 50% and LV diastolic dysfunction (LVDD) as grade I to IV, on the basis of mitral inflow and tissue Doppler parameters, and they also measured LV mass indexed to body surface area. The authors then followed patients for a mean of 26 months and tracked traditional major adverse cardiovascular events, as well as chart documentation of acute pulmonary edema, heart failure hospitalization, and cardiologist-defined diagnosis of clinical heart failure (at least stage C heart failure). At baseline, patients with three or more cardiovascular risk factors were more likely to have LVSD but not LVDD. On follow-up, multivariate predictors of adverse heart failure outcomes included gender, LVSD, and LVDD. Combined end points were predicted by age, gender, obesity, and LVSD. LVEF < 50% provided incremental prognostic value in predicting major adverse cardiac events and pulmonary edema in patients with three or more cardiovascular risk factors. It also predicted heart failure hospitalization and diagnosis of clinical heart failure in patients with two or more cardiovascular risk factors.
“Big S, Little S, What Begins With S?”
Whether asymptomatic patients with risk factors for cardiovascular disease and/or the development of clinical heart failure should be summarily screened for sinister structural shortcomings remains a contentious issue. Congruent with previous studies, the work by Carerj et al. suggests that echocardiographic screening of stage A patients with risk factors may identify a population with reduced LVEFs who are at risk for major cardiovascular events and may provide incremental prognostic information to clinical risk scores. Furthermore, patients with stage B heart failure may be candidates for different therapies than stage A patients, including angiotensin-converting enzyme inhibitors or angiotensin receptor blockers and β-blockers, which may forestall progression to stage C heart failure. Thus, a screening study that reclassifies patients from stage A to stage B should have a significant impact on both prognosis and therapy.
The 2007 appropriateness criteria for transthoracic and transesophageal echocardiography, however, either do not address or specifically recommend against the use of echocardiography in asymptomatic patients without physical examination findings suggestive of cardiac disease. Exceptions include those also mentioned by the heart failure guidelines, namely, patients who have first-degree relatives with inherited cardiomyopathy or recipients of cardiotoxic therapies. Patients with isolated premature atrial or ventricular beats and those with hypertension without suspicion for hypertensive heart disease are specifically labeled inappropriate candidates for echocardiography. There is no mention of patients with multiple risk factors for cardiovascular disease.
An important issue involves what to measure when screening. Different cutoffs for defining abnormal LVEF have been used in different trials. In the present study, LVEF < 50% appeared to provide more prognostic information than diastolic dysfunction, perhaps because diastolic dysfunction (especially the predominant early stages) may represent a very early stage in the heart failure continuum, which progresses slowly or not at all. Despite its powerful predictive capacity demonstrated in other trials, LV mass fell out in a multivariate prediction model in this study, possibly, as the authors point out, because of differing rates of underlying disease (namely, hypertension and coronary artery disease). These findings suggest that LVSD may be the best target of screening in the stage A heart failure population, but more data comparing the feasibility and utility of different ways to measure systolic and diastolic (including strain and strain rate), are needed.
The prevalence of LVSD is low (0.4%–2%) in the general population and highly dependent on risk factors, especially age (as well as the LVEF cutoff). In the present study, the prevalence of asymptomatic LVSD was significantly higher but still <10% for patients with one and two cardiovascular risk factors and 15% for those with three or more factors. The presence of any degree of diastolic dysfunction did not vary across the number of risk factors and was present in one third of patients, consistent with the finding of a prior screening study. Although these findings had prognostic value, the actual percentages of patients who experienced different adverse events were small. The number of events appears too small to correlate with specific degrees of either systolic or diastolic LV dysfunction. Because different studies have used a relatively wide variety of cutoff values for defining LVSD and LVDD, it is difficult to find a consistent level of dysfunction with prognostic and therapeutic significance.
A substantial number of patients in the present study were already taking angiotensin-converting enzyme inhibitors and β-blockers (although perhaps their doses would have been more aggressively escalated if the patients were known to have structural or functional heart disease). And, as the authors appropriately acknowledge, whether these and other therapies definitely forestall or ameliorate the progression to stage C heart failure is not well established. No measure of cost-effectiveness was performed in this study, but with such a low event rate and somewhat uncertain benefit, it seems unlikely that performance of full screening echocardiography on every stage A patient would yield an improvement in quality-adjusted life-years, or any other outcome metric, to justify the cost.
One possible solution to the cost problem is the use of hand-carried ultrasound devices by novice operators to perform limited screening in selected populations. The reduced acquisition cost, simplicity, and smaller size of the latest generation hand-carried ultrasound devices should lower screening costs and improve feasibility. Limited training programs have proven adequate to produce operators capable of detecting LVSD with acceptable accuracy. On the basis of findings from the current study, it may be that using a hand-carried ultrasound device to screen patients with three or more risk factors may prove cost effective, but this remains to be seen.
Another strategy involves the use of brain natriuretic peptide (BNP) as a screening tool for development of heart failure. Although BNP has limitations, especially in obese patients, it has demonstrated prognostic value. BNP screening is mentioned specifically in the most recent version of the heart failure guidelines, though there is no official recommendation regarding its use. The present study did not include BNP measures and therefore could not test whether echocardiography provides incremental prognostic benefit. The selective use of echocardiography in a strategy using BNP and clinical risk predictors may prove more cost effective than more broad screening with echocardiography.
The study by Carerj et al. also did not address the question of how often patients with stage A heart failure should be screened for development of LVSD or LVDD. The “warranty period” of an echocardiographic study with normal ejection fraction and no diastolic dysfunction is uncertain. It is also not clear whether or how often patients who are found to have cardiac disease should receive follow-up studies to monitor for progression. The appropriateness criteria propose that only change in symptoms or clinical status should prompt follow-up echocardiography. The heart failure guidelines state that “routine periodic assessment of LV function in other patients [without strong family histories of familiar cardiomyopathy or cardiotoxic therapy] is not recommended,” although this statement was not included in the official recommendations.
Another intriguing issue is the use of stress testing in high-risk stage A patients with unremarkable or minimally abnormal resting studies. Such tests might elicit structural or functional abnormalities. Because coronary artery disease without clinical heart failure is considered in the guideline to be a risk factor for heart failure, patients with stress-induced ischemic wall motion abnormalities would still be in stage A. It is unclear whether this reasoning would apply to other stress-induced findings such as ventricular dilatation, worsening valvular regurgitation, elevations in pulmonary artery systolic pressure, or worsening diastolic dysfunction, which might be evidence of stage B heart failure. And patients who are sedentary (whether because of lifestyle choices or conditions such as arthritis) may develop heart failure symptoms at low levels of exertion, presumably advancing them to stage C (especially if accompanied by structural or functional abnormalities). Stress testing may therefore have increased the number of patients in the study of Carerj et al. who were reclassified as having stage B or stage C heart failure, and it may have predicted adverse outcomes. But it is unclear how many additional advanced stage patients could be discovered by such a strategy, and at what cost. The 2008 appropriateness criteria for stress echocardiography rate the performance of stress tests in asymptomatic patients with low and intermediate Framingham risk scores to be inappropriate stress testing in asymptomatic patients with high-risk scores to be of uncertain appropriateness. This designation, however, was specifically for the detection of coronary artery disease, not the identification of stage B or C heart failure (not addressed in the appropriateness criteria document). Presumably, the appropriateness task force would apply similar considerations to stress testing in heart failure.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
