Who Has Advanced Heart Failure? Definition and Epidemiology

Introduction

Heart Failure (HF) has been termed an “epidemic” but its epidemiology is complex due to the multiple factors which interact in a complex manner to impact the prevalence and incidence of HF.¹ Most HF-related hospitalizations and deaths are incurred by a subgroup of patients that is refractory to guideline-based medical management, a group categorized as having “advanced HF”. This cohort constitutes an important and rapidly expanding patient population that warrants comprehensive review as outlined in this series.

Epidemiology of HF

670,000 new cases of HF are diagnosed annually². Epidemiologic studies indicate that the age and sex adjusted incidence of HF has plateaued^{3, 4} over the last several decades. In contrast, the age and sex specific prevalence of HF appears to have increased from the 1970s to the early 2000's owing to improved survival of HF patients. Finally, the population at risk for HF has increased due to the aging of those persons who comprised the surge in births after the second World War (the baby boomers) and due to the improved management of cardiovascular and non-cardiovascular disease with more people surviving into the 7^th and 8^th decades where their risk of HF is highest⁵. This combination of influences on population demographics in the United States will result in a doubling of the number of people over 65 years of age (the population at risk for HF) over the next few decades and will result in a marked increase in the number of persons with HF despite stable age and sex specific incidence and only modest increases in age and sex specific prevalence due to increased survival of HF patients.

Natural History of HF defines advanced HF

The American College of Cardiology (ACC)/American Heart Association (AHA) task force on practice guidelines categorize the development and progression of the HF syndrome in to 4 stages termed Stage A-D of HF⁶ (Figure 1). Asymptomatic patients with cardiovascular risk factors but without any structural heart disease are considered to be high risk for developing HF and are classified as being in Stage A HF. Once structural heart disease develops, patients are considered to have Stage B HF despite the absence of recognized HF symptoms. Over time, the heart fails to maintain systemic perfusion commensurate with the metabolic requirements of organs or to do so only at the price of elevated filling pressures⁷ and patients develop overt symptoms of HF (Stage C). Etiologic factors are addressed and optimal medical therapy (OMT)⁶ is titrated resulting in a variable duration of compensation. A minority of patients have a very rapid progression and never achieve sustained compensation. A larger subset enjoy a prolonged period of relative compensation. However, over the long term, a progressive decline to advanced HF refractory to OMT (Stage D HF, defined below) is the rule. A subset of patients may experience marked improvement in symptoms with cardiac resynchronization therapy (CRT) for variable lengths of time resetting their progressive course upward. Another subset of patients may exit the “HF natural history” after receiving cardiac transplantation or a LV assist device (LVAD). Unfortunately, due to the realities of donor availability and ability to tolerate LVAD placement, this group currently represents a tiny subset of the HF population.

Figure 1. Natural History of HF defines advanced HF.

The stages in the natural history of HF clarify the relationship between cardiovascular disease (Stage A), asymptomatic abnormalities in cardiac structure and function (Stage B), overt symptomatic HF (Stage C) and advanced HF (Stage D). The natural history of over HF is further characterized in the insert where the fluctuating but progressive clinical course after HF presentation and commencement of optimal medical therapy (OMT) or later, with cardiac resynchronization therapy (CRT) or advanced HF therapies such as cardiac transplantation or ventricular assist devices (LVAD) is shown. Progression to poorer functional status associated with higher risk of hospitalization and overall health care resource utilization occurs over time. The potential modes of death over the course of the natural history of HF are emphasized where patients may die of non-HF related cardiovascular or non-cardiovascular causes (non-HF death, NHFD) or sudden cardiac death (SCD) at any time with the risk of SCD increasing as HF worsens. Patients who avoid NHFD or SCD die of progressive HF (HF).

Health care resource utilization increases dramatically during the Stage D. A recent community based study assessed health care utilization over the lifetime of HF patients after their initial HF diagnosis (Figure 2)⁸. This study demonstrated that in patients who survived the initial months after their HF diagnosis, resource utilization was most intense at the time of diagnosis and then in the last few months of life. The intensity of resource utilization was due to frequent hospitalizations and intensive out-patient visits. It is important to recognize that in the community, the most common mode of death in patients with a diagnosis of HF is non-HF related death (NHFD) due to coronary artery disease and non-cardiovascular comorbidities (Figure 1 and 3A⁹). Similarly, the most common causes of hospitalization over the course of a patient's life after a HF diagnosis are non-CV diseases followed by non-HF cardiovascular causes and HF (Figure 3B¹⁰). However, a significant subset of patients will die of fatal arrhythmias (sudden cardiac death, SCD) or progressive HF. The risk of SCD increases as the severity of HF increases.

Figure 2. Health care resource utilization over life after a HF diagnosis in the community.

The average monthly cost of health care is shown for patients over their lifetime after a diagnosis of HF. Note that resource utilization is most intense at the time of diagnosis (often made during a hospitalization) and again years after the diagnosis, where the advanced HF patient is at high risk for repeated hospitalizations for HF or other conditions⁸.

Figure 3. Causes of death and hospitalization for HF patients in the community.

The “epidemic” of HF occurs in elderly patients where the progression to advanced HF does not occur in isolation from a host of other cardiovascular and non-cardiovascular comorbidities. While HF clinical trials usually exclude patients with other cardiovascular and non-cardiovascular comorbidities, these conditions potently contribute to mortality⁹ (A) and hospitalizations¹⁰ (B) in the community as demonstrated in studies from a community based HF surveillance study in Olmsted County, Minnesota. In B, the cause of all hospitalizations over the patients life after a HF diagnosis (excluding the index HF hospitalization in patients in whom the diagnosis was made during a hospitalization for HF) were assessed.

What is Stage D “Advanced” HF?

Stage D HF designates patients with truly refractory HF symptoms and is heralded by a tenuous clinical course of progressive debilitating symptoms with decreasing level of activity (NYHA IIIB, IV), recurrent hospitalizations for volume overload, rhythm management and complications of HF and HF therapy (cardiorenal syndrome, medication side effects, pulmonary emboli, anti-coagulation complications and others) and marked increases in the level of outpatient visits in efforts to avoid hospitalizations. Advanced HF makes patients more susceptible to destabilization of other medical conditions increasing the hospitalization burden. Progression to advanced HF may be gradual and it is often difficult to determine whether decompensation in a previously stable HF patient is an isolated incident due to reversible precipitants or the transition to a refractory stage. To facilitate recognition of this phase so that advanced HF therapy (in eligible patients) or a more palliative approach to care can be considered, a number of definitions of advanced HF have been proposed.

Definitions of advanced HF in guidelines

The ACC/AHA guidelines and the Heart Failure Society of America guidelines have attempted to define advanced HF as summarized in Table 1.^{6, 11}. The Heart Failure Association of the European Society of Cardiology provides additional details in their criteria for advanced HF¹² (Table 2).

Table 1. Definition of Advanced HF Across Cardiovascular Societies.

	Refractory symptoms					Exercise Intolerance	Objective evidence of severe cardiac dysfunction
	Severe Symptoms	Multiple Hospitalizations	Optimal Therapy	Inotropic Support	Fluid Retention and/or Peripheral Hypoperfusion	Severe Functional Capacity Impairment*	Reduced Ejection Fraction	Doppler Echo†	Hemodynamics§	Elevated Natriuretic Peptides
ACC/AHA	x	x	x			x	x
HFSA	x	x	x	x		x
ESC	x	x	x		x	x	x	x	x	x

^*Assessed by cardiopulmonary exercise testing or six minute walk test

^†Pseudonormal or restrictive filling patterns

^§High left ventricular filling pressures

Abbreviations-ACC/AHA: American College of Cardiology/American Heart Association, HFSA: Heart Failure Society of America, ESC: European Society of Cardiology

Table 2. European Society of Cardiology Definition of Advanced HF.

• NYHA Class III-IV Symptoms

• Episodes of volume overload and/or peripheral hypoperfusion

• Objective evidence of severe cardiac dysfunction (EF<30%, Doppler Pseudonormal or Restrictive filling pattern, PCWP>16mmHg or RAP >12 mmHg)

• Severely impaired functional capacity (Inability to exercise, 6MWD<300m, Peak VO2<12-14 ml/kg/min)

• HF Hospitalizations (≥1 in past 6 months)

• Above occurring despite attempts to optimize diuretics, RAAS antagonists, BB, CRT or in the setting of intolerance to OMT

Abbreviations: NYHA, New York Heart Association; LV, left ventricular; EF, ejection fraction; PCWP, pulmonary capillary wedge pressure; RAP, right atrial pressure; 6MWD, 6 minute walk distance; VO₂ oxygen consumption; RAAS, renin-angiotensin-aldosterone system, BB, beta blockers, CRT, cardiac resynchronization therapy, OMT, optimal medical therapy

It is important to establish that Acute Decompensated HF (ADHF) is not always advanced HF. Patients commonly present with NYHA IV symptoms at their index HF presentation yet are not considered to have advanced disease; rather, they respond to treatment of etiologic factors and optimization of medical therapy and improve to NYHA classes I-II (Figure 1).

Secular trends in the definition of advanced HF used in clinical trials and registries

Evident in examination of advanced HF clinical trials^13-29,84 and registries^30-32 is the significant variability in the definition of advanced HF over the years, the higher event rates in registries than clinical trials and the marked variability in event rates in clinical trials of advanced HF - despite efforts to select more advanced HF patients by utilizing additional entry criteria (Table 3). The marked variability in outcomes in these trials and studies speaks to the difficulty in defining a homogenous population of advanced HF patients as well as the bias introduced by studying the “natural history” of HF in clinical trial populations.

Table 3. Secular Trends in Definition of Advanced HF in Clinical Trials and Registries.

STUDY	Intervention	YEAR	Background Therapy (used or available)	Definition of Advanced Heart Failure					Reported Mortality %				SCD
				EF	NYHA	Other criteria†	HF Hosp	Inotrope Use	3 mo	6 mo	≈ 12 mo	≈ 24 mo	% of total deaths
CONSENSUS(13)	RAS	1987	Diuretic/Dig	N/A	IV					44	52		21
PROMISE(14)	Oral Milrinone	1991	Diuretic/Dig/RAS	≤ 35 %	III-IV					24			na
PRAISE(15)	Amlodipine	1996	Diuretic/Dig/RAS	< 30 %	III-IV						38		43
FIRST(29)	Epoprostenol	1997	Diuretic/Dig/RAS	≤ 25 %	IIIB-IV					37			na
CIBIS-II (16)	BB	1999	Diuretic/Dig/RAS	≤ 35 %	III-IV						17		31
EPICAL(30)	ADVANCED HF REGISTRY	1999	Diuretic/Dig/RAS	<30%	III-IV	x	x				35		na
RALES(17)	Aldo Antag	1999	Diuretic/Dig/RAS	≤ 35 %	III-IV							46	21
BEST (28)	BB	2001	Diuretic/Dig/RAS	≤ 35 %	III-IV						19		45
COPERNICUS(18)	BB	2001	Diuretic/Dig/RAS	< 25 %	III-IV						17		na
MUSTIC(19)	CRT	2001	Diuretic/Dig/RAS	< 35 %	III	x				5			na
REMATCH(20)	LVAD	2001	Diuretic/Dig/RAS/BB/Inotrope	≤ 25 %	IV	x		x			75		0
MIRACLE (21)	CRT	2002	Diuretic/Dig/RAS/BB	≤ 35 %	III-IV	x				7			na
MIRACLE-ICD(22)	CRT-D	2003	Diuretic/Dig/RAS/BB	≤ 35 %	III-IV	x				8			21
COMPANION(23)	CRT or CRT-D	2004	Diuretic/Dig/RAS/BB/Aldo Antag	≤ 35 %	III-IV	x	x				19		na
CARE-HF(84)	CRT	2005	Diuretic/Dig/RAS/BB/Aldo Antag	≤ 35 %	III-IV	x						30	33
ESCAPE (24)	PA CATH Management	2005	Diuretic/Dig/RAS/BB/Aldo Antag/Inotrope	≤ 30 %	III-IV		x			17			na
FUSION II (25)	Outpatient Nesiritide	2008	Diuretic/Dig/RAS/BB/Aldo Antag/ ICD/CRT	< 40%	III-IV	x	x		10				na
PERSIST(26)	Levosimendan	2008	Diuretic/Dig/RAS/BB/Aldo Antag/Inotrope	≤ 30 %	IIIB-IV	x	x	x	4				52
ADHERE LM (32)	ADVANCED HF REGISTRY	2008	Diuretic/Dig/RAS/BB//Inotrope/ICD/CRT	mean 30%	III-IV	x	x	x		28			17

^†QRS duration, LV dilatation, Cardiopulmonary exercise testing, ectopy, renal dysfunction, BNP

Abbreviations: RAS, renin angiotensin system antagonist; BB, beta blocker; LVAD, left ventricular assist device; CRT, cardiac resynchronization therapy; PA CATH, pulmonary artery catheterization; CPXT, cardiopulmonary exercise test; mo, month; SCD, sudden cardiac death;; Hosp, hospitalization, Aldo, aldosterone; ICD, implantable cardioverter defibrillator; Antag, antagonist, CRT-D, CRT-defibrillator; dig, digoxin

Registries specifically aimed at advanced HF patients allow insight into the characteristics of the advanced HF patient outside the context of clinical trials. ADHERE-LM was a multicenter, observational registry of 1433 patients with advanced HF defined as adults (≥18) with chronic HF and refractory symptoms on oral medical therapy that were required: (1) to be in NYHA functional class III or IV for ≥60 consecutive days and (2) hospitalized ≥2 times in the preceding year with either a primary diagnosis of HF or a secondary diagnosis of HF treated with ≥2 consecutive days or IV diuretic, vasoactive, or inotropic medications, or to have required either 2 complete IV infusions of a vasoactive or inotropic agent, each lasting ≥2 hrs or 3 IV diuretic treatments, given either as a bolus or continuous drip, during the preceding 60 days.³² As compared to the previous ADHERE ADHF patients, advanced HF patients tended to be younger, more often male and have more coronary artery disease and chronic renal insufficiency. Fatigue rather than dyspnea and edema were more common on history while physical examination revealed a lower blood pressure with less evidence of volume overload. Although systolic dysfunction was more severe in advanced HF, serum BNP levels appeared similar between the acutely decompensated and advanced HF groups. Advanced HF patients were more aggressively treated. With the exception of the REMATCH study²⁰, the patients in ADHERE-LM had a much higher event rate than patients enrolled in clinical trials of advanced HF.

These registry and cohort studies highlight that advanced HF is a discrete clinical syndrome that deserves recognition as its own entity. However, the marked variability in event rates despite similar entry criteria in clinical trials and registries of advanced HF indicate the difficulty in prospectively identifying patients with advanced HF.

Recognizing the advanced HF patient

There is no one feature which reliably identifies the advanced HF patient. Recognizing advanced HF requires integration of clinical, imaging, hemodynamic, functional and biomarker data. The presence of multiple clinical, imaging, hemodynamic, functional and biomarker parameters associated with increased risk in the setting of OMT should raise concern over progression to Stage D HF. Risk scores may provide a more objective tool to facilitate such integration.

Symptoms

While dyspnea is often a characteristic feature, many advanced HF patients have a more prominent component of right ventricular failure and may complain more prominently of fatigue, edema or abdominal distension due to ascites. Sleep disturbances due to HF symptoms or central sleep apnea, depression and cardiac cachexia are common. Intolerance (hypotension, worsening renal function, hyperkalemia) to previously tolerated doses of RAAS antagonists and beta blockers is another feature suggestive of advanced HF.

Hospitalizations

Repeated HF hospitalizations are an important but not invariant feature of advanced HF. The ADHERE LM registry revealed that 59% of advanced HF patients were hospitalized during the mean follow-up of 364 days. Of hospitalized patients, 37% were hospitalized once, 23% twice, 14% 3 times, 9% 4 times and 18% 5 or more times³². The European EPICAL registry of 2,577 patients with advanced HF revealed that patients were admitted to the hospital an average of 2.05 times per year spending 27.6 days per year in the hospital over a mean follow-up of 18 months³⁰. Each subsequent HF hospitalization is associated with incremental increase in risk of death.³³ Thus, one or more HF hospitalizations in a previously stable chronic HF patient may herald the onset of advanced HF and should alert the care provider to a potential change in the course of the patient's HF.

Doppler Echocardiography

The prognostic implications of a variety of Doppler echocardiographic features in variable HF populations have been previously described^34-53. Ejection Fraction (EF) has been shown to maintain prognostic power even among those with severe systolic dysfunction (LVEF<25%)^{41, 42}. Left ventricular size has also been shown to offer prognostic information among advanced HF patients^{40, 43-45}.

Right ventricular (RV) function is a potent predictor of survival in advanced HF patients when invasively assessed with radionuclide ventriculography^{46, 47}. As RV EF is difficult to assess echocardiographically, a number of newer quantitative indices are being investigated. Tricuspid Annular Plane Systolic Excursion (TAPSE), RV shortening, RV fractional area change, and peak systolic tricuspid annular velocity have each been shown to classify a high risk group of HF patients with recurrent HF hospitalizations, poor functional capacity and high mortality^{38, 48-50}. Repeatedly, the presence of pulmonary hypertension as measured by tricuspid regurgitant velocity ≥ 2.5 m/s (particularly in the presence of RV dysfunction) has been shown to be a potent predictor of poor outcomes in HF patients⁵⁴.

Severe diastolic dysfunction is a potent prognostic factor. A meta-analysis of over 3500 patients (18 studies) revealed that regardless of ejection fraction and age, a restrictive mitral inflow pattern is a powerful predictor of mortality in HF³⁶.

The presence of left and right atrial and ventricular dilatation and in ischemic patients, tethering of the mitral chordal apparatus, results in functional or ischemic mitral and functional tricuspid regurgitation in the absence of intrinsic valve disease. Distortion of the tricuspid valvular apparatus by ICD leads can dramatically increase tricuspid regurgitation. Both mitral and tricuspid regurgitation are associated with worse outcomes in HF⁵⁵ and are markers of advanced HF. Whether correction of functional AV valve regurgitation will change the course of advanced HF remains an important unresolved controversy.

Hemodynamics

Hospitalizations and deaths are predicted by elevated filling pressures as reflected by right atrial pressures (>10 mmHg) or pulmonary capillary wedge pressures (> 20 mmHg)^{56, 57}. Moreover, recent data from the COMPASS-HF cohort revealed that many advanced HF patients on OMT have elevated ambulatory intracardiac pressures that translates into a higher risk of hospitalization.⁵⁸ Other parameters such as pulmonary vascular resistance (>3-4 wood units), mean pulmonary arterial pressures (> 35 mmHg) and cardiac index (< 2.0) may also be used to risk stratify the advanced HF patient. The presence of such hemodynamic perturbations in the presence of all attempts to maximize medical therapy is a key feature of the advanced HF state.

Biomarkers

Hyponatremia

Hyponatremia is a simple and long-recognized poor prognostic factor in HF⁵⁹.

Natriuretic Peptides

Brain-type natriuretic peptide (BNP) and its amino-terminal fragment, NT-proBNP have been shown to predict outcomes in a variety of HF cohorts with higher values conferring increased risk⁶⁰, although some smaller studies have suggested that natriuretic peptide (NP) activity is exhausted in advanced HF⁶¹. As yet, no study has identified a single value of BNP or NT-proBNP that reliably identifies advanced HF ⁶²or discriminates between an acute decompensation versus a progressive down-hill course as underscored by the ADHERE-LM study summarized above³².

Uric acid (UA)

A product of xanthine oxidase (XO) and marker of oxidative stress, UA is elevated in advanced HF patients and has been strongly associated with worsening clinical status and impaired survival⁶³.

Renal dysfunction

Renal dysfunction has repeatedly been demonstrated to be a potent independent predictor of mortality in HF^{64, 65} and was present in over 50% of advanced HF patients in ADHERE-LM³² (mean creatinine:1.8, mean BUN:42) and REMATCH⁶⁶ (mean Cr. 1.8).

Other biomarkers reflecting inflammation (C-reactive protein, tumor necrosis factor receptor), neurohormonal activation (norepinephrine, angiotensin II and plasma rennin activity) and myocyte injury (troponin) may be elevated in advanced HF. However these biomarkers are currently not recommended for prognostic purposes in clinical practice. Anemia has long been recognized to be independently associated with adverse HF outcomes⁶⁷.

Functional capacity

Peak VO2 (<50% of age and sex specific normal values), decreased anaerobic threshold (<10ml/kg/min) and increased VE/VCO2 slope (>34) identifies high-risk HF patients with poor 1 year survival. A 6 minute walk test distance <300 m identifies severe impairment and poor outcomes but may be less reliable in elderly patients or those with comorbidities. Frailty indices may be particularly helpful in the elderly but have yet to be adopted widely in clinical practice.

Risk scores

Seattle Heart Failure Model (SHFM)

The SHFM is a mortality prediction model based on a broad range of clinical, pharmacological, device and laboratory characteristics⁶⁸ and has been validated in a number of advanced HF cohorts ^{69, 70, 85}. While additional consideration of inotrope or intra-aortic ballon pump use and care in elderly patients may be needed, the SHFM shows promise as an aid in the prospective identification of advanced HF.

Heart Failure Survival Score (HFSS)

Derived in HF patients referred for evaluation of severe HF and/or cardiac transplantation⁷¹, the HFSS has since been validated in several cohorts^72-75. This model incorporates resting heart rate, mean blood pressure, EF, serum sodium, peak VO2, HF etiology (ischemic vs nonischemic), and intraventricular conduction delay (QRS ≥120 msec). The HFSS stratifies patients into low (HFSS ≥ 8.10), medium (7.20 ≤ HFSS < 8.09) or high (HFSS ≤ 7.19) risk with an average annual VAD or transplant free survival of approximately 87%, 68%, and 44%^72-74.

ESCAPE score

The Evaluation Study of Congestive Heart Failure and Pulmonary Artery Catheterization Effectiveness Risk Model and Discharge Score (ESCAPE) trial of patients hospitalized with advanced HF²⁴ defined a model with 8 clinical variables: age (>70), BUN(>40 and >90), 6-min walk (<300 m), serum sodium (<130 meq/L), CPR/mechanical ventilation, diuretic dose at discharge (>240mg), absence of beta blocker at discharge and discharge BNP (>500 pg/mmol or >1300 pg/mmol)⁷⁶. The ESCAPE model was found to identify risk of death in the 6 months following discharge where patients with a score of 0 had a 5% 6-mortality rate and those with a score of 8 hand a 94% 6-mortality rate.

Further stratification of patients with advanced HF

Derived from patients receiving FDA approved mechanical circulatory support devices, the Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) database⁷⁷ defines specific subsets within advanced HF patients (Table 4). Subsequent studies have shown the prognostic significance of these profiles for patients undergoing LAVD ^{78, 79,80} with patients in patient profile 4-7 having significantly better survival and shorter lengths of stay post-implantation compared to those in patient profiles 1-3⁷⁸.

Table 4. INTERMACS Advanced HF Profiles.

Profile 1: Critical cardiogenic shock Patients with life-threatening hypotension despite rapidly escalating inotropic support, critical organ hypoperfusion, often confirmed by worsening acidosis and/or lactate levels. “Crash and burn.”

Profile 2: Progressive decline Patient with declining function despite intravenous inotropic support, may be manifest by worsening renal function, nutritional depletion, inability to restore volume balance “Sliding on inotropes.” Also describes declining status in patients unable to tolerate inotropic therapy.

Profile 3: Stable but inotrope dependent Patient with stable blood pressure, organ function, nutrition, and symptoms on continuous intravenous inotropic support (or a temporary circulatory support device or both), but demonstrating repeated failure to wean from support due to recurrent symptomatic hypotension or renal dysfunction “Dependent stability.”

Profile 4: Resting symptoms Patient can be stabilized close to normal volume status but experiences daily symptoms of congestion at rest or during ADL. Doses of diuretics generally fluctuate at very high levels. More intensive management and surveillance strategies should be considered, which may in some cases reveal poor compliance that would compromise outcomes with any therapy. Some patients may shuttle between 4 and 5.

Profile 5: Exertion intolerant Comfortable at rest and with ADL but unable to engage in any other activity, living predominantly within the house. Patients are comfortable at rest without congestive symptoms, but may have underlying refractory elevated volume status, often with renal dysfunction. If underlying nutritional status and organ function are marginal, patient may be more at risk than INTERMACS 4, and require definitive intervention.

Profile 6: Exertion limited Patient without evidence of fluid overload is comfortable at rest and with activities of daily living and minor activities outside the home but fatigues after the first few minutes of any meaningful activity. Attribution to cardiac limitation requires careful measurement of peak oxygen consumption, in some cases with hemodynamic monitoring to confirm severity of cardiac impairment. ”Walking wounded.“

Profile 7: Advanced NYHA III A placeholder for more precise specification in future, this level includes patients who are without current or recent episodes of unstable fluid balance, living comfortably with meaningful activity limited to mild physical exertion.

Epidemiology of Advanced HF

Experts have estimated the prevalence of advanced HF to range from 6% to 25% of the HF population^{81, 82, 80}. With the paucity of contemporary epidemiologic studies specifically assessing the prevalence of advanced HF patients using a uniform definition, these claims are not supported by rigorous data. ADHERE LM, the largest available registry of patients with chronic advanced HF, suggested that roughly 5% of patients with HF have end-stage disease with symptoms refractory to guideline-based medical therapy³² while expert opinion places this number at few hundred thousand Americans^{81, 83}. EPICAL, an observational, community-based, European cohort study of advanced HF patients reported a crude incidence rate of 225 per million, that increases dramatically with increasing age with men more susceptible than women³⁰. Given the exclusion of patients over 80 in EPICAL, this is likely an underestimation and extrapolation to the overall advanced HF population should be done with caution. While the true overall and within HF prevalence of advanced HF remains uncertain, the intense health care utilization and poor outcomes of advanced HF patients establish advanced HF as a major public health burden. While not specifically emphasized in this review, many of the principals related to the recognition and the societal impact of advanced HF are likely also pertinent to patients with HF and preserved EF (HFpEF) although the epidemiology and characteristics of the advanced HFpEF patient are even more poorly characterized.