- 9 Downloads
Heart failure (HF) is a clinical syndrome resulting from structural or functional impairment of ventricular filling or ejection of blood, characterized by typical symptoms (dyspnea, fatigue, and fluid retention) and/or signs (elevated jugular venous pressure, pulmonary rales, and peripheral edema).
Classifications of HF
HF is classified in different ways according to its development, etiology, pathophysiology, and so on. According to the anatomical location, HF is classified as three types: left heart failure, right heart failure, and whole heart failure. Based on the course of disease, HF is divided into two types: acute heart failure and chronic heart failure and is classified systolic heart failure and diastolic heart failure according to its cardiac function.
ACCF/AHA (American College of Cardiology Foundation /American Heart Association) classification of HF emphasizing the development of the disease: (a) Stage A: At high risk for HF without structural heart disease or symptoms of HF; (b) Stage B: Structural heart disease without signs or symptoms of HF; (c) Stage C: Structural heart disease with prior or current symptoms of HF; and (d) Stage D: Refractory HF.
The NYHA(New York Heart Association)classification for HF is based on cardiac function: (a) Class I: No limitation of physical activity; (b) Class II: Slight limitation of physical activity; (c) Class III: Marked limitation of physical activity; and (d) Class IV: Unable to carry on any physical activity.
According to the time course, HF with symptoms and signs generally remain unchanged for over 1 month is called “chronic stable HF”; deteriorated chronic stable HF is described as “decompensated chronic HF”; HF diagnosed for the first time is called “new-onset HF.”
The etiology of HF is diverse within and among world regions. There is no agreed single classification system according to the causes of HF. Briefly, the causes of HF can be classified by diseased myocardium (e.g., ischemic heart disease, toxic damage, metabolic derangements, and genetic abnormalities), abnormal loading conditions (e.g., hypertension, valve and myocardium structural defects, pericardial and endomyocardial pathologies, high output states, and volume overload), and arrhythmias. Infection and arrhythmia are the most common and major inducing factors.
Currently, in clinical medicine, HF is often divided by left ventricular ejection fraction (LVEF), which can be objectively measured. HF patients with normal LVEF (≥50%) are called HF with preserved EF (HFpEF); those with reduced LVEF (<40%) are called HF with reduced EF (HFrEF). Patients with an LVEF in the range of 40–49% represent a “gray area,” which are now defined as HF with mid-range EF (HFmrEF) (Ponikowski et al. 2016).
Characteristics of HF in Older People
The prevalence of HF can be estimated at 1–2% in developed countries and the incidence approaches 5–10 per 1000 persons per year, and it rises among people aged >65 years (Dharmarajan and Rich 2017). The mortality of HF is apparently increased as age increases evident from a recent study showing that 1 and 5-year-mortality increased from 7.4% and 24.4% for 60-year-olds to 19.5% and 54.4% for 80-year-olds, respectively (Dharmarajan and Rich 2017). As for hospitalization, data from Medicare beneficiaries have shown that hospitalization rates for HF have decreased for older people from 2000 to 2010, while hospitalization for HF continues to predominantly affect older people. In 2010, more than 70% of hospitalizations for HF were among adults aged 65 years and older (Dharmarajan and Rich 2017). Older people have more risk factors for HF, more comorbidities, and are more prone to have recurrent admissions for acute decompensations of cardiac function (Bader et al. 2017). Almost 25% of older people with HF are rehospitalized within 1 month of discharge and almost 70% are rehospitalized within 1 year. However, readmissions after hospitalization for HF are usually not for HF but for non-cardiovascular conditions.
Comorbidities of HF in Older People
A study revealed that 60% of older HF patients have three or more comorbidities while only 2.5% suffer none (Murad et al. 2015). The superimposed effect of comorbidities makes HF in older people progress more rapidly under an equivalent clinical severity. Hypertension is the biggest culprit as associated comorbidity with 82% prevalence. Coronary heart disease has been proved to be associated with increased mortality in both HFrEF and HFpEF with 60% comorbidity prevalence (Murad et al. 2015; Hwang et al. 2014). Other cardiovascular comorbidities include arrhythmias, peripheral vascular disease, and cardiac valvular disease. As for non-cardiac comorbidities which are equally crucial, diabetes mellitus and obesity are the most common, and other comorbidities including chronic kidney disease, sleep apnea, anemia, malnutrition, depression, gout and arthritis, and cognitive impairment. Diabetes mellitus, depression, and chronic kidney disease increase the mortality in older HF patients independently.
Key Research Findings
Treatments of HF
Pathophysiology of HF
Ventricular remodeling is the responsive pathophysiology of HF to stresses in different disease settings. Maladaptive activations of the sympathetic nervous system (SNS) and the renin-angiotensin-aldosterone system (RAAS) trigger the initiation of ventricular remodeling by activating the downstream molecular signaling pathways. Ventricular remodeling includes the alterations of cardiomyocytes, interstitial space, and vascular system (Burchfield et al. 2013). (1) Hypertrophic growth of cardiomyocytes is featured as thickening ventricular wall, increased cardiac mass, myocytes area, protein synthesis, and fetal gene reexpression. Persistent occurrence of cardiac hypertrophy results in cellular death that contributes most to the transformation of ventricular remodeling pattern from concentric to eccentric and phenotype of low cardiac output in HF. (2) Myofibroblasts differentiated from fibroblasts secrete collagen I and III that deposit in interstitial space, leading to diminished oxygen diffusion capacity and decreased compliance. (3) Vascular remodeling manifested as decreased capillary density and increased vessel stiffness, renders the failing heart an ischemic organ by reducing myocardial perfusion. The pathophysiologies above are the basis of therapeutic targets explorations of HF. The drugs widely used clinically have been proven to ameliorate the mortality of HF via inhibiting the progression of ventricular remodeling.
Pharmacological Treatments of HF
As the SNS and the RAAS activated in HF exert detrimental effects, blockade of the neurohormonal system is the foundation of HFrEF treatment. Angiotensin-converting enzyme inhibitors (ACEIs) reduce the production of angiotensin II to prevent cardiac remodeling and thereby significantly reduce the risk of adverse clinical outcomes among HFrEF patients (Garg and Yusuf 1995). Angiotensin II type I receptor blockers (ARBs) can bind the AT1 receptor of angiotensin and block its detrimental effects. ARBs treatment also reduces the risk of mortality or HF hospitalization in HFrEF (Granger et al. 2003).
Beta-blockers attenuate the effect of sympathetic nervous system activation. Selective β1-receptor blockers metoprolol and bisoprolol, and nonselective beta-blocker carvedilol have been proved to further improve survival of HFrEF patients in addition to ACEIs (Packer et al. 2001; Hjalmarson et al. 2000; CIBIS-II Investigators and Committees 1999).
Aldosterone is also a strong inducer of cardiac remodeling. Spironolactone, a mineralocorticoid receptor antagonist (MRA), blocks the receptor of aldosterone, and reduces the risk of mortality in HFrEF patients, despite treatment of an ACEI and a beta-blocker (Pitt et al. 1999). In addition, eplerenone is a newly developed selective MRA that also reduces the risk of cardiovascular mortality or hospitalization for HF in HFrEF patients (Zannad et al. 2011).
Ivabradine blocks the If channel in the sinus node. On the base of optimal beta-blocker treatment, it can further reduce heart rate. A large randomized trial demonstrated that ivabradine treatment improves clinical outcomes in HFrEF with sinus rhythm and a heart rate >70 bpm (Swedberg et al. 2010).
The discovery of angiotensin receptor neprilysin inhibitor (ARNI) is a major breakthrough in HFrEF treatment. Inhibition of neprilysin not only raises the circulating natriuretic peptides but also activates the RAAS (Richards et al. 1993), which is the rationale of dual-acting compounds that inhibit neprilysin and block the effect of angiotensin. LCZ696 consists of the neprilysin inhibitor sacubitril and the ARB valsartan. The recent PARADIGM-HF trial demonstrated that LCZ696 is superior to enalapril, an ACEI, in reducing risks of death and hospitalization for HF (Ruilope et al. 2010; McMurray et al. 2014).
Except for blockade of RAAS and SNS, improving cardiac metabolism by Coenzyme Q10 can also improve the clinical outcome of chronic HF (Mortensen et al. 2014). The effect of diuretics on long-term outcome in HFrEF has not been evaluated in any random clinical trial, but they are the cornerstone in symptom-relieving treatment and maintenance of euvolemia.
HFmrEF and HFpEF
As most of the trials designed for HFpEF included HFmrEF patients, we use “HFpEF” to describe those patients in this section. Targeting RAAS and SNS has been successful in HFrEF, but no clinical trials of these therapies revealed mortality or morbidity benefit (Yamamoto et al. 2013; Cleland et al. 2006; Yusuf et al. 2003; Pitt et al. 2014).
In a phase 2 trial, sacubitril/valsartan significantly reduces the plasma level of NT-proBNP compared with valsartan from baseline to 12 weeks (Solomon et al. 2012). The ongoing large outcome trial PARAGON-HF (NCT01920711) might be able to illustrate the effect of ARNI on long-term prognosis of HFpEF.
A new paradigm has been put forward for HFpEF, in which inactivation of the nitric oxide (NO) – cyclic guanosine 3,5-monophosphate (cGMP) – protein kinase-G (PKG) pathway caused by a systematic pro-inflammatory state is believed to play a central role (Paulus and Tschope 2013). Sildenafil that block cGMP degradation did not result in the improvement of peak oxygen consumption after 24 weeks in RELAX trial in HFpEF (Guazzi et al. 2011).
Up to date, natriuretic peptides are still the most useful biomarkers in terms of HF diagnosis. B-type natriuretic peptide (BNP) is secreted by cardiac ventricles in response to volume expansion and cardiomyocyte stretch. It is synthesized as a 134-amino acid preproBNP encoded by the human gene NPPB. Removal of the 25-residue N-terminal peptide generates proBNP which is subsequently cleaved into NT-proBNP and the biologically active BNP. Both BNP and NT-proBNP show capability in establishing or excluding HF in patients with acute dyspnea (Maisel et al. 2002; Januzzi et al. 2006). However, as NT-proBNP has high sensitivity, but not specificity, it is more reasonable to use it for ruling out HF, instead of establishing a diagnosis.
BNP and NT-proBNP are also the most commonly used biomarker for prognosis evaluation in HF patients (Masson et al. 2008). Numerous novel biomarkers have been identified to evaluate risk stratification of HF. These biomarkers reflect different aspects of HF pathophysiology, such as myocardial injury (troponin) (Latini et al. 2007), fibrosis (soluble ST2) (Ky et al. 2011), and apoptosis (growth differentiation factor-15) (Anand et al. 2010). Most of these biomarkers had prognostic value independent of NT-proBNP. However, the use of these novel biomarkers is still limited in clinical practice.
While NT-proBNP is one of the most useful biomarkers in both HF diagnosis and prognosis, researchers also believed that adjusting HF treatment to achieve a low natriuretic peptide level will further reduce the adverse outcomes. Clinical trials on this issue showed conflicting results (Troughton et al. 2014). However, a recent randomized trial that included more than 900 patients showed that NT-proBNP-guided treatment was actually not more effective than usual care in improving outcomes in HFrEF patients (Felker et al. 2017).
Nondrug Therapy of HF
Despite of adequate drug therapy, the mortality of patients with advanced HF is still high. Therefore, nondrug therapy also plays an important role in medical care of HF through improving electrical activity and cardiac contractile function with the utilization of assist devices and heart transplantation.
Ventricular contractility is not synchronized due to bundle branch block which could worsen HF. Cardiac resynchronization therapy (CRT) can be used to improve the cardiac pumping efficiency by pacing two ventricles to promote the synchronism of left and right ventricular and alleviate the symptoms of HF patients. The key indications for CRT are QRS prolongation of >130 ms while left bundle branch block or >150 ms while non-left bundle branch block on the electrocardiogram and the reduced left ventricle ejection fraction (<35%) (Ponikowski et al. 2016).
The left-ventricle assist device (LVAD) which is an electromechanical device can assist cardiac circulation in advanced HF. It pumps left ventricular blood flow into the auxiliary pump body then drives the blood flow into the aorta to improve the left heart blood pump function. For the patients who shall not undergo heart transplantation, the LVAD is what they live for the remainder of their life. Other cardiac assist devices such as Impella, extracorporeal membrane oxygenation (ECMO), and centriMag are also used in certain conditions for mechanical circulatory support in HF.
For end-stage HF, heart transplantation is a surgical transplant procedure when medical or surgical treatments have failed. Post-operation survival period averages of 15 years. Heart transplantation is not considered to be a cure for heart disease but a life-saving treatment intended to improve the quality of life for recipients.
With the deep understanding of mechanism and risk factors, HF is not only a treatable but also a preventable disease. Blood pressure control, smoking cessation, and enhanced physical exercise are believed to reduce the risk of HF development. Given the clinical utility of BNP in HF management, its value in primary prevention of HF was also investigated. The STOP-HF trial showed that among patients at risk of heart failure, BNP-based screening, and collaborative care reduced the combined rates of LV systolic dysfunction, diastolic dysfunction, and heart failure (Ledwidge et al. 2013). Several medications, such as statins, ACEI, and inhibitors of sodium-glucose cotransporter 2 (SGLT2), also had evidence for HF prevention in specific populations.
Future Directions of Research
Metabolic Remodeling of HF
Metabolic remodeling emerges in the failing heart (Bertero and Maack 2018). Under normal conditions, cardiac ATP is mainly derived from fatty acid oxidation (FAO) while glucose oxidation comes second, and glycolysis provides least. Basic research evidence showed that FAO decreases in metabolic remodeling, therefore, glucose oxidation and glycolysis which require less oxygen to produce the same amount of ATP increases as compensation. Downregulation of transcriptional regulators such as peroxisome proliferator-activated receptor alpha (PPARα) and PPARG coactivator 1 (PGC-1) is involved in metabolic remodeling. Given the important role of metabolic remodeling in HF, therapeutic approach targeting metabolic derangements seems to promise a considerable prospect. Decreasing oxygen requirements or increasing ATP might be the two therapeutic goals of HF treatment. Drugs inhibiting FA metabolism or promoting glucose oxidation so that oxygen requirement would be less to produce the same amount of ATP might be profitable theoretically. However, clinical experiments failed to inspire. FA metabolism inhibitor perhexiline (Beadle et al. 2015) and trimetazidine (Fragasso et al. 2006) improved symptoms and LVEF, respectively, in HF patients but effective clinical trials are lacking. PPAR agonists reducing FA supply, like fibrates and thiazolidinediones, failed to improve cardiac function (Jun et al. 2012; Dormandy et al. 2005). Based on the disappointing results above, directly promoting glucose oxidation rather than inhibiting FA metabolism seems to be another research direction in the future. However, dichloroacetate, a pyruvate dehydrogenase activator that stimulates glucose oxidation, failed to have consistently clinical benefit in clinical trials (Lewis et al. 1998). More evidence and target explorations are needed in this regard. Further study should not only be focused on the typical metabolic substrate itself but also be broadened to new substrate exploration and comprehensive control of cardiac metabolism.
Treatment of HFpEF
HFpEF accounts for up to 50% of all HF cases, with similar prognosis to HFrEF (Owan et al. 2006). Unfortunately, there is still no effective treatment that can improve the clinical outcome of HFpEF. Unlike HFrEF, HFpEF patients could not benefit from the blockade of RAAS. It would be possible that it is less a neurohormonal-driven disease compared with HFrEF (Sharma and Kass 2014). Pathophysiology of HFpEF is complicated and therefore the mechanism of HFpEF deserves more research in the future. However, animal models of HFpEF available focusing on only one or two features of HFpEF (Sharma and Kass 2014) and finding a truly representative disease model could be the priority of this field.
A great number of studies reported that HFpEF is a highly heterogeneous disease. First, the cut-off value of EF in HFpEF varied from 40%, 45%, to 50% (Komajda and Lam 2014). More uniformed inclusion criteria should be used in future HFpEF trials. Second, a great proportion of HFpEF patients do not have the same abnormalities (such as left ventricular hypertrophy, left atrial dilation, and diastolic dysfunction) as HFrEF (Ponikowski et al. 2016). Identification of distinct phenotypes is of great importance in future HFpEF research. Some researchers proposed treatment strategies according to clinical presentation and predisposition of HFpEF (Senni et al. 2014; Shah et al. 2016) but most of them lack convincing evidence. The differences in the underlying mechanism and response to treatment need to be further explored. Given its great heterogeneity, future laboratory and clinical studies should be focused on a specific HFpEF subgroup based on previous clues.
Nondrug Therapy of HF
CRT has improved the symptom and electrical dyssynchrony in HF patients. In the future, ongoing studies will expand the patient population benefited from CRT by exploring clinical factors influencing the therapeutic response. Furthermore, studies might further identify novel pacing technologies that augment the efficacy of resynchronization.
Heart transplantation has become the preferred standard therapy for patients with end-stage HF. But problems including rejection, a supply of donor, infection, and else limit the prevalence of heart transplantation. Many relative studies are in progress to solve these problems. As for rejection, cardiac magnetic resonance with high sensitivity and high negative predictive value in predicting biopsy-positive heart transplant rejection might be utilized for screening before endomyocardial biopsy (Butler et al. 2015). To solve the donor shortage problem, research on xenotransplantation are also in progress (Ekser et al. 2017).
HF is one of the common cardiovascular diseases in older people. The incidence, mortality, and rehospitalization rate of HF rise as age increases. HF in older people has various comorbidities, which aggravates the difficulty of treatments and worsens its prognosis. Natriuretic peptides are still the most useful biomarkers in terms of HF diagnosis and prognosis, although numerous novel biomarkers are identified to evaluate risk stratification. Ventricular remodeling and metabolic remodeling are the pathophysiological basis of HF pharmacological therapeutic targets. Although blockage of the SNS and the RAAS is the foundation of HFrEF treatment, there is still no effective medication for HFpEF. Nondrug therapy is another option for HFrEF or end-stage HF. In the future, the extra effort might be focused on the treatment for HFpEF and the improvement of nondrug therapy, and metabolic remodeling might be another direction that the new therapy should target.
- Anand IS, Kempf T, Rector TS, Tapken H, Allhoff T, Jantzen F et al (2010) Serial measurement of growth-differentiation factor-15 in heart failure: relation to disease severity and prognosis in the Valsartan heart failure trial. Circulation 122:1387–1395. https://doi.org/10.1161/CIRCULATIONAHA.109.928846CrossRefGoogle Scholar
- Burchfield JS, Xie M, Hill JA (2013) Pathological ventricular remodeling: mechanisms: part 1 of 2. Circulation 128:388–400. https://doi.org/10.1161/CIRCULATIONAHA.113.00.1878CrossRefGoogle Scholar
- Butler CR, Savu A, Bakal JA, Toma M, Thompson R, Chow K et al (2015) Correlation of cardiovascular magnetic resonance imaging findings and endomyocardial biopsy results in patients undergoing screening for heart transplant rejection. J Heart Lung Transplant 34:643–650. https://doi.org/10.1016/j.healun.2014.12.020CrossRefGoogle Scholar
- Dormandy JA, Charbonnel B, Eckland DJ, Erdmann E, Massi-Benedetti M, Moules IK et al (2005) Secondary prevention of macrovascular events in patients with type 2 diabetes in the PROactive study (PROspective pioglitAzone clinical trial in macroVascular events): a randomised controlled trial. Lancet 366:1279–1289. https://doi.org/10.1016/S0140-6736(05)67528-9CrossRefGoogle Scholar
- Felker GM, Anstrom KJ, Adams KF, Ezekowitz JA, Fiuzat M, Houston-Miller N et al (2017) Effect of natriuretic peptide-guided therapy on hospitalization or cardiovascular mortality in high-risk patients with heart failure and reduced ejection fraction: a randomized clinical trial. JAMA 318:713–720. https://doi.org/10.1001/jama.2017.10565CrossRefGoogle Scholar
- Fragasso G, Perseghin G, De Cobelli F, Esposito A, Palloshi A, Lattuada G et al (2006) Effects of metabolic modulation by trimetazidine on left ventricular function and phosphocreatine/adenosine triphosphate ratio in patients with heart failure. Eur Heart J 27:942–948. https://doi.org/10.1093/eurheartj/ehi816CrossRefGoogle Scholar
- Granger CB, McMurray JJ, Yusuf S, Held P, Michelson EL, Olofsson B et al (2003) Effects of candesartan in patients with chronic heart failure and reduced left-ventricular systolic function intolerant to angiotensin-converting-enzyme inhibitors: the CHARM-alternative trial. Lancet 362:772–776. https://doi.org/10.1016/S0140-6736(03)14284-5CrossRefGoogle Scholar
- Hjalmarson A, Goldstein S, Fagerberg B, Wedel H, Waagstein F, Kjekshus J et al (2000) Effects of controlled-release metoprolol on total mortality, hospitalizations, and well-being in patients with heart failure: the Metoprolol CR/XL randomized intervention trial in congestive heart failure (MERIT-HF). MERIT-HF study group. JAMA 283:1295–1302CrossRefGoogle Scholar
- Januzzi JL, van Kimmenade R, Lainchbury J, Bayes-Genis A, Ordonez-Llanos J, Santalo-Bel M et al (2006) NT-proBNP testing for diagnosis and short-term prognosis in acute destabilized heart failure: an international pooled analysis of 1256 patients: the international collaborative of NT-proBNP study. Eur Heart J 27:330–337. https://doi.org/10.1093/eurheartj/ehi631CrossRefGoogle Scholar
- Ky B, French B, McCloskey K, Rame JE, McIntosh E, Shahi P et al (2011) High-sensitivity ST2 for prediction of adverse outcomes in chronic heart failure. Circ Heart Fail 4:180–187. https://doi.org/10.1161/CIRCHEARTFAILURE.110.958223CrossRefGoogle Scholar
- Ledwidge M, Gallagher J, Conlon C, Tallon E, O’Connell E, Dawkins I et al (2013) Natriuretic peptide-based screening and collaborative care for heart failure: the STOP-HF randomized trial. JAMA 310:66–74. https://doi.org/10.1001/jama.2013.7588
- Mortensen SA, Rosenfeldt F, Kumar A, Dolliner P, Filipiak KJ, Pella D et al (2014) The effect of coenzyme Q10 on morbidity and mortality in chronic heart failure: results from Q-SYMBIO: a randomized double-blind trial. JACC Heart Fail 2:641–649. https://doi.org/10.1016/j.jchf.2014.06.008CrossRefGoogle Scholar
- Murad K, Goff DJ, Morgan TM, Burke GL, Bartz TM, Kizer JR et al (2015) Burden of comorbidities and functional and cognitive impairments in elderly patients at the initial diagnosis of heart failure and their impact on total mortality: the cardiovascular health study. JACC Heart Fail 35:42–550. https://doi.org/10.1016/j.jchf.2015.03.004CrossRefGoogle Scholar
- Paulus WJ, Tschope C (2013) A novel paradigm for heart failure with preserved ejection fraction: comorbidities drive myocardial dysfunction and remodeling through coronary microvascular endothelial inflammation. J Am Coll Cardiol 62:263–271. https://doi.org/10.1016/j.jacc.2013.02.092CrossRefGoogle Scholar
- Pitt B, Zannad F, Remme WJ, Cody R, Castaigne A, Perez A, Palensky J, Wittes J (1999) The effect of spironolactone on morbidity and mortality in patients with severe heart failure. Randomized Aldactone evaluation study Investigators. N Engl J Med 341:709–717. https://doi.org/10.1056/NEJM199909023411001CrossRefGoogle Scholar
- Ruilope LM, Dukat A, Bohm M, Lacourciere Y, Gong J, Lefkowitz MP (2010) Blood-pressure reduction with LCZ696, a novel dual-acting inhibitor of the angiotensin II receptor and neprilysin: a randomised, double-blind, placebo-controlled, active comparator study. Lancet 375:1255–1266. https://doi.org/10.1016/S0140-6736(09)61966-8CrossRefGoogle Scholar
- Solomon SD, Zile M, Pieske B, Voors A, Shah A, Kraigher-Krainer E et al (2012) The angiotensin receptor neprilysin inhibitor LCZ696 in heart failure with preserved ejection fraction: a phase 2 double-blind randomised controlled trial. Lancet 380:1387–1395. https://doi.org/10.1016/S0140-6736(12)61227-6CrossRefGoogle Scholar
- Troughton RW, Frampton CM, Brunner-La RH, Pfisterer M, Eurlings LW, Erntell H, Persson H et al (2014) Effect of B-type natriuretic peptide-guided treatment of chronic heart failure on total mortality and hospitalization: an individual patient meta-analysis. Eur Heart J 35(23):1559–1567. https://doi.org/10.1093/eurheartj/ehu090CrossRefGoogle Scholar
- Yusuf S, Pfeffer MA, Swedberg K, Granger CB, Held P, McMurray JJ et al (2003) Effects of candesartan in patients with chronic heart failure and preserved left-ventricular ejection fraction: the CHARM-preserved trial. Lancet 362:777–781. https://doi.org/10.1016/S0140-6736(03)14285-7CrossRefGoogle Scholar