Clinical predictors of preserved left ventricular ejection fraction in decompensated heart failure

Goldraich, Livia; Clausell, Nadine; Biolo, Andréia; Beck-da-Silva, Luís; Rohde, Luís Eduardo

doi:10.1590/S0066-782X2010000300017

Abstracts

BACKGROUND: Identification and clinical impact of preserved left ventricular ejection fraction (LVEF) on in-hospital outcomes in patients with acute decompensated heart failure (HF) remain poorly defined. OBJECTIVE: To describe clinical predictors and in-hospital outcomes of acute decompensated HF patients and preserved LVEF, and to develop a clinically-based predictive rule based on data acquired on admission. METHODS: Consecutive admissions for HF (n=721) at a tertiary care hospital were followed up to discharge or death. More than 80 clinical variables were evaluated to identify predictors of preserved LVEF upon admission. RESULTS: Preserved LVEF (>50%) was identified in 224 (31%) hospitalizations. Clinical predictors of preserved LVEF were age > 70 years old (p=0.04), female gender (p<0.001), non-ischemic etiology (p<0.001), atrial fibrillation or flutter (p=0.001), anemia (p=0.001), pulse pressure > 45 mmHg (p<0.01) and absence of EKG conduction abnormalities (p<0.001). A clinical score based on these variables was accurate to predict preserved LVEF upon hospital admission (area under ROC curve of 0.76). No significant differences were observed on in-hospital mortality or clinical complications according to quintiles of LVEF. CONCLUSION: Preserved LVEF is a prevalent and morbid condition among hospitalized HF patients. Simple clinical data obtained on admission might be useful for predicting preserved LVEF.

Acute decompensated heart failure; preserved ejection fraction and in-hospital mortality

FUNDAMENTO: Identificação e impacto clínico da fração de ejeção preservada em desfechos intra-hospitalares em pacientes com insuficiência cardíaca (IC) descompensada permanecem pouco conhecidos. OBJETIVO: Descrever preditores clínicos e desfechos intra-hospitalares de pacientes com IC descompensada e fração de ejeção de ventrículo esquerdo (FEVE) preservada, e desenvolver um escore preditivo baseado em dados clínicos obtidos no momento da admissão. MÉTODOS: Internações consecutivas devido a IC descompensada (n=721) em um hospital terciário foram acompanhadas até a alta ou óbito. Mais de 80 variáveis clínicas foram avaliadas para identificar os preditores da FEVE preservada no momento da admissão. RESULTADOS: Identificou-se FEVE preservada (>50%) em 224 (31%) internações. Os preditores clínicos de FEVE preservada foram: idade >70 anos (p= 0,04); sexo feminino (p<0,001); etiologia não-isquêmica (p<0,001); fibrilação ou flutter atrial (p=0,001); anemia (p=0,001); pressão de pulso >45 mmHg (p<0,001); e ausência de anormalidades de condução intraventricular verificadas no eletrocardiograma (p<0,001). Um escore clínico baseado nessas variáveis foi capaz de predizer com acurácia a presença de FEVE preservada no momento da admissão hospitalar (área sob a curva ROC de 0,76). Nenhuma diferença significativa foi observada na taxa de mortalidade intra-hospitalar ou de complicações clínicas de acordo com os quintis de FEVE. CONCLUSÃO: A FEVE preservada é uma condição prevalente e mórbida entre pacientes hospitalizados por IC. Dados clínicos simples obtidos no momento da internação podem ser utilizados para predizer FEVE preservada.

Insuficiência cardíaca; fração de ejeção preservada; mortalidade hospitalar

ORIGINAL ARTICLE

^IHospital de Clínicas de Porto Alegre

^IIFaculdade de Medicina da UFRGS, Porto Alegre, RS - Brazil

Mailing address

ABSTRACT

BACKGROUND: Identification and clinical impact of preserved left ventricular ejection fraction (LVEF) on in-hospital outcomes in patients with acute decompensated heart failure (HF) remain poorly defined.

OBJECTIVE: To describe clinical predictors and in-hospital outcomes of acute decompensated HF patients and preserved LVEF, and to develop a clinically-based predictive rule based on data acquired on admission.

METHODS: Consecutive admissions for HF (n=721) at a tertiary care hospital were followed up to discharge or death. More than 80 clinical variables were evaluated to identify predictors of preserved LVEF upon admission.

RESULTS: Preserved LVEF (>50%) was identified in 224 (31%) hospitalizations. Clinical predictors of preserved LVEF were age > 70 years old (p=0.04), female gender (p<0.001), non-ischemic etiology (p<0.001), atrial fibrillation or flutter (p=0.001), anemia (p=0.001), pulse pressure > 45 mmHg (p<0.01) and absence of EKG conduction abnormalities (p<0.001). A clinical score based on these variables was accurate to predict preserved LVEF upon hospital admission (area under ROC curve of 0.76). No significant differences were observed on in-hospital mortality or clinical complications according to quintiles of LVEF.

CONCLUSIONS: Preserved LVEF is a prevalent and morbid condition among hospitalized HF patients. Simple clinical data obtained on admission might be useful for predicting preserved LVEF.

Key words: Acute decompensated heart failure, preserved ejection fraction and in-hospital mortality.

Introduction

Heart failure (HF) remains a major health care problem in Brazil and worldwide^1-4. Heart failure (HF) with preserved systolic function has been recognized for more than 30 years, but it was only in the last decade that a broader understanding of its epidemiology, clinical presentation and prognosis has emerged^5-8. It is now well recognized that a significant number of patients presenting HF symptoms have a normal or only mildly reduced left ventricular ejection fraction (LVEF). Prevalence of preserved LVEF in HF cohorts, however, may vary greatly, depending on diagnostic criteria, study setting and design. It is estimated that approximately 30-50% of hospital admissions attributed to decompensated HF occur in patients without systolic dysfunction^9-12. Furthermore, hospitalizations for HF with preserved LVEF have become increasingly more frequent in the last two decades¹³.

Diagnostic criteria for HF associated with normal LVEF are not universally accepted and may be confusing and unpractical for clinicians evaluating patients with decompensated HF¹⁴. In the emergency department, rapid identification of HF patients with preserved LFEF is particularly important, as treatment strategies may change dramatically according to different classes of left ventricular function¹⁵. Diagnostic and prognostic data derived from post-hoc analysis of clinical trials that enrolled HF outpatients with preserved LVEF may not be applicable to this specific clinical scenario^16-19. Furthermore, prospective data evaluating the impact of left ventricular function on in-hospital outcomes are scarce and conflicting^5,20,21.

The purpose of this prospective cross-sectional study was to

I) describe the prevailing and independent clinical predictors of preserved LVEF in patients admitted for decompensated HF at a tertiary teaching hospital;

II) develop a simple clinically-based predictive rule of HF with preserved LVEF based on a set of clinical features acquired within the first hours from hospital admission; and

III) compare clinically relevant in-hospital outcomes of patients with preserved LVEF to those with left ventricular systolic dysfunction.

Methods

Study setting and identification of cases

The study protocol was conducted at Hospital de Clínicas de Porto Alegre, a tertiary care teaching hospital located in the South Region of Brazil, with 749 beds. Consecutive patients admitted to our institution with suspected acute decompensated HF between August 2000 and January 2004 were eligible for enrollment, irrespective of their subsequent in-hospital destination (whether clinical wards, intensive care units or discharged directly from the emergency department). The protocol of this registry has been previously reported²². In brief, a study investigator or a trained research assistant from the HF team inquired members of medical teams about potentially eligible patients on a daily basis. HF diagnosis and inclusion of individuals were based on the Boston criteria²³. Boston criteria are based on clinical history findings (maximum of 4 points: dyspnea at rest [4 points], orthopnea [4 points], paroxysmal nocturnal dyspnea [3 points], dyspnea while walking on level area [2 points] or dyspnea while climbing [3 points]), physical examination (maximum 4 points: heart rate abnormality [1-2 points], high jugular venous pressure [2-3 points], lung crackles [1-2 points], wheezing [3 points] or third heart sound [3 points] and chest radiography (maximum 4 points: alveolar pulmonary edema [4 points], interstitial pulmonary edema [3 points], bilateral pleural effusion [3 points], cardiothoracic ratio greater than 0.50 [3 points], or upper zone flow redistribution [2 points]). Patients admitted with a Boston score equal or greater than 8 points (definite HF) were considered for inclusion if there was no evidence of an alternative medical diagnosis to which the clinical picture presented could be attributed. Most relevant exclusion conditions were chronic obstructive pulmonary disease, primary pulmonary hypertension, pericardial disease, obesity, physical deconditioning, anxiety, active respiratory infections or pulmonary emboli. In addition, exclusion criteria included history of an acute cardiac event (acute coronary syndrome, myocardial revascularization or cardiac surgery) within a 3-month period from index HF admission and inability to give written informed consent. The research protocol was reviewed and approved by the Human Research Committee from our institution and written informed consent was obtained from all patients prior to enrollment. This investigation conforms to the principles outlined in the Declaration of Helsinki. Registry entries from the current analysis reflect that individual admissions and multiple hospitalizations of the same patients may be entered into the registry as separated records. The complete registry includes 779 consecutive admissions for decompensated HF, but only 721 had a recent evaluation of left ventricular function by echocardiography and constituted the study population.

Data collection

After inclusion, patients were followed throughout hospital stay until discharge or death. A complete medical history was obtained from all patients. Trained investigators inquired the patients on admission and daily during hospitalization in order to collect data including demographics, past medical history, initial presentation (clinical, physical examination and laboratory data), functional capacity, hospital management and hospital outcomes. After death or discharge, missing data were reviewed directly from electronic charts.

Standard M-mode, two-dimensional, and Doppler transthoracic echocardiography were performed at the discretion of the patients' physicians and were interpreted by experienced staff cardiologists. The non-invasive cardiac laboratory of our institution follows the current recommendations of the American Society of Echocardiography for LVEF estimations²⁴. Echocardiographic data analyzed in the present study were obtained directly from electronic reports, in order to reproduce data typically available in routine clinical practice. Preserved systolic function was defined as LVEF equal or above 50% in the absence of significant valvular or pericardial disease. Devereaux formula was used to estimate left ventricular mass, while diastolic dysfunction was defined by standardized echocardiographic criteria²⁵. Most patients had an assessment of left ventricular function performed during index admission; for those who did not, the most recent assessment was recorded from the hospital chart.

Statistical analysis

Continuous variables with normal distribution were expressed as mean standard-deviation, while those with non-normal distribution were described as median and interquartile ranges. Categorical data were described as number and percentages. Clinical characteristics and outcomes were compared among different quintiles of LVEF. More than 80 clinical variables based on medical history, physical examination, laboratory and echocardiographic data were evaluated to determine independent predictors of preserved LVEF. Pearson chi-squared test (or Fisher's exact test when appropriate) was used for categorical data analysis. Comparisons among continuous variables were performed using unpaired Student's t-test and Mann-Whitney U-test, as appropriate. To examine the significance of overall comparisons of groups regarding LVEF quintiles, ANOVA test for linearity and Mantzel-Haenzel test were used for quantitative and qualitative variables, respectively. Logistic regression analysis was performed to identify independent predictors of preserved LVEF among the clinical characteristics presented. Only variables with less than 5% missing data and with a p value < 0.10 in the univariate analysis were accepted for multivariate model (age > 70 years, female gender, non-ischemic etiology, pulse pressure > 45 mm Hg, absence of left bundle branch block or non-specific intratraventricular conduction delay, hemoglobin < 11.5 mg/dl, non-sinus rhythm on EKG, sodium, blood urea nitrogen (BUN) and third heart sound). Variables considered to represent potential colinearity were excluded from the model. Cut-off points from continuous variables associated with preserved LVEF in univariate analysis were determined by looking for the best discriminatory value in individual receiver operator characteristic (ROC) curves. After multivariate analysis, a clinical score was created using the regression coefficient estimates for those independent predictors of preserved LVEF identified by logistic regression selected model. Its diagnostic properties (sensitivity, specificity and predictive values) were then calculated for different score ranges using Pearson chi-squared test. Finally, score global accuracy was estimated by the area under the ROC curve. A two-sided p value smaller than 0.05 was considered to indicate statistical significance. All statistical analyses were performed using SPSS 12.0 software program for Windows (SPSS Inc., Chicago IL).

Results

Overall population

In the present analysis, 721 consecutive admissions for decompensated HF were studied (mean age of 66 ± 13 years old). LVEF was normally distributed as depicted in Figure 1, with mean LVEF of 42 ± 17% (median = 39%). Overall, 50% of patients were male and the predominant etiologies were ischemic and hypertensive (Table 1). A substantial proportion of the study population (approximately 23%) had a history of more than 3 previous HF hospitalizations. Also, as expected, more than 90% of them were in NYHA functional classes III and IV at admission. At hospital admission, 67% of patients were taking ACEIs (Angiotensin-converting enzyme inhibitors), but only 21% were on beta-blockers. In-hospital deaths occurred in 71 patients (10%), mostly due to progression of HF and cardiogenic shock (41[58%]). The majority of the remaining deaths were due to sepsis, acute cerebral and coronary events. During hospitalization, most patients were treated in the regular clinical ward, with intermittent or continuous use of intravenous loop diuretics and incremental doses of vasodilators. Inotropic support was used in less than 5% of admitted patients.

Thumbnail

Preserved LVEF

In this cohort of acute decompensated HF, preserved LVEF was observed in 223 (31%) admissions. Table 1 demonstrates demographic and clinical characteristics among different LVEF quintiles. Patients with higher LVEF were older and predominantly female (p<0.001). The prevalence of diabetes and chronic obstructive pulmonary disease, as well as the Charlson comorbidity index, did not differ across LVEF quintiles. Patients with higher LVEF were more likely to have hypertension as HF etiology, atrial fibrillation and higher baseline systolic and pulse pressure. Among traditional clinical signs and symptoms of HF, only the presence of third heart sound was significantly different among LVEF categories (p<0.001).

Echocardiographic evidence of diastolic dysfunction was more frequently reported in patients with preserved LVEF, but electrocardiographic conduction abnormalities (left-bundle branch block and intraventricular conduction delay) were far more common in those with LV systolic dysfunction (55[36%] versus 11[8%] comparing the lowest and highest quintiles of LVEF, p<0.001). Serum hemoglobin and sodium levels were also significantly different according to LVEF quintiles, whereas renal function was similar in patients with or without systolic dysfunction. The proportion of patients treated with diuretics and beta-blockers before HF admission was similar in different LVEF categories, but ACEIs and digoxin use increased as LVEF decreased (p<0.01).

Clinical predictors of preserved LVEF

In multivariate analysis, 7 independent predictors remained significantly associated with a LVEF greater or equal to 50%, as described in Table 2. Older age, female gender, non-ischemic etiology, non-sinus rhythm, lower hemoglobin levels, higher pulse pressure levels and the absence of electrocardiographic conduction abnormalities were significantly and independently associated with preserved LVEF. A clinical score (ranging from 0 - 6 points) based on logistic regression coefficients of the multivariate model was built. Diagnostic properties of the clinical score rule for prediction of preserved LVEF are displayed in Table 3. Scores equal or lower than 1 point showed a negative predictive value of 100% for the presence of preserved LVEF, while scores higher than 5 points determined a specificity of at least 97%. Each 1-point increment in the score increased 2.5 times the chance of identifying an HF patient with preserved LVEF (95% CI 2.1 - 2.9; p < 0.001). Overall accuracy of the score for the prediction of preserved LVEF was 76% (95% CI 72% - 79%; p < 0.0001) (Figure 2).

Thumbnail

In-hospital outcomes

In-hospital mortality and in-hospital complications was similar across different LVEF quintiles (Table 4). These findings did not change substantially after adjustment for age and gender. Both in-hospital mortality and complications rates were statistically similar when LVEF equal or above 40% (11.5% for LVEF > 40% versus 8% for LVEF < 40%, p= 0.1 for in-hospital mortality; and 45.5% for LVEF > 40% versus 38.5% for LVEF < 40%, p= 0.06 for complications) or 50% (10.5% for LVEF > 50% versus 9.5% for LVEF < 50%, p= 0.7; and 44% for LVEF > 50% versus 41% for LVEF < 50%, p= 0.4, respectively) were considered as cut-off values for preserved systolic function. Hospital length-of-stay among patients discharged alive was also similar among LVEF categories (median 11 days; interquartile range: 6-19 days). Cardiac arrhythmias and fever or evidence of infection during hospital-stay was more often observed in patients with higher LVEF values, but GI bleeding or renal function impairment did not differ according to LVEF.

Thumbnail

Discussion

In this study, we have demonstrated that preserved LVEF is a prevalent and morbid condition among hospitalized HF patients in a Brazilian teaching hospital. Patients with acute decompensated HF and preserved LVEF presenting distinctive clinical characteristics: older age, female gender, non-ischemic etiology, chronic atrial fibrillation, anemia, a large pulse pressure and narrow QRS complexes. A simple clinical score based on these findings was properly performed to rapidly identify subgroups of patients with a higher and lower likelihood of having preserved ejection fraction when presenting acute decompensated HF. Finally, patients with preserved LVEF demonstrated similar in-hospital mortality rates and in-hospital morbidity compared to those with LV systolic dysfunction.

Several reports demonstrated that HF with preserved LVEF is a prevalent condition both in community-based and hospital cohorts and that these patients present similar clinical features, although not always as severe, when compared with the typical HF patients with systolic dysfunction^4,5,7,14. The overall prognosis of these patients, although probably better than that of those with HF and systolic dysfunction, is, nonetheless, far worse when compared to normal individuals¹⁰. Furthermore, although mortality for HF with systolic dysfunction has decreased over the last 15 years, fatal outcome rates for HF with preserved LVEF have remained stable¹³.

Among patients hospitalized due to acute decompensated HF, data on clinical course and outcomes of patients with preserved LVEF is limited. Previous reports demonstrated conflicting findings about the clinical impact on long-term outcomes, and scarce data are available on in-hospital event rates^10,11. A Canadian cohort of hospitalized HF patients did not observe significant differences for both in-hospital and 1-year mortality rates between patients with systolic dysfunction and those with preserved ejection fraction (4.9% and 3.8%, respectively, for in-hospital death)²⁶. However, recent analysis from the ADHERE Registry, including over 100,000 HF admissions, demonstrated that in-hospital death rates were significantly lower for patients with preserved LVEF (2.8% vs. 3.9%; p = 0.005), although duration of intensive care unit stay and total hospital length-of-stay were similar²⁷. Considering these latter findings, it is conceivable to assume that differences in some clinical outcomes may in fact be present in patients with different degrees of LV systolic function, but the magnitude of these differences may not be as great as previously considered. Nonetheless, our data support the concept that patients with preserved LVEF hospitalized with acute decompensated HF have a considerably morbid condition, associated with substantial in-hospital mortality and high clinical event rates during hospital stay. In-hospital death rates in our sample were significantly higher, irrespective of LVEF strata, when compared to data previously published in other countries, particularly in North American reports^28,29. Our data derive from a cohort of consecutive HF patients diagnosed by clinical criteria, representing the typical ''real-world'' patient that seeks hospital care. Although widely accepted to diagnose HF in the outpatient setting, neither the Framingham nor the Boston criteria can differentiate patients with systolic dysfunction from those with preserved LVEF¹⁴. In addition, B-type natriuretic peptide levels, which have been extensively validated as a diagnostic and prognostic tool in acute decompensated HF, are high in patients with a congestive state regardless of LVEF status³⁰. All current diagnostic criteria proposed to identify HF with preserved systolic function incorporate imaging modalities to objectively define diastolic dysfunction, a rather limiting step in the acute setting¹⁴. In this scenario, simple clinical characteristics that may suggest the presence of LV dysfunction may be instrumental to physicians facing the clinical dilemma of treating patients with HF symptoms in the emergency department.

Several studies tried to identify clinical variables that could be associated with preserved LVEF, and therefore help in its identification before imaging assessment of systolic and diastolic parameters^11,19-21,27. Some of these characteristics have been systematically reported in cohorts of hospitalized HF patients, such as older age, female gender and lower prevalence of previous history of myocardial infarction or coronary artery disease among subjects with preserved LVEF^11,20,21,29. More recently, anemia³¹ and atrial fibrillation²⁷ have also been demonstrated to be more common in admitted HF patients without systolic dysfunction. On the contrary, left bundle branch blocks are significantly more frequent in the presence of LV dysfunction^21,31. Furthermore, high pulse pressure values, as identified in the present study, have been associated to higher LVEF both in outpatients and in acute decompensated HF patients^32,33. In this study, we proposed a simple predictive rule to identify LV function based on easily obtained clinical characteristics at presentation. The overall diagnostic performance of this clinical score was adequate with an area under ROC curve of 0.76. For instance, patients with a score < 2 points represent a subgroup with a very high probability of having LV dysfunction (LVEF < 50%). Our data corroborate the clinical characteristics that define this profile in a cohort of acute decompensated HF patients outside North America. The therapeutic implications of these findings cannot be underscored. Collins et al³⁴, in a recent report, elegantly suggested several specific goals to be kept in mind when treating patients with acute decompensated HF³⁴. For instance, use of inotropic negative drugs to manage blood pressure and control ventricular rate is an everyday clinical dilemma in the management of acute decompensated HF that might be greatly influenced by left ventricular function.

Limitations

Some potential limitations of our study deserve consideration. Our population was "highly selected" in order to avoid incorrectly including other diagnoses that could simulate HF symptoms, based on relatively high cut-off for Boston criteria (>8). Such inclusion strategy was defined a priori and was chosen to avoid "contamination" of the studied sample. We acknowledge that this may be a potential limitation and that our results are applicable only to a selected group of acute decompensated HF patients. We cannot exclude the possibility that the selection criteria used in the present protocol (high Boston score) could have disproportionately exclude some HF patients with preserved ejection fraction, that might have less severe HF burden and therefore lower complication rates. The proposed score derived from a cohort of consecutive HF patients representing typical "real world" patients seeking hospital care. We have previously shown that several clinical characteristics are similar between HF patients admitted to our institution and those admitted to a tertiary-care university hospital in the US²⁸. We acknowledge that external prospective validation is warranted to assure that the proposed score also performs adequately in different HF cohorts; this aspect represents a major limitation of our findings.

Conclusions

In patients hospitalized for decompensated HF, preserved LVEF is a common finding associated with substantial short-term mortality and high rates of clinically relevant in-hospital morbid events. At clinical presentation, simple clinical characteristics were helpful in identifying distinctive patient profiles that predicted LV function strata with a reasonable accuracy. This predictive rule, however, needs further prospective validation in different cohorts of HF patients.

Potential Conflict of Interest

No potential conflict of interest relevant to this article was reported.

Sources of Funding

This study was partially funded by CNPq, Fapergs and Fipe-HCPA.

Study Association

This article is part of the thesis of master submitted by Lívia Goldraich, from Programa de Pós Graduação em Ciências da Saúde: Cardiologia e Ciências Cardiovasculares da UFRGS.

References

1. Rohde LE. O prognóstico da insuficiência cardíaca no Brasil: a busca de dados confiáveis e representativos. Arq Bras Cardiol. 2005; 84: 281-2.
2. Barreto AC, Del Carlo CH, Cardoso JN, Morgado PC, Munhoz RT, Eid MO, et al. Hospital admissions and death rate from heart failure: rates still alarming. Arq Bras Cardiol. 2008; 91: 335-41.
3. Mangini S, Silveira FS, Silva CP, Grativvol PS, Seguro LF, Ferreira SM, et al. Decompensated heart failure in the emergency department of a cardiology hospital. Arq Bras Cardiol. 2008; 90: 400-6.
4. Moutinho MA, Colucci FA, Alcoforado V, Tavares LR, Rachid MB, Rosa ML, et al. Heart failure with preserved ejection fraction and systolic dysfunction in the community. Arq Bras Cardiol. 2008; 90: 132-7.
5. Hogg K, Swedberg K, McMurray J. Heart failure with preserved left ventricular systolic function: epidemiology, clinical characteristics and prognosis. J Am Coll Cardiol. 2004; 43: 317-27.
6. Muller JE, Brendorp B, Ottesen M, Kober L, Egstrup K, Poulsen SH, et al. Bucindolol evaluation in acute myocardial infarction trail group. Congestive heart failure with preserved left ventricular systolic function after acute myocardial infarction: clinical and prognostic implications. Eur J Heart Fail. 2003; 5 (6): 811-9.
7. Lenzen MJ, Scholte op Reimer WJ, Boersma PJ, Vantrimpont PJ, Follath F, Swedberg K, et al. Differences between patients with a preserved and a depressed left ventricular function: a report from the EuroHeart Failure Survey. Eur Heart J. 2004; 25 (14): 1214-20.
8. Pedersen F, Raymond I, Madsen LH, Mehlsen J, Atar D, Hildebrandt P. Echocardiographic indices of left ventricular diastolic dysfunction in 647 individuals with preserved left ventricular systolic function. Eur J Heart Fail. 2004; 6: 439-47.
9. Peyster E, Norman J, Domanski M. Prevalence and predictors of heart failure with preserved systolic function: community hospital admissions of a racially and gender diverse elderly population. J Card Fail. 2004; 10: 49-54.
10. Dauterman KW, Go AS, Rowell R, Gebretsadik T, Gettner S, Massie BM. Congestive heart failure with preserved systolic function in a statewide sample of community hospitals. J Card Fail. 2001; 7: 211-8.
11. Ahmed A, Roseman JM, Duxbury AS, Allman RM, DeLong JF. Correlates and outcomes of preserved left ventricular systolic function among older adults hospitalized with heart failure. Am Heart J. 2002; 144: 365-72.
12. Cohen-Solal A, Desnos M, Delahaye F, Emeriau JP, Hanania G. A national survey of heart failure in French hospitals. The Myocardiopathy and Heart Failure Working Group of the French Society of Cardiology, the National College of General Hospital Cardiologists and the French Geriatrics Society. Eur Heart J. 2000; 21: 763-9.
13. Owan TE, Hodge DO, Herges RM, Jacobsen SJ, Roger VL, Redfield MM. Trends in prevalence and outcome of heart failure with preserved ejection fraction. N Eng J Med. 2006; 355: 251-9.
14. Yturralde FR, Gaasch WH. Diagnostic criteria for diastolic heart failure. Prog Cardiovasc Dis. 2005; 47: 314-9.
15. Little WC, Brucks S. Therapy for diastolic heart failure. Prog Cardiovasc Dis. 2005; 47: 380-8.
16. Pocock SJ, Wang D, Pfeffer MA, Yusuf S, McMurray JJ, Swedberg KB, et al. Predictors of mortality and morbidity in patients with chronic heart failure. Eur Heart J. 2006; 27: 65-75.
17. Solomon SD, Anavekar N, Skali H, McMurray JJ, Swedberg K, Yusuf S, et al. Influence of ejection fraction on cardiovascular outcomes in a broad spectrum of heart failure patients. Circulation. 2005; 112: 3738-44.
18. Jones RC, Francis GS, Lauer MS. Predictors of mortality in patients with heart failure and preserved systolic function in the Digitalis Investigation Group trial. J Am Coll Cardiol. 2004; 44: 1025-9.
19. Philbin EF, Hunsberger S, Garg R, Lader E, Thadani U, McSherry F, et al. Usefulness of clinical information to distinguish patients with normal from those with low ejection fractions in heart failure. Am J Cardiol. 2002; 89: 1218-21.
20. Malki Q, Sharma ND, Afzal A, Ananthsubramaniam K, Abbas A, Jacobson G, et al. Clinical presentation, hospital length of stay and readmission rate in patients with heart failure with preserved and decreased left ventricular systolic function. Clin Cardiol. 2002; 25: 149-52.
21. Macín SM, Perna ER, Canella JP, Cimbaro Canella JP, Alvarenga P, Pantich R, et al. Differences in clinical profile and outcome in patients with decompensated heart failure and systolic dysfunction or preserved systolic function. Rev Esp Cardiol. 2004; 57: 45-52.
22. Rohde L, Goldraich L, Polanczyk C, Borges AP, Biolo A Rabelo E, et al. A simple clinically-based predictive rule for heart failure in-hospital mortality. J Card Fail. 2006; 12: 587-93.
23. Remes J, Miettinen H, Reunanen A, Pyorala K. Validity of clinical diagnosis of heart failure in primary health care. Eur Heart J. 1991; 12: 315-21.
24. Lang RM, Bierig M, Devereux RB, Flachskampf EA, Foster E, Pellikka PA, et al. Recommendations for chamber quantification: a report from the American Society of Echocardiography's Guidelines and the Standards Committee and the Chamber Quantification Writing Group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology. J Am Soc Echocardiogr. 2005; 18: 1440-63.
25. Foppa M, Duncan BB, Rohde LE. Echocardiography-based left ventricular mass estimation: how should we define hypertrophy? Cardiovasc Ultrasound. 2005; 3:17.
26. Bhatia RS, Tu JV, Lee DS, Austin PC, Fang J, Haouzi A, et al. Outcome of heart failure with preserved ejection fraction in a population-based study. N Eng J Med. 2006; 355: 260-9.
27. Yancy CW, Lopatin M, Stevenson LW, De Marco T, Fonarow GC. Clinical presentation, management and in-hospital outcomes of patients admitted with acute decompensated heart failure with preserved systolic function. A report from the Acute Decompensated Heart Failure National Registry (ADHERE) Database. J Am Coll Cardiol. 2006; 47: 76-84.
28. Rohde LE, Clausell N, Ribeiro JP, Goldraich L, Netto R, William Dec G, et al. Health outcomes in decompensated congestive heart failure: a comparison of tertiary hospitals in Brazil and United States. Int J Cardiol. 2005; 102: 71-7.
29. Komajda M, Follath F, Swedberg K, Cleland J, Aguilar JC, Cohen-Solal A, et al. The EuroHeart Failure survey programme- a survey on the quality of care among patients with heart failure: Part 2. Treatment. Eur Heart J. 2003; 24: 464-74.
30. Maisel AS, McCord J, Nowak RM, Hollander JE, Wu AH, Duc P, et al. Bedside B-type natriuretic peptide in the emergency diagnosis of heart failure with reduced or preserved ejection fraction. Results from the Breathing Not Properly Multinational Study. J Am Coll Cardiol. 2003; 41: 2018-21.
31. Latado AL, Passos LCS, Darzé ES, Lopes AA. Comparison of the effect of anemia on in-hospital mortality in patients with versus without preserved left ventricular ejection fraction. Am J Cardiol. 2006; 98: 1631-4.
32. Beck-da-Silva L, Goldraich L, Bonzanini L, Rossi G, Nuernberg G, Camey SA, et al. Pulse pressure and QRS width evaluation as an inexpensive tool for heart failure assessment. Congest Heart Fail. 2009; 15 (5): 222-7.
33. Aronson D, Burger AJ. Relation between pulse pressure and survival in patients with decompensated heart failure. Am J Cardiol. 2004; 93: 785-8.
34. Collins S, Storrow SB, Kirk JD, Pang PS, Diercks DB, Gheorghiade M. Beyond pulmonary edema: diagnostic, risk stratification, and treatment challenges of acute heart failure management in the emergency department. Ann Emerg Med. 2008; 51: 45-57.