European Journal of Internal Medicine 20 (2009) 313–318www.elsevier.com/locate/ejim

Original article

Prediction clinical profile to distinguish between systolic and diastolic heartfailure in hospitalized patients

Ana Maestrea,⁎, Vicente Gilb, Javier Gallegoa, Miguel Garcíac,Fernando García de Burgosc, Alberto Martín-Hidalgoa

a Internal Medicine Department. Hospital General Universitario de Elche, Spainb Universidad Miguel Hernández, Spain

c Cardiology Section, Internal Medicine Department, Hospital General Universitario de Elche, Spain

Received 3 December 2007; received in revised form 12 August 2008; accepted 3 September 2008Available online 26 October 2008

Abstract

Background: In recent decades, the growing incidence of patients with heart failure who have preserved systolic function, underlines the need todifferentiate between heart failure due to diastolic dysfunction and that due to systolic dysfunction.Objective: To develop a prediction profile of clinical parameters that enables clinicians to differentiate between patients with systolic and diastolicheart failure.Methods: 164 patients admitted for congestive heart failure to the cardiology department of an academic tertiary care hospital, whose leftventricular systolic and diastolic function had been evaluated echocardiographically and who satisfied the Framingham criteria for heart failure,were prospectively recruited. All patients answered a questionnaire which included, in addition to other clinical variables, the Framingham criteria.Results: Patients with diastolic heart failure (61.6%) were more likely to be older, female, and to present left ventricular hypertrophy (LVH), with alower proportion of smokers, alcohol drinkers, coronary disease, q wave and left bundle branch block (all pb0.005). The predicting modelobtained on the logistic regression analysis was very significant, with three variables and 72.3% of correct predictions (x2 value=40,457,pb0.001). These three variables, predictors of diastolic as opposed to systolic heart failure, were female sex (OR=3.546), left ventriclehypertrophy (OR=4.011) and absence of coronary disease (OR=3.547).Conclusion: Three variables which can be easily evaluated, female sex, left ventricular hypertrophy and presence or absence of coronary disease,may enable clinicians to differentiate between patients with systolic or diastolic heart failure.© 2008 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved.

Keywords: Clinical characteristics; Systolic and diastolic dysfunction; Heart failure; Hospitalized patients

1. Introduction

Heart failure is one of the most serious public healthproblems in the Western world and most authors consider thatwe are facing the greatest cardiovascular epidemic of the 21stcentury [1]. This has an increasing impact on the health of the

⁎ Corresponding author. Servicio de Medicina Interna, Hospital GeneralUniversitario de Elche, Camino de la Almazara s/n. 03203, Elche, Spain. Tel.:+34 966679318.

E-mail address: amaestrep@gmail.com (A. Maestre).

0953-6205/$ - see front matter © 2008 European Federation of Internal Medicine.doi:10.1016/j.ejim.2008.09.001

population since not only the incidence but also the prevalenceof heart failure is raising, with the resulting increase inmorbidity, mortality and healthcare costs [2–4].

Approximately 1.5–2% of the population have heart failure,and the prevalence rises to 6–10% in patients over 65 years ofage [2,5,6], in whom it is the main reason for hospital admission[3,7]. The annual incidence found in the Framingham study rosefrom 0.3% in men aged 50 to 59 years to 2.7% in men aged 80to 89 [6]. Despite medical advances, the mortality is still highand heart failure is currently the third cause of cardiovascularmortality in developed countries.

In Spain, there is no data available on the true incidence ofheart failure in the community. With regard to prevalence, a

Published by Elsevier B.V. All rights reserved.

Table 1Clinical characteristics of patients with systolic heart failure (SHF) and diastolicheart failure (DHF).

Characteristics Total(n=164)

SHF(n=63)

DHF(n=101)

P value

Age 73.02 70.11 74.84 0.003Gender Male 79 (48.2%) 43 (68.3%) 36 (35.6%) b0.001

314 A. Maestre et al. / European Journal of Internal Medicine 20 (2009) 313–318

population based study from Asturias found a prevalence of 5%,ranging from less than 1% in patients under 50 years old to 18%in those over 80 years old [8]. There is more information aboutmorbidity, based mainly on hospital records and series [9].Hospital admission for heart failure increased by 47% between1980 and 1993. This increase was most pronounced in the over65-year-old population, in which it was the main reason foradmission to hospital and accounted for 5% of all hospitaladmissions. Thus, heart failure is also a significant demographicand healthcare burden for the Spanish population [1].

Few diseases have experienced so many changes in theirepidemiology, physiopathological basis and therapeuticapproach in recent decades as heart failure. One of the mainepidemiological changes is the increasing prevalence of heartfailure in which systolic function is preserved [10,11], as shownby numerous reports published in recent years, based on theincidence in the community [7,12], transversal populationstudies on prevalence [5,8,13–15] or hospital cohorts [16–20].The heterogeneity of published studies, the use of differentdiagnostic criteria and cut-off points for left ventricular ejectionfraction, and the fact that left ventricular diastolic function israrely evaluated, are only some of the reasons why theepidemiology of heart failure has not been clearly established.

Although clinical features and physical examination havefailed to consistently discriminate between diastolic and systolicheart failure in previous studies [21], in clinical practice it couldbe useful to be able to differentiate between the two conditionsby means of clinical signs and symptoms.

The objective of this study was to develop a predictionprofile of clinical parameters that could make it possible todifferentiate between patients with systolic heart failure andthose with diastolic heart failure in real healthcare conditions.

Female 85 (51.8%) 20 (31.7%) 65 (64.4%)Hypertension Yes 105 (64.0%) 35 (55.6%) 70 (69.3%) 0.094

No 59 (36.0%) 28 (44.4%) 31 (30.7%)DM Yes 65 (39.6%) 22 (34.9%) 43 (42.6%) 0.412

No 99 (60.4%) 41 (65.1%) 58 (57.4%)Hyperlipidemia Yes 33 (20.1%) 14 (22.2%) 19 (18.8%) 0.690

No 131 (79.9%) 49 (77.8%) 82 (81.2%)Smoker Yes 14 (8.5%) 9 (14.3%) 5 (5.0%) 0.004

No 106 (64.7%) 31 (49.2%) 75 (74.2%)Ex-smoker

44 (26.8%) 23 (36.5%) 21 (20.8%)

Alcohol Yes 11 (6.7%) 8 (12.7%) 3 (3.0%) 0.014No 149 (90.9%) 52 (82.5%) 97 (96.0%)Ex-drinker

4 (2.4%) 3 (4.8%) 1 (1.0%)

CD Yes 51 (31.1%) 28 (44.4%) 23 (22.8%) 0.005No 113 (68.9%) 35 (55.6%) 78 (77.2%)

Anaemia Yes 31 (18.9%) 12 (19.0%) 19 (18.8%) 1.000No 133 (81.1%) 51 (81.0%) 82 (81.2%)

SBP 146.01 134.79 157.23 b0.001DBP 82.41 80.33 84.49 0.225LVH Yes 65 (39.8%) 13 (21.0%) 52 (51.5%) b0.001

No 98 (60.2%) 49 (79.0%) 49 (48.5%)Q wave Yes 17 (11.1%) 11 (18.6%) 6 (6.3%) 0.032

No 137 (88.9%) 48 (81.4%) 89 (93.7%)LBBB Yes 30 (19.2%) 17 (27.9%) 13 (13.7%) 0.037

No 126 (80.8%) 44 (72.1%) 82 (86.3%)

DM:Diabetesmellitus; CD: coronary disease; SBP: systolic blood pressure (mmHg);DBP: diastolic blood pressure (mm Hg); LVH: Left ventricular hypertrophy; LBBB:Left bundle branch block

2. Patients and methods

This was a prospective observational study. We included allpatients referred to the cardiology section of the UniversityGeneral Hospital of Elche who were admitted to hospital forcongestive heart failure between 1 June 2002 and 31 May 2003,whose left ventricular systolic and diastolic function wasevaluated echocardiographically within two days of admission,and who satisfied the modified Framingham criteria forcongestive heart failure [22]. The criteria advocated by theEuropean Study Group on Diastolic Heart Failure were used tomeasure diastolic dysfunction. This study group proposes arestrictive approach in which diagnosis of diastolic heart failurerequires a combination of clinical signs and symptoms of heartfailure, preserved or slightly depressed systolic function andevidence of anomalies in ventricular relaxation, filling ordistension. Systolic dysfunction was based on a left ventricularejection fraction of less than 45% [23,24].

Exclusion criteria were those derived from the patient (seniledementia, being bed-ridden for a long time due to non-cardiological problems, cor pulmonale or primary state ofvolume overload) and those derived from echocardiography(poor echogenic window or moderate-severe valvulopathy).

All patients answered a questionnaire which included, inaddition to other sociodemographic and clinical variables, theFramingham criteria for heart failure (Appendix A.1). Eachpatient underwent a thorough physical examination, anelectrocardiogram, chest radiography, specific laboratory testsand a transthoracic M-mode, 2-dimensional, Doppler echocar-diography. The echocardiograms were performed by a trainedcardiologist who determined whether there were valve abnorm-alities, left ventricular hypertrophy or pulmonary hypertension.In addition, volumes, left ventricle diameters and a series ofparameters of ventricular dysfunction, such as left ventricularejection fraction and the main indexes of diastolic dysfunction,were calculated by analysing the morphology of the maximumtransmitral flow velocity curve (Appendix A.2).

Univariate tests of statistical significance for differences inclinical characteristics were performed. Data for continuousvariables were expressed as means and compared using theStudent's t test. Data for categorical or dichotomous variableswere expressed as percentages and compared using the x2 testor Fisher's exact test. Multiple logistic regression analysis wasused to determine the strength and significance of clinicalcharacteristics as predictors of normal versus decreased systolicfunction. All statistical tests were 2-sided and a p value of 0.05was selected for the threshold of statistical significance.

Table 3Final predicting model for diastolic dysfunction in patients admitted to hospitalwith heart failure.

Variable Odds ratio CI of 95% p value

Female sex 3.546 (1.724–7.297) 0.001LVH 4.011 (1.916–8.399) b0.001Absence of coronary disease 3.547 (1.712–7.347) b0.001

x2 value=40,457, pb0.001Correctly classified: 72.3%

LVH: Left ventricular hypertrophy.Logistic regression analysis.

315A. Maestre et al. / European Journal of Internal Medicine 20 (2009) 313–318

Analyses were performed using SPSS statistical software,version 12.0.

3. Results

The final sample consisted of 164 patients, of whom 85 werewomen (51.8%) and 79 men (48.2%), with a mean age of73 years.

With regard to the prevalence of classical vascular riskfactors, 105 patients (64.0%) were hypertensive, 65 (39.6%)were diabetic, 33 (20.1%) had hyperlipidemia, 14 (8.5%) wereactive smokers, and 44 (26.8%) ex-smokers. In addition, 11patients (6.7%) had alcohol abuse and 61 (37.2%) were obese.Regarding main associated co-morbidities, coronary diseasewas present in 31.1% of the population studied, whereas chronicobstructive pulmonary disease and anaemia were both presentin 18.9%.

Echographic data showed systolic dysfunction in 63 (38.4%)and diastolic dysfunction in 101 (61.6%) of the 164 patients.

The clinical characteristics of the study patients are shown inTable 1. Patients with diastolic heart failure were older(p=0.003), more likely to be women (pb0.001), with higherlevels of systolic blood pressure (pb0.001), more likely to haveleft ventricle hypertrophy (pb0.001) and with a lowerproportion of smokers (p=0.004), alcohol abuse (p=0.014),coronary disease (p=0.005), total left bundle branch block(p=0.037) and q waves (p=0.032) compared with patients withsystolic failure.

Table 2Distribution of the Framingham criteria in patients with systolic heart failure(SHF) and diastolic heart failure (DHF).

Framingham criteria Total (n=164) SHF (n=63) DHF (n=101) p value

PND Yes 147 (89.6%) 56 (88.9%) 91 (90.1%) 0.798No 17 (10.4%) 7 (11.1%) 10 (9.9%)

NVD Yes 34 (21.0%) 14 (22.2%) 20 (20.2%) 0.844No 128 (79.0%) 49 (77.8%) 79 (79.8%)

Crackles Yes 135 (82.3%) 51 (81.0%) 84 (83.2%) 0.834No 29 (17.7%) 12 (19.0%) 17 (16.8%)

Cardiomegaly Yes 144 (88.3%) 59 (95.2%) 85 (84.2%) 0.043No 19 (11.7%) 3 (4.8%) 16 (15.8%)

APE Yes 19 (11.6%) 5 (7.9%) 14 (13.9%) 0.320No 145 (88.4%) 58 (92.1%) 87 (86.1%)

S3-Gallop Yes 52 (37.7%) 26 (46.4%) 26 (31.7%) 0.107No 86 (62.3%) 30 (53.6%) 56 (68.3%)

HJR Yes 27 (16.6%) 12 (19.0%) 15 (15.0%) 0.522No 136 (83.4%) 51 (81.0%) 85 (85.0%)

Ankle oedema Yes 111 (67.7%) 39 (61.9%) 72 (71.3%) 0.233No 53 (32.3%) 24 (38.1%) 29 (28.7%)

Nocturnal cough Yes 32 (19.5%) 11 (17.5%) 21 (20.8%) 0.688No 132 (80.5%) 52 (82.5%) 80 (79.2%)

Dyspnoea Yes 164 (100%) 63 (100%) 101 (100%)No 0 (0%) 0 (0%) 0 (0%)

Hepatomegaly Yes 42 (30.7%) 19 (35.8%) 23 (27.4%) 0.343No 95 (69.3%) 34 (64.2%) 61 (72.6%)

PE Yes 62 (38.0%) 19 (30.6%) 43 (42.6%) 0.138No 101 (62.0%) 43 (69.4%) 58 (57.4%)

Tachycardia Yes 43 (26.4%) 17 (27.4%) 26 (25.7%) 0.856No 120 (73.6%) 45 (7.6%) 75 (74.3%)

PND: Paroxysmal nocturnal dyspnoea; NVD: Neck vein distention; APE: Acutepulmonary oedema; HJR: Hepatojugular reflux; PE: Pleural effusion.

The presence of the Framingham clinical criteria wascompared in both groups by means of a univariate analysis(Table 2). The distribution of the Framingham criteria onlyshowed significant differences in cardiomegaly, which wasmore frequent in the group of systolic heart failure (p=0.043).

The prediction model obtained on the multiple logisticregression analysis (Table 3) was very significant (pb0.001),and made it possible to distinguish between patients withdiastolic heart failure and systolic heart failure using threevariables with a good prediction potential (correct classificationin 72.3% of cases). These three variables, predictors of diastolicas opposed to systolic heart failure, were female sex(OR=3.546), left ventricular hypertrophy (OR=4.011) andabsence of coronary disease (OR=3.547).

4. Discussion

In recent decades, the growing incidence of heart failure withpreserved systolic function underlines the need to differentiatebetween heart failure due to diastolic dysfunction and that dueto systolic dysfunction [25]. In most of the published studies,both groups are defined using the same echocardiographicparameter—left ventricular ejection fraction—without specifi-cally evaluating the indexes of diastolic function. This is themain reason why diastolic heart failure is not referred to as suchbut rather as heart failure with preserved ejection fraction[11,16,17].

Very few studies objectively evaluate diastolic function[14,15] and all that do, except for one recently published studythat assess this parameter in hospitalized patients [26], havebeen performed in a community setting. In European popula-tion, we found only one study by the German group of Fischeret al. which evaluated the prevalence of left ventricular diastolicdysfunction in the community using echocardiography [13].

Our study differs from others in that it includes only patientswith heart failure confirmed on echocardiography, and determinesnot only the ejection fraction but also diastolic ventriculardysfunction. In the present study, more than half (61.6%) of theindividuals had diastolic heart failure. Patients with diastolic heartfailure were older, there was a greater proportion of women, morelikely to have left ventricular hypertrophy, fewer smokers andalcohol drinkers, less coronary artery disease and fewer q wavesor total left bundle branch block, than in patients with systolicheart failure. In addition, patients with diastolic heart failure weremore likely to have arterial hypertension and diabetes mellitus,

316 A. Maestre et al. / European Journal of Internal Medicine 20 (2009) 313–318

although the differences did not attain the significance expected.With regard to the distribution of the Framingham criteria, nosignificant differences were found except for cardiomegaly.However, the presence of third sound and hepatomegaly did tendto be greater in the systolic dysfunction group (46.4%), as hasbeen reported in prior studies [27,28].

When all the previously mentioned characteristics arecombined, in our sample, female sex, left ventricular hypertrophyand the absence of coronary artery disease explain almost 75% ofthe variability between the diagnosis of diastolic and systolic heartfailure.

Most of these data correspond to a large extent with thosepublished so far, although some differences should be pointedout. In our study, the proportion of patients with diastolic heartfailure was 61.6% versus 38.4% with systolic heart failure. Thisgreater prevalence of patients with diastolic heart failure may bedue to the fact that one of our exclusion criteria was theexistence of moderate or severe valvulopathies, since theirpresence prevent the indexes of diastolic dysfunction frombeing correctly determined [11,16,29]. However, in otherstudies which simply evaluated whether left ventricular ejectionfraction was depressed or preserved, such patients could beincluded. Since most of these patients had systolic dysfunction,this group of patients could increase in number [18,28]. Inaddition, the advanced age of our patients might havecontributed to the greater prevalence of diastolic dysfunction.

The first study performed in Spain to evaluate the percentageof patients with altered or preserved systolic dysfunction and todescribe the clinical characteristics of both groups was carriedout in Santiago de Compostela [9]. In this study, investigatorsincluded all the patients admitted to a cardiology department forcongestive heart failure who fulfilled the Framingham clinicalcriteria and whose left ventricular systolic function had beenevaluated. Therefore, the criteria for inclusion were very similarto ours, except that both groups were defined using a singleechocardiographic parameter, left ventricular ejection fraction,and without specifically evaluating the diastolic dysfunctionindexes. They reported a mean age of 66.7 years, with apredominance of men (58.5%) and the presence of arterialhypertension in 52.2% of the cases, followed by coronarydisease in 45.4%. Fewer than 30% of the patients had preservedsystolic function. When comparing these results with those ofour study, in the former the patients were younger, there weremore men, a lower prevalence of arterial hypertension and morecoronary disease, as well as a smaller proportion of patients withdiastolic heart failure (28.8%). The authors attribute thedifferences between their results and those published in otherstudies to the fact that, in their hospital, elderly patients withheart failure not thought to be secondary to coronary disease arenot referred to the cardiology service [9]. It is possible that theseelderly patients, who are controlled by other services, make upthe population sub-group most often described in other studies,with a predominance of women and greater prevalence ofarterial hypertension. Very recently, this same group publishedanother study which was an extension of the first [28], over astudy period of 10 years, with results somewhat different. Themean age in this case was higher (69.4 years versus 66.7 years),

with a higher proportion of patients with preserved systolicfunction (39.8% versus 28.8%), although the rest of clinicalcharacteristics remained the same.

Comparing our results with those reported in other hospitalcohort studies carried out abroad, many similarities are found.The first retrospective studies showed a clear predominance ofwomen with heart failure and normal systolic function, whichwas consistent throughout the various sub-groups of patients[18,27]. McDermott et al. published a retrospective review andprovided a prediction model with which they obtained a correctclassification (76%) very similar to ours (72.3%). As in ourstudy, patients with heart failure and normal left systolicfunction were older, more often women and less likely to have ahistory of coronary artery disease [27]. Unfortunately, leftventricular diastolic dysfunction was not assessed again.

Thomas et al. evaluated the usefulness of the clinical history,physical examination, electrocardiogram and chest X-ray indifferentiating between patients with normal and depressedsystolic function (specific diastolic indexes were not evaluatedin this study). Patients with preserved systolic function weregenerally female, older and obese, with higher levels of diastolicand systolic blood pressure; whereas tachycardia, clinicalsymptoms of angina pectoris and alcohol consumption weremore frequent in patients from the other group. On multivariateanalysis, sex and tachycardia were the only clinical variablesshowing significant association [16].

The first multicentre prospective study to characterize the cli-nical profile, hospital stay and treatment of heart failurewith normalejection fraction concluded that the majority of patients werewomen (73%), older than themen in this group and therewas a highpercentage of arterial hypertension (78%), left ventricular hyper-trophy (82%), diabetes mellitus (46%), and obesity (46%) [17].

Recently, results from two other studies in hospital setting withsimilar results have been published. Owan et al. carried out aretrospective study in patients hospitalized for heart failure todefine secular trends in the prevalence and survival of heart failurewith preserved ejection fraction [26]. These authors did usespecific echocardiographic parameters for diastolic dysfunctionand found that 53% of patients had reduced ejection fraction and47% normal ejection fraction. In the other study carried out byBhatia et al. [20], of the 2802 patients admitted for heart failure,31% had an ejection fraction above 50%. These patients weremore likely to be older, female, and had a significantly higherincidence of hypertension and atrial fibrillation than those withdepressed ejection fraction. However, complications, rates of re-admission and mortality were similar in both groups.

Finally, potential limitations to our study should be acknowl-edged to facilitate the interpretation of the results. In order to avoidany possible measurement bias, extreme care was taken when per-forming the echocardiography. All the echocardiograms wereperformed by an experienced cardiologist using the same equip-ment. However, we realise that there are a series of circumstancessuch as atrial fibrillation, tachycardia or poor echographic windowthat can make it difficult to evaluate left ventricular function.Another measurement bias could arise from the reliability andconsistency of the interviewer when performing the anamnesis andphysical examination of the patients. All the interventions were

317A. Maestre et al. / European Journal of Internal Medicine 20 (2009) 313–318

made by the same observer, who underwent training in taking thesemeasurements in clinical practice. The study might also be limitedby the fact that only echocardiography was used to diagnose dias-tolic abnormalities, since echographic diastolic indexes are depen-dent on the cardiac frequency, after-load and pre-load, or the time itis performed [13]. However, the alternative techniques are invasiveor require exposure to radioisotopes, which means they are notapplicable to all the population. Another limitation could be the useof a cut-off point of 45% for the left ventricular ejection fraction asthe only index of normal systolic function, since regional or slightlyimpaired systolic dysfunction could be overlooked. We used thisvalue because it was recommended in the Spanish and Europeanguidelineswedecided to usewhen starting the study [23,24].A finallimitation could be that treatment and medication of the enrolledpatients was not considered and that results apply to a tertiaryreferral centre and may not be applicable to other populations.

In conclusion, in this study three variables that may be easilyassessed, female sex, left ventricular hypertrophy and absenceof coronary disease, enable us to differentiate between patientswith systolic or diastolic heart failure. This clinical predictingprofile of diastolic heart failure is the first to be obtained in aEuropean population admitted to hospital for heart failure.

Although clinical assessment and non-invasive cardiacinvestigations (chest radiography or electrocardiography) arenot a substitute for an objective evaluation of left ventriculardysfunction, these results may help to make an initial differentialdiagnosis between systolic and diastolic heart failure, especiallyregarding primary healthcare or non-cardiological specialitiesthat depend on cardiologists to carry out echocardiograms. Theresults could so enhance the clinician's confidence in making adiagnosis of diastolic heart failure and confirm the character-istics of these patients in the hospital setting.

In the future, it could be of clinical worth to reliablydistinguish these two populations clinically up-front to stratifytreatment strategies appropriately.

5. Learning points

• Although clinical features and physical examination havefailed to discriminate consistently between diastolic andsystolic heart failure by clinical assessment in previousstudies, in clinical practice it could be useful to differentiatebetween these two conditions.

• In this study, three variables easily evaluated: female sex, leftventricular hypertrophy and presence or absence of coronarydisease, enabled clinicians to differentiate between patientswith systolic or diastolic heart failure.

Appendix A

A.1. Questionnaire

Clinical variables:AgeGenderHypertensionDiabetes mellitus

HyperlipidemiaSmokerAlcohol intakeCoronary artery diseaseAnaemiaSymptoms:DyspnoeaAcute pulmonary oedemaNocturnal coughDyspnoea on exertionPhysical examination:WeightHeightBlood pressureHeart rateTemperatureNeck vein distensionCracklesS3 gallopHepatojugular refluxAnkle oedemaHepatomegalyRadiological data:CardiomegalyPleural effusionElectrocardiographical data:RhythmLeft ventricular hypertrophyQ waveLeft bundle branch blockLaboratory tests:Serum cholesterol (mg/dl)Fibrinogen (mg/dl)Serum triglycerides (mg/dl)Haemoglobin (g/dl)Serum creatinine (mg/dl)Partial pressure of oxygen (pO2) (mm Hg)

2. Echographic diagnostic criteria for left ventriculardysfunction

1) Systolic dysfunction:- Depressed left ventricular ejection fraction (LVEF) b45%.

2) Diastolic dysfunction:- Evidence of abnormal diastolic function indexes:

Slow early left ventricular filling:On Doppler echocardiography of mitral diastolic flow, in the

apical projection of four cavities:E/Ab1 and DTN220 ms, in patients under 50 years of ageE/Ab0.5 and DTN280 ms, in patients over 50 years of ageIn Doppler flow velocity of pulmonary veins⁎:S/DN1.5, in patients under 50 years of age.S/DN2.5, in patients over 50 years of age.Slow isovolumetric left ventricular relaxation:IVRTN92 ms, in patients under 30 years of age.IVRTN100 ms, in patients between 30 and 50 years of age.

318 A. Maestre et al. / European Journal of Internal Medicine 20 (2009) 313–318

IVRTN105 ms, in patients over 50 years of age.⁎ S/D: ratio of systolic to diastolic pulmonary venous flow

velocity; this was not determined due to the difficulties involvedin this technique.

Diastolic ventricular dysfunction required the presence ofone or more of the above criteria.

–Normal or mildly reduced ejection fraction (LVEFN45%)

References

[1] Conthe P. De la hipertensión arterial a la insuficiencia cardíaca. Cuándo ycómo iniciar tratamiento. Rev Clin Esp 2002;202(Extr2):9–16.

[2] Cowie MR, Mosterd A, Wood DA, Deckers JW, Poole-Wilson PA, SuttonGC, et al. The epidemiology of heart failure. Eur Heart J 1997;18(2):208–25.

[3] Rodriguez-Artalejo F, Guallar-Castillón P, Banegas JR, del Rey J. Trendsin hospitalizations and mortality for heart failure in Spain, 1980–1993. EurHeart J 1997;18:1771–9.

[4] Massie BM, Shah NB. Evolving trends in the epidemiologic factors ofheart failure: rationale for preventive strategies and comprehensive diseasemanagement. Am Heart J 1997;133:703–12.

[5] Mosterd A, Hoes A, Bruyne M, Deckers JW, Linker DT, Hofman A, et al.Prevalence of heart failure and left ventricular dysfunction in the generalpopulation. The Rotterdam Study. Eur Heart J 1999;20:447–55.

[6] Ho K, Pinsky J, Kannel W, Levy D. The epidemiology of heart failure: theFramingham study. J Am Coll Cardiol 1993;22(Supl A):6A–13A.

[7] Cowie MR, Fox K, Metcalfe C, Thompson SG, Coats AJ, Poole-WilsonPA, et al. Hospitalization of patients with heart failure. Eur Heart J2002;23:877–85.

[8] Cortina A, Reguero J, Segovia E, Rodriguez Lambert JL, Cortina R, AriasJC, et al. Prevalence of heart failure in Asturias (a region in the north ofSpain). Am J Cardiol 2001;87:1417–9.

[9] Varela Román A, Gonzalez Juanatey JR, Basante P, Trillo R, García-SearaJ, Martinez-Sande JL, et al. Clinical characteristics and prognosis ofhospitalised inpatients with heart failure and preserved or reduced leftventricular ejection fraction. Heart 2002;88:249–54.

[10] Dauterman K, Massie B, Gheorghiade M. Heart failure associated withpreserved systolic function: a common and costly clinical entity. Am HeartJ 1998;135:S310–319.

[11] Hogg K, Swedberg K, McMurray J. Heart failure with preserved leftventricular systolic function. Epidemiology, clinical characteristics andprognosis. J Am Coll Cardiol 2004;43:317–27.

[12] Senni M, Tribouilloy C, Rodeheffer R, Jacobsen SJ, Evans JM, Bailey KR,et al. Congestive heart failure in the community. A study of allincident cases in Olmsted County, Minnesota, in 1991. Circulation1998;98:2282–9.

[13] Fischer M, Baessler A, Hense H, Hengstenberg C, Muscholl M, Holmer S,et al. Prevalence of left ventricular diastolic dysfunction in the community.Results from a doppler echocardiographic-based survey of a populationsample. Eur Heart J 2003;24:320–8.

[14] Redfield M, Jacobsen S, Burnett J, Mahoney DW, Bailey KR, RodehefferRJ. Burden of systolic and diastolic ventricular dysfunction in thecommunity. JAMA 2003;289:194–202.

[15] Bursi F, Weston S, Redfield M, Jacobsen S, Pakhomov S, Nkomo V, et al.Systolic and diastolic heart failure in the community. JAMA2006;296:2209–16.

[16] Thomas JT, Kelly R, Thomas SJ, Stamos TD, Albasha K, Parrillo JE, et al.Utility of history, physical examination, electrocardiogram and chestradiograph for differentiating normal from decreased systolic function inpatients with heart failure. Am J Med 2002;112:437–45.

[17] Klapholz M, Maurer M, Lowe A, Messineo F, Meisner JS, Mitchell J,Kalman J, et al. Hospitalization for heart failure in the presence of a normalleft ventricular ejection fraction. J Am Coll Cardiol 2004;43:1432–8.

[18] Masoudi F, Havranek E, Smith G, Fish RH, Steiner JF, Ordin DL, et al.Gender, age, and heart failure with preserved left ventricular systolicfunction. J Am coll Cardiol 2003;41:217–23.

[19] Cleland JG, Swedberg K, Follath F, Komajda M, Cohen-Solal A, AguilarJC, et al. The EuroHeart Failure survey programme—a survey on thequality of care among patients with heart failure in Europe. 1: patientcharacteristics and diagnosis. Eur Heart J 2003;24(5):442–63.

[20] Bhatia R, Tu J, Lee D, Austin P, Fang J, Haouzi A, et al. Outcome of heartfailure with preserved ejection fraction in a population-based study. N EnglJ Med 2006;355:260–9.

[21] Vasan R, Benjamin E, Levy D. Prevalence, clinical features and prognosisof diastolic heart failure: an epidemiologic perspective. J Am Coll Cardiol1995;26:1565–74.

[22] McKee P, Castelli W, McNamara PM, Kannel WB. The natural history ofcongestive heart failure: The Framingham Study. N Engl J Med1971;285:1441–6.

[23] European Study Group on Diastolic Heart Failure of the European Societyof Cardiology. How to diagnose diastolic heart failure. Eur Heart J1998;19:990–1003.

[24] Navarro-López F, de Teresa E, López-Sendón JL, Castro-Beiras A. Guíasdel diagnóstico, clasificación y tratamiento de la insuficiencia cardíaca ydel shock cardiogénico. Informe del Grupo de trabajo de InsuficienciaCardíaca de la Sociedad Española de Cardiología. Rev Esp Cardiol1999;52(Supl 2):1–54.

[25] Vasan R, Levy D. Defining diastolic heart failure: a call for standardizeddiagnostic criteria. Circulation 2000;101:2118–21.

[26] Owan T, Hodge D, Herges R, Jacobsen S, Roger V, Redfield M. Trends inprevalence and outcome of heart failure with preserved ejection fraction. NEngl J Med 2006;355:251–9.

[27] McDermott MM, Feinglass J, Sy J, Gheorghiade M. Hospitalizedcongestive heart failure patients with preserved versus abnormal leftventricular systolic function: clinical characteristics and drug therapy. AmJ Med 1995;99:629–35.

[28] Varela Román A, Grigorian L, Barge E, Bassante P, de la Peña MG,Gonzalez-Guanatey JR. Heart failure in patients with preserved anddeteriorated left ventricular ejection fraction. Heart 2005;91(4):489–94.

[29] Zile M, Gaasch W, Carroll J, Feldman M, Aurigemma G, Schaer G, GhaliJ, Liebson P. Heart failure with a normal ejection fraction: is measurementof diastolic function necessary to make the diagnosis of diastolic heartfailure? Circulation 2001;104:779–82.