Download Congestive Heart Failure

Normal Left Ventricular Ejection
Fraction in Older Persons With
Congestive Heart Failure*
Wilbert S. Aronow, MD, FCCP; Chul Ahn, PhD; and Itzhak Kronzon, MD
Study objectives: To investigate in older patients with congestive heart failure (CHF) associated
with prior myocardial infarction or hypertension the relationship between normal left ventricular
(LV) ejection fraction and age, gender, hypertension, prior myocardial infarction, and atrial
fibrillation.
Design: A prospective study was performed in 572 older patients (age >60 years) with CHF
associated with prior myocardial infarction or hypertension and technically adequate twodimensional echocardiograms for measuring LV ejection fraction.
Setting: A long-term health-care facility.
Patients: One hundred seventy-seven men and 395 women, mean age 82 ±8 years, with CHF
associated with prior myocardial infarction or hypertension.
Measurements and results: Normal LV ejection fraction (>50%) occurred in 66 of 177 men (37%)
and in 221 of 395 women (56%) (p<0.0001). Multiple logistic regression analysis showed that
independent risk factors for normal LV ejection fraction in patients with CHF were no prior
myocardial infarction (p=0.0001; odds ratio=3.048), female gender (p=0.0004; odds ratio=1.978), and age (p=0.016; odds ratio=1.029).
Conclusions: Normal LV ejection fraction occurred in 50% of 572 older patients with CHF
associated with prior myocardial infarction or hypertension. Independent risk factors for normal
LV ejection fraction in patients with CHF were no prior myocardial infarction, female gender,
and age. (CHEST 1998; 113:867-69)
age; atrial fibrillation; congestive heart failure; female gender; hypertension; left ventricular ejection
fraction; myocardial infarction
Abbreviations: CHF=congestive heart failure; LV=left ventricular
Key words:
/^ ongestive heart failure (CHF) patients with norsystolic function should also be considered
for appropriate therapy.12 The prevalence of normal
left ventricular (LV) ejection fraction associated with
CHF in older patients has been reported to be 41%
^^ mal
in a study
including 54 patients,3 41% in 166 patients
with coronary artery disease,4 47% of 247 patients,4
and 34% of 501 patients.5 We are reporting in 572
older patients (age >60 years) with CHF associated
with prior myocardial infarction or systemic hyper¬
tension the relationship between normal LV ejection
*From the Hebrew Hospital Home (Dr. Aronow), Bronx, NY; the
Department of Geriatrics and Adult Development, Mount Sinai
School of Medicine (Dr. Aronow), New York; the Division of
Clinical Epidemiology, University of Texas Medical School (Dr.
Ahn) at Houston, Houston; and the Department of Medicine,
New York University School of Medicine (Dr. Kronzon), New
York.
Manuscript received June 24, 1997; revision accepted Septem¬
ber 9.
requests: Wilbert S. Aronow, MD, Medical Director,
Reprint
Hebrew Hospital Home, 801 Co-op City Blvd, Bronx, NY 10475
gender, systemic hypertension,
prior myocardial infarction, and atrial fibrillation.
fraction and age,
Materials
and
Methods
In a prospective study, CHF associated with prior myocardial
infarction or systemic hypertension was diagnosed in 677 of 2,535
persons (27%) in a long-term health-care facility. CHF was
if two criteria were satisfied: (1) pulmonary basilar
diagnosed
rates were heard by two physicians, including the senior author;
and (2) pulmonary vascular congestion was present on the chest
radiographs interpreted by both an experienced radiologist and
the senior author.
Technically adequate two-dimensional echocardiograms for
measuring LV ejection fraction were obtained in 572 of 677
patients (84%) at the time CHF was diagnosed. Except for
diuretic therapy, therapy for CHF was instituted after the
two-dimensional echocardiograms were obtained. The 572 per¬
sons included 177 men and 395 women, with a mean age of 82 ±8
years. All echocardiograms were interpreted by an experienced
echocardiographer (I.K.). LV volumes at end-diastole and endsystole were calculated by planimetry from the two-dimensional
study. LV ejection fraction was calculated as follows: (LV enddiastolic volume minus LV end-systolic volume)/LV end-diastolic
CHEST/113/4/APRIL, 1998
Downloaded From: http://publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21763/ on 05/07/2017
867
volumeX100%. A normal LV ejection fraction was >50%.4
Ethical standards were used in performing this research.
Patients were considered at entry into the study to have a prior
if they had a documented clinical history of
myocardial infarction
of Q-wave myocardial
myocardial infarction or ECG evidence
infarction. A systolic BP ^160 mm Hg on three occasions was
considered systolic hypertension. A diastolic BP >90 mm Hg on
three occasions was considered diastolic hypertension.
For group comparisons among patients with normal or abnor¬
mal LV ejection fraction, Fisher's Exact Tests or x2 tests were
used for dichotomous variables. Student's t tests were used for
continuous variables. The Cochran-Armitage trend test was used
to see if there is an increasing trend of normal LV ejection
fraction in CHF as the age increases in men, in women, and in
men plus women. Multiple logistic regression analysis was done
to examine the relationship between normal LV ejection fraction
associated with CHF and the baseline characteristics.
Results
Table 1 lists the baseline characteristics of patients
with CHF associated with normal vs abnormal LV
ejection fraction and p values. Except for having a
lower LV ejection fraction, the patients with prior
infarction and no hypertension were
myocardial
similar to the patients with hypertension and no prior
infarction, Atrial fibrillation was as fre¬
myocardial
quent in older patients with CHF and normal LV
ejection fraction as in older patients with CHF and
abnormal LV ejection fraction. Table 2 shows the
association of CHF with normal LV ejection fraction
with gender for different age groups and p values.
Table 3 shows the multiple logistic regression anal¬
ysis for relationship between normal LV ejection
fraction associated with CHF and baseline charac¬
teristics. Patients without a prior myocardial infarc¬
tion had a three times higher chance of having
normal LVEF than those with prior myocardial
infarction after controlling the confounding effects
of other baseline variables (p=0.0001). Women had
a two times higher chance of having normal LV
ejection fraction than men after controlling the
Table 1.Baseline Characteristics of Patients With
CHF Associated With Normal vs Abnormal LV Ejection
Fraction
Variable
Normal LVEF*
Abnormal LVEF
(n=287)
(n=285)
No.
Atrial fibrillation
Hypertension
Prior myocardial
221
66
86
202
215
infarction
*LVEF=LV
No.
(77)
(23)
(30)
(70)
(75)
174
111
102
173
(61)
(39)
(36)
(61)
261
(92)
81±8
83±7
Age, yr
Women
Men
ejection fraction.
(%)
(%)
p Value
0.002
<0.0001
0.138
0.015
<0.0001
Table 2.Association of CHF With Normal LV Ejection
Fraction With Age and Gender
Normal LV Ejection Fraction
Women
Men
No.
Age, yr
60-69
70-79
80-89
>90
All ages
4/18
18/54
35/86
9/19
66/177
(%)
(22)
(33)
(41)
(47)
(37)
No.
14/38
35/79
129/219
43/59
221/395
(%)
(37)
(44)
(59)
(73)
(56)
p Value*
0.364
0.204
0.004
0.040
<0.0001
*p values by Cochran-Armitage trend test for normal LV ejection
fraction with increasing age=0.085 for men, <0.0001 for women,
and <0.0001 for men plus women.
effects of other baseline variables
confounding There
times higher chance of
(p=0.0004). LV was a 1.3fraction
for an increment
normal
ejection
having
of 10 years of age after controlling the confounding
effects of other baseline variables (p=0.016). Atrial
fibrillation and hypertension were not independent
factors for LV
fraction in older
risk
with CHF.
patients
ejection
Discussion
Wong et al3 found in 54 older persons with CHF
that the mean LV ejection fraction increased with
age and that there was a female preponderance in
persons with normal LV ejection fraction. The prev¬
alence of atrial fibrillation was higher in persons with
normal LV ejection fraction than in persons with
abnormal LV ejection fraction. There was no signif¬
icant difference in the prevalence of hypertension or
coronary artery disease between persons with normal
or abnormal LV ejection fraction.
Pernenkil et al5 demonstrated that older persons
with CHF and normal LV ejection fraction had a
mean age, a higher prevalence of women, and
higher
a higher prevalence of no prior myocardial infarction
than older persons with abnormal LV ejection frac¬
tion. The prevalence of systemic hypertension was
similar in older persons with normal vs abnormal LV
ejection fraction, but the mean systolic BP was
LV ejection
higher in olderin persons with normal
fraction than older persons with abnormal LV
ejection fraction.
The present study showed that no prior myocar¬
dial infarction, female gender, and increasing age
were independently associated with normal LV ejec¬
tion fraction in older persons with CHF. Systemic
normal LV ejection
hypertension was associated with
fraction by univariate analysis but not by multivariate
atrial fibrillation was
analysis.in The prevalencewithofnormal
vs abnormal LV
similar older persons
ejection fraction.
868
Downloaded From: http://publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21763/ on 05/07/2017
Clinical
Investigations
Table
Between Normal LV Ejection Fraction Associated
Analysis for Relationship
3.Multiple Logistic RegressionWith
CHF and Baseline Characteristics
Baseline Characteristic
Age
Atrial fibrillation
Hypertension
No prior myocardial infarction
Sex*
Parameter Estimate
SE
p Value
Odds Ratio
95% Confidence Intervals
0.028
-0.225
0.145
1.115
0.012
0.187
0.192
0.271
0.191
0.016
0.229
0.450
0.0001
0.0004
1.029
0.798
1.156
(1.005, 1.053)
(0.553, 1.152)
(0.794, 1.684)
(1.792, 5.181)
(1.360, 2.876)
0.682
3.048
1.978
*Sex=0 if men and 1 if a
Older patients with CHF and normal LV ejection
fraction have LV diastolic dysfunction. In addition to
a decrease in LV diastolic relaxation and early dia¬
stolic filling caused by aging, older persons are more
to have LV diastolic dysfunction because they
likely
have an increased prevalence of hypertension, myo¬
cardial ischemia due to coronary artery disease, and
LV hypertrophy associated with hypertension, coro¬
nary artery disease, valvular aortic stenosis, hypertro¬
phic cardiomyopathy, and other cardiac disorders.6
The increased stiffness of the LV and prolonged LV
relaxation time impair LV early diastolic filling and
cause higher LV end-diastolic pressures at rest and
exercise in older persons.
during
A normal LV ejection fraction is frequent in older
patients with CHF. A normal LV ejection fraction
was present in 215 of 476 older patients (45%) with
CHF and a prior myocardial infarction, in 202 of 375
older patients (54%) with CHF and hypertension
with and without prior myocardial infarction, and in
72 of 96 older patients (75%) with CHF, hyperten¬
sion, and no prior myocardial infarction. Women
more frequently have a normal LV ejection fraction
than men in CHF associated with prior myocardial
infarction or with hypertension. The therapy of CHF
associated with normal LV ejection fraction is differ¬
from the therapy of CHF associated with abnor¬
mal LV ejection fraction and is discussed elsewhere.7
ent
References
McGough MF. Heart failure with normal systolic
disorder in older people. J Am Geriatr
Soc 1995; 43:1035-42
2 Aronow WS. Echocardiography should be performed in all
elderly patients with congestive heart failure. J Am Geriatr
Soc 1994; 42:1300-2
3 Wong WF, Gold S, Fukuyama O, et al. Diastolic dysfunction
in elderly patients with congestive heart failure. Am J Cardiol
1 Tresch DD,
function:
a common
1989; 63:1526-28
4 Aronow WS, Ahn C, Kronzon I.
Prognosis of congestive heart
failure in elderly patients with normal versus abnormal left
ventricular systolic function associated with coronary artery
disease. Am J Cardiol 1990; 66:1257-59
5 Pernenkil R, Vinson JM, Shah AS, et al. Course and prognosis
in patients ^70 years of age with congestive heart failure and
normal versus abnormal left ventricular ejection fraction.
Am J Cardiol 1997; 79:216-19
6 Aronow WS. Effects of aging on the heart. In: Brocklehurst
JC, Tallis RC, Fillit HM, eds. Textbook of geriatric medicine
and gerontology. 5th ed. London, UK: Harcourt, Brace, Intl
1997; 255-62
7 Aronow WS. Treatment of congestive heart failure in older
persons. J Am Geriatr Soc 1997; 45:1252-58
CHEST/113/4/APRIL, 1998
Downloaded From: http://publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21763/ on 05/07/2017