Download Preserved right ventricular ejection fraction predicts exercise

1143
JACC gol. 25, No. 5
April 1995:1143-53
HEART FAILURE
Preserved Right Ventricular Ejection Fraction Predicts Exercise
Capacity and Survival in Advanced Heart Failure
THOMAS
G. DI S A L V O , M D , M I C H A E L
MATHIER,
M D , M A R C J. S E M I G R A N ,
MD,
G. W I L L I A M D E C , M D , F A C C
Boston, Massachusetts
Objectives. This study was undertaken to determine which
exercise and radionuclide ventriculographic variables predict
prognosis in advanced heart failure.
Background. Although cardiopulmonary exercise testing is
frequently used to predict prognosis in patients with advanced
heart failure, little is known about the prognostic significance of
ventriculographic variables.
Methods. The results of maximal symptom-limited cardiopulmonary exercise testing and first-pass radionuclide ventriculography in patients with advanced heart failure referred for evalution for cardiac transplantation were analyzed.
Results. Sixty-seven patients with advanced heart failure (mean
[_+SD]; age 51 -+ 10 years, New York Heart Association functional
classes III (58%) and IV (18%); mean left ventricular ejection
fraction 0.22 -+ 0.07) underwent simultaneous upright bicycle
ergometric cardiopulmonary exercise testing and first-pass rest/
exercise radionuclide ventriculography. Mean peak oxygen consumption 0(02) was 11.8 -+ 4.2 ml/kg per min, and mean peak ageand gender-adjusted percent predicted oxygen consumption
(%Vo2) was 38 +- 11.9%. Univariate predictors of overall survival
included right ventricular ejection fraction >0.35 at rest and
>0.35 at exercise and %Vo2 _>45% (all p < 0.05). In a multivariate
proportional hazards survival model, right ventricular ejectioa
fraction >0.35 at exercise (p < 0.01) and %Vo2 >45% (p = 0.01)
were selected as independent predictors of overall survival. Univariate predictors of event-free survival included right ventricular
ejection fraction >0.35 at rest (p = 0.01) and >0.35 at exercise
(p < 0.01), functional class II (p < 0.05) and %Vo2 ->45% (p =
0.05). Right ventricular ejection fraction >0.35 at exercise (p ---0.01) was the only independent predictor of event-free survival ia
a multivariate proportional hazards model. Cardiac index at resl:,
Voz, left ventricular ejection fraction at rest, and exercise-relatot
increase or decrease >0.05 in left or right ventricnlar ejection
fraction were not predictive of overall or event-free survival in any
univariate or multivariate analysis.
Conclusions. 1) Right ventricular ejection fraction >0.35 at rest
and exercise is a more potent predictor of survival in advanced
heart failure than Vo2 or %Vo2; 2) %Voz rather than Vo2 predicts
survival in advanced heart failure; 3) neither %Vo2 nor Vo2
predicts survival to the combined end point of death or admissioa
for inotropic or mechanical support in patients with advanced
heart failure.
(J Am CoU Cardiol 1995;25:1143-53)
Despite recent advances in medical and surgical therapy,
patients with symptomatic congestive heart failure still have a
generally poor prognosis (1). A variety of clinical, hemodynamic and ventriculographic variables, including age, presence
of a third heart sound ($3) ventricular gallop, left ventricular
ejection fraction, left ventricular end-diastolic volume index,
pulmonary capillary wedge pressure and pulmonary vascular
resistance, have been shown to predict prognosis in large
groups of patients with heart failure with symptoms of varying
severity (2-16). Markers of reflex neurohumoral activation,
such as plasma norepinephrine, plasma renin activity and atrial
naturietic peptide, are elevated in heart failure and have also
been shown to portend a poorer prognosis in similar groups
(17-20). However, considerable difficulty remains in estimating
the prognosis of a subgroup of patients at greatest risk--those
with advanced heart failure referred for evaluation for cardiac
transplantation (21). In this more homogeneous patient group,
conventional clinical, hemodynamic and neurohumoral predictors of prognosis have not been shown to be as useful in predicting outcome.
Functional capacity assessed by peak oxygen consumption
(peak Vo2) during cardiopulmonary exercise testing has been
shown to predict survival in advanced heart failure (21-231,.
Peak Vo 2 <10 ml/kg per min is associated with the poorest
prognosis, and such patients are often recommended for
cardiac transplantation (21). Peak Vo 2 >14 ml/kg per rain
predicts overall 2-year survival at least equivalent to that for
cardiac transplantation, and such patients are usually followed
on medical therapy. Patients with peak Vo 2 between 10 and 14
ml/kg per min have an intermediate prognosis.
Despite the utility of peak Vo: in risk stratification ~f
patients with advanced heart failure, little is known about
additional prognostic features that can be identified during
From the Cardiac Unit, Department of Medicine, Massachusetts General
Hospital and Harvard Medical School, Boston, Massachusetts 02114.
Manuscript received August 3, 1994; revised manuscript received November
4, 1994, accepted November 7, 1994.
Address for corresnondence: Dr. Thomas G. Di Salvo, Massachusetts
General Hospital Heart Failure Center, Bulfinch 211, Massachusetts General
Hospital, Boston, Massachusetts 02114.
©1995 by the American College of Cardiolo~,
0735-1097/95/$9.50
0735- 1097(94)0051 I-N
1144
D1 SALVO ET AL.
R I G H T V E N T R I C U L A R FUNCTION IN A D V A N C E D H E A R T F A I L U R E
cardiopulmonary exercise testing. This study attempted to
determine whether rest and peak exercise first-pass radionuclide ventriculography performed during simultaneous cardiopulmonary exercise testing might provide any additional prognostic information beyond that obtained by measurement of
peak Vo2.
Methods
Study patients. The study group included 67 ambulatory
patients with advanced, symptomatic heart failure referred for
evaluation for cardiac transplantation between 1988 and 1992
to the Massachusetts General Hospital Heart Failure Center.
All patients underwent simultaneous cardiopulmonary exercise testing and rest and peak exercise first-pass radionuclide
ventriculography as part of initial evaluation for cardiac transplantation. Patients unable to exercise because of a requirement for intravenous inotropic agents or mechanical circulatory support were excluded. Mean (_+SD) patient age was 51 _+
10 years (range 17 to 61), and 79% were male. Heart failure
etiology was ischemic (46%) or dilated (54%) cardiomyopathy;
24% of patients were in New York Heart Association functional class II, 58% in class III, and 18% in class IV. Before
cardiopulmonary exercise testing, medical therapy with
digoxin, diuretic drugs and angiotensin-converting enzyme
inhibitors had been optimized on the basis of symptoms,
physical examination, renal function and, in most instances,
invasive hemodynamic monitoring. Therapy was adjusted to
achieve a pulmonary capillary wedge pressure <18 mm Hg and
cardiac index >2.2 liters/rain per ma whenever possible.
Exercise testing and radionuclide ventriculography. Exercise testing was performed after an overnight fast; oral medications were not withheld. Upright symptom-limited cycle
ergometry was performed on an electronically braked bicycle
ergometer (pedal-mode ergometer, Warren E. Collins, Inc.) at
a constant cycle speed of 60 rpm using a continuous ramp
protocol in which work load increased continuously by
12.5 W/min up to peak tolerance. Heart rate, blood pressure
and the 12-lead electrocardiogram were monitored continuously. Predicted maximal heart rate was estimated as
220 beats/rain minus age. Peak work load (W) and exercise
duration (rain) were noted. Breath-to-breath respiratory gas
exhange analysis was performed by means of a mouthpiece and
nose clips and a metabolic cart (2001 System, Medical Graphics Corp.).
Rest ventilatory and gas exchange data were averaged over
the last 2 rain of the rest period. During exercise, data were
averaged over contiguous 30-s intervals, except at peak exercise, when 15-s averaging was used to improve time resolution.
Oxygen uptake, carbon dioxide output, respiratory quotient,
minute ventilation and ventilatory equivalent for oxygen were
calculated by on-line computer as previously reported (24).
The anaerobic threshold was determined by the V slope
method (25). Maximal oxygen uptake (peak Vo2) was defined
as the highest Vo 2 measured during the last minute of
symptom-limited exercise. Predicted values for maximal oxy-
JACC Vol. 25, No. 5
April 1995:1143-53
gen consumption during upright bicycle exercise were calculated from a gender-specific regression equation incorporating
age, height and weight (26).
Rest and peak exercise first-pass radionuclide ventriculography were performed with a multicrystal gamma camera
(Baird-Atomic System 77, Baird Corporation) in the anterior
view with the patient upright at rest and at peak exercise. Red
blood cells were pretreated with 3 ml of intravenous stannous
pyrophosphate 20 min before first-pass radionuclide ventriculographic imaging. Images were acquired at 0.025 s/frame after
the intravenous bolus administration of 10 mCi of technetium99m (Tc-99m) pertechnetate at rest and after a second intravenous bolus of 15 mCi Tc-99m pertechnetate at peak exercise.
Each study was completed within 50 s.
Data were stored and analyzed with the use of MAC77
software (Scinicor Inc.). A region of interest was selected over
each ventricle, and a time-activity curve was generated. Peak
activity was considered to be end-diastole, and the activity
minimum was considered to be end-systole. Only cycles with
>70% of the maximal end-diastolic activity in the end-diastolic
frame were included for analysis. Background was taken as the
activity within the ventricular region of interest before the first
ventricular beat. The background-subtracted ventricular beats
were summated to generate a single representative cardiac
cycle. The right and left ventricular ejection fractions were
derived from time-activity curves as (End-diastolic counts End-systolic counts)fEnd-diastolic counts. Previous studies
from our laboratory (27) have demonstrated an ejection fraction estimation interobserver error of _+4% for this method of
ejection fraction calculation. The left ventricular end-diastolic
and end-systolic volumes were calculated from the regions of
interest with the use of a single-plane area-length method and
divided by the body surface area to obtain the left ventricular
end-diastolic and end-systolic volume indexes. Left ventricular
stroke volume index was calculated as end-diastolic volume
index minus end-systolic volume index.
Preserved right ventricular ejection fraction at rest was
defined as right ventricular ejection fraction ->0.35, the lower
limit for right ventricular ejection fraction within 2 SD of the
mean right ventricular ejection fraction measured in our
laboratory in 16 healthy sedentary volunteer control subjects.
In accord with previous published studies (28-31), an increase
in rest left or right ventricular ejection fraction ->0.05 at peak
exercise was considered normal.
Clinical follow-up and outcome. Medical therapy was optimized before hospital discharge. After initial evaluation for
transplantation, patients were followed up every 3 to 6 months
at the Massachusetts General Hospital Heart Failure Center.
Outpatient medical therapy was adjusted as necessary to
achieve an edema-free state and maximally tolerated dose of
vasodilator therapy.
Follow-up data were collected by review of all Massachusetts General Hospital and clinic records and telephone interview of the referring physician, patient or patient's family.
Follow-up data collected included date of death, date of
transplantation, need for admission to hospital for inotropic or
JACC Vol. 25, No. 5
April 1995:1143~3
DI SALVO ET AL.
R I G H T VENTRICULAR FUNCTION IN ADVANCED H E A R T F A I L U R E
mechanical support before transplantation and number and
timing of hospital admissions for decompensated heart failure.
Follow-up information up to the time of analysis was available
for all 67 patients.
Deaths occurring within 24 h without antecedent symptomatic progression of heart failure were defined as sudden.
Deaths from progressive symptoms of hemodynamic deterioration were defined as due to progressive heart failure. Two
study end points were defined: 1) death (either sudden or from
progressive heart failure) and 2) a combined end point of
either death or pretransplantation admission for inotropic or
mechanical bridging to cardiac transplantation.
Statistical analysis. Univariate and multivariate analyses
were performed to determine which demographic, exercisederived and radionuclide ventriculographic variables were
predictive of the two study end points, 1) death, and 2) the
combined end point. Univariate and multivariate analyses for
each end point were performed separately. Univariate and
multivariate analyses of survival to each end point were also
performed separately. Overall survival was defined as freedom
from death (sudden or due to progressive heart failure);
event-free survival was defined as freedom from the combined
end point (death or intotropic/mechanical bridging to transplantation).
To allow clinically meaningful univariate and multivariate
comparisons between groups, the following continuous variables were analyzed as stratified ordinal variables: peak Voz
(-<10, >10 to <14 and ->14 ml/kg per min); peak percent
predicted age- and sex-adjusted Vo 2 (-<30, >30 to <45 and
->45 ml/kg per min); right ventricular ejection fraction at rest
or exercise (<0.35 or ->0.35); left ventricular ejection fraction
at rest (-<0.20, >0.20 to <0.30, and ->0.30); left ventricular
end-diastolic volume index at rest (<150 or ->150 ml/m2);
left ventricular end-systolic volume index at rest (<120 or
->120 ml/m2); and cardiac index at rest (<2.8 or ->2.8 liters/
m2). A right ventricular ejection fraction at rest or exercise
>-0.35 was considered preserved right ventricular ejection
fraction in accord with studies at our and other institutions
(27). Right and left ventricular ejection fraction discordance
("ventricular discordance") was defined as right ventricular
ejection fraction at rest minus left ventricular ejection fraction
at rest ->0.10. Peak Vo2 and peak %Vo 2 were stratified into
three groups in accord with previous studies (22) and to allow
maximal opportunity to detect significant univariate and multivariate associations with other variables. Left ventricular
end-diastolic volume index, left ventricular end-systolic volume
index and cardiac index were stratified by the mean values for
the study group. All analyses were performed using these
stratified variables, which were recorded for all 67 patients.
Because of radionuclide first-pass bolus dispersion, seven
patients did not have accurate determination of left ventricular
ejection fraction at peak exercise or left ventricular enddiastolic volume index at peak exercise; these variables were
therefore not included in any univariate or multivariate anaLyses.
Univariate differences between groups were compared by
1145
Fisher exact tests for dichotomous variables, Mantel-Haenszel
chi-square tests for trend for nondichotomous ordinal variables, nonpaired t tests for normally distributed continuous
variables and Wilcoxon rank-sum tests for nonnormally distributed continuous variables. Multivariate logistic regression was
used to identify the most important predictors of death and the
combined end point. Multiple linear best possible all-subsets
regression was used to identify the predictors of peak %Vo 2.
Life-table analysis was used to determine overall and
event-free survival for the entire study group, Differences in
overall and event-free survival between groups were detected
by the Kaplan-Meier product-limit method and compared
by the Peto-Prentice generalized log-rank test for two groups
or the Peto-Prentice generalized log-rank test for trend for more
than two groups. To identify the most important predictors of
overall and event-free survival, multivariate Cox proportional
hazards analysis was used. In analyses of overall survival, cardiac
transplantation was considered a censored observation; in analyses of event-free survival, hospital admission for inotropic or
mechanical bridging to transplantation was considered a noncensored observation, and elective admission for transplantation
without inotropic/mechanical bridging was considered a censored
observation.
A p value <0.05 was considered statistically significant. All
analyses were performed by BMDP statistical software.
Results
Clinical characteristics, cardiopuimonary exercise testing
results and radionuclide ventriculographic findings. The clinical characteristics and results of cardiopulmonary exercise
testing and first-pass radionuclide ventriculography are presented in Table 1. Medical therapy included digoxin (79%),
diuretic drugs (83%), angiotensin-converting enzyme inhibitots (77%), hydralazine (8%) and vesnarinone (5%); 29% of
patients were receiving type 1A or type 3 antiarrhythmic
agents. An automatic implantable cardioverter-defibrillator
had previously been implanted in 17% of patients.
There were no significant adverse outcomes during cardiopulmonary exercise testing; three patients experienced transient angina. At peak exercise, mean systolic blood pressure
was 120 _+ 24 mm Hg, mean peak heart rate was 134 _+
37 beats/min, and mean age-adjusted peak heart rate achieved
82 +_ 22% (Table 1). Mean peak Vo 2 was 11.8 _+ 4.2 ml/kg per
min (range 6.2 to 28.7). Division into strata revealed that 24
patients (36%) had a peak Vo 2 <10 ml/kg per min, 26 (39%)
had a peak Vo 2 between 10 and 14 ml/kg per rain, and 17
(25%) had a peak Vo2 >14 ml/kg per min. Mean peak %Vo2
was 38 _+ 12%. Further division into strata revealed that 15
patients (22%) had peak %Vo 2 <30%, 32 (48%) had peak
%Vo 2 between 30% and 45%, and 20 (30%) had peak %Vo2
>45%. Mean respiratory gas-exchange ratio at peak exercise
was 1.13 _+0.19; mean anaerobic threshold was 7.9 +_ 2.3 ml/kg
per min; and mean anaerobic threshold as a percent of peak
Vo z was 66.7 _+ 11%.
Mean left ventricular ejection fraction at rest was 0.22 _+
1146
D~ SALVO ET AL.
R I G H T V E N T R I C U L A R FUNCTION IN ADVANCED H E A R T F A I L U R E
JACC Vol. 25, No. 5
April 1995:1143-53
Table 1. Clinical Characteristics and Exercise and Radionuclide
Findings in 67 Patients
Age (yr)
Gender (%)
Male
Female
NYHA (%)
II
III
IV
Etiology (%)
Ischemic CM
Nonischemic CM
SBP, rest (ram Hg)
SBP, ex (ram Hg)
HR, rest
HR, ex
MPHR (%)
Watts
Vo z (ml/kg per min)
%Vo2
R (ex)
AT (ml/kg per min)
AT% peak Vo 2
LVEF, rest
LVEF, ex
RVEF, rest
RVEF, ex
LVEDVI, rest (ml/m z)
LVESVI, rest (ml/m 2)
CI, rest (liters/rain per m z)
C u m u l a t i v e Survival
51 +_ 10 (17-61)
79
21
0.8
24
58
18
0.6
46
54
104 _+ 17 (80-170)
120 _+ 24 (75 174)
87 _+ 20 (54-143)
134 _+ 37 (60-300)
82 _+ 22 (39-176)
73 _+ 34 (23-188)
11.8 ~ 4.2 (6.2-28.7)
38 _+ 12 (17-62)
1.13 _+ 0.19 (0.68-1.65)
7.9 _+ 2.3 (2.9-14.2)
66 _+ ll (44-99)
0.22 + 0.07 (0.9-0.39)
0.24 = 0.07 (0.9-0.40)
0,29 + 0.ll (0.10-0.53)
0.28 _+ 0.ll (0.13-0.56)
155 -+ 48 (90-388)
121 _+ 47 (41-352)
2.9 _+ 0.8 (1.8-4.8)
Data presented are mean value _+ SD (range) or percent of patients. AT =
anaerobic threshold; AT %Vo 2 = anaerobic threshold as a percent of measured
peak Vo2; CI = cardiac index; CM = cardiomyopathy; ex - exercise; HR heart rate; LVEDVI = left ventricular end-diastolic volume index; LVEF - left
ventricular ejection fraction; LVESVI - left ventricular end-systolic volume
index; MPHR = maximal predicted heart rate; NYHA = New York Heart
Association functional class; R = respiratory exchange ratio; RVEF = right
ventricular ejection fraction; SBP = systolic blood pressure; Voz = peak oxygen
consumption; %Vo 2 - peak percent predicted oxygen consumption.
0.07 (range 0.09 to 0.39) and did not increase significantly
during exercise (left ventricular ejection fraction at peak
exercise 0.24 _+ 0.07, p = NS). At peak exercise left ventricular
ejection fraction increased ->0.05 in 11 patients (17%) and
decreased ->0.05 in 4 (6%). Mean right ventricular ejection
fraction at rest was also depressed at 0.29 _+ 0.11 (range 0.10 to
0.53) and did not increase significantly with exercise (right
ventricular ejection fraction at peak exercise 0.28 _+ 0.11, p =
NS). At peak exercise, right ventricular ejection fraction
increased >0.05 in 12 patients (18%) and decreased ->0.05 in
11 (17%). A total of 24 patients (36%) had a right ventricular
ejection fraction at rest ->0.35 (preserved right ventricular ejection fraction)• There was no difference in left ventricular ejection fraction at rest between patients with preserved
or nonpreserved right ventricular ejection fraction at rest (left
ventricular ejection fraction 0.22 + 0.07 vs. 0.23 _+ 0.07,
respectively, p = 0.60). Overall, 29 patients (43%) exhibited
ventricular discordance at rest. There was no difference in
\
0.4
0.2
0
0
25
50
75
100
125
150
Weeks
Figure 1. Life-table analysis of survival for the entire study group:
solid line = overall survival; dotted line = event-flee survival.
duration of heart failure symptoms or diagnosis (ischemic vs.
dilated cardiomyopathy) between those patients with and
without ventricular discordance. Mean left ventricular endsystolic volume index was 121 _+ 47 ml/m 2 at rest.
Outcome and survival. Sixty patients (90%) were deemed
suitable candidates and were eventually listed for cardiac
transplantation. Median duration of follow-up was 58 weeks
(range 1 to 180). Seventeen patients (23%) were admitted to
the hospital for exacerbation of heart failure (excluding admissions for inotropic/mechanical support as bridging to transplantation); 30 patients (45%) required an increased dose of
an established or a new diuretic drug; and 13 patients (20%)
required an increased dose of angiotensin-converting enzyme
inhibitor during follow-up. Only 1 of the 17 patients with an
automatic implantable cardioverter-defibrillator experienced
an automatic implantable cardioverter-defibrillator firing during follow-up.
Sixteen patients (24%) died, of whom 14 (88%) died out of
hospital and suddenly. Two nonsudden deaths occurred in
hospitalized patients awaiting transplantation (progressive
heart failure [n = 1]; intractable heart failure during an episode of
catheter-related sepsis [n = 1]).
A total of 29 patients (42%) reached the combined end
point of death (n = 16) or hospital admission for inotropic/
mechanical bridging to transplantation (n = 13). Among the 23
patients (35% of the study group) who underwent transplantation, 13 (56% of all who underwent transplantation) were
admitted for inotropic/mechanical bridging before transplantation; 10 patients (44% of all who underwent transplantation)
underwent transplantation as outpatients.
By life-table analysis, median overall survival was estimated
to be 140 weeks and median event-free survival 98 weeks
(Fig. 1).
JACC Vol. 25, No. 5
April 1995:1143-53
DI SALVO ET AL.
RIGHT VENTRICULAR FUNCTION IN ADVANCED HEART FAILURE
Table 2. Univariate and Multivariate Predictors of Death and
Combined End Point
Predictor
Univariate
Age
Gender
Diagnosis
NYHA
MPHR
Peak Vo 2
Peak %Vo2
RVEF <0.35, rest
RVEF <0.35, ex
LVEF, rest
Discordance
LVEDVI, rest
LVESVI, rest
CI, rest
AICD
Multivariate
RVEF, rest
RVEF, ex
Death
(p value)
Combined End Point
(p value)
0.06
0.18
1.0
0.05
0.71
0.34
0.04*
0.04*
0.05
0.10
0.04"
0.38
0.93
0.39
0.27
0.16
0.78
0.81
0.05
0.37
0.47
0.31
0.01"
0.01"
0.29
0.22
0.06
0.22
0.63
0.22
0.04*
--
-0.03*
*p < 0.05. AICD = automatic implantable cardioverter-defibrillator; other
abbreviations as in Table 1.
Predictors of death and combined end point. From among
the clinical, exercise and radionuclide variables recorded
during cardiopulmonary exercise testing (age, gender, diagnosis, functional class, percent predicted heart rate maximum
achieved, peak Vo 2, peak %Vo2, watts achieved, right ventricular ejection fraction at rest and at exercise, left ventricular
ejection fraction at exercise, right and left ventricular ejection
fraction discordance, left ventricular end-diastolic volume index and left ventricular end-systolic volume index at rest and
cardiac index at rest), four significant univariate predictors of
death were identified: right ventricular ejection fraction <0.35
at rest and <0.35 at exercise, peak %Vo 2 <30% and ventricular ejection fraction discordance (Table 2). Peak Vo 2 was not
a significant predictor of death in this patient group when
analyzed as a continuous, binary (<14 or ->14 ml/kg per min)
or stratified ordinal (-<10, >10 to <14, ->14 ml/kg per min)
variable. In multivariate logistic regression analysis, right ventricular ejection fraction <0.35 at rest was selected as the most
potent predictor of death.
From among the same candidate clinical, exercise-derived
and radionuclide ventriculographic variables, three significant
predictors of the combined end point were identified: right
ventricular ejection fraction <0.35 at rest and <0.35 at exercise and functional class IV (Table 2). Left ventricular enddiastolic volume index >175 ml/m2 was a nearly significant
univariate predictor of the combined end point. Neither peak
Vo 2 nor peak %Vo 2 analyzed as either continuous, discrete
binary or discrete stratified variables were significant predictors of the combined end point. In multivariate logistic regression analysis, right ventricular ejection fraction <0.35 at exercise was the most potent predictor of the combined end point.
1147
Table 3. Univariateand MultivariatePredictors of Overall and
Event-Free Survival
Predictor
Univariate
Age
Gender
Diagnosis
NYHA
MPHR
Peak Vo z
Peak %V%
RVEF ->0.35, rest
RVEF ->0.35, ex
LVEF rest
Discordance
LVEDVI, rest
LVESVI, rest
CI, rest
AICD
Multivariate
RVEF ->0.35, ex
Peak %Vo 2
Age
Overall Survival
(p value)
Event-Free Survival
(p value)
0.30
0.38
0.84
0.06
0.36
0.44
0.02*
0.03*
0.02*
0.24
0.06
0.83
0.65
0.34
0.11
0.16
0.22
0.83
0.03"
0.36
0.93
0.05
0.02*
0.01'
0.30
0.10
0.08
0.17
0.62
0.20
0.006*
0.010'
0.013"
0.016'
NS
NS
*p < 0.05. Abbreviations as in Tables 1 and 2.
An increase or decrease in either right or left ejection
fraction at peak exercise was not predictive of death or the
combined end point in any univariate or multivariate analysis.
Survival analysis. Univariate and multivariate predictors
of overall survival are presented in Table 3. Significant univariate predictors of overall survival included age, peak %Vo2
->45%, right ventricular ejection fraction ->0.35 at rest and
->0.35 at exercise and functional class II. Peak Vo2 was not
predictive of overall survival as either a discrete binary or
stratified variable. Overall survival stratified by peak Vo 2, peak
%Vo 2 and right ventricular ejection fraction at rest is displayed
in Figure 2. In a Cox proportional hazards model of overall
survival, age, peak %Vo2 and right ventricular ejection fraction
->0.35 at exercise were selected as independent predictors of
overall survival (Table 3).
Univariate and multivariate predictors of event-free survival are presented in Table 3. Significant univariate predictors
of event-free survival included functional class II, right ventricular ejection fraction ->0.35 at rest and ->0.35 at exercise
and peak %Vo 2 ->45%. Event-free survival stratified by peak
Vo2, peak %Vo2, and right ventricular ejection fraction at rest
is displayed in Figure 3. In a Cox proportional hazards model,
only right ventricular ejection fraction ->0.35 at exercise was
found independently to predict event-free survival (Table 3).
An increase or decrease in right or left ejection fraction at
peak exercise was not predictive of overall or event-free
survival in any univariate or multivariate analysis.
Predictors of peak age- and gender-adjusted peak %Vo2
achieved. The relation among right and left ejection fraction
at rest, peak V% and peak % g o 2 is displayed in Figure 4. Significant univariate predictors of peak %Vo2 included percent
1148
DI S A L V O E T AL.
R I G H T V E N T R I C U L A R F U N C T I O N 1N A D V A N C E D H E A R T F A I L U R E
Figure 2. Log-rank analysis of overall survival stratified by peak
oxygen uptake (Vo2), peak age- and gender-adjusted %Vo 2 and rest
right ventricular ejection fraction. Top, Overall survival stratified by
peak Vo2: solid line = Vo2 -<10 ml/kg per rain; dotted line = Vo 2 10
to 14 ml/kg per min; dashed line = Vo 2 ->14 ml/kg per min. Middle,
Overall survival stratified by peak %Vo2: solid line = %Vo 2 <30%;
dotted line = % V o 2 30% to 45%; dashed line = % V o 2 ->45%.
Bottom, Overall survival stratified by right ventricular ejection fraction
at rest: solid line = right ventricular ejection fraction <0.35; dotted
line = right ventricular ejection fraction ->0.35.
Cumulative
.
1
J A C C Vol. 25, No. 5
April 1995:1143 53
Figure 3. Log-rank analysis of event-free survival stratified by peak
oxygen uptake (Vo2), peak age- and gender-adjusted %Vo2 and rest
right ventricular ejection fraction. Top, Event-free survival stratified by
peak Voz: solid line = Vo z <10 ml/kg per rain; dotted line - Vo 2 10
to 14 ml/kg per min; dashed line = Vo 2 ->14 ml/kg per min. Middle,
Event-free survival stratified by peak %Vo2: solid line = %Vo 2
-<30%; dotted line = Vo 2 30% to 45%; dashed line = % V o 2 ->45%.
Bottom, Event-free survival stratified by right ventricular ejection
fraction at rest: solid line - right ventricular ejection fraction <0.35;
dotted line - right ventricular ejection fraction ->0.35.
survival
Cumulative
survival
1
m
0.8
_
0.8
• ,.I..
0.6
I
0.6
-I.
'
0.4
0,4
0.2
0.2
0
25
50
75
1O0
125
150
0
25
50
75
Weeks
Cumulative
100
125
150
Weeks
Cumulative
survival
survival
1
- 'I._ _[ .
0.8
. . . . . .
I
O.
"-_--_1 .....
L
__
I
O.
0.6
L
0.4
O.
0.2
O.
,
__
I
I'
,
.
.....
_-.
l
0
0
0
25
50
75
100
125
25
50
50
Cumulative
75
1 O0
125
150
Weeks
Weeks
survival
1
Cumulative
survival
1
0.8
%
03
[
0.6
0.6
0.4
I
0.4
0.2'
0.2
0
0
25
50
75
Weeks
1 O0
125
150
0
25
50
75
Weeks
100
125
150
JACC Vol 25, No. 5
April 1995:1143-53
1149
Dz SALVO ET A L
R I G H T VENTRICULAR FUNCTION IN ADVANCED H E A R T F A I L U R E
RVEF
RVEF
60
60
50
50
|
,=
•
40
== I
40
...... !"........
•
30
30
,,%
•
20
•
.,
=.
10
9
12
15
18
21
24
27
30
==
,.,
°
°,
10
0
3
•=
20
=
m
0
10
30
20
40
V02
50
60
70
60
70
%V02
LVEF
LVEF
50
50
40
40
•
°
==
30
30
• -.:- -...
20
.°
•* =
10
m,
°,
I
0
3
6
9
12
• "
•
•
°
="
",
•
°
•
."
lol
I
0
•
20
"," • ° , "
15
18
21
24
27
V02
maximal heart rate achieved, right ventricular ejection fraction
->0.35 at rest and ->0.35 at exercise, ventricular ejection fraction
discordance, left ventricular end-systolic volume index, functional
class II and systolic blood pressure at peak exercise (Table 4). Left
ventricular ejection fraction at rest, an increase or decrease in
either left or fight ejection fraction at peak exercise, left ventricular end-diastolic volume index, systolic blood pressure at rest
and cardiac index at rest were not predictive of peak %Vo2. A
multivariate linear regression analysis from all significant univariate candidate variables identified right ventricular ejection fraction ->0.35 at rest and percent maximal heart rate achieved as
independent predictors of peak %Vo 2.
Comparison of patients by right ventricular ejection fraction. There was no difference in age, gender, diagnosis, functional class, peak V02, percent predicted heart rate achieved,
left ventricular ejection fraction, left ventricular end-diastolic
volume index, left ventricular end-systolic volume index, cardiac index at rest or duration of symptoms in those patients
with or without preserved right ventricular ejection fraction
(Table 5). Patients with preserved right ventricular ejection
fraction achieved a significantly higher peak %Vo 2 but not
peak Vo 2 and had longer median overall and event-free
survival.
30
ol
10
20
30
40
50
%V02
Figure 4. Plots of rest right (RVEF) and left ventricular ejection
fraction (LVEF) versus peak oxygen uptake (Vo2) and peak %Vo2.
Top left, Right ventricular ejection fraction versus peak Vo2: r = 0.27,
p = 0.02. Top right, Right ventricular ejection fraction versus peak
%Vo2: r = 0,37, p = 0.00l. Bottom left, Left ventricular ejection
fraction versus peak Vo2: r = 0.05, p = 0.7. Bottom right, Left
ventricular ejection fraction versus peak %Vo2: r = 0.17, p = 0.15.
Discussion
The major finding of this study is the importance of right
ventricular function in the prediction of both overall and
event-free survival in patients with advanced heart failure.
Preserved rest or exercise right ventricular ejection fraction
was found to better predict both overall and event-free survival
than peak Vo2. Although peak %Vo2 did predict overall and
event-free survival, it was a less potent predictor than right
ventricular ejection fraction. In addition, ventricular ejection
fraction discordance was identified as an important predictor
of functional capacity in patients with advanced heart failure.
Right ventricular ejection fraction. Many studies have
shown that left ventricular ejection fraction, diastolic or systolic dimensions and filling pressures predict survival but not
functional capacity in patients with advanced heart failure (32).
1150
DI SALVO ET AL.
RIGHT VENTRICULAR FUNCTION IN ADVANCED HEART FAILURE
Table 4. Univariate and Multivariate Predictors of Peak Age- and
Gender-Adjusted Peak Oxygen Consumption During
Cardiopulmonau Exercise Testing
Predictor
Univariate
NYHA
SBP, rest
SBP, ex
MPHR
RVEF ->0.35, rest
RVEF >0.35, e×
LVEF, rest
Change in RVEF
Change in LVEF
Discordance
LVEDVI, rest
LVESVI, rest
CI, rest
Multivariate
RVEF >0.35, rest
MPHR
p Value
11.05"
0.08
0.02*
0.004*
0.001'
0.001 *
0.15
0.94
0.86
0.003*
0.08
0.04:'
0.74
0.001 *
0.009*
JACC Vol. 25, No. 5
April 1995:1143-53
Table 5. Clinical Characteristics of Patients With Nonpreserved
Versus Preserved Right Ventricular Ejection Fraction
Age (yr)
Gender (%)
Ischemic CM (%)
NYHA
Duration (too)
Peak Vo z
Peak %Vo2
R (ex)
AT
AT %peak Vo 2
MPHR
LVEF, rest
LVEDVI, rest
LVESVI, rest
CI, rest
Nonpreserved
Preserved
p Value
51 + 9
80
54
3 + 0.06
29 _+ 34
ll.3 = 4
36 _+ 11
1.15 _+ 0.21
7.7 + 2.5
67.5 _+ 11.5
81 + 22
11.22 _+ 0.07
160 2 52
126 + 50
3.0 -+ 0.0
51 + 11
76
56
3 _+ 0.04
25 + 28
12.8 + 4
43 ± 13
1.12 + 0.15
8.2 ± 2.1
65.2 - 10
80 -+ 24
0.23 + 0.07
145 + 42
111 ± 43
2.7 + 0.7
NS
NS
NS
NS
NS
0.13
0.009*
NS
NS
NS
NS
NS
NS
0.20
0.10
*p < 0.05. Data presented are mean value _+SD, unless otherwise indicated.
Abbreviations as in Table 1.
*p < 0.05. Abbreviations as in Table 1.
Only a few relatively small studies have reported the prognostic significance of right ventricular ejection fraction in patients
with advanced heart failure. Polak et al. (33) reported a
significantly higher 2-year mortality in patients with a right
ventricular ejection fraction <0.35 at rest (71% vs. 23%).
Baker et al. (34) reported that radionuclide right ventricular
ejection fraction at rest correlated with peak Vo2 (r = 0.7, p <
0.001) and was superior to left ventricular ejection fraction in
predicting exercise capacity. By contrast, Szlachcic et al. (35)
found a significant univariate but not multivariate correlation
between right ventricular ejection fraction and exercise capacity. To our knowledge, the present study is the largest reported
to date to show that in a homogeneous group of patients with
advanced heart failure, right ventricular ejection fraction at
rest or exercise predicts both survival and functional capacity
as assessed by peak V% during exercise. In addition, the
prognostic importance of right ventricular ejection fraction was
not dependent on duration or etiology of heart failure.
That the right ventricle plays an important role in determining exercise capacity in advanced heart failure is supported
by previous clinical trials. Primary venodilators such as nitrates, which improve right ventricular loading conditions (in
the case of nitrates by decreasing both right ventricular preload
and afterload), increase exercise capacity in advanced heart
failure more than primary arterial vasodilators such as enalapril, which affect right ventricular loading conditions considerably less (36,37). Patients treated with long-term captopril who
experience a greater decrease in pulmonary vascular resistance
(and reduction in right ventricular afterload) have a greater
improvement in cardiac index, stroke volume index and left
ventricular stroke work index than those who do not, suggesting that increased right ventricular output as a result of
decreased right ventricular afterload leads to increased left
ventricular filling and output in advanced heart failure (38).
Observations from these studies have led to the hypothesis
that exercise capacity in patients with advanced heart failure
may in fact be more closely related to right ventricular than left
ventricular performance (38). Both a decrease in pulmonary
vascular resistance (with concomitant decrease in right ventricular afterload) and recruitment of the contactile reserve of
right ventricular free wall contribute to the normal increase in
right ventricular ejection fraction seen during exercise in
healthy volunteers (28,29). Inadequate right ventricular free
wall systolic performance (from previous infarction or ischemia) or contractile reserve (from increased right ventricular
afterload secondary to elevated pulmonary vascular resistance
or pulmonary hypertension) both may prohibit the normal
increase in right ventricular ejection fraction and right-sided
cardiac output.
Peak Voz. Peak Vo 2 has been shown in multiple studies
(5,21,32) to predict survival in patients with symptomatic heart
failure. Mancini et al. (22) classified patients with advanced
heart failure referred for evaluation for cardiac transplantation
into two groups on the basis of peak Vo 2. Group 1 had peak
Vo: <14 ml/kg per min (n = 35, mean Vo: 11.2 + 2.5), and
group 2 had peak Vo 2 >14 ml/kg per rain (n = 52, mean Vo:
19.0 _+ 4.1) during symptom-limited treadmill cardiopulmonary exercise testing. The two groups were comparable in age,
functional class, left ventricular ejection fraction and cardiac
index at rest; pulmonary capillary wedge pressure was significantly lower in group 2. One-year actuarial survival was
reported to be significantly lower in group 1 than group 2 (70%
vs. 94%, p < 0.005). This influential study serves as the basis
for the current recommendation that patients with peak Vo 2
<14 ml/kg per rain be considered for transplantation (21).
Symptom-limited upright bicycle exercise testing as used in
the present study may underestimate peak Vo 2 by 5% to 10%
compared with symptom-limited treadmill testing as used in
the study by Mancini et al. (22) and others (21,23). However,
JACC Vol. 25, No. 5
April 1995:1143-53
DI SALVO ET AL.
R I G H T V E N T R I C U L A R FUNCTION IN A D V A N C E D H E A R T F A I L U R E
the lack of predictive power of peak Vo 2 in the present study
probably results from differences in the study patients rather
than the method of peak Vo 2 measurement. Although our
patients did not differ from those studied by Mancini et al. (22)
with regard to age, left ventricular ejection fraction or heart
failure etiology, mean peak Vo2 was significantly lower in our
patients (11.8 vs. 14.7 ml/kg per rain). Further, our patients
appeared to be more limited in the distribution of their peak
Vo2: Only five patients (7.5%) in our study achieved a Vo2
>19 ml/kg per rain, the mean value for group 2 in the study by
Mancini et al. (22). Although peak Vo 2 is of great importance
in predicting overall survival in large groups of patients with
impaired left ventricular ejection fraction but varying degress
of functional limitation, it appears to lose its predictive value in
more impaired and homogeneous patient groups. A similar
loss of predictive value has been reported for left ventricular
ejection fraction whenever it decreases <25% (1,32).
In the present study, age- and gender-adjusted peak %Vo2
but not nonadjusted peak Vo 2 predicted overall and event-free
survival. Peak Vo 2 and peak %Vo 2 assess not only cardiac
reserve, but also the adequacy of circulatory compensatory
mechanisms in heart failure (32). Such mechanisms include the
level of general conditioning status, muscle mass and perfusion
and vasoregulatory autonomic reflexes (39). Given the age and
gender dependence of these compensatory mechanisms, peak
%Vo2 is a more sensitive indicator of functional capacity than
peak Vo2, in a manner analogous to the superiority of ageadjusted rather than nonadjusted maximal heart rates during
exercise. In the present study, peak %Vo 2 <30% was associated with the worst, and %Vo2 >45% was associated with the
best, overall and event-free survival.
Exercise-related change in left or right ventricular ejection
fraction and ventricular ejection fraction discordance. Although it has been hypothesized that the ability to augment
right or left ventricular ejection fraction may predict functional
capacity and prognosis in heart failure, the present study
confirms previous smaller reports in demonstrating that the
change in left ventricular ejection fraction or right ventricular
ejection fraction from rest to peak exercise does not predict
either functional capacity or prognosis. An increase or decrease in right or left ventricular ejection fraction with exercise
was not associated in univariate or multivariate analysis with
peak Vo2, peak %Vo2, or overall or event-free survival.
Among the previous smaller studies, Szalchcic et al. (35) found
no correlation with peak Vo 2 and change in left or right
ventricular ejection fraction in patients with advanced heart
failure. Higginbotham et al. (30) found no univariate or
multivariate association between change in left ventricular
ejection fraction with peak Vo2 or metabolic equivalents
(METS) achieved during bicycle ergometry. As in these two
previous studies, the present study also found no association
between left ventricular end-diastolic volume index at rest or
change in left ventricular end-diastolic volume index during
exercise with peak Vo 2 or overall or event-free survival (30,35).
Ventricular ejection fraction discordance did not predict
overall and event-free survival but did predict functional
1151
capacity. The right and left ventricles share common muscle
bundles, a common interventricular septum and the pericardial
space (40-42). Acute decline in right ventricular systolic
function and acute increase in right ventricular diastolic volume have been shown to impair the systolic and diastolic
function of the left ventricle, respectively, as a result of
ventricular interaction (43-45). The predominant mechanism
or mechanisms of ventricular interaction are not known, but
constraint of ventricular diastolic filling by the presence of an
intact pericardium or diastolic bowing of the interventricular
septum from preferential right or left ventricular diastolic
filling probably play an important role (46). Preserved right
ventricular systolic function and resultant smaller right ventricular end-diastolic volume may minimize the adverse effects of
right ventricular interaction on left ventricular systolic performance in advanced heart failure and result in improved
functional capacity. If validated in larger studies, ventricular
ejection discordance, a noninvasive, easily acquired measure of
functional capacity, may be of potential use in the initial
stratification of patients with advanced heart failure.
Contrary to studies of patients with varying degrees of heart
failure and functional limitation, the present study of a homogeneous patient group with advanced heart failure and marked
functional limitation found that left ventricular volumetric
indexes such as left ventricular end-diastolic volume index and
left ventricular end-systolic volume index did not predict
functional capacity or overall or event-free survival.
Determinants of age- and gender-adjusted predicted peak
%Voz achieved. Analysis of the univariate and multivariate
determinants of peak %Vo 2 corroborated the important functional role of the right ventricle in advanced heart failure. In
multivariate analysis, right ventricular ejection fraction ->0.35
at exercise emerged as the most important predictor of peak
%Vo 2. In multivariate models, including a smaller patient
group with complete catheterization and first-pass rest and
exercise radionuclide ventriculography data, ventricular discordance and right ventricular stroke work index entered the
model in place of right ventricular ejection fraction ->0.35 at
exercise as the most important predictors of peak %Vo2. Thus,
right ventricular function, measured by either right ventricular
ejection fraction, right and left ventricular ejection fraction
discordance or right ventricular stroke work index, is an
important determinant of peak %Vo 2 in these patients.
Study limitations. The present study included a relatively
small number of patients. Medical therapy had been only
recently maximized in these patients, many of whom were
previously suboptimally treated. Exercise testing and peak Vo2
determinations were not performed serially in these patients.
Symptom-limited bicycle ergometric exercise testing was used,
which may underestimate peak Vo 2 by 5% to 10% compared
with symptomqimited treadmill exercise. In the present study,
right heart cardiac catheterization within 3 weeks of exercise
testing was performed in only 66% of patients after optimization of digoxin, diuretic drugs and angiotensin-converting
enzyme inhibitors; because not all patients underwent catheterization, hemodynamic variables were not included in this
1152
DI SALVO ET AL.
RIGHT VENTRICULAR FUNCTION IN ADVANCED HEART FAILURE
analysis. Further, the results of this study of patients with
advanced heart failure may not be readily generalizable or
applicable in predicting outcome or Vo2 in patients with a
milder degree of heart failure.
Summary. The present study identified rest and exercise
right ventricular ejection fractions as important predictors of
overall and event-free survival in patients with advanced heart
failure. Preservation of right ventricular ejection fraction was
the single most important predictor of overall and event-free
survival in this referral group of patients. Peak %Vo 2 but not
peak Vo 2 was predictive of overall and event-free survival in
this study; this observation probably reflected the more homogeneous patient group in this study compared with those in
previous studies. The present study also identified rest or
exercise ventricular ejection fraction and ventricular ejection
fraction discordance as important predictors of functional
capacity in advanced heart failure. Larger studies using firstpass radionuclide ventriculography in both advanced and
milder degrees of heart failure are necessary to determine the
prognostic value of right ventricular function and ventricular
ejection fraction discordance in such patients.
References
1. Francis G, Kubo S. Prognostic factors affecting diagnosis and treatment of
congestive heart failure. Curr Prob Cardiol 1989;14:631-7I.
2. Stevenson L, Tillisch J, Hamilton M, et al. Importance of hemodynamic
response to therapy in predicting survival with ejection fraction <20%
secondary to ischemic or nonischemic dilated cardiomyopathy. Am J Cardiol
1990;66:1348-54.
3. Anguita M, Arizon J, Bueno G, et al. Clinical and hemodynamic predictors
of survival in patients aged <65 years with severe congestive heart failure
secondary to ischemic or nonischemic dilated cardiomyopathy. Am J Cardiol
1993;72:413-7.
4. Cleland J, Dargie H, Ford I. Mortality in heart failure: clinical variables of
prognostic value. Br Heart J 1987;58:572-82.
5. Cohn J, Johnson G, Shabetai R, et al. Ejection fraction, peak exercise oxygen
consumption, cardiothoracic ratio, ventricular arrhythmias, and plasma
norepinephrine as determinants of prognosis in heart failure. Circulation
1993;87:Suppl VI:VI-5-16.
6. Diaz R, Obasohan A, Oakley C. Prediction of outcome in dilated cardiomyopathy. Br Heart J 1987;58:393-9.
7. Franciosa J, Wilen M, Zieshce S, Cohn J. Survival in men with severe chronic
left ventrieular failure due to either coronary artery disease or idiopathic
dilated cardiomyopathy. Am J Cardiol 1983;51:831-6.
8. Fuster V, Gersh B, Giulianai E, Tajik A, Brandenburg R, Frye R. The
natural history of idiopathic dilated cardiomyopathy. Am J Cardiol 1981;47:
525-31.
9. Grilfin B, Shah P, Ferguson J, Rubin S. Incremental prognostic value of
exercise hemodynamic variables in chronic congestive heart failure secondary to coronary artery disease or to dilated cardiomyopathy. Am J Cardiol
1991;67:848-53.
10. Kelly T, Cremo R, Nielsen C, Shabetai R. Prediction of outcome in
late-stage cardiomyopathy. Am Heart J 1990;119:1111-21.
11. Keogh A, Baron D, Hickie J. Prognostic guides in patients with idiopathic or
ischemic dilated cardiomyopathy assessed for cardiac transplantation. Am J
Cardiol 1990;65:903-8.
12. Komajda M, Jais J, Reeves F, et al. Factors predicting mortality in diopathic
dilated cardiomyopathy. Eur Heart J 1990;11:824-31.
13. Likoff M, Chandler S, Kay H. Clinical determinants of mortality, in chronic
congestive heart failure secondary to idiopathic dilated or to ischemic
cardiomyopathy. Am J Cardiol 1987;59:634-8.
14. Middlekanff H, Stevenson W, Stevenson L. Prognostic significance of atrial
fibrillation in advanced heart failure: A study of 390 patients. Circulation
1991;84:40-8.
JACC Vol. 25, No. 5
April 1995:1143-53
15. Saxon L, Stevenson W, Middlekauff H, et al. Predicting death from
progressive heart failure secondry to ischemic or idiopathic dilated cardiomyopathy. Am J Cardiol 1993;72:62-5.
16. Unverferth D, Magorien R, Moeschberger M, Baker P, Fetters J, Leier C.
Factors influencing the one-year mortality of dilated cardiomyopathy. Am J
Cardiol 1984;54:147-52.
17. Swedberg K, Eneroth P, Kjekshus J, Wilhelmsen L. Hormones regulating
cardiovascular function in patients with severe congestive heart failure and
their relation to mortality. Circulation 1990;82:1730-6.
18. Lee W, Packer M. Prognostic importance of serum sodium concentration
and its modification by converting-enzyme inhibition in patients with severe
chronic heart failure. Circulation 1986;73:257-67.
19. Francis G, Benedict C, Johnstone D, et al. Comparison of neuroendocrine
activation in patients with left ventricular dysfunction with and without
congestive heart failure: a substudy of the Studies of Left Ventricular
Dysfunction (SOLVD). Circulation 1990;82:1724-9.
20. Cohn J, Levine B, Olivari M, et al. Plasma norepinephrine as a guide to
prognosis in patients with chronic congestive heart fialure, N Engl J Med
1984;311:819-23.
21. Mudge G, Goldstein S, Addonizio L, et al. 24th Bethesda Conference on
Cardiac Transplantation. Task Force 3: Recipient guidelines/prioritization.
J Am Coil Cardiol 1993;22:21-31.
22. Mancini D, Eisen H, Kussmaul W, Mull R, Edmunds L, Wilson J. Value of
peak exercise oxygen consumption for optimal timing of cardiac transplantation in ambulatory patients with heart failure. Circulation 1991;83:778-86.
23. Itoh H, Taniguichi K, Koike A, Doi M. Evaluation of severity of heart failure
using ventilatory gas analysis. Circulation 1990;81:Suppl II:II-31-7.
24. Boucher C, Kanareck D, Okada R. Exercise testing in aortic regurgitation:
comparison of radionuclide left ventricular ejection fraction with exercise
performance at the anaerobic threshold and at peak exercise. Am J Cardiol
1983;52:801-8.
25. Yeh M, Gardner R, Adams T, Yarowitz F, Crapo R. Anarobic threshold:
problems of determination and validation. J Appl Physiol 1983;55:1178-86.
26. Hansen JE. Exercise instruments, schemes, and protocols for evaluating the
dyspneic patient. Am Rev Respir Dis 1984;129:$25-7.
27. Kaul S, Boucher C, Okada R. Sources of variability in the radionuclide
angiographic assessment of ejection fraciton. Am J Cardiol 1984;53:823-8.
28. Maddahi J, Berman D, Matsuoka D, Waxman A, Forrester J, Swan H. Right
ventricular ejection fraction during exercise in normal subjects and in
coronary artery disease patients: assessment by multiple-gated equilibrium
scintigraphy. Circulation 1980;62:133-40.
29. Morrison D, Sorensen S, Caldwell J, et al. The normal right ventricular
response to supine exercise. Chest 1982;82:686-91.
30. Higginbotham M, Morris K, Conn E, Coleman E, Cobb F. Determinants of
variable exercise performance among patients with severe left ventricular
dysfunction. Am J Cardiol 1983;51:52-60.
31. Berger H, Johnstone D, Sands J, Gottschalk A, Zaret B. Response of right
ventricular ejection fraction to upright bicycle exercise in coronary artery
disease. Circulation 1979;60:1292-9.
32. Dell'Italia L, Freeman G, Gaasch W. Cardiac function and functional
capacity: implications for the failing heart. Curr Proh Cardiol 1993;18:71158.
33. Polak J, Holman L, Wynne J, Colucci W. Right ventricular ejection fraction:
an indicator of increased mortality in patients with congestive heart failure
associated with coronary artery disease. J Am ColI Cardiol 1983;2:217-24.
34. Baker B, Wilen M, Boyd C, Dinh H, Franciosa J. Relation of right
ventricular ejection fraction to exercise capacity in chronic left ventricular
failure. Am J Cardiol 1984;54:596.
35. Szlachcic J, Massie B, Kramer B, Topic N, Tubau J. Correlates and
prognostic implications of exercise capacity in chronic congestive heart
failure. Am J Cardiol 1985;55:1037-42.
36. Leier C, Huss P, Magorien R, Unverferth D. Improved exercise capacity and
differing arterial and venous tolerance during chronic isosorbide dinitrate
therapy for congestive heart failure. Circulation 1983;67:817-22.
37. Cohn J, Archibald D, Zieschek S, et al. Effect of vasodilator therapy on
mortality in chronic congestive heart failure: Results of a Veterans Administration cooperative study. N Engl J Med 1986;314:547-52.
38. Packer M, Lee W, Medina N, Yushak M. Hemodynamic and clinical
significance of the pulmonary vascular response to long-term captopril
therapy in patients with severe chronic heart failure. J Am Coll Cardiol
1985;6:635- 45.
JACC Vol. 25, No, 5
April 1995:114333
DI SALVO ET AL.
RIGHT VENTRICULAR FUNCTION IN ADVANCED HEART FAILURE
39. Massie B. Exercise tolerance in congestive heart failure: role of cardiac
function, peripheral blood flow, and muscle metabolism and effect of
treatment. Am J Med 1988;84:Suppl 3A:75-82.
40. Dell'Italia LJ, Freeman GL, Gaasch WH. Cardiac function and functional
capacity: implications for the failing heart. Curr Prob Cardiol 1993;18:70758.
41. Bernard D, Alpert JS. Right ventricular function in health in disease. Curr
Prob Cardiol 1987;13:423-49.
42. Dell'Italia L. The right ventricle: anatomy, physiololgy, and clinical importance. Curr Prob Cardiol 1991;16:658-720.
43. Feneley MP, Gavaghan PT, Baron DW, Branson JA, Roy PR, Morgan JJ.
1153
Contribution of left ventricular contraction to the generation of right
vcntricular systolic pressure in the human heart. Circulation 1985;7l:473-80.
44. Kelly DT, Spomitz HM, Beiser GD, Pierce JE, Epstein SE. Effects of chronic
right ventricular volume and pressure loading on left ventricular performance. Circulation 1971;44:403-18.
45. Konstam MA, Cohen SR, Salem DN, et al. Comparison of left and right
ventricular end-systolic pressure-volume relations in congestive heart failure. J Am Coll Cardiol 1985;5:1326-34.
46. Janicki JS. Influence of the pericardium and ventricualr interdependence on
left ventricular diastolic and systolic function in patients with heart failure.
Circulation 1990;81:Suppl III:III-15-20.