Download Non-invasive Measurement of Cardiac Output and

Clinical Science ( 1997) 93, 195-203 (Printed in Great Britain)
I95
Non-invasive measurement of cardiac output and ventricular
ejection fractions in chronic cardiac failure: relationship to
impaired exercise tolerance
Ian C. STEELE, Ann MOOREm,Anne-Marie NUGENT, Marshall S. RILEY, Norman P. S. CAMPBELLmand
D. Paul NICHOLLS
Department of Medicine, Royal Victoria Hospital, Grosvenor Road, Belfast BTI 2 6BA, Northern Ireland, U.K.,
and *Regional Medical Cardiology Centre, Royal Victoria Hospital, Grosvenor Road, Belfast BT12 6BA,
Northern Ireland, U.K.
(Received 3 Februaty/23 May 1997; accepted 29 May 1997)
1. The role of cardiac output limitation in the
pathophysiology of exercise in patients with chronic
failure remains undefined. During steady-state submaximal exercise, oxygen uptake is similar in
patients and control subjects, but it is not known if
cardiac output is also similar. We wished to determine if the reduced exercise tolerance of patients
with chronic cardiac failure during such exercise is
related to reduced cardiac output, or to peripheral
factors.
2. Ten male patients with stable chronic failure and
ten age-matched male normal controls were studied
at rest and during exercise. Each subject performed
a familiarization exercise test, a symptom-limited
maximal exercise test and two submaximal exercise
tests. Cardiac output was measured by a carbon
dioxide rebreathing method. We also measured oxygen consumption, ventilation, Borg score of perceived exertion and venous lactate concentration,
and ejection fractions.
3. As expected, patients had lower peak oxygen consumption [median (range) 1.18 (0.98-1.76) versus
1.935 (1.53-2.31) Vmin; P <0.0011, lower peak
venous lactate concentration but a similar overall
level of perceived exertion. At the same submaximal
workload, patients and control subjects had similar
oxygen consumption [0.67 (0.59-0.80) versus 0.62
(0.52-0.82) l/min] and cardiac output [6.92
(5.79-9.76) versus 7.3 (5.99-10.38) l/min] but the
patients had a greater perceived level of exertion
[Borg score: 4 (1-6) versus 3 (1-5); P<0.005],
higher venous lactate concentration C1.6 (1-3.3)
versus 1.14 (0.7-1.7) mmoV1; P<0.05] and higher
heart rate [lo6 (89-135) versus 87 (69-112) beats/
min; P <0.0051.
4. During submaximal exercise at a similar
absolute workload, patients with cardiac failure
have a similar oxygen uptake and cardiac output
but greater anaerobiosis and increased fatigue when
compared with normal subjects. These findings
appear to relate predominantly to changes that
occur in the periphery rather than abnormalities of
central cardiac function.
INTRODUCTION
Chronic cardiac failure (CCF) is classically
defined as “a state in which the heart fails to maintain an adequate circulation for the needs of the
body despite a satisfactory venous filling pressure”
[l]. This definition identifies a reduction in cardiac
output (Qt) both at rest and during exercise as the
key problem in heart failure [2-61. However, treatments which improve central cardiac function may
not increase exercise capacity [7], and increased
limb blood flow produced by dobutamine infusion
does not improve exercise capacity or reduce lactate
production [8]. These observations indicate that
peripheral factors may also be important in exercise
limitation in CCF.
We have shown that patients with CCF adapt to
increased exercise workloads more slowly, but
achieve the same oxygen uptake (vo2) as control
subjects [9]. They also recover more slowly [lo]. voz
is closely related to Qt in normal subjects [ll, 121,
but in patients with CCF the relationship is less
clearly defined [13]. Patients with previous myocardial infarction and asymptomatic left ventricular
dysfunction also demonstrate a delay in attainment
of Po2 during constant workload exercise [14], and
Key words: cardiac output, chronic cardiac failure, ejection fraction, exercise capacity.
Abbreviations: Cacoco,, arterial COz concentration: CCF, chronic cardiac failure; CJO,, mixed venous C02 concentration; LVEF, left ventricular ejection fraction: Pma,end-tidal
COXconcentration; PFR, peak filling rate; PVG, peak achieved oxygen consumption: Qt, cardiac output; RVEF, right ventricular ejection fraction: VD, dead space: VE, minute
ventilation; VT, tidal volume; \‘co, COz production; VQ, Oz consumption.
Correspondence: Dr D. P. Nicholls.
I96
I.C.Steele et al.
the relationship between voz and Qt appears to be
quite different during incremental exercise compared with normal subjects [15].
The present study was therefore designed to compare the Qt responses to exercise in patients with
CCF and matched controls at a steady-state submaximal work rate. Non-invasive technology was
employed [121, as invasive haemodynamic monitoring may in itself produce circulatory changes [16,
171. In addition, so as to define central cardiac function further, right and left ventricular responses to
exercise were measured by first-pass radionuclide
angiography, using a new multiwire camera to give
high-definition images.
METHODS
Patients
Ten men (median age 66 years, range 37-75
years) with compensated CCF took part in the
study. All had been clinically stable for a minimum
of 3 months before the study. The mean duration
from time of diagnosis was 3.6 (range 1-6) years.
The cardiothoracic ratio was >0.50 in all cases and
all patients had a history of at least one episode of
pulmonary oedema. Six patients were in New York
Heart Association Class I1 and four in Class 111. The
aetiology of CCF was ischaemic heart disease in
eight and dilated cardiomyopathy in two. Two
patients had diabetes mellitus, one was insulindependent. All patients were in sinus rhythm. All
were being treated with diuretics [median dose 80
(range 40-200) mg of frusemide]. Four were taking
flosequinan. Six were taking angiotensin-converting
enzyme inhibitors (two taking captopril, two taking
enalapril, one taking lisinopril and one taking trandolapril). None had significant pulmonary disease
from history or spirometry (defined as forced expiratory volume in 1 s <75% or forced expiratory
volume is 1 s/functional vital capacity <75% predicted), intermittent claudication, angina or musculoskeletal disease leading to premature cessation of
exercise. Clinical evidence of fluid overload (peripheral oedema, elevated venous pressure, basal
rates) was absent at the time of the study.
Control subjects
Ten men (median age 67 years, range 65-69
years), with no evident cardiac or pulmonary disease
or other limitations to exercise, acted as control subjects. They were determined to be healthy on the
basis of normal history, examination, ECG and exercise test. All were sedentary and were taking no
medication. They were recruited by advertising in
the local newspaper for volunteers to help with
medical research. They were not hospital employees
or health care workers, and before the study, were
not acquainted with exercise testing.
Ethical approval was granted by the Ethics Committee of The Queen's University of Belfast. Written
informed consent was given by all subjects.
Protocol
Patients and controls each attended the department on 3 separate days and all were studied in an
identical manner. Studies on day 2 and day 3 were
performed after a 12 h overnight fast, as taking food
may alter ventricular function [18] and decrease
exercise tolerance in patients with CCF despite
increasing Qt [19].
Day 1. At the initial visit, history, examination,
resting ECG and familiarization bicycle exercise test
were performed. We have previously shown high
reproducibility for peak achieved oxygen consumption (PVo,) in our laboratory during treadmill exercise, in both patients with CCF and healthy
individuals after an initial familiarization test [20].
All exercise testing was performed on an upright
electronically braked cycle ergometer (Seca Cardiotest 100).
Day 2. At least 1 week later, subjects attended the
exercise laboratory in the morning after a 12 h overnight fast. No tobacco or drink other than water was
allowed for the duration of the fast, and patients
omitted their normal morning medications. A
Teflon cannula was inserted into the antecubital
vein of each arm. Thirty minutes after arriving at the
laboratory the subject sat on the exercise bicycle.
After a further 5 min the resting radionuclide scan
was performed using the cannula in the right arm
for the bolus injection (see below). There then
followed a further 30 min rest. The subject then sat
on the bicycle again for 5 min, before blood being
taken for measurement of venous lactate concentration. After this resting period the subject performed
a symptom-limited exercise test using a standardized
exponential exercise protocol, which has previously
been validated for use in patients with CCF [21].
During exercise the subject maintained a pedalling
rate of 60 rev./min, and was encouraged to exercise
as long as possible. Blood samples for lactate
measurement were taken at peak exercise and
after 3 and 6 min of recovery. Continuous online measurement of gas exchange was performed
throughout the 5 min before exercise, during excercise and for the 6 min recovery period (see below).
This test was used to obtain Pvoz (the highest 6'0,
value averaged over 15 s during the final minute of
exercise).
Duy 3. Between 3 and 10 days later subjects attended the laboratory under the same conditions as for
day 2. On arrival in the laboratory, the subjects
rested for 30 min after having a Teflon catheter
inserted into the antecubital vein of the right arm
only. They then sat on the bicycle for measurement
of resting Qt using the COz rebreathing method (see
below). Three rebreathing manoeuvres were per-
Cardiac output in chronic cardiac failure
formed at rest to obtain three measurements of Qt.
Subjects then exercised at a level corresponding to
approximately 30% of the PVoz obtained from day
2. After 6 rnin of steady-state exercise, cardiac ejection fractions and Qt were measured simultaneously.
A second rebreathing manoeuvre was performed
immediately after the first. Blood samples were
taken for lactate concentration after 6 min of exercise. After a 30 min rest period the exercise was
repeated at a level corresponding to 50% of PVo,.
Cardic ejection fraction determination, Qt measurement and blood sampling were performed as for the
30% level.
During all of the tests the 12-lead ECG was monitored continuously and blood pressure was recorded
at 3 min intervals using a mercury sphygmomanometer. After each exercise test the subjects were
invited to indicate the overall perceived level of
exertion by means of a Borg score [22].
Measurement of gas exchange
Minute ventilation (VE) was measured with a vane
turbine placed on the inspiratory side of a nonrebreathing respiratory valve circuit (dead space
88 ml) in conjunction with a ventilometer (PK Morgan, Chatham, Kent, U.K.). Interruptions of a light
beam by the vane were counted to measure inspired
volume that was then converted to expired volume
with the Haldane correction and standard formulae.
Expired gas was led through lightweight tubing into
a 5 litre mixing chamber. This was sampled continuously and expired 0 2 and C02 concentrations were
determined by paramagnetic and IR analysis respectively. The outputs from the ventilometer and gas
analysers were fed through an analog-to-digital converter to an Amstrad PC for on-line calculation of
Vo2,C02 production (Vco,), and VE. Data points
were averages of 15 s periods. Calibration of the
ventilometer was carried out each day with multiple
strokes of a standard 1 litre syringe, and the gas
analysers were calibrated before each test with gases
of known concentration. Validation of the system
has been described previously [23].
I97
sively. Vco2is measured from the mixed expired air.
Arterial C02 concentration ( C ~ C O ~
is) estimated
from end-tidal COz (PETcoz), which is measured by
a probe situated at the mouthpiece. The mixed
venous C02 (Cvcoz) is estimated by a manoeuvre
which involves rebreathing a high concentration of
C02 in oxygen to achieve an equilibrium [24]. C02
diffuses across the lungs into the bloodstream until
an equilibrium is achieved which reflects mixed
venous C02 levels.
The patient breathed through a mouthpiece with
the nose clipped. Vco2 was measured over 1.5 min.
PETCO~
was measured over approximately 30s. A
rotary valve was then activated allowing the subject
to breathe in and out of a bag containing a COz and
0 2 mixture for 10-15 s. At rest the bag contained
approximately 2 litres of 9% COz (balance 0 2 ) . For
exercise higher levels of COz and greater gas
volumes were required.
The method has previously been validated for use
in chronic heart failure [25]. Such patients however
have abnormal ventilatory responses to exercise [26],
and so we examined the effect of changes in dead
ratio on the computation
space/tidal volume (VD/VT)
of Qt. This calculation assumes a Cvco2 of 62.6
m1/100 ml and a PETCO~)
of 45 mmHg. The calculation was performed by deriving the Pco2of arterial
using the empirical forblood (Paco2) from PETCO~
mula of Jones et al. [27]:
PaCOz = 5.5 +0.90 PETcoz -0.0021 VT
and then correcting Paco2 (Paco2cor) for the alteration of VT in patients with C C F
Paco2cor = Pacoz x [(1 -VD/VTin CCF)/(1- VD/VT
in controls)]
The value for VD/VTwas taken as 0.2 in normal controls and as 0.3, 0.4 and 0.5 in patients with CCF.
The C02 content was derived for Paco2 and Paco2cor using the logarithmic equation from the tabular
data of McHardy [28]:
+2.38
lOgJ2COz = 0.396 loge PCO~
The error was then calculated as: error (%)=
[ l - (C,co, - CcoJCvcoz- Cco2cor)]x 100, for VD/VT
from 0.5 to 21 (Fig. l), and was consistently ~ 5 % .
Measurement of Qt
Qt was measured using a Cardiac Output Module
(P K Morgan) and COZrebreathing technique. The
module is connected to gas analysers for OZ (paramagnetic) and C02 (Engstrom Eliza breathby-breath IR analyser), and measurements are fed
to the computer as described above. Validation of
the system has been described previously [12]. The
principle for measuring Qt is based on the direct
Fick method but uses COz as the indicator rather
than oxygen:
Qt = VCOZ/C~CO~-C~CO~.
These three parameters are estimated non-inva-
Measurement of cardiac ejection fractions
Measurement of left and right ventricular ejection
fractions (LVEF, RVEF) was carried out using a
newly developed digital multiwire camera (Xenos
Medical Systems, Houston, TX, U.S.A.) [29]. This
camera is capable of high maximal count rates (850
kc/s) allowing much higher resolution images to be
obtained than with conventionally used single-crystal
and multi-crystal gamma cameras. It utilizes the isotope Ta178,which is produced from W178 by a generator system contained within the camera [30].
Ta178 has a very short half-life of 9.3 min, which
I. C.Steele et al.
I98
_I Fl
-
ney U-test. Comparism within groups was performed using Wilcoxon’s matched pairs signed rank
test. Where repeated measurements were made with
time, a Friedman two-way analysis of variance by
ranks test was performed. Those results that were
significantly different from the baseline value were
then compared using the Wilcoxon’s matched pairs
signed rank test. A probability value of less than
0.05 was taken as the level of statistical significance.
Results are expressed as median and range. Correlations were calculated between mean Qt measurement, Voz and work rate for each subject in each
group using Pearson’s rank correlation. The relationships between these parameters were compared
for the two subject groups using analysis of covariance.
VdNt 0.5
5
Fig. 1. Effect of increased VDNT ratio on calculated Qt as expressed
as percentage error on the y-axis. The x-axis is VT (ml)
allows scans to be repeated after a short time interval. High-resolution images of both right and left
ventricles are obtained by giving high doses of Ta17*,
but the patient receives a low total radiation dose
because the isotope half-life is short. The usual
radiation dose for each measurement in this study
was 20-40 mCi. The ejection fractions are measured
by a ECG-gated first-pass method, with a total data
acquisition time for RVEF then LVEF of 30s.
Images are acquired at 25 ms intervals. The computer-generated images are displayed on screen and
regions of interest are defined around the left and
right ventricles. The computer then determines
which frames are used to calculate the ejection fractions. LVEF obtained with this system has previously been shown to be similar to that obtained by
contrast ventriculography (r = 0.72, P = 0.005);
RVEF is similar to that obtained using a singlecrystal camera (r = 0.77, P < O . O O l ) [31]. In addition,
the peak filling rate (PFR) in diastole was computed
for both ventricles as a measure of diastolic function.
Blood sampling and assays
The venous cannula was flushed with 0.9% NaCl
(saline) avoiding the use of heparin. Before each set
of samples was taken the dead space volume of the
cannula was discarded. Samples for lactate were precipitated immediately in 8% perchloric acid and the
supernatant was assayed by an enzymic colorimetric
method (Sigma, St Louis, MO, U.S.A.). The coefficient of variation for the assay was 0.8%.
Statistical analysis
Non-parametric tests were used. Differences
between groups were assessed by the Mann-Whit-
RESULTS
All subjects completed the study. One of the control subjects felt faint after the symptom-limited
exercise, but otherwise the exercise tests were completed uneventfully. No subject had ECG evidence
of exercise-induced cardiac ischaemia. Table 1
shows the results from the symptom-limited exercise
test performed on day 2. Exercise time and PVoz
were much lower for the patients than for the control subjects but perceived exertion was similar in
the two groups. The reason for stopping exercise
was similar in the two groups. Heart rate was similar
at rest but lower at peak exercise in the patients
compared with the control subject. Ventilation was
significantly higher at rest in patients but was significantly lower *at peak exercise. Respiratory exchange
ratio (VcoJVo2)was significantly higher at rest but
was similar at peak exercise. Plasma concentrations
Table I . Results of the symptom-limited exercise test. Results are
expressed as median (range). Abbreviations: RER, respiratory exchange
ratio; NS, not significant.
(re4
Weight (kg)
Exercise time (s)
80rg score
Fatigue (n)
(4
Dyspnoea
PVo2(I/min)
Heart rate (beatslmin)
Rest
Peak
VE (I/min)
Rest
Peak
Lactate (mmoVI)
Rest
Peak
RER
Rest
Peak
Patients
Controls
P
66 (37-75)
69 (59-91)
426 (300-574)
3 (3-6)
8
2
1.18 (0.98-1.76)
67 (65-69)
71 (57-86)
634 (502-743)
3 (2-7)
8
2
1.94 (I .53-2.3 I)
NS
NS
81 (56-99)
134 (108-160)
74 (52-87)
151 (122-172)
NS
I I .8 (6.0- 19.6)
49.2 (39.9-77.8)
10.1 (8.8-14.0)
72.2 (58.2-86.9)
0.9 (0.5-1.4)
2.4 (I.6-3.5)
0.8 (0.6- I.I)
5.5 (2.4-6.9)
0.81 (0.78-1.22)
I.I3 (0.93-1 .l9)
0.78 (0.73-0.88)
1.13 (1.04-1.21)
<0.00I
NS
NS
NS
<0.001
<O.Ol
<0.05
<0.05
NS
<0.00I
<0.05
NS
Cardiac output in chronic cardiac failure
of lactate were similar at rest but the control group
showed a significantly greater rise on peak exercise
which persisted through recovery. Table 2 shows the
results of the tests performed on day 3. At rest the
heart rate, VE, Vo, and Qt were similar in the two
groups. The minimum workload the bicycle could be
set for was 25. W, and as a result some of the
patients had a Vo2greater than 30% or peak during
the first steady-state exercise level, even when pedalling rates below 60 rev./min were employed. It may
be that reducing the pedalling rate below 60 rev./
min had little impact on the workload, as the bicycle
is designed to produce a constant resistivity for all
rates above 40 rev./min. For two patients the test
was performed with the bicycle switched off in a further attempt to minimize the workload being performed. This difficulty has resulted in the mean Vo2
for the patients being 45% of their peak. This level
Table 2. Results of day 3 tests at rest and for steadystate exercise.
Restingejection fraction and PFR are shown in italics as they were measured
on day 2 and not day 3. Results are expressed as median (range).
Abbreviation: NS, not significant; LV, left ventricular; RV, right ventricular.
Patients
Heart rate (beatshin)
Rest
Low
Medium
Vo, (mVmin)
Rest
Low
Medium
VE (Ilmin)
Rest
Low
Medium
Qt(Vmin)
Rest
Low
Medium
LVEF (%)
Rest
Low
Medium
RVEF (%)
Rest
Low
Medium
LV PFR
Rest
Low
Medium
RV PFR
Rest
Low
Medium
Lactate (mmoVI)
Low
Medium
Borg score
Low
Medium
Controls
P
82 (57-103)
101 (81-131)
106 (89- 135)
76 (47-90)
87 (69- I 12)
105 (83-1 12)
NS
<0.05
NS
204 (94-289)
555 (333-795)
669 (588-797)
I80 (I 32-226)
624 (519-815)
910 (729-1214)
NS
<0.01
<O.OOl
12.2 (5.4-19.0)
23.6 (I 9.0-34.6)
27.4 (24.8-32.7)
12.4 (7.8- 17.6)
25.4 (I9.0-32.0)
33.2 (26.9-43.3)
NS
NS
= 0.01
3.32 (2. I 1-4.97)
6.59 (3.90-8.14)
6.92 (5.79-9.76)
3.26 (2.49-4.13)
7.3 (5.99-10.38)
9.63 (7.31-13.4)
NS
<0.05
<0.001
19 (5-40)
22 (6-40)
21 (5-39)
55 (48-80)
35 (27-59)
39 (24-54)
30 (22-40)
42 (33-49)
47 (37-60)
45 (39-64)
<0.05
1.1 (0.5-1.8)
I.4 (0.6-2.3)
I.4 (0.4-1.7)
2.1 (1.4-3.2)
2.8 (I.8-3.9)
3.1 (2.3-5.2)
<0.0005
2.0 (1.1-9.5)
2.9 (I.7-9.3)
2.4 (I.5-5.5)
1.8 (1.1-3.4)
2.4 (I .7-3.7)
2.6 (I .6-6.0)
NS
1.4 (0.8-2. I)
I.6 (I -3.3)
1.14 (0.7-1.7)
I .8 (I.O-2.5)
NS
3 (0.5-4)
4 (1-6)
3 (0.5-4)
3 (1-5)
57 (46-77)
58 (49-83)
<0.001
<0.001
<0.001
<0.05
<O.OOl
of exercise will be referred to as the ‘low level’.
During exercise at this low level the Qt was significantly lower in patients than in control subjects, the
heart rate significantly higher, but VE was similar.
For the 50% level the patients were excercised at a
greater workload than for the previous level, even if
it had given a value around 50%. This resulted in
the mean value for the group being 56% of peak.
This exercise will be referred to as the ‘medium
level’. At the medium level of exercise VE and Qt
were significantly lower in the patients but the heart
rate was similar.
The Vo2levels at the low and medium levels were
significantly different between the two groups, as
would have been expected from the different work
rates performed. When the medium exercise level
for the patient group and the low exercise level for
the control group were compared there was no significant difference between them for pedalling rate
[60 (0) versus 57 (2.1) rev./min; mean (SD)] or
workload in W (see Table 3). At this similar level of
work there was no significant difference between
groups for either Vo, or Qt, but heart rate, Borg
score, lactate and VE were all significantly higher in
patients than in normal subjects.
As expected, LVEF was greatly impaired in
patients (Table 2), and also RVEF to a lesser
extent. There was little change with submaximal
exercise. Left ventricular PFR, as a measure of diastolic function, was also reduced in the patient
group, with some impairment of the exercise
response. In contrast, the right ventricular PFR both
at rest and during submaximal exercise was similar
in the two groups.
The relationship between i.0, and Qt at each
stress level is shown in Fig. 2 for the two groups.
There was a significant correlation between these
two parameters for both groups (patients: Y = 0.90,
P <0.0001; controls: Y = 0.96, P <0.0001; Pearson’s
rank correlation). Analysis of covariance was carried
out for this comparison, and showed that the intergroup difference in Qt could be accounted for by the
differences in yo,. The relationship between workload (W) and Vo2is shown in Fig. 3, and the relaTable 3. Comparison of results from medium level exercise in
patients and low level exercise in controls. Results are shown as median
(range). Abbreviations: LV PFR, left ventricular peak filling rate (end-diastolic
volume s-I); RV PFR, right ventricular peak filling rate.
<0.0005
<0.0005
NS
NS
NS
NS
<0.05
I99
Workload (w)
Cycling speed (revhin)
VO, (rnI/min)
Qt (Ilrnin)
VE (I/min)
Heart rate (beatshin)
Borg score
Lactate (mmoVI)
LV PFR
RV PFR
Patients medium level
Controls low level
P
25 (25-40)
60 (60-60)
669 (588-797)
6.92 (5.79-9.76)
27.4 (24.8-32.7)
I06 (89-1 35)
4 (1-6)
I.6 (I .O-3.3)
I.4 (0.4- I.7)
2.4 (I.5-5.5)
U(25-35)
60(40-60)
624 (5 I9-8 I5)
7.3 (5.99-10.38)
25.4 (19.0-32.0)
87 (69- I 12)
3 (0.5-4)
1.1 (0.7-1.7)
2.8 (1.8-3.9)
2.4 (1.7-3.7)
NS
NS
NS
NS
<0.05
<0.005
<0.005
<0.05
<0.0005
NS
I. C.Steele et a].
200
14
T
I
0
l4
12
--
lo
--
A
10
4'
A
8
I2
A
T
I
I
0
0
A
CCFmed
4
0 1
0
Control rest
A
Control low
200
6W
OM)
1200
--
0
0
8
*
0
0
A
.
A
o
o
A
A
A
Control IOW
0
Control med
2 --
l
ID00
800
A
CCFmed
Control med
I
0
A
A
A CCFlow
0
0
0
I
I
1
+
1400
V O (~d m i n )
Fig. 2. Relationship between Qt and V& for patients with CCF and
controls at rest and during steady-state exercise
l4O0
I200
T
i
tionship between W and Qt in Fig. 4. As discussed
previously, the pedalling rate was regarded as not
affecting workload if it was 40 rev./min or more. For
one patient who pedalled at 30 rev./min with a 25 W
workload the value used was 18.75 (i.e. 25 x 30/40).
For the two patients who performed unloaded cycling with the bicycle switched off the workload has
been estimated as 15 W. There was a significant
correlation between W and Vo2 for both groups
(patients: r = 0.60, P = 0.005; controls: r = 0.89,
P <0.0001; Pearson's rank correlation) and this was
also found for the relationship between W and Qt
(patients: r = 0.48, P<0.05; controls: r = 0.79,
P <0.0001; Pearson's rank correlation). For these
relationships analysis of variance showed that the
inter-group differences in Vo2and Qt could both be
accounted for by differences in work rate.
0
0
0
0
0
0
OM)
I
PA
A
DISCUSSION
CCFmed
A Control low
0
0
10
20
30
40
50
Control med
60
'10
Workload (watts)
Fig. 3. Relationship between workload (W) and V& for patients with
CCF and controls during steady-state exercise
80
This study shows that during submaximal exercise
below the anaerobic threshold patients with CCF
develop a Qt and vo2 appropriate to the work rate
being performed. Non-invasive techniques were
used to avoid any interference with cardiac function
that may occur as a result of invasive procedures
[16] and to avoid the potential hazards of central
venous cannulation [32]. Most non-invasive techniques, such as echocardiography, Doppler flow or
bioimpedance, are difficult to perform during exer-
Cardiac output in chronic cardiac failure
20 I
cise, whereas COz rebreathing is more reliable
patients and the ‘low’ workload in the controls was
during exercise than at rest [12]. However, there are
the same. Voz and Qt were the same in the two
limitations with this technique. First, the COz
groups for this similar absolute workload but
rebreathing method cannot be used to measure Qt
patients had a higher heart rate, lactate level and
Borg score.
at peak exercise, as it would not be possible to mainPrevious studies of ventricular function in CCF
tain peak Voz for the length of time required to
have concentrated on resting LVEF, which correcomplete the measurements (typically about 2 min).
lates poorly with symptoms [44]. Resting RVEF may
Secondly, submaximal exercise levels where the Voz
correlate better with exercise tolerance than LVEF
is less than 60% of peak should be chosen due to
[45], but not all studies have confirmed this finding
the increased COz production which occurs above
[46, 471. The advanced multiwire camera using gated
the anaerobic threshold. Above this level the COz
first-pass data acquisition used in our study allowed
dissociation curve is shifted downwards. Finally, the
to look at both LVEF and RVEF during exercise.
use of PETCO~
to estimate CaCOz also requires further
Patients had very depressed LVEF at rest, with no
consideration. It is known that in normal subjects
significant change during exercise. There was, in
during exercise at the level of the anaerobic threshaddition, evidence of diastolic dysfunction in
old the PETCO;!
is approximately 3 mmHg greater
patients, with reduced left ventricular PFR both at
than the arterial Caco2[33]. In patients with CCF
rest and during the submaximal work loads. The
during exercise the PETCO~
is typically 6-8 mmHg
response of RVEF to exercise in patients with CCF
less than it would be in normal controls [26]. It is
has not been reported before. RVEF was only
not clear if this reduction represents a true differmoderately impaired at rest, but tended to fall with
ence in CaCOz or if it is a result of the pulmonary
exercise. Right ventricular PFR was similar to conabnormalities that occur in cardiac failure [34]. If
trol subjects, indicating that right ventricular functhe equation used to derive the Pacoz from the
tion was relatively preserved in the patient group.
PETCO~
[35] is corrected assuming values of VD/VT
It has been shown that therapeutic interventions
greater than the 0.2 typically found in normal subthat
improve central haemodynamics do not proQt
1)
the
error
in
the
value
of
jects (see Fig.
duce immediate changes in exercise capacity or in
obtained remains small (<5%). It is therefore
unlikely that any differences in the PETCOJC~CO~Vo2or lactate production, even if they increase limb
blood flow as well as Qt [8, 48, 491. After cardiac
relationship in patients with CCF would affect the
transplantation, despite the virtually immediate
conclusion that Qt at submaximal workloads is the
restoration of resting Qt, the expected increase in
same in patients and controls.
exercise capacity is delayed over a number of
The reason some previous studies have shown a
months [50]. Exercise programmes can produce
difference in Vo, in patients [36-401 may be that
improvements in peak Vo2 in patients with CCF
there was insufficient time for steady state to be
compared with non-exercising control groups [51].
achieved, as it appears there is a delay in kinetics
This improvement in exercise performance has been
due to a slowed circulatory response to exercise in
shown to be associated with a reduction in the venpatients with CCF [9, 411. Roubin et al. [42] showed
tilatory abnormalities in CCF [52], an increase in
no significant difference in Vozbetween patients and
blood flow to exercising skeletal muscle, a more efficontrols performing exercise at similar absolute
cient peripheral oxygen extraction at maximal exerwork rates, but found that the Qt was significantly
cise and a decrease in lactate accumulation during
lower in the patients. The investigators used an
sub-maximal exercise [53].
incremental exercise protocol with 3 min stages,
In the present study whole-body oxygen consumprather than a steady-state test. If the time at each
tion has been measured, but it is known that skeletal
exercise level had been extended the patients may
muscle oxygen uptake during exercise constitutes
have achieved the same steady-state Qt as the con8 0 4 5 % of whole-body uptake [54]. Previous work
trol group, on the basis of the delayed circulatory
examining the metabolism of skeletal muscles using
response [9].
31P-NMR has shown that there is excessive acidosis
There has been much debate as to whether comand an elevated phosphate/phosphocreatine ratio
parison of the submaximal exercise responses of
during exercise in CCF [55, 561. These changes are
patients with CCF with controls should be at similar
present at similar relative workloads and at similar
absolute or similar relative workloads [43]. This
levels of muscle blood flow to control subjects [57].
study has examined both, but it is difficult to comExercise under ischaemic conditions [58] produces
pare two such widely disparate groups at the same
similar abnormalities, and these findings point
absolute level. A very light workload for normal subtowards an abnormality in skeletal muscle metabojects may be quite demanding for patients and be
lism. Recent work has suggested that this may
above the anaerobic threshold, and hence unsuitable
involve small myelinated and unmyelinated nerve
for the measurement of Qt by the COz rebreathing
fibres which arise from work-sensitive receptors in
method. Equally, the exercise capacity of some
skeletal muscle [59]. These mediate an ergoreflex
patients was. so low that even unloaded pedalling
effect constituting an increase in sympathetic outproduced a Voz 45% of peak. However, the average
flow, which results in vasoconstriction in distant
exercise workload at the ‘medium’ workload in the
202
I. C.Steele et al.
vascular beds and possibly a small increase in heart
rate [60]. These ergoreceptors are active in patients
with cardiac failure and they may link abnormal
skeletal muscle function to the fatigue, dyspnoea,
hyperpnoea and sympathoexcitation of CCF [61].
In conclusion, patients with CCF have decreased
exercise tolerance but at a similar absolute level of
exercise the Vo2and Qt are the same as for normal
controls. The heart rate is greater in patients, enabling patients to maintain Qt in the face of a reduced
stroke volume by compensatory tachycardia. Despite
the normal Qt in patients with CCF at this submaximal exercise level, lactate, respiratory exchange
ratio and perceived exertion were all greater in
patients than in controls. This suggests that peripheral factors are important in the abnormal exercise responses of patients with CCF. Abnormalities
in skeletal muscle metabolism are likely to be linked
to central abnormalities in the failing heart [62] via
chronic hypoperfusion, deconditioning and malnutrition.
ACKNOWLEDGMENTS
We are grateful to Sister E. Crawford for her help
with patient care, and to Dr C. Patterson for advice
with statistics. This study was supported by the
Royal Victoria Hospital through the award of
research fellowships to I.C.S. and A.M.
REFERENCES
I. Wood P. Diseases of the heart and circulation. London: Eyre and Spottiswoode,
1969; 291.
2. Donald KW, Bishop JM, Wade OL. A study of minute to minute changes of
arteriovenous oxygen content difference, oxygen uptake and cardiac output and
rate of achievement of a steady state during exercise in rheumatic heart disease.
J, Clin Invest 1954; 33: 1146-7.
3. Epstein SE, Beiser GD, Stampfer M, Robinson BF. Braunwald E. Characterization
of the circulatory response to maximal upright exercise in normal subjects and
patients with heart disease. Circulation 1967; 3 5 1049-62.
4. Franciosa JA,Ziesche S, Wilen M. Functional capacity of patients with chronic left
ventricular failure. Am J Med 1979; 67: 460-6.
5. Rubin SA, Siemienauk D, Nathan MD, Prauss J, Swan HJC. Accuracy of cardiac
output, oxygen uptake, and arteriovenousoxygen difference at rest, during
exercise, and after vasodilator therapy in patients with severe chronic heart
failure. Am J Cardiol 1982; 5 0 973-8.
6. Weber KT, Kinasewitz GT, JanickiJS, Fishman AP. Oxygen utilization and
ventilation during exercise in patients with chronic cardiac failure. Circulation
1982; 65: 1213-23.
7. Kramer BL, Massie BM, Topic N. Controlled trial of captopril in chronic heart
failure: a rest and exercise hemodynamic study. Circulation 1983; 6E 807-1 6.
8. Wilson JR, Martin JL, Ferraro RN. Impaired skeletal muscle nutritive flow during
exercise in patients with congestive heart failure: role of cardiac pump
dysfunction as determined by the effect of dobutamine. Am J Cardiol 1984; 53:
1308-15.
9. Riley M, P6rszkz J, Stanford CF, Nicholls DP. Gas exchange responses to
constant work rate exercise in chronic cardiac failure. Br Heart J 1994; 7 2
150-5.
10. Riley M, Standford CF, Nicholls DP. Ventilatory and heart rate responses after
exercise in chronic cardiac failure. Clin Sci 1994; 87: 23 1-8.
I I. McDonough JR, Danielson RA, Wills RE, Vine DL. Maximal cardiac output during
exercise in patients with coronary artery disease. Am J Cardiol 1974; 3 3 23-9.
12. Nugent AM, McParland J, McEneaney DJ, Steele I, Campbell NPS, Stanford CF,
Nicholls DP. Non-invasive measurement of cardiac output by a carbon dioxide
rebreathing method at rest and during exercise. Eur Heart J 1994; I 5 36 1-8.
13. Franciosa JA,Leddy CL, Wilen M, S c h w m DE. Relation between hemodynamic
and ventilatory responses in determining exercise capacity in severe congestive
heart failure. Am J Cardiol 1984; 5 3 127-34.
14. Koike A, Hiroe M, Adachi H, et al. Cardiac output-02 uptake relation during
incremental exercise in patients with previous myocardial infarction. Circulation
1992;85 1713-9.
15. Koike A, Hiroe M, Adachi H. et al. Oxygen uptake kinetics are determined by
cardiac function at onset of exercise rather than peak exercise in patients with
prior myocardial infarction. Circulation 1994; 9 0 2324-32.
16. Packer M, Medina N, Yushak M. Haemodynamic changes mimicking a vasodilator
drug response in the absence of drug therapy after right heart catheterisation in
patients with chronic heart failure. Circulation 1985; 71: 761-6.
17. Cowley AJ, Fullwood LJ, Muller AF, Stainer K, Skene AM, Hampton JR Exercise
capability in heart failure: is cardiac output important after all?Lancet 1991; 337:
771-3.
18. Brown JM, White CJ, Sobol SM, Lull 4.Increased left ventricular ejection fraction
after a meal: potential source of error in performance of radionuclide
angiography. Am J Cardiol 1983; 5 I: I 709- I I.
19. Muller AF, Hawkins M, Batin P, Evans S, Cowley A]. Food in chronic heart
failure: improvement in central haemodynamics but deleterious effects on
exercise tolerance. Eur Heart J 1992; 1 3 1460-7.
20. Elborn IS, Stanford CF, Nicholls DP. Reproducibilityof cardiopulmonary
parameters during exercise in patients with chronic cardiac failure. The need for
a preliminary test. Eur Heart J 1990 II: 75-81.
21. Riley M, Northridge DB, Henderson E, Stanford CF, Nicholls DP, Dargie HJ. The
use of an exponential protocol for bicycle and treadmill exercise testing in
patients with chronic cardiac failure. Eur Heart J I992 13: 1363-7.
22. Borg GAV. Psychophysical bases for perceived exertion. Med Sci Sports Exer
1982; 1 4 377-81.
23. Riley M, Elborn IS,McKane WR, Bell N, Stanford CF, Nicholls DP. Resting
energy expenditure in chronic cardiac failure. Clin Sci 1991; 8 0 633-9.
24. Collier CR Determination of mixed venous CO2 tensions by rebreathing. J. Appl
Physiol 1956; 9 25-9.
25. Franciosa ]A, Ragan DO, Rubenstone SJ. Validation of the CO1 rebreathing
method for measuring cardiac output in patients with hypertensionor heart
failure. J Lab Clin Med 1976; 8 8 672-82.
26. Wasserman K, B a n g Y-Y, Riley MS. Ventilation during exercise in chronic heart
failure. Basic Res Cardiol 1996; 91 (Suppl. I): I- I I.
27. Jones NL, Robertson DG, Kane JW. Difference between end-tidal and arterial
PCO, on exercise. J Appl Physiol 1979; 47: 954-60.
28. McHardy GJR The relationship between the difference in pressure and content
of carbon dioxide in arterial and venous blood. Clin Sci 1967; 3 2 299-309.
29. Lacy JL, LeBlanc AD, Babich JW,et at. A gamma camera for medical applications
using a proportionalcounter. J Nuclear Med 1984; 2 5 1003-12.
30. lacy JL, Ball ME, Verani MS, et al. An improved tungsten-178/tantalum-178
generator system for high volume clinical applications. J Nuclear Med 1988 2 9
1526-38.
3 I. Vallely SR, Campbell NPS, Laird JD, Ferguson R, Anderson C. Radionuclide
angiography using a multiwire camera and tantalum-I 78 (la-178) - an advance
over conventional imagingtechniques [abstract]. Eur Heart J 1992; I 3 (Suppl.):
144.
32. Anon. Complicationsof pulmonary artery balloon flotation catheters. Lancet
1983; i:37-8.
33. Jones NL, McHardy GJR, Naimark A, Campbell EJM. Physiological dead space
and alveolar-arterial gas pressure differences during exercise. Clin Sci 1966; 3 I:
19-29.
34. Rubin SA, Brown HV. Ventilation and gas exchange during exercise in severe
chronic heart failure. Am Rev Respir Dis 1984; 129 (Suppl.): S63-4.
35. JonesNL, Robertson DG, Kane JW, Campbell EJM. Effect of Pco, level on
alveolar-arterial PCO, difference during rebreathing. J Appl Physiol 1972; 3 2
782-7.
36. Roberts JM, Sullivan M, Froelicher M,Genter F. Myers J.Predictingoxygen
consumption from treadmill testing in normal subjects and coronary artery
disease subjects. Am Heart J 1984; 108 1454-60.
37. Lipkin DP, Perrins J, Poole-Wilson PA. Respiratory gas exchange in the
assessment of patients with impaired ventricular function. Br Heart J 1985; 5 4
321-8.
38. Cowley A], Stainer K, Rowley JM, Hampton JRAbnormalities of the peripheral
circulation and respiratory function in patients with severe heart failure. Br
HeartJ 1986; 5 5 75-80.
Cardiac output in chronic cardiac failure
39. Cohen-Solal A, ChabernaudJM, Gourgon G. Comparison of oxygen uptake
during bicycle exercise in patients with chronic heart failure and in normal
subjects. J Am Coll Cardiol. I990 I 6 80-5.
40. Meyer K, Westbrook S,Schwaibold M, Hajric R, Lehmann M, Roskamm H.
Cardiopulmonarydeterminants of functional capacity in patients with chronic
heart failure compared with normals. Clin Cardiol 1996; 1 9 944-8.
41. Sullivan MJ, Green HI, Cobb FR Altered skeletal muscle metabolic response to
exercise in chronic heart failure: relation to skeletal muscle aerobic enzyme
activity. Circulation I991; 8 4 1597-607.
42. Roubin GS. Anderson SA, Shen WF, et al. Hemodynamic and metabolic basis of
impaired exercise tolerance in patients with severe left ventricular dysfunction.
J Am Coll Cardiol 1990; 15 986-94.
43. Francis GS, Goldsmith SR, Ziesche S, Nakajima H, Cohn IN. Relative attenuation
of sympathetic drive during exercise in patients with congestive heart failure.
J Am Coll Cardiol 1985; 5: 832-9.
44. Franciosa ]A, Park M, Levine TB. Lack of correlation between exercise capacity
and indexes of resting left ventricular performance in heart failure. Am J Cardiol
1981; 47: 33-9.
45. Baker BJ, Wilen WM, Boyd CM, Dinh HA, Franciosa JA. Relation of right
ventricular ejection fraction to exercise capacity in chronic left ventricular failure.
Am J Cardiol 1984; 5 4 596-9.
46. Braun S, Deedwania P, Massie B for the captopril-digoxin Study Group.
Interrelationof right and left ventricular EF and exercise capacity in CHF.
Circulation 1987; 76 (Suppl. IV): 357.
47. Clark AL, Swan JW,Laney R, Connelly M, Somerville J, Coats A$. The role of
right and left ventricular function in the ventilatory response to exercise in
chronic heart failure. Circulation 1994; 8 9 2062-9.
48. Massie BM, h e r B, Haughom F. Acute and long-term effects of vasodilators
on resting and exercise hemodynamics and exercise tolerance. Circulation 1981;
6 4 I 2 18-26.
49. Mancini DM, Schwartz M, Ferraro N, Seestedt R, Chance B, Wilson JR Effect of
dobutamine on skeletal muscle metabolism in patients with congestive heart
failure. Am J Cardiol 1990 6 5 I 121-6.
50. Astrom H. Exercise testing in patients with heart disease. Acta Pharmacol
Toxicol (Copenhagen) 1986; 59 (Suppl.) 6 65-70.
203
5 I. Jette M, Heller R, Landry F, Blumchen G. Randomised 4-week exercise program
in patients with impaired left ventricular function. Circulation I99 I; 8 4 I561-7.
52. Davey P, Meyer T, Coats A, Adamopoulos S, Casadei B, Conway J, Sleight P.
Ventilation in chronic heart failure: effects of physical training. Br Heart J 1992;
6 8 473-7.
53. Sullivan MJ, Higginbotham MB, Cobb FR Exercise training in patients with severe
left ventricular dysfunction: Hemodynamic and metabolic effects. Circulation
1988; 78: 506-15.
54. Atrand P-0, Rodahl K. Textbook of work physiology. 3rd edn. New York
McGraw-Hill, 1986.
55. Amolda L, Conway M, Dolecki M, et al. Skeletal muscle metabolism in heart
failure: a )'P nuclear magnetic resonance spectroscopy study of leg muscle. Clin
Sci I990 7 9 593-9.
56. Wilson JR, Fink L, Maris J, Ferraro N, Power-VanwartJ, Eleff S, Chance B.
Evaluation of energy metabolism in skeletal muscle of patients with heart failure
with gated phosphorus-3I nuclear magnetic resonance. Circulation 1985; 71:
57-62.
57. Massie B, Conway M, Yonge R et al. Skeletal muscle metabolism in patients with
congestive heart failure: relation to clinical severity and blood flow. Circulation
1987; 7 6 1009- 19.
58. Massie BM, Conway M, Rajagopalan B, et al. Skeletal muscle metabolism during
exercise under ischaemic conditions in congestive heart failure. Evidence for
abnormalities unrelated to blood now. Circulation 1988; 7 8 320-6.
59. hang C-S. Hood JrWB. Afferent neural pathway in the regulation of
cardiopulmonary responses to tissue hypermetabolism. Circ Res 1976; 3 8
209- 14.
60. Piepoli M, Coats AJS. Skeletal muscle ergoreflex: a possible role in the ventilatory
control. Circulation 1992; 86 (Suppl. I): 1-368.
61. Coats AJS, Clark AL, Piepoli M. Voltenani M, Poole-Wilson PA. Symptoms and
quality of life in heart failure: the muscle hypothesis. Br Heart J 1994; 72 (Suppl.):
S36-9.
62. Wasserman K. Reduced aerobic enzyme activity in skeletal muscles of patients
with heart failure. A primary defect or a result of limited cardiac output?
Circulation 199 I; 8 4 1868-70.