Vol. 88, Issue 4, 1228-1238, April 2000
O2 kinetics reveal a
central limitation at the onset of subthreshold exercise in heart
transplant recipients
Bertrand
Mettauer1,
Quan Ming
Zhao1,
Eric
Epailly1,
Anne
Charloux1,
Eliane
Lampert1,
Bernadette
Heitz-Naegelen1,
François
Piquard1,
Pietro E.
di
Prampero2, and
Jean
Lonsdorfer1
1 Département de Physiologie, Jeune
Equipe 2105 Centre National de la Recherche Scientifique, Services
des Explorations du Système Circulatoire et des Explorations
Fonctionnelles Respiratoires, Hôpital Central, F-67091
Strasbourg Cedex, France; and 2 Dipartimento di
Scienze e Tecnologie Biomediche, Universita di Udine, I-33100 Udine,
Italy
 |
ABSTRACT |
Because the cardiocirculatory response of
heart transplant recipients (HTR) to exercise is delayed, we
hypothesized that their O2 uptake
(
O2) kinetics at the onset
of subthreshold exercise are slowed because of an impaired early
"cardiodynamic" phase 1, rather than an abnormal
subsequent "metabolic" phase 2. Thus we
compared the O2 kinetics in
10 HTR submitted to six identical 10-min square-wave exercises set at
75% (36 ± 5 W) of the load at their ventilatory threshold (VT) to
those of 10 controls (C) similarly exercising at the same absolute (40 W; C40W group) and relative load (67 ± 14 W; C67W group).
Time-averaged heart rate, breath-by-breath
O2, and O2
pulse (O2p) data yielded monoexponential time constants of
the O2 (s) and
O2p increase. Separating phase 1 and 2 data
permitted assessment of the phase 1 duration and phase
2 O2 time
constant
(
). The O2 time constant was
higher in HTR (38.4 ± 7.5) than in C40W (22.9 ± 9.6; P 
0.002) or C67W (30.8 ± 8.2; P 0.05), as was the
O2p time constant, resulting from a lower phase 1 O2 increase (287 ± 59 vs.
349 ± 66 ml/min; P 0.05), O2p increase (2.8 ± 0.6 vs. 3.6 ± 1.0 ml/beat; P 0.0001), and a longer
phase 1 duration (36.7 ± 12.3 vs. 26.8 ± 6.0 s; P 0.05), whereas the
was similar in HTR and C (31.4 ± 9.6 vs. 29.9 ± 5.6 s; P = 0.85). Thus the HTR have slower subthreshold
O2 kinetics due to an
abnormal phase 1, suggesting that the heart is unable to
increase its output abruptly when exercise begins. We expected a faster
in HTR because of their prolonged phase 1 duration. Because
this was not the case, their muscular metabolism may also be impaired at the onset of subthreshold exercise.
heart transplantation; pulmonary gas exchange; oxygen consumption; heart rate
| |
INTRODUCTION |
AS HEART TRANSPLANTATION HAS evolved to an established
procedure for the management of end-stage heart failure with extended survival rates, physicians have become more concerned about the patients' quality of life and, therefore, exercise capacity. The latter remains impaired after heart transplantation to ~60% of that
of age-matched sedentary controls. The mechanisms of this exercise
limitation are only partially elucidated, with some central components
due to the transplanted heart's denervation and diastolic impairment
(5, 18) and with some peripheral alterations thought to be due to
deconditioning, the effects of immunosuppressive therapy (8, 19, 23),
and heart failure-induced dysfunctions that would have persisted after
transplantation (33).
Among the different mechanisms that may be involved, the role of
delayed O2 uptake
(O2) transients has been
suggested (8, 9, 15, 26, 27). The
O2 kinetics at the onset of
work depend on numerous factors: phase 1 during the first
15-30 s of work is assumed to result mainly from an abrupt
increase in pulmonary blood flow, whereas the following phase 2 depends on the increase in muscular oxidative metabolism (1) and,
therefore, is influenced by the type of exercise, age, and fitness (11,
39). The gas exchange kinetics are also influenced by the energetic
requirements of work: below the anaerobic threshold (AT) the
O2 kinetics reflect the onset
of intracellular metabolism and the mechanisms of O2
transfer but remain only slightly affected by the exercise intensity
(32, 36). Above the AT, these kinetics slow down with the work level
(32), and their behavior is complicated by the superimposition of a
slow component (12, 36, 39), which increases
O2 above the O2
requirements of subthreshold work, probably because of the recruitment
of type II glycolytic fibers (12). Accordingly, the
O2 kinetics follow a
first-order transfer function only for exercises below the AT (12, 36, 39). Therefore, complexities associated with the
O2 slow-component behavior
may confound accurate calculation and interpretation of the
O2 deficit, as well as the utilization of simple
mathematical models to characterize the
O2 kinetics (12, 36, 39). Yet most previous studies assessing
O2 kinetics in
heart-transplanted patients [heart transplant recipients
(HTR)] were performed at or above the AT (8, 9, 26).
Paterson et al. (27) reported
O2 kinetics below the AT in a
limited number of HTR. By the observation of faster
O2 kinetics during a second
square-wave forcing, when the patients' heart rate (HR) was higher,
they suggested that the O2
transients depend on O2 delivery kinetics in HTR.
Nevertheless this study concerned patients early (2.3 ± 0.2 mo) after
surgery, at a time when the chronotropic response and exercise capacity
of the HTR are still expected to improve greatly (24). The conclusions of Paterson et al. (27) have been recently questioned by Grassi et al.
(15), who studied a larger number of patients at different delays after
transplantation and concluded that the slower
O2 kinetics of HTR depend on
their impaired muscular oxidative capacity and not on a delayed
increase in blood flow. Therefore, the issue of the control mechanisms
of O2 kinetics in HTR is not
yet settled. To our knowledge, the respective contributions of
phases 1 and 2 to the rest-to-work
O2 transition have never been
assessed in HTR.
Therefore, the aim of the present study was to compare the
O2 kinetics at the onset of
subthreshold exercise in long-term stabilized HTR to that of matched
sedentary normal controls (C), to examine the respective roles of
phases 1 and 2 in these kinetics, and to determine whether the
delayed O2 kinetics depend on
graft and/or muscular dysfunctions in HTR. We hypothesized that, if delayed O2 kinetics persist
for subthreshold exercises late after transplantation, they may also be
due to an abnormal "cardiodynamic" phase 1 response and
not only to a sluggish "metabolic" phase 2 response.
Moreover, we attempted to correlate the phase 1 O2 parameters with cardiac
variables (i.e., indexes of diastolic function and chronotropic
response), which, if abnormally delayed, may prevent the cardiac output
from abruptly increasing at the onset of exercise in HTR.
| |
METHODS |
Subjects.
Ten male HTR (age 43.3 ± 7.6 yr, weight 77.9 ± 11.4 kg), considered
as rehabilitated but not enrolled in a formal retraining program or
sporting activities, agreed to participate in the study. It was >6 mo
after surgery for all of them (delay since surgery: 32.2 ± 27.1 mo,
range: 6.1-87.5 mo). Heart transplantation was performed for ischemic heart disease in three, dilated hypokinetic cardiomyopathy in five, and valvular heart disease in two subjects. They were all under triple drug immunosuppressive therapy with the
cyclosporine dose adapted to ensure whole blood through levels of
150-200 ng/ml. None suffered peripheral vascular disease, and no
negative chronotropic drugs were administered. All patients were free
of significant graft rejection or accelerated atherosclerosis as
disclosed by a recent endomyocardial biopsy and their yearly coronary
angiograms. All patients gave informed consent to the study, which had
been approved by the local institutional ethics committee. Ten healthy
sedentary male subjects (age 44.0 ± 5.2 yr, weight 74.6 ± 11.3 kg)
volunteered to serve as C. This C group was similar to the group of
patients in terms of age, weight, and fitness level. They all followed
the same experimental protocol as the patients. None was taking any
medication, and, although they were all professionally and
recreationally active, none participated in any regular sporting activity.
Exercise tests.
To assess the peak exercise capacity of the patients and C and the
position of their ventilatory threshold (VT) taken as an approach
of their aerobic-anaerobic transition, both groups were submitted
to an incremental, symptom-limited, maximal exercise test while
measurements of the gas exchanges were taken. During this test, the
work rate was increased by 20-W, 2-min steps up to the point of
exhaustion. Thereafter, the
O2 kinetics were assessed
in each subject by means of six consecutive, identical constant-rate
exercise tests. To assess the
O2 transients at a level of
exercise at which they follow first-order kinetics, the work rate of
the constant-rate exercises was set at 75% of the work rate of each
patient's individual VT. In the C, two series of six consecutive
constant-rate exercise tests were performed: the first series was
realized at 75% of the work rate of each subject's individual VT, as
in the patients, and the second series was realized at a work rate (40 W) that was the closest possible to the mean of the absolute work rates
performed by the HTR. Thus we were able to compare HTR and C at same
absolute and relative levels of exercise. In HTR and C, the 10-min
constant-rate exercise began abruptly after 4 min of rest sitting on
the cycle-ergometer, after a brief vocal signal given without warning
by the operator. Each series of six subsequent constant-rate exercises
was performed on 3 test days, each separated by a maximum of 2 days;
the two tests performed the same day were done at least 1 h apart to
ensure complete recovery. Before the beginning of exercise, the cycle ergometer (CardiO2 cycle, Medical Graphics, St. Paul, MN) was programmed to electrically drive the flywheel at 60 rotations/min, so
as to obviate the need to overcome the flywheel inertia.
Measurements.
The minute ventilation (E) and gas
exchange parameters were measured on a breath-by-breath basis by means
of an open-circuit metabolic cart with rapid O2 and
CO2 analyzers (CardiO2 apparatus, Medical Graphics) during
both the incremental and constant-rate tests. Before each individual
exercise test, the pneumotachograph was calibrated with several stokes
given by a 3-liter calibration syringe (Hans Rudolph), and the gas
analyzers were calibrated by means of reference gases of known
O2 and CO2 concentrations. During the
incremental test, the breath-by-breath data were smoothed by a
six-point moving averaging algorithm. The breath-by-breath values,
acquired during the six successive, identical constant-rate exercises
performed by each subject, were not smoothed but were time aligned to
the beginning of exercise, interpolated, and time averaged according to
the technique described by Whipp et al. (37).
By multiple exercise repetitions, with subsequent time alignment and
averaging, the noise was divided by the square root of the number of
repetitions (22, 37). Accordingly, as the noise was reduced by a factor
of 2.45 in our study, we expected to separate adequately the phase
1 and phase 2 data (37). Because our metabolic cart's
software calculates the O2
and the CO2 production
(CO2) by standard formulas
based on measurements of the expiratory gas volume only (3), and
without possibilities to measure expired nitrogen, we were not able to
correct for the breath-by-breath changes in pulmonary gas stores
because of changes in the functional residual capacity. Therefore, we
assessed the breath-by-breath gas exchanges at the mouth and not true
alveolar exchanges (2). The HR was continuously recorded by a
cardiotachometer included in the metabolic cart.
As part of the patients' routine follow-up, echocardiographic and
pulsed Doppler examinations were performed at rest on a monthly basis
to record parameters of systolic left ventricular function (ejection
fraction) and of diastolic left ventricular function [isovolumic
relaxation time (IVRT); transmitral pressure decay half time] as
potential indexes of graft rejection (34); these two latter echographic
indexes can also be taken to represent, respectively, the amount of the
patients' left ventricular relaxation and diastolic compliance
impairments, which have been shown to exist in HTR (18, 28).
Considering that the main factors of the phase 1 response are
hemodynamic, we attempted to correlate the patients' phase 1 amplitude and duration with the previously defined echographic indexes,
with the half time of the chronotropic response, with the parameters of
maximal exercise capacity, and with the lapse of time between the
operation and the exercise tests.
Measurements during the incremental exercise.
The rest and peak exercise
O2,
CO2, and
E were measured by standard,
open-circuit ergometric techniques (3). The rest values were 1-min
averages of the breath-by-breath values, after stabilization of the
O2,
CO2,
E, and respiratory exchange ratio (RER = CO2/O2).
The peak values were the averages over the last 30 s of the incremental
exercise, with both patients and C being encouraged to push the
exercise to the point of exhaustion. The VT was assumed to occur when
CO2 related to
O2 changed slope (V-slope method).
Measurements during the constant-rate exercises.
O2, O2 pulse
(O2p = O2/HR),
CO2,
E, and HR were continuously measured
during the 4 min of rest, the 10 min of exercise, and the first 5 min
of recovery, but only the
O2, O2p, and HR were analyzed. As representing the rest values, we averaged the measured values over 1 min, with the subjects quietly sitting on the
ergometer. All the subjects reached a true steady state during the
constant-rate exercises (as the data of the last 3 min of exercise were
no longer correlated with time). For the steady-state values, we also
averaged the measured data over the last minute of exercise.
The kinetics of the parameters during the overall rest-to-exercise
transition were assessed by fitting all of the 10-min time-averaged, breath-by-breath data to a monoexponential model forced to start at the
beginning of exercise without a time delay, according to the method
described as "model 1" in the work by Whipp et al. (37).
Applying the formula 
Y(t) = Y(steady
state)[1 
e
(t/
)],
where Y may be O2,
CO2, or
E; Y(t) is the
increase in Y above the prior steady-state value at time
t; and Y(steady state) is the steady-state increase
in Y, yielded the time constant 
for the overall
O2 and O2p
transition, which characterizes the rest-to-exercise transition
regardless of its phases.
Because the slowed HR response to the exercise of HTR has been reported
to increase linearly for some authors (8, 9) and exponentially by
others (29), we characterized our patients' HR response by the half
time of the HR increase, defined as the time taken from the beginning
of exercise for the HR to reach one-half of its increase observed after
10 min of exercise. Because the
O2 increase during the
transition is assumed to follow first-order kinetics, the
O2 deficit was calculated as equal to the for the
overall O2 transition, as
previously defined, multiplied by the increment in
O2 during the square-wave
exercise (O2 at steady
state O2 at rest)
(1, 11, 36).
The end of phase 1 was assumed to occur at the time when a
decrease in end-tidal PO2 with a
simultaneous increase in end-tidal
PCO2 appeared, as well as when a
sharp decrease in the RER occurred (32, 36, 37). As a rule, this method of separating phase 1 and phase 2 corresponds also to
the end of the small initial plateau of the
O2 (36, 37). To illustrate the components of phase 1 and the phase 1-phase 2 discrimination method, an example of the initial behavior of the RER,
the end-tidal PO2,
PCO2, and its correspondence to the
initial O2 at the
onset of exercise of two representative patients, one with a short and
one with a long phase 1, is presented in Fig.
1. The duration of phase 1 was
determined as the time between the start of exercise and the phase
1-phase 2 transition, assessed with the preceding criteria
by agreement between two observers (Q. Zhao, E. Epailly) unaware of the
other's results. The phase 1 amplitude for
O2, O2p, and HR
was calculated as the average of all the values throughout phase
1 as in the work by Sietsema et al. (32). The exercise increase in
O2, O2p, and HR
was calculated for phase 1 or the steady state as the
corresponding O2,
O2p, or HR values minus the corresponding average rest
values. The phase 2 O2 was calculated
after fitting the time-averaged breath-by-breath
O2 data to the same
monoexponential formula as described previously, but with the fitting
time t starting at the phase 1-phase 2 transition point. This corresponds to "model 3" in the
work by Whipp et al. (37).
View larger version (34K):
[in this window]
[in a new window]
|
Fig. 1.
Examples of breath-by-breath expired gas analysis data in 2 representative heart transplant patients. Shown are end-tidal
O2 partial pressure
(PETO2;  ),
end-tidal CO2 partial
pressure (PETCO2;  ),
respiratory exchange ratio (RER;  ), and O2 uptake
(O2;  ). These
curves represent time-averaged data of first 180 s of 6 identical
constant-rate exercises in each patient. Simultaneous changes in
PETO2,
PETCO2, RER, and
O2 that characterize
phase 1-to-phase 2 transition are shown in 1 patient with a short (26.7 s) and in 1 with a long (41.0 s) phase
1 duration. Rest value (or x-axis = 0) represents average
of data over last minute of rest.
|
|
Statistical analysis.
Values are expressed as means ± SD. The monoexponential fitting of
the O2, HR, and
O2p data was performed by multiple iterations and the least
squares technique by using commercially available software (Prism
Graphpad software, San Diego, CA). Owing to the small populations
studied, nonparametric tests were used to perform comparisons. The
incremental exercise and kinetic measurements performed in HTR were
compared with those of the C by means of the Mann-Whitney U
test for unpaired values; the steady-state exercise values were
compared with the rest values within each group by means of the
Mann-Whitney U test for paired values; and the rest and
steady-state exercise values were compared between the HTR and the C
groups, working at similar absolute and relative loads to the HTR, by
means of the Kruskal-Wallis test followed by Dunn's procedure. A
P 0.05 was taken to represent a significant difference.
| |
RESULTS |
Incremental exercise.
During the 20-W, 2-min incremental maximal test, the HTR reached 126 ± 21 W, whereas the C reached 206 ± 54 W (P 0.0004). The
rest, the peak values, and values recorded at the VT are presented in
Table 1. The resting values are similar in
HTR and C, except that the HR is higher in HTR, as expected. At the VT,
the work rate and, therefore, the gas exchange values are lower in HTR, but the RER is similar [P = 0.59, not significant
(NS)], the E is only
insignificantly lower (P = 0.12, NS), and the HR is
insignificantly higher (P = 0.27, NS). The VT is reached for a
similar percentage of the peak
O2 in both groups: 55 ± 11% in HTR and 51 ± 6% in C (P = 0.85, NS). As expected,
the peak O2 of our HTR is
~60% that of our C (P 0.0004). The peak exercise RER
appears similar in both groups and appears indicative of maximal
exercise (P = 0.94, NS), whereas the peak
E and HR are only insignificantly lower
in HTR than in C (P = 0.35 and 0.17, respectively).
View this table:
[in this window]
[in a new window]
|
Table 1.
Results of incremental maximal exercise in heart transplant recipients
and controls with values recorded at rest, ventilatory threshold,
and maximal effort
|
|
Constant-rate exercise.
Because the HTR exercised at 75% of the work rate of their individual
VT, four patients exercised at 30 W and six at 40 W, corresponding to a
mean work rate of 36 ± 5 W. Therefore, all the C exercised at 40 W to
serve as controls at a similar absolute work rate. We label this latter
group of measurements as C40W. Similarly to the HTR, all C also
performed a set of constant-rate exercises at 75% of the work rate of
their individual VT, corresponding to an average of 67 ± 14 W. We
label this group of measurements as C67W. The resting and steady-state
exercise values are presented in Table 2.
At rest, all the gas exchange values are similar, and the RER values
remain within the range of usual resting values in all three groups of
measurements. The resting HR is higher in HTR, but with only a slightly
and insignificantly lower O2p (Table 2). During
steady-state exercise, all the ventilation and gas exchange parameters
are similar in HTR and C at the same absolute work rate (HTR vs. C40W;
Table 2). At these loads, the patients' HR tends to be only
insignificantly higher and the O2p slightly lower than that
in C. On the other hand, the C67W group of measurements is higher for
the ventilation, gas exchanges, and O2p, because of the
higher work rate performed. Because during steady-state exercise the
RER values remain <1 for the three groups of measurements, we assume
that the subjects really exercised below their VT.
View this table:
[in this window]
[in a new window]
|
Table 2.
Rest and steady-state values during the constant-rate exercises
in heart transplant recipients and controls
|
|
O2 and HR kinetics.
The values of the rest-to-constant-rate-exercise transition are
presented in Table 3. At a similar absolute
work rate (HTR and C40W), when the overall data of the 10-min
square-wave forcing are considered regardless of its phases [as
in model 1 in the work by Whipp et al. (37)], we observed
a higher for the monoexponentially fitted
O2 data in HTR than in C. Accordingly, the calculated O2 deficit is higher in HTR
than in C (Table 3) at the same absolute work rate. The of the
monoexponentially fitted O2p data is also greater in HTR
than in C at the same work rate (Table 3). The half time of the HR
increase is much greater in HTR than in C because of the at least
partially denervated state of the transplanted heart. Interestingly,
the dispersion of these half times is large in the HTR (range:
16-266 s), with some patients having a HR response almost as fast
as that of the C. The patients' individual HR responses and the
corresponding O2 are depicted
in Fig. 2 and are sorted by increasing
phase 1 durations.
View larger version (57K):
[in this window]
[in a new window]
|
Fig. 2.
Individual recordings of heart rate (HR) and
O2 variations
(O2) during exercise.
Patients are sorted by increasing phase 1 duration to exemplify
relationships between chronotropic response and phase 1 characteristics.
|
|
At a similar relative work rate (HTR vs. C67W), the HTR still exhibited
a slightly but significantly higher than the C (38.4 ± 7.5 vs.
30.8 ± 8.2 s; P 0.05) for the monoexponential fits of the
entire 10-min O2 data. At
these work rates, the O2 deficit is only insignificantly
higher in C, despite their higher work rate (378 ± 87 vs. 474 ± 230 ml in HTR and C67W, respectively; P = 0.47, NS).
The comparison of the HTR and C exercise transitions at the same
absolute work rate shows that the differences concern essentially the
phase 1 rather than the phase 2 response (Fig.
3, Table 3). The phase 1 duration
is significantly longer and the
O2 and O2p
increases during phase 1 are significantly smaller in HTR (Table 3). On the other hand, as shown in Table 3 and Fig.
4, the values of monoexponential fits
of the phase 2 O2
() or O2p data that start at the end of phase 1 [as in model 3 in the work by Whipp et al. (37)]
are similar in HTR and C at the same absolute work rate.
View larger version (32K):
[in this window]
[in a new window]
|
Fig. 3.
Comparison between breath-by-breath
O2 response to first 160 s of
exercise in heart transplant recipients and normal controls. A:
means ± SD of O2
response of each group is presented side by side on similarly
scaled axes. B: mean values are superimposed (heart transplant
recipients, ; normal controls, ), showing that kinetic
differences concern essentially the initial phase 1 response.
Rest value (or x-axis = 0) represents average of data over last
minute of rest.
|
|
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 4.
Comparison between mean breath-by-breath
O2 response during phase
2 in heart transplant recipients () and controls (). These
means of individual breath-by-breath data (after time averaging over 6 identical transitions for each subject) time aligned to beginning of
phase 2 show the identical kinetic behavior of
O2 during phase 2 in
patients and controls.
|
|
Whereas no significant correlations were found with the phase 1 amplitude values, we found an inverse correlation between the phase
1 duration and the peak
O2 (r = 0.67, P 0.04; Fig. 5). We
also observed a slight but significant direct correlation between the
phase 1 duration and the IVRT (r = 0.63, P 0.05; Fig. 5) but not with the transmitral pressure decay half time (r = 0.32, P = 0.36) or with the ejection fraction
(r = 0.07, P = 0.84). The phase
1 duration is correlated to the HR half time (r = 0.81, P 0.005; Fig. 6) and is
inversely correlated to the HR variation during phase 1 (r = 0.72, P 0.02; Fig. 6). The delay since
the operation only tends to be inversely related to the phase 1 duration (r = 0.58, P = 0.08).
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 5.
Correlation of phase 1 duration with peak
O2 (A) and with
echo-Doppler isovolumic relaxation time (IVRT; B) in the 10 heart transplant patients. Peak
O2 is inversely correlated
with the phase 1 duration, suggesting that a slow phase
1 may have a negative impact on exercise capacity. Phase 1 duration is directly correlated with echo-Doppler IVRT estimations,
suggesting that a slowed ventricular relaxation may lengthen phase
1.
|
|
View larger version (14K):
[in this window]
[in a new window]
|
Fig. 6.
Correlation of phase 1 duration with parameters of chronotropic
response to rectangular exercise in the 10 heart transplant patients.
A: phase 1 duration is positively correlated with the
half time of heart rate increase during constant-rate exercise,
suggesting that a rapid heart rate response may shorten the phase
1 duration. B: phase 1 duration is negatively
correlated with amplitude of heart rate response during phase
1, suggesting that appearance of a heart rate increase during early
exercise may shorten phase 1 duration.
|
|
| |
DISCUSSION |
In accordance with our working hypothesis, we found delayed
O2 kinetics in HTR even long
after the operation, leading to an increased O2 deficit at
the onset of moderate subthreshold exercise. When partitioned in the
initial cardiodynamic phase 1 and in the subsequent metabolic
phase 2, it appears that most of the delayed
O2 kinetics are due to a less
pronounced and longer phase 1 rather than to a slowed phase
2. Because this phase 1 response depends essentially on the
capacity of the cardiovascular system to abruptly increase the
pulmonary blood flow and, therefore, cardiac output at the beginning of
exercise, our findings represent an element of central exercise
limitation that operates even at moderate subthreshold work rates.
Moreover, we found that the phase 1 duration correlates with
the IVRT, taken as an index of ventricular relaxation, and inversely
with indexes of rapidity and importance of the chronotropic response at
the onset of exercise. By slowing the cardiac output kinetics to
various extents, the degrees of chronotropic incompetence and
relaxation abnormalities may combine their deleterious effects on
O2 kinetics, thus explaining in part the wide differences in the early gas exchange response to
exercise from one patient to another.
Maximal and steady-state submaximal exercise.
In our patients, we observed a peak exercise capacity (8, 18, 19, 24),
a O2 at the VT, and a maximal
HR (24) within the range of what has been previously shown in long-term HTR. Yet the HR dispersion at maximal exercise was higher in HTR than
in C, suggesting that the chronotropic incompetence varies widely among
patients. Because the peak RER is similar in both groups and clearly
>1, we assume that we adequately assessed peak exercise capacity in
our subjects. During the constant-rate exercises, the steady-state RERs
remained <1 in all HTR and C, indicating that they exercised below
their VT. There is no clear explanation for the slightly higher RER in
the C40W group at rest than in the HTR and the C67W groups. This C40W
group of measurements might have been done in a slightly more stressful
environment owing to the higher HR observed. However, this should not
affect our conclusions.
Overall O2 kinetics.
When the rest-to-subthreshold-exercise increase in
O2, O2p, and HR
are considered regardless of the transition phases, we found higher values in HTR than in C exercising at the same absolute (Table 3) and
relative levels. Only a few studies have compared the
O2 kinetics in HTR and C,
either in terms of half times (8, 9, 15, 26) or values (27). Our C
kinetics agree with the previously published values in normal subjects (32, 37). On the other hand, our HTR
O2 is shorter than that
previously reported. This may be explained as follows:
1) The O2 kinetics
are affected by work rate. Below the VT, the
O2 is almost independent
of work rate (32), but above the VT it increases greatly as a function
of work rate (32, 36). In previous studies on
O2 kinetics in HTR, one was
designed to assess the transition below the VT (27). Others were either slightly (9, 26) or far (8) above it.
2) Above the VT, the steady state no longer holds and the
O2 increases further with a
slow component (12, 36, 39). At these work-rate levels, because the
O2 does not follow
first-order kinetics, monoexponential fitting of the data is not
accurate, and the half-time estimates increase with the duration of
exercise. This is the case in the studies by Cerretelli et al. (8, 9), Meyer et al. (26), and probably of some patients in the study by Grassi
et al. (15) when working at 50 W. In this latter study, the
O2 kinetics were
also slower in HTR at 25 W, a work rate below the VT in both HTR and C;
nevertheless, at such low exercise levels, close to 80% of the
O2 increase occurs during
phase 1 (32), potentially obscuring the phase 2 O2 kinetics in some subjects,
especially in the normal C.
3) The large difference between the values of Paterson et al.
(27) (77 ± 26 s) and ours (38.4 ± 7.5 s) might be explained by the
fact that for their patients it was <3 mo (2.3 ± 0.2 mo) after
surgery, whereas for our HTR it was >6 mo after transplantation. Increases in the exercise capacity within the first 6 mo have been
consistently reported (5, 24) because of improvements in the
cardiocirculatory response to exercise (5, 24) and the retraining
effect of daily activities (19). These mechanisms might have speeded up
O2 kinetics within the first
6 mo after the operation by shortening the phase 1, as well as
by accelerating the phase 2, response. For time periods later
than 6 mo, Grassi et al. (15) showed that the lapse of time after
surgery does not affect the
O2 kinetics any more. As
previously suggested (15), because the of the
O2 increase is higher in our
patients than in the C at similar absolute and relative work rates, the long-term HTR keep slightly but significantly delayed
O2 kinetics, despite an
improvement with time in their exercise capacity, and, accordingly,
accumulate a higher O2 deficit at the onset of subthreshold exercise.
O2 kinetics during phases
1 and 2.
This study is the first to examine the respective contributions of
phases 1 and 2 to the
O2 kinetics in HTR. Grassi et
al. (14) already examined in HTR the phase 1 ventilatory
response, but they did not report
O2 during this phase. As we
separated the O2 transition
into its two phases, we observed that the delayed overall
O2 kinetics of HTR depend on
an abnormal phase 1 response, in amplitude as well as in
duration, whereas the phase 2 kinetics are similar in HTR and
C. We found that the patients' phase 1 duration is negatively
correlated with their peak
O2, whereas neither the
phase 1 duration nor its amplitude is correlated with the peak
O2 in C. Because we were not
able to correct for the acute changes in functional residual capacity,
these changes may have affected the breath-by-breath
O2 pattern during phase 1 (2). Nevertheless, the artifacts due to lung gas stores changes should be of the same magnitude in patients and C, because both groups
have been assessed with the use of the same technology and protocol.
Should the differences in phase 1 gas exchanges that we
observed be due only to lung gas stores changes, then a systematic
difference between the HTR and C functional residual capacity changes
would have occurred during phase 1. This is not likely the case
because Grassi et al. (14) observed that the changes in inspiratory and
expiratory air flows are the same during phase 1 in HTR as well
as in matched C.
The importance of the O2
response during phase 1 can be taken as a noninvasive marker of
the abrupt circulatory adjustments at the beginning of exercise. Krogh
and Lindhard (21) already hypothesized in their pioneering work that
the increase in O2 that
follows the initiation of exercise must be due to an abrupt increase in
cardiac output. Afterwards, several lines of evidence confirmed that
the major determinant of the
O2 increase during phase 1 is the initial increase in pulmonary blood flow, either by indirect arguments (36) or by direct comparisons of the kinetics of
O2 and cardiac output
increase (40). Moreover, phase 1 has been found to be blunted
in diseases that prevent the pulmonary blood flow from increasing with
exercise (30, 31). The duration of phase 1 represents also the
circulatory delay taken by the desaturated muscular blood to reach the
lung exchanger (1, 36). For instance, this duration has been found to
be longer than normal in pulmonary vascular disease, in which the
increase in pulmonary blood flow is abnormally low at the onset of
exercise (30). Accordingly, our results suggest that the HTR are unable to increase their cardiac output abruptly at the onset of exercise. At
present, one group only measured the cardiac output kinetics in HTR by
using Kubiceck's impedance method (9, 26). Although the validity of
Kubicek's impedance method has been questioned (35), this group found
neither blunted cardiac output kinetics in HTR nor a significant effect
of these kinetics on the O2 response (9, 26). The fact that the phase 1 duration is
inversely correlated with the peak
O2 in HTR suggests that the
abnormal phase 1 may contribute to limit their maximal exercise
capacity. On the other hand, the phase 2 O2 increase reflects, at the lung exchanger level, the increase in muscular
O2 (1). McCreary et
al. (25) showed that the subthreshold phase 2 O2 kinetics reflect the
kinetics of muscular phosphocreatine decrease. Nevertheless, as shown
by Barstow et al. (1), phases 1 and 2 are not independent processes: in the face of unchanged muscular
O2 kinetics, phase 2 should be speeded up when phase 1 is altered by delayed blood flow kinetics (1, 7). Thus our similar phase 2 values in
HTR and C may in fact reflect slower than normal muscular
O2 kinetics in patients. This
agrees with the impaired muscular oxidative capacity that we and others
have reported after heart transplantation (18, 23).
The issue of whether muscular
O2 is limited at the
beginning of exercise, either by O2 delivery or by
intramuscular mechanisms beyond the capillary level, is not yet
completely clarified and might be affected by the subject's fitness or
pathological situations. On the one hand, Grassi et al. (16) observed
in well-trained cyclists that the initial increase in muscular blood
flow is accompanied by a temporary decrease in muscular arteriovenous
O2 difference, showing that muscular
O2 temporarily lags behind
local O2 delivery. On the other hand, several authors
reported delayed O2 kinetics with interventions that blunt the cardiac output response to exercise (39).
By assessing the O2 kinetics
with increasing levels of carboxyhemoglobin in sedentary subjects,
Koike et al. (20) showed that the O2 delivery can limit the
phase 2 O2 kinetics
at the diffusion level, even at the onset of subthreshold exercise.
Cochrane and Hughson (10) suggested that, if the muscular
O2 utilization determines the
O2 kinetics in the normal
situation, then the balance between O2 delivery and
O2 utilization is very delicate during the unsteady state
after exercise onset. Thus even subtle changes in O2
transport, diffusion gradients, or blood flow requirements in the
nonworking tissues would cause the
O2 increase to follow O2 delivery.
Such considerations, together with methodology differences, might
explain the differences between our results and those of others, as
well as the apparent contradictions between the data reported by
Paterson et al. (27) and Grassi et al. (15). Our group has shown that
HTR have a decreased muscular capillary density (23); thus their
phase 2 O2 kinetics
might be more sensitive to O2 transport, especially if the
chronotropic response is blunted (24). This explains the observation of
Paterson et al. of a slower than normal phase 2 O2 response in patients early
after surgery that speeds up to near normal values during a second
square-wave exercise when the HR response is faster (27). The data of
Paterson et al. suggest also that the phase 2 of exercise
does not differ in HTR and C during the second square-wave exercise
when O2 delivery is no longer limited. This agrees with our
similar phase 2 kinetics in HTR and C. The fact that the half
times of the O2 increase failed to shorten at the time of a second square-wave forcing, despite
slightly faster cardiac output kinetics in the study by Grassi et al.
(15), may be explained by the fact that several of their patients must
have exercised close to their VT, as suggested by their lactate and RER
data. A transient anaerobic lactic O2 deficit may have
occurred at the onset of the second square-wave forcing (11), thus
masking the effects of the improved O2 delivery during that
second exercise. Therefore, we believe that the observations by Grassi
et al. (15) still concur with those of Paterson et al. (27) and our conclusions.
Potential mechanisms of the delayed
O2 transition in HTR.
Because the patients' cardiodynamic phase 1 is abnormal, we
attempted to correlate its duration
and O2 or O2p
amplitude with parameters of graft function to characterize some of its mechanisms. We found the phase 1 duration to be correlated with the IVRT, as a marker of the graft relaxation, and with the half time
of the HR response, as an index of the chronotropic response kinetics.
Despite a normal systolic function, the transplanted heart is
characterized by slowed relaxation (28) and decreased compliance (18),
which both worsen during acute rejection (34). These diastolic
abnormalities may affect the phase 1 kinetics by preventing the
cardiac output from increasing abruptly at the beginning of exercise,
as suggested by our correlation between the IVRT and the phase
1 duration. However, because this result offers indirect evidence
at most, and the stroke volume has been observed to increase after
exercise onset in HTR (9, 26), the impact of the impaired graft
relaxation should be investigated further.
The correlations between the phase 1 duration and the
parameters of chronotropic response that we found in our patients
suggest more firmly the existence of a relationship between the
phase 1 O2 kinetics
and the rapidity of HR increase after the onset of exercise in the HTR.
Although the mean HR kinetics are consistent with previous reports
(29), we found a wide range of HR half times (16-266
s). This opens the question of the occurrence of partial
orthosympathetic reinnervation in some patients. Some HTR have a HR
response almost as fast as normal, whereas others exhibit a widely
delayed response (Fig. 2). Several lines of evidence suggest that
orthosympathetic reinnervation (38) and a characteristic variability of
the R-R interval (4) can occur with time in the transplanted cardiac
graft but with no clear beneficial consequences for the peak exercise
capacity (5, 24). The observation of a rapid HR response accompanied by
a short phase 1 in certain of our patients could reflect
partial reinnervation. Nevertheless, this important issue should be
investigated further.
Potential clinical consequences of the delayed
O2 kinetics.
Because the delay in the O2
increase at the beginning of subthreshold exercise concerns essentially
the early cardiodynamic phase 1, which depends on graft
characteristics, we speculate that this delay might be only partially
shortened by retraining but might be sensitive to chronotropic
interventions. The patients might tolerate differently the onset of
exercise, depending on their own
O2 , chronotropic
response, and ventricular diastolic function. The impaired ventricular
relaxation of the graft depends on numerous factors (28, 34), including
some pericardial constraint due to surgery and the effects of
hypertension, denervation, and interstitial fibrosis presumably due to
rejection and cyclosporine therapy. At present, the effects that these
factors may have on retraining are unknown, but they are unlikely to
occur. The chronotropic response to exercise of the graft
depends on atrial stretch, the rise of circulating catecholamines (29),
atrial 
-receptor sensitivity (13), and reinnervation (38). Although
training is unlikely to affect reinnervation, the catecholamine
response and receptor sensitivity may be increased by regular exercise,
thus improving the chronotropic response (19). Moreover, those patients
who have a slow O2 response
would benefit from gradual increases in exercise levels and from
techniques aimed at accelerating the HR before the exercise, if they
are willing to engage in sports.
Conclusion.
Our study shows that the O2
transition remains delayed at the onset of moderate subthreshold
exercise in long-term stabilized HTR, leading to an increased
O2 deficit, but to a smaller extent than early after
surgery. It also shows that this slowed
O2 transition is due to an
abnormally low and delayed early cardiodynamic phase 1. This
indicates that the graft is less able than normal to abruptly increase
its output at the onset of exercise, even if moderate. Our data also
suggest that the abnormal phase 1 might result from an impaired
relaxation of the graft and from a delayed chronotropic response to
exercise. Therefore, the abnormal response of the graft during exercise
transitions introduces a central factor of limitation that operates
even at moderate subthreshold exercise levels. Nevertheless, as the
patients' phase 2 kinetics should have appeared faster with
respect to their lower phase 1 but were found to be similar to
that of normal subjects, an impairment of the muscular
metabolism may also operate in HTR at the onset of exercise. As a
consequence of the phase 1 limitation, which appears to be
inherent in the transplanted graft function, exercise training may only
partially improve the transitory gas exchange kinetics at the onset of
moderate exercise but should have a greater impact on
exercise-to-exercise than on rest-to-exercise transients.
Therefore, a progressive onset of work should be better tolerated by
the HTR than an abrupt onset.
| |
ACKNOWLEDGEMENTS |
This study was supported in part by the Institut National
de la Santé et de la Recherche Médicale network
"Activité physique muscle et handicap."
| |
FOOTNOTES |
Address for reprint requests and other correspondence: B. Mettauer,
Département de Physiologie, Service des Explorations
Fonctionnelles du Système Circulatoire, Pavillon Chirurgical A,
Hôpital Central, 1, Place de l'Hôpital, F-67091 Strasbourg
CEDEX, France (E-mail:
Bertrand.Mettauer{at}physio-ulp.u-strasbg.fr).
Received 22 September 1997; accepted in final form 22 November
1999.
| |
REFERENCES |
1.
Barstow, TJ,
Lamarra N,
and
Whipp BJ.
Modulation of muscle and pulmonary O2 uptakes by circulatory dynamics during exercise.
J Appl Physiol
68:
979-989,
1990[Abstract/Free Full Text].
2.
Beaver, WL,
Lamarra N,
and
Wasserman K.
Breath-by-breath measurement of true alveolar gas exchange.
J Appl Physiol
51:
1662-1675,
1981[Abstract/Free Full Text].
3.
Beaver, WL,
Wasserman K,
and
Whipp BJ.
On-line computer analysis and breath-by-breath graphical display of exercise function tests.
J Appl Physiol
34:
128-132,
1973[Free Full Text].
4.
Bernardi, L,
Bianchini B,
Spadacini G,
Leuzzi S,
Valle F,
Marchesi E,
Passino C,
Calciati A,
Vigano M,
Rinaldi M,
Martinelli L,
Finardi G,
and
Sleight P.
Demonstrable cardiac reinnervation after human heart transplantation by carotid baroreflex modulation of RR interval.
Circulation
92:
2895-2903,
1995[Abstract/Free Full Text].
5.
Bussières, LM,
Pflugfelder PW,
Menkis AH,
Novick RJ,
McKenzie N,
Taylor AW,
and
Kostuk WJ.
Basis for aerobic impairment in patients after heart transplantation.
J Heart Lung Transplant
14:
1073-1080,
1995[ISI][Medline].
6.
Casaburi, R,
Daly J,
Hansen JE,
and
Effros RM.
Abrupt changes in mixed venous blood gas composition after the onset of exercise.
J Appl Physiol
67:
1106-1112,
1989[Abstract/Free Full Text].
7.
Casaburi, R,
Spitzer S,
and
Wasserman K.
Effects of altering heart rate on oxygen uptake at exercise onset.
Chest
95:
6-12,
1989[Abstract/Free Full Text].
8.
Cerretelli, P,
Marconi C,
Meyer M,
Ferretti G,
and
Grassi B.
Gas exchange and metabolic transients in heart transplant recipients.
Respir Physiol
74:
355-371,
1988[ISI][Medline].
9.
Cerretelli, P,
Marconi C,
Meyer M,
Ferretti G,
and
Grassi B.
Gas exchange kinetics in heart transplant recipients.
Chest
101:
199S-205S,
1992[Abstract/Free Full Text].
10.
Cochrane, JE,
and
Hughson RL.
Computer simulation of O2 transport and utilization mechanisms at the onset of exercise.
J Appl Physiol
73:
2382-2388,
1992[Abstract/Free Full Text].
11.
Di Prampero, PE,
Mahler PB,
Giezendanner D,
and
Cerretelli P.
Effects of priming exercise on O2 kinetics and O2 deficit at the onset of stepping and cycling.
J Appl Physiol
66:
2023-2031,
1989[Abstract/Free Full Text].
12.
Gaesser, GA,
and
Poole DC.
The slow component of oxygen uptake kinetics in humans.
Exerc Sports Sci Rev
24:
35-71,
1996.
13.
Gilbert, EM,
Eiswirth CC,
Mealey PC,
Larrabee P,
Herrick CM,
and
Bristow MR.
-Adrenergic supersensitivity of the transplanted human heart is presynaptic in origin.
Circulation
79:
344-349,
1989[Abstract/Free Full Text].
14.
Grassi, B,
Ferretti G,
Xi L,
Rieu M,
Meyer M,
Marconi C,
and
Cerretelli P.
Ventilatory response to exercise after heart and lung denervation in humans.
Respir Physiol
92:
289-304,
1993[ISI][Medline].
15.
Grassi, B,
Marconi C,
Meyer M,
Rieu M,
and
Cerretelli P.
Gas exchange and cardiovascular kinetics with different exercise protocols in heart transplant recipients.
J Appl Physiol
82:
1952-1962,
1997[Abstract/Free Full Text].
16.
Grassi, B,
Poole DC,
Richardson RS,
Knight DR,
Erickson BK,
and
Wagner PD.
Muscle O2 uptake kinetics in humans: implications for metabolic control.
J Appl Physiol
80:
988-998,
1996[Abstract/Free Full Text].
17.
Jones, PW,
French W,
Weissman ML,
and
Wasserman K.
Ventilatory response to cardiac output changes in patients with pacemakers.
J Appl Physiol
51:
1103-1107,
1981[Abstract/Free Full Text].
18.
Kao, AC,
Van Trigt P, III,
Shaeffer-McCall GS,
Shaw JP,
Kuzil BB,
Page RP,
and
Higginbotham MB.
Central and peripheral limitations to upright exercise in untrained cardiac transplant recipients.
Circulation
89:
2605-2615,
1994[Abstract/Free Full Text].
19.
Kavanagh, T,
Yacoub MH,
Mertens DJ,
Kennedy J,
Campbell RB,
and
Sawyer P.
Cardiorespiratory response to exercise training after orthotopic cardiac transplantation.
Circulation
77:
162-171,
1988[Abstract/Free Full Text].
20.
Koike, A,
Wasserman K,
McKenzie DK,
Zanconato S,
and
Weiler-Ravell D.
Evidence that diffusion limitation determines oxygen uptake kinetics during exercise in humans.
J Clin Invest
86:
1698-1706,
1990.
21.
Krogh, A,
and
Lindhard J.
The regulation of respiration and circulation during the initial stages of muscular work.
J Physiol (Lond)
47:
112-136,
1913.
22.
Lamarra, N,
Whipp BJ,
Ward SA,
and
Wasserman K.
Breath by breath "noise" and parameter estimation of gas exchange kinetics.
J Appl Physiol
62:
2003-2012,
1987[Abstract/Free Full Text].
23.
Lampert, E,
Mettauer B,
Hoppeler H,
Charloux A,
Charpentier A,
and
Lonsdorfer J.
Structure of skeletal muscle in heart transplant recipients.
J Am Coll Cardiol
28:
980-984,
1996[Abstract].
24.
Mandak, JS,
Aaronson KD,
and
Mancini DM.
Serial assessment of exercise capacity after heart transplantation.
J Heart Lung Transplant
14:
468-478,
1995[ISI][Medline].
25.
McCreary, CR,
Chilibeck PD,
Marsh GD,
Paterson DH,
Cunningham DA,
and
Thompson RT.
Kinetics of pulmonary oxygen uptake and muscle phosphates during moderate-intensity calf exercise.
J Appl Physiol
81:
1331-1338,
1996[Abstract/Free Full Text].
26.
Meyer, M,
Cerretelli P,
Marconi C,
Rieu M,
and
Cabrol C.
Cardiorespiratory adjustment to exercise after cardiac transplantation.
In: Clinical Aspects of O2 Transport and Tissue Oxygenation, edited by Reinhart K,
and Eyrich K.. Berlin: Springer Verlag, 1989, p. 476-499.
27.
Paterson, DH,
Cunningham DA,
Pickering JG,
Babcock MA,
and
Boughner DR.
Oxygen uptake kinetics in cardiac transplant recipients.
J Appl Physiol
77:
1935-1940,
1994[Abstract/Free Full Text].
28.
Paulus, WJ,
Bronzwaer JGF,
Felice H,
Kishan N,
and
Wellens F.
Deficient acceleration of left ventricular relaxation during exercise after heart transplantation.
Circulation
86:
1175-1185,
1992[Abstract/Free Full Text].
29.
Perini, R,
Orizio C,
Gamba A,
and
Veicsteinas A.
Kinetics of heart rate and catecholamines during exercise in humans, the effects of heart denervation.
Eur J Appl Physiol
66:
500-506,
1993[ISI].
30.
Sietsema, KE.
Oxygen uptake kinetics in response to exercise in patients with pulmonary vascular disease.
Am Rev Respir Dis
145:
1052-1057,
1992[ISI][Medline].
31.
Sietsema, KE,
Cooper DM,
Perloff JK,
Rosove MH,
Child JS,
Canobbio MM,
Whipp BJ,
and
Wasserman K.
Dynamics of oxygen uptake during exercise in adults with cyanotic congenital heart disease.
Circulation
73:
1137-1144,
1986[Abstract/Free Full Text].
32.
Sietsema, KE,
Daly JA,
and
Wasserman K.
Early dynamics of O2
TOP
ABSTRACT
INTRODUCTION
