Aerobic fitness effects on exercise-induced low-frequency
diaphragm fatigue
Mark A. Babcock, David F. Pegelow, Bruce D. Johnson and Jerome A. Dempsey
J Appl Physiol 81:2156-2164, 1996. ;
You might find this additional info useful...
This article cites 1 articles, 1 of which you can access for free at:
http://jap.physiology.org/content/81/5/2156.full#ref-list-1
This article has been cited by 11 other HighWire-hosted articles:
http://jap.physiology.org/content/81/5/2156#cited-by
Updated information and services including high resolution figures, can be found at:
http://jap.physiology.org/content/81/5/2156.full
This information is current as of November 28, 2012.
Journal of Applied Physiology publishes original papers that deal with diverse area of research in applied physiology,
especially those papers emphasizing adaptive and integrative mechanisms. It is published 12 times a year (monthly) by
the American Physiological Society, 9650 Rockville Pike, Bethesda MD 20814-3991. Copyright © 1996 the American
Physiological Society. ISSN: 8750-7587, ESSN: 1522-1601. Visit our website at http://www.the-aps.org/.
Downloaded from http://jap.physiology.org/ by guest on November 28, 2012
Additional material and information about Journal of Applied Physiology can be found at:
http://www.the-aps.org/publications/jappl
Aerobic fitness effects on exercise-induced
low-frequency diaphragm fatigue
MARK A. BABCOCK, DAVID F. PEGELOW, BRUCE D. JOHNSON,
AND JEROME A. DEMPSEY
John Rankin Laboratory of Pulmonary Medicine,
University of Wisconsin, Madison, Wisconsin 53705
low-frequency fatigue; aerobic capacity; diaphragm force output
THIS STUDY WAS AIMED at determining the effects of
aerobic capacity or maximal O2 consumption (V̇O2 max)
on exercise-induced diaphragm fatigue. Heavy endurance exercise to exhaustion has been shown to cause
significant reductions in the force production of the
diaphragm in response to low-frequency supramaximal
phrenic nerve stimulation (2, 3, 18). This fatigue of the
diaphragm is in part determined by the amount of force
output by the diaphragm during exercise (3). Because
the highly trained subject is capable of sustaining
endurance exercise at greater-than-average metabolic
rates, ventilatory outputs, and inspiratory muscle force
outputs, these increased requirements may make the
athlete more susceptible to exercise-induced diaphragm fatigue. On the other hand, the endurance of
the respiratory muscles, as judged by the maximum
ventilation sustainable to the point of volitional task
failure, is generally greater in the endurance athlete (8,
10, 13) and increases in formerly sedentary subjects
2156
who undergo whole body physical training (27). These
adaptations of the respiratory muscles to whole body
training may protect the athlete from exercise-induced
diaphragm fatigue. In fact, Coast et al. (11) claim, on
the basis of measurements of maximal volitional inspiratory occlusion pressure before and after exercise in
the trained and untrained subject, that highly trained
subjects were completely protected from exerciseinduced respiratory muscle fatigue.
We hypothesized that the subjects with greater aerobic capacity will be protected completely or at least
partially from exercise-induced diaphragm fatigue. In
the present study, we have used bilateral phrenic nerve
stimulation (BPNS) (5) to provide more objective and
specific measures of diaphragm fatigue in 24 normal
subjects of widely varying levels of habitual activity
and V̇O2 max who were subjected to whole body endurance exercise of comparable intensity and duration.
METHODS
Subjects. Twenty-four subjects (20 men and 4 women) gave
informed consent to participate in the study. All procedures
were approved by the Institutional Review Board of the
University of Wisconsin-Madison. These subjects had participated in one of three previous studies on the effects of
exercise-induced diaphragm fatigue in our laboratory (2, 3,
18). These data were combined in the present study to provide
a larger sample size for analysis of fitness effects on exerciseinduced diaphragm fatigue.
BPNS. The BPNS procedure has been described previously
(5, 18); therefore, only a brief description follows. Two balloontipped catheters were passed intranasally: one was positioned in the stomach to measure gastric pressure (Pga), and
one was positioned in the lower one-third of the esophagus to
measure esophageal pressure (Pes). The algebraic sum of Pga
and Pes gave transdiaphragmatic pressure (Pdi). Surface
electromyogram electrodes were placed over each hemidiaphragm in the sixth or seventh intercostal space near the
costal margin to record the compound muscle action potential
(M wave) resulting from phrenic nerve stimulation. After
location and marking of the surface stimulation site on the
neck, ,2 cm above the clavicle, the maximal M wave was
determined by increasing the stimulation current until no
change in M wave amplitude was found. To ensure supramaximal stimulation, the current of both stimulators was increased a further 50% above this level. This procedure was
carried out before each BPNS data collection during preexercise control and 6–12, 30, and 60 min after exercise. Lung
volumes were continuously monitored throughout the stimulation tests by connecting the subject to a wedge spirometer
via a mouthpiece that was occluded before stimulation and
providing the subject with visual feedback by means of an
oscilloscope display of the lung volume.
During each BPNS session, 9–12 repeats of ‘‘twitch’’ stimulation at functional residual capacity (FRC) and 3–5 repeat
0161-7567/96 $5.00 Copyright r 1996 the American Physiological Society
Downloaded from http://jap.physiology.org/ by guest on November 28, 2012
Babcock, Mark A., David F. Pegelow, Bruce D.
Johnson, and Jerome A. Dempsey. Aerobic fitness effects
on exercise-induced low-frequency diaphragm fatigue. J.
Appl. Physiol. 81(5): 2156–2164, 1996.—We used bilateral
phrenic nerve stimulation (BPNS; at 1, 10, and 20 Hz at
functional residual capacity) to compare the amount of exercise-induced diaphragm fatigue between two groups of healthy
subjects, a high-fit group [maximal O2 consumption (V̇O2max) 5
69.0 6 1.8 ml · kg21 · min21, n 5 11] and a fit group (V̇O2 max 5
50.4 6 1.7 ml · kg21 · min21, n 5 13). Both groups exercised at
88–92% V̇O2 max for about the same duration (15.2 6 1.7 and
17.9 6 2.6 min for high-fit and fit subjects, respectively, P .
0.05). The supramaximal BPNS test showed a significant
reduction (P , 0.01) in the BPNS transdiaphragmatic pressure (Pdi) immediately after exercise of 223.1 6 3.1% for the
high-fit group and 223.1 6 3.8% (P . 0.05) for the fit group.
Recovery of the BPNS Pdi took 60 min in both groups. The
high-fit group exercised at a higher absolute workload, which
resulted in a higher CO2 production (126%), a greater
ventilatory demand (116%) throughout the exercise, and an
increased diaphragm force output (128%) over the initial
60% of the exercise period. Thereafter, diaphragm force
output declined, despite a rising minute ventilation, and it
was not different between most of the high-fit and fit subjects.
In summary, the high-fit subjects showed diaphragm fatigue
as a result of heavy endurance exercise but were also
partially protected from excessive fatigue, despite high ventilatory requirements, because their hyperventilatory response
to endurance exercise was reduced, their diaphragm was
utilized less in providing the total ventilatory response, and
possibly their diaphragm aerobic capacity was greater.
AEROBIC FITNESS EFFECTS ON DIAPHRAGM FATIGUE
resistor at five flow rates. A multiple regression equation for
each subject was computed from these data, and PcapI could
be calculated at any given lung volume and flow rate. During
tidal inspiration in exercise, the peak inspiratory Pes was
identified, and we applied the corresponding volume and flow
rate to the multiple regression equation to determine the
available PcapI. The difference between the peak pressure
reached during tidal inspiration and PcapI represented an
approximation of the demand-capacity margin for inspiratory
pressure development during exercise.
Exercise test protocols. The first exercise test determined
the subject’s V̇O2 max by use of a progressive short-term test to
volitional exhaustion, as outlined previously (19). Nineteen of
the subjects ran on a treadmill, and five exercised on a
stationary cycle ergometer.
The endurance exercise session was conducted on a separate day. The various BPNS measures were completed before
exercise (baseline measures). Subjects warmed up briefly
with light exercise and were then quickly brought up to a
workload that required 95% V̇O2 max, which was maintained
until volitional exhaustion. Immediately after exercise the
BPNS measures were repeated (within 6–12 min); this was
also done 30 min after exercise and then every 30 min until
the BPNS Pdi measurements returned to the baseline.
During the exercise at 3-min intervals and at exercise
termination, the subjects were asked to rate whole body effort
using the Borg 10-point scale (7). Each test was terminated at
the subject’s volitional fatigue. During exercise, expired gases,
flow, volumes, Pes, Pga, Pdi, and mouth pressure were
monitored continuously. Blood O2 saturation was measured
throughout exercise by ear oximetry (Hewlett-Packard). At 2to 3-min intervals, inspiratory capacity efforts were made in
duplicate to estimate EELV (20). For analysis, data from
20–30 consecutive breaths were averaged at 3-min intervals
during the exercise. A mean value for the time integral of the
inspiratory Pdi (ePdi) and the mean time integral of the
inspiratory Pes (ePes) were calculated over the 20- to 30breath sample by the computer. Each of the time integrals
was multiplied by the breathing frequency (f) to provide
results of force output of the diaphragm (ePdi · f) and all the
inspiratory muscles (ePes · f).
Statistical analyses were done using the statistical program Systat. Values are means 6 SE. One-way analysis of
variance with repeated measures was used to determine
differences in mean values over the duration of the exercise
and recovery period. Student’s unpaired t-test was used to
detect differences between mean values of the high-fit and fit
groups. The level of significance was set at P , 0.05.
RESULTS
The subjects were divided into two groups on the
basis of a V̇O2 max of 60 ml · kg21 · min21 (Table 1). The
high-fit group V̇O2 max (n 5 11) ranged from 61.1 to 78.6
ml · kg21 · min21, and the fit group V̇O2 max (n 5 13)
ranged from 39.5 to 58.6 ml · kg21 · min21. The results of
the routine pulmonary function tests are also reported
in Table 1. There was no difference between the high-fit
and fit groups, except the high-fit group had a higher
MVV than the fit group.
Response to BPNS. At 6–12 min after whole body
endurance exercise, high-fit and fit groups showed a
significant fall (P , 0.05) in the BPNS Pdi at all
stimulation frequencies (Fig. 1). The mean BPNS Pdi
remained below control values at 30 min into recovery
Downloaded from http://jap.physiology.org/ by guest on November 28, 2012
stimulations at 10 and 20 Hz at FRC with the mouthpiece
occluded were completed, and the resultant Pes, Pga, Pdi, and
M waves were collected on computer and magnetic tape for
later analysis. The 10- and 20-Hz stimulations were delivered
using a constant 400-ms train, so that four stimuli were given
at 10 Hz and eight stimuli were given at 20 Hz. A particular
stimulation was repeated if 1) the M wave amplitude changed
by greater than 615% of the maximal value, 2) the stimulated BPNS Pdi was 610% from the maximal stimulated
value, or 3) the measured lung volume varied by greater than
610%.
The stimulated BPNS Pdi values were analyzed for peak
pressure, contraction time, and half-relaxation time (RT‰ ).
The peak pressure was defined as the maximum increase in
tension from the Pdi baseline at the onset of the stimulation.
Contraction time represented the time interval from the
initiation of the stimulation until Pdi reached its peak value.
RT‰ was the time required for the Pdi to decline from peak
pressure to one-half of the peak pressure. A fatigue index was
calculated using the mean percent change in the BPNS Pdi at
each stimulation frequency (twitch, 10 Hz, and 20 Hz) and
then computing the arithmetic mean of these three values.
We have calculated two types of within-subject coefficient
of variation (CV): 1) CV for Pdi amplitude within a single trial
of 7–10 repeated stimulations (average CV 5 5.4 6 0.6%) and
2) CV for Pdi amplitude and M wave amplitude between two
mean values before and after subjects changed position (CV
for M wave 5 15% and CV for Pdi 5 10%). The latter values
were intended to test reproducibility of the BPNS technique
under conditions that required repositioning of the stimulating electrode and reestablishment of lung volume, as would
occur before vs. after exercise.
Potential complications of the BPNS technique include
twitch potentiation, changes in diaphragm length, changes in
chest wall-lung configuration and compliance after exercise,
and maintenance of supramaximal phrenic nerve stimulation. We have dealt with these problems previously (2, 3, 18;
see RESULTS ).
Pulmonary function tests. Vital capacity and inspiratory
capacity were determined using a Collins 13.5-liter watersealed spirometer. Thoracic gas volume and FRC were determined in a Collins body plethysmograph. Maximal voluntary
ventilation (MVV) was measured using a 12-s test, and the
best of three repeated trials was reported.
Measurement of mechanical limitation. The open-circuit
system for breath-by-breath measurement of inspiratory and
expiratory flow rate, volumes, pressures, end-tidal PO2 and
PCO2, and end-expiratory lung volume (EELV) has been
described in detail (20).
The level of expiratory flow limitation was determined for
each subject by averaging tidal exercise flow-volume loops
over 20–30 breaths and then plotting the loops within a
maximal volitional flow-volume envelope by use of a measured EELV (20). The maximal volitional flow-volume envelope represented the average of the pre- and postexercise
loop. The maximal effective expiratory pressure generation
was measured using the esophageal balloon before exercise
over a range of lung volumes to determine whether and when
the expiratory pressures during exercise exceeded their effective limit for flow generation (19).
The dynamic capacity (PcapI ) of all the inspiratory muscles
to generate Pes at different flow rates (velocity of shortening)
and at various lung volumes (muscle lengths) was determined
at rest by use of techniques similar to those previously
described (19). In essence, the seated subject produced maximal volitional inspiratory efforts against an occluded airway
at six to eight different lung volumes and against an external
2157
2158
AEROBIC FITNESS EFFECTS ON DIAPHRAGM FATIGUE
Table 1. Descriptive statistics for high-fit
and fit groups
n
Age, yr
Height, cm
Weight, kg
V̇O2 max , ml · kg21 · min21
V̇E, l/min
FRC, liters
IC, liters
TLC, liters
MEF50 , l/s
FEV1.0 , %FVC
MVV, l/min
High Fit
Fit
11
30.0 6 3.0
175.3 6 2.5
66.9 6 1.3
69.0 6 1.8
144.8 6 3.8
3.43 6 0.2
3.61 6 0.2
7.04 6 0.2
5.8
86.8 6 2.0
156.2 6 8.8
13
34.0 6 3.0
175.3 6 2.5
69.4 6 3.3
50.4 6 1.7*
121.6 6 7.9
3.30 6 0.2
3.21 6 0.2
6.51 6 0.3
4.8
82.7 6 1.1
130.1 6 8.1*
except in the fit group at 20 Hz. After 60 min of recovery
the BPNS Pdi values at all frequencies of stimulation
were still reduced but were not different from control
values for both groups. Immediately after exercise the
mean percent change in the BPNS Pdi averaged for the
three stimulation frequencies (i.e., diaphragm fatigue
index) was 223.7 6 3.1% in the high-fit group and
223.1 6 3.8% in the fit group (P . 0.05). All but 3 of 24
subjects showed a .15% decrease in the diaphragm
fatigue index after endurance exercise at an intensity
.90% V̇O2 max.
The percent change in the BPNS Pdi at 6–12 min
after exercise at each of the three stimulation frequencies is shown for all subjects in Fig. 2 as a function of
their V̇O2 max. The magnitude of the reduction in BPNS
Pdi after exercise varied between subjects and with
different stimulation frequencies. However, at any given
BPNS stimulation frequency, the exercise-induced decrease in BPNS Pdi was not systematically different
among subjects with different V̇O2 max.
The group mean values for the supramaximal BPNS
Pdi and the relative contributions of Pga and Pes to Pdi
at each stimulation frequency are shown in Table 2. No
changes were found after exercise in the amplitude of
the left or the right M wave; nor were there changes in
the lung volume at which the stimulations were done
(data not shown). The fall in the BPNS Pdi after the
endurance exercise was due to a greater decrease in the
absolute Pes component, but the relative contribution
of Pes to Pdi remained constant at all three stimulation
frequencies, as did the Pga-to-Pdi ratio (Table 2). The
ratios reported here were similar to reports in the
literature obtained in control conditions and after
inspiratory resistive loading to the point of task failure
(31). Changes in the time to peak tension and RT‰ of
the ‘‘twitch’’ after exercise were small and not significant.
Ventilatory response to exercise. The mean values for
ventilation and metabolic rate during exercise in highfit and fit groups are shown in Table 3 (over the final 3
Fig. 1. Mean response of high-fit (r) and fit (k) groups to supramaximal bilateral phrenic nerve stimulation (BPNS) with use of singletwitch (A): 10-Hz (B); and 20-Hz (C) frequencies before exercise and
during postexercise (Post-Exer) recovery. Pdi, transdiaphragmatic
pressure. Values are group means 6 SE. w Significantly different from
preexercise (pre-exerc) value, P , 0.05.
Downloaded from http://jap.physiology.org/ by guest on November 28, 2012
Values are means 6 SE; n, no. of subjects. V̇O2 max , maximal O2
consumption; V̇E, minute ventilation; FRC, functional residual capacity; IC, inspiratory capacity; TLC, total lung capacity; MEF50 ,
maximum expiratory flow at 50% of TLC; FEV1.0 , forced expiratory
volume in 1 s; MVV, maximum voluntary ventilation. * Significantly
different from high-fit group (P , 0.05).
min of exercise) and in Fig. 3 throughout exercise. Both
groups exercised at a similar average intensity of 92%
of V̇O2 max (88.8% at start of exercise and 97.8% at end of
exercise). There was also no difference in exercise
duration between the two groups: the high-fit group
exercised for 15.2 6 1.7 min and the fit group for 17.9 6
2.6 min (P . 0.40). Even though the high-fit and fit
groups exercised at the same relative intensity (percentage of V̇O2 max), the absolute O2 consumption and
CO2 production (V̇CO2) were 25.6 and 26.7% higher,
respectively, in the high-fit group (P , 0.05). The higher
metabolic cost of the exercise for the high-fit group
required a 16% higher minute ventilation (V̇E; P ,
0.056). Both groups showed an increase in V̇E-to-V̇CO2
ratio with time during exercise, although the mean
V̇E-to-V̇CO2 ratio (Fig. 3B) averaged 20% lower in the
high-fit group throughout exercise (P , 0.05). At the
end of exercise the rating of perceived exertion, on a
10-point Borg scale, was 9.8 6 0.1 for the high-fit group
and 9.9 6 0.1 for the fit group.
2159
AEROBIC FITNESS EFFECTS ON DIAPHRAGM FATIGUE
Table 2. BPNS Pdi and Pga-to-Pdi and Pes-to-Pdi
ratios for three stimulation frequencies before exercise
and during recovery period for both groups combined
Postexercise
Preexercise
Immediately
24.5 6 1.5
38.7 6 2.2
62.1 6 2.2
19.3 6 1.8
40.5 6 2.3
59.6 6 2.1
30 min
60 min
20.1 6 1.4
36.1 6 2.1
65.2 6 2.2
21.3 6 1.6
34.7 6 1.6
65.1 6 1.6
37.5 6 1.8
41.4 6 3.2
62.4 6 3.8
40.9 6 2.8
46.9 6 2.6
55.2 6 3.2
68.5 6 6.3
45.4 6 3.6
54.9 6 3.6
71.2 6 8.3
48.6 6 3.5
51.5 6 3.3
Twitch
Pdi, cmH2O
Pga/Pdi, %
Pes/Pdi, %
10 Hz
Pdi, cmH2O
Pga/Pdi, %
Pes/Pdi, %
46.9 6 3.5
47.2 6 1.7
52.3 6 1.9
36.1 6 3.7
46.7 6 2.7
55.8 6 2.5
20 Hz
80.5 6 6.9
49.1 6 2.2
51.2 6 2.2
74.9 6 7.4
48.9 6 3.2
51.8 6 3.4
Values are means 6 SE. Pdi, transdiaphragmatic pressure; Pes,
esophageal pressure; Pga, gastric pressure; BPNS, bilateral phrenic
nerve stimulation.
Fig. 2. Individual values for amount of exercise-induced diaphragm
fatigue, as represented by mean percent change (% n) in BPNS Pdi
from before to (immediately) after exercise in response to singletwitch (n 5 24), 10-Hz (n 5 24), and 20-Hz (n 5 21) frequencies. r,
High fit; k, fit. V̇O2 max, maximal O2 uptake.
Inspiratory muscle force output during exercise. Exercise effects on the time integrals for Pes and for Pdi are
shown for the fit and high-fit groups throughout exercise in Fig. 3, and mean values for the last 3 min of
exercise are summarized in Table 3. The mean ePes · f
increased in both groups throughout exercise, but it
was substantially greater in the high-fit group throughout the exercise time. ePdi · f was greater in the high-fit
group than in the fit group up to 60% of the total
exercise time. Thereafter ePdi · f declined with time in
the high-fit group but remained constant in the fit
group, so that mean values over the final 40% of the
exercise duration were not different between groups.
Figure 4 shows ePdi · f during tidal breathing vs.
V̇O2 max for each subject at the beginning, in the middle,
and at the end of exercise. We found significant correlations of force output with V̇O2 max over the time course of
the exercise. These plots, across the continuum of
V̇O2 max, emphasize the substantial overlap in ePdi · f
among subjects throughout the duration of exercise.
Also five of the high-fit subjects experienced exceptionally high ePdi · f over the latter 40% of the exercise.
These same five subjects did not show any greater
diaphragm fatigue than the other high-fit or fit subjects
(Fig. 2).
Flow- and pressure-volume relationship. Figure 5
shows the mean flow-volume and Pes-volume relationships during exercise for the high-fit and fit groups. The
high-fit group was progressively more flow limited
during expiration throughout the exercise [8.0% of tidal
volume (VT ) flow limited at 3 min vs. 50.8% at the end
of exercise]. These subjects also reached or slightly
exceeded their maximal effective expiratory pressure
generation over 50% of their expired VT during the last
Table 3. Comparison of selected parameters between
high-fit and fit groups over last 3 min of exercise
Exercise intensity, %V̇O2 max
Exercise time, min
V̇O2 , l/min
V̇E, l/min
ePdi · f, cmH2O · s · min21
ePes · f, cmH2O · s · min21
ePdi/ePes
V̇E/V̇CO2
EELV, %TLC
EILV, %TLC
Pi/PcapI , %
PcapI , cmH2O
Flow limitation, %VT
High Fit (n 5 11)
Fit (n 5 13)
92.3 6 3.8
15.2 6 1.7
4.40 6 0.2
145.5 6 6.7
571.2 6 78.3
814.7 6 77.8
0.74 6 .08
31.9 6 1.0
47.6 6 2.1
87.7 6 2.7
90.0
291.6
50.8
92.4 6 2.0
17.9 6 2.6
3.26 6 0.2*
121.7 6 7.9†
418.3 6 35.0‡
562.1 6 43.8*
0.79 6 .09
40.3 6 3.3
51.4 6 2.1
87.7 6 3.7
40.0
253.3
0.0*
Values are means 6 SE; n, no. of subjects. V̇O2 , absolute O2
consumption; V̇O2 max , maximal O2 consumption; V̇E, minute ventilation; ePdi · f, diaphragm force output; ePes · f, force output of all
inspiratory muscles; V̇CO2 , CO2 production; EELV and EILV, endexpiratory and end-inspiratory lung volume; PI, inspiratory pressure;
PcapI , dynamic capacity of inspiratory muscles to generate Pes; VT,
tidal volume. * Significantly different from high-fit group (P , 0.05).
† P , 0.07. ‡ P , 0.06.
Downloaded from http://jap.physiology.org/ by guest on November 28, 2012
Pdi, cmH2O
Pga/Pdi, %
Pes/Pdi, %
2160
AEROBIC FITNESS EFFECTS ON DIAPHRAGM FATIGUE
3 min of exercise. In the fit group VT was not flow
limited during expiration throughout exercise.
The high-fit group reached 31% (at the beginning of
exercies), 49% (in the middle of exercise), and 90% (in
the last 3 min of exercise) of their dynamic PcapI during
tidal inspiration at peak Pes and at high lung volumes.
In the fit group, peak inspiratory Pes reached during
exercise were 37% (beginning), 38% (middle), and 40%
(end) of PcapI.
DISCUSSION
Our data showed that the high-fit group was not
protected by an increased aerobic capacity from exhibit-
Fig. 4. Individual ePdi · f at 3 different times of exercise [20 (A), 60 (B), and 100% (C) of total exercise time]. Total
exercise time 5 15.2 6 1.7 and 17.9 62.6 min for high-fit and fit groups, respectively. ePdi · f and V̇O2 max are
expressed per kilogram of body weight. r, High fit; k, fit.
Downloaded from http://jap.physiology.org/ by guest on November 28, 2012
Fig. 3. Ventilatory and metabolic response to whole
body endurance exercise (92% V̇O2 max) for high-fit (r)
and fit (k) groups. V̇CO2, CO2 production; V̇E, minute
ventilation; VT/TI, mean inspiratory flow rate; ePdi · f,
diaphragm force output; ePes · f, total inspiratory
muscle force output. w Significant difference between
groups, P , 0.05.
AEROBIC FITNESS EFFECTS ON DIAPHRAGM FATIGUE
2161
ing exercise-induced low-frequency diaphragm fatigue
after intense whole body endurance exercise. This is in
contrast to other reports that showed no changes in
maximal volitional tests of inspiratory muscle force
output (6, 11) in highly trained athletes compared with
untrained subjects after severe exercise to exhaustion.
On the basis of this evidence, we had hypothesized that
the high-fit group would not have exhibited a substantial amount of low-frequency diaphragm fatigue after
intense whole body endurance exercise. We reject this
hypothesis now, because the high-fit and fit groups
experienced similar amounts of exercise-induced lowfrequency diaphragm fatigue at all stimulation frequencies immediately after whole body endurance exercise.
Why were the high-fit subjects able to exercise at a
higher absolute workload and a resultant greater ventilatory requirement and not incur a greater level of
low-frequency diaphragm fatigue? There are at least
two possibilities: 1) despite the higher ventilatory
requirement, the high-fit group may have utilized their
diaphragm during endurance exercise to the same
extent as the fit group, and 2) the aerobic capacity of
the diaphragm of the high-fit group was appropriately
increased similar to that of the limb locomotor muscles.
We now discuss each of these possibilities with reference to the current literature.
Diaphragm force production during whole body exercise. Did all the subjects regardless of V̇O2 max experience
the same amount of exercise-induced low-frequency
diaphragm fatigue because the ePdi · f was similar? We
know that the ePdi · f increased four to five times above
rest values in the first few minutes after exercise onset
(2, 3, 18); as exercise continued the ePdi · f declined
slightly, despite further time-dependent increases in
V̇E and ePes · f. The group mean data for ePdi · f during
exercise showed that the high-fit group produced more
force (128%) during the first 60% of the exercise time,
but over the last 40–50% of the exercise diaphragm
force production was not different from the fit group
(Figs. 3 and 4). Thus the higher ventilatory demand in
most high-fit subjects was (with some notable exceptions, see below) dependent on increased recruitment of
accessory inspiratory muscles as exercise time progressed. Because force development by the diaphragm
is an important determinant of exercise-induced diaphragm fatigue (3), this apparent sparing of ePdi · f in
many highly fit subjects would be expected to alleviate
some of the fatigue, despite a higher ventilatory requirement.
We also emphasize that the ventilatory output in the
high-fit subjects was not increased in proportion to
their higher absolute work rate and V̇CO2. That is, the
V̇E-to-V̇CO2 ratio remained lower throughout exercise in
the high-fit group. This reduced ventilatory requirement means that the requirement for force output by
all the inspiratory muscles would also not be increased
in proportion to their higher exercise V̇CO2 in the highly
trained subject. Our study does not address the cause of
this reduced V̇E-to-V̇CO2 ratio in the high-fit athletes,
Downloaded from http://jap.physiology.org/ by guest on November 28, 2012
Fig. 5. Group mean flow-volume (left) and esophageal
pressure-volume loops (right) for high-fit (A) and fit (B)
groups. Flow-volume loops during tidal breathing are
enclosed in maximum volitional flow-volume envelope,
which was determined at rest. Flow-volume and esophageal pressure-volume tidal loops were plotted using
end-expiratory lung volumes measured during rest and
exercise. Each tidal flow-volume and pressure-volume
loop represents ensemble average of 12 fit subjects and
10 high-fit subjects, 20 breaths/subject.
2162
AEROBIC FITNESS EFFECTS ON DIAPHRAGM FATIGUE
We emphasize that our supramaximal BPNS test
does not evaluate all the important characteristics of
diaphragm fatigue. Fatigue has been defined as a
reduction in the force-generating capacity of the muscle
resulting from activity under load that is reversible by
rest (22a). By this definition, most high-fit and fit
subjects have clearly shown low-frequency exerciseinduced diaphragm fatigue. However, the BPNS technique does not provide information on the velocity of
shortening, inasmuch as all stimulations are done at
fixed lung volumes and are assumed to be ‘‘quasiisometric’’ contractions. Therefore, any changes in the
velocity of shortening of the diaphragm resulting from
the whole body endurance exercise remains to be
determined, as do any changes in this important characteristic of ‘‘fatigue’’ between subjects of different
aerobic capacity.
Response of the diaphragm to whole body exercise
training. A second explanation as to why the high-fit
group could exercise at a higher absolute workload and
exhibit the same level of low-frequency diaphragm
fatigue as the fit group may be an enhanced diaphragmatic aerobic capacity in the high-fit group. This
explanation might apply especially to those five high-fit
subjects who clearly generated greater force output of
the diaphragm over most of the duration of the endurance exercise and yet experienced the same amount of
diaphragm fatigue (Fig. 2). We would predict that the
high-fit subjects also had a greater Pdi relative to their
absolute available capacity to generate Pdi throughout
heavy exercise, because the capacity for force development (at any given lung volume or velocity of shortening) was similar in the high-fit and fit subjects, and
during tidal breathing in exercise the lung volumes
were similar and the inspiratory flow rates were higher
in the high-fit group.
Evidence has accumulated in animal models that
supports the idea that the aerobic capacity of the
diaphragm does increase with intense and prolonged
physical training. Three types of changes in response to
whole body physical training have been documented to
occur in the diaphragm: 1) increased oxidative enzyme
activity (14, 16, 24), 2) decreased diffusion distance
from capillaries to muscle due to decreased crosssectional area of type I and type IIa muscle fibers (14,
24, 25, 30), and 3) increased capillary density (15).
Whether the human diaphragm responds in the same
way to whole body physical training or to increased
aerobic capacity has not been documented. The human
studies have shown improvements in volitional ventilatory muscle endurance performance after whole body
physical training, as shown by increased maximal
sustainable ventilation (10, 27) and greater MVV in
trained than in untrained subjects (8–10, 13). On the
other hand, in the present study and in others, no
changes were found between normal-fit and high-fit
subjects in maximal inspiratory pressure generation at
a fixed lung volume (10, 23) or in the pressuregenerating capacity of the inspiratory muscles at any
given flow rate (19).
Downloaded from http://jap.physiology.org/ by guest on November 28, 2012
although a similar finding has been reported with
short-term heavy exercise and attributed to reduced
levels of metabolic acidosis (21).
What does this leveling off of the ePdi · f mean? It
may be indicative of a changing recruitment pattern of
the inspiratory muscles during intense endurance exercise in response to the onset of diaphragm fatigue.
Sieck and Fournier (28) found that the recruitment
pattern of the diaphragm muscle fibers followed the
size principle, inasmuch as the most fatigue-resistant
type I fibers were recruited at low ventilatory loads and
moderately fatigue-resistant type IIa fibers were recruited at moderate loads. The least fatigue-resistant
type IIb fibers were only recruited for nonventilatory
behaviors such as sneezing or gagging. Thus the very
high ventilatory requirements during exercise requiring greater and greater force development from the
diaphragm might make this muscle more susceptible to
fatigue and compromise its role as the major inspiratory muscle (see below). It is appropriate, then, that
diaphragm force production is inhibited perhaps by
feedback inhibition (17) and accessory inspiratory and
expiratory muscles are recruited during prolonged exercise. Similar patterns of selective recruitment of accessory muscles and derecruitment of the diaphragm have
been observed in animals undergoing severe resistive
loading (4). Furthermore, during prolonged exercise,
the diaphragm (along with the accessory inspiratory
muscles) may act as a significant force generator to
displace the lungs and chest wall, but the diaphragm,
with its longer fibers, mixed fiber types, convex shape,
and high aerobic capacity, is ideally suited to serve as a
major generator of high velocities of shortening and
flow rates. Hence, during the latter stages of the
endurance exercise, the diaphragm may actually further increase its velocity of shortening and make
greater contributions to increasing flow rate at a time
when its relative contribution to force development is
decreasing.
Does the diaphragm force production level off in
prolonged exercise because the diaphragm cannot produce more force or because it will not produce the force
it is capable of producing? Theoretically, at the higher
flow rates achieved during exercise, the velocity of
shortening of the diaphragm would also be quite high
and thus would compromise the muscle’s capability for
force production (1). However, we think this is unlikely,
because we have shown (3) that resting subjects were
able to volitionally increase and maintain diaphragm
force production 78% above the levels needed during
endurance exercise and to sustain this force for the
same time period as the exercise. Therefore, the diaphragm was capable of producing and sustaining a
force production and velocity of shortening at rates that
were well above those experienced during exercise
without experiencing task failure. Whether the diaphragm force production could be increased and sustained this much during whole body exercise when
other fatiguing influences are present (3) remains to be
determined.
AEROBIC FITNESS EFFECTS ON DIAPHRAGM FATIGUE
reduction in force output of a single, albeit primary,
inspiratory muscle clearly did not cause global respiratory muscle task failure or inadequate alveolar ventilation (2, 3, 18).
In a recent study in which the inspiratory muscles
were partially mechanically unloaded, no differences
were found in V̇E or exercise performance time between
normal and unloaded endurance exercise (22). Assuming that this partial unloading may have alleviated
diaphragm fatigue, the authors interpreted these data
to show that inspiratory muscle fatigue had no effect on
the ventilatory response or breathing pattern during
heavy endurance exercise. We agree and would speculate that the significant consequence of the exerciseinduced diaphragm fatigue might be in providing a
reordering of the pattern of respiratory muscle recruitment.
Finally, the fact that the diaphragm and inspiratory
muscles do show significant increases in aerobic capacity in response to physical training also distinguishes
the chest wall from the lung in terms of malleability.
These chronic adaptations mean that the ratio of
demand to capacity in the diaphragm (during exercise)
remains about the same in trained and untrained
subjects. Training effects on the respiratory muscles
may also be important in preventing a more marked
exercise-induced respiratory muscle fatigue and perhaps even task failure, especially given the very high
ventilatory requirements faced in the highly trained
during high-intensity endurance exercise.
This study was funded by the National Heart, Lung, and Blood
Institute. M. A. Babcock is a Parker B. Francis Fellow of Pulmonary
Research.
Present address of B. D. Johnson: Mayo Clinic, Rochester, MN
55905.
Address for reprint requests: M. A. Babcock, Dept. of Preventive
Medicine, 504 N. Walnut St., Madison, WI 53705.
Received 12 May 1995; accepted in final form 31 May 1996.
REFERENCES
1. Agostoni, E., and W. O. Fenn. Velocity of muscle shortening as
a limiting factor in respiratory air flow. J. Appl. Physiol. 15:
349–353, 1960.
2. Babcock, M. A., B. D. Johnson, D. F. Pegelow, O. E. Suman,
D. Griffin, and J. A. Dempsey. Hypoxic effects on exerciseinduced diaphragmatic fatigue in normal healthy humans. J.
Appl. Physiol. 78: 82–92, 1995.
3. Babcock, M. A., D. F. Pegelow, S. R. McClaran, O. E. Suman,
and J. A. Dempsey. Contribution of diaphragmatic power
output to exercise-induced diaphragm fatigue. J. Appl. Physiol.
78: 1710–1719, 1995.
4. Bazzy, A. R., and G. G. Haddad. Diaphragmatic fatigue in
unanesthetized adult sheep. J. Appl. Physiol. 57: 182–190, 1984.
5. Bellemare, F., and B. Bigland-Ritchie. Assessment of human
diaphragm strength and activation using phrenic nerve stimulation. Respir. Physiol. 58: 263–277, 1984.
6. Bender, P. R., and B. J. Martin. Maximal ventilation after
exhausting exercise. Med. Sci. Sports Exercise 17: 164–167,
1985.
7. Borg, G. A. V. Psychophysical bases of perceived exertion. Med.
Sci. Sports Exercise 14: 377–381, 1982.
8. Boutellier, U., R. Buchel, A. Kundert, and C. Spengler. The
respiratory system as an exercise limiting factor in normal
trained subjects. Eur. J. Appl. Physiol. Occup. Physiol. 65:
347–353, 1992.
Downloaded from http://jap.physiology.org/ by guest on November 28, 2012
We are uncertain whether increased aerobic capacity
of the diaphragm alone can explain our findings of
similar diaphragm fatigue, despite elevated ePdi · f in
many of the high-fit subjects, because we know that
more than just ePdi · f per se causes exercise-induced
diaphragm fatigue (3). Other factors such as blood flow
distribution to diaphragm vs. locomotor muscle during
endurance exercise and circulating metabolites produced by the locomotor muscles also contribute, and
these factors might also very well be different in the
highly fit subject.
Pulmonary system limitations in the high-fit group:
demand vs. capacity. Key determinants of the pulmonary system’s capacity for maximum gas transport
during exercise include alveolar-capillary diffusion surface, the flow-volume maximum envelope, and aerobic
capacity of the respiratory muscles. All these functions
are placed under considerable stress during heavy
exercise in the high-fit subjects but appear to have
quite different susceptibilities to reaching limitation
because of their different capacities and degree of
malleability in response to physical training. In the
case of diffusion limitation, many very highly fit humans (1.5–2 times normal V̇O2 max) show significant
exercise-induced hypoxemia, presumably because their
extraordinary demand for O2 transport is not matched
by enhanced diffusion surface area in the lung (12).
This hypoxemia presents a significant limitation to
systemic O2 transport and to V̇O2 max (26). Similarly, the
maximum flow-volume envelope is also unaltered in
most highly fit subjects (19). Accordingly, with their
high metabolic and ventilatory requirements, the highly
fit individuals experience significant expiratory flow
limitation and increased ventilatory work, even in
moderately heavy exercise (Fig. 5), and many subjects
will show complete mechanical flow limitation to ventilation at maximum exercise (19). Incurring these high
mechanical loads during exercise may contribute to
exercise limitation in the high-fit subjects, perhaps via
high metabolic and blood flow requirements by the
respiratory muscles or by mechanical constraint on
alveolar ventilation. However, the actual contribution
of these factors to exercise and ventilatory limitation
remains controversial and unresolved.
Exercise-induced diaphragm fatigue as a third potential pulmonary system limitation presents quite differently in these comparisons of demand to capacity than
do the diffusion or flow-volume limitations. First,
healthy young adult subjects of all fitness levels and
V̇O2 max tested to date experienced significant exerciseinduced diaphragm fatigue (as shown by a reduced
force production in response to BPNS), so long as
exercise was of sufficient intensity and duration (Fig.
2). In other words, there was no specific threshold of
V̇O2 max, ventilatory requirement, or ePdi · f during the
heavy endurance exercise below which diaphragm fatigue did not occur in healthy subjects (Fig. 1). In a
sense then, the demand for sustained force output by
the diaphragm during exercise exceeded the muscle’s
aerobic capacity in all healthy subjects. However, this
2163
2164
AEROBIC FITNESS EFFECTS ON DIAPHRAGM FATIGUE
21. Jones, N. L. Determinants of breathing pattern. In: Exercise:
Pulmonary Physiology and Pathophysiology, edited by K. Wasserman and B. J. Whipp. New York: Dekker, 1991, vol. 52, p. 99–119.
(Lung Biol. Health Dis. Ser.)
22. Marciniuk, D., D. Mckim, R. Sanii, and M. Younes. Role of
central respiratory muscle fatigue in endurance exercise in
normal subjects. J. Appl. Physiol. 76: 236–241, 1994.
22a. National Heart, Lung, and Blood Institute Workshop.
Respiratory muscle fatigue: report of the respiratory muscle
fatigue workshop group. Am. Rev. Respir. Dis. 142: 474–480,
1990.
23. O’Kroy, J. A., and J. R. Coast. Effects of flow and resistive
training on respiratory muscle endurance and strength. Respiration 60: 279–283, 1993.
24. Powers, S. K., D. Criswell, J. Lawler, D. Martin, L. L. Ji,
R. A. Herb, and G. Dudley. Regional training-induced alterations in diaphragmatic oxidative and antioxidant enzymes.
Respir. Physiol. 95: 227–237, 1994.
25. Powers, S. K., D. Criswell, F.-K. Lieu, S. Dodd, and H.
Silverman. Diaphragmatic fiber type specific adaptation to
endurance exercise. Respir. Physiol. 89: 195–207, 1992.
26. Powers, S. K., J. Lawler, J. A. Dempsey, S. Dodd, and G.
Landry. Effects of incomplete pulmonary gas exchange on
V̇O2 max. J. Appl. Physiol. 66: 2491–2495, 1989.
27. Robinson, E. P., and J. M. Kjeldgaard. Improvement in
ventilatory muscle function with running. J. Appl. Physiol. 52:
1400–1406, 1982.
28. Sieck, G. C., and M. Fournier. Diaphragm motor unit recruitment during ventilatory and nonventilatory behaviors. J. Appl.
Physiol. 66: 2539–2545, 1989.
30. Tamaki, N. Effect of endurance training on muscle fiber composition and capillary supply in rat diaphragm. Eur. J. Appl.
Physiol. Occup. Physiol. 56: 127–131, 1987.
31. Yan, S., T. Similowski, A. P. Gauthier, P. T. Macklem, and F.
Bellemare. Effect of fatigue on diaphragmatic function at
different lung volumes. J. Appl. Physiol. 72: 1064–1067, 1992.
Downloaded from http://jap.physiology.org/ by guest on November 28, 2012
9. Boutellier, U., and P. Piwko. The respiratory system as an
exercise limiting factor in normal sedentary subjects. Eur. J.
Appl. Physiol. Occup. Physiol. 64: 145–152, 1992.
10. Clanton, T. L., G. F. Dixon, J. Drake, and J. E. Gadek. Effects
of swim training on lung volumes and inspiratory muscle conditioning. J. Appl. Physiol. 62: 39–46, 1985.
11. Coast, J. R., P. S. Clifford, T. W. Henrich, J. StrayGunderson, and R. L. Johnson, Jr. Maximal inspiratory
pressure following maximal exercise in trained and untrained
subjects. Med. Sci. Sports Exercise 22: 811–815, 1990.
12. Dempsey, J. A. Is the lung built for exercise? Med. Sci. Sports
Exercise 18: 143–155, 1986.
13. Fairburn, M. S., K. C. Coutts, R. L. Pardy, and D. C.
McKenzie. Improved respiratory muscle endurance of highly
trained cyclists and the effects on maximal exercise performance.
Int. J. Sports Med. 12: 66–70, 1991.
14. Green, H. J., M. Plyley, D. M. Smith, and J. G. Kile. Extreme
endurance training and fiber type adaptation in rat diaphragm.
J. Appl. Physiol. 66: 1914–1920, 1989.
15. Hoppeler, H., H. Howald, K. E. Conley, S. J. Lindstedt, H.
Claassen, P. Vock, and E. R. Weibel. Endurance training in
humans: aerobic capacity and structure of skeletal muscle. J.
Appl. Physiol. 59: 320–327, 1985.
16. Ianuzzo, C. D., E. G. Noble, N. Hamilton, and B. Dabrowski.
Effects of streptozotocin diabetes, insulin treatment, and training on the diaphragm. J. Appl. Physiol. 52: 1471–1475, 1986.
17. Jammes, Y., B. Buchler, S. Delpierre, A. Rasidakis, C.
Grimard, and C. Roussos. Phrenic afferents and their role in
inspiratory control. J. Appl. Physiol. 60: 854–860, 1986.
18. Johnson, B. D., M. A. Babcock, O. E. Suman, and J. A.
Dempsey. Exercise-induced diaphragmatic fatigue in healthy
humans. J. Physiol. Lond. 460: 385–405, 1993.
19. Johnson, B. D., K. W. Saupe, and J. A. Dempsey. Mechanical
constraints on exercise hyperpnea in endurance athletes. J.
Appl. Physiol. 73: 874–886, 1992.
20. Johnson, B. D., K. C. Seow, D. F. Pegelow, and J. A.
Dempsey. Adaptation of the inert gas FRC technique for use in
heavy exercise. J. Appl. Physiol. 68: 802–809, 1990.