Physical Fitness and Performance
Inspiratory muscle training improves rowing
performance
STEFANOS VOLIANITIS, ALISON K. MCCONNELL, YIANNIS KOUTEDAKIS, LARS MCNAUGHTON,
KARRIANNE BACKX, and DAVID A. JONES
School of Sport and Exercise Sciences, The University of Birmingham, Edgbaston, Birmingham B15 2TT, UNITED
KINGDOM; School of Health Sciences, University of Wolverhampton, Wolverhampton WV1 1SB, UNITED KINGDOM;
and School of Life Sciences, Kingston University, Kingston upon Thames, Surrey KT1 2EE, UNITED KINGDOM
ABSTRACT
VOLIANITIS, S., A. K. MCCONNELL, Y. KOUTEDAKIS, L. MCNAUGHTON, K. BACKX, and D. A. JONES. Inspiratory muscle
training improves rowing performance. Med. Sci. Sports Exerc., Vol. 33, No. 5, 2001, pp. 803–809. Purpose: To investigate the effects
of a period of resistive inspiratory muscle training (IMT) upon rowing performance. Methods: Performance was appraised in 14 female
competitive rowers at the commencement and after 11 wk of inspiratory muscle training on a rowing ergometer by using a 6-min all-out
effort and a 5000-m trial. IMT consisted of 30 inspiratory efforts twice daily. Each effort required the subject to inspire against a
resistance equivalent to 50% peak inspiratory mouth pressure (PI
max
) by using an inspiratory muscle training device. Seven of the
rowers, who formed the placebo group, used the same device but performed 60 breaths once daily with an inspiratory resistance
equivalent to 15% PI
max
. Results: The inspiratory muscle strength of the training group increased by 44 ⫾ 25 cm H
2
O (45.3 ⫾ 29.7%)
compared with only 6 ⫾ 11 cm H
2
O (5.3 ⫾ 9.8%) of the placebo group (P ⬍ 0.05 within and between groups). The distance covered
in the 6-min all-out effort increased by 3.5 ⫾ 1.2% in the training group compared with 1.6 ⫾ 1.0% in the placebo group (P ⬍ 0.05).
The time in the 5000-m trial decreased by 36 ⫾ 9 s (3.1 ⫾ 0.8%) in the training group compared with only 11 ⫾ 8 s (0.9 ⫾ 0.6%)
in the placebo group (P ⬍ 0.05). Furthermore, the resistance of the training group to inspiratory muscle fatigue after the 6-min all-out
effort was improved from an 11.2 ⫾ 4.3% deficit in PI
max
to only 3.0 ⫾ 1.6% (P ⬍ 0.05) pre- and post-intervention, respectively.
Conclusions: IMT improves rowing performance on the 6-min all-out effort and the 5000-m trial. Key Words: RESPIRATORY
MUSCLE TRAINING, PERFORMANCE ENHANCEMENT, INSPIRATORY MOUTH PRESSURE, RESPIRATORY FATIGUE,
DYSPNEA
H
istorically, exercise performance has not been con-
sidered to be limited by ventilation or respiratory
muscle function. However, occurrence of respira-
tory muscle fatigue after prolonged submaximal exercise
(23), as well as short-term maximal exercise (19,25), has
suggested that the ventilatory system might contribute to
exercise limitation. Some studies in which the inspiratory
muscles were partially unloaded during prolonged exercise,
and respiratory muscle fatigue was supposedly alleviated,
reported no effect on ventilation or exercise performance
(11,20), whereas other studies show significant improve-
ments in performance (14,15).
In addition, several studies in recent years have examined
the effects of specific respiratory muscle training upon ex-
ercise performance, but the literature is inconclusive; some
have shown improvements (4,5,30), whereas others show no
effect on performance (13,24). The discrepancies between
studies may reflect differences in the exercise intensities and
durations used for testing, as well as differences in experi-
mental design and fitness level of the subjects.
Rowing is a sport requiring large aerobic power and a
high minute ventilation, typically greater than 200 L·min
-1
in elite males (26). Peak expiratory flow rates can reach
values up to 15 L·s
-1
in elite male rowers (7). The entrain
-
ment of breathing in rowing (31) places additional demands
on the respiratory muscles, which must stabilize the thorax
during the stroke, as well as bringing about breathing related
excursions of the thorax. If respiratory muscle fatigue oc-
curs during competitive rowing, it might be of physiological
significance to the regulation of ventilation and breathing
pattern, and to respiratory muscle recruitment and hence
respiratory sensation. Furthermore, an alteration of the re-
cruitment pattern could have an effect on the mechanical
efficiencies of breathing and rowing, with detrimental con-
sequences for performance.
In view of the unique respiratory demands of rowing and
the discrepancies in the literature with regard to the benefits
of inspiratory muscle training, this study investigated the
effect of inspiratory muscle training upon rowing
performance.
0195-9131/01/3305-0803/$3.00/0
MEDICINE & SCIENCE IN SPORTS & EXERCISE
®
Copyright © 2001 by the American College of Sports Medicine
Received for publication December 1999.
Accepted for publication July 2000.
803
METHODS
Subjects. Fourteen female competitive rowers (mean ⫾
SD, age 23.8 ⫾ 3.8 yr, height 173.4 ⫾ 3.8 cm, weight 68.2
⫾ 4.6 kg, maximal oxygen uptake (V
˙
O
2max
) 3.56 ⫾ 0.17
L·min
-1
, maximal power output (P
max
229 ⫾ 22 W)) were
assigned randomly to either an inspiratory muscle training
(IMT) or placebo group. The subjects were informed about
the nature and risks involved in participation in the exper-
iments. The experimental protocol was approved by the
local ethics committee, and all subjects acknowledged vol-
untary participation through written informed consent. The
subjects were instructed to adhere to their usual diet and not
to engage in strenuous activity the day before an exercise
test. On test days, the subjects were asked not to drink coffee
or other caffeine-containing beverages. The tests were per-
formed at similar times of the day. The initial performance
assessment took place at the end of October, which is the
first month of the preparatory period of the rowing season.
All the subjects where either national team members or
candidates for the national team and had been competing for
a minimum of 3–4 yr.
Procedure. At the beginning of the study, the subjects
performed a submaximal incremental load test followed by
a 6-min all-out test on a rowing ergometer (model c, Con-
cept II, Nottingham, UK). On the same occasion, baseline
spirometry values and maximum respiratory mouth pres-
sures were taken before and after the rowing tests. Both
groups commenced an 11-wk period of inspiratory muscle
training. The effects of the intervention were evaluated, with
the same battery of tests, at 4 wk and after completion of the
training period. Mouth pressure measurements, for evalua-
tion of respiratory muscle function during rowing, took
place on all occasions. The maneuvers were performed
within 30 s after the completion of the maximum effort.
Submaximal incremental load test. The test proto-
col consisted of five stages of 4 min each with a 1-min
interruption for blood sampling. The initial work rate was
individualized based on known work capacity. The rowers
where asked to start rowing with a frequency of 18
strokes·min
-1
at a work rate that they usually perform their
daily warm-up. The work rate increments for each subse-
quent stage was 20 or 25 W, depending on the rower’s
capacity. Once the protocol for a particular rower was es-
tablished at the beginning of the study, it was not varied
thereafter. Heart rate was monitored via a short-range te-
lemetry system (Polar Sporttester, Polar Electro, Kempele,
Finland). A preexercise and poststage blood sample was
collected from the earlobe and analyzed for lactate concen-
tration. Stroke ratings (st·min
-1
) and power output (W) were
recorded for each stage. Continuous analysis of expired
gases and static spirometry (flow-volume loops) were per-
formed with an Oxycon Alpha diagnostic system (Jaeger
b.v., Manheim, Germany).
Maximal performance tests. After the submaximal
incremental load test, the rowers performed a 6-min all-out
effort, which is a simulation of the competitive rowing
duration. Rowing events last between 5.5 and 7.5 min,
depending on boat type, category, and gender of the rowers.
We chose 6-min for our test as it represents the duration of
the women’s eight events. The rest period between the
submaximal test and the 6-min test was standardized at
8–10 min to minimize any fatiguing effect of the submaxi-
mal test but at the same time to maintain readiness of the
rowers. Additional performance data have been obtained at
baseline and after 4 wk of inspiratory muscle training by
means of a 5000 m ergometer trial that the subjects per-
formed as part of their training control.
Maximum inspiratory pressure measurement.
Maximal static inspiratory mouth pressure (PI
max
)iscom
-
monly used to measure inspiratory muscle strength. A por-
table hand held mouth pressure meter (Precision Medical,
London, United Kingdom) was used for this measurement.
This device has been shown to measure inspiratory and
expiratory pressures accurately and reliably (12). A mini-
mum of five technically satisfactory measurements were
conducted and the highest of three measurements with less
than 5% variability or within 5 cm H
2
O(1kPa⫽ 10.3 cm
H
2
O) difference was defined as maximum (34). The initial
length of the inspiratory muscles was controlled by initiat-
ing each effort from residual volume (RV). This procedure
was adopted because, from our experience, RV is more
reproducible than functional residual capacity (FRC). Sub-
jects were instructed to take their time and to empty their
lungs slowly to RV, thereby avoiding problems associated
with variability in lung volumes and dynamic airway com-
pression. All maneuvers were performed in the upright
standing position, and verbal encouragement was given to
help the subjects perform maximally. The subjects had been
familiarized with the nature of the maneuvers to reduce any
learning effect.
Respiratory muscle fatigue. For practical purposes,
“fatigue” was defined as the inability to continue to generate
a given pressure with the same motor command as when the
muscle was still fresh. A condition like this does not nec-
essarily imply any “task failure” in the form of inadequate
pressure generation for the required ventilation, but it is an
indication that the functional capacity is compromised and it
will eventually lead to “task failure.” Therefore, the original
definition of Edwards (9) of skeletal muscle fatigue as a
“failure to maintain the required or expected force” has been
extended for respiratory fatigue to include also the state of
muscle weakness (27).
Perception of dyspnea. A category scale, the modi-
fied Borg (3) scale, was chosen to evaluate the respiratory
effort during exercise. The scale consisted of a series of
integers from 0 to 10. The rower was asked to estimate the
effort required to breathe but not the effort of the exercise.
During rowing, the Borg scale remained in front of the
rower and an assessment was made at the end of every stage
and after the all-out effort.
Inspiratory muscle training. The training group per-
formed 30 inspiratory efforts twice daily. Each effort re-
quired the subject to inspire against a resistance equivalent
to 50% peak inspiratory mouth pressure (PI
max
) by using an
inspiratory muscle trainer (POWERbreathe®, IMT Tech-
804
Official Journal of the American College of Sports Medicine http://www.acsm-msse.org
nologies Ltd., Birmingham, UK). POWERbreathe® is a
pressure-threshold device that requires continuous applica-
tion of inspiratory pressure throughout inspiration for the
inspiratory regulating valve to remain open while it allows
unrestricted expiration. Subjects were instructed to initiate
each breath from RV and to continue the inspiratory effort
up to the lung volume where the inspiratory muscle force
output for the given load limited further excursion of the
thorax. Because of the increased tidal volume, a decreased
breathing frequency was adopted to avoid hyperventilation
and the consequent hypocapnia. Previous studies from our
lab (6) have suggested that the protocol used by the training
group is successful in eliciting an adaptive response. The
placebo group trained using the same device, but they per-
formed 60 breaths once daily, at a resistance to inspiration
equivalent to 15% PI
max
, a load known to elicit a negligible
training effect (5). The two seemingly different training
protocols were designed to maintain the naivety of the
subjects who were told that one group was training for
strength and the other for endurance of the inspiratory
muscles. All subjects kept a training diary recording their
adherence to the program. Each of the two daily sessions of
the training group lasted approximately 5 min, whereas the
single training session of the placebo group lasted approx-
imately 10–12 min, depending on the breathing frequency
that each subject adopted.
Blood lactate. Arterialized capillary blood samples
were taken from the ear lobe before the incremental load test
and at the end of each stage. Analysis was done with an
Analox GM7 (London, UK). The within-run precision was
1.6% at a whole blood lactate concentration of 5.0
mmol·L
-1
. At low levels of lactate concentration, measure
-
ment errors exceeding ⫾ 0.2 mmol·L
-1
were rare. Thus, a
measured rise of more than 0.4 mmol·L
-1
during the course
of a progressive test was likely to represent a real increase
in lactate concentration.
Statistical analysis. Results were analyzed using non-
parametric repeated measures analysis of variance (Fried-
man’s test) and Wilcoxon signed ranks test for intra- and
inter-group comparisons, respectively. Probability values of
less than 0.05 were considered significant. All results are
expressed in means ⫾ SD unless otherwise stated.
RESULTS
Respiratory Muscle Function: PI
max
After the initial 4 wk of the training period, PI
max
in
-
creased by 40 ⫾ 25 cm H
2
O (40.7 ⫾ 25.1%; P ⬍ 0.01) and
by 5 ⫾ 6cmH
2
O (4.6 ⫾ 6.0%; P ⫽ 0.083) from baseline,
in the IMT and placebo groups, respectively. After 11 wk of
IMT, PI
max
increased slightly more to a total increase of 44
⫾ 25 cm H
2
O (45.3 ⫾ 29.7%; P ⬍ 0.01) and 6 ⫾ 11 cm
H
2
O (5.3 ⫾ 9.8%; P ⫽ 0.21) from baseline, in the IMT and
placebo groups, respectively (see Table 1.) The PI
max
im
-
provements of the training group, expressed in percentage,
were significantly different both between groups and across
time within the group. Analysis of the training diaries re-
vealed that both groups compliance with the prescribed
training was between 96–97%.
Rowing Performance
6 min all-out. After the first 4 wk of the training period
the performance in the 6-min all-out test improved, from
baseline, by 3.4 ⫾ 1.0% (P ⬍ 0.05) in the IMT group, and
by 1.1 ⫾ 0.4% (P ⬍ 0.05) in the placebo group. Upon
completion of the training period, performance had in-
creased from baseline a total of 3.5 ⫾ 1.2% (P ⬍ 0.05) in
the IMT group and 1.6 ⫾ 1.0% (P ⬍ 0.05) in the placebo
group from their baseline values (see Table 1.). These im-
provements were also significantly different between the
two groups after 4 wk (P ⬍ 0.05) and after 11 wk (P ⬍
0.05).
5000 m. The time for the completion of the 5000-m test,
after the first 4 wk of IMT, decreased by 36 ⫾ 9 s (3.1 ⫾
0.8%; P ⬍ 0.05), whereas the placebo group’s time de-
creased by 11 ⫾ 8 s (0.9 ⫾ 0.6%; P ⬍ 0.05). The difference
in the improvement between the two groups was also sig-
nificant (P ⬍ 0.05). There were no data available for the
5000-m test upon completion of the 11-wk IMT period.
Lactate
After 4 wk of inspiratory muscle training blood lactate
was lower relative to baseline values by 0.3 ⫾ 0.3 mmol·L
-1
(P ⬍ 0.05) in the third stage and 1.3 ⫾ 1.3 mmol·L
-1
(P ⬍
0.05) in the fifth stage of the submaximal incremental test
for the IMT group. Even though there was also a decreasing
trend in the placebo group, it did not reach significance (P ⫽
0.11, in the fifth stage). In the interval between the 4th and
11th week of inspiratory muscle training blood lactate de-
creased a further 0.37 ⫾ 0.32 mmol·L
-1
(P ⬍ 0.05) in the
IMT group at the second stage of the incremental test with
no significant changes in the placebo group. Overall, both
IMT and placebo groups had a significant decrease in lactate
of 1.3 ⫾ 1.47 mmol·L
-1
and 1.3 ⫾ 1.2 mmol·L
-1
, respec
-
tively (P ⬍ 0.05) in the fifth stage of the incremental test.
There was no significant difference between the groups. No
changes occurred in the blood lactate response to the 6-min
all-out effort throughout the study.
Respiratory Muscle Fatigue
Baseline fatigue, defined as the decrease of maximum
mouth pressure generating capacity, after the baseline 6-min
all-out rowing effort, was 11.2 ⫾ 2.6% (P ⬍ 0.05) and 11.1
TABLE 1. PI
max
in centimeters of H
2
O (mean ⫾ SE), and performance, in meters
(m), during the 6-min all-out rowing effort for the training (IMT) and placebo
groups, throughout the 11 wk of inspiratory muscle training.
PI
max
(cm H
2
O)
Performance (m)
IMT Placebo IMT Placebo
Baseline 104 ⫾ 8 130 ⫾ 12 1561 ⫾ 9.3 1566 ⫾ 20.7
4 wk 144 ⫾ 10** 135 ⫾ 11 1613 ⫾ 12.2** 1582 ⫾ 21.4*
11 wk 148 ⫾ 10** 136 ⫾ 12 1616 ⫾ 13.4** 1592 ⫾ 21.1**
*Significantly different from baseline (
P
⬍ 0.05); **significantly different from baseline
(
P
⬍ 0.01).
INSPIRATORY MUSCLE TRAINING IN ROWING Medicine & Science in Sports & Exercise
姞
805
⫾ 0.8% (P ⬍ 0.05) for the IMT and the placebo groups,
respectively. After the first 4 wk of the training period the
fatigue after the 6-min all-out effort in the IMT group
decreased to 3.1 ⫾ 1.1% (P ⬍ 0.01), whereas the placebo
group remained at 10.7 ⫾ 2.8%. Upon completion of the
training period the fatigue for the IMT and the placebo
groups did not change any further (4.5 ⫾ 4.7%, P ⬍ 0.01
and 10.7 ⫾ 2.2%, NS, respectively). Between-group differ-
ences in fatigue where also significant for both the 4 and 11
wk comparisons (P ⬍ 0.05; see Fig. 1).
Perception of Dyspnea
Significant improvements in the perception of respiratory
effort during the incremental test were found in the IMT
group throughout the training period (Fig. 2). However, no
change was found in the dyspnea after the 6-min all-out
effort. There were no significant changes in the control
group either during the incremental test or the 6-min all-out
effort (Fig. 2).
Ventilation and Breathing Pattern
After the completion of the training period, there were no
significant changes in the ventilatory volumes at any stage
of the incremental test, for either the IMT or the placebo
group. However, during the 6-min all-out effort, minute
ventilation increased for the placebo group, from a baseline
of 120.3 ⫾ 18.5 to 129.6 ⫾ 13.4 L·min
-1
(P ⬍ 0.05). The
IMT group also increased minute ventilation from a baseline
of 119.9 ⫾ 12.8 to 122.5 ⫾ 12.3 L·min
-1
, a difference that
just failed to reach significance (P ⫽ 0.051). The breathing
pattern of the IMT group at the 6-min all-out effort changed
after the completion of the training period. There was a shift
to a significantly deeper breathing pattern with an increase
of the tidal volume from 2.01 ⫾ 0.16 to 2.16 ⫾ 0.16 L (P
⬍ 0.01). Breathing frequency did not change significantly.
The placebo group did not exhibit any significant changes in
breathing pattern, but there was a tendency toward a more
tachypneic pattern with an increase of 4.5% in their breath-
ing frequency compared with only 1.5% of the IMT group
(see Table 2).
DISCUSSION
The most important finding of this study is that inspira-
tory muscle training improved rowing performance to a
greater extent than conventional training alone. To our
knowledge, ours is the only study investigating the effect of
inspiratory muscle training upon an index of sports perfor-
mance rather than a marker of physiological capacity such
as the time-limit test (Tlim). In the reports of Caine and
McConnell (5) and Lisboa et al.(22), cycling time to ex-
FIGURE 2—Dyspnea-power curves after 4 and 11 wk of inspiratory
muscle training for the IMT (top) and placebo (bottom) groups. Values
are means ⴞ SD. * Significantly different (P < 0.05); ** significantly
different (P < 0.01); ⴙ significantly different (P < 0.05) power output
for the same dyspnea.
TABLE 2. A summary of statistical significance for within- and between-group
comparisons after 11 wk of IMT in selected parameters.
Parameter IMT Group
Placebo
Group
Between-Group
Comparisons
Resting PI
max
Improved No change Yes
PI
max
after exercise
Improved No change Yes
Lactate incremental test Decreased Decreased No difference
Borg scale 6-min test Decreased No change No difference
??
E
Incremental test No change No change No difference
??
E
6 min test No change Increased No difference
??
T
6 min test Increased No change No difference
Bf 6 min test No change No change No difference
P
ETCO
2
No change No change No difference
P
ETO
2
Increased No change Yes
6-min test power Improved Improved Yes
5000-m trial time Improved Improved Yes
FIGURE 1—Decrement in inspiratory muscle strength after the 6-min
all-out test, in percentage decrease from resting mouth pressure gen-
erating capacity, throughout the 11 wk of inspiratory muscle training
in the training and placebo groups. Values are mean ⴞ SD ** P < 0.01
different from the placebo group. IMT, inspiratory muscle training
group.
806
Official Journal of the American College of Sports Medicine http://www.acsm-msse.org
haustion, a 6-min walk, or an incremental test are used for
evaluation of exercise tolerance. Even though the 6-min
walk might be argued to be representative of a task encoun-
tered by patients with COPD, it is still not a simulation of
any known sport. In contrast, the 6-min and the 5000-m time
trial represent very close simulations of competitive rowing
events and are therefore one step closer to actual sports
performance than any test attempted in previous studies.
Because the early report of Leith and Bradley (21), many
different groups have demonstrated that ventilatory muscle
training increases maximal voluntary ventilation, ventila-
tory muscle strength, ventilatory muscle endurance, and
functional exercise capacity. Our results of 45.3% improve-
ment in PI
max
are similar in magnitude to other studies (see
Smith and colleagues’ (29) meta-analysis on patients with
COPD) ranging from 32% to 53% (22,32,33). However, be-
cause there may well be important differences between healthy
subjects and those with COPD, a more appropriate comparison
would be with studies using healthy subjects (5,13) where
PI
max
also increases in the range of 34% to 45.3%, respectively,
after 4 wk of inspiratory muscle training.
Previous reports (4,5,30) have shown that after inspira-
tory muscle training a submaximal power output can be
maintained for longer (Tlim test). However, the intensity
used for the Tlim test in these studies was associated either
with the anaerobic threshold (Th
an
) or the maximum lactate
steady state (MLSS). Even though these physiological
markers correlate very well with endurance performance,
this approach is one step removed from competitive sports
performance. Our study shows that inspiratory muscle train-
ing can improve performance in two tests that simulate
competitive performance as closely as possible in the lab-
oratory context, viz. the 6-min all-out effort and the 5000-m
trial. Both tests are routinely used for rowing-specific per-
formance evaluation by coaches. Both IMT and placebo
groups improved their performance after 11 wk of training.
The margin of their improvement was expected because the
study commenced at the beginning of the preparatory train-
ing period and lasted for the bigger part of it. Even though
we acknowledge the possibility that the responses observed
may have occurred as a result of the subjects’ regular
training, the 1.9% improvement of the IMT group in the
6-min all-out effort over and above the improvement of the
placebo group suggests that this is unlikely. Therefore, the
data suggest that the inspiratory muscle training had an
additional effect upon rowing performance beyond that ex-
pected by regular training. The significance of this differ-
ence can be appreciated more within the context of com-
petitive rowing where Olympic medals are decided with a
much smaller margin than 1.9%.
We believe that there are a number of reasons why other
studies have not reported any significant improvements in
performance after IMT. Arguably, the most important of which
is the low reliability of the tests used to evaluate performance
in other studies, compared with the 6-min all-out effort used in
our study, made the detection of a meaningful effect difficult.
For example, the coefficient of variation for the Tlim test has
been reported to be anything between 25% and 40%, whereas
the 6-min all out test is only 2.4% (17). Therefore, much larger
improvements were required to assure that the observations
were not due to the variability of the test itself. Other studies
(13,24) have reported improvements in performance but failed
to reach significance. We suspect that insufficient statistical
power, due to the small sample size of these studies, may have
introduced a type II error and failed to reject the null hypoth-
esis. Support of our findings is provided by studies using
isocapnic hyperpnea training protocols, which suggest that
respiratory muscle training induces significant improvements
in cycling performance (Tlim) (4,30). In addition, a recently
completed study showed that after 5 wk of respiratory muscle
training, using a high velocity (flow) and a high resistance
(pressure) training protocol, cycling time trials improved sig-
nificantly by approximately 5% (J. Dempsey, personal com-
munication). In the absence of any clear insight into the hard
evidence of the underlying physiological mechanisms for the
observed effects, we are forced to speculate on possible mech-
anisms, three of which are discussed below.
Respiratory Muscle Fatigue
First, even though respiratory muscle fatigue of the IMT
group was diminished, there was no evidence for signifi-
cantly different ventilatory response between the two
groups. These data support the notion that respiratory mus-
cle fatigue was without significant consequence for the
ventilatory response. This is consistent with the suggestion
that when the diaphragm is confronted by fatiguing contrac-
tion patterns, the accessory inspiratory muscles become
more active and the overall ventilation is not compromised.
Therefore, since the respiratory pump did not fatigue to the
point of “task failure,” it is unlikely that the improvements
in performance were the result of improved gas exchange or
a better compensation for metabolic acidosis. However, the
altered breathing pattern observed after IMT suggests that
respiratory muscle fatigue might have been of physiological
significance to the regulation of the breathing pattern. In the
IMT group, tidal volume increased significantly, whereas
the placebo group resorted to a more tachypneic breathing
pattern, characteristic of respiratory muscle fatigue for the
maintenance of minute ventilation. Indeed, as the breathing
pattern during exercise seems to be optimized to avoid
exhaustive fatigue and “task failure” of the respiratory mus-
cles, the increased strength of the IMT group might have
enabled them to increase tidal volume without fatiguing. In
contrast, the placebo group, which was susceptible to fa-
tigue, resorted to an increased breathing frequency. Even
though we did not assess the degree of entrainment between
breathing and stroke rate, it is possible that the prevention of
a tachypneic breathing pattern in the IMT group enhanced
the mechanical efficiency of the rowing work by enabling
the maintenance of entrainment. Indeed, our data are in
agreement with previous suggestions that breathing in row-
ing occurs at times where muscle synergy produces larger
ventilatory volumes for a given amount of respiratory work,
or alternatively, the same volume for less respiratory work
(28); consequently performance may be improved.
INSPIRATORY MUSCLE TRAINING IN ROWING Medicine & Science in Sports & Exercise
姞
807
