Downloaded from www.ajronline.org by 190.221.255.156 on 10/05/15 from IP address 190.221.255.156. Copyright ARRS. For personal use only; all rights reserved
765
Acute Effects of Exercise on MR Imaging of Skeletal Muscle: Concentric
Frank G. Shellock1 Tetsuo Fukunaga2’3 Jerrold H. Mink1 V. R. Edgerton2
vs Eccentric
Actions
Eccentric (lengthening) muscle actions involve the forced lengthening of active muscles. Compared with concentric (shortening) muscle actions subjected to the same relative work load, eccentric actions have lower oxygen consumption requirements, fewer activated motor units, and less lactate production. This study was conducted to determine if T2-weighted MR could show any difference in muscles performing these specific types of actions and, therefore, be useful for physiologic investigations of eccentric and concentric actions. Five subjects performed exhaustive exercise by doing isolated concentric actions (raising a dumbbell, flexing at the elbow) and eccentric muscle actions (lowering a dumbbell, extending the contralateral arm). T2-weighted MR images of the arms were obtained immediately before and after exercise. Muscles that performed concentric actions had increases in signal intensity, whereas muscles that performed eccentric actions showed little or no change. T2 relaxation times increased significantly (p < .01) in all volunteers, but T2 relaxation times for the muscles that performed concentric actions were significantly higher than those for muscles that performed eccentric actions (p < .01). Therefore, T2 times increased with both concentnc
and
eccentric
actions,
but
the
images
failed
to show
the
changes
in the
muscles
that performed the eccentric actions. These data demonstrate that assessment of T2 values can be used to distinguish between muscles that perform concentric actions and those that perform eccentric actions, and this phenomenon may be useful for further physiologic investigations of these specific types of muscle actions. AJR
1
Tower
Musculoskeletal
1990;
accepted
Imaging
after
Center,
Ce-
dars-Sinai Medical Center, Los Angeles, CA 90048 and Department of Radiological Sciences, UCLA School of Medicine, Los Angeles, CA 90024. Address reprint requests to F. G. Shellock, CedarsSinai Medical Center, MRI, 8700 Beverly Blvd., Los
Angeles, CA 90048. 2
fomia,
Department
of Kinesiology,
Los Angeles,
Los Angeles,
3 Present address: Department ences, University of Tokyo, Tokyo,
0361 -803X/91/1 564-0765 0 American Roentgen Ray Society
University
of Cali-
CA 90024. of Sports Japan.
Sci-
April
1991
Eccentric (lengthening) muscle actions involve the forced lengthening of active muscles and the transfer of external power from the environment to the subject [1 2]. Examples include the use of resistance equipment, such as weight-lifting devices, or deceleration processes during activity, such as running downhill. Therefore, eccentric muscle actions produce active tension while the muscle is being lengthened. Various investigations have shown that, compared with concentric (shortening) muscle actions, eccentric actions are associated with reduced oxygen consumption, fewer activated motor units, and less lactate production for the same power output [3-7]. Therefore, concentric actions apparently require more energy than eccentric actions do when subjected to the same relative work load [3-7]. Fleckenstein et al. [8] first reported that T2-weighted spin-echo images of active skeletal muscle showed increased signal intensity immediately after exercise. Other studies have also investigated this phenomenon [9-1 1 }. This contrast enhancement of exercised muscle was suggested to be caused by increased vascular and extracellular volumes [8], as well as by changes in intracellular water [1 1 ]. Research has additionally shown that T2 relaxation times of active skeletal muscles depend on exercise intensity [8, 1 1]. In consideration of these facts, the present study was conducted to determine if ,
Received September 24, revision October 23, 1990.
156:765-768,
SHELLOCK
766
T2-weighted acute
MR images
exercise
vs eccentric
actions
information
technique
of muscles
show
of muscles
is available
perform
relative
because,
performing
any difference that
at the same
is important
accurate Downloaded from www.ajronline.org by 190.221.255.156 on 10/05/15 from IP address 190.221.255.156. Copyright ARRS. For personal use only; all rights reserved
could
response
work
currently,
and MR could be used to precisely involved muscles.
direct
a long
TE
Study
samples
actions
sampling
[1],
of the
images
are
were
used,
T2-weighted
obtained
and
frorn the mid-forearms
immediately
proton-
to the deltoids,
jects’ arms placed together over their head loosely applied cloth tape to inhibit motion.
and Methods
and held
before
and
with the subin place
with
Subjects
Five
two
women; in this
normal
volunteer
gation. These arm resistance the study.
Exercise
subjects
(three
men
and
aver-
any in
Protocol
weighted
to a normalized
percentage
(i.e., 15%
for women and 20% for men) of the subject’s body weight was used for the resistance exercise. The weights ranged from 1 2 to 1 5 lb. for the women and from 25 to 30 lb. for the men. The subjects performed a “biceps curl” movement for the exercise while in a standing position. One arm was selected randomly to perform concentric (i.e. , shortening) actions. This was accomplished with strict form, by bending the arm at the elbow and raising the dumbbell with the palm up, moving from extension to full flexion. The weight was then passed to the contralateral arm by an assistant to perform the eccentric (i.e., lengthening)
actions.
This
was
bell with strict form,
beginning
it to extension.
extremities
The
accomplished
by lowering
with the arm at full flexion were
in supinated
out the range of motion of each movement. concentric and eccentric actions took place each movement took approximately 2 sec. subjects performed isolated concentric and alternating manner. Exercise was performed by each of the perceived exhaustion and “failure” (i.e., the raise
the dumbbell)
with
each
and the
extremity.
same
Therefore,
number the
the
dumb-
and lowering
positions
to the point of no longer could
of repetitions
concentric
and
were
achieved
eccentric
actions
were performed to the same relative work level. This exercise protocol was similar to those used by others to study concentric vs eccentric actions
[12]
techniques
and
was
also
frequently
designed
to
used
by body
work
performed
simulate
builders
resistive
and strength
by measuring
total
power,
(kg) x distance time
yielded
iOO kg m/sec
the
results
arm was calculated
subjects
in this
1 09 kg rn/sec (woman), 1 55 kg m/ sec (man), 142 kg rn/sec (man), and 155 kg rn/sec (rnan).
(woman),
MR Protocol MR was performed ture-dniven,
with a 1 .5-T, 64-MHz
transmit/receive
body
coil
MR imager
(General
determined vided
for these
by General
Electric
that
performed
the
care
to avoid
inclusion
or bony anatorny. regions
of interest
concentric
and
eccentric
of subcutaneous
T2 relaxation by using
fat,
times were
the software
pro-
Electric.
T2 relaxation times measured before and after exercise for each muscle group (e.g. , biceps before exercise concentric action vs biceps after exercise concentric action, triceps before exercise concentric action extrernity vs triceps after exercise concentric action extremity) were compared by using a paired t-test. A comparison between the postexercise concentric action biceps data and the postexercise eccentric action biceps data also was made by using a paired t-test.
Results Visual inspection of the images differences in the signal intensities
and quadraCompany,
Milwaukee, WI). Inasmuch as Fleckenstein et al. [8] showed that active muscles are most conspicuous on MR when pulse sequences
(Fig. 1) showed no apparent of any of the muscle groups
before exercise (i.e., the biceps and triceps). Immediately after exercise, increases were apparent in the signal intensity of the biceps and brachialis muscles that performed concentric actions, intensity eccentric
whereas little or no change occurred in the signal of the biceps and brachialis muscles that performed actions (Fig. 1). The signal intensity of the triceps
(i.e., inactive concentric
muscles)
and eccentric
of the extremities
that per-
actions
to be un-
appeared
changed from that on the preexercise images. The data on T2 relaxation times are summarized in Figure 2. The T2 relaxation times were not significantly different for the triceps
the
special
fascia, blood vessels,
muscles
(m)
for
arms
with
formed
(sec)
following
mid-upper
actions,
training
as follows:
mass
calculation
by each
ject’s
trainers
113]. The estimated
The images were filmed with standardized window settings, and the relative signal intensities of the active muscles in the arrns involved in concentric and eccentric actions were compared visually with those of the nonexercising muscles (i.e. , the triceps). Regions of interest were selected in the center of the biceps (i.e., the active muscles) and triceps (i.e., inactive muscles) of each sub-
through-
The rate at which the was monitored, so that Using this protocol, the eccentric actions in an subjects subject
Data Analysis
investi-
individuals were untrained and had not performed training for at least 6 months before involvement
A single dumbbell
study:
TR
load. This
age age, 36 years; range, 21 -48 years) were involved
This
long
after exercise with the following pararneters: axial imaging plane; 2000/80, 20 (TR/TE); 128 x 256 matrix; two excitations; 44-cm field of view; 1 0-mm slice thickness; and 2-mm interslice gap. These pararneters were selected to minimize data acquisition tirne while still obtaining the necessary image information from both exercising limbs for analysis. The subjects were placed supine in the MR irnager, and images
were obtained Subjects
and
spin-echo
April 1991
density,
biopsy
vs eccentric
with
AJR:156,
concentric
no sufficiently
for obtaining
concentric
in the
ET AL.
and SD, 0.4 29.8
muscles
of the extremities
performing
concentric
eccentric actions (note: all reported values are mean ± triceps preexercise concentric action extremity, 29.5 ± msec, triceps postexercise concentric action extremity, ± 0.5 msec, p = N.S.; triceps preexercise eccentric
action extremity, 29.6 ± 0.3 msec, triceps postexercise centric action extremity, 29.6 ± 0.5 msec, p = N.S.). Statistically significant increases in T2 relaxation times
ecwere
seen for the biceps muscles performing concentric actions (biceps preexercise concentric action, 29.1 ± 0.6 msec, biceps postexercise concentric action, 38.5 ± 0.9 msec, p = .0001
)
and the biceps
performing
eccentric
actions
(biceps
AJR:156,
EXERCISE
April 1991
AND
MR OF SKELETAL
centric action,
action, 33.1
MUSCLE
767
38.5 ± 0.9 msec,
1 .1 msec, p
±
=
biceps .0006).
postexercise
eccentric
Downloaded from www.ajronline.org by 190.221.255.156 on 10/05/15 from IP address 190.221.255.156. Copyright ARRS. For personal use only; all rights reserved
Discussion MR imaging has developed into the foremost noninvasive imaging technique for examining normal and diseased conditions of the musculoskeletal system [1 4]. When MR imaging is used, biological tissues and fluids with excess amounts of
free water have long T2 values and, therefore, tend to be the most conspicuous on T2-weighted images [1 4, 15]. Exercise-induced
water content
changes
produce
in extracellular
changes
and intracellular
in proton relaxation
times [8-
1 1] and correlate with the level of exertion [8, 1 1]. Fleckenstein et al. [8] first showed that exercise is associated with statistically significant changes in Ti T2, and spin-density relaxation times observed on MR. The greatest effect was ,
seen on images obtained with T2-weighted pulse sequences and, to a lesser degree, on those obtained with gradientreversal techniques in which the flip angle was reduced to decrease the effect of Ti differences on the images [8]. In consideration of the results of previous studies [8, 1 1 ], we decided to use T2-weighted MR selectively for our investigation to maximize the suspected differences between muscles performing concentric actions and those performing eccentric actions.
The preexercise
occurs
In signal
actions,
and they are not differentiated
ifty
of active
muscles
that
performed
eccentric
as easily.
actions
-
PREEXC BICEPS POST
EX.C
BICEPS PREEXC
‘
45
TRICEPS
40
II
POST EX.C TRICEPS
35
__
30
PRE EX-E
25
BICEPS
!
POST EX.E BICEPS
CJ
15
___
10
5
C’”’”?,
PRE EX-E TRICEPS POST EX-E TRICEPS
0
Fig. 2.-Graph
shows T2 values of biceps and triceps measured before extremities performed concentric and eccentric actions. Values are means ± SD. EX-C = concentric actions; EX-E = eccentric actions. and after
preexercise exercise addition, concentric biceps
eccentric action, 29.5 ± 0.5 msec, biceps posteccentric action, 33.1 ± 1 .1 msec, p = .004). In the T2 relaxation times for the biceps performing action were significantly higher than those for the
performing
eccentric
actions
(biceps
postexercise
con-
T2 values for the muscles
evaluated
in this
study were comparable to those reported by Fleckenstein et al. [8] and others [1 3, 1 6, 1 7]. The immediate postexercise T2 values for the muscles performing concentric and eccentric also
were
within
the
ranges
reported
by
previous
investigators for active muscles [8, 1 1]. Additionally, we observed that muscles performing eccentric actions had a statistically significant lower T2 value than did muscles performing concentric actions. Research studies have shown that muscle actions that involve eccentric actions require less oxygen, produce less lactate, and use fewer muscle fibers than do concentric actions subjected to the same relative work load [1 -7]. Of further note, Fisher et al. [1 1 ] reported a strong correlation between the increase in T2 values and the mean force during exercise. Therefore, considering that T2 values are related to exercise intensity and eccentric action requires less of an energy expenditure than does concentric action, it is not surprising that we found statistically significant differences between T2 values for each of these specific types of muscle action. Exercise produces rapid alterations in the content of water found in skeletal muscle [1 8, 1 9]. Several mechanisms appear to be involved in this process. The water content in the exercising muscles increases as a result of the movement of water across the capillary wall, which is mediated by the hydrostatic pressure in the capillary and the osmotic forces in both the capillary blood and interstitial fluid [1 8, 19]. The increase in muscle lactate during exercise causes an increased tissue osmolality that also contributes to the increased water content of skeletal muscle observed with exercise [18, 19]. Additionally, the vascular bed of active muscles has an enlarged functional capillary surface area and
SHELLOCK
768
an increased
creased interstitial
Downloaded from www.ajronline.org by 190.221.255.156 on 10/05/15 from IP address 190.221.255.156. Copyright ARRS. For personal use only; all rights reserved
mean
capillary
pressure
filtration of water space [1 8, 19].
from
that
produces
the capillary
an in-
crease in muscle water content occurs in the intracellular space [19]. Fleckenstein et al. [8] indicated that the increased
water,
also may be involved.
in intracellular
The prelim-
formed the concentric action, with a minor contribution from an increase in extracellular water, as our subjects performed
protocol.
tion of extracellular maximal concentric
studied
However,
the relative
et al. [9] localization
proposed technique
scopic studies of active muscles anatomy
and individual
from our data.
that exercise-enhanced for guiding MR spectro-
because
differences
variations
in muscle
pat-
terns exist, and palpation techniques used to identify the sample volume are relatively inaccurate. Therefore, the use of exercise-induced portant for avoiding
active
contrast enhancement is particularly admixtures of spectroscopic signals
and inactive
muscles
[9]. However,
intensity
changes
are not visually
images
of muscles
that perform
may
not
be possible
to rely
apparent
isolated
simply
because
imfrom
signal
on T2-weighted
eccentric
on visual
actions,
inspection
sensitive
volume
actions.
for MR spectroscopy
studies
involving
only
There is currently no sufficiently accurate technique available for obtaining biopsy samples of muscles performing concentric vs eccentric actions [1] and, therefore, MR could be used to precisely direct the sampling of the involved active muscles. MR also may be helpful in further elucidation of physiologic differences between these specific types of mus-
cle actions.
Day
Ohira, for their
John
Hodgson,
involvernent
in this
Tracey
Meeks,
study.
Armstrong
RB. Mechanisms
of exercise-induced
delayed
onset
muscular
5. Bigland-Ritchie B, Woods M. Integrated electromyogram and oxygen uptake during positive and negative work. J Physiol 1976;260:267-277 6. Davies CTM, Bames C. Negative (eccentric) work. II. Physiological responses to walking uphill and downhill on a motor-driven treadmill. Ergonomics 1972;15: 121 -131 7. Schwane JA, Watrous BG, Johnson SR. Armstrong RB. Is lactic acid related to delayed-onset muscle soreness? Physician Sportsmed 1983;1 1: 8. Fleckenstein JL, Canby RC, Parkey exercise on MR imaging of skeletal
RW, Peshock RM. Acute effects of muscle in normal volunteers. AJR
1988;151 :231 -237 9. Fleckenstein JL, Bertocci LA, Nunnally RL, Parkey RW, Peshock RM. Exercise-enhanced MR imaging of variations in forearm muscle anatomy and use: importance in MR spectroscopy. AJR 1989;153:693-698 1 0. Peshock R, Fleckenstein J, Payne J, Lewis 5, Mitchell J, HaIler R. Muscle
11 .
usage pattems during cycling: MRI evaluation (abstr.). Magn Reson Imaging 1990;7:23 Fisher MJ, Meyer RA, Adams GR, Foley JM, Potchen EJ. Direct relationship
between proton T2 and exercise intensity in skeletal muscle MR images. Invest Radiol 1990;25:480 12. Clarkson PM, Tremblay I. Exercise-induced muscle damage, repair, and adaption in humans. J AppI Physiol 1988;65: 1-6 13. Fleck SJ, Kraemer WJ. Designing resistance training programs. Champaign, IL: Human Kinetics Books, 1987 1 4. Mink JH, Deutsch A. MRI of the musculoskeletal system: a teaching file.
it of
images to identify transient increases in signal intensity. Accordingly, it may be necessary to measure T2 relaxation times or to quantify the signal intensity changes to identify the eccentric
Yoshio
Kathy
REFERENCES
in muscle
recruitment
thank
124-131
contribu-
and intracellular water changes related to or eccentric muscle actions has not been
before and could not be determined
Fleckenstein MR is a useful
and
1952;28:364-382
free water related
images in our present study depended primarily on an increase in intracellular water of the active muscles that per-
exercise
gratefully Slimp,
a brief review. Med Sci Sports Exer 1984;16:529-538 2. Stauber WT. Eccentric action of muscles: physiology, injury, and adaption. In: Exercise and sports sciences reviews. Philadelphia: The Franklin Institute, 1988:157-185 3. Armstrong RB, Laughlin MH, Rome L, Taylor CR. Metabolism of rats running up and down an incline. J Appl Physiol 1983;55:518-521 4. Asmussen E. Positive and negative muscular work. Acta Physiol Scand
to maximal exercise. In consideration of this, we presume that the signal intensity changes observed on T2-weighted
a maximal
April 1991
soreness:
mary work by Fisher et al. [1 1] suggested that the increased signal intensity seen on T2-weighted images of active muscles
was largely due to increases
We Gina
1.
in muscles that performed fatiguing mainly by an increase in extracellular
but other processes
AJR:156,
ACKNOWLEDGMENTS
bed into the
Mild to moderate levels of exercise appear to be associated with a large (i.e., up to 1 00%) increase in extracellular water and a slight increase (i.e., about 1 0%) in intracellular water [1 8]. However, with maximal work loads, the greatest in-
signal intensity seen exercise were caused
ET AL.
New York: Raven, 1990 15.
Fisher MR, Dooms GC, Hncak H, Reinhold C, Higgins CB. Magnetic resonance imaging of the normal and pathologic muscular system. Magn Reson Imaging 1986;4:491-496 16. Hazelwood CF, Chang DC, Nichols BL, Woessner DE. Nuclear magnetic resonance transverse relaxation times of water protons in skeletal muscle. Biophys J 1974;14:583 1 7. Pettersson H, Fitzsimmons J, Krop D, Hamlin D. Magnetic resonance imaging of the extremities. II. Ti and T2 relaxation times of muscle and fat: normal values, reproducibility and dependence on physiologic variatins. Acta Radiol Diagn 1985;26:413-4i6 18. Sjogaard G, Saltin B. Extra- and intracellular water spaces in muscles of man at rest and with dynamic exercise. Am J Physiol 1982;243: R27i-R280 1 9. Sjogaard G, Adams RP, Saltin B. Water and ion shifts in skeletal muscle of humans with intense 1985;248:R190-R196
dynamic
knee
extension.
Am
J
Physiol