Eur J Appl Physiol (1992) 64:51-55

Applied
European
Journal of

Physiology and Occupational Physiology

@ Springer-Vedag1992

Hypertrophy without increased isometric strength

after weight training
D. G. Sale, J. E. Martin, and D. E. Moroz
Department of Physical Education, McMaster University, Hamilton, Ontario, Canada L8S 4KI

Accepted October 15, 1990

Summary. Eight men (20-23 years) weight trained 3 the nervous system (Komi 1986; Sale 1988). Alterna-
days. week-1 for 19 weeks. Training sessions consisted tively, training could cause changes within the muscles
of six sets of a leg press exercise (simultaneous hip and that increase the force that can be developed per unit
knee extension and ankle plantar flexion) on a weight muscle CSA (Jones et al. 1989).
machine, the last three sets with the heaviest weight A far less common observation, and one more diffi-
that could be used for 7-20 repetitions. In comparison cult to explain, has been a training-induced increase in
to a control group (n = 6) only the trained group in- muscle size without a corresponding increase in
creased (P<0.01) weight lifting performance (heaviest strength. In this pattern of results, strength measured
weight lifted for one repetition, 29%), and left and right with a test mode (isometric) different from the training
knee extensor cross-sectional area (CAT scanning and mode (weight lifting) has not increased significantly,
computerized planimetry, 11%, P<0.05). In contrast, despite a significant increase in muscle size. Thus,
training caused no increase in maximal voluntary isom- weight training in older men failed to increase the
etric knee extension strength, electrically evoked knee isometric strength of the right knee extensors, despite a
extensor peak twitch torque, and knee extensor motor 12% increase in knee extensor CSA (Frontera et al.
unit activation (interpolated twitch method). These data 1988). A recent study in our laboratory found that
indicate that a moderate but significant amount of hy- weight training in older men failed to increase elbow
pertrophy induced by weight training does not neces- flexion isometric strength, despite a 17% increase in el-
sarily increase performance in an isometric strength bow flexor CSA (data to be published elsewhere). A
task different from the training task but involving the failure to increase strength despite hypertrophy might
same muscle group. The failure of evoked twitch torque be the result of training-induced decrease in specific
to increase despite hypertrophy may further indicate tension (reduced force per unit muscle CSA), or coun-
that moderate hypertrophy in the early stage of strength terproductive neural adaptations such as inhibition of
training may not necessarily cause an increase in intrin- agonists or increased co-contraction of antagonists.
sic muscle force generating capacity. In the present study we provide further evidence of
training-induced hypertrophy without a corresponding
Key words: Muscle size - Strength training - Contrac- increase in voluntary isometric strength. The data sug-
tile properties gest that a decrease in muscle specific tension may have
occurred. On the other hand, there was no evidence of
inhibition after training.

Introduction
Methods
A common observation in short-term strength training
studies has been an increase in strength far exceeding Subjects
any increase in muscle cross-sectional area (CSA) (Ikai
and Fukunaga 1970; Tesch et al. 1983; Jones and Ru- Fourteen young men were randomly assigned to a control (n = 6)
therford 1987; Davies et al. 1988; Narici et al. 1989). A or training group (n = 8). The control group had an age of 21.5, SE
greater increase in voluntary strength than muscle size 0.6 years, a height of 1.74, SE 0.002 m, and a body mass of 75.5,
has most often been attributed to adaptive changes in SE 3.0 kg. The training group had an age of 21.4, SE 0.4 years, a
height of 1.706, SE 0.003 m, and a body mass of 71.t, SE 2.5 kg.
There were no significant differences between the groups in these
physical characteristics. The men were physically active, most be-
Offprint requests to: D. G. Sale ing physical education students, but had no previous experience
52

with intense weight training. The subjects participated with in- tractions were evoked by percutaneous electrical stimulation.
formed, written consent and the project carried the approval of After preparing the skin with alcohol and conducting gel, two
McMaster University's Human Ethics Committee. large (100x 45 mm) stimulating electrodes (carbon impregnated
rubber) were taped to the skin, one over the femoral nerve in the
inguinal region, the second over the distal third of the quadriceps.
Training Stimuli were rectangular voltage pulses, 100 or 200 ~ts in dura-
tion, delivered from a stimulator (Devices Medical Systems). In a
test a series of stimuli of progressively increasing intensity were
Weight training was done 3 days. week-1 (Mon., Wed., Fri.) for
delivered. The twitch response was considered to be maximal
19 weeks. There was a 2-week break at the midpoint of the train-
when further increases in stimulus intensity caused no further in-
ing to accomodate a Christmas recess. A single training exercise
creases in twitch torque, as monitored on the oscilloscope. A max-
was used, a bilateral leg press done on a "Universal" type weight
imal twitch response was then dispatched to the computer for de-
machine (Global Gym, Downsview, Ontario). A leg "press" con-
termination of the peak torque attained. The twitch measurements
sisted of simultaneous hip and knee extension and ankle plantar
were always completed before any voluntary contractions were
flexion. The training device had a seat with a back rest that was
done, in order to avoid the confounding potentiating effect of the
inclined backward at an angle of 0.17 tad (10 °) from the vertical.
MVCs (Vandervoort et al. 1983).
The exercise began with the subject seated with knee and hip
joints at 1.57 rad (90°) and 1.05 rad (60°), respectively. The feet
Motor unit activation. The extent of motor unit activation during
rested on a footplate that was pushed in the horizontal plane until
the MVCs was assessed using the interpolated twitch technique
the knee joint was fully extended and the hip joint was 1.92 rad
(Belanger and McComas 1981). A supramaximal stimulus was de-
(110°). Each "repetition" of the exercise consisted of a concentric
livered to the quadriceps when the torque was maximal during the
and eccentric contraction phase of the involved muscle groups.
MVCs. If an increment occurred on the MVC torque recording,
Each training session consisted of three "warm-up" sets with in-
the magnitude of the increment, expressed as a percentage of the
creasing percentages of the greatest mass that could be used for
maximum twitch magnitude evoked at rest, represented that por-
the designated number of repetitions. These sets were followed by
tion of the muscle mass not activated by voluntary effort. The per-
three "work" sets with the greatest mass that could be used for the
centage value of the increment, subtracted from 100, gave the per-
designated number of repetitions (i.e. repetitions maximum, RM).
centage value of voluntary motor unit activation. The method can-
The masses used were increased whenever the upper limit of the
not distinguish between incomplete recruitment or insufficient
prescribed repetition range could be done on two successive train-
motor unit firing rates as being responsible for the increment on
ing sessions. In the first session of each week, the first work set
the torque recording; hence, the term "activation" is used (Belan-
was done at 15-20 RM, the second at 10-15 RM, and the third at
ger and McComas 1981).
7-10 RM. In the second session all three sets were done at 15-20
RM. In the third session the first set was at 15-20 RM, and the
Muscle size. CSA of the right and left knee extensors was mea-
last two sets were at 10-15 RM.
sured by computerized planimetry of photographs of computered
tomography (CT) scans obtained with a CT scanner (Technicare
2020HR). CT scans were made at the mid-thigh level between the
Measurements most lateral point of the greater trochanter and the most lateral
point of the lateral epicondyle of the femur (Sale et al. 1990).
All measurements were made in both groups in a 2-week period
before and after the training period.

Voluntary strength. Voluntary strength was measured in two ways. Statistics

First, weight-lifting strength was measured as the greatest mass
that could be lifted (concentric contraction phase) for one repeti- Data were analysed with a two-factor (group, time) ANOVA with
tion (1 RM) in the leg press movement on the training device. In repeated measures on the time factor (before and after the train-
this test the subject did successive sets of single repetitions with ing period). A significant effect of training would be indicated as
progressively greater masses until the 1 RM was determined. a significant group by time interaction. Where significant interac-
Three-minute rest periods were given between sets to minimize tions occurred the Tukey post hoc test was used to determine sig-
the possibility of fatigue. The test was repeated on a second occa- nificant differences between mean values. Significance was ac-
sion on a separate day and the greater of the two test results was cepted at P < 0.05. Descriptive statistics included mean and stand-
taken as the pre-training value. ard error of the mean (SE).
Second, isometric strength of right and left knee extension was
measured on a custom-made dynamometer. The dynamometer
consisted of a chair that supported the subject with the thighs in
the horizontal plane, the trunk-hip angle at 1.92 rad (110°), and Results
the knee joint angle at 1.57 rad (90°). Hip and thigh restraining
straps prevented extraneous movements during the isometric con- T h e effect o f t r a i n i n g o n weight lifting p e r f o r m a n c e (1
tractions. The lower leg was strapped firmly to a large alumini'um R M ) a n d m u s c l e C S A is s h o w n in Fig. 1. T h e r e was a
plate. The proximal end of the plate was clamped to a steel shaft s i g n i f i c a n t g r o u p b y time i n t e r a c t i o n for b o t h leg press
whose axis was coincident with the axis of the knee joint. The 1 R M ( P < 0 . 0 0 1 ) a n d left ( P - - 0 . 0 3 3 ) a n d right
shaft was instrumented with a strain gauge that sensed the torque
( P = 0.034) k n e e e x t e n s o r CSA. I n the t r a i n i n g group, 1
developed by the isometric contractions. The signal from the
strain gauge was amplified and sent to a storage oscilloscope R M i n c r e a s e d 29.1% ( P < 0 . 0 1 ) a n d left a n d right knee
(Hewlett Packard 120 1B) for display and to a computer (PDP e x t e n s o r C S A i n c r e a s e d b y 10.8% a n d 11.1%, respec-
11-03, Digital Equipment) for analysis. The subject did 2-3 maxi- tively ( P < 0.05). I n the control g r o u p , there were n o sig-
mal voluntary contractions; the contraction producing the n i f i c a n t c h a n g e s in these measures.
greatest torque was taken as the measurement of isometric T h e effect of t r a i n i n g o n v o l u n t a r y i s o m e t r i c
strength (MVC).
strength, e v o k e d twitch torque, a n d m o t o r u n i t activa-
Evoked twitch torque. Twitch torque measurements were made on t i o n is s h o w n i n Fig. 2. T h e r e were n o g r o u p or time
the same system used for the MVC measurements. Twitch con- m a i n effects, n o r g r o u p b y time i n t e r a c t i o n s i n a n y of
 53

300
T
p r e post I
o~ 300 Z
:/
"E 200 %
Z %
tic // - % //

•"- 200 0
//

% g /1 %
O} #1 % //
"~ 100 • //
% g I
// %
IJkl /t 11
:/
%
n,"
100 z z z
,%
//
(J
% z %
ILl
.-I
CONTROL TRAIN CONTROL TRAIN

CONTROL TRAIN

50 " i

1"I
A

E T
_T__
z 40 • ?, ~_ ~ - :/,v7 ]__
LU
12500- ~ -- 1/
E 0 30 - / /
IT" /1

0 Z
m
< 10000- ~/ -- ~/~ I'- 2 0 - :%
"
"T-
o -
//
O 11
o" 5oo- _ 1o • Z
/I
//

Z 5000 - // I--
/I
0 '

y'/ CONTROL TRAIN CONTROL TRAIN

"//

i
Z 0 -- -- -- 100 -
control train control train

LEFT LEG RIGHT LEG

75_
Fig. 1. Effect of weight training on leg press 1 RM (top) and knee // -- y/

extensor cross-sectional area (CSA). Only the training group in- //

creased 1 RM (** P<0.01) and left and right knee extensor CSA <: ~o - "-" /~ - ~
(* P< 0.05). RM, Repetitions maximum
25 -

these measures; that is, there were no changes in either

o
the control or training group after the training period. control train control train

LEFT LEG RIGHT LEG

Discussion Fig. 2. Effect of weight training on voluntary knee extensor isom-

etric strength (MVC, top), evoked isometric knee extensor twitch
The present study has shown that a weight-training pro- torque (middle), and knee extensor motor unit activation (MUA,
bottom). There were no significant changes in any of the measures
gramme which increased leg press weight-lifting per- after the training period
formance by 29% and knee extensor CSA by 11% failed
to increase knee extensor voluntary isometric strength
and evoked twitch torque. The observation of an in-
crease in muscle CSA without a corresponding increase ceps fibre area. In a second trial men did the same
in a measure o f strength is u n c o m m o n but not unpre- weight training as before but also regularly practised
cedented. We recently found no increase in the elbow isokinetic contractions. This time the muscle and mus-
flexor isometric strength of elderly men after a weight- cle fibre size increases were the same as before, but iso-
training programme, despite significant increases in kinetic strength had increased markedly (MacDougall
weight-lifting performance and muscle CSA (data sub- et al. 1977, 1980). Finally, hypertrophy of animal mus-
mitted for publication elsewhere). Another group has cles by functional overload has sometimes not been as-
made a similar observation in the knee extensor mus- sociated with an increase in isometric strength (Kan-
cles of elderly men: relatively large increases in weight- darian and White 1989).
lifting performance and muscle CSA, but a smaller or If it is assumed that evoked twitch torque is a valid
no increase in isometric strength (Frontera et al. 1988). measure of muscle force generating capacity, then this
Some years ago we had observed no increase in elbow capacity did not increase after about 5 months of
extensor low-velocity isokinetic concentric contraction weight training, despite significant hypertrophy. Thus,
strength in young men after 6 months of weight train- the hypertrophy during this period was associated with
ing, despite significant increases in arm girth and tri- a decrease in specific tension. There is also the implica-
54

tion that hypertrophy made no contribution to the in- the possibility that in the initial stages of strength train-
crease in weight-lifting performance; instead, this in- ing (several weeks to a few months) observed increases
crease was entirely due to neural adaptations such as in muscle mass may not necessarily be contributing to
improved coordination and learning (Rutherford and the increases in voluntary strength observed.
Jones 1986). However, twitch torque may be an equivo- The mechanisms responsible for the force deficit as-
cal measure of muscle force generating capacity. The sociated with hypertrophy are not known. In a recent
twitch response would be very sensitive to any adapta- animal study it was shown that the deficit could occur
tion that altered excitation-contraction coupling. Some without connective tissue proliferation, an increase in
investigators use twitch torque as a measure of the sta- interstistial space, or a decrease in protein concentra-
tus of excitation-contraction coupling (Duchateau and tion. The authors suggested that future studies incorpo-
Hainaut 1984). Tetanic torque would be the best meas- rating ultrastructural and myofibrillar protein analyses,
ure of intrinsic muscle strength. (We elected not to use and investigations of excitation-contraction coupling
this measure because of the extreme discomfort caused and the cytoarhitecture, may help to uncover the mech-
by tetanic stimulation.) We cannot exclude the possibil- anisms (Kandarian and White 1989). With specific ref-
ity that in the present study tetanic torque may have erence to the human quadriceps, it has been suggested
increased despite no increase in twitch torque. This pat- that hypertrophy, by increasing the angle of pinnation
tern of results could happen if training impaired excita- of muscle fibres, would decrease the net force acting
tion-contraction coupling enough to prevent twitch tor- along the line of the tendon (Maughan 1984).
que from increasing, but not so much so as to prevent
hypertrophy from being expressed as an increase in te- Acknowledgements. The authors are indebted to John Moroz for
tanic torque. technical assistance, Mary Cleland for secretarial assistance, and
Mauris Cecca for supervising the training of the subjects. The
We are inclined to discount the possibility that te- study was supported by the Natural Sciences and Engineering Re-
tanic torque increased, because if it had, there would search Council of Canada (NSERC).
have been an increase in voluntary isometric strength
(MVC). There was no change in motor unit activation
pre- to post-training; therefore, MVC torque should
References
have reflected any change in tetanic torque generating
capacity. However, we must concede a possible mecha-
Baratta R, Solomonow M, Zhou BH, Letson D, Chuinard R,
nism whereby tetanic torque could have increased with- D'Ambrosia R (1988) Muscular coactivation. The role of the
out an increase in MVC or a change in motor unit acti- antagonist musculature in maintaining knee stability. Am J
vation. The mechanism would be a training-induced in- Sports Med 16:113-122
crease in the co-contraction of the antagonistic knee Belanger AY, McComas AJ (1981) Extent of motor unit activation
flexors during voluntary knee extension MVCs. An in- during effort. J Appl Physiol 51:160-167
creased co-contraction of antagonists (i.e. increased Davies J, Parker DF, Rutherford OM, Jones DA (1988) Changes
in strength and cross sectional area of the elbow flexors as a
counteracting flexor torque) would reduce net extensor result of isometric strength training. Eur J Appl Physiol
torque, perhaps enough to prevent the latter's increase 57: 667-670
despite an increase in quadriceps tetanic torque. Knee Draganich LF, Jaeger RJ, Kralj AR (1989) Coactivation of the
flexors do co-contract during knee extension MVCs hamstrings and quadriceps during extension of the knee. J
(Baratta et al. 1988; Draganich et al. 1989). The antag- Bone Joint Surg [Am] 71:1075-1081
Duchateau J, Hainaut K (1984) Isometric or dynamic training:
onist knee flexor torque can be as great as 10% of total
differential effects on mechanical properties of a human mus-
extensor torque (Baratta et al. 1988). The literature is cle. J Appl Physiol Respir Environ Exercise Physiol 56:296-
divided on whether training causes an increase in co- 301
contraction of antagonists. A study of sprinters and dis- Frontera WR, Meredith CN, O'ReiUy KP, Knuttgen HG, Evans
tance runners showed that the former had greater co- WJ (1988) Strength conditioning in older men: skeletal muscle
contraction of knee flexors in high-velocity knee exten- hypertrophy and improved function. J. Appl Physiol 64:1038-
1044
sions (Osternig et al. 1986). In contrast, strength and Ikai M, Fukunaga T (1970) A study on training effect on strength
power athletes vs endurance athletes had less knee per unit cross-sectional area of muscle by means of ultrasonic
flexor co-contraction during slow knee extensions de- measurement. Eur J Appl Physiol 28:173-180
signed to stretch the knee flexors (Osternig et al. 1990). Jones DA, Rutherford OM (1987) Human muscle strength train-
Athletes with hypertrophied quadriceps show less co- ing: the effects of three different regimes and the nature of the
contraction of knee flexors during maximal low-veloc- resultant changes. J Physiol (Lond) 391:1-11
Jones DA, Rutherford OM, Parker DF (1989) Physiological
ity isokinetic knee extensions than control subjects, but changes in skeletal muscle as a result of strength training. Q J
if athletes are given specific knee flexion training exer- Exp Physiol 74:233-256
cises, flexor co-contraction increases to control levels Kandarian SC, White TP (1989) Force deficit during the onset of
within a few weeks (Baratta et al. 1988). We cannot rule muscle hypertrophy. J Appl Physiol 67:2600-2607
out the possibility that increased co-contraction o f an- Komi PV (1986) Training of muscle strength and power: interac-
tagonists may have occurred after training. tion of neuromotoric, hypertrophic and mechanical factors.
Int J Sports Med 7 [Suppl]: 10-15
Notwithstanding the equivoval nature of twitch tor- MacDougall JD, Ward GR, Sale DG, Sutton JR (1977) Biochemi-
que as a measure of intrinsic muscle force generating cal adaptation of human skeletal muscle to heavy resistance
capacity, and the possible confounding effect of altered training and immobilization. J Appl Physiol 43:700-703
co-contraction of antagonists, the present study raises MacDougall JD, Elder GCB, Sale DG, Moroz JR, Sutton JR
 55

(1980) Effects of strength training and immobilization on hu- ential responses to proprioceptive neuromuscular facilitation
man muscle fibers. Eur J Appl Physiol 43:25-34 (PNF) stretch techniques. Med Sci Sports Exerc 22:106-111
Maughan RJ (1984) Relationship between muscle strength and Rutherford OM, Jones DA (1986) The role of learning and coor-
muscle cross-sectional area. Implications for training. Sports dination in strength training. Eur J Appl Physiol 55:100-105
Med 1:263-269 Sale DG (1988) Neural adaptation to resistance exercise. Med Sci
Narici MV, Roi GS, Minetti AE, Cerretelli P (1989) Changes in Sports Exerc 20:S135-S145
force, cross-sectional area and neural activation during Sale DG, MacDougall JD, Jacobs I, Garner S (1990) Interaction
strength training and detraining of the human quadriceps. Eur between concurrent strength and endurance training. J Appl
J Appl Physiol 59:310-319 Physiol 68: 260-270
Osternig LR, Hamill J, Lander JE, Robertson R (1986) Co-activa- Tesch PA, Hjort H, Balldin UI (1983) Effects of training on G
tion of sprinter and distance runner muscles in isokinetic exer- tolerance. Aviat Space Environ Med 54:691-695
cise. Med Sci Sports Exerc 18:431-435 Vandervoort AA, Quinlan J, McComas AJ (1983) Twitch potentia-
Osternig LR, Robertson RN, Troxel RK, Hansen P (1990) Differ- tion after voluntary contraction. Exp Neurol 81:141-152