Eur J Appl Physiol (1992) 64:381-386
um.,of Applied
Hormonal adaptations and modelled responses in elite weightlifters
during 6 weeks of training
T. Busso 1, K. H~ikkinen 2, A. Pakarinen 3, H. Kauhanen 2, P. V. Komi 2, and J. R. Lacour 4
Laboratoire de Physiologie - GIP Exercice, Facult6 de M6decine Saint-Etienne, F-42023 Saint-Etienne Cedex 2, France
2 Department of Biology of Physical Activity, University of Jyv~iskyla, SF-40100 Jyv~iskyl~i,Finland
3 Department of Clinical Chemistry, University of Oulu, SF-90220 Oulu, Finland
4 Laboratoire de Physiologie - GIP Exercice, Facult6 de M6decine Lyon-Sud, F-69921 Oullins Cedex, France
Accepted January 14, 1992
Summary. The concentrations of serum testosterone,
sex-hormone-binding-globulin (SHBG) and luteinizing
hormone (LH) were examined throughout 1-year of
training in six elite weightlifters. A systems model, pro-
viding an estimation of fatigue and fitness, was applied
to records of training volume and performance levels in
clean and jerk. The analysis focused on a 6-week train-
ing period during which blood samples were taken at 2-
week intervals. A 4-week period of intensive training
(period I) could be distinguished from the following 2-
week period of reduced training (period II). During peri-
od I, decreases in serum testosterone (P<0.05) and in-
creases in serum L H concentrations ( P < 0 . 0 1 ) were ob-
served; a significant correlation (r=0.90, P < 0 . 0 5 ) was
also observed between the changes in serum L H concen-
tration and in estimated fitness. The magnitude of L H
response was not related to the change in serum andro-
gens. On the other hand, the change in testoste-
rone : SHBG ratio during period II was significantly cor-
related (r--0.97, P < 0 . 0 1 ) to the L H variations during
period I. These finding suggested that the L H response
indicated that the decrease in testosterone concentration
was not primarily due to a dysfunction of the hypothal-
amic-pituitary system control, and that the fatigue/fit-
ness status of an athlete could have influenced the L H
response to the decreased testosterone concentration.
The negative effect of training on hormonal balance
could have been amplified by its influence on the hypo-
thalamic-pituitary axis. A decrease in physiological
stress would thus have been necessary for the comple-
tion of the effect of L H release on androgenic activity.
Key words: Strength training - Systems model - Fatigue
- Fitness - Serum hormones
Offprint requests to: T. Busso, Centre Hospitalier Universitaire,
Pavillon 12, F-42650 Saint Jean Bonnefonds, France
European
Physiology
and OccupatLonalPhysiology
© Sprfnger-Verlag1992
Introduction
Systems theory is currently applied to describe the effects
of training on physical performance (Banister and Ham-
ilton 1985; Busso et al. 1990; Calvert et al. 1976; Morton
et al. 1990). The subject or athlete is characterized by a
system in which the output is the performance reacting to
a set of stimuli, i.e. from physical training, which com-
prises the system input. The model parameters are deter-
mined by fitting the model performance to the recorded
ones. The assumptions on the working of the system lead
to an estimate of the profiles of fatigue and fitness during
the period studied from the fitted parameters.
A study of the application of a systems model in elite
weightlifters and of the comparison of the modelled re-
sponses to hormonal variables, external to the model
computation, have provided a biological validation of
the model (Busso et al. 1990). The variations of fitness
were statistically related to the changes in serum testos-
terone concentration. There is some experimental data
concerning an association of improvement of fitness to
an enhancement a n d / o r periodical alteration of the ana-
bolic-androgenic activity in young boys (Mero et al.
1990), in initially low-trained subjects during endurance
training (Remes et al. 1979) or during strength training
(Alen et al. 1988; H~ikkinen et al. 1985). In elite weight-
lifters, a period as long as 2 years may be needed to ob-
serve an increase in serum testosterone concentration
(H~ikkinen et al. 1988a).
The fatigue level estimated from the systems model
remains to be clarified. In the above mentioned study
(Busso et al. 1990), a positive relationship was observed
between the variations of estimated fatigue and testoste-
rone concentrations. A negative relationships was ex-
pected: the decrease of serum androgen concentration
has been associated with the physiological stress of
training (H~ikkinen et al. 1988b; Kuoppasalmi et al.
1980). The positive relationship obtained was inter-
preted as resulting from the parallel between fitness and
fatigue profiles in the formulation of the model.
To provide a better representation of short-term
adaptations, the interpretation of the systems model was
382
r e v i s e d . M o r e o v e r , a statistical analysis was a d d e d t o t h e
m o d e l c o m p u t a t i o n to test t h e c o n f i d e n c e o f e a c h m o d e l
parameter.
T h e p u r p o s e o f this s t u d y was to a p p l y t h e r e v i s e d
m o d e l o f t r a i n i n g r e s p o n s e s to studies o f h o r m o n a l
adaptations related to prolonged strength training. The
analysis was f o c u s e d o n a 6 - w e e k t r a i n i n g p e r i o d p r e p a -
r a t o r y to t h e p r i m a r y w e i g h t l i f t i n g c o m p e t i t i o n .
Methods
Experimental methods. The data used in this paper were derived
from the study concerning the relationships between modelled re-
sponses and hormonal alterations in elite weightlifters (Busso et
al. 1990). The subjects were included in a larger group involved in
a study of hormonal changes during 1-year strength training (H/~k-
kinen et al. 1987). Only the six subjects who had a complete fol-
low-up of their training parameters were selected for the present
study. The experimental methods have been described in detail in
reports of those previous studies.
Briefly, the six subjects were elite weightlifters with a homoge-
nous background of several years of weightlifting training. The
subjects had not been taking exogenous anabolic-androgenic ste-
roids for several months before this study or during it. Their train-
ing consisted of the normal weight exercises used by elite weight-
lifters such as the olympic lifts, powerlifts, pulling exercises and
squat lifts.
Blood samples used in the present study were taken at the
weeks numbered 15, 17, 19 and 21 of the period of that year stud-
ied. Samples were drawn at 8 a.m. from the antecubital vein after
12-h fasting and a day of reduced training. They were kept at
- 20 ° C until the serum concentrations of testosterone and lutein-
izing hormone (LH) and sex-hormone-binding globulin (SHBG)
were assayed (H/~kkinen et al. 1987). The sensitivity, the coeffi-
cient of intra-assay variation and that of interassay variation were
respectively 0.26nmol.l -~, 7.0%, 9.7% for the testosterone as-
say, 1.0U.I -~, 3.2%, 6.3% for the LH assay and 0.5 n m o l ' l -a,
3.2%, 5.0% for the SHBG assay.
Throughout the 1-year period, the amounts of training were
calculated daily from the number of lifts, averaged by the week,
weighted by their average intensity referred to one maximal repeti-
tion (1 RM). For example, a session composed of 25 lifts perform-
ed at 80% of 1 RM corresponded to 20 arbitrary training units
(TU). The average daily training load was 28.5 (SD 6.3) TU during
the entire training period studied. The performance level was esti-
mated from the maximal result in clean and jerk tests performed
regularly during the training period (Busso et al. 1990).
Quantification of fatigue and fitness. The amount of training (sys-
tem input) and the performance (system output), defined above,
were used in a systems model of training effects on performance.
The data concerning the entire 1-year period were used in the
model computation for each subject. In previous studies concern-
ing modelling of training responses (Banister and Hamilton 1985;
Morton et al. 1990; Busso et al. 1990), the athlete was considered
as a system in which the behaviour was described by control of
two antagonistic systems associated with first order transfer func-
tions. The negative function represented the fatigue response and
the positive one the fitness response. In the present model, the be-
haviour of the athlete system has been described by a set of first-
order transfer functions, the number and the type of components
being determined by the model computation. The model formula-
tion led to a mathematical relationship of the system output (per-
formance) with the system input (amount of training). The model
performance at day n ~n) was estimated from the successive
amounts of training (w ~, at day number i) by:
rt--I ~ 0 --n/Tp
/~n=p.+ kr" Z w'.e (n--O/~r+ qp'e (1)
r=i t--I p--1
The system was described by R training components and P in-
itial components. The terms corresponding to the training compo-
nents were obtained by convolving the training loads to the im-
pulse response (kr" e - t/Tr), where kr was a multiplying factor in ar-
bitrary units depending of the units used in the quantification of
training load and performance; and Zr was a decay time constant
expressed in days. The initial components were introduced in the
model formulation by taking into account the events prior to the
beginning of the period of the study, where %0 was the initial lev-
el in arbitrary units and Tp was the decay time constant in days.
Thus the model performance was completely defined with the
following set of parameters:
1. A multiplying factor k and a decay time constant r for each
training component
2. An initial level qo and a decay time constant r for each initial
component
3. An additive term p* (called basic fitness)
The degrees of freedom (df) of the model depended on the to-
tal number of components. For a given number of training and
initial components, respectively R and P, the model parameters
are determined by fitting the model performance to the measured
one. The researched parameters were obtained by minimizing the
residual sum of squares (RSS) between modelled and measured
performance.
The model was solved for an increasing number of compo-
nents. Initially the simpler model, with only one training compo-
nent (df= 2), was applied to the training data. Then more complex
models were applied by introducing one by one further compo-
nents (training or initial components). Each additional component
corresponded to an increase in the model degrees of freedom
(Adf=2). The confidence limit for each added component was
tested by an analysis of variance on the decrease of the residual
sum of squares (ARSS). The F test was used to evaluate the level
of confidence. The components which decreased RSS significantly
(with a statistical level of 5%) were the only ones retained in the
final model formulation. The number of model parameters could
thus be different according to the subject. Goodness-of-fit indica-
tors were given for the more complex models obtained by this
analysis.
Then fitness and fatigue profiles were estimated from the set
of model parameters. The amount of training completed on day i
had an effect on the performance on day n quantified by
R
E(i/n) = ~ k~. w''e--~n--0/~r (2)
r=l
When E(i/n) was positive, it was considered as a fitness gain.
Conversely, when E(i/n) was negative, it was considered as a fa-
tiguing effect. By recurrence, the profiles of fatigue and fitness
were determined for the entire period studied as
n--I
fn= ~, ]E(i/n)l, when E(i/n)<0 (3)
1=1
a n =fin + f n (4)
where a n a n d f n were respectively the fitness and fatigue levels on
the day number n. The performance output was the difference be-
tween fitness and fatigue levels.
The 6-week period before the primary competition (weeks 15 to
21). The analysis focused on weeks 15 to 21, during which blood
samples were taken at 2-week intervals. This period corresponded
to the period of preparation before the primary competition. The
average amount of training between weeks 15 and 19 was statisti-
cally different (P<0.05) to the average amount of training be-
tween weeks 19 and 21:26.9 (SD 9.0) TU vs 19.2 (SD 4.5) TU.
Therefore, a 4-week period of intensive training (period I, between
weeks 15 and 19) could be distinguished from a 2-week period of
reducing training (period II, between weeks 19 and 21).
 383

Stat&tical analys&. The statistical significance of the performance Table 2. Serum concentrations of testosterone (T), luteinizing
fit was tested by an analysis of variance on RSS. The df of RSS hormone (LH), sex hormone binding globulin (SHBG) and
depended on the number of parameters included in the model T : SHBG ratio
computation. The F-test was used to estimate the level of signifi-
cance of the fit. Variables Weeks of training
Means and standard deviations (SD) were calculated for the
different variables. Linear regressions were used for calculation of 15 17 19 21
the correlation coefficients. Two-way analyses of variance were
used to evaluate whether significant changes occurred with time, T (nmol'l 1) mean 25.2 19.0 17.5 17.2
and changes were located by Scheffe's procedure. SD 8.6 5.6 4.4 4.8
LH (U.1-1) mean 8.8 12.0 11.7 10.1
SD 8.3 8.6 8.0 8.7
SHBG (nmol'l -~) mean 28.3 30.5 20.3 20.1
SD 5.1 9.2 3.7 3.8
Results
T : SHBG mean 0.90 0.65 0.88 0.87
SD 0.26 0.19 0.26 0.24
Performance fit over the year

T h e g o o d n e s s - o f - f i t i n d i c a t o r s are listed in T a b l e 1. T h e
fit was statistically significant in each subject. T h e final
m o d e l e q u a t i o n was c o m p o s e d o f one t r a i n i n g c o m p o - II0
i
nent for subject 5 a n d two a n t a g o n i s t i c t r a i n i n g c o m p o - I
nents in the others. O n e initial c o m p o n e n t was o b t a i n e d
for s u b j e c t 5 a n d t w o for subjects 1, 3 a n d 6. T h e coef- o
9O

ficients o f d e t e r m i n a t i o n r 2 r a n g e d f r o m 0.29 to 0.85.

[~ 70

Hormonal and modelled responses during the 6-week

period 50 i l i ~ i
13 15 17 19 21 23

T a b l e 2 shows the m e a n s e r u m c o n c e n t r a t i o n o f testoste-

- - * g o - -
r o n e , L H , S H B G a n d the t e s t o s t e r o n e : S H B G r a t i o dur- 240 "
ing the p e r i o d studied. C o m p a r e d with the values b e f o r e
the f o l l o w - u p p e r i o d , the s e r u m t e s t o s t e r o n e c o n c e n t r a -
200"
tions were d e p r e s s e d d u r i n g p e r i o d s I a n d II ( P < 0 . 0 5 ,
Fig. 1). The m e a n decreases in t e s t o s t e r o n e c o n c e n t r a -
t i o n were 2 0 % , 24% a n d 30% in weeks 17, 19 a n d 21,

/
160"

respectively. T h e s e r u m L H c o n c e n t r a t i o n was higher

( P < 0 . 0 1 ) d u r i n g weeks 17 a n d 19 t h a n d u r i n g week 15 120

(Fig. 1). T h e m e a n increase in L H c o n c e n t r a t i o n was

62% a n d 58% in weeks 17 a n d 19, respectively. T h e tes- 8O i i i i

t o s t e r o n e : S H B G r a t i o was l o w e r e d ( m e a n 2 5 % ) o n l y at 13 15 17 19 2T 23
week 17 ( P < 0 . 0 5 , Fig. 2). T h e s e r u m S H B G c o n c e n t r a -
t i o n was statistically lower at weeks 19 a n d 21 t h a n at
weeks 15 a n d 17 ( P < 0 . 0 1 , Fig. 2). T h e m e a n increase in
S H B G c o n c e n t r a t i o n was 2 3 % a n d 2 7 % in weeks 19 a n d
2O

Table 1. Goodness-of-fit indicators of performance

B
13 15 17 19 21 23
Subject R P TSS (df) MSS (df) F ratio r2
Week number
1 2 2 152 (41) 61 (8) 2.75* 0.40
Fig. 1. Serum concentrations of testosterone and luteinizing hor-
2 2 0 752 (42) 640 (4) 54.14"** 0.85
mone (LH), referred to a week-15 level of 100 and daily averaged
3 2 2 317 (29) 155 (8) 2.52* 0.49
training loads (means and SD) during the 6-week preparatory peri-
4 2 0 210 (38) 73 (4) 4.56** 0.35
od. Asterisks denote significant changes, * P<0.05; **P<0.01;
5 1 1 520 (40) 151 (4) 3.69* 0.29
TU, training unit
6 2 2 723 (40) 474 (8) 7.61"** 0.66

TSS, total sum of square; MSS, model sum of square; df, degrees
of freedom; r 2, coefficient of determination of performance fit;
R, number of training components in the final model equation; P,
number of initial components in the final model equation; aster-
isks denote the statistical significance of the fit, * P < 0 . 0 5 ,
** P < 0 . 0 1 , *** P < 0 . 0 0 1
21, respectively, c o m p a r e d to week 15. T h e v a r i a t i o n o f
L H c o n c e n t r a t i o n d u r i n g the p e r i o d I c o r r e l a t e d statisti-
cally ( r = 0 . 9 0 , P < 0 . 0 5 ) to the v a r i a t i o n s o f testoste-
r o n e : S H B G r a t i o d u r i n g the p e r i o d II (Fig. 3).
384

0
FI 80"
120"

100 "

50"
4¢-- 8#
120

80

40

40 i I i i |
13 15 17 19 21 23

-40

r4o -1,5 -1 -0,5 0 0.5

Level of estimated fitness
120 (A% change during period I)

100 Fig. 4. Comparisons between changes in estimated fitness and se-

rum luteinizing hormone (L_H) concentration during period I
80 (r=0.97, P<0.01, n=6)

60

40 i i i i i
No significant changes were observed in modelled fa-
13 15 17 19 21 23
tigue and fitness levels, nor in performance, during the 6
weeks of the study. The changes in serum L H concentra-
tions in individuals were significantly correlated
" 30 ( r = 0 . 9 7 , P < 0 . 0 1 ) to the fitness variations during the
period I (Fig. 4).
2o
=
.~ 10

"~ 0 I Discussion
~'~ 13 15 17 19 21 23
Week number
The data used in the present study were derived from a
Fig. 2. Serum concentration of SHBG, testosterone : SHBG (T: S) larger study concerning hormonal responses to weight-
ratio, referred to a week-15 level of 100 and daily averaged train-
ing loads (means and SD) during the 6-week preparatory period. lifting training (H~ikkinen et al. 1987) and only the sub-
Asterisks denote significant changes, *P<0.05; ** P<0.01. For jects who had a complete follow up of their training pa-
definitions see Table 2 rameters were selected for the present study, reducing
the group to six subjects. However, the hormone
changes observed in this reduced group were consistent
with those observed in the larger group studied by H~ik-
4O
kinen et al. 0987).
O
A decrease in serum testosterone concentration was
observed during the 4-week period of intensive training.
"~. 20
The observation of a depression in serum androgens was
in agreement with previous findings (Kuoppasalmi and
Adlercreutz 1985; Adlercreutz et al. 1986). A decrease in
serum testosterone concentration has already been ob-
served after a short period of intensive and extensive
training over several days (H~kkinen et al. 1988b) with-
out a significant change in performance. This decrease
-30 in the serum testosterone concentration during a short
training period of a few days could serve as an index of
-40 the physiological stress of training. In the present study,
0 20 40 60 80 1O0 120 140
Serum LH concentration the L H response during the 4-week intensive training pe-
(A% change during the period I) riod suggested that the depression in serum testosterone
Fig. 3. Comparisons between changes in serum L H concentration concentration was not mediated by an impairment in the
during period I and changes in testosterone:SHBG ratios during hypothalamic-pituitary function. The decrease in serum
period II (r= 0.90, P<0.05, n = 5). For definitions see Table 2 testosterone concentration seems to have been caused
primarily by an increased metabolism in the tissues or a
default in secretion. Some arguments have been pro-
posed in favour of a depressed production (Kuoppasal-

mi and Adlercreutz 1985) or a depletion in the testoste-
rone stores in the testis (Adlercreutz et al. 1986).
The decrease in serum testosterone concentration in-
duced a t e m p o r a r y decrease in the testosterone:SHBG
ratio at week 17. This ratio could be interpreted as an
index of the serum free testosterone concentration,
which has been shown to be the biologically active frac-
tion of the total testosterone (Kuoppasalmi 1980). The
t e s t o s t e r o n e : S H B G ratio regained its initial level at
week 19 and remained statistically unaltered during the
2 weeks of reduced training (weeks 19 to 21). H/ikkinen
et al. (1987) have observed a significant correlation be-
tween the changes in performance and in the testoste-
r o n e : S H B G ratio in the larger group of subjects. These
relationships were obtained between the weeks 15 and 17
and between the weeks 17 and 21. This was in agreement
with a previous observation (H/ikkinen et al. 1985). At
the end of an intensive training programme, the increase
in the testosterone : S H B G ratio seemed to be mainly due
to the decrease in serum S H B G concentration that oc-
curred between the weeks 17 and 19.
The nonsignificant change in serum androgen con-
centrations during the 2 weeks of reduced training
masked the behaviour in individuals during this period.
The systems model was used to determine the individu-
al's responses to training to examine the relationships
between the fatigue/fitness status of the subjects and the
individual's hormone responses. For this purpose period
I, composed of 4 weeks of intensive training, was distin-
guished from period II which was composed of 2 weeks
of reduced training.
A significant relationship was observed between the
changes in fitness and serum L H levels during the period
I. This finding would seem to indicate that the pituitary
response could be related to fitness level. But, the varia-
tions in fitness during the 6-week period resulted f r o m
the accrued effects of the previous training period. A
gain in fitness reflected a good adaptation to the amount
of training. Conversely, a decrease of fitness denoted a
weak adaptation to or an incomplete recovery f r o m pre-
vious training. Thus the magnitude of pituitary stimula-
tion induced by the depression of serum testosterone
concentration seemed to be dependent on the fatigue/
fitness status of the athlete. A incomplete recovery f r o m
preceding training could have reduced the L H release
during intensive training. This interpretation would bet-
ter match the results of Barron et al. (1985) who have
observed a dysfunction of the hypothalamic-pituitary
system in overtrained athletes. A 4-week period of rest
allowed them to recover normal hypothalamic-pituitary
function.
The differences in L H response were not associated
with differences in testosterone responses during period
I. However, the changes in L H during these 4 weeks
were significantly correlated to the changes in testoste-
r o n e : S H B G ratio during the following 2-weeks of re-
duced training. The enhancement of L H concentrations
seemed to be beneficial for androgenic-anabolic activity
only during the period of decreased physiological stress.
A low pituitary response during intensive training could
impair this positive effect on androgen concentrations.
385
This is in agreement with the depressed pituitary re-
sponse which has been noted in athletes suffering f r o m
persistently lowered testosterone concentrations (Hack-
ney 1989). Thus the negative effect of training on hor-
monal balance could be enhanced by the influence of
training on the hypothalamic-pituitary axis. However,
Viru (1984) has found the contrary to be the case in that
well-conditioned subjects were able to recover more
quickly and the supercompensation phenomenon oc-
curred more efficiently.
In conclusion, intensive training in elite weightlifters
was observed to be linked to a depression in serum tes-
tosterone concentrations. This depression was not pri-
marily mediated by an impairment of the hypothalamic-
pituitary system. However, the magnitude of the pituita-
ry response could have been related to the initial fati-
gue/fitness status of the subjects. An incomplete recove-
ry f r o m the preceding training loads could diminish the
L H response to depressed testosterone concentrations
and thus reduce the recovery capacity of some athletes.
However, in contrast, well-conditioned subjects exhibit
a better recovery of androgenic activity during a period
of reduced training.
References
Adlercreutz H, Harkonen M, Kuoppasalmi K, Naveri H, Huhta-
niemi I, Tikkanen H, Remes K, Dessypris A, Karvonen J
(1986) Effects of training on plasma anabolic and catabolic
steroid hormones and their response during physical exercise.
Int J Sports Med 7 : 27-28
Alen M, Pakarinen A, H/~kkinen K, Komi PV (1988) Responses of
serum androgenic-anabolic and catabolic hormones to pro-
longed strength training. Int J Sports Med 9:229-233
Banister EW, Hamilton CL (1985) Variations in iron status with
fatigue modelled from training in female distance runners. Eur
J Appl Physiol 54:16-23
Barron JL, Noakes TD, Levy W, Smith C, Millar RP (1985) Hy-
pothalamic dysfunction in overtrained athletes. J Clin Endo-
crinol Metab 60: 803-806
Busso T, H~kkinen K, Pakarinen A, Carasso C, Lacour JR, Komi
PV, Kauhanen H (1990) A systems model of training responses
and its relationship with hormonal responses in elite weightlift-
ers. Eur J Appl Physiol 61:48-54
Calvert TW, Banister EW, Savage MV, Bach T (1976) A systems
model of the effects of training on physical performance.
IEEE Trans Syst Man Cybern 6:94-102
Hackney AC (1989) Endurance training and testosterone levels.
Sports Med 8 : 117-127
H/~kkinen K, Pakarinen A, Alen M, Komi PV (1985) Serum hor-
mones during prolonged training of neuromuscular perform-
ance. Eur J Appl Physiol 53:287-293
Hakkinen K, Pakarinen A, Alen M, Kauhanen H, Komi PV
(1987) Relationships between training volume, physical per-
formance capacity, and serum hormone concentrations during
prolonged training in elite weightlifters. Int J Sports Med 8
[Suppl] : 61-65
H~tkkinen K, Pakarinen A, Alen M, Kauhanen H, Komi PV
(1988a) Neuromuscular and hormonal adaptations in athletes
to strength training in two years. J Appl Physiol 65:2406-
2412
H/~kkinen K, Pakarinen A, Alen M, Kauhanen H, Komi PV
(1988b) Daily hormonal and neuromuscular responses to inten-
sive strength training in 1 week. Int J Sports Med 9:422-428
386
Kuoppasalmi K (1980) Plasma testosterone and sex-hormone-
binding-globulincapacity in physical exercise. Scand J Lab In-
vest 40 : 411-418
Kuoppasalmi K, Adlercreutz H (1985) Interaction between cata-
bolic and anabolic steroid hormones in muscular exercise. In:
Fotherby K, Pal SB (eds) Exercise endocrinology, de Gruyter,
Berlin, pp 65-98
Kuoppasalmi K, Naveri H, Harkonen M, Adlercreutz H (1980)
Plasma cortisol, androstenedione, testosterone and luteinizing
hormone in running exercise of different intensities. Scand J
Lab Invest 40: 403-409
Mero A, Jaakkola L, Komi PV (1990) Serum hormones and
physical performance capacity in young boy athletes during a
1-year training period. Eur J Appl Physiol 60:32-37
Morton RH, Fitz-Clarke JR, Banister EW (1990) Modeling hu-
man performance in running. J Appl Physiol 69:1171-1177
Remes K, Kuoppasalmi K, Adlercreutz H (1979) Effect of long-
term physical training on plasma testosterone, androstene-
dione, luteinizing hormone and sex-hormone-binding-globulin
capacity. Scand J Lab Invest 39:743-749
Viru A (1984) The mechanism of training effects: a hypothesis. Int
J Sports Med 5:219-227