Endurance Training and the Menstrual Cycle
Effects of menstrual cycle based-training on physiologic
measures of endurance capacity and on muscle cell
parameters in women with and without oral contraception
Dissertation
zur Erlangung des Grades eines
Doktors der Sportwissenschaft (Ph.D. Exercise Science)
im Fach Sportmedizin
vorgelegt von
Ahreum Han
Ruhr-Universität Bochum
Fakultät für Sportwissenschaft
im Mai 2012
Erste Gutachterin: Prof. Dr. med. Petra Platen
Zweiter Gutachter: Prof. Dr. med. Wilhelm Bloch
ABSTRACT
ABSTRACT
PURPOSE: The menstrual cycle shows fluctuations in various endogenous
hormones between the follicular and the luteal phase in women, who do
not take oral contraceptives (OC). With the intake of OC biosynthesis and
secretion of the endogenous hormones estrogen and progesterone are
suppressed and other sex steroids are altered in different ways. Variations
of these hormonal milieus might influence trainability of endurance performance differently either between the menstrual cycle phases or between
women with oral contraceptives and women without oral contraceptives.
Therefore, this thesis aimed to investigate hormone profiles during the
menstrual cycle and the effects of menstrual cycle phase-based endurance training on physiologic and microscopic measures of aerobic capacity in eumenorrheic women, who do not take any oral contraceptive (nonOC users) and in women, who take a combined monophasic oral contraceptive (OC users).
Study 1 investigated follicular phase-based (FT) vs. luteal phase-based
(LT) endurance training in non-OC users.
Study 2 investigated quasi-follicular phase-based (qFT) vs. quasi-luteal
phase-based (qLT) endurance training in monophasic OC users.
Study 3 compared hormonal profiles and parameters of endurance performance from both studies between non-OC users and OC users.
METHODS: Thirteen non-OC users and fourteen OC users completed oneleg endurance training on a cycle ergometer for three menstrual cycles.
One leg was trained mainly in the first half of the menstrual cycle (follicular
phase training (FT) and quasi-follicular phase training (qFT), respectively)
and the other leg mainly in the second half of the cycle (luteal phase training (LT) and quasi-luteal phase training (qLT), respectively). Venous blood
samples were taken on day 11 of the menstrual cycle in the follicular
phase (FP) / quasi-follicular phase (qFP) and on day 25 of the menstrual
cycle in the luteal phase (LP) / quasi-luteal phase (qLP) to analyze values
of 17-beta estradiol (E2), progesterone (P4), total testosterone (T), free
i
ABSTRACT
testosterone (free T) and DHEA-s. Peak oxygen uptake (VO2peak), maximal
workload (Wattmax) and power output at a lactate concentration of 4mmol/l
(Wattlac4) were analyzed before and after training intervention as well as
muscle fiber composition (number of type I and type II fibers), fiber diameter (Fdm) and cell nuclei to fiber ratio (N/F) in subgroups of five and three
subjects, respectively.
RESULTS: Study 1: Prior to training, concentration of free T was higher in
FP compared to LP (P < 0.05). VO2peak increased after FT (P = 0.033) and
after LT (P = 0.038) without any difference between FT and LT. Wattmax
also increased significantly after both FT and LT. Moreover, we found a
noticeably higher increase (P = 0.038) of Wattmax after FT (+ 34.8 W) compared to LT (+ 27.8 W). There seemed to be a slight higher increase in
type I Fdm compared to type II Fdm with a slightly more pronounced increase in type I Fdm after FT compared to LT.
Study 2: DHEA-s was higher (p < 0.05) in qFP compared to qLP. E2, T
and P4 were not significantly different between two phases. VO2max and
Wattmax increased significantly after qFT and qLT without any difference
between both training periodizations. Mdm did not change after qFT and
qLT. Number of type I fibers decreased after qFT whereas number of type
I fibers increased after qLT. No relevant changes were found in N/F and in
Fdm of type I and type II fibers neither after qFT nor after qLT.
Study 3: Concentration of E2, T and free T were significantly higher in
non-OC users compared to OC users (p<0.05). P4 level was highest in LP
compared to all other phases (p<0.05). Absolute increase of Wattmax was
the lowest after LT in non-OC users (+ 27.8 W) compared to FT (+ 34.8 W),
qFT (+ 32.3 W) and qLT (+ 35.9 W) in OC users. VO2peak tended to increase after FT (+3.3 ml/min/kg, P = 0.033) and LT (+2.8 ml/min/kg, P =
0.038) and increased significantly after qFT (+5.7 ml/min/kg) and qLT
(+5.3 ml/min/kg) without any differences between non-OC and OC-users.
CONCLUSIONS: In non-OC users, FT showed a slight and delayed higher effect on maximum power output on a bicycle ergometer compared to LT
without any different effect on VO2peak and muscle diameter. This might be
due to the specific hormonal milieu during each phase of the cycle. In OCusers, however, no differences were found between the two training interii
ABSTRACT
ventions qFT and qLT. This is presumably due to the constant doses of
estrogen and progestin in monophasic OC. As a result, OC users had a
more stable hormonal milieu for training adaptation processes at least during the consumption phase of 21 days, resulting in comparable trainability
of endurance performance throughout the cycle.
Further studies with longer lasting training periods are needed in order to
analyze if the late response in aerobic training adaptation becomes more
pronounced after more than three months of menstrual cycle-based training in non-OC users. As no menstrual cycle-specific training responses
have been observed, we recommend that untrained and moderately
trained OC users perform their endurance training independently from
their pill cycle. Further studies are necessary in order to understand possible effects of androgenicity of OC pills on development of endurance performance. Furthermore, more subjects have to be included in muscle biopsy analyses in order to understand possible underlying mechanisms of
cycle-dependent aerobic training adaptations.
iii
ACKNOWLEDGMENTS
ACKNOWLEDGMENTS/DANKSAGUNG
Mit der Fertigstellung der Dissertationsschrift ist es an der Zeit, nochmals
denjenigen zu danken, die mich begleitet und unterstützt haben. Daher
möchte ich diese Gelegenheit nutzen, um meinen tiefen Dank zum
Ausdruck zu bringen.
Besonderer Dank gilt dabei meiner Doktormutter Frau. Prof. Dr. med.
Petra Platen, für die Begleitung meines Promotionsverfahrens und
wissenschaftlichen Betreuung meines Dissertationsprojektes. Sie hat nicht
nur dessen Fortgang durch kritische wie inspirierende Fachdiskussionen
bereichert, sondern mich auch im Hinblick auf meine fachliche und
berufliche Weiterentwicklung stets gefördert.
Ganz großer Dank geht an Herrn Dr. med. Timo Hinrichs, der mit seiner
Betreuung und seinem Engagement großen Anteil am Zustandekommen
dieser Dissertation hat.
Ein herzliches Dankeschön geht ebenso an Herrn Dr. Ulrich Bartmus
sowie Frau Ella Rau, welche in privaten Gesprächen immer dafür gesorgt
hat, dass ich meinen Geist angestrengt habe und die intensiven Zeiten
erfolgreich meistern konnte.
Ein
ganz
besonderer
Dank
geht
an
meine
Mitdoktorandin/
Projektmitarbeiterin, Eunsook Sung, für die außerordentlich fruchtbare
Zusammenarbeit. Diese Arbeit wäre ohne ihre Mitarbeit nicht möglich
gewesen, weshalb ich mich herzlich bedanke.
Die Menschen, die sicherlich am geduldigsten mit mir während des
Promotionsverfahrens waren, und mich privat am meisten unterstützten,
waren meine Eltern und meine Schwester. Ihnen danke ich von Herzen,
dass sie mir das Studium in Deutschland ermöglichten und mir immer
wieder die Kraft zum Durchhalten in schwierigen Phasen gegeben haben.
Allen meinen lieben Freunden danke ich für die Ausdauer, Ruhe und
Geduld, womit sie mir stets zur Seite standen und mich beständig
aufgemuntert haben.
Bochum, den 15. Mai 2012
Ahreum Han
iv
ACKNOWLEDGMENTS
감사의 글
짧지 않은 10 년이라는 시간을 독일에서 보내면서 석사과정을 마치고 또
박사논문을 무사히 제출할수 있게 여러방면으로 힘이 되어주신 한국에
계신분들께 감사의 글을 올립니다.
먼저 저에게 독일에서 학업의 문을 열게 해주신 이화여자대학교 모교에
계신 김경숙 교수님께 진심으로 감사드립니다. 유학의 첫발을 딛을 때부터
지금까지 아낌없는 조언과 격려가 큰 힘이 되었습니다.
석사과정을 하는동안 하늘나라로 가신 외할아버지, 외할머니, 그리고
유학의 마무리를 끝까지 지켜보지 못하시고 얼마전 이 세상과 작별을 하신
친할머니께 이 논문을 바칩니다. 사랑하고 너무도 보고싶습니다.
그 무엇보다도, 또 그 누구보다도 가장 힘이되준 가족들. 긴 유학생활을
믿음과 사랑으로 인내해주고 지원해주고 또 항상 용기와 힘을 붇돋아 주신
부모님과 동생이 있었기에 10 년이라는 독일의 유학생활을 견디고
해낼수있었습니다. 엄마 아빠 감사합니다. 그리고 사.랑. 합니다.
그리고 옆에서 가족처럼 물심양면으로 항상 큰 도움이 되준 고은양,
지치고
힘들때
마음의
활력소가
되어준
사랑하는
이화여자대학교
사회체육학과 97 학번 동기들, 선후배님들, 엘베가족 그리고 그 외 멀리서
힘을 전해준 모든 친구들에게도 고맙다는 말을 전하고 싶습니다.
2012 년 5 월 15 일 독일 보훔에서
한 아 름
v
CONTENTS
CONTENTS
ABSTRACT………………………………………………………………………….i
ACKNOWLEDGMENTS…………………………………………………...…...……iv
CONTENTS………………………………………………………………….……..vi
LIST OF ABBREVIATIONS……………...…….………………………………….…ix
LIST
OF
TABLES………………………………………………………….………..x
LIST OF FIGURES…….………………………….….………………..….…….….xi
INTRODUCTION……………………..………….…….……………………………1
STUDY 1: EFFECTS OF FOLLICULAR VERSUS LUTEAL PHASE-BASED ENDURANCE
TRAINING IN UNTRAINED WOMEN
1.1
Introduction…………………………….…………...…………...…...4
1.2
Methods…………………………………….……..………………….7
1.2.1 Subjects…………………………………………………………..7
1.2.2 Experimental design…………………………………….………7
1.2.3 Study schedule………………………………..……………….8
1.2.3.1
Monitoring of menstrual cycle integrity………………8
1.2.3.2
Endurance training program….…………………….…..9
1.2.3.3
Hormone analysis……………………………………….9
1.2.3.4
Physiologic measures of endurance capacity………10
1.2.3.5
Measurement of isometric muscle strength…………10
1.2.3.6
Determination of muscle diameter……………………11
1.2.3.7
Histochemical analysis of muscle samples………….12
1.2.4 Statistical Analysis……………………………………………..13
1.3
Results…………………………………….………………………...14
1.3.1 Menstrual cycle integrity………………………………………14
1.3.2 Number of training sessions…………………………….…….14
1.3.3 Hormone concentrations…..………….………………………14
1.3.4 Peak oxygen uptake (VO2peak) …………………….…………15
1.3.5 Maximum workload (Wattmax)…………………………………16
vi
CONTENTS
1.3.6 Submaximal power output at a lactate concentration of
4 mmol/l (Wattlac4)……………………………………….……..18
1.3.7 Isometric muscle strength (Fmax)………………………..……19
1.3.8 Muscle diameter (Mdm)……………………………………….19
1.3.9 Muscle fiber characteristics………………………………...…20
1.4
Discussion…………………………………………………….…….23
STUDY 2: EFFECTS
OF
MENSTRUAL PHASE-BASED ENDURANCE TRAINING
IN
ORAL CONTRACEPTION USERS
2.1
Introduction…………………………….…………...……………....30
2.2
Methods…………………………………….……..………………..33
2.2.1 Subjects…………………………………………………………33
2.2.2 Experimental design………………………..……………….. 34
2.2.3 Study schedule………………………………..………………35
2.2.3.1
Endurance training program….…………………….…35
2.2.3.2
Hormone analysis…………………………………….35
2.2.3.3
Physiologic measures of endurance capacity………36
2.2.3.4
Measurement of isometric muscle strength…………36
2.2.3.5
Determination of muscle diameter……………………37
2.2.3.6
Histochemical analysis of muscle samples………….38
2.2.4 Statistical Analysis……………………………………………39
2.3 Results…………………………………….………………………..40
2.3.1 Number of training sessions…………………………….…….40
2.3.2 Hormone concentrations…..………….………………………40
2.3.3 Peak oxygen uptake (VO2peak) …………………….…………41
2.3.4 Maximum workload (Wattmax)…………………………………43
2.3.5 Submaximal power output at a lactate concentration of
4 mmol/l (Wattlac4)……………………………………………45
2.3.6 Isometric muscle strength (Fmax)……………………………47
2.3.7 Muscle diameter (Mdm)……………………………………….48
2.3.8 Muscle fiber characteristics………………………………..…48
2.4
Discussion…………………………………………………….…….49
vii
CONTENTS
3
STUDY 3: COMPARING NON-OC USERS VERSUS OC USERS
3.1
Introduction..……………………….……..……...…………….......55
3.2
Methods…………………………………….……..………………..59
3.2.1 Statistical Analysis………………………………………….…59
3.3
Results…………………………………….……………………….60
3.3.1 Number of training sessions…………………………….…….60
3.3.2 Hormone concentrations…..………….………………………60
3.3.3 Peak oxygen uptake (VO2peak) …………………….…………61
3.3.4 Maximum workload (Wattmax)…………………………………61
3.3.5 Submaximal power output at a lactate concentration of
4 mmol/l (Wattlac4)……………………………………………..63
3.3.6 Isometric muscle strength (Fmax)…………………………..…64
3.3.7 Muscle diameter (Mdm)……………………………………….65
3.4
Discussion…………………………………………………….…….66
4 S UMMARY…………...…………………………………………………….69
5 R EFERENCES ……………………………………….…………………….71
2.
viii
LIST OF ABBREVIATIONS
LIST OF ABBREVIATIONS
∆
AMPK
ATP
DHEA-s
E
E2
E/P
Fdm
FFA
Fmax
FP
Free T
FT
GLUT-4
ICC
LP
LPL
LT
m.
Mdm
min
mTOR
N
(N)
No%
N/F
OC
P4
post
PPAR-d
pre
qFP
qFT
qLP
qLT
rpm
s
SD
T
VO2max
VO2peak
W
Wattlac4
Wattmax
yr
absolute difference
5’-AMP-activated protein kinase
adenosine triphosphate
dehydroepiandrosterone-sulfate
estrogen
17-beta estradiol
estrogen to progesterone ratio
muscle fiber diameter
free fat acid
maximum isometric strength
follicular phase
Free testosterone
follicular phase-based endurance training
glucose transporter type 4
intraclass correlation coefficient
luteal phase
Lipoprotein lipase
luteal phase-based endurance training
muscle
diameter of m.quadriceps femoris
minute
mammalian target of rapamycin
number
Newton
Muscle fiber composition
cell nuclei-to-fiber ratio
oral contraceptive
progesterone
after
peroxisome proliferation activator receptor-d
before
quasi-follicular phase
quasi-follicular phase-based endurance training
quasi-luteal phase
quasi-luteal phase-based endurance training
round per minute
second
standard deviation
total testosterone
maximum oxygen uptake
peak oxygen uptake
watt
workload by lactate concentration of 4mmol/l
maximum workload
year
ix
LIST OF TABLES
LIST OF TABLES
TABLE 1-1: Serum concentrations of E2, P4, DHEA-s, T and free T
TABLE 1-2: Maximum isometric force (Fmax) and Sum of muscle diameter (Mdm) (m. rectus femoris, m. vastus intermedius, m. vastus lateralis)
TABLE 1-3a: Muscle fiber type distribution: relative number of fiber (F%no) and relative
fiber area (F%area)
TABLE 1-3b: Muscle fiber type distribution: fiber diameter (Fdm) and nuclei-to-fiber ratio
(N/F)
TABLE 2-1: Monophasic oral contraceptive pills used by the subjects of this study including doses of ethinylestradiol and gestagen and their possible androgenicity index
TABLE 2-2: Serum concentrations of E2, P4, DHEA-s, T and free T
TABLE 2-3: Maximum isometric force (Fmax) and sum of muscle diameter (Mdm) of m.
rectus femoris, m. vastus intermedius and m. vastus lateralis. Fmax and Mdm in
the two groups of subjects taking OC without any androgenicity or with known
androgenicity
TABLE 2-4: Muscle fiber type distribution (No in %), fiber diameter (Fdm) and nuclei-tofiber ratio (N/F)
TABLE 3-1: Serum concentrations of E2, P4, DHEA-s, T and free T
TABLE 3-2: Absolute increase of maximum isometric force (Fmax) and Sum of muscle diameter (Mdm) (m. rectus femoris, m. vastus intermedius, m. vastus lateralis)
x
LIST OF FIGURES
LIST OF FIGURES
FIGURE 1-1: VO2peak FT, LT
FIGURE 1-2: Increase in Wattmax compared to the pre-training values FT, LT
FIGURE 1-3: Maximum workload (Wattmax) FT, LT
FIGURE 1-4: Increase in Wattlac4 compared to the pre-training values
FIGURE 1-5: Power output at a lactate concentration of 4 mmol/l (Wattlac4) FT, LT
FIGURE 2-1: VO2peak qFT, qLT
FIGURE 2-2: VO2peak in two groups of subjects taking OC without any androgenicity (N
= 7) or with known androgenicity (N = 7) qFT, qLT
FIGURE 2-3: Increase in Wattmax compared to the pre-training values qFT, qLT
FIGURE 2-4: Maximum workload (Wattmax) qFT, qLT
FIGURE 2-5: Maximum workload (Wattmax) in two groups of subjects taking OC without
any androgenicity (N = 7) or with known androgenicity (N = 7) qFT, qLT
FIGURE 2-6: Increase in Wattlac4 compared to the pre-training values qFT, qLT
FIGURE 2-7: Power output at a lactate concentration of 4 mmol/l (Wattlac4) qFT, qLT
FIGURE 2-8: Power output at a lactate concentration of 4 mmol/l (Wattlac4) in two
groups of subjects taking OC without any androgenicity (N = 7) or with known
androgenicity (N = 7) qFT, qLT
FIGURE 3-1: Absolute increases (∆) of VO 2peak
FIGURE 3-2: Increase in Wattmax compared to the pre-training values
FIGURE 3-3: Absolute increases (∆) of maximum workload (Watt max)
FIGURE 3-4: Progressive Increases in Wattlac4 compared to the pre-training values
FIGURE 3-5: Absolute increases ∆)
( of Power output at a lactate concentration of 4
mmol/l (Wattlac4)
xi
INTRODUCTION
INTRODUCTION
The fluctuation of hormones during the menstrual cycle may influence exercise performance and trainability of muscle endurance (Lebrun 1994,
Janse de Jonge, 2003, Constantini, Dubnov & Lebrun, 2005).
Women between the ages of approximately 13 and 50 experience a
circamensal rhythm referred to as the menstrual cycle, in which the ovarian hormones fluctuate predictably over 23–38 days on average (Oosthyse
et al. 2010, Reilly 2000). In addition to ovarian hormones such as 17-beta
estradiol (E2) and progesterone (P4), androstenedione and testosterone
also fluctuate over the menstrual cycle (Longcope 1986).
Recently, the use of oral contraceptives (OC) is increasing. The number of
female athletes using OC is also increasing for reasons like birth control,
management of premenstrual symptoms, dysmenorrhea, less menstrual
blood loss, lower risk of musculoskeletal injury and time-shifting of the
menstrual cycle, which could provide benefits for the female athletes
(Bennell, White & Crossley, 1999; Constantini, Dubnov & Lebrun, 2005;
Wojtys, Huston, Boynton, Spindler & Lindenfeld, 2002). Due to the intake
of fixed doses of synthetic E2 and P4 in OC, endogenous E2 and P4 are
suppressed in women using OC.
Since E2, P4 and other sex steroids are discussed to be important factors
for endurance capacity, there might be yet unknown different influences on
endurance training adaptation in both non-OC users and OC users. To the
authors’ knowledge, there are no controlled training interventional studies
that have assessed the trainability of endurance capacity in two phases of
menstrual cycles in both non-OC users and OC users.
The aim of the present thesis, therefore, was to investigate the hormone
profile in the follicular phase and the luteal phase of the menstrual cycle
and the effects of two different menstrual phase-based endurance trainings - follicular phase-based training (FT) versus luteal phase-based training (LT) - on physiologic and microscopic measures of aerobic capacity.
1
INTRODUCTION
This thesis on ‘Endurance Training and the Menstrual Cycle’ is one part of
a series of studies on ‘Trainability and the Menstrual Cycle’. The other part
was on ‘Strength Training and the Menstrual Cycle’ and was carried out by
Ms Eunsook Sung.
The present thesis on ‘Endurance Training and the Menstrual Cycle’ is divided into three single studies. These three studies have the same experimental designs and methods in different subject groups.
Study 1 examined thirteen eumenorrheic women who did not take any
oral contraceptives (non-OC users).
Study 2 examined fourteen women who took combined monophasic oral
contraceptives (OC users).
Study 3 compares the results of both studies between non-OC users and
OC users.
2
STUDY 1-ABSTRACT
STUDY 1: EFFECTS
OF
FOLLICULAR
VERSUS
LUTEAL PHASE-BASED
ENDURANCE TRAINING IN UNTRAINED WOMEN
Abstract
Purpose: Hormonal variations during the menstrual cycle may influence
trainability of endurance. For this reason, we investigated the effects of follicular phase-based (FT) endurance training on aerobic capacity, power
output and microscopic muscle parameters, comparing it to luteal phasebased (LT) endurance training.
Methods: Eumenorrheic women without oral contraception (N = 13) completed one-leg endurance training on a cycle ergometer for three menstrual
cycles. They trained one leg mainly in the follicular phase (FP) and the other leg in the luteal phase (LP). Concentrations of 17-beta estradiol (E2),
progesterone (P4), total testosterone (T), free testosterone (free T) and
DHEA-s were analyzed in blood samples taken during FP and LP. Peak
oxygen uptake (VO2peak), maximal workload (Wattmax), power output at a
lactate concentration of 4mmol/l (Wattlac4), maximum isometric strength of
knee extension (Fmax) and diameter of m. quadriceps femoris (Mdm) were
analyzed before and after training, as well as muscle fiber composition
(No%), fiber diameter (Fdm) and cell nuclei-to-fiber ratio (N/F) in a subgroup of three subjects.
Results: Prior to training, concentration of free T was higher in FP compared to LP (P < 0.05). VO2peak increased after FT (P = 0.033) and after LT
(P = 0.038) without any difference between FT and LT. Wattmax also increased significantly after both FT and LT. Moreover, we found a noticeably
higher increase (P = 0.038) of Wattmax after FT (+ 34.8 W) compared to LT
(+ 27.8 W). Fmax tended to decline after FT and LT, and Mdm remained unchanged. There might be a slightly higher increase in type I Fdm compared
to type II Fdm with a slightly more pronounced increase in type I Fdm after
FT compared to LT.
Conclusions: FT showed a slightly higher but delayed effect on maximum
power output on a bicycle ergometer compared to LT without any different
effect on VO2peak and muscle diameter. This might be due to the specific
hormonal milieu during each phase of the cycle. Further studies with longer
3
STUDY 1-INTRODUCTION
lasting training periods are needed in order to analyze if this late response
in aerobic training adaptation becomes more pronounced after three
months of training. Furthermore, more subjects have to be included in muscle biopsy analyses in order to understand possible underlying mechanisms
of cycle-dependent aerobic training adaptations on the molecular level of
the female skeletal muscle.
1.1
Introduction
Women between the ages of approximately 13 and 50 experience a
circamensal rhythm referred to as the menstrual cycle, in which the ovarian
hormones fluctuate predictably over 23–38 days on average (Oosthuyse &
Bosch, 2010, Reilly, 2000). 17-beta estradiol (E2) peaks prior to ovulation
and during the luteal phase (LP), while progesterone (P4) reaches its highest values during LP after ovulation (Van Look & Baird, 1980). In both sexes, androgens are produced by the reproductive organs and the adrenals.
The most important androgen secreted is testosterone; the adrenal glands
and the ovaries produce very little testosterone but secrete weaker androgens. In particular, dehydroepiandrosterone (DHEA; and its sulfoconjugate)
secreted by the adrenals, and androstenedione secreted by the adrenals
and the ovaries are of physiological importance in women (Enea, Boisseau,
Fargeas-Gluck, Diaz & Dugue, 2011). In addition to E2 and P4, androgens
also fluctuate over the menstrual cycle. The levels of androstenedione and
testosterone, for instance, reach their peaks prior to, or at the time of ovulation (Longcope, 1986).
The fluctuation of hormones during the menstrual cycle may influence exercise performance and trainability of muscle endurance (Lebrun 1994, Janse
de Jonge, 2003, Constantini, Dubnov & Lebrun, 2005). Studies on variations of endurance performance throughout the menstrual cycle are contradictory. Although a number of studies have found endurance performance
to vary between menstrual phases, there are an equal number of such
studies reporting no differences. However, strong evidence from animal research, specifically for estrogen-induced promotion of better endurance ca4
STUDY 1-INTRODUCTION
pacity, and various menstrual phase-associated metabolic perturbations
strongly suggest the assumption of menstrual-cycle dependent endurance
capacity (Oosthuyse & Bosch, 2010). Several studies have investigated the
effect of estrogen and, to a lesser extent, progesterone on substrate metabolism during exercise. Men and women to whom estrogen is administered show improved endurance performance, while muscle and liver glycogen are spared and fat oxidation is increased. Consistent with these findings, high estrogen levels during the LP are associated with the sparing of
muscle glycogen in comparison to the follicular phase (FP) (Isacco, Duche
& Boisseau, 2012). In animal studies, ovarian hormones have shown to
play a key role in the activity of enzymes of fatty acid oxidation.
Ovariectomization in rats is associated with a decrease in enzymes involved in fatty acid oxidation, which are restored by estrogen treatment but
not progesterone treatment. When the two hormones were combined, the
effect on enzymatic activities was the same as ovariectomization, showing
the ability of progesterone to inhibit estrogenic effect (Isacco, Duche &
Boisseau, 2012).
Endurance exercise results in the integration of a vast number of physiological and anatomical systems, including the cardiorespiratory, circulatory,
musculoskeletal, and neuroendocrine systems, to produce coordinated,
sustained movement while minimizing perturbations to homeostatic equilibrium. These systems adaptations occur as a result of changes on the tissue
and cellular level, which result in a greater ability to engage in prolonged
exertion that primarily represents greater uptake, delivery, and use of oxygen and nutrients, and removal of metabolic byproducts. Adaptations on the
cellular level in turn depend on qualitative and quantitative shifts in gene
expression resulting in accumulation of specific proteins. The gene expression that allows for these changes in protein concentration is pivotal to the
training adaptation (Allen, Harrison & Leinwand, 2002, Hansen et al., 2005).
Muscle glycogen, which might be affected by estrogens and progesterone,
is a determining factor for the transcription of some genes. Exercising,
when muscle glycogen concentration is low results in a greater transcriptional activation of enzymes and other factors compared with when muscle
5
STUDY 1-INTRODUCTION
glycogen concentration is high or normal at the start of exercise (Hansen et
al., 2005).
5’-AMP-activated protein kinase (AMPK) is the major cellular energy regulator driving metabolic processes to promote ATP production. Increased
AMPK activity corresponds with increased GLUT4 content, contractionstimulated glucose uptake and increased cellular fatty acid uptake. Estrogen is thought to stimulate AMPK activity. Furthermore, estrogen upregulates the transcription factor peroxisome proliferation activator receptor-d (PPAR-d) in muscle, which leads to the increased expression of various enzymes (LPL, pyruvate dehydrogenase kinase, acyl-CoA oxidase,
and uncoupling protein 2 and 3), which promotes energy dissipation and
the oxidation of FFA (Oosthuyse & Bosch, 2010).
All these data indicate, that steroid hormones might influence adaptation
processes during periods of endurance training via alterations in substrate
metabolism and activation of regulatory factors in humans. However, no
such endurance training studies with eumenorrheic women are available in
the literature. We could recently demonstrate in an interventional three
months strength training study, that follicular phase-based resistance training showed a higher effect on muscle strength adaptation, muscle diameter
and diameter of fiber type ΙΙ compared to a luteal phase-based training
(Sung et al., 2012). As the adaptive responses of muscle to exercise are
specific to the training mode (Gravelle & Blessing, 2000) aerobic training intervention studies are mandatory for the further investigation of possible influences of the fluctuations of steroid hormones throughout the menstrual
cycle on the development of endurance performance.
The aim of this study was to investigate the effects of a longer-lasting follicular phase-based (FT) endurance training on metabolic, macroscopic and
microscopic parameters of skeletal muscle adaptations compared to luteal
phase-based (LT) endurance training in an in vivo controlled training intervention study in healthy young females.
6
STUDY 1-METHODS
1.2
Methods
1.2.1 Subjects
Thirteen healthy eumenorrheic women, with a mean (± SD) age of 25.1 ±
3.6 yr, height of 164.8 ± 6.8 cm and weight of 67.7 ± 17.7 kg volunteered to
participate in this study. Subjects were untrained or moderately trained and
they were not currently performing endurance training. Moreover they had
not been taking oral contraceptives or any other hormonal treatments during the year prior to participation in this study and had no history of any endocrine disorders. Only women who reported a regular menstrual cycle
were recruited. At the beginning, 14 subjects participated in this study;
however, we had to drop one subject out because of a low P4 level in LP,
indicating luteal-phase insufficiency or anovulatory menstrual cycles.
Prior to the study, participants were informed about the purpose, procedures and risks of the study and written informed consent was obtained
from each participant. Approval for the experimental protocol was obtained
from the Ethics Committee of the Ruhr-University Bochum, Germany.
1.2.2 Experimental design
Participants performed an endurance training program on a cycle ergometer, separately for each leg, over a period of three menstrual cycles each.
Subjects were randomly divided to two groups in order to reduce effects of
leg preference: one group (N = 8) mainly trained the left leg during the follicular phase (FT), while the right leg was mainly trained during the luteal
phase (LT). The other group (N = 5) mainly trained the right leg during FP
(FT), while the left leg was trained during LP (LT). For further analysis, both
follicular phase-trained leg groups and both luteal phase-trained leg groups
were taken together in the FT- or LT-trained leg group, respectively.
The duration of the study for each participant was based on the individual
length of the menstrual cycle. The entire study took five menstrual cycles
(two control cycles followed by three training cycles). During the overall
7
STUDY 1-METHODS
study period, individual cycle integrity was analyzed by daily measurements
of basal body temperature.
In the second control cycle and in the third training cycle, blood samples for
hormone analysis were taken from a cubital vein on day 11 (late FP) and on
day 25 (late LP) of the menstrual cycle. Additionally, incremental bicycle ergometer tests until exhaustion were carried out on days 11 and 23 (single
leg tests) and on day 25 (both leg test), and maximum isometric strength of
knee extension (Fmax) was measured on day 25 after two-leg incremental
bicycle ergometer test. Furthermore, the diameter (Mdm) of three single
muscles of the quadriceps muscle was measured and muscle biopsies
were taken from the vastus lateralis muscle on day 27 (late LP) in a subgroup of three subjects. During the first and second training cycle, the incremental bicycle ergometer tests were carried out as maximal ergometer
tests on days 11 and 23 (single leg tests) and on day 25 (both leg test).
1.2.3 Study schedule
1.2.3.1
Monitoring of menstrual cycle integrity
The fluctuation of basal body temperature was used to identify the phases
of the menstrual cycle including ovulation in order to individually determine
the exact training and testing schedule. Subjects were instructed to measure their basal body temperature orally with a digital thermometer for one
minute every morning throughout the entire study period at the same time
before getting out of bed. The occurrence of ovulation was defined when an
increase in basal body temperature of at least 0.3 ºC was measured (Kelly
2006, Owen 1975). A subject was excluded from the study if no significant
increase in basal body temperature, i.e. no ovulation, was detected during
any of the five menstrual cycles.
8
STUDY 1-METHODS
1.2.3.2
Endurance training program
The subjects completed three menstrual cycles of a one-leg endurance
training program with different training quantities of the right and left leg in
FP and LP, respectively, while the total number of single-leg training sessions in one menstrual cycle remained the same. The training was performed three times a week (typically on Monday, Wednesday and Friday)
under supervision on a bicycle ergometer (Ergoselect 100, Ergoline GmbH,
Germany). One leg was mainly trained in FP (FT) and the other leg mainly
in LP (LT). In FT, subjects trained six times in the follicular phase and
around ovulation (typically between day 1 and day 14) and just twice in the
luteal phase. For LT, they trained six times in LP (typically between day 15
and day 28) and just twice in FP. When the individual cycle lasted less than
28 days, the number of training sessions was adapted accordingly so that
the total number of sessions was the same for both legs. When the cycle
lasted longer than 28 days and the number of single-leg training sessions in
LT reached the number in FT (e.g. typically N = 8), subjects continued their
endurance training with both legs for another one or two sessions to avoid
differences in the total number of training sessions for one leg between FT
and LT.
Subjects performed one-leg training for 60 minutes at the work load corresponding to 75% of the predetermined one-leg power corresponding to 4
mmol/l blood lactate concentration. On the days on which both legs had to
be trained, subjects performed two-leg exercise with a work load adjusted
to 75% of the predetermined two-leg power corresponding to 4 mmol/l
blood lactate concentration. The workloads were adjusted upwards every
four weeks according to the results of the last respective maximal bicycle
ergometer tests.
1.2.3.3
Hormone analysis
Venous blood was centrifuged after blood clotting, and the serum was kept
frozen at -80° C until analysis. Each sample was analyzed for E2, P4, total
testosterone (T) and free T, and dihydrotestosterone-sulfate (DHEA-s). E2,
9
STUDY 1-METHODS
P4, T, and DHEA-s were assayed by immunochemistry (Elecsys® 1010
System, Roche Diagnostics GmbH), and free T was assayed by radioimmunoassay (Multi-Crystal LB 2111 gamma counter, Berthold Technologies
GmbH & Co. KG).
1.2.3.4
Physiologic measures of endurance capacity
Right and left leg and two-leg incremental tests to exhaustion were performed on a cycle ergometer (Ergoselect 100, Ergoline GmbH, Germany)
with an open air spirometry system (ZAN 600 USB, nSpire Health,
Oberthulba, Germany) in the second control cycle and the third training cycle in order to determine 1. peak oxygen uptake (VO2peak), 2. maximum
workload (Wattmax) and 3. power output at a lactate concentration of 4
mmol/l in capillary blood (Wattlac4). Each right and left leg was measured on
separate days. The leg which mainly trained during the follicular phase was
measured in the late FP (day 11). The other leg which mainly trained in the
luteal phase was analyzed in the late LP (day 23). The two-leg test was
performed two days after the second test (day 25). The initial workload for
the one-leg incremental test was 20 W and the work rate was increased by
10 W every one minute until the subject could not maintain the required
pedaling frequency (>50 rpm). For the two-leg test the initial workload was
set at 25 W and increased by 25 W every two minutes. Capillary blood
samples for the measurement of blood lactate concentration (Ebio plus,
Eppendorf AG, Hamburg, Germany) were taken from an earlobe at rest,
every two minutes during the test and immediately after termination of the
test. During the first and second training cycle right and left leg and two-leg
maximal incremental tests were performed without spirometry as described
above in order to determine (Wattlac4) and to adjust training intensity, respectively.
1.2.3.5
Measurement of isometric muscle strength
Maximum isometric knee extension muscle strength (Fmax) of the right and
left leg was measured in the late LP (day 25) in the second control cycle
10
STUDY 1-METHODS
and the third training cycle. Fmax was determined on a leg press machine
(Medizinische Sequenzgeräte, Compass, Germany) using a combined
force and load cell (GSV-2ASD, ME-Messsysteme GmbH, Hennigsdorf,
Germany). The intraclass correlation coefficient of repeated measurements
(ICC) was 0.998, indicating a high internal consistency (reliability) of the
system. Subjects were familiarized with the test procedure and the testing
position (knee angle: 90°, ankle angle: 90°) on the leg press 30 min after finalization of the bicycle ergometer tests. No further warm-up was necessary. Each measurement was repeated three times with 30 s rest between
the tests. The best result was selected for data analysis.
1.2.3.6
Determination of muscle diameter
Mdm of rectus femoris, vastus intermedius and vastus lateralis muscle of
the right and left leg was measured by real-time ultrasound imaging prior to
and after training at day 25 in LP of the second control cycle and the third
training cycle analyzing the distances between the outer and inner muscle
fasciae. Previous studies showed that muscle cross-sectional area might
reliably be measured using real-time ultrasound imaging (Martinson &
Stokes, 1991). We used a Vivid I CE 0344 ultrasound device (GE Medical
System, Solingen, Germany) with a parallel scanner (8L-RS, 4.0–13.3
MHz), which provides 10 cm penetration depth of the sound wave and enables high quality analysis of deeper lying muscles. Subjects prevented longlasting static muscular tension for at least 30 minutes prior to the measurement in order to avoid alterations in Mdm (Reimers, 2004). All subjects lay
supine with outstretched legs on an examination table without any pad,
cushion or pillow underneath. Ultrasound images were obtained exactly
half-way between the spina iliaca anterior superior and the upper margin of
the patella. The transducer was placed gently on the skin to avoid compression and distortion of the underlying tissue (Reimers, 2004). The transducer was held at angles of 90° towards the skin and towards the longitudinal direction of the muscles to ensure a clear cross-sectional image. The
images were frozen on the screen to measure muscle diameter. The position of the transducer was recorded for each muscle to reproduce the exact
11
STUDY 1-METHODS
position after training intervention. The mean of three measurements of
each of the three analyzed muscles was taken for both legs and the sum of
the 3 Mdm was calculated for both sides of the body. Reliability analysis
was performed for Mdm determination. The obtained ICC was 0.997, indicating a high reliability of the ultrasound imaging of Mdm used in this study.
1.2.3.7
Histochemical analysis of muscle samples
Three subjects volunteered to participate in muscle needle biopsies taken
on day 27 of the second control cycle and of the third training cycle. After
local anesthesia with 1% lidocaine and incision of the skin and fascia, percutaneous muscle biopsy samples (70 - 300 mg) were obtained from the
vastus lateralis muscle of both the right and left leg by a standard needle
biopsy technique (Bergström, 1962). Directly after sampling, the tissue was
removed from the needle, mounted cross-sectionally in a Tissue-TEK®
embedding medium, frozen in isopentane, put into an aluminum container,
cooled further with liquid nitrogen, and stored at -80°C for subsequent analysis.
Thin sections (10 μm) of the frozen tissue were cut in a cryostat at -20°C
and mounted on cover glasses for further staining. Histochemical analysis
for the determination of muscle fiber types (types Ι and ΙΙ) was performed
with adenosine-triphosphatase (ATPase) staining procedures using an alkaline pre-incubation at pH 4.3 and 9.6 (Brooke & Kaiser, 1970). Moreover,
muscle cell nuclei were stained with hematoxylin and eosin for nuclei-tofiber ratio analysis (Yan, 2000). Fiber type counting and measurements
were performed on photographs by two investigators to standardize the
procedure. All fibers of one sample were counted and measured twice and
the average of the two counts was taken for statistical analysis. If the variation between the two counts or measurements was greater than 1%, fibers
were counted a third time and the average of the two counts with the smaller variation was used for analysis. For muscle fiber type classification, an
average of 327 fibers from each sample was counted, the fiber type (type I
or type II) identified, and the percentage of each type was calculated. For
the determination of muscle fiber diameters (Fdm), an average of 62 fibers
12
STUDY 1-METHODS
(range 21–94) from each fiber type was selected. Cellular diameters were
determined using cell life science documentation software (Olympus Life
and Material Science Europe GmbH, Germany).
1.2.4 Statistical analysis
Data are presented as mean values with SD. Normality of distributions was
proved by the Kolmogorov-Smirnov test. A one-tailed paired t-test was used
to evaluate differences in training workload, VO2peak, Wattmax, Wattlac4, Fmax
and Mdm between values before (pre) and after the training intervention
(post) (see below: a, b) and between FT and LT (see below: c), respectively.
In all cases, P values < 0.025 were taken to indicate statistical significance.
Statistics were tested with a hierarchical procedure: a) FTpost better than
FTpre; b) LTpost better than LTpre; c) if a) significant:∆FT better than ∆LT; if
b) significant: ∆LT better than ∆FT (∆FT: absolute difference between FTpre
and FTpost, ∆LT: absolute difference between LT pre and LTpost). A two-tailed
paired t-test was used to compare hormone concentration between FP and
LP and between prior to and after training and to compare training units between FT und LT for three training cycles. Significance was defined as P <
0.05. The intraclass correlation coefficient of repeated measurements (ICC)
(McGraw & Wong, 1996) was determined to evaluate reliability of the determination of Fmax and Mdm.
13
STUDY 1-RESULTS
1.3
Results
1.3.1 Menstrual cycle integrity
Basal body temperature showed a significant increase during LP compared to FP in all three training cycles of the 13 subjects included in the
study.
1.3.2 Number of training sessions
The total number of single-leg training sessions was approx. 20 sessions
per leg and was not different between FT and LT (FT: N = 19.9 ± 1.9; LT:
N = 20.5 ± 1.9; P > 0.05).
1.3.3 Hormonal concentrations
We did not find any significant differences in the serum concentrations of
E2, DHEA-s and T between day 11 day 25 of the menstrual cycle, while
P4 was significantly higher in LP compared to FP prior to and after training
(Table 1-1). Prior to training free T was significantly higher on day 11 compared to day 25; after the endurance training period, free T tended to decline in FP (P=0.061) but not in LP, resulting in similar concentrations of
free T in both phases of the menstrual cycle after the training period. The
kind of training (FT vs. LT) did not have any different effect on any of the
hormones (data not shown).
14
STUDY 1-RESULTS
TABLE 1-1: Serum concentrations of E2, P4, DHEA-s, T and free T in the follicular phase
(FP, day 11) and the luteal phase (LP, day 25) before and after endurance training
(N=13)
Pre-Training
E2
(pg/ml)
P4
(ng/ml)
DHEA-s
(ug/ml)
T
(ng/ml)
Free T
(pg/ml)
Post-Training
FP
LP
FP
LP
58 ± 26
105 ± 82
95 ± 75
92 ± 73
0.69 ± 0.48 †
6.84 ± 3.90
0.61 ± 0.27 †
5.99 ± 5.36
1.99 ± 0.89
1.94 ± 0.63
1.91 ± 0.86
1.85 ± 0.71
0.38 ± 0.26
0.33 ± 0.24
0.29 ± 0.20
0.29 ± 0.15
1.91 ± 0.59 †
1.52 ± 0.32
1.56 ± 0.35 *
1.50 ± 0.55
E2: estradiol, P4: progesterone, T: testosterone, pre/post-training: Before/after three
months of endurance training,
FP: follicular phase, LP: luteal phase, *: p = 0.061 post training vs. pre training, †: p <
0.05 FP vs. LP
1.3.4 Peak oxygen uptake (VO2peak)
Three month of one-leg endurance training induced a clear trend for an increase in VO2peak both, after FT (∆ +3.1 ± 5.3 ml/min/kg, P = 0.033) and after LT (∆ +2.8 ± 4.8 ml/min/kg, P = 0.038) without any difference between
both kinds of training periodization. VO2peak of the two-leg test increased
significantly by 2.8 ± 4.1 ml/min/kg after three months of one-leg endurance training (Figure 1-1).
15
STUDY 1-RESULTS
Pre
Post
50
*
VO2peak (ml/min/kg)
40
#
##
30
20
10
28.0 31.1
27.5 30.3
32.5 35.3
∆ 3.1
∆ 2.8
∆ 2.8
0
FT
LT
Two-leg
FIGURE 1-1: VO2peak before and after three months of follicular phase-based (FT) or
luteal phase-based (LT) endurance training (N=13); Two-leg: Test with both legs, Pre: before training, Post: after training, #: P = 0.033, ##: p = 0.038, *: P < 0.025 post-training vs.
pre-training.
1.3.5 Maximum workload (Wattmax)
Wattmax of each single leg increased continuously during both types of
training periodization (Figure 1-2). During the third training cycle increase
in Wattmax was slightly more pronounced in FT resulting in a trend towards
higher absolute Wattmax after FT compared to LT ∆FT:
(
+34.8 ± 19.2 W,
∆LT: +27.8 ± 16.6 W, P = 0.038) (Figure 1-3).
16
STUDY 1-RESULTS
*
60
34.8
50
*
20.1
∆ Wattmax (W)
40
*
15.9
†
30
20
n.s.
n.s.
10
19.7
0
27.8
*
*
15.3
FT
LT
*
Pre
1st Training
2nd Training
3rd Training
Training Cycle s
FIGURE 1-2: Increase in Wattmax compared to the pre-training values during follicular
phase-based (FT) or luteal phase-based endurance training (LT) (N=13); Two-leg: Test
with both legs, Pre: before training, Training: training cycle, n.s.: not significant, *: P <
0.025 compared to pre-training, †: P = 0.038 FT vs. LT
Pre
Post
250
*
200
Wattmax (W)
*
*
150
100
50
95.1 129.9
98.4 126.2
149.0 172.0
∆ 34.8
∆ 27.8
∆ 23.0
FT
LT Two-leg
0
FIGURE 1-3: Maximum workload (Wattmax) before and after three months of follicular
phase-based (FT) or luteal phase-based (LT) endurance training (N=13); Two-leg: Test
with both legs, Pre: before training, Post: after training, *: P < 0.025 post-training vs. pretraining, †: P = 0.038 FT vs. LT
17
STUDY 1-RESULTS
1.3.6 Submaximal power output at a lactate concentration of 4
mmol/l (Wattlac4)
Wattlac4 of each single leg and of both legs together increased during both
types of training periodization (Figure 1-4). Absolute increase of Wattlac4
was comparable between FT and LT (∆FT: 27.6 ± 22.3 W, ∆LT: 28.6 ± 18.6
W, P > 0.025; Figure 1-5).
*
28.6
50
∆ Wattlac4(W)
40
*
*
28.5
21.0
30
n.s.
n.s.
20
n.s.
10
18.4
0
*
0
0
*
15.5
*
Pre
27.6
FT
LT
1st Training
2nd Training
3rd Training
Training Cycle s
FIGURE 1-4: Increase in Wattlac4 compared to the pre-training values during follicular
phase-based (FT) or luteal phase-based (LT) endurance training (N=13); Two-leg: Test
with both legs, Pre: before training, Training: training cycle, n.s.: not significant, *: P <
0.025 compared to pre-training
18
STUDY 1-RESULTS
Pre
Post
160
*
140
*
120
Wattlac4 (W)
*
100
80
60
40
20
62.9
∆
90.5
72.0 100.6
27.6
∆
28.6
110.3 123.2
∆
12.9
0
FT
LT Two-leg
FIGURE 1-5: Power output at a lactate concentration of 4 mmol/l (Wattlac4) before and after three months of follicular phase-based (FT) or luteal phase-based (LT) endurance
training (N=13); Two-leg: Test with both legs, Pre: before training, Post: after training, *: P
< 0.025 post-training vs. pre-training
1.3.7 Isometric muscle strength (Fmax)
Fmax tended to decline after three months of FT (P = 0.034) and LT (P =
0.05) without any difference between both training periodization protocols
(Table 1-2).
1.3.8 Muscle diameter (Mdm)
The sum of Mdm of the three muscles remained unchanged (P > 0.025)
after both types of training periodization compared to the pre-training level
(Table 1-2).
19
STUDY 1-RESULTS
TABLE 1-2: Maximum isometric force (Fmax) and Sum of muscle diameter (Mdm) (m. rectus femoris, m. vastus intermedius, m. vastus lateralis) before and after three months of
follicular phase-based or luteal phase-based endurance training (N=13)
Pre-Training
Post-Training
FT
LT
FT
LT
Fmax (N)
685 ± 165
676 ± 158
649 ± 149 #
633 ± 163 ##
Mdm (cm)
6.37 ± 0.75
6.50 ± 0.88
6.48 ± 0.95
6.47 ± 0.89
FT: follicular phase-based training, LT: luteal phase-based training
#: P = 0.034, ##: P = 0.05
1.3.9 Muscle fiber characteristics
Muscle fiber characteristics of a subgroup of three subjects revealed broad
inter-individual variation (Table 1-3a,b). Although data have to be taken
cautiously, there might be a slightly higher increase in type I fiber diameter
compared to type II fiber diameter after three months of endurance training
with a slightly more pronounced increase in type I fiber diameter after FT
compared to LT (Type I: ∆FT: 2.5 ± 6.9 µm, ∆LT: 1.5 ± 3.4 µm; Type II: 0.1 ±
8.8 µm, ∆LT: -0.0 ± 6.0 µm). Fiber type distribution showed some inconsistent results, but remained nearly the same after both kinds of endurance training periodization. The nuclei-to-fiber ratio also remained unchanged after both types of training.
20
48.1±5.0%
Mean
44%
Sub. 1
54%
46.1±5.2%
Mean
Sub. 3
50%
Sub. 3
46%
48%
Sub. 2
Sub. 2
40%
Sub. 1
51.9±5.0%
46%
54%
56%
53.9±5.2 %
50%
52%
60%
Type ΙΙ
56.4±6.2%
63%
51%
55%
54.8±4.6%
59%
50%
56%
Type Ι
LT
43.6±6.2%
37%
49%
45%
45.2±4.6%
41%
50%
44%
Type ΙΙ
54.5±2.3%
57%
53%
53%
50.3±3.7%
53%
46%
51%
Type Ι
FT
45.5±2.3%
43%
47%
47%
49.7±3.7%
47%
54%
49%
Type ΙΙ
55.6±1.0%
54%
56%
56%
52.3±2.2%
52%
50%
55%
Type Ι
Post Training
LT
44.4±1.0%
46%
44%
44%
47.7±2.2%
48%
50%
45%
Type ΙΙ
before and after three months of follicular phase-based (FT) or luteal phase-based (LT) endurance training (N=3)
TABLE 1-3a: Muscle fiber type distribution: relative number of fiber (F%no) and relative fiber area (F%area) for single subject and mean values of three subjects
F%area
F%no
Type Ι
FT
Pre Training
52.4±8.2
58.1±12.5
51.5±7.1
Sub. 2
Sub. 3
Mean
2.5
3.3
2.8±0.4
Sub. 3
Mean
2.7
Sub. 2
Sub. 1
44.0±6.9
Sub. 1
Type Ι
49.6±8.0
54.3±9.0
54.0±9.2
40.4±9.0
Type ΙΙ
2.3
3.7
2.7
Type ΙΙ
50.4±2.4
52.7±9.8
47.9±9.1
50.5±9.9
2.9±0.7
52.1±4.6
57.3±10.7
48.7±7.9
50.4±8.8
Type Ι
4.1
2.6
2.2
Type ΙΙ
49.7±0.8
49.1±8.5
49.4±8.5
50.5±9.0
2.9±1.0
54.0±2.3
53.0±8.8
56.6±7.3
52.4±9.1
Type Ι
53.6±6.8
4.1
2.0
2.3
Type ΙΙ
50.4±8.3
56.8±14.1
41.0±8.5
53.3±7.7
2.8±1.1
59.9±11.9
46.4±8.5
54.6±6.6
Type Ι
three months of follicular phase-based (FT) or luteal phase-based (LT) endurance training (N=3)
TABLE 1-3b: Muscle fiber type distribution: fiber diameter (Fdm) and nuclei-to-fiber ratio (N/F) for single subject and mean values of three subjects before and after
N/F
(µm)
Fdm
STUDY 1-DISCUSSION
1.4
Discussion
This study is the first one about planning endurance training with respect to
hormonal fluctuations during the menstrual cycle. The main finding of this
study is a clear increase in all parameters of endurance performance after
three months of one-leg endurance bicycle ergometer training, showing that
this type of training was effective. A slightly different effect of training periodization, however, could only be demonstrated for Wattmax after the third
training cycle (Figure 1-2 and 1-3), when Wattmax after FT tended to be
higher compared to LT. Neither in VO2peak nor in Wattlac4 cycle-specific adaptation occurred.
In another study from our group we recently could demonstrate that follicular phase-based one-leg strength training showed a higher effect on muscle
strength, muscle diameter and diameter of type ΙΙ fibres than luteal phasebased one-leg strength training (Sung et al., 2012). In this study, the higher
increase in maximal isometric muscle strength was more pronounced and
occurred already after one month of strength training, and we concluded
that alterations in anabolic hormones like testosterone and estrogens
throughout the menstrual cycle triggered the higher training adaptation in
FT. In this study, again, we found a higher level of free T in FP compared to
LP prior to the endurance training intervention, and a significant decline of
free T in FP throughout the training period, while total T did not differ between both phases or after the training period (Table 1-1). Since androgen
secretion from the ovary is under luteinizing hormone control at least in part,
it is not unexpected that ovarian androgen secretion varies through the cycle: the blood levels of T have been described to be lowest in the early follicular phase and then to rise to their highest levels just prior to, or at the
time of, ovulation and then gradually fall during the luteal phase (Alexander,
Sherwin, Bancroft & Davidson, 1990, Longcope, 1986). In females, serum T,
however, might also origin from the adrenal gland or from peripheral conversion (Enea, Boisseau, Fargeas-Gluck, Diaz & Dugue, 2011).
Early studies have shown the production rate of T from the adrenals being
about 50 µg/day and the ovaries secreting an additional 50 µg/day, but the
major source of T being the peripheral conversion of androstenedione
23
STUDY 1-DISCUSSION
(around 100 µg/day) (Longcope, 1986). This mixture and production interrelationship might explain why some studies did not find any changes in serum T concentration throughout the menstrual cycle (Jabbour, Kelly, Fraser
& Critchley, 2006), and why free T was no longer different between FP and
LP after three months of one-leg endurance training in our study, or even
declined over time in FP after endurance training compared to FP prior to
training. DHEA-s, the main metabolite of the adrenal glands but not of the
ovaries, remained completely unaffected from the phase of the cycle and
throughout the training intervention period in this investigation.
In a very recent review of physical exercise-induced changes in the concentration of circulating androgens in women the authors concluded that studies are still contradictory regarding the effect of endurance exercise on circulating androgens in women (Enea, Boisseau, Fargeas-Gluck, Diaz &
Dugue, 2011).
Endurance exercise training induces increase in energy metabolism and
oxidative capacity, as well as changes in muscle fibre type (Gravelle &
Blessing, 2000). Testosterone has been shown to effect mitochondrial function and oxidative metabolism in men (Pitteloud et al., 2005). No data in
women are available concerning these effects. The ovarian hormones, estrogen and progesterone, have important roles in regulating substrate metabolism during exercise in women. In animal models, estrogen promotes
lipolysis and increases fatty acid availability while decreasing the rate of
gluconeogenesis and sparing muscle and liver glycogen use. The addition
of progesterone has been reported to antagonize the lipolytic effects of estrogen and reduce fatty acid availability. Conversely, the addition of progesterone appears to accentuate the carbohydrate-sparing actions of estrogen
by decreasing hepatic glycogenolysis. Furthermore, estrogen up-regulates
mitochondrial enzymes favoring fat oxidation, whereas progesterone opposed these actions. Taken together, alterations in substrate use and regulation of enzyme activities during exercise across the menstrual cycle are
dependent on the relative changes in both estrogen and progesterone
(D'Eon et al., 2002). Therefore, the amount of adaptation processes to endurance training in our study was expected to also depend on the endocrine milieu when the training stimulus was set. In contrast to the strength
24
STUDY 1-DISCUSSION
training study (Sung et al., 2012), however, differences between the increase of parameters of endurance performance after FT and LT were only
small (Wattmax) or not detectable (VO2peak, Wattlac4).
We conducted one-leg endurance training with both legs training differently
in the same subject in order to exclude inter-individual variation. The leg
which should perform follicular phase-based training performed 75% of all
training load (six from eight sessions) in the follicular phase without any influence of progesterone, while 25% of training was done during the luteal
phase under the influence of both, estrogen and progesterone. These two
training sessions were planned to maintain endurance capacity on the
reached level after six sessions in the follicular phase. The training of the
other leg was mainly performed under the combined influence of estrogen
and progesterone accordingly, and the additional two sessions were under
the influence of estrogen only in order to maintain the reached level of adaptation. An overlapping influence of the endocrine milieu of the two respective maintaining training sessions on adaptation processes, however,
cannot be excluded and might at least partly explain why only minor effects
of the periodization of the endurance training were observed. A follow-up
study without any training session in the second phase of the menstrual cycle might clarify the possible overlapping effects.
Regarding the molecular signals, the change in the energy state of the
muscle accompanying contraction is involved with a role of AMPK in inducing adaptations to endurance training (Winder, Taylor & Thomson, 2006).
AMPK is activated in response to muscle contraction; chronic chemical activation of AMPK results in increases in GLUT4, hexokinase 2, UCP-3, and
citric acid cycle enzymes; and muscle contraction and chemical activation
of AMPK both result in increases in PGC-1α, a transcriptional co-activator
involved in stimulation of mitochondrial biogenesis. In our study, the general
increase in parameters representing endurance performance (VO2peak,
Wattmax, Wattlac4) after three month of one-leg endurance training indicates
that at least some of these mechanisms have been activated.
25
STUDY 1-DISCUSSION
In skeletal muscle estrogen stimulates AMPK activity and up-regulates
transcription factors that are responsible for increased expression of various enzymes, and estrogen’s activation of AMPK is the major key to the
metabolic perturbations of estrogen (Oosthuyse & Bosch, 2010). Data on
progesterone effects on AMPK are not available in the literature. Furthermore, it is not clear whether and how the effects of exercise and estrogens
on AMPK interfere. Recent findings have shown that the expression of estrogen receptors in skeletal muscle increases with the level of endurance
training (Wiik et al., 2005), and hence one might speculate that a greater
estrogen stimulated AMPK response could be expected in skeletal muscles
after endurance training. However, the amount of AMPK activation when
estrogen and acute training load are superimposed remains unclear. Further studies should analyze dose-response relationships with variations in
both of these parameters.
In our study one-leg endurance training did not lead to any changes in
muscle diameter and maximal isometric muscle strength even tended to
decrease after both training periodization (Table 1-2). The adaptive responses of muscle to exercise are specific to the training mode (Gravelle &
Blessing, 2000). Endurance training induces increases in mitochondrial
density and enzymes of the tricarbonacid cycle and electron transport chain
as well as increases in capillary density, myoglobin, and VO2max. The end
result is an increase in the ability of an individual to perform prolonged exercise with little or no increase in strength. In contrast, strength training results in enhanced force production, hypertrophied muscle cells, increased
glycolytic enzyme activity, augmented intracellular fuel stores of ATP and
phosphocreatine, and reductions in mitochondrial and capillary density.
Changes in VO2max are usually minimal or nonexistent (Gravelle & Blessing,
2000). Activation of AMPK through endurance training might even decrease
the mammalian target of rapamycin (mTOR) signaling and consecutively
protein synthesis (Kimball, 2006). As we did not find any increase in muscle
diameter and even a decline in muscle strength, our findings indicate that
the main purpose of the specific sub-maximal one-leg bicycle ergometer
training regime, which was to induce specific stimuli for increasing parame26
STUDY 1-DISCUSSION
ters of endurance performance probably via activation of AMPK was fully
accomplished.
To check for ovarian hormone fluctuation in our study E2 and P4 were analyzed on day 11 (pre-ovulation) and on day 25 (luteal phase). On these
days, both hormones showed high inter-individual variations. P4 clearly increased in all subjects in the luteal phase, indicating that ovulation had occurred in all of them and that the training period had not induced any severe
alteration in menstrual cycle integrity like anovulation or luteal phase insufficiency. The similar concentrations in E2 on days 11 and 25 prior to the
training period are probably due to the fact that day 11 represents a phase
prior to ovulation, when E2 is already elevated compared to early and middle FP (Van Look & Baird, 1980).
This study is not only the first to conduct an endurance training program
based on the phases of the menstrual cycle but we also investigated muscle fiber parameters after the two types of menstrual-cycle based endurance training. Unfortunately, only three from 13 subjects agreed in muscle
biopsy. Therefore, results are very preliminary and have to be interpreted
carefully. It has long been demonstrated that endurance training sessions
might stimulate hypertrophy of type Ι muscle fibers (Tegtbur, Busse & Kubis,
2009). This is in line with our study, where we observed a slight increase in
type I but not in type II fiber diameter, again indicating the specificity of the
training stimulus. Additionally, we observed a slightly higher increase in
type I fiber diameter after FT compared to LT. Alterations in fiber type distribution were inconsistent and nuclei-to-fiber ratios remained unaffected.
No data on menstrual cycle dependent adaptations of skeletal muscle fibers
are available in the literature. We therefore recommend further studies including more subjects in order to get more insight into possible underlying
mechanisms in menstrual cycle phase-specific training adaptations.
To summarize, this study could demonstrate a slightly higher but delayed
effect of follicular phase- compared to luteal phase-based endurance training on maximum power output on a bicycle ergometer compared to LT
without any different effect on VO2peak and muscle diameter. This might be
due to the specific hormonal milieu during each phase of the cycle. Further
studies with longer lasting training periods are needed in order to analyze if
27
STUDY 1-DISCUSSION
this late response in aerobic training adaptation becomes more pronounced
after three months of training. Furthermore, more subjects have to be included in muscle biopsy analyses in order to understand possible underlying mechanisms of cycle-dependent aerobic training adaptations.
28
STUDY 2-ABSTRACT
STUDY 2 – EFFECTS
OF
MENSTRUAL PHASE-BASED ENDURACNE
TRAINING IN ORAL CONTRACEPTION USERS
Abstract
Purpose Modern monophasic oral contraceptives (OC) suppress circulating endogenous sex steroid hormone concentrations. OC provide the women with fixed doses of estrogen and progestogen over 21 days (consumption phase), followed by a 7-days break (withdrawal phase). Theses hormone fluctuations together with alterations in endogenous sex steroids
might be important influence factors for endurance trainability. We investigated the effects of quasi-follicular phase-based endurance training (qFT,
day 1-14 of the menstrual cycle) on physiologic and microscopic measures
of aerobic capacity comparing it to quasi-luteal phase-based endurance
training (qLT, day 15-28) in OC users.
Methods: Fourteen women using monophasic OC completed one-leg endurance training on a cycle ergometer for three menstrual cycles. They
trained one leg mainly in the quasi-follicular phase (qFP) and the other leg
mainly in the quasi-luteal phase (qLP). Concentrations of 17-beta estradiol
(E2), progesterone (P4), total testosterone (T), free testosterone (free T)
and DHEA-s were analyzed in blood samples taken during qFP and qLP.
Peak oxygen uptake (VO2peak), maximal workload (Wattmax), power output at
a lactate concentration of 4mmol/l (Wattlac4), maximum isometric strength of
knee extension (Fmax) and muscle diameter (Mdm) were analyzed before
and after training, as well as (No%), fiber diameter (Fdm) and cell nuclei-tofiber ratio (N/F) in a subgroup of five subjects.
Results: DHEA-s was higher (p < 0.05) in qFP compared to qLP before
and after the endurance training period. E2, total and free T and P4 did not
differ between the two phases or prior compared to after training. VO2max,
Wattmax and Wattlac4 increased significantly after qFT and qLT without any
difference between both training periodizations. Fmax increased after FT and
LT, and Mdm remained unchanged. Androgenicity of the progestin of the
OC pills led to lower increase in VO2peak after qLT, while in users of OC pills
without androgenicity higher increase in Wattlac4 after qLT was observed.
Skeletal muscle fiber distribution and fiber diameter remained unaffected by
29
STUDY 2-INTRODUCTION
the endurance training period, and nucleus-to-fiber ratio tended to increase
after both types of training periodization.
Conclusions: One-leg endurance training induced significant increments of
parameters of aerobic performance without any differences between the
two training interventions. This is presumably due to the constant doses of
estrogen and progestin in monophasic OC, interfering with alterations in
endogenous sex steroid hormones. Effects of androgenicity index on some
parameters of endurance performance are inconsistent. In summary, we
recommend that untrained and moderately trained OC users perform their
endurance training independently from their pill cycle, and that further studies are necessary in order to understand possible effects of OC pills on development of endurance performance.
2.1
Introduction
Nowadays the use of oral contraceptives (OC) is increasing and about 100
million women are using OC worldwide (Erkkola, 2007). OC are utilized for
contraceptive and for non-contraceptive purposes. The number of female
athletes using OC is also increasing for reasons like birth control, management of premenstrual symptoms, dysmenorrhea, less menstrual blood loss,
lower risk of musculoskeletal injury and time-shifting of the menstrual cycle,
which could provide benefits for the female athletes (Bennell et al., 1999;
Constantini et al., 2005; Wojtys et al., 2002).
Among OC, monophasic preparations are the most used in Germany. Monophasic OC contain fixed doses of ethinylestradiol and progestin which
are taken for 21 days (the consumption phase), followed by 7 days of OC
break (the withdrawal phase). OC lead to follicle-stimulation hormone and
luteinizing hormone inhibition by suppressing gonadotropin-releasing hormone and prevent ovulation (Greydanus, Patel & Rimsza, 2001). As a result endogenous estradiol (E2) and progesterone (P4) are suppressed and
their concentrations remain constant during 21 days of the OC consumption
phase (Rechichi, Dawson & Goodman, 2009).
As sexual steroid hormones play a considerable role in training adaptation
processes, OC-induced alterations in their blood concentrations might lead
30
STUDY 2-INTRODUCTION
to alterations in the amount of training adaptation in OC users compared to
non-OC users. Due to the exogenous regulation of endogenous sex steroids biosynthesis and secretion, levels of testosterone, free testosterone
and DHEA-s decrease significantly with usage of OC (Graham, Bancroft,
Doll, Greco & Tanner, 2007; Rickenlund et al., 2004). E2 was also found to
be significantly lower in OC users in comparison to non-OC users (Vaiksaar
et al., 2011). Moreover, progesterone remains low in the quasi-luteal phase
(qLP) in OC users, whereas it increases significantly in the luteal phase
(LP) as compared to in the follicular phase (FP) in non-OC users (Vaiksaar
et al., 2011).
All kinds of OC can influence fat and carbohydrate metabolism (Dooley &
Brincat, 1994; Kiley & C., 2007; Rechichi et al., 2009). They might cause
deterioration in glucose tolerance, increase of peripheral resistance to insulin and increases of lipoprotein metabolism. However, these effects depend
on the doses of exogenous estrogen and progesterone (Dooley & Brincat,
1994; Kiley & C., 2007). Furthermore, monophasic OC might affect endurance performance (Burrows & Peters, 2007) with possible effects on body
composition, metabolism, aerobic capacity and cardiovascular responses.
OC pill induced increase in basal body temperature, for instance, could
cause decline in long duration performance. There are some studies to
suggest that endurance performance is not affected by changes from active
to inactive OC pill phase, although there is also a potential for variation in
endurance performance throughout an OC cycle (Vaiksaar et al., 2011).
Differences in ventilation, oxygen consumption and substrate metabolism
between studies appear to relate to variations in the types of OC used and
the dosage of progestogen administered (Rechichi et al., 2009). Altogether,
research in the area of OC and exercise capacity is sparse and much has
been plagued by poor research design, methodology and small sample size
(Burrows & Peters, 2007). Furthermore, no study on androgenicity of the
progestin in oral contraceptive pills is available in the literature.
In a previous study we could demonstrate that follicular phase-based endurance training (FT) showed slightly higher but delayed effect on maximum power output on a bicycle ergometer compared to luteal phase-based
training (LT) without any different effect on VO2peak and muscle diameter in
31
STUDY 2-INTRODUCTION
eumenorrheic non-OC users (Han et al., 2012), which was probably due to
the specific hormonal milieu during each phase of the cycle. In contrast to
this investigation in non-OC users, there are no training intervention studies
available in OC users that have differentially assessed the trainability of endurance in the two respective phases of OC use. Additionally, the possible
influences of other interacting anabolic hormones like T and DHEA-s in
training adaptation processes in OC users are not clear until now.
The aim of this study was to investigate the effects of longer-lasting quasi
follicular phase-based endurance training (qFT) on macroscopic and microscopic parameters of skeletal muscle adaptations compared to quasi luteal
phase-based endurance training (qLT) in an in vivo controlled training intervention study in healthy young females and to analyze for possible differences of OC pills with or without androgenicity.
32
STUDY 2-METHODS
2.2
Methods
2.2.1 Subjects
Fourteen healthy women, with a mean (± SD) age of 23.9 ± 3.0 yr, height of
168.4 ± 6.9 cm and weight of 61.9 ± 8.8 kg volunteered to participate in this
study. Subjects were untrained or moderately trained and they were not
currently performing endurance training. Moreover they had been taking
OC at least for a year prior to participation in this study and had no history
of any endocrine disorders. The monophasic combined oral contraceptives
are taken for 21 days followed by a 7 day pill-free interval and a withdrawal
bleed occurs in this pill-free break week. Constant concentrations of synthetic estrogen (20-35 µg of ethinylestradiol) and progestin (125-150 and
2000-3000 µg of gestagen depending on brands) are contained in the 21
pills, which inhibit fertility. The kind and number of preparation used by the
subjects of this study including values of their assumed androgenic effects
is given in Table 2-1. Prior to the study, participants were informed about
the purpose, procedures and risks of the study and written informed consent was obtained from each participant. Approval for the experimental protocol was obtained from the Ethics Committee of the Ruhr-University Bochum, Germany.
TABLE 2-1: Monophasic oral contraceptive pills used by the subjects of this study including
doses of ethinylestradiol and gestagen and their possible androgenicity index
Dose of Ethinylestradiol and Gestagen in OC
Name of
Ethinylestradiol
Gestagen
Trades
Belara
(µg)
30
(µg)
2000
Desmin 20
20
Femigoa
Type of Progestin
Androgenicity
Number of
chlormadinone acetate
0
subjects
1
150
Desogestrel
0.51
1
30
150
Levonorgestrel
1.25
1
Juilette
35
2000
Cyproteron acetate
0
1
Lamuna 20
20
150
Desogestrel
0.51
3
Monostep
30
125
Levonorgestrel
1.04
2
Vallete
30
2000
Dienogest
0
2
Yasmin
30
3000
Drospirenone
0
2
Yasminelle
20
3000
Drospirenone
0
1
33
STUDY 2-METHODS
2.2.2 Experimental design
Participants performed an endurance training program on a cycle ergometer, separately for each leg, over a period of three menstrual cycles each.
Subjects were randomly divided to two groups in order to reduce effects of
leg preference: one group (N = 8) mainly trained the right leg during qFP,
while the left leg was mainly trained during qLT. The other group (N = 6)
mainly trained the left leg during qFP, while the right leg was trained during
qLP. For further analysis, both follicular phase-trained legs and both luteal
phase-trained legs were taken together in the qFT- or qLT-trained leg group,
respectively. The entire study took four menstrual cycles (1 control cycle
followed by 3 training cycles), equivalent to 112 days considering that one
menstrual cycle always took 28 days. The first day of menstrual bleeding in
the withdrawal phase was defined as day 1 of the cycle. Day 1 to day 14
was defined as qFP, and day 15 to the first day of the following menstrual
bleeding as qLP, oriented on the terminology of menstrual phase classification in eumenorrheic women.
In the control cycle and in the third training cycle, blood samples for hormonal analysis were taken from a cubital vein on day 11 (late qFP) and day 25
(late qLP) of the menstrual cycle. Additionally, incremental bicycle ergometer tests until exhaustion were carried out on day 11 and 23 (single leg
tests) and on day 25 (both leg test). Furthermore, the maximum isometric
leg strength was measured on day 25 after the two-legged incremental bicycle ergometer test. On day 27 (late qLP), the diameter of the quadriceps
muscle was measured and muscle biopsies were extracted from the vastus
lateralis muscle in a subgroup of five subjects. During the first and second
training cycle, the incremental bicycle ergometer tests were carried out as
maximal ergometer tests on days 11 and 23 (single leg tests) and on day
25 (both leg test).
34
STUDY 2-METHODS
2.2.3 Study schedule
2.2.3.1
Endurance training program
The subjects completed three menstrual cycles of one-leg endurance training program with different training quantities of the right and left leg in qFP
and qLP, respectively, while the total number of single-leg training sessions
in one menstrual cycle remained the same. The training was performed
three times a week (typically on Monday, Wednesday and Friday) under
supervision on a bicycle ergometer (Ergoselect 100, Ergoline GmbH, Germany). One leg was mainly trained in qFP (qFT) and the other leg mainly in
qLP (qLT). In qFT, subjects trained six times in the quasi-follicular phase
(typically between day 1 and day 14) and just twice in the quasi-luteal
phase. For qLT, they trained six times in qLP (typically between day 15 and
day 28) and just twice in qFP. The total number of sessions was the same
for both legs.
Subjects performed one-leg training for 60 minutes at the work load corresponding to 75% of the predetermined one-leg power corresponding to 4
mmol/l blood lactate concentration. On the days on which both legs had to
be trained, subjects performed two-leg exercise with a work load adjusted
to 75% of the predetermined two-leg power corresponding to 4 mmol/l
blood lactate concentration. The workloads were adjusted upwards every
four weeks according to the results of the last respective maximal bicycle
ergometer tests.
2.2.3.2
Hormone analysis
Venous blood was centrifuged after blood clotting, and the serum was kept
frozen at -80° C until analysis. Each sample was analyzed for E2, P4, total
testosterone (T) and free T, and dihydrotestosterone-sulfate (DHEA-s). E2,
P4, T, and DHEA-s were assayed by immunochemistry (Elecsys® 1010
System, Roche Diagnostics GmbH), and free T was assayed by radioimmunoassay (Multi-Crystal LB 2111 gamma counter, Berthold Technologies
GmbH & Co. KG).
35
STUDY 2-METHODS
2.2.3.3
Physiologic measures of endurance capacity
Right and left leg and two-leg incremental tests to exhaustion were performed on a cycle ergometer (Ergoselect 100, Ergoline GmbH, Germany)
with an open air spirometry system (ZAN 600 USB, nSpire Health,
Oberthulba, Germany) in the control cycle and the third training cycle in order to determine 1. peak oxygen uptake (VO2peak), 2. maximum workload
(Wattmax) and 3. power output at a lactate concentration of 4 mmol/l in capillary blood (Wattlac4). Each right and left leg was measured on separate days.
The leg which mainly trained during qFP was measured in late qFP (day
11). The other leg which mainly trained in qLT was analyzed in late qLP
(day 23). The two-leg test was performed two days after the second test
(day 25). The initial workload for the one-leg incremental test was 20 W and
the work rate was increased by 10 W every one minute until the subject
could not maintain the required pedaling frequency (>50 rpm). For the twoleg test the initial workload was set at 25 W and increased by 25 W every
two minutes. Capillary blood samples for the measurement of blood lactate
concentration (Ebio plus, Eppendorf AG, Hamburg, Germany) were taken
from an earlobe at rest, every two minutes during the test and immediately
after termination of the test. During the first and second training cycle right
and left leg and two-leg maximal incremental tests were performed without
spirometry as described above in order to determine (Wattlac4) and to adjust
training intensity, respectively.
2.2.3.4
Measurement of isometric muscle strength
Maximum isometric knee extension muscle strength (Fmax) of the right and
left leg was measured in the late qLP (day 25) after the two-leg ergometer
test in the control cycle and the third training cycle. Fmax was determined on
a leg press machine (Medizinische Sequenzgeräte, Compass, Germany)
using a combined force and load cell (GSV-2ASD, ME-Messsysteme
GmbH, Hennigsdorf, Germany). The intraclass correlation coefficient of repeated measurements (ICC) was 0.998, indicating a high internal consistency (reliability) of the system. Prior to testing the subjects were famil36
STUDY 2-METHODS
iarized with the test procedure and the testing position (knee angle: 90°,
ankle angle: 90°) on the leg press. Each measurement was repeated three
times with 30 s rest between the tests. The best result was selected for data analysis
2.2.3.5
Determination of muscle diameter
Mdm of rectus femoris, vastus intermedius and vastus lateralis muscle of
the right and left leg was measured by real-time ultrasound imaging prior to
and after training at day 25 in qLP of the control cycle and the third training
cycle analyzing the distances between the outer and inner muscle fasciae.
Previous studies showed that muscle cross-sectional area might reliably be
measured using real-time ultrasound imaging (Martinson & Stokes, 1991).
We used a Vivid I CE 0344 ultrasound device (GE Medical System, Solingen, Germany) with a parallel scanner (8L-RS, 4.0–13.3 MHz), which provides 10 cm penetration depth of the sound wave and enables high quality
analysis of deeper lying muscles. Subjects prevented long-lasting static
muscular tension for at least 30 minutes prior to the measurement in order
to avoid alterations in Mdm (Reimers, 2004). All subjects lay supine with
outstretched legs on an examination table without any pad, cushion or pillow underneath. Ultrasound images were obtained exactly half-way between the spina iliaca anterior superior and the upper margin of the patella.
The transducer was placed gently on the skin to avoid compression and
distortion of the underlying tissue (Reimers, 2004). The transducer was
held at angles of 90° towards the skin and towards the longitudinal direction
of the muscles to ensure a clear cross-sectional image. The images were
frozen on the screen to measure muscle diameter. The position of the
transducer was recorded for each muscle to reproduce the exact position
after training intervention. The mean of three measurements of each of the
three analyzed muscles was taken for both legs and the sum of the 3 Mdm
was calculated for both sides of the body. Reliability analysis was performed for Mdm determination. The obtained ICC was 0.997, indicating a
high reliability of the ultrasound imaging of Mdm used in this study.
37
STUDY 2-METHODS
2.2.3.6
Histochemical analysis of muscle samples
Five subjects volunteered to participate in muscle needle biopsies taken on
day 27 of the control cycle and of the third training cycle. After local anesthesia with 1% lidocaine and incision of the skin and fascia, percutaneous
muscle biopsy samples (70 - 300 mg) were obtained from the vastus
lateralis muscle of both the right and left leg by a standard needle biopsy
technique (Bergström, 1962). Directly after sampling, the tissue was removed from the needle, mounted cross-sectionally in a Tissue-TEK® embedding medium, frozen in isopentane, put into an aluminum container,
cooled further with liquid nitrogen, and stored at -80°C for subsequent analysis.
Thin sections (10 μm) of the frozen tissue were cut in a cryostat at -20°C
and mounted on cover glasses for further staining. Histochemical analysis
for the determination of muscle fiber types (types Ι and ΙΙ) was performed
with adenosine-triphosphatase (ATPase) staining procedures using an alkaline pre-incubation at pH 4.3 and 9.6 (Brooke & Kaiser, 1970). Moreover,
muscle cell nuclei were stained with hematoxylin and eosin for nuclei-tofiber ratio analysis (Yan, 2000). Fiber type counting and measurements
were performed on photographs by two investigators to standardize the
procedure. All fibers of one sample were counted and measured twice and
the average of the two counts was taken for statistical analysis. If the variation between the two counts or measurements was greater than 1%, fibers
were counted a third time and the average of the two counts with the smaller variation was used for analysis. For muscle fiber type classification, an
average of 262 fibers from each sample was counted, the fiber type (type I
or type II) identified, and the percentage of each type was calculated. For
the determination of muscle fiber diameters (Fdm), an average of 35 fibers
(range 20–65) from each fiber type was selected. Cellular diameters were
determined using cell life science documentation software (Olympus Life
and Material Science Europe GmbH, Germany).
38
STUDY 2-METHODS
2.2.4 Statistical Analysis
Data are presented as mean values with SD. Normality of distributions was
proved by the Kolmogorov-Smirnov test. A one-tailed paired t-test was used
to evaluate differences in training workload, VO2peak, Wattmax, Wattlac4, Fmax
and Mdm between values before (pre) and after the training intervention
(post) (see below: a, b) and between qFT and qLT (see below: c), respectively. In all cases, P values < 0.025 were taken to indicate statistical significance. Statistics were tested with a hierarchical procedure: a) qFTpost better than qFTpre; b) qLTpost better than qLTpre; c) if a) significant: ∆qFT be tter
than ∆qLT; if b) significant: ∆qLT better than ∆qFT (∆qFT: absolute diffe rence between qFTpre and qFTpost, ∆qLT: absolute difference between qLTpre
and qLTpost). A two-tailed paired t-test was used to compare hormone concentration between qFP and qLP and between prior to and after training
and to compare training units between qFT und qLT for three training cycles. Significance was defined as P < 0.05. The intraclass correlation coefficient of repeated measurements (ICC) (McGraw & Wong, 1996) was determined to evaluate reliability of the determination of Mdm.
39
STUDY 2-RESULTS
2.3
Results
2.3.1 Number of training sessions
The total number of single-leg training sessions was approximately 20 sessions per leg and was not different between qFT and qLT (qFT: N = 20.2 ±
1.5; qLT: N = 20.4 ± 1.7; P > 0.05).
2.3.2 Training load
Mean training load did not differ between qFT and qLT at the beginning of
the training period (qFT: 53.9 ± 13.2 W; qLT: 55.0 ± 11.4 W, P > 0.05).
Training load was elevated continuously according to the increase in aerobic capacity from the beginning of the training period to the last training
session in qFT and qLT. Training load in the last study phase did not differ
between qFT and qLT (qFT: 67.5 ± 15.3 W; qLT: 70.0 ± 15.4 W, P > 0.05)
2.3.3 Hormone concentrations
We did not find any significant difference in concentrations of E2, P4, T and
free T between day 11 and day 25 of the menstrual cycle prior to and after
training intervention. DHEA-s showed significantly (p < 0.05) higher concentration in qFP than in qLP before and after training, respectively (Table
2-2).
40
STUDY 2-RESULTS
TABLE 2-2: Serum concentrations of E2, P4, DHEA-s, T and free T in the quasifollicular phase (qFP, day 11) and the quasi-luteal phase (qLP, day 25) before and
after endurance training (N=14)
Pre-Training
Post-Training
qFP
qLP
qFP
qLP
E2
(pg/ml)
8.6 ± 7.6
10.8 ± 10.2
12.1 ± 10.1
12.1 ± 11.0
P4 (ng/ml)
0.47 ± 0.20
0.42 ± 0.23
0.50 ± 0.23
0.43 ± 0.21
DHEA-s
(ug/ml)
1.46 ± 0.52
1.24 ± 0.54 †
1.50 ± 0.60
1.19 ± 0.39 †
T
(ng/ml)
0.12 ± 0.07
0.12 ± 0.07
0.16 ± 0.12
0.12 ± 0.09
Free T
(pg/ml)
0.83 ± 0.23
0.77 ± 0.22
0.87 ± 0.20
0.76 ± 0.17
E2: estradiol, P4: progesterone, T: testosterone, pre-/post-training: before/after
three months of endurance training,
qFP: quasi-follicular phase, qLP: quasi-luteal phase, †: p < 0.05 FP vs. LP
2.3.4 Peak oxygen uptake (VO2peak)
Three month of one-leg endurance training induced a significant increase in
VO2peak both, after qFT (∆ +5.7 ± 2.3 ml/min/kg) and after qLT (∆ +5.3 ± 3.5
ml/min/kg, P < 0.025) without any difference between both kinds of training
periodization. VO2peak of the two-leg test increased significantly by 4.3 ± 4.9
ml/min/kg after three months of one-leg endurance training (Figure 2-1).
Furthermore, VO2peak in the subjects taking OC pills with known
androgenicity tended to increase to a lesser extend in qLT compared to
qFT, while VO2peak increased the same in the subjects taking OC without
any androgenecity (Figure 2-2).
41
STUDY 2-RESULTS
Pre
Post
50
*
*
VO 2peak (ml/min/kg)
40
*
30
20
27.7 33.4
10
26.6 31.9
33.1 37.4
∆ 5.3
∆ 4.3
∆ 5.7
0
qFT qLT
Two-leg
FIGURE 2-1: VO2peak before and after three months of quasi-follicular phase-based (qFT)
or quasi-luteal phase-based (qLT) endurance training (N=14); Two-leg: Test with both legs,
Pre: before training, Post: after training, *: P < 0.025 post-training vs. pre-training
50
*
VO 2peak(ml/min/kg)
*
*
*
40
30
20
10
26.9 32.4
26.4 32.6
∆ 5.5
∆ 6.2
qFT
qLT
28.5 34.4
pre
post
∆ 5.9
26.8 31.2
∆ 4.4 †
0
qFT
Androge nicity = 0
qLT
Androge nicity > 0
FIGURE 2-2: VO2peak before and after three months of quasi-follicular phase-based
(qFT) or quasi-luteal phase-based (qLT) endurance training in two groups of subjects
taking OC without any androgenicity (N = 7) or with known androgenicity (N = 7)
42
STUDY 2-RESULTS
Pre: before training, Post: after training, *: P < 0.025 post-training vs. pre-training, †: P =
0.043 ∆qFT vs. ∆qLT in group of subjects taking OC with androgenicity group.
2.3.5 Maximum workload (Wattmax)
Wattmax of each single leg increased continuously during both types of training periodization (Figure 2-3). After the three cycles of endurance training
Wattmax of both single legs increased significantly without any difference between training periodization (Figure 2-4). Furthermore, Wattmax in the subjects taking OC pills without any androgenicity increased about the same as
in the subjects taking OC pills with known androgenicity (Figure 2-5).
*
50
35.9
*
28.1
40
∆ Wattmax (W)
n.s.
*
15.3
30
n.s.
20
32.3
25.5
n.s.
10
0
*
*
12.8
0
0
*
Pre
1st Training
qFT
qLT
2nd Training
3rd Training
Training Cycle s
FIGURE 2-3: Increase in Wattmax compared to the pre-training values during quasi-follicular
phase-based (qFT) or quasi-luteal phase-based (qLT) endurance training (N=14); Two-leg:
Test with both legs, Pre: before training, Training: training cycle, n.s.: not significant, *: P <
0.025 compared to pre-training
43
STUDY 2-RESULTS
Pre
Post
250
*
Wattmax (W)
200
*
*
150
100
50
91.8 124.1
89.9 125.8
148.7 173.7
∆ 32.3
∆ 35.9
∆ 25.0
qLT
Two-leg
0
qFT
FIGURE 2-4: Maximum workload (Wattmax) before and after three months of quasi-follicular
phase-based (qFT) or quasi-luteal phase-based (qLT) endurance training (N=14); Two-leg:
Test with both legs, Pre: before training, Post: after training, *: P < 0.025 post-training vs.
pre-training
200
*
*
Wattmax (W)
150
*
*
100
50
94.1 128.3
93.7 133.1
∆ 34.2
∆ 39.4
qFT
qLT
89.5 120.0 86.1 118.6
pre
post
∆ 30.5
∆ 32.5
qFT
qLT
0
Androge nicity = 0
Androge nicity > 0
44
STUDY 2-RESULTS
FIGURE 2-5: Maximum workload (Wattmax) before and after three months of quasifollicular phase-based (qFT) or quasi-luteal phase-based (qLT) endurance training in
two groups of subjects taking OC without any androgenicity (N = 7) or with known androgenicity (N = 7)
Pre: before training, Post: after training, *: P < 0.025 post-training vs. pre-training
2.3.6 Submaximal power output at a lactate concentration of 4
mmol/l (Wattlac4)
Wattlac4 of each single leg increased continuously during both types of training periodization (Figure 2-6). After the three cycles of endurance training
Wattlac4 of both single legs increased significantly without any difference between training periodization (Figure 2-7). Furthermore, Wattlac4 in the subjects taking OC pills without any androgenicity tended to increase more after qLT compared to qFT (P = 0.041), while Wattlac4 increased about the
same in the subjects taking OC pills with known androgenicity after both
training periodizations (Figure 2-8).
40
*
∆ Wattlac4(W)
23.4
*
30
17.3
*
n.s.
12.9
n.s.
20
n.s.
10
22.0
*
0
0
0
Pre
12.1
17.2
*
*
1st Training
2nd Training
qFT
qLT
3rd Training
Training Cycle s
FIGURE 2-6: Increase in Wattlac4 compared to the pre-training values during quasi-follicular
phase-based (qFT) or quasi-luteal phase-based (qLT) endurance training (N=14); Two-leg:
Test with both legs, Pre: before training, Training: training cycle, n.s.: not significant, *: P <
0.025 compared to pre-training
45
STUDY 2-RESULTS
Pre
Post
*
160
140
Wattlac4 (W)
*
*
120
100
80
60
40
72.5 94.5
73.9 97.3
113.7 131.4
20
∆ 22.0
∆ 23.4
∆ 17.7
qFT
qLT Two-leg
0
FIGURE 2-7: Power output at a lactate concentration of 4 mmol/l (Wattlac4) before and after
three months of quasi-follicular phase-based (qFT) or quasi-luteal phase-based (qLT) endurance training (N=14); Two-leg: Test with both legs, Pre: before training, Post: after
training, *: P < 0.025 post-training vs. pre-training
140
*
*
120
*
*
Wattlac4 (W)
100
80
60
40
20
75.8 98.1 71.9 100.1
∆ 22.3
∆ 28.2†
pre
post
69.7 90.4
75.9 94.5
∆ 20.7
∆ 18.6
qFT
qLT
0
qFT
qLT
Androgenicity = 0
Androgenicity > 0
46
STUDY 2-RESULTS
FIGURE 2-8: Power output at a lactate concentration of 4 mmol/l (Wattlac4) before and
after three months of quasi-follicular phase-based (qFT) or quasi-luteal phase-based
(qLT) endurance training in two groups of subjects taking OC without any androgenicity
(N = 7) or with known androgenicity (N = 7)
Pre: before training, Post: after training, *: P < 0.025 post-training vs. pre-training, †: P =
0.041, ∆qFT vs. ∆qLT in group of subjects taking OC without androgenicity.
2.3.7 Isometric muscle strength (Fmax)
Fmax increased both after qFT (P < 0.025) and qLT (P = 0.027) without any
difference between the two training periodizations. The differentiation into subjects with OC pills without or with known androgenicity revealed an increase
in Fmax after qLT and a trend for an increase after qFT in the subgroup taking pills without androgenicity, but no significant change in the subgroup
taking pills with known androgenicity (Table 2-3).
TABLE 2-3: Maximum isometric force (Fmax) and sum of muscle diameter (Mdm) of m.
rectus femoris, m. vastus intermedius and m. vastus lateralis before and after three
months of quasi-follicular phase-based (qFT) or quasi-luteal phase-based (qLT) endurance training (N=14), and mean values of Fmax and Mdm in the two groups of subjects
taking OC without any androgenicity (N = 7) or with known androgenicity (N = 7)
Pre-Training
N = 14
Androgenicity = 0
Fmax (N)
N=7
Androgenicity > 0
N=7
N = 14
Androgenicity = 0
Mdm (cm)
N=7
Androgenicity > 0
N=7
Post-Training
qFT
qLT
qFT
qLT
627 ± 160
615 ± 173
680 ± 206 *
657 ± 200 †
625 ± 188
640 ± 215
673 ± 236 ††
698 ± 228 *
628 ± 141
590 ± 130
687 ±189
615 ± 175
5.94 ± 0.70
5.79 ± 0.61
6.02 ± 0.56
6.01 ± 0.56
5.65 ± 0.65
5.52 ± 0.65
5.81 ± 0.59
5.79 ± 0.54
6.22 ± 0.66
6.05 ± 0.48
6.22 ± 0.48
6.22 ± 0.52 *
*: p < 0.025, post-training versus pre-training, †: p = 0.027, post-training versus pre-training,
†† p = 0.053, post-training versus pre-training
47
STUDY 2-RESULTS
2.3.8 Muscle diameter (Mdm)
The sum of Mdm of the three muscles remained unchanged after both
types of training periodization compared to the pre-training level. The differentiation into subjects with OC pills without or with known androgenicity revealed an increase in Mdm only in the subgroup taking pills with known
androgenicity after qLT (Table 2-3).
2.3.9 Muscle fiber characteristics
Freeze damage, created by freeze-thawing during preparation, is a major
artifact that affects morphological analysis in this type of study. Although
the least-damaged fibers in well-preserved regions were selected, there
was still evidence of minor damage. The volume of artifacts varied between
individuals, but pre- and post-training sample quality was similar so the results were not affected.
Muscle fiber characteristics of a subgroup of five subjects revealed broad
inter-individual variation. Although data have to be taken cautiously, there
was a trend for an increase in nuclei-to-fiber ratio after both types of training
periodization after three months of one-leg endurance training. Muscle fiber
type distribution and fiber diameter remained unaffected by the endurance
training intervention (Table 2-4).
TABLE 2-4: Muscle fiber type distribution (No%), fiber diameter (Fdm) and nuclei-to-fiber
ratio (N/F) before and after three months of quasi-follicular phase-based or quasi-luteal
phase-based endurance training (N=5)
Pre-Training
Post-Training
qFT
Type Ι
qLT
Type ΙΙ
No%
45.9± 8.5
Fdm
(μm)
N/F
50.3 ± 5.5 46.0 ± 4.0
Type Ι
qFT
Type ΙΙ
Type Ι
qLT
Type ΙΙ
Type Ι
Type ΙΙ
54.1 ± 8.5 41.9 ± 10.5 58.1 ± 10.5 39.0 ± 4.8 61.0 ± 4.8 45.9 ± 11.3 54.1 ± 11.3
48.7 ± 8.6
2.5 ± 0.5
46.3 ± 9.5 49.8 ± 9.4 45.0 ± 6.5 48.9 ± 5.2
2.6 ± 0.5
2.9 ± 0.7 †
45.8 ± 4.7
2.7 ± 0.5 †
FT: follicular phase-based training, LT: luteal phase-based training; †: p < 0.05, posttraining versus pre-training
48
STUDY 2-DISCUSSION
2.4
Discussion
This study is the first one about planning endurance training with respect
to hormonal fluctuations during the menstrual cycle in monophasic OC users. The main findings of the present investigation are comparable increments of VO2peak, Wattmax and Wattlac4 after both, quasi-follicular phasebased and quasi-luteal phase-based endurance training in OC users, a
slightly smaller increase in VO2peak in the subgroup of OC users taking a
pill with known androgenicity in qLT, and a slightly more pronounced increase in Wattlac4 in the subgroup of OC users taking a pill without
androgenicity in qLT.
OC are the main form of birth control in the general population and with
the introduction of low dose OC preparations, their use has increased in
athletic women (Rechichi et al., 2009). OC pill use in athletic women
matches the prevalence of use within the general community. OC pill reduces cycle-length variability and provides a consistent 28-day cycle by
systematically controlling concentrations of endogenous sex hormones,
reducing the natural production of estrogens and progesterone through inhibition of the pituitary secretion of gonadotropins, thus inhibiting ovulation
and preventing pregnancy. Monophasic pills provide the woman with fixed
doses of estrogen and progestogen over 21 days, followed by 7 days of
placebo (Bennell, White & Crossley, 1999, Burrows & Peters, 2007, SitrukWare, 2006).
Only one synthetic estrogen (ethinylestradiol) is found in today’s monophasic OC pills, compared with one of several progestogens. Ethinylestradiol
is hormonally effective by activating the estrogen receptor and thus is an
estrogen. While ethinylestradiol is considered to be responsible for insulin
resistance, progestins are associated with changes in the insulin half-life
and increased insulin response to glucose (Sitruk-Ware & Nath, 2011).
However, a review of studies in women without diabetes suggests limited
effects of hormonal contraceptives on carbohydrate metabolism, indicating
that there is no strong evidence of a diabetogenic effect of OC pills, but
the few studies with limited sample size and poor reporting of methods are
not conclusive (Lopez, Grimes & Schulz, 2012).
49
STUDY 2-DISCUSSION
Endurance exercise training induces increase in energy metabolism and
oxidative capacity, as well as changes in muscle fiber type (Gravelle &
Blessing, 2000). Testosterone has been shown to effect mitochondrial
function and oxidative metabolism in men (Pitteloud et al., 2005). No data
in women are available concerning these effects. The ovarian hormones,
estrogen and progesterone, have important roles in regulating substrate
metabolism during exercise in women. In animal models, estrogen promotes lipolysis and increases fatty acid availability while decreasing the
rate of gluconeogenesis and sparing muscle and liver glycogen use. The
addition of progesterone has been reported to antagonize the lipolytic effects of estrogen and reduce fatty acid availability. Conversely, the addition of progesterone appears to accentuate the carbohydrate-sparing actions of estrogen by decreasing hepatic glycogenolysis. Furthermore, estrogen up-regulates mitochondrial enzymes favoring fat oxidation, whereas
progesterone opposed these actions.
Taken together, alterations in substrate use and regulation of enzyme activities during exercise across the menstrual cycle in OC users and in nonOC users are dependent on the relative changes in both estrogen and
progesterone and other anabolic sex steroids, and possibly also on the
andogenic potential of the gestagen in the pill (D'Eon et al., 2002, Rechichi
et al., 2009). Therefore, the amount of adaptation processes to endurance
training in our study was expected to also depend on the endocrine milieu
when the training stimulus was set.
In contrast to the endurance training study with subjects without taking OC
(Han et al., 2012), where Wattmax showed a significant but delayed sharper
increase after FT compared to LT, the increase in Wattmax in this study
was the same in OC users after qFT and qLT. VO2peak and Wattlac4 increased significantly after three month of endurance training in both studies, with and without OC usage, without any different effect of training periodization. The increase in Wattmax in this study with OC usage was in line
with a slight increase in Fmax after the training period in the mainly untrained subjects, again independent from the periodization of endurance
training. Despite the above mentioned hormonal factors possibly influencing adaptation to endurance training depending on the phase of the (pill)
50
STUDY 2-DISCUSSION
cycle when the stimulus is set, no gradually different adaptations occurred
in our well-controlled one-leg endurance training intervention.
Hormone analysis in this study revealed that concentration of E2 and P4
did not differ significantly between day 11 and day 25 of qFP and qLP
(Table 2-2). We do not have any information about these hormone concentrations throughout the rest of the cycle in this study. Other investigations have shown that after OC intake, endogenous E2 and P4 are suppressed for 21 days (Rechichi et al., 2009). Nevertheless, the level of endogenous E2 is significantly higher in the late OC withdrawal phase (6-7
days post active OC cessation) than in the early OC withdrawal phase (2-3
days post active OC cessation) and in the OC consumption phase. Both
endogenous E2 and P4 remain suppressed during the early withdrawing
phase, however, later in the withdrawing phase endogenous E2 increases
while P4 continues to be suppressed, indicating that it takes some time for
E2 to increase after OC withdrawing. We, therefore, assume that the similar behavior of any of the parameters of endurance performance after qFT
and qLT might be due to the more or less constant concentrations of the
sex steroid hormones during the OC consumption phase, or to an overlapping of effects of the withdrawing phase and the early consumption
phase which occurred both during the 14 days of qFT.
The only hormone showing significant differences between qFP and qLP
in our study was DHEA-s with significantly lower values in qLP compared
to qFP both, prior to and after the training period (Table 2-2). This is partly
in line with findings in the literature, where DHEA-s, testosterone and free
testosterone have been shown to increase significantly during the withdrawal phase after ceasing OC (Wiegratz, Jung-Hoffmann & Kuhl, 1995).
Increase in DHEA-s might be more prolonged as possible increments of T
and free T, as it still was higher on day 11 compared to day 25 in our study,
while T and free T were not different between these days.
The form of progestogen used in the OC pill will oppose estrogen to varying levels depending on its potency and androgenicity. Apart from binding
to the progesterone receptor, most progestins according to their chemical
structure could also interact with the androgen receptor, estrogen receptor,
glucocorticoid receptor or mineralocorticoid receptor and by these mechanisms can influence metabolic parameters (LaGuardia, Shangold,
51
STUDY 2-DISCUSSION
Fisher, Friedman & Kafrissen, 2003) and, consecutively, training adaptation processes. Androgenicity refers to the ability of the progestogen to
produce masculine characteristics and is calculated by multiplying the
progestogen dose within the OC pill by its androgenic activity. As the more
androgenic progestogens oppose the estrogen effects, one would expect
the
OC
pills
containing
progestogens
with
higher
potency
and
androgenicity to have a more significant impact on performance then
those
OC
pills
containing
progestogens
with
low
potency
and
androgenicity (Burrows & Peters, 2007).
In our investigation, subjects taking monophasic OC have been included.
However, as we wanted to include a representative group of women taking
different monophasic preparations into the study, we did not further reduce
inclusion criteria to some specific monophasic OC preparations. Therefore,
the type of monophasic preparation varied between subjects concerning
both, the amount of ethinylestradiol (20 – 35 µg) and the type and amount
of progestogen with their different levels of androgenicity (Table 2-1). Analyzing for potential effects of androgenicity of the OC pill, we divided subjects into two groups. One group used OC without any androgenic potential, and the other group used OC with known androgenicity. Effects of
androgenicity on the development of parameters of endurance performance, however, were inconsistent. While VO2max increased less under
androgenic
influence
in
qLT,
Wattmax
remained
unaffected
by
androgenicity, and the submaximal performance parameter Wattlac4 increased more without any influence of androgenicity in qLT. The few studies in the literature on the effects of OC on aerobic capacity throughout a
menstrual cycle are also inconsistent. When looking at the studies utilizing
the newer low dose monophasic OC pills and more sophisticated study
designs, the evidence suggests that there is no effect of monophasic OC
pills on aerobic capacity in untrained women (Burrows & Peters, 2007).
Somewhat conflicting data could be due to the different progestogens utilized and their associated progestational and androgenic activity, or the
fact that different progestogens take different lengths of time to exert their
effect on aerobic capacity. Further research is necessary to elucidate the
exact effects of the various monophasic OC pills on aerobic capacity and even more important – on trainability of endurance performance.
52
STUDY 2-DISCUSSION
This study is the first to investigate muscle fiber parameters depending on
menstrual cycle phase-based endurance training in OC users. Only five of
14 subjects volunteered to participate in needle muscle biopsies. Therefore, data have to be interpreted carefully. All muscle cell parameters
showed broad inter-individual variation (Table 2-4).
We could demonstrate a clear trend for an increase in N/F after both training periodization without any significant changes in fiber type composition
and fiber diameter. N/F was analyzed by means of hematoxylin and eosin
staining, which might represent satellite cell activation and incorporation
into the muscle cell. Endurance training is known to mainly stimulate type I
fibers and to induce conversion of type II into type I fibers (Tegtbur, Busse
& Kubis, 2009). However, no such changes could be detected in this study.
We therefore assume that training volume and/or training intensity was not
high enough in order to induce remarkable alterations in these specific
muscle cell parameters. We conclude that specific training induced activation of molecular pathways and/or anabolic hormonal influence on skeletal
muscle cell adaptation was either inconsistent or not sufficient. To further
analyze possible effects of menstrual cycle-based endurance training in
OC users, further studies including larger samples are required. Moreover,
further studies should not only differentially analyze fiber type Ι and ΙΙ but
they also should include the analysis of fiber type IΙ subtypes (Kraemer et
al., 1995; Scott, Stevens & Binder-Macleod, 2001).
In summary, we could not demonstrate any significantly different effect on
physiologic and microscopic measures of endurance capacity between
qFT and qLT in monophasic OC users. We assume that overlapping effects between suppression and stimulation of endogenous sex steroids,
exogenous steroid hormones within the pill, and androgenicity of the
gestagen component of OC pills exist and interfere with adaptation processes to endurance training. We therefore conclude that untrained or
moderately trained women taking monophasic OC do not need to consider
the phase of the menstrual cycle when planning their endurance training
program. More studies, however, are necessary in order to deeper understand the underlying mechanisms of endurance training adaptation in OC
users.
53
STUDY 3-ABSTRACT
STUDY 3 – EFFECTS
OF
MENSTRUAL CYCLE PHASE-BASED ENDUR-
ANCE TRAINING IN NON-OC USERS VERSUS OC USERS
Abstract
Purpose: Hormonal variations during the menstrual cycle may influence
trainability of endurance. However oral contraception (OC) alters the profile of these hormones. For this reason, we compared the effect of menstrual cycle-based endurance training between eumenorrheic females
(non-OC users) and oral contraceptive using females (OC users).
Methods: Females (N = 27: non-OC users = 13, OC users = 14) completed one-leg endurance training on a cycle ergometer for three menstrual
cycles. They trained one leg mainly in the first half of the menstrual cycle
(follicular phase training (FT) or quasi-follicular phase training (qFT)) and
the other leg mainly in the second half of the cycle (luteal phase training
(LT) or quasi-luteal phase training (qLT). Concentrations of 17-beta estradiol (E2), progesterone (P4), total testosterone (T), free testosterone (free
T) and DHEA-s were analyzed in blood samples taken during follicular
phase (FP)/ quasi-follicular phase (qFP) and luteal phase (LP)/ quasiluteal phase (qLP). Peak oxygen uptake (VO2peak), maximal workload
(Wattmax) and power output at a lactate concentration of 4mmol/l (Wattlac4)
were analyzed before and after training.
Results: Concentration of E2, T and free T were significantly higher in
non-OC users compared to OC users (p<0.05). P4 level was highest in LP
compared to all other phases (p<0.05). Absolute increase of Wattmax was
the lowest after LT in non-OC users (+ 27.8 W) compared to FT (+ 34.8 W)
and qFT (+ 32.3 W) and qLT (+ 35.9 W) in OC users. VO2peak tended to increase after FT (+3.3 ml/min/kg, P = 0.033) and LT (+2.8 ml/min/kg, P =
0.038) and increased significantly after qFT (+5.7 ml/min/kg) and qLT
(+5.3 ml/min/kg) without any differences between non-OC and OC-users.
Conclusions: Anabolic hormones are clearly higher in non-OC users
compared to OC users. Amount of training adaptation, however, was not
reflected by hormone differences. Further studies with more subjects are
54
STUDY 3-INTRODUCTION
needed in order to understand the underlying mechanisms of training adaptations in non-OC users and OC users.
3.1
Introduction
Fluctuation of hormones, such as estradiol (E2), progesterone (P4) and
testosterone over the course of the menstrual cycle has been repeatedly
reported in eumenorrheic females (non-OC users). 17-beta estradiol (E2)
peaks prior to ovulation and during the luteal phase (LP), while progesterone (P4) reaches its highest values during LP after ovulation (Van Look et
al. 1980). In both sexes, androgens are produced by the reproductive organs and the adrenals. The most important androgen secreted is testosterone; the adrenal glands and the ovaries produce very little testosterone
but secrete weaker androgens. In particular, dehydroepiandrosterone
(DHEA;
and
its
sulfoconjugate)
secreted
by
the
adrenals,
and
androstenedione secreted by the adrenals and the ovaries are of physiological importance in women (Enea, Boisseau, Fargeas-Gluck, Diaz &
Dugue, 2011). Moreover, the levels of androstenedione and testosterone,
for instance, reach their peaks prior to, or at the time of ovulation
(Longcope, 1986).
Oral contraceptive users (OC users) have different hormone concentrations compared to non-OC users due to the intake of fixed doses of synthetic E2 and progestin. The exogenous estrogen and progestin in oral
contraceptive pills suppress the production and secretion of endogenous
estrogen and progesterone (Dooley & Brincat, 1994; Rechichi et al., 2009).
The endogenous estrogen and progesterone are suppressed for 21 days
while oral contraceptives are taken in the consumption phase. After 21
days of the consumption phase, no oral contraceptives are taken for 7 day
(the withdrawal phase) and menstrual bleeding occurs. The concentration
of both hormones are continually suppressed in the early withdrawal
phase and estrogen starts to increase in the late withdrawal phase, while
progesterone remains further suppressed (Dooley & Brincat, 1994;
Rechichi, Dawson & Goodman, 2008; Rechichi et al., 2009).
Due to this exogenous and endogenous hormonal regulation, levels of testosterone, free testosterone and dehydroepiandrosterone (DHEA; and its
55
STUDY 3-INTRODUCTION
sulfoconjugate) decrease significantly with usage of oral contraceptives
(Graham et al., 2007; Rickenlund et al., 2004). E2 was found as well to be
significantly lower in OC users in comparison to non-OC users (Vaiksaar
et al., 2011). Moreover, P4 remains low in the quasi luteal phase in OC
users, whereas it increases significantly in the luteal phase compared to
the follicular phase in non-OC users (Vaiksaar et al., 2011).
The main sex hormones such as E2 and P4 are known to influence substrate metabolism during endurance exercise. An animal study demonstrated that E2 increases glucose availability and promotes contractionstimulated glucose uptake into type Ι muscle fibers during short duration of
high intensity aerobic exercise (Campbell & Febbraio, 2002). Moreover,
ovariectomized rats with E2 injection showed significant higher glycogen
sparing in skeletal muscle and liver compared to control rats and they ran
significantly longer and completed more work on the treadmill during submaximal exercise (Kendrick, Steffen, Rumsey & Goldberg, 1987).
D´Eon et al. (D'Eon et al., 2005) indicated from an animal study with ovariectomized rodent that E2 promotes fat oxidation in muscle and the use of
lipid as fuel. They also found an increased AMP-activated protein kinase
(AMPK) activity in skeletal muscle of E2-treated mice. They suggested that
the activation of AMPK leads to higher free fat acid uptake into mitochondria and increase fat oxidation which might benefit exercise performance.
Rogers et al. (Rogers, Witczak, Hirshman, Goodyear & Greenberg, 2009)
as well found a rapid increase in AMPK activation in skeletal muscle after
E2 treatment and they stated this increased AMPK activity stimulates
muscle sensitivity to insulin.
Progesterone was reported to play an anti-estrogenic roll (Oosthuyse &
Bosch, 2010) and progesterone is suspected to have a negative influence
on protein metabolism which might reduce performance (Kriengsinyos,
Wykes, Goonewardene, Ball & Pencharz, 2004; Oosthuyse & Bosch,
2010). Progesterone inhibits estrogenic action of promoting glucose uptake during short endurance exercise (Campbell & Febbraio, 2002).
Campbell and Febbraio (Campbell & Febbraio, 2002) examined glucose
uptake in ovariectomized rats treated with either E2 (E), progesterone (P),
same dose of progesterone and E2 (P+E), progesterone with high dose of
E2 (P+HiE). During the aerobic exercise, glucose uptake restored only in E
56
STUDY 3-INTRODUCTION
and P+HiE rats whereas it decreased in P and P+E rats. Another study
(Campbell & Febbraio, 2001) from the same author indicated that E2 increases the maximal activity of key enzymes in the fat oxidative pathway
of skeletal muscle and this lipolytic effect is inhibited by progesterone.
These two studies demonstrated that E2 has a positive effect only in the
absence of progesterone or when concentrations of E2 were higher than
progesterone (P+HiE). Therefore, despite increased concentrations of E2
in LP, all the potential benefits of E2 might be antagonized by elevated
concentrations of progesterone.
All kinds of OC can influence fat and carbohydrate metabolism (Dooley &
Brincat, 1994; Kiley & C., 2007; Rechichi et al., 2009). They might cause
deterioration in glucose tolerance, increase of peripheral resistance to insulin and increases of lipoprotein metabolism. However, these effects depend on the doses of exogenous estrogen and progesterone (Dooley &
Brincat, 1994; Kiley & C., 2007). Furthermore, monophasic OC might affect endurance performance (Burrows & Peters, 2007) with possible effects on body composition, metabolism, aerobic capacity and cardiovascular responses. OC pill induced increase in basal body temperature, for instance, could cause decline in long duration performance. There are some
studies to suggest that endurance performance is not affected by changes
from active to inactive OC pill phase, although there is also a potential for
variation in endurance performance throughout an OC cycle (Vaiksaar et
al., 2011). Differences in ventilation, oxygen consumption and substrate
metabolism between studies appear to relate to variations in the types of
OC used and the dosage of progestogen administered (Rechichi et al.
2009). Altogether, research in the area of OC and exercise capacity is
sparse and much has been plagued by poor research design, methodology and small sample size (Burrows & Peters, 2007).
Since estrogen, progesterone and other sex steroids are discussed to be
important factors for endurance capacity, there might be yet unknown different influences on endurance training adaptation between non-OC users
and OC users. Nevertheless, to the authors’ knowledge, there are no training interventional studies that have compared the trainability of endurance
capacity of menstrual cycle-based training between OC users and non-OC
users.
57
STUDY 3-INTRODUCTION
Therefore, the purpose of this study was to compare the hormone profile
and the endurance training effects on physiologic and microscopic
measures of aerobic capacity depending on the menstrual cycle phase,
when the training stimuli are set, e.g. mainly in the follicular phase or in the
luteal phase, respectively, between non-OC users and OC users.
58
STUDY 3-METHODS
3.2
Methods
Study 3 compares the results of both Study 1 and Study 2. For the experimental design, the study schedule, the training program and analyzing
methods please refer to Study 1 and Study 2
3.2.1 Statistical Analysis
Data are presented as mean values with SD. Normality of distributions
was proved by the Kolmogorov-Smirnov test. A one-tailed independent ttest was used to evaluate differences in training workload, VO2peak, Wattmax,
Wattlac4, Fmax and Mdm between values before (pre) and after the training
intervention (post) (see below: a, b) and between FT and LT (see below:
c), respectively. In all cases, P values < 0.025 were taken to indicate statistical significance. Statistics were tested with a hierarchical procedure:
1. FT vs. qFT: a) FTpost better than FTpre; b) qFTpost better than qFTpre;
c) if a) significant∆FTpost -pre better than∆qFTpost -pre; if b) significant
∆qFTpost-pre better than ∆FTpost-pre
2. LT vs. qLT: a) LTpost better than LTpre; b) qLTpost better than qLTpre;
c) if a) significant∆LTpost -pre better than ∆qLTpost -pre; if b) significant
∆qLTpost-pre better than ∆LTpost-pre.
A two-tailed independent t-test was used to compare hormone concentration between two groups (FP vs. qFP, LP vs. qLP). Significance was defined as p < 0.05. The intraclass correlation coefficient of repeated measurements (ICC) (McGraw & Wong, 1996) was determined to evaluate reliability of the determination of Fmax and Mdm.
59
STUDY 3-RESULTS
3.3
Results
3.3.1 Number of training sessions
The total number of single-leg training sessions was approx. 20 sessions
per leg and was not different between FT, LT, qFT and qLT (FT: N = 19.9
± 1.9; LT: N = 20.5 ± 1.9; qFT: N = 20.2 ± 1.5; qLT: N = 20.4 ± 1.7; P >
0.05).
3.3.2 Hormone concentrations
The hormone values before training and after training were calculated together for each FP, LP, qFP and qLP to compare the concentrations between phases and between non-OC users and OC users independently
from training effect.
Concentration of E2, T and free T were significantly higher in non-OC users (FP and LP) compared to OC users (qFP und qLP). P4 level was the
highest in LP compared to other phases (p < 0.05) whereas DHEA-s and
free T showed the lowest values in qLP (p < 0.05) (Table 3-1).
TABLE 3-1: Serum concentrations of E2, P4, DHEA-s, T and free T in the (quasi-) follicular phase (FP/qFP, day 11) and the (quasi-) luteal phase (LP/qLP, day 25) (N=27)
non-OC users
E2
(pg/ml)
P4
(ng/ml)
DHEA-s
(ug/ml)
T
(ng/ml)
Free T
(pg/ml)
OC users
FP
LP
qFP
qLP
76 ± 44 *
100 ± 74 #
10.2 ± 7.9
12.2 ± 11.4
0.65 ± 0.31 †
6.53 ± 4.17#
0.49 ± 0.19
0.42 ± 0.20
1.95 ± 0.80
2.07 ± 0.86 #
1.48 ± 0.52 ††
1.24 ± 0.48
0.34 ± 0.21 *
0.32 ± 0.17#
0.14 ± 0.08
0.13 ± 0.08
1.73 ± 0.38 †*
1.50 ± 0.40#
0.85 ± 0.16 ††
0.77 ± 0.17
E2: estradiol, P4: progesterone, T: testosterone.
†: P < 0.05 FP vs. LP, †† : P < 0.05 qFP vs. qLP, * : P < 0.05 FP vs. qFP, # : p < 0.05 LP
vs. qLP
60
STUDY 3-RESULTS
3.3.3 Peak oxygen uptake (VO2peak)
Three months of one-leg endurance training showed a significant increase
in VO2peak after qFT (∆ +5.7 ± 2.3 ml/min/kg) und qLT (∆ +5.3 ± 3.5
ml/min/kg), respectively, whereas it showed only a tendency to increase
after FT (∆ +3.1 ± 5.3 ml/min/kg, p = 0.033) and LT (∆ +2.8 ± 4.8 ml/min/kg,
p = 0.038). When absolute increases (∆ post -pre) after training intervention were compared among the four groups (∆FT, ∆LT, ∆qFT and ∆qLT),
there was no significant differences between them (Figure 3-1).
10
*
#
∆ VO 2peak (ml/min/kg)
8
##
*
6
4
2
∆ 3.1
∆ 2.8
∆ 5.7
∆ 5.3
FT
LT
qFT
qLT
0
non-OC
OC
FIGURE 3-1: Absolute increases (∆) of VO2peak after three months of follicular phasebased (FT), luteal phase-based (LT), quasi-follicular phase-based (qFT) or quasi-luteal
phase-based (qLT) endurance training (N=27); #: p = 0.033, ##: p = 0.038, *: p < 0.025
post-training vs. pre-training
3.3.4 Maximum workload (Wattmax)
Wattmax of each single leg increased continuously during all types of training periodization (Figure 3-2). Comparing the absolute increases ∆post
(
pre) after training intervention among the four groups, FT tended to induce
higher increase of Wattmax comared to LT (p = 0.038).
61
STUDY 3-RESULTS
Moreover, LT seemed to have the lowest increase of Wattmax compared to
FT, qFT and qLT, however, the difference was not significant (Figure 3-3).
60
50
∆ Wattmax (W)
40
30
20
10
FT
LT
qFT
qLT
0
Pre
1st Training
2nd Training
3rd Training
Training Cycle s
FIGURE 3-2: Progressive increases in Wattmax compared to the pre-training values during
follicular phase-based (FT), luteal phase-based (LT), quasi-follicular phase-based (qFT)
or quasi-luteal phase-based (qLT) endurance training (N=27); Pre: before training, Training: training cycle; values are not shown in the graph
60
*
50
*
*
*
∆ Wattmax (W)
40
30
20
∆ 34.8
∆ 27.8 †
∆ 32.3
∆ 35.9
FT
LT
qFT
qLT
10
0
non-OC
62
OC
STUDY 3-RESULTS
FIGURE 3-3: Absolute increases ∆
( ) of m aximum workload (Wattmax) before and after
three months of follicular phase-based (FT), luteal phase-based (LT), quasi-follicular
phase-based (qFT) or quasi-luteal phase-based (qLT) endurance training (N=27), *: p <
0.025 post-training vs.pre-training, †: p = 0.038 ∆FT vs. ∆LT
3.3.5 Submaximal power output at a lactate concentration of 4
mmol/l (Wattlac4)
Wattlac4 of each single leg increased significantly during all four training
programs (Figure 3-4). Absolute increase of Wattlac4 was not significantly
different among FT, LT, qFT and qLT (∆FT: 27.6 ± 22.3 W, ∆LT: 28.6 ± 18.6
W, ∆qFT: 22.0 ± 37.6 W, ∆qLT: 23.4 ± 12.5 W, P > 0.025; Figure 3-5).
50
∆ Wattlac4(W)
40
30
20
10
0
FT
LT
qFT
qLT
-10
Pre
1st Training
2nd Training
3rd Training
Training Cycle s
FIGURE 3-4: Progressive Increases in Wattlac4 compared to the pre-training values during
follicular phase-based (FT), luteal phase-based (LT), quasi-follicular phase-based (qFT)
or quasi-luteal phase-based (qLT) endurance training (N=27), Pre: before training, Training: training cycle; values are not shown in the graph
63
STUDY 3-RESULTS
60
∆ Wattlac4 (W)
50
*
*
40
*
*
30
20
10
∆ 27.6
∆ 28.6
∆ 22.0
∆ 23.4
FT
LT
qFT
qLT
0
non-OC
OC
FIGURE 3-5: Absolute increases (∆) of Power output at a lactate concentration of 4
mmol/l (Wattlac4) before and after three months of follicular phase-based (FT), luteal
phase-based (LT), quasi-follicular phase-based (qFT) or quasi-luteal phase-based (qLT)
endurance training (N=27), *: P < 0.025 post-training vs. pre-training
3.3.6 Isometric muscle strength (Fmax)
Fmax tended to decline after three months of FT (P = 0.034) and LT (P =
0.05) without any difference between both training periodization protocols.
However, increases were observed after qFT (P < 0.025) and qLT (P =
0.027) without any significant difference between the two training interventions
(Table 3-2).
64
STUDY 3-RESULTS
TABLE 3-2: Absolute increase of maximum isometric force (Fmax) and Sum of muscle diameter (Mdm) (m. rectus femoris, m. vastus intermedius, m. vastus lateralis) compared to
the pre-training values during (quasi-) follicular phase-based (FT/qFT) or (quasi-) luteal
phase-based endurance training (LT/qLT) (N=27).
∆Post- Pre
∆FT
Fmax (N)
∆LT
∆qFT
-36 ± 65 # -43 ± 88 ## 54 ± 75 *
∆qLT
42 ± 74 †
Mdm (cm) 0.11 ± 0.34 -0.03 ± 0.37 0.08 ± 0.37 0.22 ± 0.31
*: p < 0.025, post-training versus pre-training, †: p = 0.027, post-training versus pretraining, #: P = 0.034, ##: P = 0.05
3.3.7 Muscle diameter (Mdm)
The sum of Mdm of the three muscles remained unchanged (P > 0.025)
after all types of training periodization compared to the pre-training level
(Table 3-2).
65
STUDY 3-DISCUSSION
3.4
Discussion
This study is the first one about comparing endurance trainability between
non-OC users and OC users with respect to hormonal fluctuation during
the menstrual cycle. The main findings of this study are significantly higher
concentrations of anabolic hormones in non-OC users compared to OC
users. However, none of the parameters of endurance capacity showed
significant differences between two groups. A trend for higher increase in
VO2peak was observed in OC users (qFT and qLT) compared to non-OC
users (FT and LT) and with the lowest increase after LT. Moreover, LT
showed the lowest increase in Wattmax as well in comparison to FT (P =
0.038), qFT and qLT.
The form of progestogen used in the OC pill is known to oppose estrogen
to varying levels depending on its potency and androgenicity. Apart from
binding to the progesterone receptor, most progestins according to their
chemical structure could also interact with the androgen receptor, estrogen receptor, glucocorticoid receptor or mineralocorticoid receptor and by
these mechanisms can influence metabolic parameters (LaGuardia,
Shangold, Fisher, Friedman & Kafrissen, 2003) and, consecutively, training adaptation processes. As the more androgenic progestogens oppose
the estrogen effects, one would expect the OC pills containing
progestogens with higher potency and androgenicity to have a more significant impact on performance then those OC pills containing progestogens
with low potency and androgenicity (Burrows & Peters, 2007). Nevertheless, the newer low does monophaseic OC pills, which were used in present study, seem to have no effect on aerobic capacity in untrained women (Burrows & Peters, 2007).
In another study from our group, we compared strength training adaptation
between non-OC users and OC users. In this study, the lowest increase (p
< 0.025) in maximal isometric muscle strength was observed after luteal
phase-based strength training (non-OC) compared to follicular phasebased (FT), quasi-follicular phase-based (qFT) and quasi-luteal phasebased (qLT) training. This is in the line with the present study showing the
lowest increase in Wattmax in LT.
66
STUDY 3-DISCUSSION
Pronounced differences between non-OC users and OC users were the
concentration of androgen hormones. Significantly lower concentrations of
E2, T and free T were observed in OC users in both qFP and qLP compared with FP and LP in non-OC users. Amount of training adaptation,
however, was not reflected by anabolic hormone differences.
An important hormone to focus on is P4 showing the highest level in LP
(6.53 ± 4.17 ng/ml) of non-OC users. It was significantly (p < 0.05) higher
compared to all other phases (FP: 0.65 ± 0.31 ng/ml, qFP: 0.49 ± 0.19
ng/ml, qLP: 0.42 ± 0.20 ng/ml). As gestagens have been shown to induce
protein catabolism and to have other influences on metabolism, this clearly
gestagenic condition might expain the lowest increase of Wattmax in LT.
Estrogen (E) is repeatedly reported to be the main important factor for increasing endurance capacities. Previous studies demonstrated that E
promotes contraction-stimulated glucose uptake into type Ι muscle fibers
during short duration of high intensity aerobic exercise (Campbell & Febbraio, 2002), advances time to exhaustion in a prolonged submaximal
treadmill run (Kendrick et al., 1987) by sparing glycogen and by increasing
the availability and oxidation of lipid substrate during exercise (Kendrick et
al., 1987) and E also stimulates AMP-activated protein kinase (AMPK) in
skeletal muscle (Rogers et al., 2009).
In contrast, progesterone (P4) was reported to display anti-oestrogenic effects (Oosthuyse & Bosch, 2010) by inhibiting estrogenic action of promoting glucose uptake during a short endurance exercise (Campbell & Febbraio, 2002). Moreover, P4 was suspected to enhance amino acid oxidation (Landau & Lugibihl, 1961, 1967), which cause significant higher protein catabolism in the luteal phase of non-OC users (Kriengsinyos et al.,
2004; Lamont, Lemon & Bruot, 1987).
Taking this into account, Janse de Jonge (2003) suggested to consider the
ratio of estrogen (pg/ml) and progesterone (ng/ml) (E/P) due the interactive effects of estrogen and progesterone, which influence their secondary
effects. Oosthuyse and Bosch (2010), as well, proposed to investigate the
E/P ratio across the menstrual cycle, specifically in the late FP in non-OC
users, when E2 reaches its peak while P4 remains low.
67
STUDY 3-DISCUSSION
In fact, Oosthuyse et al. (2005) found a trend for a more improved time trial performance in the late FP compared to the early FP based on finishing
time (p = 0.0265). In late FP, estrogen concentration reaches its peak
without the antagonistic effects of progesterone and a high E/P ratio could
allow estrogen to mitigate the anti-estrogenic actions of progesterone
(Oosthuyse, Bosch & Jackson, 2005).
The E/P ratio in our study was highest in FP (134.0 ± 99.0) and lowest in
LP (16.4 ± 4.9) in non-OC users, whereas it was 27.7 ± 30.2 and 37.8 ±
41.3 in qFP and qLP, respectively, in OC users. This fact might explain the
tendentiously higher increase in Wattmax after FT compared to LT as well
as the no significant difference between qFT and qLT. Moreover, the relative lower E/P ratio and higher P4 concentration in LP corresponded with
relatively lower increase in endurance capacity after LT. Therefore, we
suspect that this hormonal milieu influenced training adaptation negatively
in LP.
In conclusion endurance training adaptation between non-OC users and
OC users were comparably high. However, the endurance training in the
luteal phase in non-OC users showed somewhat lower increases in parameters of endurance capacities compared to the training in the follicular
phase in non-OC users, in the quasi follicular phase in OC users and to
the training in the luteal phase in OC users. This outcome could be related
to the different hormonal milieu among menstrual phases. In order to further understand the underlying mechanisms of endurance training adaptations on the physiologic and microscopic measures of aerobic capacity,
further studies with more subjects are needed. Since there might be different training effects between non-OC users and OC users, investigators
should take this into consideration for their future research with female
subjects.
68
SUMMARY
SUMMARY
This thesis investigated endurance training adaptations in different phases
of the menstrual cycle in non-OC users as well as in OC users. Remarkable differences in trainability were observed between follicular phasebased compared to luteal phase-based trainings as well as between nonOC users and OC users.
The follicular phase-based endurance training showed remarkably higher
increase (p = 0.038) in maximal power output after 12 weeks compared to
the luteal phased based training in non-OC users whereas no difference
was observed between quasi-follicular phase- and quasi-luteal phasebased training in OC users.
The hormone analysis on day 11 and on day 25 of the menstrual cycle revealed different hormone milieus between non-OC users and OC users. In
non-OC users, the androgen hormones such as estrogen and testosterone
were significantly higher in the follicular phase as compared to in the luteal
phase. In the contrast, progesterone level was significantly higher in the
luteal phase. These different hormone milieus might have induced higher
training adaptation in the follicular phase compared to the luteal phase.
Estrogen is known to have positive influences on carbohydrate, fat and
protein metabolism and it was also repeatedly reported to be an important
factor for increasing endurance capacity. In contrast, progesterone displays anti-estrogenic effects and is suspected to enhance protein catabolism in the luteal phase. Therefore, many researchers suggest the ratio of
estrogen and progesterone (E/P) over the course of the menstrual cycle to
be taken into account. In non-OC users, ratio of E/P was the highest in FP
and the lowest in LP indicating that high E/P in FP could allow estrogen to
mitigate the anti-estrogenic actions of progesterone.
In contrast to non-OC users, concentrations of estrogen, testosterone and
progesterone did not differ significantly between the quasi-follicular phase
and the quasi-luteal phase in the subjects taking combined monophasic
oral contraceptives. Due to the intake of constant concentrations of exogenous estrogen and progesterone over 21 days, the endogenous androgen hormones are suppressed and remain nearly unchanged during
69
SUMMARY
the OC consumption phase. Therefore, OC use caused indifferent levels
of estrogen, progesterone and testosterone resulting in comparable trainability between the two phases.
Based on the above described main outcomes, we recommend that nonOC users should consider their individual menstrual cycle for scheduling
endurance training sessions, since trainability seems to be effected by the
menstrual phase when the stimulus is set with highest results after follicular phased-based training. Furthermore, OC users can perform endurance
training without concerning their menstrual cycle, because cycle phase
does not have any influences on training adaptation.
Comparing effects of endurance training between non-OC users and OC
users, the luteal phase-based training in non-OC users induced the lowest
increase in parameters of endurance capacity, while the highest concentration of progesterone was detected, all compared to the training in the
follicular phase in non-OC users, in the quasi-follicular phase in OC users
and in the quasi-luteal phase in OC users. Moreover, OC users showed a
more stable hormonal milieu over the course of the menstrual cycle, resulting in more stable endocrine condition for training adaptation compared to
non-OC users during the consumption phase.
Nevertheless, the number of subjects in both studies was small so that final conclusions should be drawn carefully. Therefore, the suspicious lower
endurance capacity in the luteal phase in non-OC users should be confirmed in studies with a larger number of subjects to clarify the underlying
hormonal influences of endurance training over the course of the menstrual cycle in non-OC users and OC users. Furthermore, endurance training
periodization studies with respect to the withdrawing phase, when endogenous production of sex steroid hormones is induced again, as well as
larger studies with OC preparations of different androgenicity should be
analyzed in future studies in order to better understand the underlying
mechanisms of endurance training adaptation in OC users.
4.
70
REFERENCES
REFERENCES
Alexander, G.M., Sherwin, B.B., Bancroft, J. & Davidson, D.W. (1990).
Testosterone and sexual behavior in oral contraceptive users and
nonusers: a prospective study. Horm Behav, 24(3), 388-402.
Allen, D.L., Harrison, B.C. & Leinwand, L.A. (2002). Molecular and genetic
approaches to studying exercise performance and adaptation.
Exerc Sport Sci Rev, 30(3), 99-105.
Bennell, K., White, S. & Crossley, K. (1999). The oral contraceptive pill: a
revolution for sportswomen? Br J Sports Med, 33(4), 231-238.
Bergström, J. (1962). Muscle electrolytes in man. Scand J Clin Lab Invest.,
14 (supply 68), 1-110.
Brooke, M.H. & Kaiser, K.K. (1970). Three "myosin adenosine
triphosphatase" systems: the nature of their pH lability and sulfhydryl dependence. J Histochem Cytochem, 18(9), 670-672.
Burrows, M. & Peters, C.E. (2007). The influence of oral contraceptives on
athletic performance in female athletes. Sports Med, 37(7), 557-574.
Constantini, N.W., Dubnov, G. & Lebrun, C.M. (2005). The menstrual cycle
and sport performance. Clin Sports Med, 24(2), e51-82, xiii-xiv.
D'Eon, T.M., Sharoff, C., Chipkin, S.R., Grow, D., Ruby, B.C. & Braun, B.
(2002). Regulation of exercise carbohydrate metabolism by estrogen and progesterone in women. Am J Physiol Endocrinol Metab,
283(5), E1046-1055.
Dooley, M.M. & Brincat, M.P. (Hrsg.). (1994). Understanding Common
Disoders in Reproductive Endocrinology. Chichester, England:
John Wiley & Sons Ltd.
Enea, C., Boisseau, N., Fargeas-Gluck, M.A., Diaz, V. & Dugue, B. (2011).
Circulating androgens in women: exercise-induced changes. Sports
Med, 41(1), 1-15.
Erkkola, R. (2007). Recent advances in hormonal contraception. Curr Opin
Obstet Gynecol, 19(6), 547-553.
Graham, C.A., Bancroft, J., Doll, H.A., Greco, T. & Tanner, A. (2007).
Does oral contraceptive-induced reduction in free testosterone ad
71
REFERENCES
versely
affect
the
sexuality
or
mood
of
women?
Psychoneuroendocrinology, 32(3), 246-255.
Gravelle, B.L. & Blessing, D.L. (2000). Physiological adaptation in women
concurrently training for strength and endurance. . The Journal of
Strength and Conditioning Research., 14(1), 5-13.
Greydanus, D.E., Patel, D.R. & Rimsza, M.E. (2001). Contraception in the
adolescent: an update. Pediatrics, 107(3), 562-573.
Han, A., Sung, E., Hinrichs, T., Vorgerd, M., Manchado, C. & Platen, P.
(2012). Effects of follicular versus luteal phase-based endurance
training in untrained women. Med Sci Sports Exerc (Submitted)
Hansen, A.K., Fischer, C.P., Plomgaard, P., Andersen, J.L., Saltin, B. &
Pedersen, B.K. (2005). Skeletal muscle adaptation: training twice
every second day vs. training once daily. J Appl Physiol, 98(1), 9399.
Isacco, L., Duche, P. & Boisseau, N. (2012). Influence of hormonal status
on substrate utilization at rest and during exercise in the female
population. Sports Med, 42(4), 327-342.
Jabbour, H.N., Kelly, R.W., Fraser, H.M. & Critchley, H.O. (2006). Endocrine regulation of menstruation. Endocr Rev, 27(1), 17-46.
Janse de Jonge, X.A. (2003). Effects of the menstrual cycle on exercise
performance. Sports Med, 33(11), 833-851.
Kelly, G. (2006). Body temperature variability (Part 1): a review of the history of body temperature and its variability due to site selection, biological rhythms, fitness, and aging. Altern Med Rev, 11(4), 278-293.
Kiley, J. & C., H. (2007). Combinde oral contraceprives: A comprehensive
Review. Clinical Obstetrics and Gynecology, 50(4), 868-877.
Kimball, S.R. (2006). Interaction between the AMP-activated protein kinase and mTOR signaling pathways. Med Sci Sports Exerc, 38(11),
1958-1964.
LaGuardia, K.D., Shangold, G., Fisher, A., Friedman, A. & Kafrissen, M.
(2003). Efficacy, safety and cycle control of five oral contraceptive
regimens containing norgestimate and ethinyl estradiol. Contraception, 67(6), 431-437.
72
REFERENCES
Lebrun, C.M. (1994). The effect of the phase of the menstrual cycle and
the birth control pill on athletic performance. Clin Sports Med, 13(2),
419-441.
Longcope, C. (1986). Adrenal and gonadal androgen secretion in normal
females. Clin Endocrinol Metab, 15(2), 213-228.
Lopez, L.M., Grimes, D.A. & Schulz, K.F. (2012). Steroidal contraceptives:
effect on carbohydrate metabolism in women without diabetes mellitus. Cochrane Database Syst Rev(4), CD006133.
Martinson, H. & Stokes, M.J. (1991). Measurement of anterior tibial muscle
size using real-time ultrasound imaging. Eur J Appl Physiol Occup
Physiol, 63(3-4), 250-254.
McGraw, K.O. & Wong, S.P.(1996). Forming inferences about some intraclass correlation coefficients. Psychological Methods, 1(1), 30-46.
Oosthuyse, T. & Bosch, A.N. (2010). The effect of the menstrual cycle on
exercise metabolism: implications for exercise performance in
eumenorrhoeic women. Sports Med, 40(3), 207-227.
Oosthuyse, T., Bosch, A.N. & Jackson, S. (2005). Cycling time trial performance during different phases of the menstrual cycle. Eur J Appl
Physiol, 94(3), 268-276.
Owen, J.A., Jr. (1975). Physiology of the menstrual cycle. Am J Clin Nutr,
28(4), 333-338.
Pitteloud, N., Mootha, V.K., Dwyer, A.A., Hardin, M., Lee, H., Eriksson,
K.F., et al. (2005). Relationship between testosterone levels, insulin
sensitivity, and mitochondrial function in men. Diabetes Care, 28(7),
1636-1642.
Rechichi, C., Dawson, B. & Goodman, C. (2009). Athletic performance and
the oral contraceptive. Int J Sports Physiol Perform, 4(2), 151-162.
Reilly, T. (2000). The menstrual cycle and human performance: An overview. Biological Rhythm Research, 31(1), 29-40.
Reimer, C.D., Gaulrapp, H., Kelle, H. (Hrsg.) (2004). Sonographie der
Muskeln, Sehnen und Nerven. : Deutscher Ärzte-Verlag, Köln.
Rickenlund, A., Carlstrom, K., Ekblom, B., Brismar, T.B., Von Schoultz, B.
& Hirschberg, A.L. (2004). Effects of oral contraceptives on body
composition and physical performance in female athletes. J Clin
Endocrinol Metab, 89(9), 4364-4370.
73
REFERENCES
Scott, W., Stevens, J. & Binder-Macleod, S.A. (2001). Human skeletal
muscle fiber type classifications. Phys Ther, 81(11), 1810-1816.
Sitruk-Ware, R. (2006). New progestagens for contraceptive use. Hum
Reprod Update, 12(2), 169-178.
Sitruk-Ware, R. & Nath, A. (2011). Metabolic effects of contraceptive steroids. Rev Endocr Metab Disord. , 12(2), 63-75.
Sung, E., Han, A., Hinrichs, T., Vorgerd, M., Manchado, C. & Platen, P.
(2012). Effects of follicular versus luteal phase-based strength training in untrained women. Med Sci Sports Exerc. (Submitted)
Tegtbur, U., Busse, M.W. & Kubis, H.P. (2009). [Exercise and cellular adaptation of muscle]. Unfallchirurg, 112(4), 365-372.
Vaiksaar, S., Jurimae, J., Maestu, J., Purge, P., Kalytka, S., Shakhlina, L.,
et al. (2011). No Effect of Menstrual Cycle Phase and Oral Contraceptive Use on Endurance Performance in Rowers. J Strength
Cond Res.
Van Look, P.F. & Baird, D.T. (1980). Regulatory mechanisms during the
menstrual cycle. Eur J Obstet Gynecol Reprod Biol, 11(2), 121-144.
Wiegratz, I., Jung-Hoffmann, C. & Kuhl, H. (1995). Effect of two oral contraceptives
containing
ethinylestradiol
and
gestodene
or
norgestimate upon androgen parameters and serum binding proteins. Contraception, 51(6), 341-346.
Wiik, A., Gustafsson, T., Esbjornsson, M., Johansson, O., Ekman, M.,
Sundberg, C.J., et al. (2005). Expression of oestrogen receptor alpha and beta is higher in skeletal muscle of highly endurancetrained than of moderately active men. Acta Physiol Scand, 184(2),
105-112.
Winder, W.W., Taylor, E.B. & Thomson, D.M. (2006). Role of AMPactivated protein kinase in the molecular adaptation to endurance
exercise. Med Sci Sports Exerc, 38(11), 1945-1949.
Wojtys, E.M., Huston, L.J., Boynton, M.D., Spindler, K.P. & Lindenfeld, T.N.
(2002). The effect of the menstrual cycle on anterior cruciate ligament injuries in women as determined by hormone levels. Am J
Sports Med, 30(2), 182-188.
Yan, Z. (2000). Skeletal muscle adaptation and cell cycle regulation. Exerc
Sport Sci Rev, 28(1), 24-26.
74