Terms and Conditions for Use of PDF The provision of PDFs for authors' personal use is subject to the following Terms & Conditions: The PDF provided is protected by copyright. All rights not specifically granted in these Terms & Conditions are expressly reserved. Printing and storage is for scholarly research and educational and personal use. Any copyright or other notices or disclaimers must not be removed, obscured or modified. The PDF may not be posted on an open-access website (including personal and university sites). The PDF may be used as follows: to make copies of the article for your own personal use, including for your own classroom teaching use (this includes posting on a closed website for exclusive use by course students); to make copies and distribute copies (including through e-mail) of the article to research colleagues, for the personal use by such colleagues (but not commercially or systematically, e.g. via an e-mail list or list serve); to present the article at a meeting or conference and to distribute copies of such paper or article to the delegates attending the meeting; to include the article in full or in part in a thesis or dissertation (provided that this is not to be published commercially). This material is the copyright of the original publisher. Unauthorised copying and distribution is prohibited.
Transcript
Terms and Conditions for Use of PDF
The provision of PDFs for authors' personal use is subject to the following Terms & Conditions:
The PDF provided is protected by copyright. All rights not specifically granted in these Terms & Conditions are expressly reserved. Printing and storage is for scholarly research and educational and personal use. Any copyright or other notices or disclaimers must not be removed, obscured or modified. The PDF may not be posted on an open-access website (including personal and university sites).
The PDF may be used as follows:• to make copies of the article for your own personal use, including for your own classroom teaching use (this includes posting on a closed website for exclusive use by course students); • to make copies and distribute copies (including through e-mail) of the article to research colleagues, for the personal use by such colleagues (but not commercially or systematically, e.g. via an e-mail list or list serve); • to present the article at a meeting or conference and to distribute copies of such paper or article to the delegates attending the meeting; • to include the article in full or in part in a thesis or dissertation (provided that this is not to be published commercially).
This material is the copyright of the original publisher.Unauthorised copying and distribution is prohibited.
This material is
the copyright of the
original publisher.
Unauthorised copying
and distribution
is prohibited.
The Effect of the Menstrual Cycleon Exercise MetabolismImplications for Exercise Performance in Eumenorrhoeic Women
Tanja Oosthuyse1 and Andrew N. Bosch2
1 School of Physiology, University of the Witwatersrand Medical School, Johannesburg, South Africa
2 UCT/MRC Research Unit for Exercise Science and Sports Medicine, University of Cape Town,
Abstract The female hormones, oestrogen and progesterone, fluctuate predictablyacross the menstrual cycle in naturally cycling eumenorrhoeic women. Otherthan reproductive function, these hormones influence many other physiolo-gical systems, and their action during exercise may have implications forexercise performance. Although a number of studies have found exerciseperformance – and in particular, endurance performance – to vary betweenmenstrual phases, there is an equal number of such studies reporting no dif-ferences. However, a comparison of the increase in the oestrogen concen-tration (E) relative to progesterone concentration (P) as the E/P ratio (pmol/nmol) in the luteal phase in these studies reveals that endurance perfor-mance may only be improved in the mid-luteal phase compared with the early
REVIEW ARTICLESports Med 2010; 40 (3): 207-227
0112-1642/10/0003-0207/$49.95/0
ª 2010 Adis Data Information BV. All rights reserved.
This material is
the copyright of the
original publisher.
Unauthorised copying
and distribution
is prohibited.
follicular phase when the E/P ratio is high in the mid-luteal phase. Further-more, the late follicular phase, characterized by the pre-ovulatory surge inoestrogen and suppressed progesterone concentrations, tends to promote im-proved performance in a cycling time trial and future studies should includethis menstrual phase. Menstrual phase variations in endurance performancemay largely be a consequence of changes to exercise metabolism stimulatedby the fluctuations in ovarian hormone concentrations. The literature sug-gests that oestrogen may promote endurance performance by altering car-bohydrate, fat and protein metabolism, with progesterone often appearing toact antagonistically. Details of the ovarian hormone influences on the meta-bolism of these macronutrients are no longer only limited to evidence fromanimal research and indirect calorimetry but have been verified by substratekinetics determined with stable tracer methodology in eumenorrhoeic wo-men. This review thoroughly examines the metabolic perturbations inducedby the ovarian hormones and, by detailed comparison, proposes reasons formany of the inconsistent reports in menstrual phase comparative research.Often the magnitude of increase in the ovarian hormones between menstrualphases and the E/P ratio appear to be important factors determining an effecton metabolism. However, energy demand and nutritional status may beconfounding variables, particularly in carbohydrate metabolism. The reviewspecifically considers how changes in metabolic responses due to the ovarianhormones may influence exercise performance. For example, oestrogen pro-motes glucose availability and uptake into type I muscle fibres providing thefuel of choice during short duration exercise; an action that can be inhibitedby progesterone. A high oestrogen concentration in the luteal phase aug-ments muscle glycogen storage capacity compared with the low oestrogenenvironment of the early follicular phase. However, following a carbo-loading diet will super-compensate muscle glycogen stores in the early folli-cular phase to values attained in the luteal phase. Oestrogen concentrationsof the luteal phase reduce reliance on muscle glycogen during exercise andalthough not as yet supported by human tracer studies, oestrogen increasesfree fatty acid availability and oxidative capacity in exercise, favouringendurance performance. Evidence of oestrogen’s stimulation of 50-AMP-activated protein kinase may explain many of the metabolic actions of oestro-gen. However, both oestrogen and progesterone suppress gluconeogenicoutput during exercise and this may compromise performance in the latterstages of ultra-long events if energy replacement supplements are inadequate.Moreover, supplementing energy intake during exercise with protein may bemore relevant when progesterone concentration is elevated compared withmenstrual phases favouring a higher relative oestrogen concentration, asprogesterone promotes protein catabolism while oestrogen suppresses pro-tein catabolism. Furthermore, prospective research ideas for furthering theunderstanding of the impact of the menstrual cycle on metabolism and ex-ercise performance are highlighted.
In many fields of physiology, sex is consideredto be a variable that should be ‘controlled for’.Therefore, men and women are expected to re-spond differently to various interventions or con-
ditions. Often, sample groups are restricted toincluding only men, possibly because male phy-siology remains relatively consistent from dayto day. Conversely, women between the ages of
208 Oosthuyse & Bosch
ª 2010 Adis Data Information BV. All rights reserved. Sports Med 2010; 40 (3)
This material is
the copyright of the
original publisher.
Unauthorised copying
and distribution
is prohibited.
approximately 13 and 50 years experience a circa-mensal rhythm termed the menstrual cycle, wherethe ovarian hormones fluctuate predictably over,on average, 23–38 days.[1] The ovarian hormones,oestrogen and progesterone, are secreted from theovaries and to a lesser extent from the adrenalglands in women.[2] Although these hormonesprimarily function to support reproduction, theyhave been reported to influence other physiologi-cal systems. For this reason, studies have beenconducted to compare established responses inmen with the response in women. However, inthese studies women are studied mostly only dur-ing the early stages of their menstrual cycle whenthe ovarian hormones are considered to be attheir lowest, so as to avoid the ‘moving target’scenario.
Women, however, function and compete insporting events at all stages of the menstrualcycle. Therefore, some researchers have en-
deavoured to compare physiological responsesin women between identified phases of themenstrual cycle, corresponding to accepted con-centration ranges for the ovarian hormones(figure 1). The menstrual cycle is broadly dividedinto two phases – the follicular phase (FP) andthe luteal phase (LP) – which are separated byovulation. The system involved in the regulationthereof is termed the hypothalamic-pituitary-ovarian axis, and is thoroughly reviewed byReilly[1] and Birch.[4]
Unfortunately, this field of research is plaguedwith many inconsistent findings, but it is possiblethat most inconsistencies can be solved by acloser examination of the hormone interactions.
Part one of this review considers studies thathave compared exercise performance betweenmenstrual phases. Part two presents a detailedreview of the effects of the ovarian hormones (asthey naturally occur during the various menstrual
Day 1(onset of menstruation)
Day 14ovulation
FSHOestrogenProgesterone LH
Day 28
MFEF LF EL ML LL
Fig. 1. Diagrammatic representation of the cyclical changes in the female sex hormones that characterize the various menstrual phases.17b-Oestradiol is the primary oestrogen secreted, but may be metabolized further to form oestrone and oestriol, which are less potentoestrogens.[2] The luteinizing hormone (LH) surge commences 36 hours before ovulation occurs.[3] EF = early follicular; EL = early luteal;FSH = follicle-stimulating hormone; LF = late follicular; LL = late luteal; MF = mid follicular; ML = mid luteal.
Menstrual Cycle, Metabolism and Performance 209
ª 2010 Adis Data Information BV. All rights reserved. Sports Med 2010; 40 (3)
This material is
the copyright of the
original publisher.
Unauthorised copying
and distribution
is prohibited.
phases) on substrate metabolism during exerciseat submaximal intensities. Although the ovarianhormones are known to influence other physio-logical systems (such as the respiratory, thermo-regulatory, and cardiovascular systems, and evenmuscle satellite cell activation), these fall outsidethe scope of the current review.
1. Effects of the Ovarian Hormones onExercise Performance
1.1 Short Duration or Maximal ExerciseIntensities
The maximum oxygen consumption (.VO2max)
and time to exhaustion in maximal ramp tests aremostly unchanged by menstrual phase.[5-14]
However, there is one report of a 2% lower.VO2max in the mid-luteal (ML) phase comparedwith the early follicular (EF) phase[14] and an-other in which a 13% decrease in
.VO2max after
4 months of oral contraceptive use was re-ported.[10] Conversely, altitude dwellers tended tohave a higher
.VO2max (p = 0.06) in the LP com-
pared with the FP,[15] possibly due to the in-creased respiratory drive in the LP, which mayfacilitate a slightly higher oxygen saturation.[9]
However, such findings are not found in womenwho are acutely exposed to altitude.[9] None-theless, it may be worthwhile to consider the po-tential for the progesterone-induced increase inrespiratory drive in the LP to benefit maximalexercise at high altitudes in well-acclimatizedathletic eumenorrhoeic women.
Conversely, others have reported that thehigher respiratory drive in the LP jeopardizesmaximal exercise performance in non-athletesdue to the increased sensation of dyspnoea,[12]
although the increased respiratory drive did notinfluence maximal performance in athletes.[12] Inthis regard, exercise-induced bronchoconstric-tion in asthmatic athletes is more severe duringthe ML phase compared with the mid-follicular(MF) phase following an incremental ramp testto exhaustion.[16] However, a consistently higherrespiratory rate throughout 90 minutes of sub-maximal exercise in theML phase compared withthe EF phase has been found to not increase
metabolic demand, and therefore should not in-fluence rate of fatigue.[17]
It appears fairly consistent from curve-fittingmethods that the exercise intensity that inducesthe point of inflection corresponding to either thelactate or ventilatory thresholds remains un-changed by menstrual phase.[7,11-13,18] However,one study has found that the ventilatory thresh-old occurs at a higher percentage of
.VO2max in the
EF phase compared with the late follicular (LF)and ML phases.[6] Moreover, Forsyth et al.[18]
found that the intensity corresponding to 4mmol/Llactate threshold was higher in the LP than FP.Similarly, others,[5,19-21] but not all,[6-8,22-25] havereported lower blood lactate concentrations dur-ing exercise in the LP compared with FP, thus sug-gesting the potential for a decreased blood lactateaccumulation during exercise and hence, by im-plication, lower anaerobic glycolysis in the LP.
Performances in all-out sprints and in mea-sures of muscle strength have been found to bebest during menstruation.[26-28] However, othershave found no differences in a Wingate perfor-mance test between menstruation and LP[29] or in10-second sprints betweenMF andML phases.[30]
In summary, menstrual phase has been foundonly occasionally to influence maximal aerobic oranaerobic performances. However, various phy-siological changes (such as respiratory drive) as-sociated with the ovarian hormones, other thansimply alterations to metabolism, may influenceexercise at such high intensities.
1.2 Submaximal Exercise Intensities
1.2.1 Time to Exhaustion
Low dose oestrogen supplementation toovariectomized rats has been shown to improvetime to exhaustion in a prolonged submaximaltreadmill run by 20% compared with sham in-jected rats.[31] Endurance time continued to im-prove with increasing oestrogen dose, resulting inup to a 42% improvement when oestrogen wasincreased within physiological concentrationscompared with sham-injected controls,[31] and50% improvements with a supraphysiological doseof oestrogen.[31] These massive improvements inendurance capacity coincidedwith glycogen-sparing
210 Oosthuyse & Bosch
ª 2010 Adis Data Information BV. All rights reserved. Sports Med 2010; 40 (3)
in the red and white vastus muscle, myocardium andliver.[31]
However, time to exhaustion at submaximalexercise intensity is a measure of endurance ca-pacity rather than a direct measure of exerciseperformance,[32] although it can provide an in-dication of an athlete’s potential for enduranceevents. Such protocols do not have a high re-producibility, and studies have reported coeffi-cients of variation as high as 30% when usingthese tests[32] thus reducing the statistical powerof comparison between interventions, in this casebetween menstrual phases. Nonetheless, in hu-mans, two studies have reported an effect ofmenstrual phase on endurance capacity. The firststudy found that following 40 minutes of sub-maximal cycling at low to moderate intensities,time to exhaustion at 90% of maximum poweroutput was doubled in the ML phase comparedwith the MF phase.[5] This coincided with lowerblood lactate levels in the ML phase.[5] The sec-ond study had a smaller sample size (n = 6);possibly because of this, the (on average) 10%longer time to exhaustion at 70%
.VO2max in the
ML phase compared with the MF phase did notquite reach significance (p < 0.07).[23] Conversely,two further studies also compared time toexhaustion at 70%
.VO2max and did not find any
difference between the EF and ML phase,[9,33]
with or without carbohydrate supplements dur-ing exercise.[33]
While all these studies demonstrated a 2-foldor greater increase in oestrogen from the follicu-lar to luteal phase, the oestrogen to progesteroneconcentration ratio (E/P; pmol/nmol) differednoticeably. Studies reporting a better perfor-mance in the LP had a higher E/P ratio, while thestudies that found no change in endurance timehad a lower E/P ratio (table I). This observationimplies that the higher relative progesteroneconcentration in the latter studies impeded themetabolic benefits of oestrogen that may havebeen more prominent during the LP of the formerstudies.
However, 6 days of transdermal oestrogensupplementation in amenorrhoeic women failedto alter time to exhaustion at 85%
.VO2max that
was preceded by 90 minutes of submaximalrunning.[36] In this study though, the transdermaloestrogen supplement resulted in only modestincreases in circulating oestrogen, to levels typi-cally experienced in the early to mid-FP. Further-more, while the duration of oestrogen exposurewas sufficient to lower glucose kinetics, it mayhave been too short to produce certain otheroestrogen effects, such as muscle glycogen-sparing during exercise.[37]
1.2.2 Time Trial Performance
Exercise protocols with a fixed endpoint –
such as time to complete a given distance, or toexpend a given amount of energy or distance
Table I. Relative changes in the ovarian hormones between the follicular (FP) and luteal phase (LP) in relation to submaximal
endurance performance
Magnitude of increase in
oestrogen in LP above FP
E/P in LP Result Reference
Time to exhaustion at submaximal intensity
2.28-fold 12.3 NS; EF vs ML 33
2.87-fold 8 NS; EF vs ML 9
2-fold 21.3 p < 0.02; ML > MF 5
3.85-fold 18 p < 0.07; ML > MF 23
Time trial performance
2.3-fold 5.5 p < 0.05 MF faster than ML without CHO supplement 22
2.5-fold 6 NS; MF vs ML with CHO supplement 22
1.4-fold 9.7 NS; MF vs ML following a normal or CHO-loading diet 34
4-fold 18.5 NS; EF vs ML but tendency for LF faster than EF (p = 0.027) 35
CHO = carbohydrate; EF = early follicular; E/P = oestrogen to progesterone ratio; LF = late follicular phase; MF = mid-follicular phase;
ML = mid-luteal phase; NS = not significant.
Menstrual Cycle, Metabolism and Performance 211
ª 2010 Adis Data Information BV. All rights reserved. Sports Med 2010; 40 (3)
covered in a fixed time period etc. – are a goodmeasure of exercise performance, having high-test-retest reproducibility as described by a lowcoefficient of variability (1–3%).[32,38]
Three studies have measured time trial per-formance between menstrual phases.[22,34,35]
Campbell et al.[22] compared the time to expend agiven amount of energy after completing a 2-hoursubmaximal session at 70%
.VO2max in the MF
and ML phase with and without carbohydratesupplements in overnight fasted subjects. Theyobserved a 13% improvement in time trial per-formance in the MF phase without carbohydratesupplementation during exercise.[22] This betterMF performance was associated with highercarbohydrate use and whole body rate of glucoseappearance (hepatic glucose production) and rateof disappearance (or glucose uptake), suggestinga better capacity for carbohydrate use in the MFphase.[22]
An increased capacity for carbohydrate utili-zation is beneficial in short duration time trialevents that take place at high intensities. Thisobservation of better time trials in the MFphase[22] coincides well with another study fromthe same authors who found oestrogen to pro-mote contraction-stimulated glucose uptake andhepatic glycogenolysis during exercise in ovar-iectomized rats during a short, high intensityrun, while progesterone antagonized these res-ponses.[39] The pre-exercise MF phase averageoestrogen concentration was relatively high(360 pmol/L)[22] and hence oestrogen may havepromoted glucose uptake into muscles duringthese trials. In addition, despite a 2.5-fold in-crease in oestrogen in the ML phase over the MFphase, the E/P ratio in the ML phase was com-paratively low (5–6) (table I).[22] Thus, the rela-tively high progesterone concentration during thetrials in the ML phase may have countered thebenefits of an elevated oestrogen concentrationand produced a worse performance. However,the use of carbohydrate supplements during ex-ercise elevated the glucose rate of appearance,disappearance and plasma glucose use, providingsufficient fuel-of-choice to promote an optimalperformance in a short duration, high intensitytime trial, regardless of menstrual phase.[22]
McLay et al.[34] also compared cycling timetrial performance (over 16 km) in the MF andML phase after a lengthy submaximal exerciseperiod (75 minutes). These authors found nodifference in finishing time between menstrualphases when subjects participated following3 days of either a normal mixed diet or a carbo-hydrate-loading diet. Unfortunately, the authorsallowed the subjects to view their power outputthroughout the time trial, and this may have beena potential shortfall as subjects would have beenable to consciously regulate their exercise in-tensity to match their previous time trial. More-over, the group average oestrogen concentrationincreased by a meagre 1.4-fold more from theMFto ML phase, while the average progesteroneconcentration in the ML phase was substantial,resulting in an E/P ratio of only 9.7 (table I),[34]
thus possibly partly explaining the lack of differ-ence in performances.
A study was also performed in our laboratoryto assess cycling time trial performance duringthe EF, LF and ML phase of the menstrualcycle.[35] The inclusion of the LF phase in thiscomparison is a novel contribution to the litera-ture and is motivated by the many oestrogen-induced metabolic effects (see section 2) thatshould promote performance in such an event.In our subjects who participated in a non-fastedstate, time trial performance was also not sig-nificantly different between the EF and MLphase.[35] However, we observed a strong ten-dency for better performance in the LF phasecompared with the EF phase (p = 0.027), but thisdid not quite reach significance with Bonferronicorrection applied for three multiple compar-isons.[35] Nevertheless, the results suggest a posi-tive influence of oestrogen on performance insuch events. Conversely, the coincident increaseof progesterone in the ML phase may have an-tagonized the benefits of an elevated oestrogenconcentration despite a high E/P ratio (18.5) inthe ML phase and average 4-fold increase inoestrogen. However, subjects in this study[35] andthe study byMcLay et al.[34] participated roughly2 hours postprandially, which may have alle-viated the metabolic demand that is thoughtnecessary to potentiate ovarian hormone influences,
212 Oosthuyse & Bosch
ª 2010 Adis Data Information BV. All rights reserved. Sports Med 2010; 40 (3)
particularly regarding carbohydrate metabolism.Nonetheless, evidence of oestrogen’s capacity topromote cycling time trial performance withoutprogesterone antagonizm was still evident in theLF phase in the study by Oosthuyse et al.,[35] des-pite subjects exercising 2 hours postprandially.
In summary, only eight studies in total haveconsidered menstrual phase variations in en-durance exercise performance. Of the eightstudies, four have reported menstrual cyclevariations. Thus, the potential for naturallycycling ovarian hormones to alter performancecan neither be excluded nor confirmed. However,strong evidence from animal research, specifi-cally for oestrogen-induced promotion of betterendurance capacity, and the various menstrualphase-associated metabolic perturbations dis-cussed in part two of this review should providemotivation for further endurance performancestudies. Future studies should consider the in-crease in oestrogen relative to progesterone in theLP and the absolute magnitude of increase inoestrogen between any two menstrual phases.Furthermore, all exercise trials to date have beenlimited to <2 hours in duration. The metabolicinfluences of oestrogen in promoting fat use andspare glycogen stores should best support per-formances in ultra-endurance events. Thus,future studies should investigate menstrual phasevariations in ultra-endurance events and considerincluding the LF phase, which coincides withmaximal increases in oestrogen independent ofchanges in progesterone.
2. Ovarian Hormones and Metabolism
Both oestrogen and progesterone are reportedto alter metabolic responses. However, in thisrespect, progesterone displays largely anti-oestrogenic effects.[39-41] D’Eon et al.[42] have pro-posed that a metabolic response to changes in theovarian hormones occurs only when the E/P ratiois sufficiently elevated and the magnitude of theincrease in oestrogen from the EF to the phase ofcomparison such as LF or ML is at least in theorder of 2-fold more. Nutritional status is alsoa determining variable, since most variations inmetabolism betweenmenstrual phases are reported
when subjects participate in a study following anovernight fast, whereas a positive nutritional statemay lessen the impact of the ovarian hormones.[22]
2.1 Carbohydrate Metabolism
2.1.1 Stable Isotopic Measures of Systemic GlucoseKinetics in Eumenorrhoeic Women
In a fasted state, glucose rate of appearance(Ra) is solely determined by endogenous glucoseproduction, which is predominantly controlledby hepatic gluconeogenesis and glycogenolysis.Glucose rate of disappearance (Rd) is depen-dent on insulin-mediated glucose uptake andcontraction-mediated glucose transport, with thelatter predominating during exercise. GlucoseRa and Rd are naturally related to each otherand primarily influenced by the rate of glucoseutilization.[25]
A number of studies have found that the Raand Rd of glucose during exercise is attenuatedby either therapeutic increases in circulating oes-trogen[36,37,42,43] or with the coincident rise inoestrogen and progesterone during the ML phaseof the menstrual cycle versus the EFphase.[19,22,44] Therefore, the ovarian hormone-induced decrease in glucose kinetics is most likelyan oestrogen-associated effect and is one thatprogesterone does not antagonize but may in factpotentiate.[42]
Glucose Ra was dependent on hepatic glucoseproduction in all of the above studies, as thesubjects did not receive any form of exogenousglucose during exercise and participated follow-ing an overnight fast (besides D’Eon et al.,[42]
where differences in glucose kinetics did not quitereach significance). Therefore, the ovarian hor-mone-induced reduction in glucose kinetics notedin the above studies is supposedly due to theability of oestrogen to hamper hepatic gluconeo-genesis.[45] This hypothesis is supported by astudy performed in carbohydrate-depleted wo-men, where blood glucose was maintained duringsubmaximal exercise in the MF phase but con-centrations dropped progressively during exercisein the ML phase.[20]
However, Horton et al.[25] hypothesized thatthe effect of oestrogen on hepatic glucose output
Menstrual Cycle, Metabolism and Performance 213
ª 2010 Adis Data Information BV. All rights reserved. Sports Med 2010; 40 (3)
appears to become noticeable only when the ex-ercise intensity is sufficient to increase the de-mands on glucose utilization to above a certain‘critical level’. At this ‘critical level’ the demandon endogenous glucose production is sufficientlyelevated such that the effect of oestrogen sup-pression on gluconeogenesis is evident in a re-duced glucose Ra. As illustrated by Hortonet al.,[25] this is clearly supported by comparingthe glucose kinetic results of Campbell et al.,[22]
Horton et al.[25] and Zderic et al.[19] (table II).When the subjects in the study by Zderic
et al.[19] exercised at approximately 50%.VO2max,
glucose Ra was significantly lower in the MLphase versus the EF phase. However, no differ-ence was noted between menstrual phases whenthese subjects exercised at only 42%
.VO2max.
[19]
Thus, when there was less of a demand on en-dogenous glucose production, and an increasedreliance on lipid utilization, no menstrual phaseeffect was evident in glucose kinetics. In the study
by Horton et al.,[25] the subjects were slightly lesstrained than those in the study by Zderic et al.[19]
Therefore, when the subject groups of the twostudies exercised at an equal intensity of 50%.VO2max, the absolute workload was lower in thestudy by Horton et al.[25] Thus, the lower abso-lute workload demanded a lower total fuel utili-zation and hence lower glucose utilization withless demand on endogenous glucose productionand consequently a lower glucose Ra. At thislower glucose Ra no difference was observedbetween the EF, MF and ML phases,[25] as wassimilarly observed in the study by Zderic et al.[19]
at 42%.VO2max. In contrast, well-trained subjects
in the study by Campbell et al.[22] exercised at70%
.VO2max. This higher intensity exercise in-
creased the demand on endogenous glucose pro-duction above the ‘critical level’ and produced anoticeable difference in glucose Ra between theEF and ML phase. A study by Devries et al.[44]
that included both eumenorrhoeic women and
Table II. Comparison of menstrual phase studies investigating plasma glucose kinetics during exercise with consideration of nutritional status
and the absolute exercise intensity
Reference (year) Menstrual phases Training status.VO2max
(mL/kg/min)
Exercise
intensity
(%.VO2max)
Absolute.VO2
(mL/kg/min)
Glucose Raa
(mmol/kg/min)
Significance
Overnight fasted studies
Horton et al.[25] (2002) EF vs MF vs ML 39.9 – 5.8 50 20.2 20 vs 20 vs 18 NS
Zderic et al.[19] (2001) EF vs ML 48.2 – 1.1 42
52
20.2
25.1
20 vs 18
33.7 vs 28.8
NS
p < 0.05
Campbell et al.[22] (2001) EF vs ML 53.5 – 0.9 69 36.8 33 vs 25 p < 0.05
Devries et al.[44] (2006) MF vs ML (part OC) 39 – 2 65 25.4 53 vs 49 p = 0.03
Ruby et al.[36] (1997) Am; PL vs 72 h E vs
144 h E
45.5 – 5.6 65 29.2 21.9 vs 18.9 vs
18.9
p < 0.05
Carter et al.[43] (2001) M; PL vs E 53.3 – 6.7 60 31.1 48 vs 42 p < 0.05
Devries et al.[37] (2005) M; PL vs E 44 – 2 65 28.6 65 vs 60 p = 0.04
Roughly 3-hours postprandial studies
D’Eon et al.[42] (2002) GnRH agonist
vs E vs E + P
42.5 – 8 54 23 (96 watts) 51 vs 45.6 vs
43.3
(0.05 < p < 0.1)
for E and
E + P < GnRHa
Suh et al.[24] (2002) EF vs ML 43.6 – 2 45
65
20 (59 watts)
29 (97 watts)
27.7 vs 28.3
38.9 vs 40.6
NS
NS
Exogenous glucose ingestion during exercise
Campbell et al.[22] (2001) EF vs ML 53.5 – 0.9 70 37.8 46 vs 43 NS
a Some glucose Ra values are only approximations as estimated from figures.
Am = amenorrhoeic women; E = exogenous oestrogen supplements; E + P = exogenous oestrogen and progesterone supplements; EF = early
follicular phase; GnRHa = gonadotropic-releasing hormone agonist that will suppress endogenous oestrogen and progesterone secretion;
M = male subjects; MF = mid-follicular phase; ML = mid-luteal phase; NS = not significantly different; OC = oral contraceptive; PL = placebo;
Ra = rate of appearance;.VO2 = oxygen uptake;
.VO2max = maximal oxygen uptake.
214 Oosthuyse & Bosch
ª 2010 Adis Data Information BV. All rights reserved. Sports Med 2010; 40 (3)
women on oral contraceptives confirms thesefindings, with glucose Ra 6% lower in the LPcompared with the FP when subjects exercised at65%
.VO2max following an overnight fast.
However, when subjects in the study byCampbell et al.[22] received an energy drinkduring exercise the difference in glucose Rabetween menstrual phases was no longer signif-icant, possibly because glucose Ra was now lar-gely determined by exogenous glucose absorption(table II).[25] This theory can be extended to in-clude subjects who exercise only a few hourspostprandially (but do not receive glucose sup-plements during exercise).[24,42] Under these con-ditions it appears that the demands on glucoseutilization need to exceed an even higher criticallevel before a difference in glucose kinetics be-tween menstrual phases becomes evident(table II). It would be interesting to challengethis hypothesis with a study in which glucosekinetics parameters are measured in subjects ex-ercising at high intensities of >70%
.VO2max fol-
lowing a short postprandial period, in variousmenstrual phases.
Thus, in summary, glucose kinetics appears tobe influenced by menstrual phase when theenergy demand of exercise is sufficiently highto pressurise endogenous glucose production.However, it appears that the postprandial periodis a major determinant of the level of the demandon endogenous glucose production that is neces-sary before the influence of menstrual phasebecomes evident. This can be explained as oes-trogen or oestrogen and progesterone impose arestriction on endogenous glucose production bysuppressing gluconeogenesis.[45] However, whenexercise follows a short postprandial period itwould be expected that hepatic glycogenolysiswould provide a greater proportional contribu-tion to endogenous glucose production than fromgluconeogenesis relative to a condition that im-poses a longer fasting period. Finally, when exo-genous glucose is provided throughout exercisethe influence of menstrual phase on glucosekinetics is negligible, as this condition minimizesthe demand on endogenous glucose produc-tion.[22] However, future studies should considerthe influence of menstrual phase or ovarian hor-
mone concentration on glucose kinetics duringultra-long events even with exogenous glucosesupplements as, ultimately, gluconeogenic outputwill become increasingly relevant.
2.1.2 Indirect Estimation of Muscle Glycogen Use
Glucose metabolic tracer studies often makethe assumption that 100% of glucose uptake (Rd)is oxidized and therefore glucose Rd approx-imates plasma glucose oxidation rate. The dif-ference between total carbohydrate oxidationestimated by indirect calorimetry and glucose Rdprovides an estimate of muscle glycogen useduring exercise. However, this is a crude as-sumption, as the percentage of glucose Rd oxi-dized is probably closer to 60–90% and may varydepending on the study conditions; thus, the cal-culation underestimates glycogen use and shouldbe considered as minimal muscle glycogen utili-zation.[46] In fact, when such indirect estimates ofmuscle glycogen use were compared with actualmuscle biopsy measures, the values did not cor-relate.[44] Furthermore, while muscle biopsy datarevealed menstrual phase differences betweenmuscle glycogen use during exercise, no differ-ences were evident when based on indirect tracerestimates in the same sample group.[44]
Possibly, for these reasons, some studies mea-suring glucose kinetics have not estimated muscleglycogen use,[24,25,43] which may explain whyothers have found no difference in estimatedglycogen use between menstrual phases[22] orwith oestrogen treatment.[24] Therefore, the re-sults from indirect muscle glycogen estimationsmust be considered in light of the possible limit-ations of the method.
However, some studies that reported a lowerglucose uptake (Rd) with elevated oestrogen[42]
or oestrogen and progesterone[19] also reportedlower estimated muscle glycogen use during ex-ercise under these conditions compared with EFphase conditions. Interestingly, D’Eon et al.[42]
found that pharmacologically elevated oestrogenplus progesterone resulted in greater estimatedmuscle glycogen use during exercise comparedwith a condition of suppressed ovarian hor-mones. Such a finding is contrary to reports fromother authors whose muscle biopsy data suggest
Menstrual Cycle, Metabolism and Performance 215
ª 2010 Adis Data Information BV. All rights reserved. Sports Med 2010; 40 (3)
muscle glycogen sparing in the LP, in which oes-trogen and progesterone concentrations are ele-vated.[47] However, the rise in ovarian hormonesin the study by D’Eon et al.[42] was not natural,and although during the oestrogen plus proges-terone condition oestrogen was elevated to withinnormal LP levels, progesterone increased toaround 151.4 nmol/L (47.6 ng/mL), which ishigher than normal LP levels. Thus, the findingsof D’Eon et al.[8] suggest that progesterone mayantagonize glycogen sparing. The muscle glyco-gen sparing that has been reported to occur in theLP[19,47] must be largely due to the elevated oes-trogen levels that occur during this phase andcould possibly be more pronounced during theLF phase where oestrogen alone is elevated.
Furthermore, D’Eon et al.[42] described an in-teresting inverse correlation in which free fattyacid (FFA) concentration explained 50% of thevariance in estimated muscle glycogen use, whereFFA concentration was greater with oestrogensupplementation. This possibly infers that anoestrogen-induced increased FFA availability pro-moted glycogen sparing during exercise.[42]
Therefore, the influence of oestrogen and pro-gesterone on muscle glycogen utilization maydepend on their influence on FFA availability oroxidation.
Estimation of muscle glycogen content bymuscle biopsy in eumenorrhoeic women suggeststhat the hormone milieu in the LP promotesmuscle glycogen storage[23,34,48] compared withthe FP. Of particular note, given the currentinterest in multistage events, Nicklas et al.[23]
reported greater muscle glycogen repletion fol-lowing a period of induced glycogen depletionin the LP compared with the FP. However, acarbohydrate-loading diet increased muscle gly-cogen stores in the FP to the higher values pre-viously attained in the LP when following anormal diet,[34] but the carbohydrate-loading dietfailed to increase muscle glycogen stores furtherin the LP.[34] Thus, carbohydrate loading bal-ances the capacity for muscle glycogen storagebetween the FP and LP.[34]
However, a recent study by Devries et al.[44]
used a subject group comprising of part eu-menorrhoeic women and part women usingtriphasic oral contraceptives and found no dif-ference in resting muscle glycogen stores betweenthe follicular and LP. Moreover, oestrogen sup-plementation in men resulted in a trend for lowerresting muscle glycogen with a moderate oestro-gen dose,[49] or significantly lower resting glyco-gen stores with a higher oestrogen dose comparedwith placebo treatment.[37] However, the oestro-gen exposure period in these latter studies mayhave been too short to promote glycogen storage,or oestrogen supplementation may have de-creased the calorie intake[50] of the men whencompared with placebo treatment, because al-though their diet was self-controlled it was notrigorously regulated.[37] Nonetheless, in supportof the latter findings, rat studies using male[51] orovariectomized female rats[31,39] where the quan-tity of food-intake was controlled, reported nochange in resting muscle glycogen stores follow-ing oestrogen and/or progesterone treatment.However, one such study did report an increase inliver glycogen stores with oestrogen supplementsin ovariectomized rats compared with progester-one, combined progesterone and oestrogen, orplacebo treatment,[39] thus demonstrating thepotential for oestrogen to maximize glycogenstores. However, we cannot exclude the possibi-lity of interspecies differences in carbohydratemetabolism.
Nevertheless, oestrogen has been shown toincrease muscle glycogen synthase activity.[52]
Moreover, oestrogen deficiency is associated withinsulin resistance[53] and higher insulin concen-trations.[42] Furthermore, intravenous oestrogenin postmenopausal women promoted insulinactions by increasing insulin-stimulated glucoseuptake at rest during a hyperinsulinaemic clamp.[54]
Other studies have also reported insulin sensitiv-ity to be heightened in the presence of oestro-gen[53,55] but often report no change in insulinresponsiveness to a large glucose load.[53,55-57]
Thus, we would expect increases in oestrogenconcentration to promote glycogen storage.Moreover, the similar carbohydrate-loaded gly-cogen stores between menstrual phases could be
216 Oosthuyse & Bosch
ª 2010 Adis Data Information BV. All rights reserved. Sports Med 2010; 40 (3)
explained by the previously reported lack of anoestrogen effect on insulin responsiveness to alarger glucose load.
Conversely, increases in progesterone con-centration are associated with a decrease in theinsulin-responsive glucose transporter (GLUT4)content in insulin-sensitive tissue[39] and insulinresistance;[58-60] hence, progesterone most likelycounters glycogen storage.
Given the controversy in the findings of rest-ing muscle glycogen stores between studies fromnaturally cycling eumenorrhoeic women andoestrogen-supplemented studies, further studiesare necessary to clarify whether the ovarian hor-mones alter resting muscle glycogen stores.
A number of studies have reported lower rates ofglycogen use during exercise in the LP comparedwith the FP based on biopsy samples taken be-fore and after 60 minutes of exercise at 70%.VO2max.
[44,47] In addition, one such study foundthat muscle glycogen use during exercise was ne-gatively correlated with oestrogen concentra-tion.[47] More specifically, Devries et al.[44] iso-lated muscle glycogen into proglycogen andmacroglycogen fractions and found women in theFP used 30% more proglycogen and 16% moremacroglycogen and together 24% more totalmuscle glycogen during exercise than when intheir LP. However, their sample group includedboth eumenorrhoeic women and women usingoral contraceptives.
Previously, the same group of researchersfound no evidence of muscle glycogen sparingduring endurance exercise in men receivingoestrogen supplements compared with placebotreatment.[37,49] However, the authors’ speculatethat the period of oestrogen treatment or sexdifferences in oestrogen receptor density mayexplain the lack of effect in the men.[44]
Animal research consistently finds that oes-trogen treatment results in skeletal muscle gly-cogen sparing during exercise.[31,39,51] However,despite the greater pre-exercise liver glycogenin oestrogen-treated ovariectomized rats in thestudy by Campbell and Febbraio,[39] after30 minutes of submaximal running at 0.35m/sec,liver glycogen stores were no longer significantlydifferent between ovarian hormone treatments.
These results suggest a greater rate of hepaticglycogenolysis with oestrogen treatment. This iscontrary to the finding of Kendrick et al.[31]
where after 2 hours of submaximal running atthe same intensity (0.37m/sec) marked hepaticglycogen sparing was reported with oestrogentreatment in oophorectomized rats. The dis-crepancy in liver glycogen metabolism with oes-trogen treatment may be related to the differencein exercise duration of the two studies. There-fore, the greater hepatic glycogen use with oes-trogen treatment in the study by Campbell andFebbraio[39] may reflect initial responses to exercisethat are characterized by an early-phase higherdependence on carbohydrate metabolism. Thus,the findings from this study reflect the capacity ofoestrogen to increase glucose availability and,moreover, uptake into specifically type I musclefibres during periods of demand.[39] In fact, twoanimal studies have found oestrogen to potenti-ate contraction-stimulated glucose uptake (50%increases have been reported with oestrogen re-placement relative to oestrogen deficiency byovariectomy).[39,55] Conversely, the findings ofKendrick et al.[31] depict metabolic preferencesof prolonged exercise where the presence of in-creased oestrogen may promote a different re-sponse that includes liver glycogen sparing,which, in that study, coincided with substantialincreases in endurance capacity.
2.1.4 Conclusion of the Influence of the OvarianHormones on Carbohydrate Metabolism
In summary, evidence from metabolic studiessuggests that oestrogen and progesterone havevarious effects on carbohydrate metabolism.Oestrogen promotes insulin sensitivity and pos-sibly greater glycogen storage, while progester-one promotes insulin resistance. Oestrogenpromotes contraction-stimulated glucose uptakeinto type I muscle fibres during short durationexercise, which should be beneficial for perfor-mance in higher intensity aerobic exercise, whileprogesterone antagonizes this action.[39] Thus,the increase in oestrogen relative to progesteronemay determine the influence of the ovarian hor-mones during the LP on insulin-stimulated andcontraction-stimulated glucose uptake and so
Menstrual Cycle, Metabolism and Performance 217
ª 2010 Adis Data Information BV. All rights reserved. Sports Med 2010; 40 (3)
variably influence glycogen storage and plasmaglucose availability during exercise. However,whole body systemic glucose kinetics is reducedduring prolonged exercise, with increases in oes-trogen alone or in combination with progester-one. Such a decrease in whole body glucosekinetics is possibly due to suppression of hepaticgluconeogenic production when the exercise in-tensity is sufficiently intense to put pressure onthe system. Gluconeogenic suppression may jeo-pardize exercise performance when glycogenstores are limited. However, muscle glycogenstores are spared during exercise in the LP or inrats with oestrogen supplementation and shouldpromote better endurance.
Although researchers investigating menstrualphase comparisons have extensively studied carbo-hydrate metabolism during exercise, isotopictracer measures of plasma glucose oxidation rateare still lacking.
2.2 Fat Metabolism
While occasionally indirect calorimetry mea-surements suggest greater whole body lipid useduring exercise in the LF or ML phase comparedwith EF or MF phase,[11,19,22,61] this is notconsistently reported.[24,25,44,62,63] However,when oestrogen supplements are administeredindependently of progesterone, the respiratoryexchange ratio (RER) is lower during exercisecompared with placebo in men[37] or comparedwith oestrogen and progesterone supplementstogether or gonadotropin-releasing hormone(GnRH) agonists (suppressed ovarian secretion)in women.[42]
2.2.1 Systemic Glycerol Kinetics as a Measureof Lipolytic Rate
Determination of glycerol Ra by tracer meth-odology is routinely used as an index of wholebody lipolytic rate.[64] This is based on the as-sumption that following triacylglycerol hydro-lysis in muscle and adipose tissue, glycerol mustbe released into the blood. Glycerol must be re-phosphorylated by the enzyme glycerol kinase,present only in the liver and to a lesser degree inthe kidneys before it can be reused in triacylgly-
cerol re-esterification. Therefore, it is assumedthat hepatic clearance of glycerol from the bloodis the only significant route of irreversible loss ofa glycerol tracer.[64] However, this assumptionhas been challenged, as some authors[65] havefound that only half of the glycerol Ra is takenup by the splanchnic bed and therefore the peri-phery must be taking up the rest. Secondly, thefindings of others[66] suggest that muscle maymetabolize a significant amount of glycerol andtherefore not all of the glycerol released by in-tramuscular triacylglycerol hydrolysis will appearin the bloodstream.
Nonetheless, glycerol kinetics have been com-pared at rest and during submaximal exercise ineumenorrhoeic women in their EF andML phaseand then after 4 months of oral contraceptivesupplementation.[67] No significant differencewas found between menstrual phases in a sub-sample (n = 5), but oral contraceptive use in-creased glycerol Ra during submaximal exercise(n = 8). Oral contraceptive use also resulted inhigher cortisol concentration, which is pre-sumably causative of the heightened lipolyticrate.[67] A further study has examined glycerolkinetics during moderate intensity exercise ineumenorrhoeic women and included the MFphase –which is associated withmoderate increasesin oestrogen independent of progesterone – intheir comparison.[62] These authors also found nosignificant variation in glycerol kinetics betweenmenstrual phases. However, the average increasein oestrogen in the MF and ML phases wasmodest (265 and 393 pmol/L, respectively).
Furthermore, exogenous oestrogen supple-ments given to amenorrhoeic women[36] ormen[43] also failed to alter glycerol kinetics duringsubmaximal exercise. However, the increase inoestrogen in the amenorrhoeic women was mod-est, approximating only FP levels. Conversely,the oral supplements in the men increased oes-trogen to supraphysiological levels, and whethersex differences in lipolytic regulation[68,69] mayhave obscured an oestrogen effect on glycerolkinetics is indeterminate. In addition, the possi-bility of a sex difference in the response to oestro-gen treatment cannot be excluded, as the receptorpopulation of the endogenous sex hormones are
218 Oosthuyse & Bosch
ª 2010 Adis Data Information BV. All rights reserved. Sports Med 2010; 40 (3)
reportedly different between sexes.[70] None-theless, the limited evidence to date suggeststhat oestrogen and the ovarian milieu of theLP do not alter whole body systemic glycerolkinetics and, by inference, lipolytic rate duringexercise.
Oestrogen’s stimulation and progesterone’santagonizm of growth hormone response toexercise,[71] however, suggest oestrogen may stim-ulate lipolysis, but consideration of the E/P ratioin the LP would be prudent. Animal researchpresents convincing evidence to suggest thatoestrogen can in fact increase lipolytic rate duringexercise. For example, Beniot et al.[72] reporteda heightened sensitivity to catecholamines inoestrogen-supplemented rats, with a correspond-ing increase in hormone-sensitive lipase activity.These authors suggest that oestrogen acts via itscatechol-oestrogen derivative to potentiate thelipolytic action of adrenaline (epinephrine) bycompeting with catecholamines for catechol-O-methyltransferase.[72] In addition, Hansenet al.[73] demonstrated an increase in lipolysis andreduced fatty acid synthesis in isolated fatcells from oestrogen-treated rats, while proges-terone had no effect compared with control/unsupplemented rats. Oestrogen supplementa-tion in male rats has also been found to alter lipo-protein lipase (LPL) activity in a tissue-specificfashion.[74] While adipocyte LPL activity was re-duced, muscle LPL activity was increased, pro-moting a redistribution of lipids from adipose tomuscle tissue. Consequently, oestrogen supplemen-tation not only elevated resting intramuscularlipid content but also promoted triacylglycerolesterification during submaximal exercise in thered vastus muscle, as triacylglycerol content waseven greater post-exercise than at rest.[74] There-fore, whole body glycerol kinetics would not beable to elucidate oestrogen’s tissue-specific actionbut instead presents the overall summated re-sponse. However, it must be considered that in-terspecies differences may occur in the regulationof lipid metabolism.[75]
Encouraged by the overwhelming evidencefrom animal studies, future studies should con-sider tissue-specific glycerol kinetics using meth-ods such as arteriovenous balance during exercise
in various menstrual phases including the LFphase, occurring approximately 2 days beforeovulation and in which oestrogen peaks.
FFA Ra provides an index of plasma FFAavailability and measures the release of fattyacids that are primarily derived from the hy-drolysis of adipose tissue triacylglycerol intoplasma.[69] When used as a measure of lipolyticresponse, FFA Ra does not account for triacyl-glycerol re-esterification.[76] FFA Rd measures therate of uptake into tissues and has been used as anestimate of plasma FFA oxidation rate;[77] how-ever, this is a crude estimate as the actual propor-tion of FFA uptake that is oxidized can vary andhas been reported to be as low as 50%.[78]
A number of studies have considered plasmaFFA kinetics,[79-81] dietary FFA uptake intobody stores[82] and plasma triacylglycerolkinetics[81] between menstrual phases,[79,81,82] andwith and without oestrogen supplements inpostmenopausal women[80] at rest. All studiesreported no differences between menstrual pha-ses or treatments. In fact, a similar FFA Ra atrest between menstrual phases is not surprising,as animal studies confirm that basal lipolysis isunchanged or even suppressed in the presenceof oestrogen compared with oestrogen defi-ciency.[52,83] Conversely, oestrogen enhancescatecholamine sensitivity as is noted by an up-regulated lipolytic response to catecholaminestimulation with oestrogen treatment.[72,83]
More recently, plasma FFA kinetics and oxi-dation have been compared during submaximalexercise during various menstrual phases in eu-menorrhoeic women.[62,63] Jacobs et al.[63] per-formed a longitudinal study comparing plasmaFFA metabolic response in the EF and MLphases and then with subsequent oral contra-ceptive use. Unfortunately, their menstrual phasecomparison was reduced to a sample size of five,which limited the statistical power of the com-parison. Considering this limitation they re-ported no significant differences in the rates ofwhole body fat oxidation, plasma FFA oxida-tion, non-plasma FFA oxidation or plasma FFA
Menstrual Cycle, Metabolism and Performance 219
ª 2010 Adis Data Information BV. All rights reserved. Sports Med 2010; 40 (3)
rate of appearance or disappearance or rate ofre-esterification.[63] The average oestrogen con-centration in the ML phase was a modest311 pmol/L and the E/P ratio was fairly low, at 9.Furthermore, they failed to make use of theacetate correction factor in their calculation ofplasma FFA oxidation, which is now establishedas necessary for more accurate estimates ofplasma FFA oxidation rate.[84]
The acetate correction factor accounts for theproportion of tracer-derived carbon label that isretained in the products of secondary exchangereactions that occur with tricarboxylic acid cycleintermediates instead of being released as carbondioxide.[84] Our laboratory has observed sig-nificant variability in the acetate correction fac-tor between menstrual phases.[85] The correctionfactor was lower in the ML phase compared withthe EF phase.[85] We speculate that this may beassociated with increased protein catabolismduring exercise in the ML phase, as reported byothers.[86] That is, the increased flux throughtransamination pathways may result in a slightlygreater proportion of FFA tracer-derived carbonlabel isotope being retained in the products ofsubsidiary reactions with tricarboxylic acid cycleintermediates. Thus, in order to further increasethe sensitivity of the comparison of plasma FFAoxidation rate between menstrual phases, itwould be necessary to simultaneously derive theacetate correction factor and plasma FFA oxi-dation for each menstrual phase.
Shortly following the study by Jacobs et al.,[63]
a second study by Horton et al.[62] consideredFFA kinetics during moderate exercise in the EFversus MF versus ML phases. The MF phase ischaracterized by a gradual increase in oestrogenconcentration independently of progesterone.The authors found no variation in FFARa or Rdbetween menstrual phases. However, the averageoestrogen concentration recorded in the MFphase of the study by Horton et al.[62] was mod-erate (264 pmol/L), and even the ML phase oes-trogen value (393 pmol/L) was fairly modest forthis menstrual phase, resulting in a low E/P ratioof 10.7. These authors, as with Jacobs et al.,[63]
agreed that themagnitude of increase in oestrogenand the oestrogen increase relative to progester-
one may be an important factor determining theimpact of the ovarian hormones on fat metabo-lism. Horton et al.[62] went on to presume thatvariations may be noticeable in the LF or perio-vulatory period when oestrogen is elevated in-dependently of progesterone.
Such speculations are based on compellingevidence from animal studies. Ovariectomy re-duces the activity of key enzymes in fat metabo-lism, i.e. carnitine palmitoyl transferase-I (CPT-I)and beta-3-hydroxyacyl-CoA dehydrogenase(b–HAD).[40] Oestrogen restores the activity ofthese enzymes, while progesterone inhibits thesepositive actions when oestrogen is at physiologi-cal concentrations.[40] However, a supraphysio-logical concentration of oestrogen overrides thenegative effects of progesterone.[40] Interestingly,the difference in b-HAD activity with oestrogentreatment was only evident in muscle sectionscomposed primarily of type I fibres[40] and not insections of predominantly type II fibres.[40,52]
Nonetheless, this rat model demonstrates theability of the ovarian hormones to alter the capa-city for skeletal muscle to oxidize FFAs by directlyimpacting on the cellular metabolic pathways.
In an initial pilot study performed in our la-boratory, we measured plasma palmitate Ra andRd with a continuous infusion of 1-13C palmitateduring moderate intensity exercise over 90 min-utes in eumenorrhoeic women (n = 5) who were3-hours postabsorptive (Oosthuyse and Bosch,unpublished observations). The women all com-pleted the trial twice, once in the EF phase andthen again in either the LF (or periovulatory)phase or ML phase or late luteal (LL) phase. Theintention was to obtain the full range of ovarianhormones and E/P ratios that may occur during amenstrual cycle. We found that plasma palmitateRa was highly correlated with E/P ratio (r = 0.85;p = 0.06) and was significant between plasmapalmitate Rd and E/P ratio (r = 0.89; p= 0.04).A trend for a relationship between palmitate Ra,Rd and oestrogen concentration was evident.However, due to the limited sample size, no de-finite conclusions can be drawn. Nonetheless, thispilot study provides motivation for further in-vestigations of exercising FFAmetabolism of thiskind that consider oestrogen and progesterone
220 Oosthuyse & Bosch
ª 2010 Adis Data Information BV. All rights reserved. Sports Med 2010; 40 (3)
variability between subjects and from one day tothe next even within a menstrual phase.
In summary, a first glance at the latest workcovering plasma FFA kinetics and oxidationduring exercise suggests no menstrual phase orovarian hormone effect. All researchers in thefield agree that considering interindividual andintraindividual variability in ovarian hormonesand the relative ratio of E/P as they change evenwithin a menstrual phase is imperative in con-vincingly identifying whether the fluctuations inoestrogen and progesterone, as occurs naturallyacross the menstrual cycle and specifically the LFphase, alter lipid metabolism during prolongedexercise.
2.2.3 Intramyocellular Stores and Use DuringExercise
No study to date has considered the influenceof the menstrual phase on intramyocellular lipid(IMCL) stores and use during exercise. Althoughan indirect estimation of IMCL use in exercisecan be attained from isotopic tracer methods(by calculating the difference between wholebody lipid oxidation and plasma FFA oxidation),it represents a rough estimate because it cannotdifferentiate IMCL oxidation from plasma triacyl-glycerol oxidation.[63]
A number of recent studies have investigatedvariability in IMCL stores and use during ex-ercise between men and women.[87-94] A full dis-cussion of the findings of these sex-comparativestudies is beyond the scope of the current review.However, researchers should carefully considerthese findings when designing studies to considervariations in IMCL stores and use during exercisebetween menstrual phases. In particular, suchfuture studies should consider menstrual varia-tions in IMCL use during ultra-long enduranceexercise, as Devries et al.[89] reported a higherpercentage association between IMCL dropletsand mitochondria following 90 minutes of ex-ercise in women compared with men, where nosex differences were found before exercise.Ovarian hormone effects on IMCL use may be-come more apparent during the latter stages ofultra-events.
2.2.4 50-AMP-Activated Protein Kinase, a KeyRegulator of Cellular Metabolism, is Influenced byOestrogen
Recent advances in the metabolic regulatoryactions of 50-AMP-activated protein kinase(AMPK) suggest that it is the major cellular energyregulator driving metabolic processes to promoteATP production.[95,96] Increased AMPK activitycorresponds with increased GLUT4 content,contraction-stimulated glucose uptake and in-creased cellular fatty acid uptake, although evi-dence for a dominant role over the regulationof fat oxidation is not yet conclusive.[95] Evidenceof oestrogen affecting AMPK activity[83] drawstogether the many previously suspected, but of-ten elusive, actions of oestrogen. Specifically,AMPK is known to increase translocation offatty acid translocase (FAT/CD36) and plasmamembrane-bound fatty acid-binding protein(FABPpm).[96] FABPpm is abundant and there-fore not limiting FFA uptake, but an increase inFAT/CD36 will increase FFA transport andoxidation and thus may limit FFA uptake. Sinceoestrogen is thought to stimulate AMPK activ-ity,[83] it could be speculated that oestrogen willincrease FFA uptake when sufficiently elevated ineumenorrhoeic women. Although AMPK’s role inenhancing fat oxidation has come under recentscrutiny,[97] AMPK is thought to inhibit the en-zyme glycerol-3-phosphate acyltransferase, whichinitiates triacylglycerol synthesis, and acetyl-CoAcarboxylase (ACC), which regulates the produc-tion of malonyl-CoA.[96] Malonyl-CoA is wellknown as a potent inhibitor of CPT-I and thusentry of long chain fatty acids (LCFAs) into mito-chondria. Furthermore, AMPK is thought toincrease the activity of malonyl-CoA decarboxy-lase, which breaks down malonyl-CoA to acetyl-CoA.[96] Thus, AMPK should promote entry ofLCFA into beta oxidation.
The study by D’Eon et al.[83] demonstrates therole of oestrogen in regulating fat metabolism viagenomic and non-genomic pathways. Oestrogenpromotes leanness and decreases adipocyte sizeby decreasing fatty acid uptake into adipose tis-sue (via decreased expression of lipoprotein lipase[LPL]), decreasing lipogenesis (via decreased ex-pression of ACC-1 and fatty acid synthase [FAS])
Menstrual Cycle, Metabolism and Performance 221
ª 2010 Adis Data Information BV. All rights reserved. Sports Med 2010; 40 (3)
and increasing catecholamine stimulated lipolysis,but not basal lipolysis (via increased expressionof the phosphoprotein, perilipin).[83] Oestrogenalters gene expression by binding to the oestrogenreceptor response elements located in the pro-moters of target genes and thereby regulates liverX receptor a (LXRa) and sterol regulatoryelement-binding protein 1c (SREBP-1c), whichregulate the transcriptional expression of ACC-1and FAS.[83]
However, the genomic influence of oestrogenin muscle and liver differ from that of adipo-cytes.[83] In muscle and liver, oestrogen up-regulates the transcription factor peroxisomeproliferation activator receptor-d (PPAR-d),which leads to the increased expression of variousenzymes (LPL, pyruvate dehydrogenase kinase,acyl-CoA oxidase, and uncoupling protein 2 and3), which promotes energy dissipation and theoxidation of FFA.[83]
Moreover, oestrogen-treated ovariectomizedmice displayed 5-fold increased AMPK activityin skeletal muscle compared with placebo-treatedmice.[83] TheAMPK response to oestrogen occurredrapidly in a time- and dose-dependent manner viabinding to membrane-bound oestrogen receptorsand thus stimulating increased fat uptake intomitochondria.[83] Interestingly, recent findingshave shown that the expression of oestrogen re-ceptors a and b in skeletal muscle increases withthe level of endurance training,[98] and hence it isinteresting to question whether a greater oestrogen-stimulated AMPK response could be expected inskeletal muscles of endurance-trained athletes.However, oestrogen’s stimulation of AMPK isnot aided by adipokines, as the study by D’Eonet al.[83] found leptin and adiponectin concentrationto be lower in oestrogen-treated mice, possiblydue to the smaller adipocyte size in the oestrogen-treated group and oestrogen’s inhibition ofSREBP-1c. Other studies have also found oestro-gen to decrease adiponectin concentration inpregnancy.[99] However, one study showed largefluctuations in adiponectin in women acrossthe menstrual cycle, but no association with theovarian hormone profile was evident.[100]
D’Eon et al.[83] suggest that the ‘oestrogen ef-fect’ is largely a result of oestrogen’s manipula-
tion of the following targets: SREBP-1c, PPAR-dand AMPK. In fact, although oestrogen may actindependently on SREBP-1c and PPAR-d, cel-lular and transcriptional regulation afforded byAMPK[96] mimics many of the observations re-ported in the oestrogen-treated mice in the studyby D’Eon et al.[83] Thus, it is tempting to proposethat oestrogen’s activation of AMPK is the majorkey to the metabolic perturbations of oestrogen.Further work in this area is suggested. Studies ineumenorrhoeic women should consider variationsin the AMPK response to exercise across the men-strual cycle and to test for associations betweenoestrogen and/or progesterone concentration andchanges in the AMPK response to exercise.
2.2.5 Conclusion of the Influence of the OvarianHormones on Fat Metabolism
Animal research presents strong evidence thatoestrogen promotes lipolysis and increases plas-ma FFA availability during exercise, increasesintramuscular lipid stores and increases cellularcapacity for FFA oxidation. Recent evidencesuggests that many of oestrogen’s metabolic ac-tions may occur through AMPK stimulation andactivation of transcription factors. Such optimi-zation of lipid metabolism with oestrogen wouldpromote an ideal metabolic response for en-durance exercise. However, isotopic tracer studiesin resting or exercising eumenorrhoeic womenhave reported no differences in systemic glycerolor FFA kinetics. Future lipid metabolic studiesshould consider the magnitude of increase inoestrogen between menstrual phases and the in-crease in oestrogen relative to progesterone dur-ing the LP. Furthermore, studies focusing ontissue-specific metabolism in women may help toexplain the divergent findings of animal and hu-man research.
2.3 Influence of Ovarian Hormoneson Protein Metabolism
Recent tracer studies have consistently foundamino acid oxidation to be greater in the LP com-pared with the FP at rest.[101-103] Kriengsinyoset al.[101] observed the strongest correlation be-tween measurements of phenylalanine oxidationand progesterone, suggesting that progesterone
222 Oosthuyse & Bosch
ª 2010 Adis Data Information BV. All rights reserved. Sports Med 2010; 40 (3)
This material is
the copyright of the
original publisher.
Unauthorised copying
and distribution
is prohibited.
has the greatest impact on amino acid catabolismin the LP. Furthermore, the greater amino acidoxidation in the LP coincided with greater dietarylysine requirement in the LP compared with theFP.[101] Reports on amino acid flux are less con-sistent. One study has reported greater flux in theLP compared with the FP at rest,[102] while othersreported no differences.[101,103] Some have alsofound non-oxidative leucine disposal (NOLD) tobe lower in the LP, suggesting less proteinsynthesis.[103] In a study in which ovarian hor-mone secretion was suppressed by administrationof a gonadotropin releasing hormone agonist, itwas found that leucine turnover and NOLD wereattenuated compared with normally cycling wo-men.[103] Thus, the ovarian hormones supportmaintenance of normal protein turnover in wo-men, and as catabolism varies across menstrualphase, protein requirement varies coincidently.
Evidence of greater protein catabolism duringexercise in the LP is provided by earlier studies.Bailey et al.,[33] found the concentration of variousamino acids (i.e. alanine, glutamine, proline andisoleucine) to be lower in theML phase comparedwith the EF phase at rest and during prolongedsubmaximal exercise in subjects exercising3 hours postabsorptively. This suggests greateramino acid catabolism during exercise in the LP.However, the amino acid concentration differ-ence between phases was smaller if a carbo-hydrate supplement was ingested during exercisecompared with a placebo drink.[33] Lamontet al.,[86] also reported that protein catabolismwas greater during the ML phase compared withthe EF phase as measured by total urea nitrogenexcretion over 4 days, including a 60-minuteperiod of exercise at 70%
.VO2max. Total urinary
urea nitrogen excretion over the exercise periodwas also greater in the ML phase compared withthe EF phase.[86]
In contrast to the seemingly catabolic stimu-lation of protein metabolism in the LP, recentevidence suggests a positive influence of oestrogenin reducing protein oxidation.[104] For example,oral oestrogen supplementation in men reducedleucine oxidation by 16% at rest and duringexercise compared with placebo treatment andso accounted for an increase in the protein bal-
ance (calculated as the difference between totalprotein synthesis and breakdown) by 8mg ofprotein/kg/h at rest and 17mg of protein/kg/hduring exercise.[104]
A recent study examined the rate of myofi-brillar and connective tissue protein synthesisfollowing one-legged kicking exercise in twogroups of eumenorrhoeic women.[105] One groupparticipated in the FP and the other in the LP. Nodifferences between phases were reported, al-though this study may have been limited by thestatistical unpaired design.[105] Further studiesregarding protein or amino acid kinetics andoxidation during exercise across the menstrualcycle are warranted.
In summary, the ovarian hormones have anoticeable influence on protein metabolism atrest and during exercise, which is often seen asincreased catabolism in the LP. It appears thatprogesterone is responsible for the consistentfinding of increased protein catabolism in theLP,[101] while oestrogen may reduce protein cat-abolism.[104] It would be interesting to investigatewhether the E/P ratio in the LP is important indetermining the extent of protein catabolismin this menstrual phase. Furthermore, studies ineumenorrhoeic women in the late follicular orpre-ovulatory phase to verify oestrogen’s capa-city to reduce protein oxidation would be valu-able. Such studies conducted during exercise arenecessary to determine the protein requirementsof female athletes participating in endurancecompetition.
3. Conclusions
The potential of the ovarian hormones toimpose major metabolic aberrations to carbo-hydrate, lipid and protein metabolism has beenreported from both human and animal studies,and suggests repercussions for exercise perfor-mance in eumenorrhoeic women (as detailed inthe summary following each subsection). How-ever, the influences of the ovarian hormones onthe various metabolic pathways appear highlycomplex and often tissue specific. Furthermore,oestrogen and progesterone have mostly opposinginfluences on the various systems, and responses
Menstrual Cycle, Metabolism and Performance 223
ª 2010 Adis Data Information BV. All rights reserved. Sports Med 2010; 40 (3)
may be dependent on the concentrations of therespective ovarian hormones. Moreover, the ex-tent of metabolic demand will also determinewhether the ovarian hormone influences arephysiologically significant. Consideration of thesedetails is important for appreciating the effect onperformance during different types of exercise.
Consequently, menstrual phase-associatedchanges to the various metabolic measurementsare not consistently identified. By inference, thismay explain the inconsistency in the reports ofmenstrual phase-associated changes to exerciseperformance. The findings of menstrual phasestudies may be confounded by the high variabilityin the concentrations of ovarian hormones be-tween subjects and from day to day within sub-jects during any particular menstrual phase.For this reason, investigating relations betweenmetabolic and exercise performance parametersand the change in the ovarian hormone con-centrations between the menstrual phases and/orthe E/P ratio should increase the sensitivity ofstudies for identifying metabolic or performancechanges caused by the naturally cycling ovarianhormones.
Acknowledgements
Studies cited in this review as unpublished findings weresupported by grants from the University of theWitwatersrandResearch Council, Medical Research Council of South Africaand the National Research Foundation. The authors have noconflicts of interest that are directly relevant to the content ofthis review.
References1. Reilly T. The menstrual cycle and human performance: an
overview. Biol Rhythm Res 2000; 31: 29-40
2. Lebrun CM. The effect of the phase of the menstrual cycleand the birth control pill on athletic performance. ClinSports Med 1994; 13: 419-41
3. Buffenstein R, Poppott SD, McDevitt RM et al. Food in-take and the menstrual cycle: a retrospective analysis, withimplications for appetite research. Physiol Behav 1995;58: 1067-77
4. Birch K. Circamensal rhythms in physical performance.Biol Rhythm Res 2000; 31: 1-14
5. Jurkowski JEH, Jones NL, Toews CJ, et al. Effects ofmenstrual cycle on blood lactate, O2 delivery, andperformance during exercise. J Appl Physiol 1981; 51:1439-99
6. Bemben DA, Salm PC, Salm AJ. Ventilatory and bloodlactate responses to maximal treadmill exercise duringthe menstrual cycle. J Sports Med Phys Fitness 1995; 35:257-62
7. Dean TM, Perreault L, Mazzeo RS, et al. No effect ofmenstrual cycle phase on lactate threshold. J Appl Physiol2003; 95: 2537-43
8. De Souza MJ, Maguire MS, Rubin K. Effects of menstrualphase and amenorrhea on exercise responses in runners.Med Sci Sports Exerc 1990; 22: 575-80
9. Beidleman BA, Rock PB, Muza SR, et al. Exercise VE andphysical performance at altitude are not affected bymenstrual cycle phase. J Appl Physiol 1999; 86: 1519-26
10. Casazza GA, Suh S-H, Miller BF, et al. Effects of oralcontraceptives on peak exercise capacity. J Appl Physiol2002; 93: 1698-702
11. Dombovy ML, Bonekat HW, Williams TJ. Exercise per-formance and ventilatory response in the menstrual cycle.Med Sci Sport Exerc 1987; 19: 111-7
12. Schoene RB, Robertson HT, Pierson DJ, et al. Respiratorydrives and exercise in menstrual cycles of athletic andnonathletic women. J Appl Physiol 1981; 50: 1300-5
13. Smekal G, Von Duvillard SP, Frigo P, et al. Menstrualcycle: no effect on exercise cardiorespiratory variables orblood lactate concentration. Med Sci Sports Exerc 2007;39 (7): 1098-106
14. Lebrun CM, McKenzie DC, Prior JC, et al. Effects ofmenstrual cycle phase on athletic performance. Med SciSports Exerc 1995; 27: 437-44
15. Brutsaert TD, Spielvogel H, Caceres E, et al. Effect ofmenstrual cycle phase on exercise performance of high-altitude native women. J Exp Biol 2002; 202: 233-9
16. Stanford KI, Mickleborough TD, Ray S, et al. Influence ofmenstrual cycle phase on pulmonary function in asth-matic athletes. Eur J Appl Physiol 96: 703-10
17. Oosthuyse T, Bosch AN. Influence of menstrual phase onventilatory responses to submaximal exercise. S Afr JSports Med 2006; 18: 31-7
18. Forsyth JJ, Reilly T. The combined effect of time of dayand menstrual cycle on lactate threshold. Med Sci SportsExerc 2005; 37 (12): 2046-53
19. Zderic TW, Coggan AR, Ruby BC. Glucose kinetics andsubstrate oxidation during exercise in the follicular andluteal phases. J Appl Physiol 2001; 90: 447-53
20. Lavoie J-M, Dionne N, Helie R, et al. Menstrual cyclephase dissociation of blood glucose homeostasis duringexercise. J Appl Physiol 1987; 62: 1084-9
21. McCracken M, Ainsworth B, Hackney AC. Effects of themenstrual cycle phase on the blood lactate response toexercise. Eur J Appl Physiol 1994; 69: 174-5
22. Campbell SE, Angus DJ, Febbraio MA. Glucose kineticsand exercise performance during phases of the menstrualcycle: effect of glucose ingestion. Am J Physiol 2001; 281:E817-25
23. Nicklas BJ, Hackney AC, Sharp RL. The menstrual cycleand exercise: performance, muscle glycogen, and sub-strate responses. Int J Sports Med 1989; 10: 264-9
24. Suh S-H, Casazza GA, Horning MA, et al. Effects of oralcontraceptives on glucose flux and substrate oxidation
224 Oosthuyse & Bosch
ª 2010 Adis Data Information BV. All rights reserved. Sports Med 2010; 40 (3)
rates during rest and exercise. J Appl Physiol 2002; 93:42-50
25. Horton TJ, Miller EK, Glueck D, et al. No effect of men-strual cycle phase on glucose kinetics and fuel oxidationduring moderate-intensity exercise. Am J Physiol 2002;282: E752-62
26. Brooks-Gunn J, Gargiulo JM, Warren MP. The effect ofcycle phase in the performance of adolescent swimmers.Physician Sportsmed 1986; 14: 182-4
27. Davies BN, Elford JC, Jamieson KF. Variations in per-formance in simple muscle tests at different phases of themenstrual cycle. J Sports Med Phys Fitness 1991; 31 (4):532-7
28. Redman LM, Weatherby RP. Measuring performanceduring the menstrual cycle: a model using oral contra-ceptives. Med Sci Sports Exerc 2004; 36 (1): 130-6
29. Bushman B, Masterson G, Nelsen J. Anaerobic powerperformance and the menstrual cycle: eumenorrheic andoral contraceptive users. J Sports Med Phys Fitness 2006;46 (1): 132-7
30. Middleton LE, Wenger HA. Effects of menstrual phase onperformance and recovery in intense intermittent activity.Eur J Appl Physiol 2006; 96: 53-8
31. Kendrick ZV, Steffen CA, Rumsey WL, et al. Effect ofestradiol on tissue glycogen metabolism in exercisedoophorectomized rats. J Appl Physiol 1987; 63: 492-6
32. Jeukendrup A, Saris WHM, Brouns F, et al. A new vali-dated endurance performance test. Med Sci Sports Exerc1996; 28 (2): 266-70
33. Bailey SP, Zacher CM,Mittleman KD. Effect of menstrualcycle phase on carbohydrate supplementation duringprolonged exercise to fatigue. J Appl Physiol 2000; 88:690-7
34. McLay RT, Thomson CD, Williams SM, et al. Carbo-hydrate loading and female endurance athletes: effect ofmenstrual-cycle phase. Int J Sport Nutr Exerc Metab2007; 17 (2): 189-205
35. Oosthuyse T, Bosch AN, Jackson S. Cycling time trialperformance during different phases of the menstrualcycle. Eur J Appl Physiol 2005; 94: 268-76
36. Ruby BC, Robergs RA, Waters DL, et al. Effects of es-tradiol on substrate turnover during exercise in amenor-rheic females. Med Sci Sports Exerc 1997; 29: 1160-9
37. Devries MC, HamadehMJ, Graham TE, et al. 17b-estradiolsupplementation decreases glucose rate of appearance anddisappearance with no effect on glycogen utilization dur-ing moderate intensity exercise in men. J Clin EndoMetab2005; 90: 6218-25
38. Palmer GS, Dennis SC, Noakes TD, et al. Assessmentof the reproducibility of performance testing on anair-braked cycle ergometer. Int J Sports Med 1996; 17:293-8
39. Campbell SE, Febbraio MA. Effect of the ovarian hor-mones on GLUT4 expression and contraction-stimulatedglucose uptake. Am J Physiol 2002; 282: E1139-46
40. Campbell SE, Febbraio MA. Effect of ovarian hormoneson mitochondrial enzyme activity in fat oxidation path-way of skeletal muscle. Am J Physiol 2001; 281: E803-8
41. Hatta H, Atomi Y, Shinohara S, et al. The effects ofovarian hormones on glucose and fatty acid oxidation
during exercise in female ovariectomized rats. Horm Me-tab Res 1988; 20: 609-11
42. D’Eon TM, Sharoff C, Chipkin SR, et al. Regulation ofexercise carbohydrate metabolism by estrogen and pro-gesterone in women. Am J Physiol 2002; 283: E1046-55
43. Carter SL, McKenzie S, Mourtzakis M, et al. Short-term17b-estradiol decreases glucose Ra but not whole bodymetabolism during endurance exercise. J Appl Physiol2001; 90: 139-46
44. Devries MC,Mazen JH, Phillips SM, et al. Menstrual cyclephase and sex influence muscle glycogen utilization andglucose turnover during moderate-intensity enduranceexercise. Am J Physiol 2006; 291: R1120-8
45. MatuteML, Kalkhoff RK. Sex steroid influence on hepaticgluconeogenesis and glycogen formation. Endocrinology1973; 92: 762-8
46. Friedlander AL, Casazza GA, Horning MA, et al. Train-ing-induced alterations of carbohydrate metabolism inwomen: women respond differently from men. J ApplPhysiol 1998; 85 (3): 1175-86
47. Hackney AC. Influence of oestrogen on muscle glycogenutilization during exercise. Acta Physiol Scand 1999; 167:273-4
48. Hackney AC. Effects of the menstrual cycle on restingmuscle glycogen content. Horm Metab Res 1990; 22: 647
49. Tarnopolsky MA, Roy BD, MacDonald JR, et al.Short-term 17-b-estradiol administration does not affectmetabolism in young males. Int J Sports Med 2001; 22:175-80
50. Shimomura K, Shimizu H, Tsuchiya T, et al. Is leptin a keyfactor which develops obesity by ovariectomy? Endocr J2002; 49 (4): 417-23
51. Rooney TP, Kendrick ZV, Carlson J, et al. Effect of es-tradiol on the temporal pattern of exercise-induced tissueglycogen depletion in male rats. J Appl Physiol 1993; 75:1502-6
52. Beckett T, Tchernof A, TothMJ. Effect of ovariectomy andestradiol replacement on skeletal muscle enzyme activity infemale rats. Metabolism 2002; 51 (11): 1397-401
53. Latour MG, Shinoda M, Lavoie J-M. Metabolic effects ofphysical training in ovariectomized and hyperestrogenicrats. J Appl Physiol 2001; 90: 235-41
54. Van Pelt RE, Gozansky WS, Schwartz RS, et al. Intra-venous estrogens increase insulin clearance and actionin postmenopausal women. Am J Physiol 2003; 285:E311-7
55. Hansen PA, McCarthy TJ, Pasia EN, et al. Effects ofovariectomy and exercise training on muscle GLUT-4content and glucose metabolism in rats. J Appl Physiol1996; 80: 1605-11
56. Cooper BC, Sites CK, Casson PR, et al. Ovarian suppres-sion with a gonadotropin-releasing hormone agonist doesnot alter insulin-stimulated glucose disposal. Fertil Steril2007; 87 (5): 1131-8
57. Toth MJ, Cooper BC, Partley RE, et al. Effect of ovariansuppression with gonadotropin-releasing hormone agoniston glucose disposal and insulin secretion. Am J Physiol2008; 294: E1035-45
58. Kalkhoff RK. Metabolic effects of progesterone. Am JObstet Gynecol 1982; 142: 735-8
Menstrual Cycle, Metabolism and Performance 225
ª 2010 Adis Data Information BV. All rights reserved. Sports Med 2010; 40 (3)
59. Elkind Hirsch KE, Sherman LD, Malinak R. Hormonereplacement therapy alters insulin sensitivity in youngwomen with premature ovarian failure. J Clin EndocrinolMetab 1993; 76: 472-5
60. Ezenwaka EC, Akanji AO, Adejuwon CA, et al. Insulinresponses following glucose administration in menstruat-ing women. Int J Gynaecol Obstet 1993; 42: 155-9
61. Hackney AC, Curley CS, Nicklas BJ. Physiological re-sponses to submaximal exercise at the mid-follicular,ovulatory and mid-luteal phases of the menstrual cycle.Scand J Med Sci Sports 1991; 1: 94-8
62. Horton TJ, Miller EK, Bourret K. No effect of menstrualcycle phase on glycerol or palmitate kinetics during 90minof moderate exercise. J Appl Physiol 2006; 100: 917-25
63. Jacobs KA, Casazza GA, Suh S-H, et al. Fatty acidreesterification but not oxidation is increased by oralcontraceptive use in women. J Appl Physiol 2005; 98:1720-31
64. Wolfe RR. Radioactive and stable isotope tracers in bio-medicine: principles and practice of kinetic analysis. NewYork: Wiley-Liss, Inc., 1992
65. Landau BR, Wahren J, Previs SF, et al. Glycerol produc-tion and utilization in humans: sites and quantification.Am J Physiol 1996; 271: E1110-7
66. Elia M, Kahn K, Calder G, et al. Glycerol exchange acrossthe human forearm assessed by a combination of tracerand arteriovenous exchange techniques. Clin Sci 1993; 84:99-104
67. Casazza GA, Jacobs KA, Suh S-H, et al. Menstrual cyclephase and oral contraceptive effects on triglyceride mo-bilization during exercise. J Appl Physiol 2004; 97: 302-9
68. Hellstrom L, Blaak E, Hagstrom-Toft E. Gender differ-ences in adrenergic regulation of lipid mobilization duringexercise. Int J Sports Med 1996; 17: 439-47
69. Mittendorfer B, Horowitz JF, Klein S. Effect of gender onlipid kinetics during endurance exercise of moderateintensity in untrained subjects. Am J Physiol 2002; 283:E58-65
70. Haffner SM, Valdez RA. Endogenous sex hormones: im-pact on lipids, lipoproteins, and insulin. Am J Med 1995;98: 40S-7S
71. Faria ACS, Bekenstein LW, Booth Jr RA, et al. Pulsatilegrowth hormone release in normal women during themenstrual cycle. Clin Endocrin 1992; 36: 591-6
72. Benoit V, Valette A, Mercier L, et al. Potentiation of epi-nephrine-induced lipolysis in fat cells from estrogen-trea-ted rats. Biochem Biophys Res Comm 1982; 109: 1186-91
73. Hansen FM, FahmyN, Nielsen JH. The influence of sexualhormones on lipogenesis and lipolysis in rat fat cells. ActaEndocrin 1980; 95: 566-70
74. Ellis GS, Lanza-Jacoby S, Gow A, et al. Effects of estradiolon lipoprotein lipase activity and lipid availability in ex-ercised male rats. J Appl Physiol 1994; 77: 209-15
75. Spriet LL. Regulation of fat/carbohydrate interaction inhuman skeletal muscle during exercise. Adv ExpMed Biol1998; 441: 249-61
76. Romijn JA, Coyle EF, Sidossis LS, et al. Substrate meta-bolism during different exercise intensities in endurance-trained women. J Appl Physiol 2000; 88: 1707-14
77. Romijn JA, Coyle EF, Sidossis LS, et al. Regulation ofendogenous fat and carbohydrate metabolism in relationto exercise intensity and duration. Am J Physiol 1993; 265:E380-91
78. Friedlander AL, Casazza GA, Horning MA, et al. Effectsof exercise intensity and training on lipid metabolism inyoung women. J Appl Physiol 1998; 275: E853-63
79. Heiling VJ, Jensen MD. Free fatty acid metabolism in thefollicular and luteal phases of the menstrual cycle. J ClinEndocrinol Metab 1992; 74: 806-10
80. Jensen MD, Martin ML, Cryer PE, et al. Effects of estro-gen on free fatty acid metabolism in humans. Am J Phy-siol 1994; 266: E914-20
81. Magkos F, Patterson BW, Mittendorfer B. No effect ofmenstrual cycle phase on basal very-low-density lipopro-tein triglyceride and apolipoprotein B-100 kinetics. Am JPhysiol 2006; 291: E1243-9
83. D’Eon TM, Souza SC, Aronovitz M, et al. Estrogenregulation of adiposity and fuel partitioning: evidence ofgenomic and non-genomic regulation of lipogenic andoxidative pathways. J Biol Chem 2005; 280 (43): 35983-91
84. Sidossis LS, Coggan AR, Gastadelli A, et al. A new cor-rection factor for use in tracer estimations of plasma fattyacid oxidation. Am J Physiol 1995; 269: E649-56
85. Oosthuyse T, Bosch AN, Jackson S. Effect of menstrualphase on the acetate correction factor used in metabolictracer studies. Can J Appl Physiol 2003; 28: 818-30
86. Lamont LS, Lemon PWR, Bruot BC. Menstrual cycle andexercise effects on protein catabolism. Med Sci SportsExerc 1987; 19: 106-10
87. White LJ, Ferguson MA, McCoy S, et al. Intramyocellularlipid changes in men and women during aerobic exercise: a1H-magnetic resonance spectroscopy study. J Clin En-docrinol Metab 2003; 88: 5638-43
88. Tarnopolsky MA, Rennie CD, Robertshaw HA, et al.Influence of endurance exercise training and sex on intra-myocellular lipid and mitochondrial ultrastructure, sub-strate use, and mitochondrial enzyme activity. Am JPhysiol 2007; 292 (3): R1271-8
89. Devries MC, Lowther SA, Glover AW, et al. IMCL areadensity, but not IMCL utilization, is higher in womenduring moderate-intensity endurance exercise, comparedwith men. Am J Physiol 2007; 293 (6): R2336-42
90. Zehnder M, Ith M, Kreis R, et al. Gender-specific usage ofintramyocellular lipids and glycogen during exercise. MedSci Sports Exerc 2005; 37 (9): 1517-24
91. Forsberg AM, Nilsson E, Werneman J, et al. Musclecomposition in relation to age and sex. Clin Sci 1991; 81:249-56
92. Roepstorff C, Steffensen CH, Madsen M, et al. Genderdifferences in substrate utilization during submaximalexercise in endurance-trained subjects. Am J Physiol 2002;282: E435-47
93. Roepstorff C, Donsmark M, Thiele M, et al. Sex differ-ences in hormone-sensitive lipase expression, activity, andphosphorylation in skeletal muscle at rest and during ex-ercise. Am J Physiol 2006; 291: E1106-14
226 Oosthuyse & Bosch
ª 2010 Adis Data Information BV. All rights reserved. Sports Med 2010; 40 (3)
94. Steffensen CH, Roepstorff C, Madsen M, et al. Myo-cellular triacylglycerol breakdown in females but not inmales during exercise. Am J Physiol 2002; 282: E634-42
95. Jørgensen SB, Richter EA, Wojtaszewski JFP. Role ofAMPK in skeletal muscle metabolism regulation andadaptation in relation to exercise. J Physiol 2006; 574 (1):17-31
96. Steinberg GR, Macaulay SL, Febbraio MA, et al. AMP-activated protein kinase: the fat controller of the energyrailroad. Can J Physiol Pharmacol 2006; 84: 655-65
97. Roepstorff C, Thiele M, Hillig T, et al. Higher skeletalmuscle a2AMPK activation and lower energy charge andfat oxidation in men than in women during submaximalexercise. J Physiol 2006; 574 (1): 125-38
98. Wiik A, Gustafsson T, Esbjornsson M, et al. Expression ofoestrogen receptor alpha and beta is higher in skeletalmuscle of highly endurance-trained than of moderatelyactive men. Acta Physiol Scand 2005; 184 (2): 105-12
99. Combs TP, Berg AH, Rajala MW, et al. Sexual differ-entiation, pregnancy, calorie restriction, and aging affectthe adipocyte-specific secretory protein adiponectin.Diabetes 2003; 52: 268-76
100. Kleiblova P, Springer D, Haluzık M. The influence ofhormonal changes during menstrual cycle on serum adi-ponectin concentration in healthy women. Physiol Res2006; 55: 661-6
101. Kriengsinyos W, Wykes LJ, Goonewardene LA, et al.Phase of menstrual cycle affects lysine requirement inhealthy women. Am J Physiol 2004; 287: E489-96
102. Lariviere F, Moussalli R, Garrel DR. Increased leucineflux and leucine oxidation during the luteal phase ofthe menstrual cycle in women. Am J Physiol 1994; 267:E422-8
103. Toth MJ, Sites CK, Matthews DE, et al. Ovarian sup-pression with gonadotropin-releasing hormone agonistreduces whole body protein turnover in women. Am JPhysiol 2006; 291: E483-90
104. Hamadeh MJ, Devries MC, Tarnopolsky MA. Estrogensupplementation reduces whole body leucine and carbo-hydrate oxidation and increases lipid oxidation in menduring endurance exercise. J Clin Endocrinol Metab 2005;90 (6): 3592-9
105. Miller BF, Hansen M, Olesen JL, et al. No effect of men-strual cycle on myofibrillar and connective tissue proteinsynthesis in contracting skeletal muscle. Am J Physiol2006; 290: E163-8
Correspondence: Dr Tanja Oosthuyse, School of Physiology,University of the Witwatersrand Medical School, PostnetSuite 19, Private Bag X8, Northriding 2162, South Africa.E-mail: [email protected]
Menstrual Cycle, Metabolism and Performance 227
ª 2010 Adis Data Information BV. All rights reserved. Sports Med 2010; 40 (3)
