novel causes and consequences of overtraining syndrome ......these novel findings may explain...

RESEARCH ARTICLE Open Access

Novel causes and consequences ofovertraining syndrome: theEROS-DISRUPTORS studyFlavio A. Cadegiani* and Claudio E. Kater

Abstract

Background: Hormonal physiology in athletes, dysfunctional paths leading to overtraining syndrome (OTS), andclinical and biochemical behaviors that are independently modified by the presence of OTS remain unclear.Although multiple markers of OTS have recently been identified, the independent influence of OTS on hormonesand metabolism have not been assessed. Hence, the objective of the present study was to uncover the previouslyunrecognized independent predictors of OTS and understand how OTS independently modifies the behaviors ofclinical and biochemical parameters.

Methods: In a total of 39 athletes (OTS-affected athletes (OTS) = 14 and healthy athletes (ATL) = 25), we performedtwo clusters of statistical analyses using the full data of the Endocrine and Metabolic Responses on OvertrainingSyndrome (EROS) study, in a total of 117 markers. We first used logistic regression to analyze five modifiableparameters (carbohydrate, protein, and overall caloric intake, sleep quality, and concurrent cognitive effort) aspotential additional independent risk factors for OTS, and OTS as the outcome. We then used multivariate linearregression to analyze OTS as the independent variable and 38 dependent variables. Training patterns were found tobe similar between OTS and ATL, and therefore excessive training was not a risk, and consequently not a predictor,for OTS.

Results: Each of the three dietary patterns (daily carbohydrate, daily protein, and daily overall calorie intake) werefound to be the independent triggers of OTS, while sleeping, social, and training characteristics depended on otherfactors to induce OTS. Once triggered, OTS independently induced multiple changes, including reductions ofcortisol, late growth hormone and adrenocorticotropic hormone responses to stimulations, testosterone-to-estradiolratio, neutrophils, neutrophil-to-lymphocyte ratio, vigor levels, hydration status, and muscle mass, while increase oftension levels and visceral fat.

Conclusions: OTS can be independently triggered by eating patterns, regardless of training patterns, while theoccurrence of OTS reduced late hormonal responses and the testosterone-to-estradiol ratio, worsened mood, andaffected the immunology panel. These novel findings may explain underperformance, which is the keycharacteristic of OTS.

Keywords: Athletes, Conditioning, Endocrine and metabolic responses on overtraining syndrome (EROS) study,Performance, Overtraining syndrome

© The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

* Correspondence: [email protected]; [email protected] and Hypertension Unit, Division of Endocrinology and Metabolism,Department of Medicine, Federal University of São Paulo (Unifesp/EPM), RuaPedro de Toledo 781 – 13th floor, São Paulo, SP 04039-032, Brazil

Cadegiani and Kater BMC Sports Science, Medicine and Rehabilitation (2019) 11:21 https://doi.org/10.1186/s13102-019-0132-x

http://crossmark.crossref.org/dialog/?doi=10.1186/s13102-019-0132-x&domain=pdf

http://orcid.org/0000-0002-2699-4344

http://creativecommons.org/licenses/by/4.0/

http://creativecommons.org/publicdomain/zero/1.0/

mailto:[email protected]

mailto:[email protected]

IntroductionOvertraining syndrome (OTS) is characterized by a pro-longed and unexplained decrease in sports performanceusually associated with severe psychological manifesta-tions [1]. It is caused by an imbalance among training,social, sleep, and eating patterns, which leads to meta-bolic, endocrine, and biochemical changes [2–7] relatedto a long-term shortage of energy and mechanisms of re-pair [1, 3, 6, 8]. Despite the name “overtraining syn-drome”, referring to excessive exercise training, othermodifiable factors may trigger OTS [3, 4, 9]. Althoughexcessive training disrupts physiological processes lead-ing to OTS, whether and how eating, social, and sleeppatterns disrupt adaptive changes in athletes remainsuncertain. Challenges and issues in the methodologyused for the assessment of OTS include the failure toidentify relevant biomarkers and pathophysiology.Given the gap in knowledge about OTS triggers, we

conducted the Endocrine and Metabolic Responses toOvertraining Syndrome (EROS) study [9–13], to en-hance our understanding of the pathophysiology andbiomarkers of OTS and assessment tools for the earlyrecognition, and prevention of OTS. The EROS study[9–13] compared OTS-affected athletes (OTS group),healthy athletes (ATL group), and non-physically activecontrols on 117 parameters. The identification of morethan 45 new biomarkers and mechanisms of OTSshowed that disruptions in the adaptive changes of ath-letes led to a breakdown in multiple pathways, therebyleading to OTS. However, despite the multiple novelfindings in OTS, we were unable to identify independentpredictors of OTS and parameters that were independ-ently disrupted by the presence of OTS without a jointanalysis of all the results of the EROS study [9–13] theuse of more complex statistical tools.The objective of the present study was to understand

whether and which factors independently trigger OTS,aside from excessive training, and how OTS independ-ently leads to changes in behaviors in multiple clinical,metabolic, and biochemical markers. The uncovering ofthese mechanisms in the present study was intended toimprove our understanding of the etiology and conse-quences of OTS, and to develop additional tools for theprevention and precise diagnosis of OTS.

MethodsFor the present analysis, we performed a comprehensivejoint statistical analysis of data from five of the arms ofthe EROS study, including four of primary findings: 1.)the EROS-HPA axis, in which we evaluated the hypo-thalamic-pituitary-adrenal axis hormonal responses inathletes [9]; 2.) the EROS-STRESS, in which we evalu-ated the prolactin and growth hormone (GH) responsesto an exercise-independent stimulation test – the insulin

tolerance test (ITT) – as well as the glucose behaviorduring this test [10]; 3.) the EROS-PROFILE, in whichwe evaluated eating, psychological, sleeping, and socialpatterns [11]; 4.) the EROS-BASAL, in which we evalu-ated: basal hormones; inflammatory; immune; and mus-cular parameters [12]; and 5.) the additional EROS-HIFTarm, in which we evaluated specific characteristics ofhealthy and OTS-affected HIFT (including CrossFit) ath-letes [3].Full descriptions of the materials and methods (i.e.,

the selection of participants and study procedures),results of the statistical analyses of data, and their re-spective discussions are available in these five of thearms of the present study [9–13], as well as in adepository (https://osf.io/bhpq9), which also has theraw data of the results of each participant.

Subject selectionWe recruited participants through calls for participation insocial media and group messages, and invitations to sportscoaches. Prior to interview, each candidate self-reportedsex, age, body mass index (BMI), and whether he intendedto participate as a healthy athlete, clinically suspected forOTS, or healthy sedentary. Aiming homogenous groups,we specified criteria for all groups, of OTS-affected athletes(OTS group), healthy athletes (ATL group), and non-phys-ically active controls (NPAC group) including sex (male),age (18–50 years old), BMI (20–29.9 kg/m2 for sedentaryand 20–32.9 kg/m2 for athletes), absence of known hormo-nal, metabolic, inflammatory, or psychiatric disorders, non-current or recent use of drugs or hormones. In order toavoid false athletes, for the two groups of athletes we re-quired a minimum amount of training per week (> 300minand > four times a week), intensity of training (at least mod-erate-to-intense, according to their sport coaches), timesince started non-stop training (> 6 months). To avoid mis-diagnosis of OTS, for athletes suspected of OTS, we re-quired a sports-coach certified reduction of at least 10% ofprevious performance, or a loss of > 20% in time to fatigue,increased sense of effort for a same training intensity andvolume, persistent fatigue that lasted > 2 weeks, unrespon-sive to resting, and lack of use of confounding drugs or hor-mones, and presence of confounding diseases. For allcandidates that fulfilled criteria for any of the two groups,we performed hormones and basic biochemical profile andavoided those who presented alterations in any of the testedparameters.

Identification of independent triggers and consequencesof OTSIn the present study, we performed a joint multivariateand logistic regression analyses for the identification ofindependent triggers and consequences of OTS, amongthose parameters that were suitable for the diagnosis or

Cadegiani and Kater BMC Sports Science, Medicine and Rehabilitation (2019) 11:21 of 13

https://osf.io/bhpq9

as characteristic of OTS, and significantly different be-tween OTS and ATL. From the 117 evaluated parame-ters in the EROS study [9–13], 31 were non-diagnostic,not useful, or unsubstantiated; nine were qualitative (yesvs no), three had missing data in > 5% of participants,and 27 had similar levels (Table 1). Hence, we selected44 hormonal responses to stimulation tests, basal andaccumulated hormonal levels, social and psychologicalaspects, specific eating patterns, and body metabolismand composition parameters, among which 38 were vari-ables dependent of five modifiable variables (eating,sleeping, and social patterns) plus the presence of OTSas an additional variable, in a total of 44 variables (38dependent and 6 independent variables) with two groupsof athletes: the OTS and ATL groups (N = 39; OTS = 12and ATL = 25) (Fig. 1). Additional analyses of the excludedparameters n were also performed, but were not includedThe variability of the biochemical markers measured in allarms of the EROS study were as low as 3.5 and 3.0% for in-ter- and intra-assay coefficients, respectively.

Statistical analysisOf 44 markers, we used logistic regression to analyzefive independent variables and one dependent variable inorder to identify independent triggers of OTS. Fivemodifiable habits, including caloric, protein, and carbo-hydrate intake, sleep quality, and the number of hoursspent working or studying were the independent predic-tors, and the presence of OTS was the outcome. Weused multivariate linear regression with six independentvariables, including the five modifiable variables plus thepresence of OTS (as the sixth independent variable), andthe markers among the 38 remaining variables that weresignificantly different between the OTS and ATL groups,as the dependent variables. The purpose was to elucidatethe role of OTS as an independent modifier of bodycomposition and metabolism, and biochemical, hormo-nal, and psychological markers.All statistical analyses were performed using SAS 9.4

(SAS Institute, Inc., Cary, NC). Logistic regression ana-lyses were performed using the five independent vari-ables and binominal codes for the presence or absenceof OTS as the dependent variable; 35.9% of the partici-pants had OTS (p < 0.05). Given the context of thepresent study and its main objective, the number of par-ticipants in the present study was found to be sufficientfor the number of variables and outcomes for thepresent logistic regression analyses. We used multivari-ate linear regression with the backward variable selectionmethod (removal criterion = p > 0.01) to analyze the sig-nificance of the contributions of the 44 variables. Thisprocess was performed until none of the predictors metthe removal criterion. The standardized residual vari-ables of the last performed model were examined for

normality, homoscedasticity, and multicollinearity. Thetolerance index of the remaining variables in the lastmodel was ≥0.403. A p-value < 0.05 for all analyses wasconsidered statistically significant.

ResultsIn the EROS study, from the 87 athletes suspected ofOTS and 46 healthy athletes that were initially recruited,14 were selected for the OTS group (83.9% of the initialcandidates for OTS were excluded due to exclusion ofthe actual diagnosis of OTS) and 25 for the ATL group.The baseline characteristics of age (OTS = 30.6 years andATL = 32.7 years) and body mass index (BMI) (OTS =26.7 kg/m2 and ATL = 24.9 kg/m2), and the training pat-terns including training intensity (OTS = 8.79 and ATL =8.76, on a scale from zero to ten), frequency (OTS =5.36 days and ATL = 5.46 days) and period (OTS = 574.3min and ATL = 550.0 min a week), and time sincestarted training non-stop were statistically similar be-tween OTS and ATL. All 14 participants selected for theOTS group had true and naturally occurring presence ofOTS, not functional or non-functional overreaching, asall athletes had a verified decrement of > 10% of previoussports performance and fatigue that were prolonged(average duration of fatigue and decreased perform-ance = 44.3 ± 23.0 days), and none had fully recovered bythe time of the study. Supplementary information re-garding the selection process and baseline characteristicshave been previously published [3–7].The independent triggers of OTS are presented in

Table 2. The variables that were independently modifiedby the presence of OTS, their degree of association, andthe estimation equation for each of these variables arepresented in Table 3.When analyzed together, at least one factor between

low carbohydrate, low protein intake, low overall caloricintake, and poor sleep quality was present in 100% ofthe study’s cases of OTS. Carbohydrate intake was foundto be an independent trigger of OTS when it was ana-lyzed together with sleep and social patterns, with anodds ratio (OR) = 1.61, [confidence limits (CL) = 1.03–2.50] for the risk of developing OTS, while its ability toinduce OTS was lost without the concurrent analysis ofsleep quality.Conversely, protein intake was shown to independently

induce OTS without the concurrent analysis of any of theother possible triggers in all scenarios. Likewise, overallcaloric intake independently induced OTS, irrespective ofthe proportions of macronutrients, indicating that if cal-oric intake (but no carbohydrate or protein intake), workhours, and sleep quality had been analyzed together as thethree modifiable habits, caloric intake would have beenthe only independent trigger (p = 0.004; OR = 1.13 [CL =1.04–1.23]) between these three variables. In contrast,


Table

1Markersinclud

edin

thepresen

tanalysis,amon

gthoseevaluatedby

theEROSstud

y

Stud

y/Tests

Markers

Totaln

umberof

markers:117

(Included:45)

Whe

ther

includ

edor

exclud

ed(and

ifexclud

ed,w

hy)

EROS-HPA

axis

Totaln

umberof

markers:20

Included

markers:7

BasalA

CTH

andcortisol

andtheirrespon

seto

aninsulin

tolerancetest(ITT)

1.Basalcortisol

(μg/dL)

2.Cortisol

durin

ghypo

glycem

ia(μg/dL)

3.Cortisol

30min

afterhypo

glycem

ia(μg/dL)

4.Cortisol

increase

durin

gITT(μg/dL)

5.BasalA

CTH

(pg/mL)

6.ACTH

durin

ghypo

glycem

ia(pg/mL)

7.ACTH

30min

afterhypo

glycem

ia(pg/mL)

8.ACTH

increase

durin

gITT(pg/mL)

9BasalA

CTH

/cortisol

ratio

10.A

CTH

/cortisol

ratio

durin

ghypo

glycem

ia11.A

CTH

/cortisol

ratio

30min

afterhypo

glycem

ia

Similarlevelsbe

tweenOTS

andATL

INCLU

DED

INCLU

DED

INCLU

DED

Similarlevelsbe

tweenOTS

andATL

Similarlevelsbe

tweenOTS

andATL

INCLU

DED

INCLU

DED

Unsub

stantiatedmarker

Unsub

stantiatedmarker

Unsub

stantiatedmarker(alth

ough

different

betw

eenOTS

andATL)

Cortisol

respon

seto

acosyntropinstim

ulation

test(CST)

12.C

ortisol

at30

min

aftersynthe

ticACTH

shot

(μg/dL)

13.C

ortisol

at60

min

aftersynthe

ticACTH

shot

(μg/dL)

14.D

ifferen

cebe

tweenbasalcortisol

onday1(CST)and

day3(ITT)

(%)

Similarlevelsbe

tweenOTS

andATL

Similarlevelsbe

tweenOTS

andATL

Not

diagno

sticor

helpful

Salivarycortisol

rhythm

(SCR)

15.Salivarycortisol

ataw

aken

ing(ng/dL)

16.Salivarycortisol

30min

afterwaken

ing(ng/dL)

17.Salivarycortisol

at4PM

(ng/dL)

18.Salivarycortisol

at11

PM(ng/dL)

19.C

ortisol

awaken

ingrespon

se(CAR)

(%)

20.D

ifferen

cebe

tween8AM

and4PM

salivarycortisol

(%)

Similarlevelsbe

tweenOTS

andATL

INCLU

DED

Similarlevelsbe

tweenOTS

andATL

Similarlevelsbe

tweenOTS

andATL

INCLU

DED

Not

diagno

sticor

helpful

EROS-STRESS

Totaln

umberof

evaluatedmarkers:12

Included

markers:7

GHandProlactin

respon

seto

anITT

1.Basal(GH)(μg

/L)

2.GHdu

ringhypo

glycaemia(μg/L)

3.GH30

min

afterhypo

glycaemia(μg/L)

4.Basalp

rolactin

(ng/mL)

5.Prolactin

durin

ghypo

glycaemia(ng/mL)

6.Prolactin

30min

afterhypo

glycaemia(ng/mL)

7.Prolactin

increase

durin

gITT(ng/mL)

8.Basalserum

glucose(m

g/dL)

9Serum

glucosedu

ringhypo

glycem

ia(m

g/dL)

10.C

apillaryglucosedu

ringhypo

glycem

ia(m

g/dL)

11.A

dren

ergicsymptom

sdu

ringhypo

glicem

ia(0–10)

12.N

euroglycop

enicsymptom

sdu

ringhypo

glicem

ia(0–10)

INCLU

DED

INCLU

DED

INCLU

DED

INCLU

DED

INCLU

DED

INCLU

DED

INCLU

DED

Not

diagno

sticor

helpful

Not

diagno

sticor

helpful

Not

diagno

sticor

helpful

Not

diagno

sticor

helpful(althou

ghdifferent

betw

eenOTS

andATL)

Not

diagno

sticor

helpful

EROS-BA

SAL

(totaln

umberof

evaluatedmarkers:32)

Included

markers:9

Hormon

almarkers

1.Totaltestosteron

e(ng/dL)

2.Estradiol(pg

/mL)

3.IGF-1(pg/mL)

4.TSH(μUI/m

L)5.Free

T3(pg/mL)

6.Totalcatecho

lamines

(μg/12

h)7.Totalm

etanep

hrines

(μg/12

h)8.Noradrenaline(μg/12

h)9.Epinep

hrine(μg/12

h)10.D

opam

ine(μg/12

h)

INCLU

DED

INCLU

DED

Similarlevelsbe

tweenOTS

andATL

Similarlevelsbe

tweenOTS

andATL

Similarlevelsbe

tweenOTS

andATL

INCLU

DED

Similarlevelsbe

tweenOTS

andATL

Similarlevelsbe

tweenOTS

andATL

Similarlevelsbe

tweenOTS

andATL

INCLU

DED


Table

1Markersinclud

edin

thepresen

tanalysis,amon

gthoseevaluatedby

theEROSstud

y(Con

tinued)

Stud

y/Tests

Markers

Totaln

umberof

markers:117

(Included:45)

Whe

ther

includ

edor

exclud

ed(and

ifexclud

ed,w

hy)

11.M

etanep

hrine(μg/12

h)12.N

ormetanep

hrine(μg/12

h)13.C

atecho

lamine-to-m

etanep

hrineratio

Similarlevelsbe

tweenOTS

andATL

Similarlevelsbe

tweenOTS

andATL

Not

diagno

sticor

helpful(althou

ghdiferentebe

tweenOTS

andATL)

Bioche

micalmarkers

14.Erythrocyte

sedimen

tatio

nrate

(ESR,m

m/h)

15.H

ematocrit

(%)

16.C

-reactiveprotein(CRP,m

g/dL)

17.Lactate

(nMol/L)

18.Vitamin

B12(pg/mL)

19.Ferritin

(ng/mL)

20.N

eutrop

hils(*mm

3 )21.Lym

phocyte(*mm

3 )22.Eosinop

hils(*mm

3 )23.C

reatinekinase

(CK,U/L)

24.M

edium

corpuscularvolume(M

CV)

25.Platelets(103/m

m)

26.Low

density

lipop

rotein

cholesterol(LD

Lc)(m

g/dL)

27.H

ighde

nsity

lipop

rotein

cholesterol(HDLc)(m

g/dL)

28.Tryglicerides

(mg/dL)

Similarlevelsbe

tweenOTS

andATL

Similarlevelsbe

tweenOTS

andATL

Similarlevelsbe

tweenOTS

andATL

INCLU

DED

Similarlevelsbe

tweenOTS

andATL

Similarlevelsbe

tweenOTS

andATL

INCLU

DED

Similarlevelsbe

tweenOTS

andATL

Similarlevelsbe

tweenOTS

andATL

INCLU

DED

Not

diagno

sticor

helpful

Not

diagno

sticor

helpful

Datamissedin

>5%

ofparticipants

Datamissedin

>5%

ofparticipants

Datamissedin

>5%

ofparticipants

Ratio

s29.Testosteron

e-to-oestradiolratio

30.Testosteron

e-to-cortisol

ratio

31.N

eutrop

hil-to-lymph

ocyteratio

32.Platelet-to-lymph

ocyteratio

s

INCLU

DED

Similarlevelsbe

tweenOTS

andATL

INCLU

DED

Similarlevelsbe

tweenOTS

andATL

EROS-PROFILE

(totaln

umberof

evaluatedmarkers:53)

Included

markers:21

Nutritionalp

atterns(7-day

diet

record,p

riorand

durin

gOTS)

1.Calorieintake

(kcal/kg/day)

2.Carbo

hydrateintake

(g/kg/day)

3.%

caloriesfro

mcarboh

ydrate

(%)

4.Proteinintake

(g/kg/day)

5.%

caloriesfro

mprotein(%)

6.Fatintake

(g/kg/day)

7.%

caloriesfro

mfat(%)

8.Carbo

hydrateintake

>3g/kg/day

(Y/N)

9.Dailywhe

yproteinconsum

ption(Y/N)

10.Followed

adiet

plan

(Y/N)

11.Post-workout

carboh

ydrate

intake

>0.5g/kg

(Y/N)

INCLU

DED

(asamod

ifiablehabita)

INCLU

DED

(asamod

ifiablehabita)

Intrinsically

linkedto

othe

rparameters

INCLU

DED

(asamod

ifiablehabita)

Intrinsically

linkedto

othe

rparameters

Similarlevelsbe

tweenOTS

andATL

Intrinsically

linkedto

othe

rparameters

Qualitativemarker

Qualitativemarker

Qualitativemarker

Qualitativemarker

Psycho

logicalp

atterns(duringOTS)

12.Profileof

Moo

dState(POMS)

questio

nnaire

(totalscore:−32

to+120)

13.A

nger

subscale(0

to48)

14.C

onfusion

subscale(0

to28)

15.D

epressionsubscale(0

to60)

16.Vigou

rsubscale(0

to32)

17.Fatigue

subscale(0

to28)

18.Ten

sion

subscale(0

to36)

19.H

owdo

youfelltoday?

(0–10)

20.H

aveyoube

ensick

inthelasttw

oweeks?(Y/N)?

21.H

owwas

your

lasttraining

sessioncomparedto

theprojectedgo

als?

(Extremelyeasy

toextrem

elyhard)

22.H

owdo

your

muscles

feel?(Nothing

atalltoextrem

elypainful)

INCLU

DED

INCLU

DED

INCLU

DED

INCLU

DED

INCLU

DED

INCLU

DED

INCLU

DED

Not

diagno

sticor

helpful(althou

ghdifferent

betw

eenOTS

andATL)

Qualitativemarker

Not

diagno

sticor

helpful(althou

ghdifferent

betw

eenOTS

andATL)

Not

diagno

sticor

helpful(althou

ghdifferent

betw

eenOTS

andATL)


Table

1Markersinclud

edin

thepresen

tanalysis,amon

gthoseevaluatedby

theEROSstud

y(Con

tinued)

Stud

y/Tests

Markers

Totaln

umberof

markers:117

(Included:45)

Whe

ther

includ

edor

exclud

ed(and

ifexclud

ed,w

hy)

23.H

owfrien

dlydo

youfeeltoday?

(0–6)

24.H

owworthless

doyoufeeltoday?

(0–6)

25.H

owmiserabledo

youfeeltoday?

(0–6)

26.H

owhe

lpfuld

oyoufeeltoday?

(0–6)

27.H

owbad-tempe

reddo

youfeeltoday?

(0–6)

28.H

owun

worthydo

youfeeltoday?

(0–6)

29.H

owpe

eved

doyoufeeltoday?

(0–6)

30.H

owcheerfu

ldoyoufeeltoday?

(0–6)

31.H

owsaddo

youfeeltoday?

(0–6)

32.N

umbe

rof

hoursof

activities

beside

sprofession

altraining

(h/day)

Not

diagno

sticor

helpful

Not

diagno

sticor

helpful

Not

diagno

sticor

helpful

Not

diagno

sticor

helpful(althou

ghdifferent

betw

eenOTS

andATL)

Not

diagno

sticor

helpful

Not

diagno

sticor

helpful

Not

diagno

sticor

helpful

Not

diagno

sticor

helpful(althou

ghdifferent

betw

eenOTS

andATL)

Not

diagno

sticor

helpful

INCLU

DED

(asamod

ifiablehabita)

Socialpatterns

(duringOTS)

33.D

urationof

nigh

tsleep(h)

34.Self-rep

ortedsleepqu

ality

(0–10)

35.Self-rep

ortedlibido(0–10)

36.Initialimnson

ia(Y/N)

37.Terminalim

nson

ia(Y/N)

38.M

orethan

twowake-up

sdu

ringsleep(Y/N)

39.W

orkand/or

stud

y(Y/N)

40.Libidodu

ringrestingpe

riods

/vacatio

ns(0–10)

Similarlevelsbe

tweenOTS

andATL

INCLU

DED

(asamod

ifiablehabita)

INCLU

DED

Qualitativemarker

Qualitativemarker

Qualitativemarker

Qualitativemarker

Not

diagno

sticor

helpful

Body

metabolism

analysis(indirect

calorim

etry)

41.M

easured-to-predicted

basalm

etabolicrate

(BMR,%)

42.Percentageof

fatbu

rningcomparedto

totalB

MR(%)

INCLU

DED

INCLU

DED

Body

compo

sitio

n(Bod

Pod,

InBo

dy770and3D

body

scanne

r)43.Bod

yfatpe

rcen

tage

(%)

44.Visceralfat

(cm

2 )45.M

usclemassweigh

t(%)

46.Bod

ywater

percen

tage

(BW,%

)47.Extracellularwater

comparedto

totalB

W(%)

48.Bod

yweigh

t(kg)

49.Che

stto

waistcircum

ference

50.W

aistcircum

ference(cm)

51.C

hestcircum

ference(cm)

52.Bicep

scircum

ference(cm)

53.H

ipcircum

ference(cm)

INCLU

DED

INCLU

DED

INCLU

DED

INCLU

DED

INCLU

DED

Not

diagno

sticor

helpful

Not

diagno

sticor

helpful

Not

diagno

sticor

helpful

Not

diagno

sticor

helpful

Not

diagno

sticor

helpful

Not

diagno

sticor

helpful

OTS

Athletesaffected

byov

ertraining

synd

rome,

ATL,H

ealth

yathletes

a For

statistical

purposes,m

odifiab

lefactorswereconsidered

asinde

pend

entvaria

bles,from

which

thede

pend

entvaria

bles

werestatistically

evalua

ted


excessive work and poor sleep quality, each failed to in-duce OTS independently, regardless of the combinationsof predictors.Among the parameters that were statistically differ-

ent between the OTS and ATL groups, the presenceof OTS independently blunted late responses of adre-nocorticotropic hormone (ACTH), cortisol, andgrowth hormone (GH) to an insulin tolerance test(ITT). This accounted for 20, 26, and 23% of their re-sponses, respectively, while later prolactin, and earlyACTH, cortisol, GH, and prolactin responses wereunaffected by OTS.

With respect to the basal hormones, OTS reduced thetestosterone-to-estradiol (T:E) ratio by 43%, while it didnot modulate total testosterone, estradiol, or any of theother hormones. Conversely, the basic immunologypanel, including neutrophils, lymphocytes, and the neu-trophil-to-lymphocyte ratio were influenced by the oc-currence of OTS, although at lower degrees ofassociation, and only when combined with othertriggers.OTS also affected tension, fatigue, and vigor levels

when evaluated through the Profile of Mood States(POMS) questionnaire, accounting for 43, 84, and 86%

Fig. 1 Variables included in the present analysis


of their levels, respectively. While OTS did not affectany aspect of body metabolism (ratio between measuredand expected basal metabolic rate (BMR) and percentageof fat oxidation), it independently led to reductions inmuscle mass and body water content to 34 and 51%, re-spectively, and an increase in visceral fat to 38%. Whilevisceral fat was increased in OTS, overall body fat wasunchanged by the presence of OTS.

DiscussionDespite the identification of multiple markers amongclinical, metabolic, and biochemical parameters in OTSathletes in the EROS study [9–13], we were unable toidentify specific patterns or a standard group of bio-markers for OTS, as each affected athlete exhibited aunique combination of altered markers. In the absenceof a unique accurate biomarker for the diagnosis ofOTS, we observed that combinations of markers thatwere significantly different between the OTS and ATLgroups could potentially lead to a precise diagnosis ofOTS, with an accuracy of 100% to distinguish OTS ath-letes from healthy athletes. Despite the successful abilityto identify affected athletes, we were unable to identifyindependent triggers of OTS, as our previous analysesdid not identify the influences on or causes of OTS atthe individual level. Moreover, the analyses did not en-hance our understanding of how each of the modifiablepatterns and the occurrence of OTS independently in-duced changes in the behaviors of multiple clinical and

biochemical markers (i.e., inherent changes caused byeach modifiable factor, and changes that were independ-ently modified by OTS itself, not by its triggers).The post-hoc use of multivariate linear regression and

logistic regression, which were not used in the previousEROS studies on OTS [9–13], identified the factors thatindependently led to OTS, and the parameters that wereinherently modulated by the presence of OTS. Tounderstand the correlations between OTS and its trig-gers, and OTS and its consequences, we investigatedwhich modifiable factors could be independent causes ofOTS, (i.e., whether a specific modifiable factor was solelyresponsible for the occurrence of some cases of OTS).We also examined which parameters might be inde-pendently modified by the presence of OTS, irrespectiveof other characteristics (i.e., even with the same caloric,protein, and carbohydrate intake, the same sleep qualityand duration, the same amount of additional sports-re-lated activity, and the same training intensity, volume,frequency, and duration). Our aim was to identifywhether and how the mere presence of OTS modifiedthe behaviors of the tested parameters. Specifically,among the intrinsic mechanisms of OTS, which were in-herently responsible for at least some of the dysfunc-tional changes found in OTS, as consequences, notcauses, of OTS. The dysfunctional adaptations in theclinical and biochemical aspects induced by the modifi-able factors, plus the changes in these parameters wereinherently due the occurrence of OTS, which was

Table 2 Independent triggers of OTS

Scenario Independentvariablesincluded

Results (* = positive forindependent risk factors andtriggers)

Interpretation

Scenario 1 – All modifiablevariables

CHO, PROT,CAL, WORK,SLEEP

PERFECT SEPARATION Together, modifiable patterns were able to explain all cases of OTS inthe athletes studied.

Scenario 2 – All modifiablevariables, except WORK

CHO, PROT,CAL, SLEEP

PERFECT SEPARATION Dietary patterns together with sleep quality were also able to fullyexplain all cases of OTS in the studied population of athletes.

Scenario 3 – All modifiablevariables, except CAL

CHO, PROT,WORK, SLEEP

CHO: p = 0.036OR/CL = 1.61 (1.03–2.50)PROT: p = 0.029OR/CL = 16.7 (1.34–208.1)WORK: p = n/sSLEEP: p = 0.069OR/CL = 2.19 (0.94–5.09)

When daily caloric intake is not accounted, not all cases of OTS may bejustified. However, in this scenario both CHO and PROT were shown tobe independent triggers of OTS.

Scenario 4 – Withoutspecification of eachmacronutrient

CAL, WORK,SLEEP

CAL: p = 0.004OR/CL = 1.13 (1.04–1.23)WORK: p = n/sSLEEP: p = n/s

When each macronutrient intake is not specified, not all cases of OTSmay be justified. However, in this scenario CAL was enough toindependent etiology of OTS.

Scenario 5 – Only dietarypatterns

CHO, PROT,CAL

CHO: p = n/sPROT: p = 0.066OR/CL = 25.85 (0.81–825.3)CAL: p = 0.045OR/CL = 1.27 (1.01–1.61)

When only dietary patterns are evaluated, we cannot explain all casesof OTS in the studied population. However, in this scenario, overallcaloric intake, but not each macronutrient, was able to

CHO Daily carbohydrate intake (g/kg/day), PROT Daily protein intake (g/kg/day), CAL Mean daily caloric intake (kcal/kg/day), WORK Average number of working orstudying hours a day, besides training sessions (h/day), SLEEP Self-reported sleep quality (0–10), OTS Overtraining syndrome, OR Odds ratio, CL 95% ConfidenceLimits, p Level of significance, n/s non-significant (p > 0.1)


triggered by the same modifiable factors that also led tochanges in the behaviors of multiple parameters. In anegative synergistic process, in which dysfunctions wereenhanced by the concurrent insufficient carbohydrate,protein, and/or caloric intake, or poor sleep quality, andthe presence of OTS, they were also induced by thesefactors, whereby both changeable factors and the pres-ence of OTS increased the dysfunctions induced by bothfactors. This vicious cycle probably plays an importantrole in the challenging recovery process of OTS, as thesefactors can have a “snowball effect,” which precludes thehealing process.The use of both healthy and OTS-affected athletes for

the logistic regression analyses was important to predictbehavior patterns prior to OTS, as the development ofOTS may be understood as a process on a continuum(i.e., the end of an unresolved mixture of attempts toadapt to chronic energy depletion and the mechanismsunderlying a recovery-deprived environment) [1–3, 8, 9,14, 15]. The significant differences in clinical, hormonal,metabolic, psychological, and biochemical behaviors be-tween the ATL and OTS groups, when all variables wereperfectly adjusted for baseline characteristics, and train-ing, eating, social, and sleep patterns, supported the

conclusion that these changes in behaviors were inher-ently due to the presence of OTS, as the occurrence ofOTS was shown to independently increase tension levelsand blunt vigor levels, while may independently enhancefatigue, as a sort of a vicious cycle, since fatigue is alsoone of the features of OTS. Given the data generated inthe present study, the relationship between physiologicaland pathological behavior patterns suggest these areearly signs of future dysfunction (OTS), and therefore,should be used as a warning signal in clinical practice.These differentiations and the pathophysiological pathshave provided us with a more comprehensive under-standing of OTS.

Independent triggers of overtraining syndrome: beyondexcessive trainingExcessive training has traditionally been viewed as themajor cause of unexplained reductions in sports per-formance, and therefore, referred to as “overtrainingsyndrome.” However, given advances in knowledgeabout the importance of periodized training, excessivetraining is now considered a minor factor in the de-velopment of OTS.

Table 3 Clinical and biochemical behaviors independently modified by overtraining syndrome (OTS)

Parameters modified by thepresence of OTS

p of theinfluence ofOTS a

Level of influence of thepresence of OTS a

(Adjusted R-Square)

Other variables thatmay also influence

Equation for the estimation of theparameter level in male athletes

Late ACTH response to an ITT(30’after hypoglycaemia) (pg/mL)

0.002 19.9% none n/a

Late cortisol response (30’afterhypoglycaemia) (μg/dL)

0.0005 26.1% none Cortisol (μg/dL) = 17.86–3.81(if OTS)

Cortisol response to an ITT (μg/dL) 0.002 22.0% none n/a

Late GH response (30’afterhypoglycaemia) (μg/L)

0.001 23.0% none n/a

Testosterone-to-oestadiol ratio (T/E) 0.0002 30.7% none T/E = 14.1 + 12.9 (if OTS)

POMS vigour subscale < 0.0001 83.6% Sleep quality POMS vigour subscale = 3.7 + 1.15x(sleepquality) – 11.96(if OTS)

POMS fatigue subscale < 0.0001 85.7% Sleep quality POMS fatigue subscale = 24.5–0.9 x(sleepquality) + 15.3(if OTS)

POMS tension subscale < 0.0001 42.8% none Not able to be estimated

Visceral fat (cm2) 0.002 38.2% Protein and overallcalorie intake

Visceral fat = 47.4–11.9x(protein intake) +1.3x(calorie intake) + 45.1(if OTS)

Muscle mass (%) 0.028 33.7% Protein intake Muscle mass = 47.84 + 1.42x(protein intake)– 3.47(if OTS)

Body water (%) 0.001 50.5% Protein and overallcalorie intake

Body water = 60.75 + 1.69x(protein intake) –0.12x(calorie intake) - 5.77(if OTS)

Neutrophils (/mm3) 0.015 13.8% Calorie intake Neutrophils = 4210–60.7x(calorie intake) +154.4x(CHO intake) -1724(if OTS)

Neutrophil-to-lymphocyte ratio 0.015 13.6% none Ratio = 2.00–1.32(if OTS)

CHO Carbohydrate, ITT Insulin tolerant test, POMS Profile of mood states, BMR Basal metabolic rate, T/E Testosterone-to-oestradiol, OTS Overtraining syndrome’, n/anon applicables (non-normal distribution)Calorie intake = kcal/kg/day, CHO intake = g(CHO)/kg/day; protein intake = g(protein)/kg/day; extra activities = working and/or studying hours besides training,sleep quality = self-reported sleep quality (0 to 10)aOther minor influences may also reflect the p-value and the level of influence


Unexpectedly, the incidence of OTS did not decreasewith the improvements in training patterns, nor did itshow a paradoxical increase; perhaps this finding is dueto the growing number of athletes. Given this context,despite the clear existence of OTS triggers other thanexcessive training, these findings had not been reportedprior to the EROS study.As all the training patterns were similar between the

healthy and OTS-affected athletes in the EROS study,excessive training was not found to be a trigger for all ofthe affected athletes, which allowed us to identify noveletiologies of OTS. In the EROS-PROFILE arm [11], diet-ary (i.e., carbohydrate, protein, and total caloric intake),social (i.e., the number of hours spent working or study-ing), and sleep (e.g., sleep quality) patterns were foundto have a role in the development of OTS, as these pa-rameters were significantly different between OTS andATL group. However, whether any of these triggers wereindependent or dependent upon a combination of trig-gers was not examined in this arm of the EROS study.The combination of OTS triggers identified in the EROS

study using logistic regression explained all cases of OTSamong the participants (i.e., the combination was shownto be “the perfect predictor”). Even without the independ-ent variable of number of hours worked, the combinationof dietary and sleep patterns was still found in all of theOTS cases. Conversely, dietary patterns alone, or the com-bination of two of the three dietary characteristics withother factors did not explain OTS in any of the affectedathletes. Therefore, all dietary patterns plus sleep qualityneed to be assessed in order to identify athletes at risk forOTS. However, not all possible triggers are needed to de-velop OTS. In addition, it is important to mention that avery high odds ratio is likely to be a statistical overesti-mation of an association of different variables when onevariable is the sole predictor of an outcome (in this case,OTS) without controlling for other variables.Carbohydrate, protein, or overall caloric intake may

each, independently disrupt physiological responses to asport; hence, OTS can be induced without the presence ofany of the other risk factors. Noteworthy, OTS is morelikely to occur after changes in eating, sleeping and/or so-cial patterns. In clinical practice, dietary characteristicsshould be assessed prior to other triggers, and wheneverthey do not indicate the presence of OTS, sleep and socialpatterns should be investigated. However, there is not aspecific threshold for each activity or habit, as each the in-fluence of them will highly depend on the combinationwith other potential triggers of OTS.

Overtraining syndrome as an independent predictor ofclinical, metabolic, and biochemical behaviorsOur findings help provide novel tools to identify athletesat risk for developing OTS and for its prevention; this

approach is more efficient than the management of thechallenges associated with recovery from OTS. Specificoutcomes related to these findings are described below.Although early hormonal responses to the ITT were

predicted independently and positively by carbohydrateintake, the presence of OTS predicted their late responses(except for prolactin). Indeed, the commencement of aphysical activity at maximum capacity for a short period,which is represented by early responses to stimulation andunaffected by OTS, is not typically observed in athleteswith OTS. Conversely, reduced time-to-fatigue, a hallmarkof OTS, can be explained by the blunted late hormonal re-sponses independently predicted by the presence of OTS.This indicates an inability to maintain hormonal responsesfor longer periods in the presence of OTS, which probablyexplains the reduced pace and impaired performance ofathletes during training sessions and competitions.Among the basal hormones, the T:E ratio [12], but not

any single hormone, was disrupted by the presence ofOTS. The T:E ratio was found to be a better predictor ofmetabolic and psychological parameters than testoster-one or estradiol alone [12, 16–22], as the benefits of in-creased estradiol in males were apparent only with aconcurrent increase in testosterone [18, 19, 22]. Testos-terone alone did not have the same benefits as the sim-ultaneous increase of both testosterone and estradiol[16–18]. The benefits of an increase in estradiol accom-panied by an increase in testosterone contrasted withthe harmful effects of increased estradiol without an in-crease in testosterone, which is explained by whether theunderlying mechanisms that raise estradiol levels arephysiological or pathological. Estradiol physiologicallyincreases in response to increased testosterone, andtherefore, both levels are higher; however, a rise in estra-diol may be a pathological increase due to an exacerba-tion of aromatase activity, which is present in metabolicand inflammatory dysfunctions, such as obesity and dia-betes. The best way to discern whether an estradiol in-crease has a physiological or pathological cause, using asingle marker, is through the T:E ratio, which is un-affected by physiological situations and reduced by aro-matase exacerbations, as in the case of an estradiolincrease, leading to a testosterone decrease. A reducedT:E ratio might be additional evidence that OTS, regard-less of its triggers, induces an anti-anabolic, dysfunc-tional, and energy-saving environment to reducetestosterone as a protective mechanism against energyexpenditure and anabolic activity by its conversion intoestradiol by the enzyme aromatase. However, the under-lying mechanisms that lead to a reduced T:E ratio inOTS are unknown. The EROS study showed that a T:Eratio should be greater than 13.7:1.0 (for total testoster-one and estradiol are expressed in ng/mL and pg/dL,respectively) [12].


The basic immunology panel was also independentlyaffected by the presence of OTS, which supports the the-ory of involvement of the immune system in the patho-physiology of OTS. Although altered immunology panels(i.e., altered when compared with healthy athletes, butsimilar to those of non-athletes) may be linked toblunted hormonal responses to stress [23, 24], the im-munology panel and the hormonal responses to stimula-tion did not exhibit linear correlations or predictions, atleast for the immunologic markers analyzed in thepresent study: neutrophils, lymphocytes, and the neutro-phil-to-lymphocyte ratio. Other mechanisms, such as anenvironment with chronic stressors leading to OTS maydirectly predict leukocyte composition [25].The relative dehydration, the decrease in muscle

mass, and the increase in visceral fat, which were in-dependently induced by OTS, may have been causedby the multiple dysfunctions associated with this syn-drome. The highly oxidative and inflammatory envir-onment that occurs in OTS might have causedincreased visceral fat without a concurrent increase inoverall body fat.The impaired mood induced by OTS may contribute

to the severe psychological effects of OTS, which aresometimes not fully recoverable. Interestingly, althoughdepression has been reported to be one of the outcomesof OTS [1, 3, 6], this parameter was not predicted byOTS. The harmful changes in both body compositionand mood also may have roles in previously unexplained

decreases in performance, which is the key and sine-quo-non characteristic of OTS.Overall, the findings of the various arms of the

EROS study led to a new understanding of the under-lying mechanisms, risk factors, and diagnosis of OTS,including its pathophysiology, as a mix of failures inthe conditioning processes that are typically observedin athletes. Our findings also showed that excessivetraining results from a combination of different trig-gers, including insufficient caloric intake, excessivephysical and concurrent cognitive effort, and poorsleep quality, instead of the traditional theory cen-tered on overtraining.We hypothesized that any type of disruption in eating,

sleep, social, or training patterns could lead to a spreadof dysfunctional reactions through multiple pathways, asa “domino effect,” leading to aberrant changes in hormo-nal, muscular, immunologic, metabolic, and/or physicalbehaviors, and ultimately, leading to OTS, if notpromptly addressed. Although not demonstrated herein,psychological dysfunctions could also play a role in thepathogenesis of OTS. The key premise of this hypothesisis that any imbalance among psychological, sleep, eating,training, or social characteristics (not only excessivetraining) may lead to OTS; this has been reported exten-sively in the different arms of the EROS study [9–13].Usually, a complex and unique combination of differ-

ent types of dysfunctions lead to OTS, suggesting thateach affected athlete should have an individual

Fig. 2 Summary of the predictions of Overtraining syndrome (OTS) and its implications


combination of parameters that are positive for OTS.Hence, OTS can be diagnosed only by using multiple in-dices, which was supported by all the study’s cases ofOTS, which may be explained only if all possible triggersare assessed, as performed in this study using logistic re-gression. We suggest that further studies on OTS shouldalways assess at least eating, training, psychological, andsocial patterns. Although we did not evaluate differentsports, the importance of each aspect as part of thepathophysiology of OTS may vary according to the typeof sport practiced. However, regardless of the type ofsport, the most important aspects of OTS impairmentare the rapid reduction in pace during long training ses-sions and reduced time-to-fatigue, which are both typic-ally found in athletes with OTS. The failure to achieveprolonged optimization of hormonal responses in OTSis likely responsible for athletes’ decreased performanceand reduced pace.The summary of the independent predictors of OTS

and its disruptions on clinical and biochemical behaviorsis illustrated in Fig. 2.

LimitationsThe EROS study only evaluated male athletes that prac-ticed either both endurance and strength modalities, orsports that demand both endurance and strength efforts.As the present study did not analyze athletes of endur-ance, strength, or explosive (“stop-and-go” sports, suchas ball games) modalities, it is uncertain whether thefindings on OTS can be replicated to these athletes, aswell as female athletes. Further studies with larger sam-ples of athletes are crucial to confirm whether our dataare reproducible; longitudinal studies are needed becausethe present study’s design precludes drawing conclusionsfrom the sequence of events in response to interventionsin modifiable patterns, including training, eating, and so-cial aspects.

ConclusionsWe identified that insufficient protein, carbohydrate, oroverall caloric intake may trigger OTS without the pres-ence of other triggers, while the combination of dietarypatterns and sleep quality may explain all cases of OTS.Once triggered, OTS leads to a failure to achieveoptimization of prolonged hormonal responses, whichmay explain reduced time-to-fatigue and decreased per-formance in long-duration sports. It also causes a reduc-tion of the T/E ratio, paradoxical worsened of bodycomposition and metabolism, failure to benefit from im-munologic adaptations observed in healthy athletes,worsening vigor, fatigue, and tension, and decreasedmuscle mass and hydration. Worse body compositionand impaired mood may also have roles in the deterior-ation of athletes’ performance, the hallmark of OTS.

AbbreviationsACTH: Adrenocorticotropic hormone; ATL: Healthy athletes; CL: Confidencelimits; EROS: Endocrine and Metabolic Responses on Overtraining Syndrome;GH: Growth hormone; ITT: Insulin tolerance test; OR: Odds ratio;OTS: Overtraining syndrome; POMS: Profile of Mood States; T:E: testosterone-to-estradiol

AcknowledgementsWe acknowledge the Medical School of the Federal University of Sao Paulo,DASA Laboratórios da América, and Corpometria Insitute.

Authors’ contributionsFAC and CEK developed the central idea of the EROS study. FAC performedthe tests of the EROS study, compilated the data, and analyzed the results.FAC and CEK actively participated in the discussion, supervised, andreviewed the results. FAC wrote the primary version of the presentmanuscript. CEK helped with the final version of the manuscript, and gavethe last word before its submission. All authors read and approved the finalmanuscript.

FundingNo funding was obtained for this study.

Availability of data and materialsThe raw data of the present study is available at https://osf.io/bhpq9/.

Ethics approval and consent to participateThis study included was approved by the ethical committee of the FederalUniversity of São Paulo, under the approval number 1093965). Allparticipants gave written consent to participate.

Consent for publicationNo individual information was used. All authors approved the submittedversion of the manuscript.

Competing interestsThe authors declare that they have no competing interests.

Received: 16 April 2019 Accepted: 22 August 2019

References1. Meeusen R, Duclos M, Foster C, et al. European College of Sport Science;

American College of Sports Medicine. Prevention, diagnosis, and treatmentof the overtraining syndrome: joint consensus statement of the EuropeanCollege of Sport Science and the American College of Sports Medicine.Med Sci Sports Exerc. 2013;45(1):186–205.

2. Kreher JB, Schwartz JB. Overtrining syndrome: a practical guide. SportsHealth. 2012;4(2):128–38.

3. Nederhof E, Zwerver J, Brink M, Meeusen R, Lemmink K. Different diagnostictools in nonfunctional overreaching. Int J Sports Med. 2008;29(7):590–7.

4. Lehmann M, Foster C, Keul J. Overtraining in endurance athletes: a briefreview. Med Sci Sports Exerc. 1993;25(7):854–62.

5. Rietjens GJ, Kuipers H, Adam JJ, et al. Physiological, biochemical andpsychological markers of strenuous training-induced fatigue. Int J SportsMed. 2005;26(1):16–26.

6. Smit PJ. Sports medicine and 'overtraining'. S Afr Med J. 1978;54(1):4.7. McTernan EJ, Leiken AM. A pyramid model of health manpower in the

1980s. J Health Polit Policy Law. 1982;6(4):739–51.8. Slivka DR, Hailes WS, Cuddy JS, Ruby BC. Effects of 21 days of

intensified training on markers of overtraining. J Strength Cond Res.2010;24(10):2604–12.

9. Cadegiani FA, Kater CE. Body composition, metabolism, sleep, psychologicaland eating patterns of overtraining syndrome: results of the EROS study(EROS-PROFILE). J Sports Sci. 2018;36(16):1902–10.

10. Cadegiani FA, Kater CE. Hypothalamic-pituitary-adrenal (HPA) axisfunctioning in overtraining syndrome: findings from endocrine andmetabolic responses on overtraining syndrome (EROS) - EROS-HPA axis.Sports Med Open. 2017;3(1):45.

11. Cadegiani FA, Kater CE. Growth hormone (GH) and prolactin responses to anon-exercise stress test in athletes with overtraining syndrome: results from


https://osf.io/bhpq9/

the endocrine and metabolic responses on overtraining syndrome (EROS) -EROS-STRESS. J Sci Med Sport. 2018;21(7):648–53.

12. Cadegiani FA, Kater CE. Basal Hormones and Biochemical Markers asPredictors of Overtraining Syndrome in Male Athletes: The EROS-BASALStudy. J Athl Train. 2019.

13. Cadegiani FA, Kater CE, Gazola M. Clinical and biochemical characteristics ofhigh-intensity functional training (HIFT) and overtraining syndrome: findingsfrom the EROS study (the EROS-HIFT). J Sports Sci. 2019;20:1–12.

14. Angeli A, Minetto M, Dovio A, Paccotti P. The overtraining syndrome inathletes: a stress-related disorder. J Endocrinol Invest. 2004;27(6):603–12.

15. Budgett R. Fatigue and underperformance in athletes: the overtrainingsyndrome. Br J Sports Med. 1998;32:107–10.

16. van Koeverden ID, de Bakker M, Haitjema S, et al. Testosterone to oestradiolratio reflects systemic and plaque inflammation and predicts futurecardiovascular events in men with severe atherosclerosis. Cardiovasc Res.2019;115(2):453–62.

17. Chan YX, Knuiman MW, Hung J, et al. Testosterone, dihydrotestosteroneand estradiol are differentially associated with carotid intima-mediathickness and the presence of carotid plaque in men with and withoutcoronary artery disease. Endocr J. 2015;62(9):777–86.

18. Colleluori G, Aguirre LE, Qualls C et al. Adipocytes ESR1 expression, Body Fatand Response to Testosterone Therapy in Hypogonadal Men VaryAccording to Estradiol Levels. Nutrients. 2018;10(9).

19. Aguirre LE, Colleluori G, Fowler KE, et al. High aromatase activity in hypogonadalmen is associated with higher spine bone mineral density, increased truncal fatand reduced lean mass. Eur J Endocrinol. 2015;173(2):167–74.

20. Xu X, Wang L, Luo D, et al. Effect of testosterone synthesis andconversion on serum testosterone levels in obese men. Horm MetabRes. 2018;50(9):661–70.

21. Xu X, Sun M, Ye J, et al. The effect of aromatase on the reproductivefunction of obese males. Horm Metab Res. 2017;49(8):572–9.

22. Rubinow KB. Estrogens and body weight regulation in men. Adv Exp MedBiol. 2017;1043:285–313.

23. Farhat K, Bodart G, Charlet-Renard C, et al. Growth hormone (GH)deficient mice with GHRH gene ablation are severely deficient invaccine and immune responses against Streptococcus pneumoniae. FrontImmunol. 2018;9:2175.

24. Bodart G, Farhat K, Renard-Charlet C, et al. The Severe Deficiency of theSomatotrope GH-Releasing Hormone/Growth Hormone/Insulin-Like GrowthFactor 1 Axis of Ghrh−/− Mice Is Associated With an Important Splenic Atrophyand Relative B Lymphopenia. Front Endocrinol (Lausanne). 2018;9:296.

25. Penz M, Kirschbaum C, Buske-Kirschbaum A, Wekenborg MK, Miller R.Stressful life events predict one-year change of leukocyte composition inperipheral blood. Psychoneuroendocrinol. 2018;94:17–24.

Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.


novel causes and consequences of overtraining syndrome ......these novel findings may explain...

Documents