This is a sel -a chi ed e sion o an o iginal a icle. This e sion
may di e om he o iginal in pagina ion and ypog aphic de ails.
Au ho (s):
Ti le:
Yea :
Ve sion:
Copy igh :
Righ s:
Righ s u l:
Please ci e he o iginal e sion:
CC BY-NC-ND 4.0
h ps://c ea i ecommons.o g/licenses/by-nc-nd/4.0/
Rela ionships be ween Hea Ra e Va iabili y, Sleep Du a ion, Co isol and Physical
T aining in Young A hle es
© 2021 Jou nal o Spo s Science and Medicine
Published e sion
Mishica, Ch is ina; Ky öläinen, Heikki; Hynynen, Esa; Nummela, A i; Holmbe g,
Hans-Ch is e ; Linnamo, Vesa
Mishica, C., Ky öläinen, H., Hynynen, E., Nummela, A., Holmbe g, H.-C., & Linnamo, V. (2021).
Rela ionships be ween Hea Ra e Va iabili y, Sleep Du a ion, Co isol and Physical T aining in
Young A hle es. Jou nal o Spo s Science and Medicine, 20(4), 778-788.
h ps://doi.o g/10.52082/jssm.2021.778
2021
©Jou nal o Spo s Science and Medicine (2021) 20, 778-788
h p://www.jssm.o g DOI: h ps://doi.o g/10.52082/jssm.2021.778
Recei ed: 03 Feb ua y 2021 / Accep ed: 20 Sep embe 2021 / Published (online): 30 Sep embe 2021
`
Rela ionships be ween Hea Ra e Va iabili y, Sleep Du a ion, Co isol and
Physical T aining in Young A hle es
Ch is ina Mishica 1, Heikki Ky öläinen 1, Esa Hynynen 2, A i Nummela 2, Hans-Ch is e Holmbe g 3
and Vesa Linnamo 1
1 Facul y o Spo and Heal h Sciences, Uni e si y o Jy äskylä, Jy äskylä, Finland; 2 KIHU – Resea ch Ins i u e o
Olympic Spo s, Jy äskylä, Finland; 3 Depa men o Heal h, Educa ion and Technology, Luleä Uni e si y o
Technology, Luleå Sweden
Abs ac
The aims o he cu en s udy we e o examine he ela ionships
be ween hea a e a iabili y (HRV), sali a y co isol, sleep
du a ion and aining in young a hle es. Eigh a hle es (16 ± 1
yea s) we e moni o ed o 7 weeks du ing aining and
compe i ion seasons. Subjec s we e aining o endu ance-based
win e spo s (c oss-coun y skiing and bia hlon). T aining was
di ided in o wo zones (K1, easy aining and K2, ha d aining).
Hea a e and blood lac a e du ing submaximal unning es s
(SRT), as well as co isol, sleep du a ion and noc u nal HRV
(RMSSD), we e de e mined e e y o he week. HRV and co isol
le els we e co ela ed h oughou he 7-week pe iod ( = -0.552,
P = 0.01), wi h he s onges co ela ion du ing week 7 ( = -0.879,
P = 0.01). The ela i e changes in K1 and HRV showed a posi i e
co ela ion om weeks 1-3 ( = 0.863, P = 0.006) and a nega i e
co ela ion du ing weeks 3-5 ( = -0.760, P = 0.029). The ela i e
change in sleep du ing weeks 1-3 we e nega i ely co ela ed wi h
co isol ( = -0.762, P = 0.028) and K2 ( = -0.762, P = 0.028). In
conclusion, HRV appea s o e lec he eco e y o young a hle es
du ing high loads o physical and/o physiological s ess. Co isol
le els also e lec ed his eco e y, bu signi ican change equi ed
a longe pe iod han HRV, sugges ing ha co isol may be less
sensi i e o s ess han HRV. Mo eo e , ou esul s indica ed ha
du ing he compe i ion season, eco e y o young endu ance
a hle es inc eased in du a ion and addi ional sleep may be
bene icial.
Key wo ds: Reco e y, endu ance aining, physiological s ess,
indi idual adap a ions, submaximal es s, au onomic ne ous
sys em.
In oduc ion
Du ing he pas ou decades, inc easing emphasis has been
placed on aining young a hle es, many o whom now ain
all yea ound (B enne , 2016). A he same ime, mos
in es iga ions on esponses o endu ance aining ha e
in ol ed adul subjec s and conside ably less is known
abou he ainabili y and de elopmen o younge
indi iduals (Naugh on e al., 2000; Mu ay, 2017).
Mo eo e , in addi ion o he s ess o daily aining, he
added s ess o hei s udies a ec s he eco e y o young
a hle es. The e o e, esea ch on he aining o young
a hle es, ocused on eaching a high le el o pe o mance,
is necessa y o a aining maximal gains and allowing
young a hle es o succeed in eli e spo s (B enne , 2016;
Mu ay, 2017).
Physiological p ocesses ha occu du ing sleep a e
a undamen al aspec o an a hle e’s eco e y and
subsequen abili y o ain and compe e a maximal
capaci y (Samuels, 2008; B and and Ki o , 2011).
Howe e , bo h he quali y and quan i y o sleep by young
a hle es has been declining (Samuels, 2008; Copenha e
and Diamond, 2017) o a numbe o easons, including
aining schedules, educa ion, social e en s and a el
plans (Copenha en and Diamond, 2017; Simpson e al.,
2017) as well as ch onic and acu e s ess. I is known ha
he human esponse o s ess is la gely egula ed by he
au onomic ne ous sys em (ANS) and, he e o e, can be
obse ed easily and non-in asi ely by measu ing bea - o-
bea a ia ion in es ing hea a es, also known as hea
a e a iabili y (HRV) (Elec ophysiology, 1996; McEwen,
2007). Nume ous s udies ha e in es iga ed he e ec s o
endu ance aining on HRV (Picho e al., 2000; Hau ala e
al., 2001; Ca e e al., 2003; Ki iniemi e al., 2007) and
ecen s udies sugges ha noc u nal eco dings help
u he e alua e an indi idual’s accumula ed aining load
(Picho e al., 2000; Hynynen e al., 2010). HRV
measu emen s du ing sleep p o ide a measu emen ha is
independen o ex e nal ac o s and he e o e, enhances
hei eliabili y (Picho e al., 2000; Buchhei e al., 2004;
Nummela e al., 2010). Fu he mo e, noc u nal HRV
appea s o ha e a dose- esponse ela ionship wi h
inc eased exe cise in ensi y causing a educ ion in
noc u nal HRV (Hynynen e al., 2010). Thus, sleep
du a ion and noc u nal HRV measu emen s may be
e ec i e measu es o moni o eco e y in young a hle es.
Co isol is one o he mos equen ly in es iga ed
ho mones as a measu e o o e aining and s ess.
Ex ended pe iods o inc eased o dec eased le els o
co isol ha e a nega i e impac on heal h and he e o e,
may hinde a hle ic pe o mance (Duclos e al., 2007). The
ela ionship be ween co isol and exe cise as well as he
di e en me hods used o measu ing co isol sec e ion
ha e con lic ing esul s (Nea y e al., 2002; Duclos e al.,
2007). Howe e , se e al s udies ha e shown a s ong
ela ionship be ween se um and sali a y co isol le els
indica ing ha sali a y co isol is a eliable measu emen
me hod, a good bioma ke o physiological s ess and a
non-in asi e op ion o moni o ing a hle es (Nea y e al.,
2002; Gus a sson e al., 2008; Hellhamme e al., 2009).
Du ing a 37-week ollow up, sali a y co isol inc eased
wi h inc eased aining bu when aining was educed, no
change was obse ed and no ela ionship was ound
be ween pe o mances (Cha a d e al., 2002). In addi ion, a
Resea ch a icle
Mishica e al.
779
epea ed exe cise p esc ip ion does no appea o elici he
same changes in es ing le els o co isol wi h p e ious
esea ch inding ha co isol dec eased conside ably
be ween wo di e en maximal es s (Hedelin e al., 2000).
Maximal es s a e a good measu e o pe o mance bu
equi e a highly in ensi e exe ion, educing hei
applica ion o e e yday aining (Lambe s e al., 2011;
Capos agno e al., 2016). As a esul , submaximal es s a e
used mo e equen ly o moni o a hle es and p edic
pe o mance (Lambe s e al., 2011). Submaximal
eadmill es s ha e been compa ed o cycle es s in a hle es
aining o ia hlon, and indings showed ha bo h es ing
modes could be used in e changeably (Basse and Boulay,
2003). Running is a common aining mode o bo h c oss-
coun y skie s and bia hle es, allowing SRT o be a alid
es o moni o aining in his s udy.
Al hough moni o ing aining load a he eli e le el
is common p ac ice, he e is no well-de ined bounda y
be ween e ec i e and ine ec i e aining adap a ions
(Gus a sson e al., 2008). P e ious esea ch has shown ha
he u iliza ion o HRV measu emen s in spo is
challenging due o inconsis en p ocedu es making
compa ison o esul s p oblema ic as well as ime
cons ain s educing he o e all a hle e compliance,
especially o e a long season (Ra e e al., 2018). Inc eases
in echnology ha e in oduced moni o ing op ions ha a e
easily accessible and collec ed wi h minimal e o .
Howe e , hese measu es a e o en pe o med in home
en i onmen s. The e o e, in es iga ing “ eal-li e” alues is
highly ele an and may p o ide u u e unde s anding ha
is highly applicable o a hle es and coaches. Acco dingly,
his s udy was designed o cha ac e ize he ela ionships
be ween HRV, sali a y le els o co isol, sleep du a ion,
and blood lac a e du ing submaximal unning es s (SRT)
in young a hle es du ing hei aining and compe i ion
seasons. Ou main hypo hesis was ha noc u nal HRV
exhibi s a nega i e ela ionship o sali a y co isol le els
in he mo ning. We also hypo hesized ha a dec ease in
sleep du a ion alone o in combina ion wi h mo e in ense
and p olonged aining educes noc u nal HRV and
ele a es mo ning co isol le els.
Me hods
Pa icipan s
Eigh well- ained young endu ance a hle es pa icipa ed in
his s udy. The pa icipan s we e all a hle es a a spo s
academy high school compe ing and aining o c oss-
coun y skiing (6 subjec s) o bia hlon (2 subjec s) yea
ound and pa icipa ing a he na ional le el.
Cha ac e is ics o he a hle es a e p esen ed in Table 1. All
subjec s we e ully in o med o he s udy p ocedu es and
ga e w i en consen o pa icipa e in he p ojec . The e hics
commi ee o he Uni e si y o Jy äskylä, Finland,
app o ed he s udy and he measu emen s we e pe o med
in acco dance wi h he decla a ion o Helsinki.
S udy Design
This s udy was pe o med du ing a 7-week pe iod
(No embe -Decembe ) ha in ol ed bo h a aining (T1)
and ea ly compe i ion (T2) aining phase. ANS s a e was
assessed wi h noc u nal HRV analysis, collec ed using a
ballis ioca diog aphic (BCG) sleep- acking de ice (Em i
QS, Jy äskylä, Finland). Addi ional assessmen es s
occu ed on ou sepa a e occasions: wice du ing he
aining season (weeks 1, 3; T1) and wice du ing he
compe i ion season (weeks 5, 7; T2). Du ing each es
week, subjec s pa icipa ed in SRT and sali a samples we e
collec ed o h ee consecu i e days; one day be o e SRT,
he es day and one day a e SRT. A hle es eco ded hei
own indi idual aining plans du ing his ime and aining
cha ac e is ics (easy aining, ha d aining and aining
load) we e e alua ed ia elec onic aining dia ies. Body
a pe cen age was measu ed using he bioimpedance
me hod (InBody 720, Inbody CO., Ce i os, Cali o nia,
USA) in he beginning o he aining pe iod.
Table 1. Cha ac e is ics o he subjec s (means ± SD).
Women (n = 5) Men (n = 3)
Age (y s) 16 ± 1 16 ± 1
Body mass (kg) 58 ± 5 67 ± 5
Body a (%)a16.2 ± 8.8 14.5 ± 11.1
T aining hou s (y)b500 ± 76 600 ± 71
a Assessed on he basis o bioimpedence measu emen s.
b eco ded in elec onic aining dia ies.
HRV and Sleep Analysis
P e ious esea ch has demons a ed ha de e mina ion o
HRV on he basis o BCG is bo h accu a e and eliable
(Shin e al., 2011; Wang e al., 2015). The Em i QS de ice
(EMFIT QS, Em i OY, Jy äskylä, Finland) consis s o a
con ac less p essu e senso (542mm x 70mm x 1.4mm) ha
u ilizes BCG o in e p e epea ed mo emen s o he human
body, such as hea bea , by ep esen ing hem g aphically
(Pinhei o e al., 2010). E alua ion o his de ice du ing one
nigh o sleep unde eal-li e condi ions e ealed good
ag eemen wi h he measu emen s p o ided by a e e ence
de ice ha employs elec oca diog aphy and has been
alida ed in labo a o y s udies, wi h only e y mino
di e ences in he mean HR and HRV alues ob ained
(Ves e inen e al., 2020). Thus, al hough he eliabili y o
his de ice o moni o ing hese pa ame e s du ing sleep
has ye o be es ablished, i would appea o p o ide a
simple and e ec i e ool o au oma ic daily analysis o
HRV.
The e o e, he ime spen sleeping, na u e o he
sleep, and associa ed HR and HRV we e moni o ed wi h
an Em i QS de ice o 7 weeks he e. To minimize he
dis ance o he hea and he eby maximize signal quali y,
his de ice was placed unde he ma ess nea he ches in
a manne such ha he subjec was unawa e o i s p esence.
The de ice began o eco d au oma ically a a sampling a e
o 100 Hz when i sensed body weigh and con inued
h oughou he nigh , s opping when he subjec go ou o
bed in he mo ning.
This moni o ing o noc u nal HRV and HR was
collec ed in con inuous 3-minu e pe iods and da a om
pe iods in which he signal was poo and/o dis up ed was
excluded. The magni ude o he HRV is exp essed ela i e
o ime, u ilizing he oo -mean-squa ed di e ence
be ween successi e RR in e als (RMSSD, ms). Fo his
pu pose, he a e age RMSSD o each 3-minu e pe iod was
calcula ed and hese a e ages used o isualize he
Hea Ra e Va iabili y and T aining
780
noc u nal HRV alues g aphically (Figu e 1). The
endpoin s o he bes linea i o each nigh we e
conside ed o be he a e age RMSSD alues o e ening
and mo ning sleep and he la e aken o be he HRV
RMSSD alue (HRV
NOC) du ing ha nigh o sleep.
Al hough p e iously he a e age 3-minu e alues o he
en i e nigh ha e been used o calcula e his alue
(Ves e inen e al., 2020), he cu en in es iga ion ocused
on an indi idual’s cu en s a e o eco e y and eadiness
o ain, so he mo ning alue was conside ed o be mo e
ele an . Fo moni o ing sleep pa e ns ou side he
labo a o y, w is ac ig aphy is he app oach mos widely
used and bes alida ed (Van De Wa e e al., 2011), bu , a
he same ime, de ices inco po a ed in o he bed a e highly
con enien . These elimina e he need o a achmen o
elec odes o senso s o he body, p o iding a aluable
op ion o longe - e m moni o ing o sleep a home. Such
de ices iden i y he di e en classes o sleep, as well as
pe iods o wake ulness, wi h good accu acy (Yi e al.,
2019).
Figu e 1. A ep esen a i e example o he 3-minu e sampling
o he oo mean squa e o successi e di e ences be ween RR
in e als (RMSSD) o hea a e a iabili y (HRV) du ing
sleep. The line shows he bes i used o calcula e noc u nal
HRV (HRVNOC).
The sleep du a ion and HRVNOC alues analyzed
he e a e p esen ed au oma ically on he use in e ace o he
Em i QS de ice, p o iding easy daily access o bo h
coaches and a hle es. These alues we e also calcula ed as
3-day a e ages o compa ison wi h co isol alues and he
pe iods o SRT es ing.
Co isol analysis
Co isol le els we e analyzed om mo ning sali a
samples. A he beginning o he es ing pe iod, all subjec s
we e ins uc ed on how o handle and collec he sali a
samples. Collec ion occu ed immedia ely a e waking up
be o e ea ing, d inking o b ushing ee h. The passi e
d ool-me hod was used o sali a collec ion and small cups
we e p o ided so ha subjec s could ake 100 mL o wa e
o wash ou hei mou h be o e collec ing he sample.
Subjec s we e asked o p o ide a leas 3 mL o sali a o
each sample as well as eco d he da e, he ime o day and
how long i ook o comple e he p ocedu e. Sali a samples
we e aken on 3 consecu i e days, allowing o a sample
he day be o e, he day o and one day a e he SRT
measu emen s occu ed. Sali a samples we e immedia ely
Placed in o subjec s’ eeze s and collec ed e e y o he
week h oughou he es ing pe iod.
Sali a samples we e analyzed using he chemilumi-
nescence me hod wi h he IMMULITE 2000 XPi Analyze
(Siemens Heal hca e Diagnos ics P oduc s L d., Glyn
Rhonwy, Llanbe is, UK). The sensi i i y o he sali a
assay o co isol was 5.5 nmol/l wi h in e -assay p ecision
8.2 % a 12.5 nmol/l. This me hod p o ides a nonin asi e,
easily epea able and p ac ical way o assess he co isol
esponse. Sali a co isol alues we e analyzed in 3-day
a e age alues o coincide wi h he HRV alues and SRT
es ing pe iods.
T aining analysis
Indi idual aining plans we e ollowed h oughou he
es ing pe iod. Subjec s we e asked o w i e down all
aining sessions daily including he in ensi y, du a ion and
exe cise mode in hei elec onic aining dia ies
(elogge .ne , Espoo, Finland). Subjec s had pa icipa ed in
p io maximal g aded exe cise es s ha p o ided
indi idually de e mined hea a e zones o guide aining
in ensi y on a daily basis wi h hei indi idual hea a e
moni o s. Endu ance aining in ensi ies we e based on a 5-
zone aining dis ibu ion wi h zone 1 and 2 ep esen ing
basic aining (es ima ed: ≤ 2 mM blood lac a e) and zone
3-5 ep esen ing all high in ensi y aining. T aining was
analyzed acco ding o he elec onic aining dia ies. Easy
aining (zone 1 and 2) was de ined as all aining below
ae obic h eshold (K1, minu es) and ha d aining (zone 3-
5) was de ined as all aining abo e ae obic h eshold (K2).
Weekly dis ibu ion o speci ic aining modes o K1 and
K2 aining du ing he 7-week es ing pe iod a e p esen ed
in Figu e 2. T aining load was quan i ied using a modi ied
e sion o Lucia’s simpli ied TRIMP sys em (An a and
Es e e-Lanao, 2011). T aining load was calcula ed wi h he
ollowing equa ion:
TL = 1 x K1 + 2.5 x K2
TL = aining load, K1 = ime aining unde ae obic h eshold,
K2 = ime aining a /abo e ae obic h eshold. T aining alues
we e exp essed in minu es. The aining load, K1 and K2 we e
calcula ed o de e mine weekly alues and h ee-day a e age
alues when STR and co isol measu emen s occu ed.
Submaximal Running Tes
The SRT was 16-minu es in leng h and included 4 s ages.
Tes s we e pe o med on a Tun u i GO Run 50 T eadmill
(Tun u i Fi ness, Fle oland, Ne he lands). The SRT was
s anda dized o speed (women: 10 km/h, men: 11.7 km/h)
wi h inclina ion inc easing e e y 4 minu es, s a ing a 2%,
hen 4%, 7%, and 9%. One amilia iza ion SRT was
conduc ed so ha all subjec s we e amilia wi h he es
p o ocol. The SRT used in his s udy was designed o
junio c oss-coun y skie s and bia hle es. Al hough we do
no ha e alida ion o his p o ocol, i is a classic me hod
used as a con ol es by a hle es and coaches in Finland.
Due o he homogenous g oup, he s anda dized p o ocol
was app op ia e and submaximal in ensi ies we e eached
o all subjec s. Hea a e (HR) was moni o ed wi h a HR-
moni o (Pola V800, Pola Elec o Oy, Kempele, Finland)
and HR alues we e eco ded when 15 s o each load
emained. Subjec s b ie ly s opped unning and blood
samples (20 µL) we e aken om he inge ip e e y 4
minu es o de e mine blood lac a e concen a ions (Biosen
C_line Lac a e Analyze . EKF Diagnos ic, Magdebu g,
Mishica e al.
781
Ge many). Sample collec ion ime (app ox. 15 s) was
included in he 4 minu es o upcoming s age.
S a is ical Analysis
All s a is ical analyses we e pe o med in he SPSS o
Windows so wa e (IBM SPSS S a is ics 24 (SPSS, Inc.,
Chicago, IL, USA)). Since he numbe o subjec s was
small, F iedman’s non-pa ame ic es o ela ed samples
was applied o analyze changes in aining, HRV and
sali a y co isol le els. Pos -hoc analyses we e pe o med
wi h he Wilcox signed ank es . Spea man’s co ela ion
coe icien was used o de e mine he ela ionship be ween
3-day a e age HRV and co isol le els, as well as be ween
HRV, sleep du a ion, co isol le els and aining
cha ac e is ics. In addi ion, he ela i e changes in HRV,
co isol le els, sleep, SRT and aining cha ac e is ics om
week o week we e also in es iga ed using Spea man’s
co ela ion coe icien . All alues o HRV, sleep du a ion
and co isol le el u ilized we e 3-day a e ages, whe eas
weekly a e ages we e employed in he case o SRT and
aining cha ac e is ics. All alues shown a e means ± SD
and s a is ical signi icance de ined as p < 0.05.
Figu e 3. A e age indi idual 3-day alues o he oo mean squa e o successi e di e ences be ween RR in e als (RMSSD)
and mo ning sali a y le els o co isol a di e en ime-poin s du ing he 7-week s udy.
Figu e 2. G aphs A and B ep esen he weekly dis ibu ion
o exe cise aining modes o easy (K1) and ha d (K2)
aining du ing all weeks o es ing.
Resul s
Du ing all weeks o es ing, HRV and sali a y le els o
co isol we e in e sely ela ed. Al hough he in e -
indi idual di e ences we e p onounced (Figu e 3), he
highes HRV and lowes le el o co isol we e obse ed
du ing week 5.
Rela ionships be ween HRV, Co isol le els, sleep and
aining
A e age alues o HRV, le els o co isol, sleep and
weekly aining alues a e shown in Table 2. Weekly
co ela ions be ween hese pa ame e s a e shown in Table
3. Figu e 4 is a sca e plo o he ela ionship be ween
HRV and co isol du ing all weeks o es ing combined.
Figu e 5 shows a e age daily alues o HRV and co isol
h oughou he es ing pe iod.
As can be seen, he mos p onounced co ela ion
be ween HRV and mo ning co isol le els was obse ed
du ing week 1 and week 7 ( = -0.833, p < 0.05) (Table 3).
The HRV alues we e signi ican ly di e en be ween week
1 and week 3 (p < 0.05) and be ween week 5 and week 7
(p < 0.05). In addi ion, sleep and noc u nal sleep HR we e
bo h signi ican ly di e en be ween week 1 and week 7 (p
< 0.05) (Table 2). Di e ences be ween es weeks o
co isol and aining alues we e no signi ican .
Th oughou he aining pe iod (T1, weeks 1-3), a
nega i e co ela ion was obse ed o ela i e changes in
co isol le els and sleep ( = -0.762, p < 0.05). The same
ela ionship was demons a ed o K2 and sleep (week 1-3,
= -0.762, p < 0.05). Du ing he shi om he aining o
he ea ly compe i ion season (week 3-5), co isol displayed
a posi i e ela ionship wi h K2 ( = 0.810, p < 0.05). When
looking a HRV, a posi i e ela ionship was displayed o
ela i e changes om week 5 o week 7 wi h sleep ( =
0.786, p < 05) and a nega i e ela ionship was obse ed o
di e ences be ween weeks 1-7 ( = -0.714, p < 0.05) wi h
sleep HR. In addi ion, he ela i e changes in K2 and HRV

Hea Ra e Va iabili y and T aining
782
showed a posi i e co ela ion weeks 3-7 ( = 0.736, p <
0.05) and he same pa e n was obse ed o ela i e
changes in HRV and TL (week 3-7, = 0.810, p < 0.05).
Associa ions be ween co isol le els, aining, and hea
a e and le els o blood lac a e du ing Submaximal
Running Tes s (SRT)
Co isol’s esponse o changes in SRT a ied h oughou
he es ing pe iod. Rela i e di e ences in week 3-5
demons a ed a nega i e ela ionship o bo h HR ( = -
0.929, p < 0.01) and blood lac a e ( = -0.857, p < 0.05).
Howe e , du ing he end o he es ing pe iod (week 5-7),
he di e ences be ween co isol le els displayed a posi i e
ela ionship wi h he changes in he SRT HR alues ( =
0.929, p < 0.01). In addi ion, a nega i e ela ionship
be ween blood lac a e and co isol was obse ed du ing
week 7 ( = -0.714, p < 0.05).
Moni o ing o he dis ibu ion o aining e ealed
ha olume o K2 was he highes du ing week 3, while K1
and aining load we e bo h g ea es du ing week 5, i.e., a
he beginning o he compe i ion season (Table 2).
Dis ibu ion o aining modes o K1 and K2 h oughou
he es ing pe iod can be iewed in Figu e 2.
Table 2. HRV, mo ning sali a y le els o co isol, noc u nal hea a e and sleep, amoun s o ha d and easy
aining, o e all aining load, and blood le el o lac a e and hea a e du ing submaximal unning es s (SRT)
a di e en ime-poin s du ing he 7-week s udy pe iod (means ± SD).
T1, week 1 T1, week 3 T2, week 5 T2, week 7
RMSSD (ms)a 66.9 ± 17.8 75.0 ± 17.0* 75.6 ± 28.7 67.0 ± 24.8*
Co isol (µg/L)a6.7 ± 2.5 6.3 ± 3.1 5.6 ± 2.0 7.0 ± 2.4
Ha d aining (min)b 44 ± 30 64 ± 28 45 ± 17 53 ± 14*
Easy aining (min)b 508 ± 190 555 ± 151 743 ± 251 577 ± 136
T aining loadb616 ± 246 714 ± 209 856 ± 283 710 ± 132*
Blood lac a e (mmol/L)c 3.8 ± 1.0 3.6 ± 0.6 3.5 ± 0.6 3.7 ± 1.1
Hea a e (bpm)c 183 ± 6 184 ± 8 180 ± 9 182 ± 9
Sleep (min)a 482 ± 29** 513 ± 36 513 ± 52 527 ± 44
Noc u nal hea a e (bpm)a 58 ± 5** 55 ± 4 55 ± 3 54 ± 5
a A e age 3-day alues o he noc u nal oo mean squa e o successi e di e ences be ween RR in e als (RMSSD),
noc u nal hea a e, nigh sleep and sali a le els o co isol. b Weekly alues. c A e age du ing he inal s age o he
SRT. * p < 0.05 compa ed o he p e ious week. ** p < 0.05 when compa ed o week 7.
Table 3. Rela ionships (Spea man’s co ela ions) be ween HRVNOC and mo ning sali a y le el o co isol, he amoun s
o ha d and easy aining, and o e all aining load du ing he di e en weeks o he s udy.
Week Co isol (µg/L) Ha d aining (min) Easy aining (min) T aining load
1 -0.833* -0.287 -0.071 -0.095
2 -0.238 -0.036 -0.048 -0.132
3 -0.524 0.024 -0.238 -0.143
4 -0.833* -0.571 -0.264 -0.143
* p < 0.05
Figu e 4. G aphs A and B ep esen he weekly dis ibu ion o exe cise aining modes o easy
(K1) and ha d (K2) aining du ing all weeks o es ing.
Discussion
The majo indings o he cu en in es iga ion we e as
ollows: 1) HRVNOC and sali a y co isol co ela ed
signi ican ly du ing es weeks ha displayed he lowes
HRVNOC and highes sali a y co isol le els. 2) The decline
in HRV om week 3 o week 7 was co ela ed wi h a
educed olume o in ense aining/ aining load,
sugges ing ha his decline was no due o aining s ess,
bu a he he inc ease in a igue/s ess associa ed wi h he
beginning o he compe i ion pe iod. 3) When he a hle es
we e ocused on aining (weeks 1-3), he change in sleep
du a ion was nega i ely co ela ed o bo h K2 and sali a y
co isol le els, indica ing ha a educ ion in he amoun o
sleep may be associa ed wi h ele a ed weekly s ain.
Finally, 4) al hough he amoun o easy aining and
aining load inc eased, a educ ion in he olume o ha d
aining is e lec ed in sali a y co isol le els wi h a
Mishica e al.
783
Figu e 5. Daily alues o he a e age oo mean squa e o successi e di e ences be ween RR in e als (RMSSD,
ms) and mo ning sali a y le el o co isol du ing he 7-week es ing pe iod.
posi i e ela ionship be ween changes in K2 and co isol
du ing weeks 3-5. Thus, young endu ance a hle es appea
o handle la ge amoun s o easy aining when he olume
o ha d aining is educed.
Du ing he 7-week s udy pe iod, HRVNOC and
co isol displayed a nega i e ela ionship (Figu e 4). When
obse ing each es week sepa a ely, we ound s ong
nega i e co ela ions be ween HRVNOC and co isol du ing
week 1 and 7 and mode a e nega i e co ela ions o all
emaining es weeks (Table 3). This indica es ha co isol
and HRVNOC ha e a nega i e ela ionship o young
endu ance a hle es.
The e alua ion o h ee-day noc u nal RMSSD o
he HRV analysis used in he cu en s udy is based on
p e ious indings. Al hough day ime eco dings a e
commonly used, i has been sugges ed ha nigh eadings
enhance eliabili y since ex e nal ac o s a e no longe
a ec ing an indi idual du ing sleep (Picho e al., 2000;
Buchhei e al., 2004; Nummela e al., 2010). Ea lie
esea ch has ound ha he ime domain a iable (RMSSD)
has shown simila eco e y imes and changes o he
commonly used high- equency a iables o HRV (Hau ala
e al., 2001; Ca e e al., 2003), indica ing RMSSD
e ec i ely e alua es change in au onomic egula ion.
Addi ionally, RMSSD measu es ha e shown a high
co ela ion o he high- equency a iabili y (O zenbe ge
e al., 1998; Esco e al., 2018) wi h a ious b ea hing
equencies ha ing minimal e ec s on RMSSD alues
(Elec ophysiology, 1996). Mo eo e , simila
ballis oca diog aphic-based measu es o noc u nal HRV
ha e been compa ed o elec oca diog aphic measu es and
esul s showed ha bo h HRV and HR da a ag eed wi h
elec oca diog aphy da a ha was also es ed in eal-li e
condi ions (Ves e inen e al., 2020). These indings suppo
he idea ha HRVNOC may be a use ul me hod o e alua e
indi idual esponse o endu ance aining. Daily (Hau ala
e al., 2001; Ca e e al., 2003; Ki iniemi e al., 2007;
Hynynen e al., 2010; He zig e al., 2017) and weekly HRV
(Picho e al., 2000; Nummela e al., 2010) alues, which
a e bo h esponsi e o aining, a e commonly analyzed.
Howe e , since daily measu emen s can be in luenced by
p onounced diu nal a ia ions, weekly a e ages may
p o ide a be e indica ion o adap a ion o aining and a e
ecommended o use (Plews e al., 2013). A he same
ime, u ilizing he combined alues o noc u nal and
mo ning HRV o e se e al days ins ead o weekly alues
p o ides a be e measu e o apid esponses by he
au onomic ne ous sys em (Nuu ila e al., 2017).
Addi ionally, al hough se e al s udies ound no sex
di e ences in es ing HRV alues esponse o aining o
c oss-coun y skie s (Hedelin e al., 2000, Schä e e al.,
2015), he a e aging o se e al HRV alues may help
diminish he e ec o indi idual con ounde s, such as
gende , on HRV (Schä e e al., 2015). The e o e, we
analyzed he noc u nal mo ning RMSSD o h ee
successi e days, an app oach ha does no equi e special
so wa e and in ol es calcula ions ha can be made easily,
allowing i s use no only in he labo a o y, bu in eal li e
as well (Hynynen e al., 2010).
One impo an ac o ha may in luence he daily
changes in HRV alues as well as he cu en s a e o
eco e y is sleep (Shina e al., 2006). Sleep and o e all
le els o a igue a e highly in e connec ed, and sleep
appea s o ha e a c i ical ole in he daily unc ioning
du ing he adolescen yea s (B and and Ki o , 2011). Sleep
du a ion is a equen ly and easily in es iga ed measu e o
o e all heal h and ecommenda ions sugges ha
adolescen s (13-18 yea s o age) should ob ain 8-10 hou s
o sleep each nigh (Pa u hi e al., 2016). In addi ion,
a hle es a e ad ised o ob ain addi ional sleep and ample
esea ch has epo ed he de imen al e ec s o sleep loss
on human pe o mance; demons a ing sleep is a aluable
ac o o obse e in a hle es (Fullaga e al., 2015; Simpson
e al., 2017). Al hough moni o ed each nigh , sleep was no
Hea Ra e Va iabili y and T aining
784
con olled du ing his s udy. The e o e, pe sonal
commi men s (i.e. socializing, s udying) and indi idual
sleeping habi s likely in luenced he ela ionships be ween
sleep and o he in es iga ed a iables.
P e ious esea ch has disco e ed ha e en eli e
a hle es a e o en unable o ob ain he ecommended
amoun o sleep (Robe s e al., 2019) wi h a ecen e iew
inding g ea e de iciencies in a hle es’ sleep du ing
compe i ion pe iods (O’Donnell e al., 2018). Ou indings
di e om his endency, wi h he g ea es olume o sleep
occu ing du ing he compe i ion pe iod (Table 2).
Addi ionally, he sleep du a ion du ing his s udy
consis en ly emained wi hin he ecommended 8-10 hou s.
Al hough he quali y o sleep was no moni o ed, p e ious
esea ch wi h young gymnas s ound he o e all quali y o
sleep was una ec ed du ing a compe i ion pe iod (Sa o e
al., 2017). The e o e, he posi i e co ela ion be ween
changes in sleep and HRVNOC obse ed du ing he
compe i ion pe iod ( = 0.786) indica es ha indi iduals
who ob ained mo e sleep may ha e also expe ienced an
enhanced eco e y.
Du ing his 7-week aining pe iod, aining
cha ac e is ics had a mixed e ec on HRVNOC . P e ious
esea ch has ound con lic ing esponses wi h inc eases,
dec eases and no changes in HRV all occu ing wi h an
inc eased aining load (Picho e al., 2000; Hau ala e al.,
2001; Ca e e al., 2003; Hynynen e al., 2010). I is
e iden ha he exac mechanisms behind he e ec s o
endu ance aining on HRV a e no well-de ined (He zig e
al., 2017). Mode a e amoun s o exe cise ha e been shown
o enhance agal- ela ed HRV indexes (Buchhei e al.,
2004). Thus, he inc ease in HRVNOC om week 1 o week
3, when physical aining inc eased, suppo s his p e ious
inding. Mo eo e , HRVNOC was he highes , du ing week
5, when aining load also eached i s highes alues,
indica ing he e was a good ole ance o he p esen aining
s imulus. Howe e , we ound ha he changes om week
3 o week 7 showed a posi i e co ela ion wi h dec eases
in olume o ha d aining, aining load and HRV.
P e ious esea ch ound a dec ease in noc u nal HRV
alues a e bo h mode a e and hea y endu ance aining
sessions (Hynynen e al., 2010). Addi ionally, a
p og essi e dec ease in HRV alues we e ound ollowing
a 3-week pe iod o in ensi e aining (Picho e al., 2000).
In he p esen s udy, he olume o ha d aining and
aining load we e educed, indica ing ha he dec ease in
HRVNOC was no associa ed o cu en aining induced
s ess. Since he pe iod o aining analyzed in his s udy
occu ed du ing he ini ial phase o compe i ion, all
subjec s also had he common goal o p epa ing o he
ea ly s age o hei compe i ion season. The e o e, as a
g oup, weekly aining ollowed simila and expec ed
aining p og ams wi h an in en ional inc ease in ski-
speci ic aining h oughou he s udy o educe aining
s ess as compe i ions app oached. Simila o physical
s ess (physical wo k, a igue, die a y s ess), physiological
s ess (emo ional, anxie y, cogni i e), such as an icipa o y
s ess, may ha e inc eased du ing he compe i ion phase
due o ace-induced p essu e and men al p epa a ions ha
occu p io o compe i ion. An independen measu e o
anxie y was no included in his s udy; as a esul , al hough
i appea s, i is ha d o iden i y i he al e a ions in HRVNOC
and co isol we e induced by compe i ion s ess.
Ne e heless, p e ious esea ch has shown ha acu e
s ess, due o an an icipa o y ask, had an impac on HRV
du ing sleep and was associa ed wi h a dec eased
pa asympa he ic modula ion, and he e o e, esul ed in
lowe HRV alues (Hall e al., 2004), u he suppo ing
ou cu en inding.
Sali a y co isol le els a e equen ly used as a
bioma ke o psychological s ess and ha e shown a
mode a e associa ion o pe cei ed s ess (Hellhamme e
al., 2009). P e ious li e a u e has in es iga ed sali a y
co isol le el’s esponse o exe cise and ound co isol only
signi ican ly inc eased a e high-in ensi y exe cise wi h no
changes occu ing a e low and mode a e exe cise
(VanB uggen e al., 2011). Addi ionally, when
in es iga ing co isol le els in o e - ained and con ol
a hle es no signi ican di e ences be ween g oups we e
ound (Hynynen e al., 2006). Fu he mo e, bo h baseline
and esponse o aining co isol le els a e in luenced by
gene ics (Fei osa e al., 2002) so indi idual a ia ion has an
added e ec on co isol alues. In he p esen s udy,
changes du ing week 3 o week 5 showed ha sali a y
co isol appea ed o espond o he educed olume o ha d
aining by demons a ing a posi i e co ela ion ( = 0.810).
The ac ha week 5 included he highes amoun o
physical aining and lowes sali a y co isol le els
sugges s ha young a hle es appea o handle la ge
olumes o easy aining and high aining loads as long as
he amoun o ha d aining is educed (Table 3). This
cu en inding suppo s p e ious esea ch ha ound, an
inc ease o low-in ensi y aining, equi alen o abou
100% inc ease in aining load, showed no changes in
co isol, al hough dec eased pe o mance occu ed
(Jü imäe e al., 2004). Ou s udy displayed he g ea es
inc ease in co isol om week 5 o week 7, du ing he ea ly
compe i ion pe iod, sugges ing changes in sali a y co isol
may be mo e ela ed o he ea ly season ace schedule
a he han amoun o ha d aining. Du ing week 6, he
compe i ion season began wi h 7/8 subjec s pa icipa ing in
hei i s ace. The e o e, when in e p e ing co isol
esul s, he s ess om acing is an impo an ac o o
conside . P e ious esea ch has ound an inc ease in
mo ning and a e noon sali a y co isol le els du ing a
compe i ion day, ega dless o simila aining olume and
in ensi y, indica ing compe i ion may al e he physiology
o s ess- ela ed ho mones (Iellamo e al., 2003).
In addi ion, a dec ease in sleep quali y and du a ion
has been associa ed wi h aised co isol concen a ions as
well as an inc ease in ac i i y o he sympa he ic ne ous
sys em (Spiegel e al., 1999). In he p esen s udy, he
aining s imulus and sleep du a ion emained simila each
week; he e o e, ou indings suppo he idea ha
compe i ion s ess may ha e inc eased mo ning co isol
le els. Howe e , he co isol alues p esen ed in ou s udy
we e no collec ed on ace day, so i is ha d o know i a
compe i ion-induced s ess was s ill p esen . The analysis
o 3-day a e age sali a y co isol le els used in he p esen
s udy shi s he ocus o he o al s ess ha was occu ing
each week a he han he s ess esponse o an indi idual
compe i ion. When he subjec s we e ocused on aining
Mishica e al.
785
(week 1 o week 3), changes in sleep du a ion e ealed a
nega i e ela ionship wi h bo h K2 and co isol, p oposing
a dec ease in sleep may be associa ed o an inc eased
amoun o weekly s ain. This ag ees wi h indings ha
ound high in ensi y aining nega i ely a ec ed bo h
subjec i e sleep pa ame e s and eco e y- ela ed a ings
(Kölling e al., 2016).
Fu he mo e, in endu ance spo s, he
pa asympa he ic o m o o e aining synd ome o en
domina es (Lehmann e al., 1993). The e o e, he inc ease
in physical aining du ing week 5, ollowed by he
compe i ion s ess du ing week 6, may ha e esul ed in a
delayed a iguing a ec ha was displayed du ing week 7.
P o-longed s ess causes an inc ease in co isol as well a
dec ease pa asympa he ic ac i i y (McEwen, 2007) which
may explain why an inc ease in co isol was ound as well
as a dec ease in HRV du ing week 7. Addi ionally, an an i-
in lamma o y p ocess occu s due o aining as well as
muscle damage. The e o e, he ele a ion o co isol may be
associa ed o he g ea e aining olume du ing week 5 o
a esul o a maximal ace e o causing added s ess and
inc eased s imula ion o glycogen e-syn hesis (Ki wan e
al., 1998).
In he p esen s udy, pe o mance/ eco e y s a us
was ollowed wi h SRT. As illus a ed in Figu e 2, unning
is a common exe cise mode in bo h ha d and easy aining;
he e o e, a SRT was applicable o moni o ing a igue
wi h his g oup o subjec s. The easily epea able design
(based on speed and inclina ion) o his es ing p o ocol
p o ides a es ha can be conduc ed in a ious aining
en i onmen s, such as a aining camps o a e long
a els, o help a hle es and coaches de e mine cu en
le els o a igue. Al hough he lack o indi idualized
exe cise in ensi ies may educe he eliabili y o he es ,
he epea ed design p o ides aluable hea a e and lac a e
da a a s anda dized exe cise in ensi ies du ing he 7-week
pe iod and signi ican changes in his da a would indica e
ha le els o a igue should be u he in es iga ed.
P e ious li e a u e suppo s he applica ion o submaximal
es s o moni o ing and p edic ing pe o mance (Lambe s
e al., 2004), bu de ails he impo ance o implemen ing
mul iple a iables so adequa e insigh o indi idual s a us
is applied when in e p e ing esul s (Capos agno e al.,
2016). As a esul , we in es iga ed he ela ionships
be ween SRT hea a e, SRT blood lac a e, mo ning
sali a y co isol, HRVNOC, and physical aining. Du ing
con olled submaximal in ensi ies, HR has shown o
emain cons an wi h he lowes a ia ion occu ing a 90%
HRmax alues (Lambe s e al., 2004). In he cu en s udy,
changes o SRT hea a e (a ound 90% VO2max) be ween
week 5 and 7 displayed a s ong ela ionship wi h changes
in co isol ( = 0.929). Co isol demons a ed an addi ional
ela ionship be ween week 3 and 7 wi h a nega i e
co ela ion o changes in SRT hea a e ( = -0.929) and
blood lac a e ( = -0.857). Common assump ions abou
changes in HR a submaximal in ensi ies sugges ha an
inc ease in ae obic i ness is linked o dec eases in HR,
while inc eases in HR a e associa ed wi h a decline in
i ness, dehyd a ion o o e aining (Lambe s e al., 2004).
Ea lie esea ch addi ionally sugges s educed submaximal
HR is only a sign o e ec i e endu ance aining when no
decline in maximal pe o mance is p esen (Hedelin e al.,
2000). The e o e, wi hou maximal HR alues i is ha d o
e alua e he ela ionship be ween SRT HR and es ing
co isol alues, which also ha e mixed esul s. In addi ion,
in o de o de ec signi ican changes in SRT, i is
ecommended ha he HR alues a e app oxima ely 7 bpm
di e en a 90% HRmax wo kload (Lambe s e al., 2004)
and he e o e, luc ua ions du ing he p esen s udy we e
oo small o in e p e any aining induced changes.
Since a ia ion in esponse o aining s ess is an
appa en di e ence, i is logical o assume ha moni o ing
a iables, such as HRV, ha also ha e an indi idualized
esponse o aining s ess would be bene icial o
op imizing pe o mance. Resea ch has in es iga ed he
esponse o endu ance aining and nume ous ac o s ha e
helped explain hese di e ences such as, geno ype, aining
backg ound, gende , age, aining load, e c (Ca e e al.,
2003; Buchhei e al., 2004; Nummela e al., 2010). In
addi ion, la ge di e ences we e obse ed despi e
p esc ibing he same amoun o olume and modi ying
in ensi y aining indi idually (Nummela e al., 2010).
Du ing his s udy, al hough physical aining was no
s anda dized, compa able aining occu ed due o g oup
aining and compe i ion schedules. P esen indings
showed simila weekly ends o bo h HRV and co isol
bu indi idual di e ences we e high, ag eeing wi h
p e ious indings (Figu e 3). Due o his high in a-
indi iduali y, p e ious esea ch has in es iga ed and
implemen ed HRV-guided aining in o endu ance aining
p og ams. HRV-guided aining esul ed in a lowe
equency o high-in ensi y exe cises and he e o e, a
dec eased aining load (Ki iniemi e al., 2007). When he
iming and amoun o high in ensi y exe cise is adjus ed, a
sligh change in he aining pe iodiza ion occu s. A la ge
aining ocus o young endu ance a hle es is building
hei ae obic capaci y and an imp o ed endu ance comes
om accumula ed yea s o e ec i e aining. As a esul ,
u he esea ch should ollow he long- e m e ec s on
HRV and endu ance aining be o e implemen ing a HRV-
guided app oach o aining in young a hle es.
Limi a ions
Limi a ions o his s udy may ha e occu ed due o he
small sample size and he g ouping o bo h gende s, as well
as he dec eased s anda diza ion due o he collec ions o
noc u nal HRV, sleep du a ion and mo ning co isol alues
occu ing a home. The e may be a ious ac o s such as
poo o dis up ed sleep ha a e no associa ed o aining
bu s ill a ec mo ning co isol and HRV alues.
Addi ionally, he assessmen o physical aining and
aining load came om sel -accessed aining dia ies;
he e o e, hey we e solely based on subjec i e es ima ions
o he aining-induced s ess ha was occu ing du ing his
pe iod. Finally, ou subjec s’ le el o psycho-physiological
s ess may ha e isen when he pe iod o compe i ion
began, due, e.g., o he p essu e o pe o m well. This
po en ial change was no e alua ed independen ly and may
ha e in luenced he HRV, sleep du a ion and sali a co isol
le els. Howe e , au onomic s ess eac ions do no di e
be ween he sou ce o s ess, and mos likely, he g ea es
in luence is he o e all s ess/ eco e y balance.