phys-11-612188 Janua y 2, 2021 Time: 15:13 # 1
ORIGINAL RESEARCH
published: 13 Janua y 2021
doi: 10.3389/ phys.2020.612188
Edi ed by:
Plamen Ch. I ano ,
Bos on Uni e si y, Uni ed S a es
Re iewed by:
Thomas Penzel,
Cha i é – Uni e si ä smedizin Be lin,
Ge many
Denis Mongin,
Gene a Uni e si y Hospi als (HUG),
Swi ze land
*Co espondence:
En ico G. Caiani
[email p o ec ed]
Special y sec ion:
This a icle was submi ed o
F ac al and Ne wo k Physiology,
a sec ion o he jou nal
F on ie s in Physiology
Recei ed: 30 Sep embe 2020
Accep ed: 16 Decembe 2020
Published: 13 Janua y 2021
Ci a ion:
Solbia i S, Ma in-Yeb a A,
Vaïda P and Caiani EG (2021)
E alua ion o Ca diac Ci cadian
Rhy hm Decondi ioning Induced by
5- o-60 Days o Head-Down Bed
Res . F on . Physiol. 11:612188.
doi: 10.3389/ phys.2020.612188
E alua ion o Ca diac Ci cadian
Rhy hm Decondi ioning Induced by
5- o-60 Days o Head-Down Bed
Res
Sa ah Solbia i1,2, Alba Ma in-Yeb a3, Pie e Vaïda4and En ico G. Caiani1,2*
1Depa men o Elec onics, In o ma ion and Bioenginee ing, Poli ecnico di Milano, Milan, I aly, 2Ins i u e o Elec onics,
Compu e and Telecommunica ion Enginee ing, Consiglio Nazionale delle Rice che, Milan, I aly, 3Cen o de In es igación
Biomédica en Red – Bioingenie ía, Bioma e iales y Nanomedicina, BSICoS G oup, Uni e sidad de Za agoza, Za agoza,
Spain, 4College o Heal h Sciences, Uni e si y o Bo deaux, Bo deaux, F ance
Head-down il (HDT) bed es elici s changes in ca diac ci cadian hy hms, gene a ing
possible ad e se heal h ou comes such as inc eased a hy hmic isk. Ou aim was o
s udy he impac o HDT du a ion on he ci cadian hy hms o hea bea (RR) and
en icula epola iza ion (QTend) du a ion in e als om 24-h Hol e ECG eco dings
acqui ed in 63 subjec s du ing six di e en HDT bed es campaigns o di e en du a ion
( wo 5-day, wo 21-day, and wo 60-day). Ci cadian hy hms o RR and QTend in e als
se ies we e e alua ed by Cosino analysis, esul ing in a alue o midline (MESOR),
oscilla ion ampli ude (OA) and ac ophase (ϕ). In addi ion, he QTc (wi h Baze co ec ion)
was compu ed, and day- ime, nigh - ime, maximum and minimum RR, QTend and
QTc in e als we e calcula ed. S a is ical analysis was conduc ed, compa ing: (1) he
e ec s a 5 (HDT5), 21 (HDT21) and 58 (HDT58) days o HDT wi h baseline (PRE); (2)
ends in eco e y pe iod a pos -HDT epochs (R) in 5-day, 21-day, and 60-day HDT
sepa a ely s. PRE; (3) di e ences a R + 0 due o bed es du a ion; (4) changes
be ween he las HDT acquisi ion and he espec i e R + 0 in 5-day, 21-day, and 60-
day HDT. Du ing HDT, majo changes we e obse ed a HDT5, wi h inc eased RR and
QTend in e als’ MESOR, mos ly ela ed o day- ime leng hening and inc eased minima,
while he QTc sho ened. A e wa d, a p og essi e end owa d baseline alues was
obse ed wi h HDT p og ession. Addi ionally, he ϕan icipa ed, and he OA was educed
du ing HDT, dec easing sys em’s abili y o eac o incoming s imuli. Consequen ly, he
es o a ion o he o hos a ic posi ion elici ed he sho ening o RR and QTend in e als
oge he wi h QTc p olonga ion, no wi hs anding he pe iod spen in HDT. Howe e , he
magni ude o pos -HDT changes, as well as he di e ence be ween he las HDT day
and R + 0, showed a end o inc ease wi h inc easing HDT du a ion, and 5/7 days
we e no su icien o eco e ing a e 60-day HDT. Addi ionally, he ϕpos poned
and he OA signi ican ly inc eased a R + 0 compa ed o PRE a e 5-day and 60-
day HDT, possibly inc easing he a hy hmic isk. These esul s p o ide e idence ha
con inuous moni o ing o as onau s’ ci cadian hy hms, and u he in es iga ions on
F on ie s in Physiology | www. on ie sin.o g 1Janua y 2021 | Volume 11 | A icle 612188
phys-11-612188 Janua y 2, 2021 Time: 15:13 # 2
Solbia i e al. Head-Down Bed Res Ci cadian Decondi ioning
possible measu es o coun e ac ing he obse ed modi ica ions, will be key o u u e
missions including long pe iods o weigh lessness and g a i y ansi ions, o p ese ing
as onau s’ heal h and mission success.
Keywo ds: head-down bed es , hea a e, en icula epola iza ion, ci cadian hy hm, ca diac decondi ioning
INTRODUCTION
Long du a ion space ligh s p esen a numbe o physiological and
psychological s esso s ha deeply challenge as onau s’ heal h
and pe o mance in space. F om he La in wo ds “ci ca diem,”
meaning “app oxima ely a day,” ci cadian hy hms a e non-
andom luc ua ions ha ing a pe iod o app oxima ely 24 h ound
in almos e e y physiological p ocesses (Richa ds and Gumz,
2013). These hy hms a e au onomous, and hus pe sis ing
in absence o ex e nal synch onize s, bu hey also en ain o
en i onmen al and social cues. In pa icula , oge he wi h he
ligh -da k cycle, a numbe o ac o s con ibu e in main aining
a co ec en ainmen o ci cadian hy hms on Ea h, such as
wo king and eeding hou s, wake/sleep and ac i i y/ es cycles,
social in e ac ion, ambien empe a u e (McKenna e al., 2017),
as well as g a i y (Fulle e al., 1994). Du ing space ligh , hese
condi ions a e emo ed: he ligh -da k cycle is al e ed, he
pull o g a i y is educed, and he egula al e na ion be ween
s anding and supine posi ion is elimina ed. A educ ion in he
s eng h o he synch onize s di ec ly impac s he cha ac e is ics
o he ci cadian hy hms, in e ms o al e ed midline, ampli ude
and phase, leading o a educed capaci y o adap a ion o
incoming en i onmen al luc ua ions (McKenna e al., 2017).
Addi ionally, ci cadian hy hm impai men has been co ela ed
on Ea h o long- e m heal h p oblems, including sleep diso de s,
obesi y, diabe es, dep ession, seasonal a ec i e diso de and
aging (Logan and McClung, 2019), wi h he se e i y o he
consequences inc easing wi h he le el o ci cadian dis up ion
(Gazendam e al., 2013).
Ale ness and pe o mance also exhibi ci cadian hy hmici y
(Johnson e al., 1992): as s udied in pilo s (Caldwell, 2005)
and shi wo ke s (Boi in and Boud eau, 2014), when hy hms
desynch oniza ion occu s, an inc ease in he le el o a igue is
obse ed, consequen ly impai ing pe o mance and accen ua ing
he isk o acciden s (Flynn-E ans e al., 2016). Episodes o
sleep loss, a igue, and ci cadian hy hms dis up ion ha e been
obse ed in as onau s du ing pas space missions (Manzey
and Lo enz, 1998;Monk e al., 1998;Casle and Cook,
1999;Dijk e al., 2001;Flynn-E ans e al., 2016), and ha e
been ela ed o he in ense wo k schedule as well as o
en i onmen al ac o s, such as mic og a i y, con inemen and
mo ion sickness (Guo e al., 2014). Since he main enance
o high le el heal h s a us and ale ness a e c ucial o he
success o a mission, he “Risk o Pe o mance Dec emen s
and Ad e se Heal h Ou comes Resul ing om Sleep Loss,
Ci cadian Desynch oniza ion, and Wo k O e load” has also
been included in he mos ecen NASA Human Resea ch
P og am In eg a ed Resea ch Plan (NASA Human Resea ch
P og am, 2020), hus highligh ing he need o iden i ying
possible consequences due o ci cadian desynch oniza ion o
indi idual heal h and in es iga ing measu es o mi iga ing his
isk, pa icula ly in he scena ios o deep space explo a ion and
plane a y ope a ions.
Due o he limi ed possibili ies o in- ligh esea ch, g ound-
based analogs a e used o ep oducing and s udying he
e ec s o mic og a i y on he human body. Fo example, limb
cas ing and limb suspension ha e been used o s udy bone
densi y educ ion and neu omuscula al e a ions due o inac i i y
and unloading (Ki aha a e al., 2003;De Boe e al., 2007;
Sanse e ino e al., 2018), while bed es and d y imme sion
a e he mos widely used g ound analogs o simula ing he
e ec s o p olonged weigh lessness exposu e on he di e en
physiological sys ems. In pa icula , Head-Down (−6◦) Til
(HDT) bed es elici s ex ensi ely educed mo o ac i i y (Sandle
and Ve nikos, 1986), as well as he elimina ion o he egula
al e na ion be ween 1 and 0 Gz along he head- o- oo axis, and
he cha ac e is ic luids edis ibu ion occu ing du ing sus ained
exposu e o mic og a i y (Mon gome y, 1993). The u he
neu aliza ion o axial loading and body suppo is achie ed by
d y imme sion p o ocols (Na asiola a e al., 2011;Wa enpaugh,
2016;Tomilo skaya e al., 2019), which in ol e imme sing he
es subjec , co e ed wi h a highly elas ic wa e p oo clo h,
in a ank o he moneu al wa e (Shulzhenko e al., 1980;
Tomilo skaya e al., 2019). Despi e simila i ies in he e ec s
induced by HDT bed es and d y imme sion (Wa enpaugh,
2016;Tomilo skaya e al., 2019), he magni ude o changes
on ca dio ascula , pos u al and neu omuscula sys ems induced
by d y imme sion esul ed up o se en imes la ge han wi h
HDT bed es (Tomilo skaya e al., 2018). Howe e , due o i s
con enience and ease o use (Wa enpaugh, 2016), also allowing
easie access o coun e measu es es ed on he subjec s, he HDT
p o ocol became he p e e ed and mos u ilized g ound-based
analog o mic og a i y.
As e idenced in he NASA Human Resea ch P og am
In eg a ed Resea ch Plan, he e is cu en ly a lack o ci cadian
phase bioma ke s easily collec ible du ing space ligh .
Acco dingly, HDT bed es can be used o his aim, by
in es iga ing ci cadian en ainmen .
In p e ious s udies du ing HDT bed es , we showed ha
ca dio ascula decondi ioning is possibly a ec ing ca diac
elec ical ac i i y by inc easing en icula epola iza ion
he e ogenei y, and hus he isk o inducing hy hm diso de s
(Caiani e al., 2016). In he same pooled g oup o subjec s
examined in his pape , we also showed ha , when he no mal
g a i y ield is es o ed, T-wa e al e nans indices inc eased
a e 60-day HDT (bu no o sho e du a ions), indica i e
o incipien elec ical ins abili y on en icula epola iza ion
(Ma ín-Yeb a e al., 2019).
In addi ion, ou p elimina y esul s on 12 subjec s
unde going a 60-day HDT bed es showed ha also he
F on ie s in Physiology | www. on ie sin.o g 2Janua y 2021 | Volume 11 | A icle 612188
phys-11-612188 Janua y 2, 2021 Time: 15:13 # 3
Solbia i e al. Head-Down Bed Res Ci cadian Decondi ioning
ci cadian hy hms o bo h bea -by-bea du a ion (RR in e al)
and en icula epola iza ion du a ion (QTend in e al)
we e a ec ed, wi h a educ ion in day/nigh di e ences
al eady a e 5 days o HDT (Solbia i e al., 2020). As in
mice he de iciency o excess o K üppel-like ac o 15,
wi h a ole in he ansc ip ional con ol o he hy hmic
genes exp ession equi ed o gene a ing he ansien
ou wa d po assium cu en , has been ela ed o loss o
hy hmic QT a ia ion, abno mal epola iza ion and enhanced
suscep ibili y o en icula a hy hmias (Jeya aj e al., 2012),
he impo ance o u he s udying in humans he ci cadiani y
o RR and QT du a ion du ing HDT bed es as bioma ke s
appea s e iden .
Acco dingly, ou aim was o in es iga e he changes in
he ci cadian hy hms o ca diac elec ical ac i i y om 24-
h Hol e ECG eco dings acqui ed du ing six HDT bed es
s udies o di e en du a ion in a la ge pooled g oup o no mal
subjec s, hus expanding he p e ious esul s in Solbia i e al.
(2020), by e alua ing he deg ee o ci cadian desynch oniza ion
along he HDT, as well as ocusing on he changes elici ed a
HDT discon inua ion, as a consequence o he es o a ion o
he g a i y ield.
This pape is s uc u ed as ollows: in sec ion “Ma e ials and
Me hods” he s udy design and popula ion, da a acquisi ion
and p ocessing, as well as he me hods u ilized o he
analysis o ci cadian hy hms and s a is ical analyses a e
desc ibed. In sec ion “Resul s” he esul s a e p esen ed,
o ganized in o h ee sub-sec ions: (1) ocusing on he
e ec s du ing HDT, conside ing i s di e en du a ion; (2)
examining he pos -HDT bed es eco e y dynamics HDT,
sepa a ely o he di e en HDT du a ions; (3) compa ing
he e ec s o g a i y ield es o a ion obse ed a e 5-
day, 21-day, and 60-day HDT. In sec ion “Discussion” he
discussion o he obse ed changes in he ca diac ci cadian
hy hm is p esen ed, ollowing he same s uc u e adop ed
in sec ion “Resul s,” oge he wi h a desc ip ion o he s udy
limi a ions, and inally in sec ion “Conclusion” he main
conclusions a e epo ed.
MATERIALS AND METHODS
The da a u ilized in his s udy consis ed o 12-lead Hol e
ECG eco dings p e iously acqui ed in se e al HDT bed es
campaigns o which ou esea ch g oup pa icipa ed du ing
he las 10 yea s. De ails on he s udy design, popula ion, and
ECG acquisi ion p o ocol will be desc ibed in he ollowing
sub-sec ions.
S udy Design and Popula ion
In he con ex o he Eu opean Space Agency head-
down bed es s a egy, an only male heal hy popula ion
was ec ui ed o wo sho -du a ion (5 days), wo mid-
du a ion (21 days), and wo long-du a ion (60 days)
HDT bed es s udies, pe o med a he Ins i u de
Médecine e de Physiologie Spa iales (MEDES) in
Toulouse (F ance), o a he Ge man Ae ospace Cen e
(Deu sches Zen um ü Lu - und Raum ah e.V, DLR)
in Cologne (Ge many). All subjec s we e no aking
medica ion o any kind.
Volun ee s we e andomly assigned o a con ol (CTRL) o
o a coun e measu e (CM, depending on he s udy, di e en
in e en ions we e applied o he subjec s du ing HDT o es
e ec i eness in p e en ing changes) g oup. Each s udy included
a pe iod o baseline da a collec ion (PRE), a pe iod o s ic
6◦HDT bed es 24 h a day, and a pe iod o pos -bed es
eco e y, whe eas lying in bed du ing he day was no allowed
be o e and a e he HDT pe iod. Fo de ailed desc ip ion
o he design ele an o each campaign, please e e o he
Supplemen a y Ma e ial.
The p esen s udy conside ed only he ECG da a acqui ed
om he subjec s assigned o CTRL g oups, o a o al o
63 subjec s. Sleeping hou s we e scheduled om 11:00 pm
o 6:30 am a DLR and om 11:00 pm o 7:00 am a
MEDES, and napping was no allowed du ing he day. All he
pe o med p ocedu es we e in acco dance wi h he 1964 Helsinki
Decla a ion and i s la e amendmen s o compa able e hical
s anda ds. Fo each bed es campaign, all olun ee s p o ided
w i en in o med consen o pa icipa e in he s udy, app o ed by
he espec i e E hical Commi ee o Human Resea ch a each o
he hos ing ins i u ions.
An h opome ic in o ma ion ele an o he subjec s en olled
as CTRL in he 5-day, 21-day, and 60-day HDT campaigns a e
epo ed in Table 1, and p esen ed as median [25 h pe cen ile;
75 h pe cen ile].
ECG Da a Acquisi ion and
P e-p ocessing
Fo each bed es campaign, 12-lead 24-h Hol e ECGs (1000 Hz,
H12+, Mo a a Ins umen Inc.) we e acqui ed o each subjec
a speci ic epochs be o e, du ing, and a e he HDT pe iod, as
schema ized in Figu e 1. O no e, longe campaigns p ese ed
he same in e media e acquisi ion days as in sho e campaigns,
hus esul ing in: 63 subjec s om 5-day, 21-day, and 60-
day HDT campaigns s udied a PRE and HDT5 (i.e., he 5 h
day o HDT bed es ), o which 41 subjec s s udied also a
HDT21 (i.e., he 21s day o HDT bed es ), and 21 subjec s
s udied also a HDT58 (i.e., he 58 h day o HDT bed es ). In
addi ion, all 63 subjec s we e s udied in he 24 h ollowing g a i y
eins a emen , namely R + 0, as well as se e al days a e , om
R + 3 (i.e., he 4 h day o eco e y) o R + 7 (i.e., he 8 h day
o eco e y), acco ding o he campaign du a ion, as speci ied
in Figu e 1.
Fo each eco ding, he iducial poin s co esponding o he Q,
R, and T wa e end (Tend) o he ECG signal we e ex ac ed, and
bea - o-bea RR and QTend in e al se ies (ms) we e compu ed
using he p op ie a y Mo a a Ins umen esea ch so wa e
Supe ECG (Mo a a, 2009). The ex ac ed a iabili y se ies we e
p e-p ocessed in o de o exclude ou lie s o a i ac s due o
acquisi ion p oblems (i.e., elec ode de aching, cable in e e ence,
o he s): an ou lie ejec ion me hod based on he s anda d
de ia ion was applied, aimed a elimina ing samples exceeding
ou imes he s anda d de ia ion o he p eceding 50 samples.
F on ie s in Physiology | www. on ie sin.o g 3Janua y 2021 | Volume 11 | A icle 612188
phys-11-612188 Janua y 2, 2021 Time: 15:13 # 4
Solbia i e al. Head-Down Bed Res Ci cadian Decondi ioning
TABLE 1 | To al numbe o male subjec s en olled in he con ol g oup o 5-day, 21-day, and 60-day HDT campaigns, oge he wi h hei age and Body Mass Index
(BMI) exp essed as median [25 h pe cen ile; 75 h pe cen ile].
5-day HDT 21-day HDT 60-day HDT p- alue
Numbe o male subjec s 22 20 21
Age 31.6[25.6;35.8] 32[28.8;40.3] 28[25;36] 0.4268
BMI 24.4[23.2;25.1] 22.9[21.8;24.5] 24.4[22.5;25.1] 0.14
The esul s (p- alue) o he K uskal–Wallis es a e also p esen ed, wi h no signi ican di e ences be ween he g oups.
FIGURE 1 | Schema iza ion o he 24 h Hol e ECG expe imen conduc ed du ing he 5-day, 21-day, and 60-day Head-down il (HDT) bed es campaigns. Speci ic
days in which he ECG acquisi ions we e pe o med a e indica ed by he a ows. PRE: baseline da a collec ion (BDC) be o e he beginning o HDT; HDTx: acquisi ion
pe o med a day x o he HDT; R + 0: acquisi ion s a ing immedia ely a e he conclusion o HDT; R+x: acquisi ion pe o med x days a e he end o HDT (see
sec ion “ECG da a acquisi ion and p e-p ocessing” and Supplemen a y Ma e ials o de ails).
Addi ionally, Baze ’s co ec ion (Baze , 1920) was applied
o compu e bea - o-bea QTc in e al se ies, acco ding o he
o mula:
QTc(i)=QTend(i)/pRR(i−1)i=1,2,...,N(1)(1)
being N he numbe o RR and QTend in e als in he se ies.
A e wa d, RR, QTend, and QTc se ies we e ealigned o he
ime o he beginning o he day, and he median alues o
consecu i e, non-o e lapping 15-min segmen s we e compu ed
o each subjec a each acquisi ion epoch. Addi ionally, he
minimum and he maximum RR, QTend, and QTc alues o e
he 24 h we e compu ed o each eco ding, as well as he median
alue du ing he diu nal and he noc u nal pe iods.
Me hods o he Analysis o Ci cadian
Rhy hms
The cha ac e is ics o ci cadian hy hm in he ca diac elec ical
ac i i y we e e alua ed by pe o ming he single componen
Cosino analysis. The Cosino analysis is a widely used eg ession
me hod o he e alua ion o non- andom hy hms in ime
se ies, which consis s in i ing he o iginal equidis an o non-
equidis an se ies wi h a pe iodic unc ion (Co nelissen, 2014),
de ined by he equa ion:
Y( )=MESOR +OA ×cos(2π /τ+ϕ)+e( )(2)
whe e MESOR (Midline S a is ic o Rhy hm) ep esen s a
hy hm-adjus ed mean, OA is he oscilla ion ampli ude,
measu ing hal a ia ion wi hin a nigh -day cycle, ϕis he
ac ophase, ha is he empo al alue a which he ampli ude o
he i ing sinusoid eaches i s maximum alue, τis he pe iod
ep esen ing he du a ion o one cycle, and e( ) is he i ing
e o e m a each ime poin , de ined as an independen andom
a iable wi h mean equal o ze o and a iance σ2.
In addi ion o he Cosino model, equency-based app oaches
such as he Fou ie spec al analysis, he Lomb–Sca gle
pe iodog am o he au o eg essi e spec al analysis, allow o
pe o m he spec al decomposi ion o he se ies, iden i ying as
ci cadian hose se ies whose spec um co ela es wi h a pe iod
o 24 h (Leise, 2017). Fu he mo e, o he mo e specialized
me hods ha e been de eloped in suppo o he analysis
o se ies cha ac e ized by ime- a ian ampli ude o phase,
including spec og ams and wa ele ans o ms (Leise, 2015),
F on ie s in Physiology | www. on ie sin.o g 4Janua y 2021 | Volume 11 | A icle 612188
phys-11-612188 Janua y 2, 2021 Time: 15:13 # 5
Solbia i e al. Head-Down Bed Res Ci cadian Decondi ioning
as well as empi ical (Huang e al., 1998) and non-linea mode
decomposi ion (Ia senko e al., 2015), in which he o iginal
se ies is decomposed in o se o non-linea modes, s a ing
om he ime- equency ep esen a ion o he se ies ob ained by
windowed Fou ie ans o m o wa ele ans o m.
Unlike he Cosino analysis, howe e , he ci ed me hods
equi e a leas 3 o p e e ably mo e consecu i e cycles acqui ed
om he same sou ce o indi idual. In he p esen s udy,
mul iple, non-consecu i e, 24-h-long se ies we e a ailable, and
hus he analysis o he changes in he ci cadian pe iod was no
allowed. Acco dingly, he ci cadian pe iod τwas assumed o be
equal o 24 h (Co nelissen, 2014), being synch onized o he
ex e nal 24-h ligh -da k cycle, which was ixed and imposed by
he expe imen al p o ocol in all he conside ed s udies. Thus,
being τassumed as known, he equa ion (2) can be e-w i en as:
Y( )=MESOR +βx+γz+e( )(3)
whe e β= A(cosϕ), γ=−A(sinϕ), x = cos(2π /τ), z = sin(2π /τ).
A e wa d, he leas squa es me hod is used o minimize he
esidual sum o squa es (RSS), i.e. he sum o squa ed di e ences
be ween measu emen s Y( i), o i = 1, 2, . . ., N, and he alues
es ima ed om he model (Co nelissen, 2014):
RSS =X
i
[Yi−
∧
MESOR +
∧
βxi+
∧
γzi]2(4)
T igonome ic o mulae a e hen used o de i e es ima es o he
ampli ude (OA) and ac ophase (ϕ) om
∧
βand ∧
γ.
Consequen ly, in he p esen wo k, he ci cadian hy hm o
24-h RR and QTend se ies was e alua ed o each eco ding
by pe o ming he single componen Cosino analysis on he
compu ed 15-min medians, o minimize i ing e o s o he leas
squa e me hod in p esence o possible ou lie s, hus esul ing in
a alue o MESOR, OA and ϕQTend and ϕRR o each subjec a
each epoch. Reco dings ha ing less han 60% o a ailable signal
o e he 24 h we e excluded om he analysis. In all o he cases,
linea in e pola ion was pe o med o ill missing samples.
Addi ionally, he di e ence in minu es be ween he RR and
QTend ac ophases was compu ed as:
1ϕ=ϕRR −ϕQTend (5)
S a is ical Analysis
The pe o med s a is ical analyses a e desc ibed in de ail in he
ollowing sub-sec ions. Fo all es s, p<0.05 was conside ed as
le el o signi icance (α) o ejec he null hypo hesis.
P esence o Ci cadian Rhy hms
The p esence o ci cadian hy hmici y in RR and QTend se ies
was assessed by he Ze o-Ampli ude es (Co nelissen, 2014),
wi h he null hypo hesis ha he ampli ude o he se ies is ze o,
and hus ha he e is no hy hm in i .
E ec o HDT Bed Res Du a ion
In o de o es he null hypo hesis ha HDT, independen ly o i s
du a ion, did no p oduce any e ec on he compu ed pa ame e s,
he non-pa ame ic K uskal–Wallis es o independen samples
(due o he di e en numbe o subjec s in each g oup o his
compa ison) was pe o med o compa e alues ob ained a PRE,
HDT5, HDT21, and HDT58, wi h he pos hoc Mann–Whi ney
es wi h Bon e oni co ec ion.
Pos -HDT Bed Res Reco e y
To es he null hypo hesis ha in he same subjec HDT did no
p oduce any e ec on he compu ed pa ame e s a i s conclusion
and in he ollowing days, he non-pa ame ic F iedman es o
epea ed measu es was applied, oge he wi h pos hoc Wilcoxon
Signed Rank es wi h Bon e oni co ec ion, o compa e alues
a PRE, a R + 0 and a he ollowing a ailable R+x epoch (wi h
x depending om he HDT du a ion – see Figu e 1), sepa a ely
a e 5-day, 21-day, and 60-day HDT. In case o missing da a, he
ele an subjec was excluded om he pai ed compa ison.
E ec s o G a i y Field Res o a ion
To es he null hypo hesis ha HDT discon inua ion,
independen ly o i s du a ion, did no p oduce immedia e
e ec s on he compu ed pa ame e s, he di e ence (exp essed
as 1%) be ween he las a ailable HDT acquisi ion and he
espec i e R + 0 was compu ed o 5-day, 21-day, and 60-day
HDT campaigns sepa a ely, acco ding o he o mulae:
15−day HDT =(R+0)−(HDT5)
HDT5×100 (6)
121−day HDT =(R+0)−(HDT21)
HDT21 ×100 (7)
160−day HDT =(R+0)−(HDT58)
HDT58 ×100 (8)
Thus, he K uskal–Wallis es (p<0.05) was pe o med o
compa e 15−dayHDT,121−dayHDT, and 160−dayHDT, oge he
wi h pos hoc Mann–Whi ney es wi h Bon e oni co ec ion.
RESULTS
Resul s a e p esen ed as median [25 h pe cen ile; 75 h pe cen ile],
and de ailed in o ma ion abou s a is ical analyses esul s a e
epo ed in he ables. In he ex , only signi ican a ia ions a e
unde lined (unless o he wise speci ied) in e ms o median o he
ela i e change (in %) compa ed o espec i e alues a PRE.
Due o acquisi ion p oblems, he eco dings speci ied in
Table 2 we e no a ailable o he analyses. Consequen ly, he
esul s a ailable om hose subjec s wi h missing eco dings we e
excluded om all he analyses in ol ing pai ed compa isons,
while hey we e s ill conside ed in unpai ed es s.
Locally missing da a wi hin he 24 h, epo ed bo h as % in
espec o he whole acquisi ion and as median [25 h pe cen ile;
75 h pe cen ile] du a ions, we e ound a PRE (1.04%, 15[0;90]
minu es), a HDT58 (2.08%, 30[0;94] minu es), and a R + 0 in
21-day HDT bed es (2.08%, 30[15;67] minu es).
E ec o HDT Bed Res Du a ion
Bo h a PRE and a each HDT bed es epoch, RR and QTend
in e al se ies exhibi ed a ma ked ci cadian pa e n, as isible in
F on ie s in Physiology | www. on ie sin.o g 5Janua y 2021 | Volume 11 | A icle 612188

phys-11-612188 Janua y 2, 2021 Time: 15:13 # 6
Solbia i e al. Head-Down Bed Res Ci cadian Decondi ioning
Figu e 2 in which he dis ibu ions o e he 24 h ( ep esen ed as
a median [25 h-75 h] alue e e y 15-min) o he RR and QTend
in e als o he pooled popula ion a e shown. In pa icula , as
con i med by he Ze o-Ampli ude es , he null hypo hesis was
ejec ed and he ci cadian hy hm was main ained in all subjec s
a bo h PRE and a all HDT epochs. As expec ed, sho e RR
in e als we e isible du ing he day, deno ing highe hea a e,
ollowed by a s eep inc ease owa d he highe alues obse ed a
nigh , wi h QTend in e al ollowing RR a ia ions.
TABLE 2 | Lis o he numbe o subjec s and co esponding head-down il (HDT)
bed es campaign (using i s ac onym as de ined in Supplemen a y Ma e ial) in
which day, nigh o Cosino analyses we e no pe o med o ce ain epochs
because o missing da a.
Epoch Day Nigh Cosino
PRE 1 SAG 1 SAG, 1 RSL
HDT5 1 MNX 1 MNX, 1 RSL
HDT21 1 MNX 1 MNX
R + 0 5-day HDT 1 BR-AG1 1 BR-AG1 1 BR-AG1
R + 0 21-day HDT 1 MEP 1 MEP 1 MNX, 1 MEP
R + 4/5 21-day HDT 4 MEP 1 MNX, 4 MEP 1 MNX, 4 MEP
R + 0 60-day HDT 1 RSL 1 RSL
PRE, days o obse a ion be o e HDT was s a ed; HDTx, acquisi ion pe o med a
day x o he HDT; R + 0, acquisi ion s a ing immedia ely a e he conclusion o
HDT; R+x, acquisi ion pe o med x days a e he end o HDT (see sec ion “ECG
da a acquisi ion and p e-p ocessing” o de ails).
Howe e , he cha ac e is ics o he ci cadian hy hm changed
du ing he HDT pe iod. Table 3 p o ides a summa y o day,
nigh , maximum and minimum alues o RR, QTend and QTc
in e als, as well as o he Cosino pa ame e s o RR and QTend,
a PRE and a each HDT epoch.
Majo changes we e obse ed a HDT5, wi h inc eased RR
and QTend MESOR compa ed o PRE (+11.4% and +2.3%,
espec i ely), mos ly ela ed o a leng hening occu ing du ing
he diu nal pe iod (RR: +16.8%; QTend: +3.7%), as also
e idenced by he inc eased minima (RR: +39.4%; QTend: +8%).
On he o he hand, no changes we e obse ed in RR and QTend
maxima and in noc u nal alues o any o he HDT epochs. Wi h
he p olonga ion o HDT a g adual eco e y owa d baseline
alues was obse ed: o he RR in e al, his was mani es ed a
HDT58 as dec eased diu nal (−13.5%), noc u nal (−9.2%), and
MESOR (−11.4%) alues compa ed o HDT5. Howe e , he RR
in e al was s ill leng hened compa ed o PRE a HDT21 du ing
he day (+9.7%), and bo h RR and QTend minima emained
highe han a PRE up o HDT58, showing only a end owa d
he baseline (RR: +35.8% a HDT21, +30.1% a HDT58; QTend:
+6.8% a HDT21, +5.9% a HDT58).
The inc eased minimum and diu nal alues, and he
simul aneous unchanged maximum and noc u nal alues, du ing
he en i e HDT caused a dampening in he ampli ude o day-
nigh oscilla ions (OA) o RR (be ween −23.8 and −30%)
and QTend (be ween −24 and −32%) om baseline as he
HDT was p olonged.
FIGURE 2 | Cumula i e esul s (each alue ep esen s he 15-min median wi h i s 25 h and 75 h pe cen iles) o hea bea du a ion (RR) and en icula epola iza ion
du a ion (QTend) compu ed om he co esponding pooled popula ion be o e (PRE) he beginning o head-down il (HDT) and a he x- h day o HDT bed es
(HDTx).
F on ie s in Physiology | www. on ie sin.o g 6Janua y 2021 | Volume 11 | A icle 612188
phys-11-612188 Janua y 2, 2021 Time: 15:13 # 7
Solbia i e al. Head-Down Bed Res Ci cadian Decondi ioning
The ϕQTend occu ed ea lie du ing he HDT compa ed o
PRE, and a simila end (signi ican only a HDT58) was
obse ed in ϕRR.
Also he QTc in e al was a ec ed du ing he HDT, being
pa icula ly sho ened a HDT5 compa ed o he baseline (Day
−2.4%; Nigh −2.8%; Maximum −5.8%) eaching alues lowe
han he 390 ms no mali y limi o sho QT synd ome
(Rau aha ju e al., 2009). A e wa d, only maximum QTc
emained sho ened a HDT21 (−4.3%), while day and nigh
alues ended owa d eco e y, being g ea e a HDT58 when
compa ed o HDT5.
Pos -HDT Reco e y
In his pa ag aph, he esul s ele an o he analysis o pos -
HDT eco e y a e p esen ed in e ms o compa ison be ween he
baseline and he eco e y epochs, o 5-day, 21-day, and 60-day
HDT campaigns sepa a ely.
Reco e y A e 5-Day HDT
The esul s ele an o he 5-day HDT bed es a e epo ed
in Table 4.
A R + 0, he diu nal and noc u nal RR in e al dec eased
wi h espec o he baseline (Day −8.3%, Nigh −3.9%), as well
as in e ms o minimum (−29.7%), maximum (−3.5%), and
MESOR (−6.7%) alues, wi h QTend ollowing RR a ia ions
(Day −4.1%, Nigh −1.1%, Minimum −9.5%, MESOR −2.6%).
Mo eo e , he OA o bo h RR and QTend was inc eased a R + 0
by +28.1 and +26.9%, espec i ely. A R + 3/4, a end owa d
eco e y was obse ed. In mos o he cases, his was isible
as a e u n owa d baseline alues, ha o RR maximum and
MESOR was mani es ed as an inc ease compa ed o PRE (+5.1
and +3.5%, espec i ely). No changes we e de ec ed in RR and
QTend ac ophases and in hei di e ence, as well as o he QTc.
Reco e y A e 21-Day HDT
The esul s ele an o he 21-day HDT bed es a e epo ed
in Table 5.
A R + 0, he RR in e al appea ed sho ened compa ed
o PRE, in e ms o day (−11.2%), nigh (−11.5%), maximum
(−8.3%), and MESOR (−11%) alues. Con e sely, he QTend
in e al esul ed sho ened only du ing he day (−2.9), while
diu nal and noc u nal QTc leng hened (+3.3 and +2.1%,
espec i ely). A e wa d, a R + 4/5, RR and QTend in e als
eco e y owa d baseline alues was eached, excep o diu nal,
noc u nal and minimum QTc ha con inued o be highe
han PRE (+2.9, +3.2, and +4%). The RR ac ophase was
sligh ly an icipa ed compa ed o PRE while ha o QTend
was no modi ied.
Reco e y A e 60-Day HDT
The esul s ele an o he 60-day HDT bed es a e epo ed
in Table 6.
A R + 0, bo h RR and QTend in e als signi ican ly sho ened
compa ed o baseline, in e ms o diu nal (RR: −27.5%; QTend:
−10%), noc u nal (RR: −22.5%; QTend: −7.0%), maximum
(RR: −21%; QTend: −6.4%), minimum (RR: −26.4%; QTend:
−11.9%), and MESOR (RR: −24.6%; QTend: −8.3%) alues.
Con e sely, he QTc esul ed p olonged a R + 0 wi h espec o
TABLE 3 | Resul s exp essed as median [25 h pe cen ile; 75 h pe cen ile] o he analysis o he e ec s o head-down il (HDT) du a ion on he pa ame e s compu ed
om hea a e (RR) and en icula epola iza ion du a ion, as o iginal alues (QTend) and co ec ed using he Baze o mula (QTc), in he pooled popula ion be o e HDT
was s a ed (PRE: 63 subjec s day, 62 subjec s nigh , 61 subjec s Cosino ) and a each a ailable HDT epoch du ing he bed es (HDT5: 63 subjec s day, 62 subjec s
nigh , 61 subjec s Cosino ; HDT21: 41 subjec s day, 40 subjec s nigh and Cosino ; HDT58: 21 subjec s).
PRE HDT5 HDT21 HDT58 p- alue
RR Day (ms) 818[763;907] 955[870;1006]*(7e−08)897[820;967]*(3e−03)826[791;908]F(1e−03)3.2e-07
Nigh (ms) 1089[992;1188] 1140[1055;1209] 1076[1016;1187] 1035[983;1120]F(7e−03)0.037
Maximum (ms) 1187[1090;1291] 1246[1149;1334] 1191[1101;1318] 1150[1068;1234] 0.070
Minimum (ms) 531[391;650] 740[696;824]*(2e−14)721[628;782]*(4e−07)691[600;770]*(6e−05)3e-14
MESOR (ms) 915[852;998] 1019[931;1081]*(4e−05)962[917;1037] 903[862;961]F(8e−04)6.3e-05
OA (ms) 160[137;201] 112[90;144]*(3e−06)122[105;160]*(1e−03)120[98;151]*(1e−03)8.3e-06
ϕ(hou :min) 3:46[3:15;4:19] 2:14[2:31;4:08] 3:30[3:03;4:09] 3:16[2:43;3:43]*(6e−03)0.032
QTend Day (ms) 374[365;388] 388[370;403]*(1e−03)381[369;401] 374[366;393] 0.014
Nigh (ms) 413[399;432] 411[401;430] 415[395;434] 409[399;427] 0.870
Maximum (ms) 428[411;451] 424[411;443] 428[405;449] 421[412;440] 0.813
Minimum (ms) 338[302;350] 365[352;380]*(1e−10)361[348;375]*(2e−06)358[343;368]*(7e−04)3.7e-10
MESOR (ms) 386[375;401] 395[380;412] 392[379;409] 387[377;402] 0.176
OA (ms) 25[20;32] 17[13;20]*(4e−10)19[16;23]*(1e−04),F(7e−03)18[15;23]*(2.7e−04)7.4e-10
ϕ(hou :min) 3:58[3:09;4:33] 3:19[2:40;3:52]*(2e−03)3:33[3:10;4:08] 3:27[2:55;3:44]*(6.3e−03)0.004
1ϕ ϕ (min) –10[–25.4;3.3] –2.1[–19.6;12.2] –5[–18.9;13.9] –8[–17.3;–1.7] 0.161
QTc Day (ms) 409[400;422] 399[388;410]*(1e−04)403[395;416] 415[397;424]F(4e−03)4.0e-4
Nigh (ms) 398[388;411] 387[378;399]*(3e−03)394[382;404] 402[385;412] 0.014
Maximum (ms) 446[428;462] 420[414;427]*(1e−10)427[417;439]*(3e−05)433[422;443] 5.6e-10
Minimum (ms) 384[374;400] 377[367;390] 385[371;397] 393[376;408] 0.022
MESOR: midline s a is ic o hy hm; OA, oscilla ion ampli ude; ϕ, ac ophase. The p- alue column e e s o he esul o he K uskal–Wallis es . Resul s o he pos hoc
Mann–Whi ney es wi h Bon e oni co ec ion a e p esen ed as: ∗(p− alue):p<0.05/6 s. PRE; F(p− alue):p<0.05/6 s. HDT5.
F on ie s in Physiology | www. on ie sin.o g 7Janua y 2021 | Volume 11 | A icle 612188
phys-11-612188 Janua y 2, 2021 Time: 15:13 # 8
Solbia i e al. Head-Down Bed Res Ci cadian Decondi ioning
TABLE 4 | Rela i e changes o he immedia e (R + 0) e ec s o eco e y, and ollowing same days (R + 3/4), a e a 5-day head-down il (HDT) on he pa ame e s
compu ed om hea a e (RR) and en icula epola iza ion du a ion, as o iginal alues (QTend) and co ec ed using he Baze o mula (QTc), compa ed o baseline
alues (PRE) as e e ence.
Rela i e changes compa ed o PRE
5-days HDT PRE R + 0 R + 3/4 p- alue
RR Day (ms) 850[770;907] −8.3[−15.8;−6.6]%*(7e−05)+0.6[−2.8;3.8]%F(6e−05)6.0e-07
Nigh (ms) 1093[984;1181] −3.9[−7.6; −1.2]%*(1e−03)+4.5[−0.9;6.9]%F(1.e−04)1.1e-05
Maximum (ms) 1159[1072;1271] −3.5[−5.0; −0.4]%*(1e−02)+5.1[1.9;9.1]%*(3e−03), F(1e−04)1.9e-05
Minimum (ms) 554[401;665] −29.7[−43.2; −3.5]%*(8e−04)+12.8[−8.5;57.7]%F(6e−05)4.4e-06
MESOR (ms) 917[851;1001] −6.7[−10.3; −4.6]%*(1e−04)+3.5[0.4;6.6]%*(8e−03), F(9e−05)1.6e-07
OA (ms) 150[108;191] +28.1[1.2;58.4]%*(3e−03)+4.7[−17.2;26.3]%F(0.01)0.022
ϕ(hou :min) h 4:00[3:35;4:47] 0.18[−0.55;0.91] hou s −0.53[−1.52;0.88] hou s 0.387
QTend Day (ms) 376[366;387] −4.1[−6.5; −1.3]%*(4e−04)+0.5[−1.6;2.3]%F(6e−05)1.8e-05
Nigh (ms) 409[398;435] −1.1[−5.0;1.1]% +1.5[−1.6;3.4]%F(7e−03)4.8e-03
Maximum (ms) 422[408;444] −0.9[−3.5;1.5]% +2.[−0.5;4.2]%F(1e−03)0.002
Minimum (ms) 340[279;349] −9.5[−16.6; −0.1]%*(4e−03)+6.3[−3.5;25.2]%F(6e−05)1.0e-05
MESOR (ms) 385[377;401] −2.6[−4.6; −1]%*(4e−03)+1.4[−0.2;3.7]%F(1e−04)1.1e-05
OA (ms) 23[18;33] +26.9[0.2;60.8]%*(5e−03)−11.8[−24.6;20.8]%F(3e−04)0.001
ϕ(hou :min) h 4:14[3:56;5:03] +0.21[−0.35;1.08] hou s −0.72[−1.28;0.42] hou s 0.445
1ϕ ϕ (minu es) −12.8[−32.7;4.4] −7.1[−25.4;13.2] minu es 0.3[−22.8;33] minu es 0.350
QTc Day (ms) 409[399;419] +2[−0.3;3.7]% +0.9[−1.0;2.7]% 0.228
Nigh (ms) 399[391;411] +0.6[−1.5;2.2]% −0.4[−1.4;1.3]% 1
Maximum (ms) 439[425;453] +3.6[−1;7.2]% +1.2[−1;3.4]% 0.100
Minimum (ms) 383[374;401] +1.5[−1.3;4.3]% +0.9[−0.7;2.6]% 0.538
MESOR, midline s a is ic o hy hm; OA, oscilla ion ampli ude; ϕ, ac ophase. Resul s a e exp essed as median [25 h pe cen ile; 75 h pe cen ile] ep esen ing he ela i e
change compa ed o PRE (in %). The p- alue column e e s o he esul o he F iedman es . Resul s o he pos hoc Wilcoxon Signed Rank es wi h Bon e oni co ec ion
a e p esen ed as: ∗(p− alue):p<0.05/3 s. PRE; F(p− alue):p<0.05/3 s. HDT5.
he baseline (Day +5.3%; Nigh +3.9%; Minimum +4.1%). These
changes in RR and QTc s ill pe sis ed a R + 5/7, showing only a
sligh end owa d eco e y.
While he OA o RR did no show signi ican a ia ions a
R + 0, he OA o QTend esul ed inc eased by +38.4% compa ed
o PRE, and hen eco e ing a R + 5/7.
Bo h RR and QTend ac ophases esul ed pos poned a R + 0,
bu eco e ed a R + 5/7, whe e a sligh an icipa ion o RR o e
QT in 1ϕ was e idenced.
E ec s o G a i y Field Res o a ion
A compa ison o RR, QTend and QTc alues among 5-day, 21-
day, and 60-day HDT campaigns a he espec i e R + 0, i.e.,
immedia ely a e g a i y es o a ion, is epo ed in Figu e 3.
A e 60 days o HDT, he RR in e al appea ed mo e educed
compa ed o a e 5-day HDT (Day −20.6[−30.1;−10.3]%,
Nigh −16.2[−20.6;−10.3]%, Maximum −13.5[−18.2;−6.4]%,
MESOR −17.1[−21.1;−12.8]%) and o a e 21-day HDT (Day
−18.3[−22.9; −11.7]%, Nigh −10.2[−27.8; −6.4]%, Maximum
−10.4[−25.4; −2.2]%, MESOR −12.9[−23.7; −9.3]%). A simila
beha io was obse ed in QTend day (−7.8[−11.4; −2.3]%
compa ed o 5-day HDT, −7.0[−12.4; −3.8]% compa ed o
21-day HDT), nigh (−3.9[−7.9; −0.1]% compa ed o 5-day
HDT), maximum (−3.8[−9.0;1.2]% compa ed o 5-day HDT),
and MESOR (−6.2[−9.3; −0.3]% compa ed o 5-day HDT,
−7.3[−11.3; −0.7]% compa ed o 21-day HDT) alues. No
di e ences we e obse ed in he OA, no in he ac ophase.
Diu nal and noc u nal QTc appea ed p olonged a e 60 days
o HDT han a e 5-day HDT (Day +3.7[1.7;6.4]%, Nigh
+2.6[0.7;7.3]%), wi h a posi i e end as HDT du a ion inc eases.
Figu e 4 displays he linea eg ession be ween he median
QTc o e he 24 h and he OA o he QTend (Figu e 4A, including
all acquisi ions), as well as be ween he espec i e a ia ion 1(%)
be ween he las HDT day and R + 0 (Figu e 4B). In pa icula ,
he o me showed ha highe OA o he QTend we e associa ed
wi h longe QTc (p<0.0001), while only a sligh , non-signi ican
ising end in OA was obse ed wi h inc easing QTc om he
las HDT day o R + 0.
The e ec o he decondi ioning pe cei ed a bed es
discon inua ion was also s udied in e ms o 1(%) be ween he
las HDT day o 5-day (HDT5), 21-day (HDT21), and 60-day
(HDT58) HDT, and he espec i e R + 0. As displayed in Figu e 5,
nega i e a ia ions we e ound o he majo i y o he pa ame e s,
hus e lec ing he dec ease a R + 0 compa ed o he las day
o HDT bed es .
In pa icula , a la ge dec ease a R + 0 in he diu nal alues o
RR and QTend was elici ed a e 60-day HDT (RR: −30.5[−37.9;
−23.3]%; QTend: −11.5[−15.8; −9.3]%) compa ed o wha
elici ed a e a 5-day HDT (RR: −17.2[−22.3; −13.8]%; QTend:
−5.9[−7.7; −4.3]%) o a 21-day HDT (RR: −19.2[−22.8;
−16.2]%; QTend: −6.6[−7.9; −3.2]%), while a end wi h
inc easing HDT du a ion was isible du ing he nigh (RR:
−7.4[−9.3; −2.6]% in 5-day, −13.1[−17.1; −10.3]% in 21-day,
−18[−23.3; −14.8]% in 60-day HDT; QTend: −1.1[−1.8;0]%
F on ie s in Physiology | www. on ie sin.o g 8Janua y 2021 | Volume 11 | A icle 612188
phys-11-612188 Janua y 2, 2021 Time: 15:13 # 9
Solbia i e al. Head-Down Bed Res Ci cadian Decondi ioning
TABLE 5 | Rela i e changes o he immedia e (R + 0) e ec s o eco e y, and ollowing same days (R + 4/5), a e a 21-day head-down il (HDT) on he pa ame e s
compu ed om hea a e (RR) and en icula epola iza ion du a ion, as o iginal alues (QTend) and co ec ed using he Baze o mula (QTc), compa ed o baseline
alues (PRE) as e e ence.
Rela i e changes compa ed o PRE
21-days HDT PRE R + 0 R + 4/5 p- alue
RR Day (ms) 838[778;915] −11.2[−18.8; −7]%*(4e−04)−5.8[−12.1;0]%F(7e−03)2.1e-05
Nigh (ms) 1078[1024;1198] −11.5[−15.5; −3.6]%*(6e−04)−1.5[−9.8;2.6]%F(0.01)0.008
Maximum (ms) 1232[1148;1314] −8.3[−12; −3.7]%*(1e−03)−2.1[−8.2;0.2]% 0.003
Minimum (ms) 402[375;416] −1.5[−6.7;3]% +23.9[4.0;56.7]%F(5e−03)0.015
MESOR (ms) 927[887;1019] −11[−13.3; −7.9]%*(1e−04)−3.3[−11.9;0.9]%F(7e−03)2.3e-04
OA (ms) 183[155;209] −8.5[−37.0;30.8]% −8.4[−27.6;1.4]% 0.168
ϕ(hou :min) 3:52[3:35;4:29] +0.1[−0.2;1.1] hou s −0.9[−1.1; −0.3] hou s*(2e−03),F(0.01)0.005
QTend Day (ms) 374[367;393] −2.9[−6.0; −0.3]%*(3e−03)−0.3[−2.5;2.8]%F(4e−03)0.003
Nigh (ms) 420[411;433] −1.4[−6.3;0.6]% +2.5[−0.4;3.7]%F(1e−04)0.001
Maximum (ms) 439[424;558] −4.1[−11.9;0.1]% −1.6[−17.6;3]% 0.305
Minimum (ms) 308[298;339] −0.8[−7.9;4.6]% +9.0[−0.7;15.3]%F(2e−03)0.003
MESOR (ms) 392[384;414] −2.8[−6.2; −1.2]% −0.6[−1.4;4.2]%F(1e−03)2.5e-03
OA (ms) 28[21;34] −14.5[−37.7;27.5]% −3.6[−22.1;18.3]% 0.807
ϕ(hou :min) 3:54[2:27;4:24] +0.90[0.28;2.65] hou s −0.48[−1.09; −0.01] hou s 0.010
1ϕ ϕ (minu es) −7.3[−15;6.6] −29.3[−55; −17.5] −10.9[−27.4;4.4] 0.076
QTc Day (ms) 408[395;425] +3.3[1.2;5.3]%*(6e−03)+2.9[0.8;4.7]%*(2e−03)0.001
Nigh (ms) 401[383;414] +2.1[0.9;3.2]%*(7e−03)+3.2[1.5;4.4]%*(1e−04)1.4e-04
Maximum (ms) 467[459;629] −0.1[−4.5;2]% −1.4[−13.7;2.1]% 0.646
Minimum (ms) 390[370;398] +2.4[0.2;3.4]% +4.0[1.5;4.8]%*(5e−04)6.7e-04
MESOR, midline s a is ic o hy hm; OA, oscilla ion ampli ude; ϕ, ac ophase. Resul s a e exp essed as median [25 h pe cen ile; 75 h pe cen ile] ep esen ing he ela i e
change compa ed o PRE (in %). The p- alue column e e s o he esul o he F iedman es . Resul s o he pos hoc Wilcoxon Signed Rank es wi h Bon e oni co ec ion
a e p esen ed as: ∗(p− alue):p<0.05/3 s. PRE; F(p− alue):p<0.05/3 s. HDT5.
in 5-day, −4.5[−6.0; −1.8]% in 21-day, −6.1[−7.0; −5.1]% in
60-day HDT). Also, a end wi h HDT du a ion was p esen
in he maximum alues, which exhibi ed a la ge dec ease a e
60-day HDT (RR: −15.6[−19.6; −12.6]%; QTend: −4.2[−7.7;
−3.1]%), bo h compa ed o 5-day (RR: −6.3[−9.1; −4.7]%;
QTend: −1.8[−2.7;1.1]%) and 21-day HDT (RR: −11.4[−14.8;
−8.4]%; QTend: −3.3[−6.4;6.8]%). As ega ds minimum alues,
no di e ence in QTend educ ion was ound, while a lowe 1was
isible o RR in 60-day HDT (−38.4[−43.3; −29]%) compa ed
o 5-day HDT (−46.3[−51.8; −40.2]%). In addi ion, 60-day HDT
elici ed a la ge MESOR educ ion (RR: −25.3[−34.0; −18]%;
QTend: −8.5[−14.2; −6.8]%) bo h compa ed o 5-day (RR:
−13.6[−16.8; −11.2]%; QTend: −4.5[−5.4; −3]%) and 21-day
HDT (RR: −16.5[−19.6; −14.2]%).
As ega ds he OA, la ge alues we e obse ed a R + 0
compa ed o he baseline, as e idenced by posi i e 1(%).
In pa icula , la ge 1(%) we e ound in 5-day HDT (RR:
+55.4[26.7;93.1]%; QTend: +87.4[50.4;113.1]%) compa ed
o 21-day HDT campaigns (RR: +13[−22;40.8]%; QTend:
+24.8[−15;73.1]%).
No di e ences in he a ia ion o ac ophases and nei he
in QTc we e ound.
DISCUSSION
In his s udy, we e ospec i ely analyzed he 24 h Hol e
ECG acqui ed a di e en epochs du ing six HDT bed es
campaigns ( wo o 5 days, wo o 21 days, and wo 60 days),
aiming a e alua ing he e ec s o he HDT du a ion on
he ci cadian hy hms o RR and en icula epola iza ion
(QT) o a la ge popula ion o no mal young males. To he
knowledge o he au ho s, his s udy includes he wides pooled
popula ion in which he analysis o he changes induced by
HDT in he ci cadian hy hms o ca diac elec ical ac i i y has
e e been conduc ed.
The cumula i e esul s con i med ou p e ious indings
obse ed on da a o a single campaign only (Solbia i e al., 2020),
showing ha he cha ac e is ics o RR and QTend ci cadian
hy hms we e a ec ed by HDT in e ms o inc eased RR
and QTend midline and sho ened QTc, educed ampli ude
o day/nigh oscilla ions and sligh ly an icipa ed ci cadian
ac ophase, wi h la ge changes isible al eady a e he i s 5 days
o HDT bed es . The conclusion o he HDT elici ed opposi e
changes, including he sho ening o RR and QTend, and he
p olonga ion o he QTc, as well as he inc eased OA and a sligh ly
delayed ac ophase du ing he eco e y s age. In addi ion, we we e
able o e alua e he deg ee o posi i e dependence o pos -HDT
changes wi h inc easing HDT du a ion.
E ec s on In e als Du a ion
Wi hin he i s 5 days o HDT, he midline alue (MESOR) o RR
and QTend in e als o e he 24 h eco ding inc eased, mos ly
due o he subs an ial leng hening o he RR in e al du ing
he day and o he inc eased minimum alues wi h una ec ed
F on ie s in Physiology | www. on ie sin.o g 9Janua y 2021 | Volume 11 | A icle 612188
phys-11-612188 Janua y 2, 2021 Time: 15:13 # 16
Solbia i e al. Head-Down Bed Res Ci cadian Decondi ioning
Ve heyden, B., Becke s, F., Couckuy , K., Liu, J., and Aube , A. E. (2007).
Respi a o y modula ion o ca dio ascula hy hms be o e and a e sho -
du a ion human space ligh . Ac a Physiol. 191, 297–308. doi: 10.1111/j.1748-
1716.2007.01744.x
Wa enpaugh, D. E. (2016). Analogs o mic og a i y: head-down il and wa e
imme sion. J. Appl. Physiol. 120, 904–914. doi: 10.1152/japplphysiol.00986.
2015
Xu, D., Shoemake , J. K., Blabe , A. P., A beille, P., F ase , K., and Hughson, R. L.
(2013). Reduced hea a e a iabili y du ing sleep in long-du a ion space ligh .
Am. J. Physiol. Regul. In eg . Comp. Physiol. 305, R164–R170. doi: 10.1152/
ajp egu.00423.2012
Yamamo o, N., O suka, K., Kubo, Y., Hayashi, M., Mizuno, K., Ohshima, H.,
e al. (2015). E ec s o long- e m mic og a i y exposu e in space on ci cadian
hy hms o hea a e a iabili y. Ch onobiol. In . 32, 327–340. doi: 10.3109/
07420528.2014.979940
Con lic o In e es : The au ho s decla e ha he esea ch was conduc ed in he
absence o any comme cial o inancial ela ionships ha could be cons ued as a
po en ial con lic o in e es .
Copy igh © 2021 Solbia i, Ma in-Yeb a, Vaïda and Caiani. This is an open-access
a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License
(CC BY). The use, dis ibu ion o ep oduc ion in o he o ums is pe mi ed, p o ided
he o iginal au ho (s) and he copy igh owne (s) a e c edi ed and ha he o iginal
publica ion in his jou nal is ci ed, in acco dance wi h accep ed academic p ac ice. No
use, dis ibu ion o ep oduc ion is pe mi ed which does no comply wi h hese e ms.
F on ie s in Physiology | www. on ie sin.o g 16 Janua y 2021 | Volume 11 | A icle 612188