1
Scien i ic REPORTS | 7: 10947 | DOI:10.1038/s41598-017-11468-7
www.na u e.com/scien i ic epo s
Volun a y liquo ice inges ion
inc eases blood p essu e ia
inc eased olume load, ele a ed
pe iphe al a e ial esis ance, and
dec eased ao ic compliance
Elina J. Hau aniemi1,2, Anna M. Tah anainen1,3, Jenni K. Koskela1,3, An i J. Tikkakoski1,4, Mika
Kähönen1,4, Ma ko Ui o1, Kalle Sipilä4, Onni Niemelä1,5, Jukka Mus onen1,3 & Ilkka H. Pö s i1,3
We in es iga ed he haemodynamic e ec s o wo-week liquo ice exposu e (glycy hizin dose 290–
370 mg/day) in 22 heal hy olun ee s du ing o hos a ic challenge. Haemodynamics we e eco ded
du ing passi e 10-minu e head-up il using adial pulse wa e analysis, whole-body impedance
ca diog aphy, and spec al analysis o hea a e a iabili y. Thi y age-ma ched heal hy subjec s
se ed as con ols. Liquo ice inges ion ele a ed adial sys olic (p < 0.001) and dias olic (p = 0.018)
blood p essu e and sys emic ascula esis ance (p = 0.037). Du ing o hos a ic challenge, hea a e
inc eased less a e he liquo ice e sus con ol die (p = 0.003) and low equency powe o hea
a e a iabili y dec eased wi hin he liquo ice g oup (p = 0.034). Liquo ice in ake inc eased cen al
pulse p essu e (p < 0.001) and augmen a ion index (p = 0.002) supine and up igh , bu in he up igh
posi ion he ele a ion o augmen a ion index was accen ua ed (p = 0.007). Liquo ice die also inc eased
ex acellula luid olume (p = 0.024) and ao ic o popli eal pulse wa e eloci y (p = 0.027), and ao ic
cha ac e is ic impedance in he up igh posi ion (p = 0.002). To conclude, in addi ion o inc eased
ex acellula luid olume and la ge a e ial s i ness, wo weeks o liquo ice inges ion ele a ed sys emic
ascula esis ance and augmen a ion index. Measu emen s pe o med a es may unde es ima e
he haemodynamic e ec s o liquo ice inges ion, as enhanced cen al wa e e lec ion and educed
ch ono opic esponse we e especially obse ed in he up igh posi ion.
The mine aloco icoid ecep o (MR)1 and he glucoco icoid ecep o (GR)2 play an impo an ole in he eg-
ula ion o blood p essu e (BP). Bo h MR and GR a e exp essed in se e al issues impo an o BP homeos asis,
including he kidney, ascula wall, cen al ne ous sys em, and he hea 2–4. Co isol binds o bo h GR and
MR, al hough aldos e one is he only physiologic agonis o he MR3. In aldos e one a ge issues he enzyme
11β-hyd oxys e oid dehyd ogenase ype 2 (11β-HSD2) p e en s co isol om binding o he MR by inac i a ing
i o co isone5. The enzyme 11β-HSD2 may be in ol ed in he pa hogenesis o hype ension, since dec eased
11β-HSD2 ac i i y inc eases he ac i a ion o he MR and GR by co isol5.
The ele a ion o BP a e liquo ice inges ion is well-known6–8. The ac i e me aboli e in liquo ice, glycy he inic
acid (GA) ha esembles he s uc u e o co isone, inhibi s he enzyme 11β-HSD26. In he kidney, 11β-HSD2
is exp essed in he dis al neph on, whe e MR ac i a ion by co isol p omo es sodium eabso p ion and po as-
sium exc e ion in o he u ine5. GA may also inhibi he hepa ic s e oid-me abolizing enzymes 5β- educ ase and
3β-hyd oxys e oid dehyd ogenase, esul ing in supp essed aldos e one ca abolism9. These mechanisms lead o
inc eased BP, sodium and wa e e en ion, dec eased plasma po assium and aldos e one concen a ion, and
dec eased plasma enin ac i i y6.
1Facul y o Medicine and Li e Sciences, FIN-33014 Uni e si y o Tampe e, Tampe e, Finland. 2Nu i ion Uni , Tampe e
Uni e si y Hospi al, Tampe e, 33521, Finland. 3Depa men o In e nal Medicine, Tampe e Uni e si y Hospi al,
Tampe e, 33521, Finland. 4Depa men o Clinical Physiology, Tampe e Uni e si y Hospi al, Tampe e, 33521, Finland.
5Depa men o Labo a o y Medicine and Medical Resea ch Uni , Seinäjoki Cen al Hospi al, Seinäjoki, 60220,
Finland. Co espondence and eques s o ma e ials should be add essed o E.J.H. (email: [email p o ec ed])
Recei ed: 16 May 2017
Accep ed: 24 Augus 2017
Published: xx xx xxxx
OPEN
www.na u e.com/scien i ic epo s/
2
Scien i ic REPORTS | 7: 10947 | DOI:10.1038/s41598-017-11468-7
The ascula wall (smoo h muscle and endo helial cells)10, he hea 4 and he b ain11 a e also exp ession si es
o he enzyme 11β-HSD2. In he ascula wall, 11β-HSD2 inhibi ion inc eases a e ial one h ough enhanced
con ac ile esponses o p esso ho mones and dec eased p oduc ion o endo helial ni ic oxide12, 13. In smoo h
muscle cells MR ac i a ion may con ibu e o ascula s i ening ia emodelling o he ascula wall1. In he
hea , glucoco icoids ha e been sugges ed o induce di ec e ec s h ough MR ac i a ion14. Fu he , he in usion
o GA in o he la e al en icle o he a b ain ele a es BP wi hou a ec ing sal o wa e homeos asis, sugges ing
a di ec in luence on he cen al ne ous sys em15.
We ecen ly epo ed ha daily liquo ice consump ion in no mo ensi e subjec s o wo weeks ele a ed
pe iphe al and cen al sys olic and dias olic BP, inc eased ex acellula luid olume, and ampli ied p essu e wa e
e lec ion om he pe iphe al a e ial ee8. Howe e , inc eased pe iphe al a e ial esis ance and la ge a e ial
s i ness, and changes in ca diac unc ion and au onomic one could also con ibu e o he ele a ion o BP ol-
lowing liquo ice inges ion. Since he up igh ca dio ascula in luences o liquo ice exposu e a e also unknown,
in he p esen s udy we in es iga ed he haemodynamic changes in heal hy subjec s du ing o hos a ic challenge
ollowing olun a y liquo ice in ake. The balance o ca diac au onomic one was e alua ed by he use o powe
spec al analysis o hea a e a iabili y (HRV)16.
Me hods
E hical s a emen . The in es iga ion was pe o med wi h he unde s anding and w i en in o med consen
o each indi idual, and was app o ed by he E hics Commi ee o he Tampe e Uni e si y Hospi al (s udy code
R07053M) con o ming o he p inciples ou lined in he Decla a ion o Helsinki. The s udy is egis e ed in he
da abase o clinical ials (ClinicalT ials.go , ID: NCT01742702, da e o egis a ion 29 o No embe 2012), and
is a pa o an in es iga ion on nonin asi e eco ding o hemodynamics (DYNAMIC-s udy; Eud aCT-numbe
2006-002065-39).
S udy subjec s. The s udy popula ion comp ised o 52 heal hy, no mo ensi e (a e age b achial o ice BP
117/71 mmHg) indi iduals aged be ween 21 and 58 yea s. Exclusion c i e ia we e o ice BP >140/90 mmHg,
ca dio ascula disease wi h egula medica ion, p egnancy, and consump ion o liquo ice >300 g ams pe week.
Subjec ec ui men and da a collec ion ha e been desc ibed p e iously8. The liquo ice g oup consis ed o 22 sub-
jec s (14 women and 8 men), and he aged-ma ched con ol g oup o 30 subjec s (17 women and 13 men). When
compa ed wi h ou p e ious epo ocused on haemodynamics a es 8, he p esen analyses included da a om
wo addi ional subjec s in he liquo ice g oup no included in he p e ious s udy.
As p e iously epo ed8, he egula medica ions in he liquo ice g oup we e 160 µg budesonide + 4.5 µg o -
mo e ol wice daily o as hma (n = 1), 5 mg esci alop am once daily o dep ession (n = 1), pos menopausal
oes ogen eplacemen he apy (n = 1) and o al con acep i es (n = 5), while in he con ol die g oup 5 emale
subjec s used o al con acep i es and 3 had ho monal in au e ine de ices. Fou subjec s we e cu en smoke s
and 1 was p e ious smoke in he liquo ice g oup, and 3 subjec s we e cu en and 4 p e ious smoke s in he
con ol g oup.
Design. The s udy design has been desc ibed in de ail elsewhe e8. B ie ly, his was an open-label s udy bu
subjec s in he con ol g oup we e no awa e o ac ing as con ols o he liquo ice g oup. In he liquo ice g oup,
he subjec s consumed comme cial liquo ice p oduc s (Hal a liquo iceTM o Kou ola liquo iceTM) daily o wo
weeks. P io o baseline measu emen s, liquo ice-con aining p oduc s we e no allowed o 3 weeks. Du ing he
in e en ion he inges ed dose o liquo ice was 120–300 g/d depending on he glycy hizin concen a ion o he
p oduc and he es ima ed dose o glycy hizin anged 290–370 mg/d. All subjec s main ained he liquo ice die
o wo weeks. In he con ol g oup, he subjec s we e asked o main ain hei habi ual die , and he epo ed e-
quency o liquo ice consump ion in each subjec was once pe mon h o lowe .
Physical and labo a o y examina ions we e pe o med o all pa icipan s o ensu e he sui abili y o he s udy,
and s uc u ed ques ionnai es we e u ilized o e iew he li es yle habi s and medical and amily his o y. A base-
line blood and u ine samples we e d awn a e abou 12 hou s o as ing wi h he excep ion ha samples we e no
ecei ed om one subjec in he con ol g oup8. Hemodynamic measu emen s we e pe o med be o e and a e
2 weeks o liquo ice inges ion o 1–3 weeks o con ol die . As excep ions, in he con ol die g oup h ee subjec s
had consecu i e measu emen s wi hin 2–4 days and wo subjec s wi hin app oxima ely 4 weeks.
Labo a o y analyses. S anda d 12-lead elec oca diog ams we e eco ded wi h MAC5000 (GE Heal hca e,
Chal on S . Giles, UK), and he eco dings we e no mal in all subjec s8. Plasma sodium, po assium, c ea inine,
glucose, choles e ol lipop o eins we e de e mined by Cobas In eg a 700/800 (F. Ho mann-LaRoche L d, Basel,
Swi ze land), and blood cell coun s by ADVIA 120 o 2120 (Baye Heal h Ca e, Ta y own, NY, USA).
Haemodynamic measu emen p o ocol. P eceding he haemodynamic eco dings, ca eine con aining
p oduc s, smoking and hea y meal o a leas 4 hou s, and alcohol o a leas 24 hou s we e o be a oided. The
measu emen p o ocol was conduc ed in a quie , empe a u e-con olled labo a o y by esea ch nu ses17. The
impedance ca diog aphy elec odes we e placed on body su ace, he onome ic senso o adial BP on he le
w is , and a b achial cu o BP calib a ion o he igh uppe a m18. The ex ended le a m was placed on an a m
suppo a he le el o he hea in supine and up igh posi ions. The haemodynamics we e eco ded con inuously
o 5 minu es in supine posi ion and o 5 minu es du ing o hos a ic challenge o 60 deg ees17. Fo he s a is ical
analyses, mean alues o each minu e o he 10-min eco dings we e calcula ed.
Pulse wa e analysis. Radial BP and pulse wa e o m we e con inuously cap u ed by a onome ic senso
(Colin BP-508T, Colin Medical Ins umen s Co p., USA). The adial BP eco dings we e calib a ed wice du -
ing bo h 5-minu e pe iods using con ala e al b achial BP measu emen s. Ao ic BP, pulse p essu e (PP) and
www.na u e.com/scien i ic epo s/
3
Scien i ic REPORTS | 7: 10947 | DOI:10.1038/s41598-017-11468-7
augmen a ion index (AIx) (augmen a ion p essu e/PP*100) we e de e mined using he SphygmoCo PWMx
pulse wa e moni o ing sys em (A co Medical, Aus alia)19.
Whole-body impedance ca diog aphy. A whole-body impedance ca diog aphy de ice (Ci cMonR, JR
Medical L d., Tallinn, Es onia), which eco ds he changes in body elec ical impedance du ing ca diac cycles,
was used o de e mine bea - o-bea hea a e (HR), s oke olume (SV), ca diac index (ca diac ou pu /body su -
ace a ea), pulse wa e eloci y (PWV), and ex acellula wa e (ECW) olume20, 21. The me hod desc ip ion and
elec ode con igu a ion ha e been p e iously epo ed20, 21. Sys emic ascula esis ance index (sys emic ascula
esis ance/body su ace a ea) (SVRI) was calcula ed om he adial BP signal and ca diac index measu ed by
Ci cMonR. Ao ic cha ac e is ic impedance was calcula ed as ollows: (cen al o wa d wa e ampli ude* en icu-
la ejec ion du a ion)/(2*s oke olume)22.
Wi h he Ci cMonR whole-body impedance ca diog aphy me hod, he eco ded PWV alues show excellen
co ela ion wi h alues measu ed using ul asound o he onome ic SphygmoCo me hod18, 20, he SV shows
good co ela ion wi h 3-dimensional echoca diog aphy eco dings23, ca diac ou pu alues a e in good ag eemen
wi h alues measu ed by he he modilu ion me hod21, and he ep oducibili y and epea abili y o he measu e-
men s a e good17, 23.
F equency domain analysis o hea a e a iabili y. The elec oca diog ams eco ded by he
Ci cMonR de ice (sampling a e 200 Hz), we e analyzed using Ma lab so wa e (Ma hWo ks Inc., Na ick,
Massachuse s, USA). No mal R-R in e als we e ecognized, and a bea was conside ed ec opic i he in e al
di e ed o e 20% om he p e ious alues. The a i ac s we e p ocessed using he cubic spline in e pola ion
me hod. The equency domain a iables we e calcula ed using he Fas Fou ie T ans o ma ion me hod: i) powe
in low equency (LF) ange (0.04–0.15 Hz), ii) powe in high equency (HF) ange (0.15–0.40 Hz), and iii) LF/
HF a io24.
S a is ical analyses. A minimum sample size o 17 expe imen al and 26 con ol subjec s was equi ed o
de ec a 9 mmHg di e ence in he change in sys olic BP om baseline wi h a s anda d de ia ion (SD) o 10,
α-le el o 0.05, and powe o 80%8. S a is ical analyses we e conduc ed using IBM SPSS S a is ics Ve sion 24
(IBM Co po a ion, A monk, NY, USA). No mally dis ibu ed da a was gi en as means wi h SD, s anda d e o
o he mean o 95% con idence in e al. LF and HF powe we e ans o med o na u al loga i hm be o e analyses
o yield no mal non-skewed dis ibu ions. The homogenei y o a iances was es ed wi h he Le ene’s es . The
mean haemodynamic alues we e calcula ed om he minu es 3–5 o he eco dings du ing supine and up igh
posi ions when he signal was mos s able, and he alues om he minu es 1–5 min we e used o analyses o LF
and HF powe . Independen samples - es was used o compa e baseline da a and changes be ween he g oups.
Analysis o a iance o epea ed measu emen s was applied o s udy in e ac ion be ween ime and g oup, and
di e ences be ween he g oups and o e ime in haemodynamic a iables and HRV du ing es and o hos a ic
challenge. In case o mul iple compa isons, he esul s we e adjus ed wi h he Bon e oni co ec ion, as app o-
p ia e. Chi-squa e es was applied o es non-con inuous a iables, and Spea man’s co ela ions ( S) we e calcu-
la ed, as app op ia e. P < 0.05 was conside ed s a is ically signi ican .
Da a a ailabili y. The da ase s gene a ed du ing and analysed du ing he cu en s udy a e no publicly
a ailable as ou clinical da abase con ains se e al indi ec iden i ie s and he in o med consen ob ained does no
allow publica ion o indi idual pa ien da a. Howe e , he da ase s a e a ailable om he co esponding au ho
on easonable eques .
Resul s
The baseline cha ac e is ics o he s udy subjec s a e p esen ed in Table1. The demog aphic da a, ou ine labo a-
o y alues, and ECW olumes did no di e be ween he g oups.
The P- alues in he igu es e e o he changes in haemodynamic a iables induced by he liquo ice die e -
sus con ols om week 0 o week 2. The o he s a is ical esul s a e gi en below in he ex .
Haemodynamic a iables a baseline. A week 0, he e we e no di e ences in adial sys olic and dias-
olic BP (Fig.1), HR (Fig.2), ao ic PP, AIx and ao ic cha ac e is ic impedance (Fig.3), o SVRI (Fig.4) du ing
he 10-min eco dings be ween he g oups. Howe e , analysis o a iance o epea ed measu emen s showed
a signi ican ime*g oup in e ac ion (p < 0.001) in he baseline analyses o SV (Fig.2), so ha he alues a es
we e highe in he liquo ice g oup han in he con ol g oup. In addi ion, a signi ican ime*g oup in e ac ion in
ca diac index (p = 0.003) a week 0 was obse ed (Fig.4) indica ing a highe up igh dec ease o ca diac ou pu
in he liquo ice g oup.
Haemodynamic in luences o 2-week liquo ice inges ion. A e wo weeks o liquo ice inges ion,
adial sys olic BP was ele a ed h oughou he 10-min eco ding p o ocol, while an inc ease in dias olic BP was
de ec ed in he supine posi ion (Fig.1). Du ing o hos a ic challenge he inc ease in HR (p = 0.003) was educed
a e liquo ice consump ion (Fig.2). Inc eased ao ic PP, AIx (Fig.3), and SVRI (Fig.4), we e also obse ed
a e he liquo ice die . The ele a ion o AIx in he up igh posi ion was e en accen ua ed a e he liquo ice die ,
as indica ed by he signi ican ime*g oup in e ac ion (Fig.3, p = 0.007), and in pa allel ao ic cha ac e is ic
impedance was inc eased (Fig.3). The e we e no signi ican di e ences in he analyses o changes in SV (Fig.2)
o ca diac index (Fig.4) om week 0 o 2 be ween he g oups. O no e, he mo e p onounced up igh dec ease in
ca diac index in he liquo ice g oup pe sis ed h oughou he s udy (Fig.4, p < 0.001 o he ime*g oup in e ac-
ion du ing he 10-min eco ding p o ocol a week 2).
www.na u e.com/scien i ic epo s/
4
Scien i ic REPORTS | 7: 10947 | DOI:10.1038/s41598-017-11468-7
The AIx is in luenced by PWV, a ma ke o a e ial s i ness, bu also o he ac o s including SVRI, HR, en-
icula ejec ion du a ion, SV, and ECW olume8, 18, 25. The e o e, we de e mined he co ela ions be ween hese
a iables in he liquo ice g oup: he inc ease o AIx in he up igh posi ion signi ican ly co ela ed wi h he
changes in HR ( S = −0.592, p = 0.004), SVRI ( S = 0.600, p = 0.003) and ejec ion du a ion ( S = 0.736, p < 0.001),
bu no wi h he changes in PWV, SV o ECW olume. Since he numbe o he s udy subjec s was a he small,
he use o mul i a ia e analysis o u he s a is ics was no easible.
A week 2 he mean changes in ECW olume we e −0.16 li es (95% CI −0.37 o 0.04) e sus + 0.57 li es
(95% CI −0.03 o 1.17) in he con ol e sus liquo ice g oups, espec i ely (p = 0.024 o be ween g oup di -
e ence). The PWV mean (SD) alues a baseline we e 7.1 (1.1) and 7.1 (0.7) m/s in he con ol and liquo ice
g oups (p = 0.880), espec i ely. Al hough he mean changes (95% CI) in PWV we e small wi hin he con ol
and liquo ice g oups; −0.18 m/s (−0.38 o 0.03) e sus 0.27 m/s (−0.11 o 0.65), espec i ely, hese changes we e
signi ican ly di e en be ween he g oups (p = 0.027).
Hea a e a iabili y a e 2-week liquo ice inges ion. HRV is dependen on a e age HR due o bo h
physiological and ma hema ical easons26. The ma hema ical dependency esul s om he nonlinea ela ionship
be ween RR in e al and HR, and as a consequence o his associa ion he HRV analysis may be biased26. In o de
o educe he HRV dependence on HR, we di ided he LF and HF powe by he a e age RR in e al squa ed26.
In he HR-co ec ed powe spec al analyses o HRV, he e we e no signi ican wi hin-g oup changes in he HF
powe o LF/HF a io ei he in supine o up igh posi ions (Fig.5, p > 0.05 o all compa isons). Al hough he e
we e no signi ican be ween-g oup di e ences in LF powe , in he liquo ice g oup he LF powe was dec eased
a e wo weeks (Fig.5, p = 0.034). As a mino be ween-g oup di e ence, he up igh LF/HF a io was lowe in
he liquo ice g oup han in he con ol g oup a e wo s udy weeks (Fig.5, p = 0.008).
Discussion
The liquo ice-induced ele a ion o BP has been a ibu ed o enal sodium and wa e e en ion6. In line wi h
hese iews, ou p e ious indings indica ed ha he ele a ion o BP a e liquo ice exposu e was associa ed wi h
inc eased ex acellula luid olume8. In he p esen s udy, we in es iga ed he haemodynamic in luences o
wo-week liquo ice in ake du ing o hos a ic challenge, and ound ha he exposu e inc eased ex acellula luid
olume and ele a ed adial sys olic and dias olic BP, SVRI, ao ic PP, and AIx. In addi ion, liquo ice die esul ed
in signi ican changes in PWV and ao ic cha ac e is ic impedance, wo indexes o ao ic s i ness22, 25, when
compa ed wi h he con ol g oup. Du ing o hos a ic challenge, liquo ice inges ion esul ed in a u he inc ease
o AIx indica ing enhanced p essu e wa e e lec ion om he pe iphe y, while a dec eased ca diac ch ono opic
esponse was also obse ed. In he powe spec al analyses o HRV, he LF powe was dec eased in he liquo ice
g oup a e wo weeks. Al oge he , he p esen esul s indica e ha he liquo ice-induced ele a ion o BP was due
o mul iple mechanisms: olume o e load, educed la ge a e ial compliance, inc eased pe iphe al ascula esis -
ance, and enhanced p essu e wa e e lec ion om he pe iphe y pa icula ly in he up igh posi ion.
BP is de ined as he p oduc o ca diac ou pu and pe iphe al ascula esis ance27. The p esen liquo ice and
con ol g oups demons a ed pe sis en unc ional di e ences in he egula ion o ca diac ou pu in he up igh
posi ion. Howe e , he be ween-g oup di e ence in ca diac index emained unchanged du ing he s udy, while
Con ol
(n = 29–30*)Liquo ice
(n = 22) P
alueMean SD Mean SD
Female/Male 17/13 14/8 0.776
Age (yea s) 33.8 8.1 34.9 9.2 0.656
Body mass index (kg/m2) 23.0 2.7 23.3 1.9 0.643
Wais ci cum e ence (cm)
Male 86.9 4.4 82.6 7.1 0.100
Female 78.5 10.3 76.3 7.4 0.521
Ex acellula wa e (l) 12.8 1.2 12.3 1.5 0.228
Hema oc i (%) 0.42 0.05 0.40 0.03 0.222
Fas ing plasma
Choles e ol (mmol/l) 4.4 0.8 4.6 0.8 0.366
T iglyce ides (mmol/l) 0.94 0.49 0.83 0.32 0.374
High densi y lipop o ein
(mmol/l) 1.72 0.39 1.83 0.36 0.305
Low densi y lipop o ein
(mmol/l) 2.2 0.7 2.4 0.6 0.478
Glucose (mmol/l) 5.0 0.4 5.2 0.4 0.151
C ea inine (µmol/l) 76 11 77 17 0.805
Sodium (mmol/l) 140 2.0 140 1.3 0.957
Po assium (mmol/l) 3.9 0.2 3.9 0.3 0.448
Table 1. Demog aphic da a and labo a o y alues a baseline. *Blood samples o as ing plasma alues we e no
ob ained om one subjec .
www.na u e.com/scien i ic epo s/
5
Scien i ic REPORTS | 7: 10947 | DOI:10.1038/s41598-017-11468-7
SVRI was ele a ed a e liquo ice inges ion. Se e al mechanisms may explain he obse ed inc ease in sys emic
ascula esis ance a e he liquo ice die . Glucoco icoid ac i i y is con olled locally and sys emically by he
ype 1 and 2 iso o ms o enzyme 11β-HSD5. Bo h 11β-HSD enzymes a e exp essed in he ascula wall, whe e
hey can in luence ascula one by egula ing ac i e glucoco icoid concen a ions13. The subsequen MR ac i a-
ion in smoo h muscle cells can esul in he emodelling o he ascula wall1. Glucoco icoids can e en po en i-
a e he asocons ic o ac ions o angio ensin II and ca echolamines in smoo h muscle, and supp ess asodila o y
sys ems including endo helial ni ic oxide syn hase and p os acyclin syn hesis28. The ac i a ion o GR may educe
Figu e 1. Radial sys olic blood p essu e (a,b) and dias olic blood p essu e (c,d) be o e and a e wo weeks o
con ol and liquo ice die . Haemodynamic da a was cap u ed con inuously du ing he 10-min eco dings, and
passi e head-up il was pe o med om 5 o 10 min. The g aphs depic mean and s anda d e o o he mean,
and he s a is ical analyses a e o he changes in he liquo ice g oup e sus con ols om week 0 o week 2.

www.na u e.com/scien i ic epo s/
6
Scien i ic REPORTS | 7: 10947 | DOI:10.1038/s41598-017-11468-7
neu onal ni ic oxide elease in he a e ies29, whe eby al e a ions in pe i ascula ni e gic unc ion may also
con ibu e o he glucoco icoid-induced inc ease in ascula esis ance.
Au onomic ne ous sys em plays a signi ican ole in he con ol o BP16. Du ing pos u al challenge, a dec ease
in BP is sensed by he ba o ecep o s o ca o id sinus and ao ic a ch, and a e en inpu o he nucleus o he
ac us soli a ies is dec eased16. Subsequen ly, e e en agal inpu o he sinoa ial node is educed, sympa he ic
inpu o he hea , a e ioles and enules is inc eased16, and he educ ion in BP is co ec ed ia inc eases in HR
and sys emic ascula esis ance16. In he p esen s udy, he HR esponse o o hos a ic challenge was a enua ed
a e liquo ice inges ion when compa ed wi h he con ol die . Simul aneously, he LF powe ha p edominan ly
e lec s sympa he ic ac i i y24, was dec eased in he liquo ice g oup. Also, he LF/HF a io, an indica o o ca diac
Figu e 2. S oke olume (a,b) and hea a e (c,d) be o e and a e wo weeks o con ol and liquo ice die .
Mean and s anda d e o o he mean, s a is ical analyses a e o he changes in he liquo ice g oup e sus
con ols om week 0 o week 2.
www.na u e.com/scien i ic epo s/
7
Scien i ic REPORTS | 7: 10947 | DOI:10.1038/s41598-017-11468-7
sympa ho agal balance24, was sligh ly lowe in he liquo ice g oup han in con ols a week 2. The esul s o he
powe spec al analyses we e in line wi h he obse ed changes in he con ol o HR a e liquo ice inges ion, and
indica e ha he e was no inc ease in ca diac sympa he ic one du ing he in e en ion. The e o e, he al e a ions
in sympa ho agal balance we e no he cause o he liquo ice-induced ele a ion o BP. P e ious expe imen al
indings sugges ha he changes in ca diac unc ion could esul om di ec MR o GA ac ion in he hea , as
aldos e one has been ound o dec ease hea a e and epola iza ion a e in abbi hea muscle cells30, while GA
has been epo ed o show a nega i e ino opic ac ion in he isola ed pe used a hea 31.
Figu e 3. Ao ic pulse p essu e (a,b), augmen a ion index (c,d), and ao ic cha ac e is ic impedance (e, )
be o e and a e wo weeks o con ol and liquo ice die . Mean and s anda d e o o he mean, s a is ical
analyses a e o he changes in he liquo ice g oup e sus con ols om week 0 o week 2.
www.na u e.com/scien i ic epo s/
8
Scien i ic REPORTS | 7: 10947 | DOI:10.1038/s41598-017-11468-7
The de e mina ion o PWV is conside ed he gold s anda d in he e alua ion o a e ial s i ness25. The p esen
wi hin-g oup changes in PWV a e he liquo ice and con ol die s we e no signi ican , bu he changes in PWV
be ween he wo g oups we e signi ican ly di e en . We also e alua ed la ge a e ial s i ness by calcula ing he
ao ic cha ac e is ic impedance, i.e. he impedance o he le en icula pulsa ile low22. Inc eased ao ic s i -
ness inc eases he cha ac e is ic impedance and o wa d wa e ampli ude as a esul om he misma ch be ween
p ope ies o he ao ic oo and peak ao ic low22. Al hough o wa d wa e ampli ude mainly ep esen s he
blood low going o wa d om he hea , he esul s mus be in e p e ed wi h cau ion, as wa e e lec ion om
he pe iphe al ci cula ion may also in luence he magni ude o o wa d wa e ampli ude32. We ound ha liquo-
ice inges ion inc eased ao ic cha ac e is ic impedance especially in he up igh posi ion. The p esen esul s
Figu e 4. Ca diac index (a,b) and sys olic ascula esis ance index (c,d) be o e and a e wo weeks o con ol
and liquo ice die . Mean and s anda d e o o he mean, s a is ical analyses a e o he changes in he liquo ice
g oup e sus con ols om week 0 o week 2.
www.na u e.com/scien i ic epo s/
9
Scien i ic REPORTS | 7: 10947 | DOI:10.1038/s41598-017-11468-7
hus sugges ha al eady wo weeks o liquo ice inges ion can inc ease la ge a e ial s i ness. This is p obably
explained by he ele a ion o BP, as he indices o a e ial s i ness signi ican ly depend on he p e ailing BP ha is
dis ending he blood essels25. I is unlikely ha s uc u al changes would ake place du ing wo weeks o liquo ice
exposu e, bu such changes a e bound o esul om longe pe iods o ele a ed BP27.
Figu e 5. Low equency powe (a,b), high equency powe (c,d) and low o high equency powe a io
(e, ) be o e and a e wo weeks o con ol and liquo ice die in he supine and up igh posi ion. Mean and
95% con idence in e al depic ed, *p = 0.034 o he change wi hin he liquo ice g oup, **p = 0.008 o he
di e ence in mean alues a week 2 be ween he g oups; analysis o a iance o epea ed measu emen s, pos -
hoc - es s we e adjus ed wi h Bon e oni co ec ions o mul iple compa isons.