nu ien s
Re iew
Classi ica ion o Hyd a ion in Clinical Condi ions:
Indi ec and Di ec App oaches Using Bioimpedance
Hen y C. Lukaski 1, Nicano Vega Diaz 2, An onio Tallu i 3and Lexa Nescola de 4,*
1Depa men o Kinesiology and Public Heal h Educa ion, Uni e si y o No h Dako a, G and Fo ks,
ND 58202-7166, USA; hen y[email p o ec ed]
2Neph ology Se ice, Uni e si y Hospi al o G and Cana y and Facul y o Science, Uni e si y Los Palmas,
35019 Los Palmas, G and Cana y, Spain; [email p o ec ed]
3An onio Tallu i, Fa by e, Inc., 50012 Bagno a Ripoli, Flo ence, I aly; [email p o ec ed]
4Depa men o Elec onic Enginee ing, Uni e si a Poli ècnica de Ca alunya, 08034 Ba celona, Spain
*Co espondence: [email p o ec ed]; Tel.: +34-93-4137286
Recei ed: 5 Ma ch 2019; Accep ed: 8 Ap il 2019; Published: 10 Ap il 2019


Abs ac :
Al hough he need o assess hyd a ion is well ecognized, labo a o y es s and clinical
imp essions a e imp ac ical and lack sensi i i y, espec i ely, o be clinically meaning ul. Di e en
app oaches use bioelec ical impedance measu emen s o o e come some o hese limi a ions and aid
in he classi ica ion o hyd a ion s a us. One indi ec app oach u ilizes single o mul iple equency
bioimpedance in eg ession equa ions and heo e ical models, espec i ely, wi h an h opome ic
measu emen s o p edic luid olumes (bioelec ical impedance spec oscopy—BIS) and es ima e
luid o e load based on he de ia ion o calcula ed o e e ence ex acellula luid olume.
Al e na i ely, bioimpedance ec o analysis (BIVA) uses di ec phase-sensi i e measu emen s o
esis ance and eac ance, measu ed a 50 kHz, no malized o s anding heigh , hen plo ed on
a bi a ia e g aph, esul ing in a ec o wi h leng h ela ed o luid con en , and di ec ion wi h
phase angle ha indexes hyd a ion s a us. Compa ison wi h heal hy popula ion no ms enables
BIVA o classi y (no mal, unde -, and o e -) and ank (change ela i e o p e- ea men ) hyd a ion
independen o body weigh . Each app oach has wide- anging uses in e alua ion and managemen
o clinical g oups wi h o e -hyd a ion wi h an e ol ing emphasis on p ognosis. This e iew
discusses he ad an ages and limi a ions o BIS and BIVA o hyd a ion assessmen wi h commen s
on u u e applica ions.
Keywo ds:
luid o e load; esis ance; eac ance; bioelec ical impedance ec o analysis; bioelec ical
impedance spec oscopy; malnu i ion
1. In oduc ion
Rou ine assessmen o hyd a ion s a us o suppo pa ien ca e pe sis s as a challenge.
The p eponde ance o me hods su e s om imp ac icali y and insensi i i y [
1
,
2
]. His o ically,
hyd a ion assessmen was synonymous wi h he use o ace dilu ion me hods o es ima e o al
body wa e (TBW), ex acellula wa e (ECW), and plasma olume (PV) wi h he calcula ion o
in acellula wa e (ICW, ICW = TBW
−
ECW). This app oach is in asi e, ime-consuming, cos ly,
and p one o ambiguous in e p e a ions o hyd a ion s a us because o eliance on he assump ion o
cons an TBW o body weigh (% TBW), which is p ejudiced by in e -indi idual di e ences in body
composi ion (adipose, muscle mass, o cachexia)
,
o TBW o a - ee mass (FFM) (% TBW/FFM) ha
equi es an independen de e mina ion o FFM [
3
–
6
]. Thus, he measu emen o luid olumes o
assess human hyd a ion s a us has limi ed alue.
Labo a o y es s o plasma o se um cons i uen s (osmolali y, sodium, na iu e ic pep ide, and
o he p omising p o eins), u ine cha ac e is ics (osmolali y, speci ic g a i y, conduc i i y, and 24-h
Nu ien s 2019,11, 809; doi:10.3390/nu11040809 www.mdpi.com/jou nal/nu ien s
Nu ien s 2019,11, 809 2 o 22
ou pu ), and sali a and ea composi ion (osmolali y) ha e inadequa e easibili y and lack sensi i i y
in a clinical se ing [
2
]. Imaging echniques ( adiog aphy and ul asound) and in asi e p ocedu es
(ca he iza ion) a e cos ly and ail o iden i y ea ly dis u bances in hyd a ion. Simple assessmen s such
as sel - epo ed hi s , pa ien his o y, and physical examina ion a e subjec i e and lack sensi i i y [
2
,
7
].
Thus, he need pe sis s o a me hod ha is non-in asi e, alid (accu a e and sensi i e), p ac ical,
cos -e ec i e, and p o ides eal- ime disc imina ion wi hin he spec um o he hyd a ion s a us.
Classi ica ion o hyd a ion as no mal, less, o g ea e han no mal is a p ac ical and alid app oach ha
can o e come hese limi a ions. Bu geoning e idence suppo s bioelec ical impedance as a unique
and p omising me hod o classi y he hyd a ion s a us o an indi idual.
Bioimpedance (BI) is he gene al e m desc ibing he sa e, non-in asi e measu emen o he
passi e elec ical cha ac e is ics o an o ganism a e in oduc ion o a painless, low-le el al e na ing
cu en in o he body [
8
–
10
]. Bioimpedance analysis (BIA), howe e , augmen s BI measu emen s
wi h heo e ical biophysical models and s a is ical ( eg ession) equa ions con ibu ing o “black box”
p edic ions o luid olumes ha can be un eliable and imp ecise. Bioimpedance pe se o e s di ec ,
uncomplica ed measu emen s ha acili a e a p ac ical and alid app oach o moni o hyd a ion.
This e iew emphasizes he use o BI o classi y hyd a ion. I desc ibes he biophysical basis
o he BI me hod, summa izes di e en echnical app oaches and hei limi a ions o es ima e luid
olumes, discusses ad an ages o BI measu emen s compa ed o calcula ed luid olumes, highligh s
clinical applica ions o BI emphasizing i s eme ging ole in p ognosis, and p o ides ecommenda ions
o imp o e and expand he use o BI in clinical applica ions.
2. Bioelec ical Impedance
2.1. Bioimpedance Basics
The physical basis o he BI me hod is he awa eness ha he human body is a ne wo k o esis o s
and capaci o s [
8
]. Physiological luids beha e as esis o s and cell memb anes ac as capaci o s
(Figu e 1). Thus, he body may be ep esen ed as a pa allel esis o -capaci o (RC) equi alen ci cui
(Figu e 2) in which he in oduced al e na ing cu en di ides in o esis i e ( luid and elec oly es)
and capaci i e (cell memb anes and issue in e aces) pa hways. The lowe speci ic esis i i y
o body luids and issues con aining wa e and elec oly es compa ed o in ac lipid-laden cell
memb anes [
11
] enables equency-dependen impedance measu emen s [
12
]. A low-le el al e na ing
cu en con inuously passes p edominan ly h ough he esis i e componen bu concu en ly delayed
( empo a ily s o ed) by capaci i e elemen s in issues and issue in e aces. Impo an ly, a a iable
amoun o e y low- equency cu en , ega dless a which equency he cu en is in oduced,
can pene a e he memb anes o muscle cells, pa icula ly when he cu en is pa allel o he muscle
ibe [
13
]. Cell memb anes a e capaci i e elemen s su ounding he in acellula luid (ICF) in a se ies
ci cui ha exis s in pa allel wi h he wa e -con aining in e s i ial gel o ex acellula luid (ECF).
Any al e na ing cu en will pene a e he capaci i e componen ( eac ance, Xc) o he cell memb ane
in p opo ion o he equency o he applied cu en . Reac ance is in e sely ela ed o equency ( )
and capaci ance (C, Fa ads): Xc (ohm,
Ω
) = 1/[(2
π•
•
C)]. Capaci ance is he abili y o a sys em o
ci cui o s o e an elec ical cha ge.
Nu ien s 2019,11, 809 3 o 22
Nu ien s 2019, 11, x FOR PEER REVIEW 3 o 21
Figu e 1. Illus a ion o he body as a ne wo k o esis o s and capaci o s in a pa allel con igu a ion.
The al e na ing cu en usually exceeds 1 kHz and ypically is 50 kHz. C
M
is memb ane capaci ance
and R
e
and R
i
is ex acellula and in acellula esis ance, espec i ely.
Figu e 2. Rep esen a ion o he body as a pa allel esis o -capaci o (RC) equi alen ci cui . Delay o
he cu en pene a ion a he cell memb ane causes an ou -phasing o cu en .
A low equencies (e.g., 1–5 kHz), al e na ing cu en lows la gely in he ECF bu no in a di ec
o ixed p opo ion ela i e o he ICF. Simila ly, a 50 kHz al e na ing cu en does no dis ibu e in
p opo ion o luid dis ibu ion (ex a- o in a-cellula luid olume) bu ela i e o capaci i e
elemen s. Thus, he phase angle (PhA) is equency-dependen , p incipally due o he amoun o Xc,
and is an index o he amoun o applied cu en ha pene a es he capaci i e elemen o cell
memb anes. Cu en ha is es ic ed o delayed a (in ac ) cell memb anes becomes ou -phased om
he ol age d op ha occu s a he cell.
Impedance (Z) is he b oad e m desc ibing he opposi ion o he low o al e na ing cu en by
any biological conduc o , and is de ined by he ype o elec ical cu en in oduced o a ci cui . When
di ec cu en is applied, he o al conduc o is e med esis ance (R) whe eas when al e na ing
cu en is in oduced, i is called Z. Di ec cu en passes only h ough esis i e elemen s bu
al e na ing cu en lows h ough esis i e and capaci i e elemen s. The esis i e (R) componen o Z
is independen o equency so i has he same measu emen alue (Ω) when ei he di ec o
al e na ing cu en is used. The capaci o in a biological ci cui se es as an insula o and may be
en isioned as an imagina y componen o Z when di ec cu en is in oduced. Howe e , when
al e na ing cu en is p esen , he imagina y componen ac s as a poo conduc o wi h capaci i e
Figu e 1.
Illus a ion o he body as a ne wo k o esis o s and capaci o s in a pa allel con igu a ion.
The al e na ing cu en usually exceeds 1 kHz and ypically is 50 kHz. C
M
is memb ane capaci ance
and Reand Riis ex acellula and in acellula esis ance, espec i ely.
Nu ien s 2019, 11, x FOR PEER REVIEW 3 o 21
Figu e 1. Illus a ion o he body as a ne wo k o esis o s and capaci o s in a pa allel con igu a ion.
The al e na ing cu en usually exceeds 1 kHz and ypically is 50 kHz. C
M
is memb ane capaci ance
and R
e
and R
i
is ex acellula and in acellula esis ance, espec i ely.
Figu e 2. Rep esen a ion o he body as a pa allel esis o -capaci o (RC) equi alen ci cui . Delay o
he cu en pene a ion a he cell memb ane causes an ou -phasing o cu en .
A low equencies (e.g., 1–5 kHz), al e na ing cu en lows la gely in he ECF bu no in a di ec
o ixed p opo ion ela i e o he ICF. Simila ly, a 50 kHz al e na ing cu en does no dis ibu e in
p opo ion o luid dis ibu ion (ex a- o in a-cellula luid olume) bu ela i e o capaci i e
elemen s. Thus, he phase angle (PhA) is equency-dependen , p incipally due o he amoun o Xc,
and is an index o he amoun o applied cu en ha pene a es he capaci i e elemen o cell
memb anes. Cu en ha is es ic ed o delayed a (in ac ) cell memb anes becomes ou -phased om
he ol age d op ha occu s a he cell.
Impedance (Z) is he b oad e m desc ibing he opposi ion o he low o al e na ing cu en by
any biological conduc o , and is de ined by he ype o elec ical cu en in oduced o a ci cui . When
di ec cu en is applied, he o al conduc o is e med esis ance (R) whe eas when al e na ing
cu en is in oduced, i is called Z. Di ec cu en passes only h ough esis i e elemen s bu
al e na ing cu en lows h ough esis i e and capaci i e elemen s. The esis i e (R) componen o Z
is independen o equency so i has he same measu emen alue (Ω) when ei he di ec o
al e na ing cu en is used. The capaci o in a biological ci cui se es as an insula o and may be
en isioned as an imagina y componen o Z when di ec cu en is in oduced. Howe e , when
al e na ing cu en is p esen , he imagina y componen ac s as a poo conduc o wi h capaci i e
Figu e 2.
Rep esen a ion o he body as a pa allel esis o -capaci o (RC) equi alen ci cui . Delay o
he cu en pene a ion a he cell memb ane causes an ou -phasing o cu en .
A low equencies (e.g., 1–5 kHz), al e na ing cu en lows la gely in he ECF bu no in a di ec
o ixed p opo ion ela i e o he ICF. Simila ly, a 50 kHz al e na ing cu en does no dis ibu e in
p opo ion o luid dis ibu ion (ex a- o in a-cellula luid olume) bu ela i e o capaci i e elemen s.
Thus, he phase angle (PhA) is equency-dependen , p incipally due o he amoun o Xc, and is
an index o he amoun o applied cu en ha pene a es he capaci i e elemen o cell memb anes.
Cu en ha is es ic ed o delayed a (in ac ) cell memb anes becomes ou -phased om he ol age
d op ha occu s a he cell.
Impedance (Z) is he b oad e m desc ibing he opposi ion o he low o al e na ing cu en by any
biological conduc o , and is de ined by he ype o elec ical cu en in oduced o a ci cui . When di ec
cu en is applied, he o al conduc o is e med esis ance (R) whe eas when al e na ing cu en is
in oduced, i is called Z. Di ec cu en passes only h ough esis i e elemen s bu al e na ing cu en
lows h ough esis i e and capaci i e elemen s. The esis i e (R) componen o Z is independen o
equency so i has he same measu emen alue (
Ω
) when ei he di ec o al e na ing cu en is used.
The capaci o in a biological ci cui se es as an insula o and may be en isioned as an imagina y
componen o Z when di ec cu en is in oduced. Howe e , when al e na ing cu en is p esen ,
he imagina y componen ac s as a poo conduc o wi h capaci i e esis ance and is e med eac ance
Nu ien s 2019,11, 809 4 o 22
(Xc), which is equency-dependen . Thus, Z is a complex numbe , Z
2
= R
2
+ Xc
2
, and cha ac e izes
he speci ic luid and cellula componen s o an o ganism. Impedance is a ec o wi h leng h and
posi ion on he bi a ia e RXc plo (Figu e 3). The PhA desc ibes he posi ion o Z and is he angula
ans o ma ion o Xc o R [a c angen (Xc/R) •(180◦/π)] exp essed in adian deg ees.
Nu ien s 2019, 11, x FOR PEER REVIEW 4 o 21
esis ance and is e med eac ance (Xc), which is equency-dependen . Thus, Z is a complex numbe ,
Z
2
= R
2
+ Xc
2
, and cha ac e izes he speci ic luid and cellula componen s o an o ganism. Impedance
is a ec o wi h leng h and posi ion on he bi a ia e RXc plo (Figu e 3). The PhA desc ibes he
posi ion o Z and is he angula ans o ma ion o Xc o R [a c angen (Xc/R) • (180°/𝜋)] exp essed in
adian deg ees.
Figu e 3. Geome ic ela ionships among he esis ance, eac ance (capaci ance, C
M
), impedance, and
phase angle.
2.2. Measu emen o Bioelec ical Impedance
Assessmen o whole-body hyd a ion uses su ace, gel (sil e -sil e chlo ide) o mode n
hyd ogel elec odes, and e apola elec ode placemen s. Whole-body BI measu emen s employ
adi ional limb placemen s wi h pai ed cu en -in oducing (sou ce) and ol age-d op-sensing
(de ec o ) elec odes placed on he dis al w is and ankle [8,10].
3. Volume Quan i ica ion in Hyd a ion Assessmen : Limi a ions and Imp ecision
A p ima y ac o o eliable hyd a ion assessmen is he echnical alidi y o he BI
measu emen . I includes he accu acy o he ins umen , which is de e mined wi h a p ecision (<1%)
ci cui o a esis o and a capaci o in a pa allel ci cui , and p ecision o ep oducibili y ha equi es
epea ed measu emen s o a alida ion ci cui and epe i i e in i o measu emen s. The echnical
alidi y should be <2% o all BI measu emen s [9]. Simila ly, he alidi y o he e e ence ace
dilu ion me hods mus be assessed wi h echnical accu acy and p ecision de e mined o be <2%.
In e -indi idual biological sou ces o e o o he ace dilu ion me hods, including hyd ogen
exchange and anion (b omine-chlo ine) shi s ha a e >4% [14,15], di ec ly a ec he a iabili y o
luid olume es ima es.
3.1. Single-F equency Bioimpedance
Clinical in es iga o s applied di e en BI app oaches o es ima e luid olumes as indica o s o
hyd a ion. They mos ly used phase-sensi i e single- equency (50 kHz) BI because o he high signal-
o-noise a io [9,16] o mul iple- equency (5 kHz o 1 MHz) measu emen s coupled wi h ei he
mul iple eg ession equa ions o heo e ical biophysical models o p edic luid olumes [10,17–19].
Each o hese app oaches has no able conce ns ha limi hei gene al applica ion in he es ima ion
o luid olumes. The elec ical olume model adap ed om Ohm’s Law is he physical basis o
es ima ion o luid olumes wi h single- and mul iple- equency BI measu emen s [10]. The olume
(V) o a conduc o ( luid plus elec oly es) depends on he leng h (L) and c oss-sec ional a ea (A) o
a cylind ical conduc o wi h cons an geome y and composi ion (speci ic esis i i y o ρ), so ha V
= ρ(L
2
/R). S anding heigh (H ) is a biological su oga e o L, so V = ρ(H
2
/R). The assump ions o his
model, which include he body is a uni o m cylinde o cons an chemical composi ion (e.g., wa e
Figu e 3. Geome ic ela ionships among he esis ance, eac ance (capaci ance, CM), impedance, and
phase angle.
2.2. Measu emen o Bioelec ical Impedance
Assessmen o whole-body hyd a ion uses su ace, gel (sil e -sil e chlo ide) o mode n hyd ogel
elec odes, and e apola elec ode placemen s. Whole-body BI measu emen s employ adi ional limb
placemen s wi h pai ed cu en -in oducing (sou ce) and ol age-d op-sensing (de ec o ) elec odes
placed on he dis al w is and ankle [8,10].
3. Volume Quan i ica ion in Hyd a ion Assessmen : Limi a ions and Imp ecision
A p ima y ac o o eliable hyd a ion assessmen is he echnical alidi y o he BI measu emen .
I includes he accu acy o he ins umen , which is de e mined wi h a p ecision (<1%) ci cui
o a esis o and a capaci o in a pa allel ci cui , and p ecision o ep oducibili y ha equi es
epea ed measu emen s o a alida ion ci cui and epe i i e
in i o
measu emen s. The echnical
alidi y should be <2% o all BI measu emen s [
9
]. Simila ly, he alidi y o he e e ence ace
dilu ion me hods mus be assessed wi h echnical accu acy and p ecision de e mined o be <2%.
In e -indi idual biological sou ces o e o o he ace dilu ion me hods, including hyd ogen
exchange and anion (b omine-chlo ine) shi s ha a e >4% [
14
,
15
], di ec ly a ec he a iabili y
o luid olume es ima es.
3.1. Single-F equency Bioimpedance
Clinical in es iga o s applied di e en BI app oaches o es ima e luid olumes as indica o s
o hyd a ion. They mos ly used phase-sensi i e single- equency (50 kHz) BI because o he high
signal- o-noise a io [
9
,
16
] o mul iple- equency (5 kHz o 1 MHz) measu emen s coupled wi h ei he
mul iple eg ession equa ions o heo e ical biophysical models o p edic luid olumes [
10
,
17
–
19
].
Each o hese app oaches has no able conce ns ha limi hei gene al applica ion in he es ima ion
o luid olumes. The elec ical olume model adap ed om Ohm’s Law is he physical basis o
es ima ion o luid olumes wi h single- and mul iple- equency BI measu emen s [
10
]. The olume
(V) o a conduc o ( luid plus elec oly es) depends on he leng h (L) and c oss-sec ional a ea (A) o
a cylind ical conduc o wi h cons an geome y and composi ion (speci ic esis i i y o
ρ
), so ha
V = ρ(L2/R)
. S anding heigh (H ) is a biological su oga e o L, so V =
ρ
(H
2
/R). The assump ions o
his model, which include he body is a uni o m cylinde o cons an chemical composi ion (e.g., wa e
con en ) o he conduc o , a e open o c i icism [
18
]. Speci ically, he body consis s o i e cylinde s
Nu ien s 2019,11, 809 5 o 22
( wo a ms, wo legs, and he ho ax) wi h di e en geome y ( a ying leng h and diame e ) and
di e en chemical composi ion (speci ic esis i i ies o di e en issues) including a dispu ed cons an
hyd a ion o he a - ee body (73%). Resea che s exploi ed his bioelec ical olume model, oge he
wi h a ious an h opome ic in o ma ion (H , weigh (W ), age, and gende ), o de elop and alida e
mul iple eg ession p edic ion equa ions using single- and mul iple- equency measu emen s o TBW
es ima es in heal hy and ill adul s [18].
A echnical issue exis s wi h any applica ion o hese eg ession models. The independen a iable
is R and no Z. Thus, only phase-sensi i e BI ins umen s should be used in he applica ion o speci ic
single- equency p edic ion models using R as an independen a iable. Al hough Z and R a e highly
co ela ed (R
2
> 0.9), he magni ude o Z is g ea e han R (1% o 2%) because Z includes he alue o
Xc, which is no negligible (~10% o R) and includes biologically impo an in o ma ion o hyd a ion
classi ica ion [
9
,
10
]. Simila ly, p edic ions based on mul i- equency BI measu emen s should ensu e
ha app op ia e de ices a e used o ob ain he equi ed Z o R measu emen s.
C i ical e alua ion o a popula ion-based model o p edic TBW highligh s a p ac ical limi a ion
o he use o BIA o es ima e he luid olume. Sun e al. [
20
] de i ed and alida ed TBW p edic ion
equa ions in a la ge sample o heal hy adul s o di e se e hnici y:
Males: TBW (L) = 0.87 + 0.43 (H 2/R) + 0.20 W
Females: TBW (L) = 3.27 + 0.45 (H 2/R) + 0.12 W
C oss- alida ion o hese equa ions e ealed a la ge p edic i e e o (s anda d e o o es ima e
(SEE)) o 3.8 L and 2.6 L, o he males and emales, espec i ely. The magni ude o hese p edic ion
e o s is oo la ge (e.g., imp ecise) o es ima e TBW o o iden i y change in TBW o an indi idual and,
he e o e, limi s hese equa ions o obse a ional o epidemiological s udies.
3.2. Bioelec ical Impedance Spec oscopy (BIS)
Some BI ins umen s measu e Z o e a wide ange o equencies (e.g., 5 kHz up o 1 MHz), epo
a ious measu emen s and hen use p og ammed so wa e o es ima e luid olumes [
17
]. Ce ain
non-phase-sensi i e de ices only measu e Z and p o ide he a io o Z a low and high equencies
(Z
low
/Z
high
) ha is equi alen o PhA a 50 kHz. Simila ly, hey also use he Z
low
/Z
high
o compu e
he phase and canno pe o m alid calcula ions o o m he impedance modulus o he Cole plo .
The BIS me hod elies on he assump ion ha low- equency cu en lows h ough he ECF
and high- equency cu en pene a es ECF and ICF. These assump ions a e open o c i icism la gely
because hey we e de i ed om
in i o
s udies o cells suspended in luid and igno e he cell-cell
in e aces ha occu in issues [
10
]. Addi ionally, in es iga o s ag ee ha a he lowes equencies
some cu en pene a es cell memb anes (e.g., ex acellula and in acellula pa hs) and ha a he
highes equencies cu en does c oss all cells (e.g., ex acellula and in acellula pa hs) [11,21–23].
The BIS me hod uses non-linea leas -squa e (polynomial) modeling o limi ed phase-sensi i e
mul i- equency measu emen s o Z and Xc, calcula ed om PhA, hen ex apola es hese alues o
gene a e a Cole plo (Figu e 4). This ma hema ical model [
24
] calcula es heo e ical esis i i y alues
ha a e used o app oxima e luid olumes. The ex apola ed (modeled) a iables a e R
0
, also e med
as Re ( he esis ance o he ex acellula luid o in e s i ial esis i i y), and R
∞
( esis ance associa ed
wi h he sum o ex acellula and in acellula luids), c i ical equency (
C
) o he equency a which
Xc is maximal, and memb ane capaci ance (Cm). The alue o calcula ed Cm depends on ano he i ed
a iable, ime delay (Td), which ep esen s he e ec o cell memb ane and o ien a ion on he cu en
ansmission in he body. These esis ance pa ame e s a e used o calcula e in acellula esis ance (Ri)
as 1/Ri = 1/R∞−1/Re.
Mix u e heo y uses hese calcula ed esis ance pa ame e s wi h an h opome ic measu emen s
and some wide- anging assump ions o es ima e luid olumes [
17
,
25
,
26
]. I elies on a heo e ical
model ha es ima es appa en conduc i i y in a he e ogeneous en i y composed o conduc i e (wa e
and ions) and non-conduc i e (anhyd ous ma e ials such as bone, a , and cell memb anes) componen s
such as he body. Fluid olumes, ECW, and ICW a e p edic ed sepa a ely and summed o ob ain

Nu ien s 2019,11, 809 6 o 22
he TBW. Calcula ion o he ECW and ICW depends on he es ima ed Cole esis ance pa ame e s,
body scala ac o s calcula ed om he H and W o an indi idual, an assumed body densi y alue
de i ed om body mass index (BMI), empi ically de i ed gende -speci ic esis i i y alues, and o he
assump ions. Ex acellula luid olume (V
ECF
) is calcula ed as V
ECF
= k
ECF •
(H
2•
W
0.5
/Re)
2/3
wi h
k
ECF
= 10
-3 •
(K
B2•ρECF2
/D
b
)
1/3
, whe e K
B
is a body geome y ac o ha ela es ela i e olume o
he legs, a ms, and unk,
ρECF
is he esis i i y o he ex acellula luid, and D
b
is o al body densi y.
In acellula luid olume (V
ICF
) also is calcula ed om he de i ed Cole esis ance a iables and
gende -dependen esis i i ies o he ECW and ICW. These calcula ions a e pe o med wi h so wa e
p o ided by he manu ac u e s o he di e en BIS ins umen s, and a e subjec o change.
Nu ien s 2019, 11, x FOR PEER REVIEW 6 o 21
o ob ain he TBW. Calcula ion o he ECW and ICW depends on he es ima ed Cole esis ance
pa ame e s, body scala ac o s calcula ed om he H and W o an indi idual, an assumed body
densi y alue de i ed om body mass index (BMI), empi ically de i ed gende -speci ic esis i i y
alues, and o he assump ions. Ex acellula luid olume (V
ECF
) is calcula ed as V
ECF
= k
ECF
• (H
2
•
W
0.5
/Re)
2/3
wi h k
ECF
= 10
-3
• (K
B2
• ρ
ECF2
/D
b
)
1/3
, whe e K
B
is a body geome y ac o ha ela es ela i e
olume o he legs, a ms, and unk, ρ
ECF
is he esis i i y o he ex acellula luid, and D
b
is o al
body densi y. In acellula luid olume (V
ICF
) also is calcula ed om he de i ed Cole esis ance
a iables and gende -dependen esis i i ies o he ECW and ICW. These calcula ions a e pe o med
wi h so wa e p o ided by he manu ac u e s o he di e en BIS ins umen s, and a e subjec o
change.
Figu e 4. Plo o eac ance and esis ance o a heal hy male ob ained wi h a Xi on 4200 and de i ed
using non-linea cu e- i ing so wa e based on he Cole model. No e ha he majo i y o alues
(dashed lines) we e es ima ed. R
0
and R
∞
we e calcula ed and hey app oxima e esis ance a 0 and
he highes equency, espec i ely.
Unce ain basic assump ions and cons an s o he o iginal BIS app oach [26–29] esul ed in
e o s in he p edic ions o luid olumes. Speci ically, signi ican e o s in he es ima ion o TBW (2
L) and ECW (~1 L) in indi iduals wi h inc easing adiposi y, cha ac e ized using BMI, we e explained
as he e ec o inc easing adipose issue on he assumed esis i i y o he ECW [30–32]. A p oposed
emedy o his limi a ion was he use o BMI as a p oxy o adiposi y [33].
A mul i-cen e alida ion ial [33] o his e ised BIS app oach included heal hy adul s and
dialysis pa ien s and ound ha his change imp o ed he sensi i i y o he e ised ECW p edic ions
by dec easing he a iabili y o he ECW es ima e by 24% (0.6 L) in heal hy and dialysis pa ien s and
educing he a iabili y o he TBW p edic ion in adul s wi h BMI alues o less han 20 and g ea e
han 30 kg/m
2
. Al hough no signi ican di e ences be ween ace dilu ion e e ence measu emen s
and BIS p edic ions o he TBW, ECW, and ICW we e ound, he e we e wide limi s o ag eemen
(95% con idence in e als—95% CI). Speci ically, he mean di e ences we e small wi h wide limi s
o ag eemen o p edic ion o he TBW (−0.2 L (95% CI: 4.6 L)), ECW (−0.4 L; 95% CI: 2.9 L)), and ICW
(0.2 L (95% CI: 4.1 L)). A limi a ion o he use o BMI as a su oga e o body a ness is he insensi i i y
o BMI o eliably di e en ia e he body composi ion (body a and muscle mass) o an indi idual,
heal hy o ill [34]. These wide limi s o ag eemen con ibu e o poo sensi i i y o he ECW and TBW
es ima es es ima ed wi h BIS o indi iduals.
3.3. Compa ison o Fluid Volumes Es ima ed wi h Single-F equency Bioimpedance and Bioelec ical
Impedance Spec oscopy
Al hough a common opinion is ha BIS, compa ed o a single- equency BIA, p o ides mo e
accu a e es ima es o luid olume [23], expe imen al da a ail o con i m his supposi ion. Impedance
p edic ions o he TBW and ICW we e equally accu a e in compa ison o iso ope dilu ion es ima es
Figu e 4.
Plo o eac ance and esis ance o a heal hy male ob ained wi h a Xi on 4200 and de i ed
using non-linea cu e- i ing so wa e based on he Cole model. No e ha he majo i y o alues
(dashed lines) we e es ima ed. R
0
and R
∞
we e calcula ed and hey app oxima e esis ance a 0 and
he highes equency, espec i ely.
Unce ain basic assump ions and cons an s o he o iginal BIS app oach [
26
–
29
] esul ed in e o s
in he p edic ions o luid olumes. Speci ically, signi ican e o s in he es ima ion o TBW (2 L) and
ECW (~1 L) in indi iduals wi h inc easing adiposi y, cha ac e ized using BMI, we e explained as he
e ec o inc easing adipose issue on he assumed esis i i y o he ECW [
30
–
32
]. A p oposed emedy
o his limi a ion was he use o BMI as a p oxy o adiposi y [33].
A mul i-cen e alida ion ial [
33
] o his e ised BIS app oach included heal hy adul s and
dialysis pa ien s and ound ha his change imp o ed he sensi i i y o he e ised ECW p edic ions
by dec easing he a iabili y o he ECW es ima e by 24% (0.6 L) in heal hy and dialysis pa ien s and
educing he a iabili y o he TBW p edic ion in adul s wi h BMI alues o less han 20 and g ea e
han 30 kg/m
2
. Al hough no signi ican di e ences be ween ace dilu ion e e ence measu emen s
and BIS p edic ions o he TBW, ECW, and ICW we e ound, he e we e wide limi s o ag eemen
(95% con idence in e als—95% CI). Speci ically, he mean di e ences we e small wi h wide limi s o
ag eemen o p edic ion o he TBW (
−
0.2 L (95% CI: 4.6 L)), ECW (
−
0.4 L; 95% CI: 2.9 L)), and ICW
(0.2 L (95% CI: 4.1 L)). A limi a ion o he use o BMI as a su oga e o body a ness is he insensi i i y
o BMI o eliably di e en ia e he body composi ion (body a and muscle mass) o an indi idual,
heal hy o ill [
34
]. These wide limi s o ag eemen con ibu e o poo sensi i i y o he ECW and TBW
es ima es es ima ed wi h BIS o indi iduals.
3.3. Compa ison o Fluid Volumes Es ima ed wi h Single-F equency Bioimpedance and Bioelec ical
Impedance Spec oscopy
Al hough a common opinion is ha BIS, compa ed o a single- equency BIA, p o ides mo e
accu a e es ima es o luid olume [
23
], expe imen al da a ail o con i m his supposi ion. Impedance
p edic ions o he TBW and ICW we e equally accu a e in compa ison o iso ope dilu ion es ima es bu
Nu ien s 2019,11, 809 7 o 22
included a signi ican bias (e o ) in single- equency and BIS p edic ion o he ECW in s eady-s a e
hemodialysis pa ien s [
35
]. Bo h impedance me hods had a p opo ional e o in es ima ion o he
ICW la gely due o assump ions o he BIS model and he single equency equa ion used o p edic
he ICW wi h calcula ions om o al body po assium, a speci ic ma ke o body cell mass. Simila
indings came om ano he s udy o hemodialysis pa ien s [
36
]. A consis en obse a ion is he wide
limi s o ag eemen be ween he impedance and e e ence me hods ha cau ions agains he use o
hese me hods o indi idual assessmen o luid olumes [37].
O he esea ch suppo s hese indings. In adul s ecei ing g ow h ho mone eplacemen he apy,
he TBW was de e mined using iso ope dilu ion and p edic ed om BI measu emen s, collec ed wi h
he same mul i- equency ins umen , using 50 kHz alues and ou published equa ions and he
Hanai mix u e [
25
] model wi h and wi hou adjus men o BMI [
38
]. Measu ed and p edic ed TBW
alues we e signi ican ly and simila ly co ela ed o linea i y and conco dance ( > 0.9). Compa isons
o measu ed and p edic ed TBW alues e ealed limi s o ag eemen wi h sligh ly g ea e a iabili y
(0.9 L o 2.7 L o 4% o 7%) o he 50 kHz p edic ions compa ed o a iabili y es ima es (0.6 L o 0.9 L
o 1% o 2%) om he BIS p edic ions. Thus, bo h he single equency p edic ion equa ions and BIS
models pe o med equally well in p edic ing he TBW a he popula ion le el.
Follow up analyses de e mined he compa abili y o hese di e en app oaches o es ima e he
TBW a he le el o an indi idual [
38
]. The mean absolu e e o o BIS was sligh ly imp o ed compa ed
o mul iple eg ession equa ions using 50 kHz alues (5% s. 7% o 2 L s. 3 L). These indings indica e
ha he p ecision o hese BIA app oaches may be app op ia e o g oup compa isons bu is oo la ge
o moni o ing changes in he TBW in clinical condi ions.
Any compa ison o BIA-p edic ed luid olumes wi h ace dilu ion e e ence measu emen s
su e s om echnical and biological limi a ions. Sou ces o e o include he alidi y (accu acy and
p ecision) o he impedance measu emen , e o o p edic ion om he eg ession equa ion, in insic
e o o he e e ence me hod ( echnical accu acy and ep oducibili y, biological e o s in assump ions
o he dilu ion me hod), elec ical- olume e o s (aniso opy and body geome y), and biological
a iabili y (in e -indi idual di e ences in he diame e o body segmen s, limb leng hs, and a ia ion
in body a ness) [
37
]. Use o he Cole model depends on he eliabili y o he i ing o he impedance
da a. Re iew o ecen BIS publica ions e eals ha indi idual subjec s a e emo ed om analyses
because o ailu e o achie e accep able ole ances (>99% ep oducibili y) o he epea ed ime delay
es ima es de i ed om cu e i ing [
35
,
36
]. This obse a ion indica es signi ican imp ecision in he
calcula ion o he cu en delay a memb anes. Finally, use o di e en single- equency ma hema ical
p edic ion equa ions complica es compa isons. As sugges ed by Seone e al. [
38
,
39
], olume p edic ion
models ha include mo e independen a iables han H
2
/R and weigh end o p oduce g ea e
e o s in es ima ed TBW ha indica es a speci ici y o ha eg ession equa ion and, hus, excludes i s
applica ion in any sample o indi idual di e ing in cha ac e is ics om he o iginal sample in which
he eg ession model was de eloped. These ac o s as well as he imp ecision o he p edic ion models,
single- equency and BIS, impedes hei use in poin -o -ca e indi idual assessmen s o hyd a ion.
4. Classi ica ion o Hyd a ion wi h Bioelec ical Impedance Vec o Analysis (BIVA)
In con as o he use o ma hema ical modeling o limi ed BI measu emen s, heo e ical and
eg ession p edic ion models, classi ica ion o hyd a ion only equi es p ecise BI measu emen s and
e e ence alues [
40
]. A 50 kHz phase-sensi i e BI ins umen di ec ly measu es R and Xc, which
a e no malized o s anding heigh o achie e a s anda d esis i i y and plo ed on he RXc g aph
(Figu e 5). The ec o (Z) has a leng h ha is in e sely ela ed o TBW and i s posi ion, desc ibed as
he PhA, and p o ides in o ma ion on issue hyd a ion.
Nu ien s 2019,11, 809 8 o 22
Nu ien s 2019, 11, x FOR PEER REVIEW 8 o 21
Figu e 5. Resis ance- eac ance (RXc) plo wi h ole ance ellipses om heal hy Caucasian males.
Bioelec ical impedance ec o analysis (BIVA) enables classi ica ion (unde -, no mal, and o e -
hyd a ion) and anking o change in hyd a ion (mo e o less han be o e ea men ), as well as
classi ica ion (mo e o less) o so issue mass by compa ing ec o posi ion o an indi idual o a
g oup o a heal hy e hnici y-, age-, and gende -ma ched popula ion [40]. Implemen a ion o BIVA
ollows he wo k o Moo e and Boyden [3] in which hyd a ion is a unc ion o luid dis ibu ion in
he ex acellula and in acellula spaces, no ably BCM. Clinically, o e -hyd a ion is exp essed as an
expansion o he ex acellula luid (in e s i ial luid plus plasma), also e med luid o e load, ela i e
o he BCM (in acellula luid); pa ien s wi h enal disease and hea ailu e a e cha ac e ized as
o e -hyd a ed. Malnu i ion, including cachexia, also is associa ed wi h an inc ease in ECW and a
dec ease in BCM. Hypo-hyd a ion is seen wi h a dec ease in ECW and li le o no change in BCM. I
can be acu e as wi h excess luid emo al he apy o sho - e m wi h diseases causing excessi e luid
loss (dia hea o emesis).
Compa ed o single- equency BIA and BIS p edic ions o luid olumes, BIVA only has minimal
e o associa ed wi h BI measu emen and ep oducibili y (1–2%) whe eas luid olume p edic ions
include addi ional sou ces o e o including eg ession e o o he p edic ion equa ion (~10%),
echnical e o in he e e ence me hod (~4%), limi a ions o he bioelec ical olume model (i.e.,
aniso opy o issues and geome y), and biological a iabili y (i.e., in e -indi idual body
composi ion di e ences) ha p opaga e. The e o e, o an indi idual, classi ica ion and anking o
hyd a ion is mo e p ecise and accu a e han is he quan i ica ion o luid olume because BIVA is
independen o eg ession equa ions and heo e ical models ha a e acqui ed wi h limi ed and
speci ic samples and, hus, a e no obus in he assessmen o hyd a ion ou side o he g oup in
which hey we e de eloped, and a e ad e sely a ec ed by illness.
The RXc g aph (Figu e 5) is a p obabili y dis ibu ion ha classi ies a ec o acco ding o i s
dis ance om he mean heal hy ec o . The a iabili y o he impedance ec o is ep esen ed in he
bi a ia e no mal dis ibu ion wi h ellip ical p obabili y a eas (50%, 75%, and 95%) in he ole ance o
e e ence ellipses. Vec o s displacemen s pa allel o he majo axis indica e changes in hyd a ion
(mo e o less luids). Vec o s wi hin he 50% ole ance ellipse a e conside ed o be a no mal
hyd a ion whe eas leng hening o ec o s om he 51% o 75% and >76% pe cen ile o he uppe
ange o pe cen iles indica e mode a e and se e e dehyd a ion, espec i ely. Con e sely, sho ening
o ec o s om he 51% o 75% and >76% pe cen ile e e ence ellipses in he lowe ange indica e
inc easing luid o e load. Vec o s posi ioned o he le o he majo axis e lec inc easing cell mass
and ec o s o he igh indica e dec easing cell mass, espec i ely. Thus, BIVA uses pa e ns o
Figu e 5. Resis ance- eac ance (RXc) plo wi h ole ance ellipses om heal hy Caucasian males.
Bioelec ical impedance ec o analysis (BIVA) enables classi ica ion (unde -, no mal, and
o e -hyd a ion) and anking o change in hyd a ion (mo e o less han be o e ea men ), as well as
classi ica ion (mo e o less) o so issue mass by compa ing ec o posi ion o an indi idual o a
g oup o a heal hy e hnici y-, age-, and gende -ma ched popula ion [
40
]. Implemen a ion o BIVA
ollows he wo k o Moo e and Boyden [
3
] in which hyd a ion is a unc ion o luid dis ibu ion in
he ex acellula and in acellula spaces, no ably BCM. Clinically, o e -hyd a ion is exp essed as an
expansion o he ex acellula luid (in e s i ial luid plus plasma), also e med luid o e load, ela i e
o he BCM (in acellula luid); pa ien s wi h enal disease and hea ailu e a e cha ac e ized as
o e -hyd a ed. Malnu i ion, including cachexia, also is associa ed wi h an inc ease in ECW and a
dec ease in BCM. Hypo-hyd a ion is seen wi h a dec ease in ECW and li le o no change in BCM.
I can be acu e as wi h excess luid emo al he apy o sho - e m wi h diseases causing excessi e luid
loss (dia hea o emesis).
Compa ed o single- equency BIA and BIS p edic ions o luid olumes, BIVA only has minimal
e o associa ed wi h BI measu emen and ep oducibili y (1–2%) whe eas luid olume p edic ions
include addi ional sou ces o e o including eg ession e o o he p edic ion equa ion (~10%),
echnical e o in he e e ence me hod (~4%), limi a ions o he bioelec ical olume model (i.e.,
aniso opy o issues and geome y), and biological a iabili y (i.e., in e -indi idual body composi ion
di e ences) ha p opaga e. The e o e, o an indi idual, classi ica ion and anking o hyd a ion is
mo e p ecise and accu a e han is he quan i ica ion o luid olume because BIVA is independen o
eg ession equa ions and heo e ical models ha a e acqui ed wi h limi ed and speci ic samples and,
hus, a e no obus in he assessmen o hyd a ion ou side o he g oup in which hey we e de eloped,
and a e ad e sely a ec ed by illness.
The RXc g aph (Figu e 5) is a p obabili y dis ibu ion ha classi ies a ec o acco ding o i s
dis ance om he mean heal hy ec o . The a iabili y o he impedance ec o is ep esen ed in he
bi a ia e no mal dis ibu ion wi h ellip ical p obabili y a eas (50%, 75%, and 95%) in he ole ance
o e e ence ellipses. Vec o s displacemen s pa allel o he majo axis indica e changes in hyd a ion
(mo e o less luids). Vec o s wi hin he 50% ole ance ellipse a e conside ed o be a no mal hyd a ion
whe eas leng hening o ec o s om he 51% o 75% and >76% pe cen ile o he uppe ange o
pe cen iles indica e mode a e and se e e dehyd a ion, espec i ely. Con e sely, sho ening o ec o s
om he 51% o 75% and >76% pe cen ile e e ence ellipses in he lowe ange indica e inc easing luid
o e load. Vec o s posi ioned o he le o he majo axis e lec inc easing cell mass and ec o s o
he igh indica e dec easing cell mass, espec i ely. Thus, BIVA uses pa e ns o impedance ec o
dis ibu ion wi hou he need o p edic ion equa ions, body weigh o eliance on he assump ion
Nu ien s 2019,11, 809 9 o 22
o s able composi ion (wa e con en ) o he FFM. BIVA enables de ec ion and anking o changes in
issue hyd a ion s a us o <500 mL in eal- ime [40].
App op ia e de i a ion and implemen a ion o he RXc g aph o classi y hyd a ion equi es some
impo an echnical conside a ions. All measu emen s should be acqui ed wi h a phase-sensi i e
(e.g., phase angle di ec ly measu ed) BI ins umen . Failu e o use a phase-sensi i e de ice esul s in
signi ican e o s in R (10
Ω
) and Xc (10–12
Ω
) measu emen s [
41
] con ibu ing o 8–10% eposi ioning
o e e ence and pa ien ec o s. Addi ionally, use o high-impedance elec odes can lead o
misclassi ica ion o hyd a ion [42].
4.1. Clinical Applica ions o BIVA
A ocal applica ion o BIVA is he assessmen o he hyd a ion s a us in pa ien s wi h luid
o e load and e alua ion o e ec s o he apeu ic p ocedu es designed o achie e homeos asis wi hou
undesi able side e ec s. Kidney disease and hemodynamic impai men s a e he p incipal condi ions in
which BIVA is used o classi y hyd a ion and o moni o changes in esponse o ea men s. T adi ional
app oaches o assess hyd a ion in hese pa ien s a e un eliable because hey ely on body weigh o
designa e excess luid and a e insensi i e o luid o e load in he p esence o edema [43].
4.1.1. Hyd a ion Assessmen in Hemodialysis (HD)
Dialysis, pa icula ly hemodialysis (HD), is amenable o BIVA because o he need o luid
emo al and pos -dialy ic luid e en ion. Rou ine e alua ion o hyd a ion includes moni o ing o
body weigh and blood p essu e changes ha a e no eliably de e mined by luid olume. Edema
is no usually de ec able un il in e s i ial luid olume inc eases 30% o e no mal le els (4–5 kg
gain in body weigh ) and se e e dehyd a ion can occu be o e appea ance o clinical signs. Thus,
adi ional indica o s o o e - and unde -hyd a ion in pa ien s wi h enal disease a e insensi i e and
inadequa e [40].
4.1.2. Hyd a ion Changes and Vec o Pa e ns Du ing he HD Cycle
The objec i e o HD is o emo e excess luid wi hou complica ions o he pa ien . Es ima ion
o he olume o luid o emo al is subjec i e and gene ally ela ed o p e ious pos -dialy ic body
weigh , which may be un eliable [
43
]. Fluid emo al occu s du ing HD. I can be symp oma ic o
uncompensa ed wi h episodes o hypo ension, malaise, and c amps o asymp oma ic (compensa ed).
Fluid o e load can occu du ing he in e -dialy ic pe iod and is symp oma ic wi h edema, exace ba ed
hype ension, and pulmona y conges ion. D y weigh is a gene al e m associa ed wi h he pos -dialy ic
body weigh a which mos o he excess luid has been emo ed. The op imal “d y weigh ” is
de e mined clinically and ope a ionally as he weigh a which a pa ien can ole a e HD wi hou
ad e se in a-dialy ic symp oms, no ably hypo ension. Body luid olumes, howe e , change
con inuously du ing he in e -dialy ic pe iod so ha euhyd a ion occu s only o a b ie pe iod.
Se ial BI measu emen s and examina ion o ec o s be o e and a e HD e eal a classic pa e n o
luid emo al and eple ion du ing a s anda d 3-d HD cycle. Fluid emo al ia ul a il a ion (UF)
accompanies ec o leng hening pa allel o he mean ec o o heal hy adul s [
44
]. Wi hin he i s 2
pos HD, ec o s a e ela i ely s able nea he pos HD ec o . Du ing he nex 24, 48, and 72 h howe e ,
ec o s p og essi ely sho en wi hin he 50% o 75% ole ance ellipse and e lec luid eple ion o 1.4,
2.6, and 3.4 L. Thus, BIVA is speci ic and sensi i e o moni o luid du ing he we and d y cycle o HD.
4.1.3. Vec o T ajec o ies: Adequacy o Ul a il a ion (UF)
Measu emen o he elec ical p ope ies o issues enables he classi ica ion o ec o s as abno mal
and iden i ica ion o HD pa ien s a isk o in a-dialy ic p oblems. Obse a ional indings indica e
ha BIVA can iden i y po en ial symp oma ic HD pa ien s. Be o e dialysis, HD pa ien s, compa ed o
heal hy con ols, exhibi e idence o luid o e load wi h sho e ec o s and lesse phase angles [
44
,
45
].
P e-HD ec o posi ion disc imina ed uns able (symp oma ic), compa ed o s able (asymp oma ic),
Nu ien s 2019,11, 809 16 o 22
signi ican ly a ec ed ec o posi ion on he RXc g aph esul ing in signi ican di e ences in hyd a ion
classi ica ion o heal hy adul s.
S anda diza ion o BI measu emen s emains a challenge o clinical use s o BI o hyd a ion
classi ica ion. The lack o in e na ional manu ac u ing s anda ds, coo dina ion o echnology, and
c oss-calib a ion o elec ical accu acy hinde s g ow h in clinical applica ions o BI. Concu ence on
app op ia e measu emen s condi ions in hyd a ion assessmen is imminen bu clinical in es iga o s
need o be awa e o ele a ed co e and pe iphe al empe a u es as well as high impac BI measu emen s.
Use o a i ac - ee con ac elec odes is ecommended o eliable classi ica ion o hyd a ion. Small
di e ences in BI measu emen s incu ed by measu emen in luences can al e hyd a ion classi ica ion
in clinical condi ions.
8. Summa y and Conclusions
Bioimpedance measu emen s a e gaining in e es o assess and moni o hyd a ion s a us in dialy ic
and c i ically ill pa ien s because hey o e come many o he limi a ions o adi ional me hods o
hyd a ion assessmen . Whe eas wo app oaches use o BI measu emen s each has speci ic ad an ages
and ce ain limi a ions.
Use o di ec 50 kHz, phase-sensi i e BI measu emen s in he BIVA model o hyd a ion
assessmen is p og essing in a a ie y o clinical condi ions. BIVA illus a es a ec o whose posi ion
is e alua ed ela i e o heal hy e e ence anges and is in e p e ed as no mal o abno mal hyd a ion
based on dis ance om he mean heal hy ec o . Mig a ion o he ec o (sho ening o leng hening)
in esponse o p og ession o a physiological p ocess, pa hology, o an in e en ion indica es changes
in hyd a ion (gain o loss o luids). Classi ica ion o hyd a ion s a us (no mal o abno mal) a oids
insensi i i ies (>10% a iabili y and imp ecision o indi idual es ima ions) associa ed wi h eg ession
equa ions and unp o en biophysical models, and does no ely on body weigh o assess hyd a ion.
Use o BIVA imp o es he p esc ip ion o UF in dialysis by moni o ing he backwa d- o wa d
displacemen o ec o s in ela ion o he we -d y cycle o HD. I u he enhances decision-making
in dialysis by acili a ing he in e p e a ion o al e a ions in blood p essu e ela i e o he hyd a ion
s a us and, hus, adjus ing UF. Among c i ically ill pa ien s, BIVA is signi ican ly and in e sely
co ela ed wi h CVP. On he RXc g aph, impedance ec o s o pa ien s wi h low, no mal, and high
CVP mo e downwa d and ou side o he 75% ole ance le el wi h inc easing CVP, which indica es
excess luid accumula ion. Iden i ica ion o hypohyd a ion in s able CAD pa ien s be o e angiog aphic
p ocedu es enabled ehyd a ion wi h app op ia e olumes o saline o a enua e isk o CI-AKI. Use o
BIVA in assessmen o malnu i ion, no ably in illnesses wi h muscle loss when hyd a ion is al e ed,
is inc easing.
Phase-sensi i e mul i- equency BIS measu emen s coupled wi h biophysical models indi ec ly
es ima e ECW ha is compa ed wi h p ojec ions o no mal ECW o calcula e excess ECW desc ibed
as FO. The es ima es o ECW, adjus ed o BMI, a e de i ed om eg ession equa ions using
gende , heigh , and body weigh as independen a iables wi h wide limi s o ag eemen and,
hence imp ecision o an indi idual ECW es ima e. Es ima es o FO >2.5 L o >15% p edic ed ECW
a e associa ed wi h inc eased mo bidi y in dialysis pa ien s bu he p ognos ic alue o mo ali y
is lacking.
9. Fu u e Di ec ions
Ad ancemen o he use o BI measu emen s o aid in luid managemen o dialysis pa ien s
equi es p ospec i e con olled s udies wi h mo e ocus on ou comes [
79
,
90
–
92
]. Ea ly e o s we e
desc ip i e emphasizing p oo -o -p inciple esea ch. Adequa e con i ma ion o posi i e indings
ela ed o alidi y BIVA and BIS is a ailable. New esea ch should c i ically e alua e he bene i o
hese BI app oaches in pa ien ca e. Impo an ly, he e is a need o mo e consis ency and sc u iny in
esea ch design wi h app op ia e sample sizes o es hypo heses and compa abili y in epo ing o
esul s. Some opics include he abili y o p edic in a- and in e -dialy ic complica ions and mo ali y

Nu ien s 2019,11, 809 17 o 22
o HD pa ien s, measu e he e ec s o he apeu ic in e en ions on changes in hyd a ion and luid
s a us on p ese a ion o enal unc ion, e alua e he speci ici y and sensi i i y o a ailable and
pu a i e clinical ma ke s o hyd a ion, de e mina ion o changes in body composi ion as well as
nu i ional s a us/in ake ela i e o longe i y o dialysis, and he e ec o in e en ions o p ese e
muscle mass on su i al. The goal is o p o ide indings ha can be applied p ac ically o imp o e
pa ien ou comes.
Au ho Con ibu ions: Each au ho con ibu ed equally in he p epa a ion o his manusc ip .
Funding: This esea ch ecei ed no ex e nal unding.
Con lic s o In e es :
The au ho s epo no con lic s o in e es . A. Tallu i was ounde and p esiden o Ake n S l,
and is no longe in ol ed in his company.
Re e ences
1.
A ms ong, L.E. Assessing hyd a ion s a us: The elusi e gold s anda d. J. Am. Coll. Nu .
2007
,26 (Suppl. 5),
575S–584S. [C ossRe ] [PubMed]
2.
Di Somma, S.; Na a in, S.; Gio dano, S.; Spadini, F.; Lippi, G.; Ce ellin, G.; Die enbach, B.V.; Maisel, A.S.
The eme ging ole o bioma ke s and bio-impedance in e alua ing hyd a ion s a us in pa ien s wi h acu e
hea ailu e. Clin. Chem. Lab. Med. 2012,50, 2093–2105. [C ossRe ]
3.
Moo e, F.D.; Boyden, C.M. Body cell mass and limi s o hyd a ion o he a - ee body: Thei ela ion o
es ima ed skele al weigh . Ann. N. Y. Acad. Sci. 1963,110, 62–71. [C ossRe ] [PubMed]
4.
Snyde , W.S.; Cook, M.J.; Nasse , E.S.; Ka hausen, L.R.; Howells, G.P.; Tip on, I.H. Repo o he Task G oup on
Re e ence Man; Pe gamon P ess: Ox o d, UK, 1975; pp. 27–32.
5.
Wang, Z.; Deu enbe g, P.; Wang, W.; Pie obelli, A.; Baumga ne , R.N.; Heyms ield, S.B. Hyd a ion o a - ee
body mass: Re iew and c i ique o a classic body-composi ion cons an . Am. J. Clin. Nu .
1999
,69, 833–841.
[C ossRe ] [PubMed]
6.
Chumlea, W.C.; Schube , C.M.; Sun, S.S.; Deme a h, E.; Towne, B.; Sie ogel, R.M. A e iew o body wa e
s a us and he e ec s o age and body a ness in child en and adul s. J. Nu . Heal h Aging
2007
,11, 111–118.
7.
A ms ong, L.E.; Ka ou as, S.A.; Walsh, N.P.; Robe s, W.O. Diagnosing dehyd a ion? Blend e idence wi h
clinical obse a ions. Cu . Opin. Clin. Nu . Me ab. Ca e 2016,19, 434–438. [C ossRe ]
8.
Lukaski, H.C. Biological indexes conside ed in he de i a ion o he bioelec ical impedance analysis. Am. J.
Clin. Nu . 1996,64 (Suppl. 3), 397S–404S. [C ossRe ]
9.
Kyle, U.G.; Bosaeus, I.; De Lo enzo, A.D.; Deu enbe g, P.; Elia, M.; Gómez, J.M.; Hei mann, B.L.; Ken -Smi h, L.;
Melchio , J.C.; Pi lich, M.; e al. Composi ion o he ESPEN Wo king G oup. Bioelec ical impedance
analysis-pa I: Re iew o p inciples and me hods. Eu . J. Clin. Nu . 2004,23, 1226–1243. [C ossRe ]
10.
Lukaski, H.C. E olu ion o bioimpedance: A ci cui ous jou ney om es ima ion o physiological unc ion
o assessmen o body composi ion and a e u n o clinical esea ch. Eu . J. Clin. Nu .
2013
,67 (Suppl. 1),
S2–S9. [C ossRe ] [PubMed]
11.
Faes, T.J.; an de Meij, H.A.; de Munck, J.C.; Hee haa , R.M. The elec ic esis i i y o human issues
(100 Hz–10 MHz): A me a-analysis o e iew s udies. Physiol. Meas. 1999,20, R1–R10. [C ossRe ]
12.
Fos e , K.R.; Lukaski, H.C. Whole-body impedance—Wha does i measu e? Am. J. Clin. Nu .
1996
,
64 (Suppl. 3), 388S–396S. [C ossRe ] [PubMed]
13.
G imnes, S.; Ma insen, O. Bioimpedance and Bioelec ici y Basics, 3 d ed.; Else ie L d.: London, UK, 2015;
pp. 1–8.
14.
Race e, S.B.; Schoelle , D.A.; Luke, A.H.; Shay, K.; Hnilicka, J.; Kushne , R.F. Rela i e dilu ion spaces o
2H-and 18O-labeled wa e in humans. Am. J. Physiol. 1994,267, E585–E590. [C ossRe ]
15.
Schoelle , D.A. Hyd ome y. In Human Body Composi ion, 2nd ed.; Heyms ield, S.B., Lohman, T.G.,
Wang, Z.M., Going, S.B., Eds.; Human Kine ics: Champaign, IL, USA, 2005; pp. 35–50.
Nu ien s 2019,11, 809 18 o 22
16.
Piccoli, A.; Pas o i, G.; Guizzo, M.; Rebeschini, M.; Naso, A.; Cascone, C. Equi alence o in o ma ion om
single e sus mul iple equency bioimpedance ec o analysis in hemodialysis. Kidney In .
2005
,67, 301–313.
[C ossRe ]
17.
Ma hie, J.R. Bioimpedance measu emen s o human body composi ion: C i ical analysis and ou look.
Expe Re . Med. De ices 2008,5, 239–261. [C ossRe ] [PubMed]
18.
Kyle, U.G.; Bosaeus, I.; De Lo enzo, A.D.; Deu enbe g, P.; Elia, M.; Gómez, J.M.; Lilien hal Hei mann, B.;
Ken -Smi h, L.; Melchio , J.C.; Pi lich, M.; e al. Bioelec ical impedance analysis-pa II: U iliza ion in clinical
p ac ice. Eu . J. Clin. Nu . 2004,23, 1430–1453. [C ossRe ] [PubMed]
19.
Mulasi, U.; Kuchnia, A.J.; Cole, A.J.; Ea hman, C.P. Bioimpedance a he bedside: Cu en applica ions,
limi a ions, and oppo uni ies. Nu . Clin. P ac . 2015,30, 180–193. [C ossRe ] [PubMed]
20.
Sun, S.S.; Chumlea, W.C.; Heyms ield, S.B.; Lukaski, H.C.; Schoelle , D.; F iedl, K.; Kuczma ski, R.J.;
Flegal, K.M.; Johnson, C.L.; Hubba d, V.S. De elopmen o bioelec ical impedance analysis p edic ion
equa ions o body composi ion wi h he use o a mul icomponen model o use in epidemiological su eys.
Am. J. Clin. Nu . 2003,77, 331–340. [C ossRe ]
21.
Ellis, K.J.; Wong, W.W. Human hyd ome y: Compa ison o mul i equency bioelec ical impedance wi h
2H2O and b omine dilu ion. J. Appl. Physiol. 1998,85, 1056–1062. [C ossRe ]
22.
Ellis, K.J.; Bell, S.J.; Che ow, G.M.; Chumlea, W.C.; Knox, T.A.; Ko le , D.P.; Lukaski, H.C.; Schoelle , D.A.
Bioelec ical impedance me hods in clinical esea ch: A ollow-up o he NIH Technology Assessmen
Con e ence. Nu i ion 1999,15, 874–880. [C ossRe ]
23.
Ellis, K.J. Human body composi ion:
In i o
me hods. Physiol. Re .
2000
,80, 649–680. [C ossRe ] [PubMed]
24.
Cole, K.S. Pe meabili y and impe meabili y o cell memb anes o ions. In Cold Sp ing Ha bo Symposia on
Quan i a i e Biology; Cold Sp ing Ha bo Labo a o y P ess: Cold Sp ing Ha bo , NY, USA, 1940; Volume 8,
pp. 110–122.
25.
Hanai, T. Elec ical p ope ies o emulsions. In Emulsion Science; She man, P.H., Ed.; Academic P ess: London,
UK, 1968; pp. 354–477.
26.
De Lo enzo, A.; And eoli, A.; Ma hie, J.; Wi he s, P. P edic ing body cell mass wi h bioimpedance by using
heo e ical me hods: A echnological e iew. J. Appl. Physiol. 1997,82, 1542–1558. [C ossRe ]
27.
Bol on, M.P.; Wa d, L.C.; Khan, A.; Campbell, I.; Nigh ingale, P.; Dewi , O.; Elia, M. Sou ces o e o in
bioimpedance spec oscopy. Physiol. Meas. 1998,19, 235–246. [C ossRe ]
28.
Buendia, R.; Seoane, F.; Lindec an z, K.; Bosaeus, I.; Gil-Pi a, R.; Johannsson, G.; Ellegá d, L.; Wa d, L.C.
Es ima ion o body luids wi h bioimpedance spec oscopy: S a e o he a me hods and p oposal o no el
me hods. Physiol. Meas. 2015,36, 2171–2187. [C ossRe ]
29.
Wa d, L.C.; Isen ing, E.; Dye , J.M.; Kagawa, M.; Essex, T. Resis i i y coe icien s o body composi ion
analysis using bioimpedance spec oscopy: E ec s o body dominance and mix u e heo y algo i hm.
Physiol. Meas. 2015,36, 1529–1549. [C ossRe ]
30.
Cox-Reij en, P.L.; Soe e s, P.B. Valida ion o bio-impedance spec oscopy: E ec s o deg ee o obesi y and
ways o calcula ing olumes om measu ed esis ance alues. In . J. Obes. Rela . Me ab. Diso d.
2000
,
24, 271–280. [C ossRe ]
31.
Cox-Reij en, P.L.; an K eel, B.; Soe e s, P.B. Accu acy o bioelec ical impedance spec oscopy in measu ing
changes in body composi ion du ing se e e weigh loss. J. Pa en e . En e . Nu .
2002
,26, 120–127. [C ossRe ]
32.
Mage , J.R.; Sibley, S.D.; Beckman, T.R.; Kellogg, T.A.; Ea hman, C.P. Mul i equency bioelec ical impedance
analysis and bioimpedance spec oscopy o moni o ing luid and body cell mass changes a e gas ic
bypass su ge y. Clin. Nu . 2008,27, 832–841. [C ossRe ]
33.
Moissl, U.M.; Wabel, P.; Chamney, P.W.; Bosaeus, I.; Le in, N.W.; Bosy-Wes phal, A.; Ko h, O.; Mülle , M.J.;
Ellegå d, L.; Malm os, V.; e al. Body luid olume de e mina ion ia body composi ion spec oscopy in
heal h and disease. Physiol. Meas. 2006,27, 921–933. [C ossRe ]
34.
Gonzalez, M.C.; Co eia, M.I.T.D.; Heyms ield, S.B. A equiem o BMI in he clinical se ing. Cu . Opin.
Clin. Nu . Me ab. Ca e 2017,20, 314–321. [C ossRe ]
35.
Raimann, J.G.; Zhu, F.; Wang, J.; Thijssen, S.; Kuhlmann, M.K.; Ko anko, P.; Le in, N.W.; Kaysen, G.A.
Compa ison o luid olume es ima es in ch onic hemodialysis pa ien s by bioimpedance, di ec iso opic,
and dilu ion me hods. Kidney In . 2014,85, 898–908. [C ossRe ]
Nu ien s 2019,11, 809 19 o 22
36.
Raimann, J.G.; Abbas, S.R.; Liu, L.; Zhu, F.; La i e, B.; Ko anko, P.; Le in, N.W.; Kaysen, G.A.; FHN T ial.
Ag eemen o single-and mul i- equency bioimpedance measu emen s in hemodialysis pa ien s: An ancilla y
s udy o he F equen Hemodialysis Ne wo k Daily T ial. Neph on Clin. P ac . 2014,128, 115–126. [C ossRe ]
37.
Piccoli, A. Es ima ion o luid olumes in hemodialysis pa ien s: Compa ing bioimpedance wi h iso opic
dilu ion me hods. Kidney In . 2014,85, 738–741. [C ossRe ]
38.
Seoane, F.; Ab ahi, S.; Ab ahi, F.; Ellegá d, L.; Johannsson, G.; Bosaeus, I.; Wa d, L.C. Sligh ly supe io
pe o mance o bioimpedance spec oscopy o e single equency eg ession equa ions o assessmen o
o al body wa e . In P oceedings o he IEEE Enginee ing in Medicine and Biology Socie y (EMBC), Milan,
I aly, 25–29 Augus 2015; pp. 3707–3710.
39.
Seoane, F.; Ab ahi, S.; Ab ahi, F.; Ellegá d, L.; Johannsson, G.; Bosaeus, I.; Wa d, L.C. Mean expec ed e o in
p edic ion o o al body wa e : A ue accu acy compa ison be ween bioimpedance spec oscopy and single
equency eg ession equa ions. Biomed. Res. In . 2015,11, 656323. [C ossRe ]
40.
Piccoli, A.; Rossi, B.; Pillon, L.; Buccian e, G. A new me hod o moni o ing body luid a ia ion by
bioimpedance analysis: The RXc g aph. Kidney In . 1994,46, 534–539. [C ossRe ]
41.
Chumlea, W.C.; Guo, S.S.; Kuczma ski, R.J.; Johnson, C.L.; Flegel, K.; Heyms ield, S.; Lukaski, H.;
Schoelle , D.; F iedl, K.; Hubba d, V.S. Body composi ion es ima es om NHANES III bioelec ical impedance
da a. In . J. Obes. 2002,26, 1596–1609. [C ossRe ]
42.
Nescola de, L.; Lukaski, H.; De Lo enzo, A.; de-Ma eo-Sille as, B.; Redondo-Del-Río, M.P.; Camina-Ma ín, M.A.
Di e en displacemen o bioimpedance ec o due o Ag/AgCl elec ode e ec . Eu . J. Clin. Nu .
2016
,
70, 1401–1407. [C ossRe ]
43.
Pillon, L.; Piccoli, A.; Low ie, E.G.; Laza us, J.M.; Che ow, G.M. Vec o leng h as a p oxy o he adequacy o
ul a il a ion in hemodialysis. Kidney In . 2004,66, 1266–1271. [C ossRe ]
44.
Piccoli, A.; Codogno o, M. Bioimpedance ec o mig a ion up o h ee days a e he hemodialysis session.
Kidney In . 2004,66, 2091–2092. [C ossRe ]
45. Piccoli, A. Whole body–single equency bioimpedance. Con ib. Neph ol. 2005,149, 150–161.
46.
Piccoli, A. Iden i ica ion o ope a ional clues o d y weigh p esc ip ion in hemodialysis using bioimpedance
ec o analysis. The I alian Hemodialysis-Bioelec ical Impedance Analysis (HD-BIA) S udy G oup.
Kidney In . 1998,53, 1036–1043. [C ossRe ]
47.
Piccoli, A.; I alian CAPD-BIA S udy G oup. Bioelec ic impedance ec o dis ibu ion in pe i oneal dialysis
pa ien s wi h di e en hyd a ion s a us. Kidney In . 2004,65, 1050–1063. [C ossRe ]
48. Guy on, A.C. Tex book o Medical Physiology; Saunde s: Philadelphia, PA, USA, 1991; p. 282.
49.
Piccoli, A.; Pi oni, G.; Facco, E.; Fa a o, E.; Pillon, L. Rela ionship be ween cen al enous p essu e and
bioimpedance ec o analysis in c i ically ill pa ien s. C i . Ca e Med. 2000,28, 132–137. [C ossRe ]
50.
Basso, F.; Be din, G.; Vi zì, G.M.; Mason, G.; Piccinni, P.; Day, S.; C uz, D.N.; Wjewodzka, M.; Giuliani, A.;
B endolan, A.; e al. Fluid managemen in he in ensi e ca e uni : Bioelec ical impedance ec o analysis
as a ool o assess hyd a ion s a us and op imal luid balance in c i ically ill pa ien s. Blood Pu i .
2013
,36,
192–199. [C ossRe ]
51.
Samoni, S.; Vigo, V.; Reséndiz, L.I.; Villa, G.; De Rosa, S.; Nalesso, F.; Fe a i, F.; Meola, M.; B endolan, A.;
Malaca ne, P.; e al. Impac o hype hyd a ion on he mo ali y isk in c i ically ill pa ien s admi ed in
in ensi e ca e uni s: Compa ison be ween bioelec ical impedance ec o analysis and cumula i e luid
balance eco ding. C i . Ca e 2016,20, 95. [C ossRe ]
52.
Valle, R.; Asp omon e, N.; Milani, L.; Peacock, F.W.; Maisel, A.S.; San ini, M.; Ronco, C. Op imizing luid
managemen in pa ien s wi h acu e decompensa ed hea ailu e (ADHF): The eme ging ole o combined
measu emen o body hyd a ion s a us and b ain na iu e ic pep ide (BNP) le els. Hea Fail. Re .
2011
,
16, 519–559. [C ossRe ]
53.
Boucha d, J.; So oko, S.B.; Che ow, G.M.; Himmel a b, J.; Ikizle , T.A.; Paganini, E.P.; Meh a, R.L.; P og am
o Imp o e Ca e in Acu e Renal Disease (PICARD) S udy G oup. Fluid accumula ion, su i al and eco e y
o kidney unc ion in c i ically ill pa ien s wi h acu e kidney inju y. Kidney In .
2009
,76, 422–427. [C ossRe ]
54.
Hise, A.C.D.R.; Gonzalez, M.C. Assessmen o hyd a ion s a us using bioelec ical impedance ec o analysis
in c i ical pa ien s wi h acu e kidney inju y. Clin. Nu . 2018,37, 695–700. [C ossRe ]
Nu ien s 2019,11, 809 20 o 22
55.
Maioli, M.; Toso, A.; Leoncini, M.; Musilli, N.; Bellandi, F.; Rosne , M.H.; McCullough, P.A.; Ronco, C.
P e-p ocedu al bioimpedance ec o ial analysis o luid s a us and p edic ion o con as -induced acu e
kidney inju y. J. Am. Coll. Ca diol. 2014,63, 1387–1394. [C ossRe ]
56.
Maioli, M.; Toso, A.; Leoncini, M.; Musilli, N.; G ippo, G.; Ronco, C.; McCullough, P.A.; Bellandi, F.
Bioimpedance-Guided Hyd a ion o he P e en ion o Con as -Induced Kidney Inju y: The HYDRA
S udy. J. Am. Coll. Ca diol. 2018,71, 2880–2889. [C ossRe ]
57.
Di Somma, S.; Lalle, I.; Mag ini, L.; Russo, V.; Na a in, S.; Cas ello, L.; A anzi, G.C.; Di S asio, E.;
Maisel, A. Addi i e diagnos ic and p ognos ic alue o bioelec ical impedance ec o analysis (BIVA)
o b ain na iu e ic pep ide ‘g ey-zone’ in pa ien s wi h acu e hea ailu e in he eme gency depa men .
Eu . Hea J. Acu e Ca dio asc. Ca e 2014,3, 167–175. [C ossRe ]
58.
Massa i, F.; Scicchi ano, P.; Ciccone, M.M.; Calda ola, P.; Asp omon e, N.; Iaco iello, M.; Ba o, S.; Pan ano, I.;
Valle, R. Bioimpedance ec o analysis p edic s hospi al leng h o s ay in acu e hea ailu e. Nu i ion
2018
,
61, 56–60. [C ossRe ]
59.
Núñez, J.; Masca ell, B.; S ubbe, H.; Ven u a, S.; Bonanad, C.; Bodí, V.; Núñez, E.; Miñana, G.; Fácila, L.;
Bayés-Genis, A.; e al. Bioelec ical impedance ec o analysis and clinical ou comes in pa ien s wi h acu e
hea ailu e. J. Ca dio asc. Med. 2016,17, 283–290. [C ossRe ]
60.
San a elli, S.; Russo, V.; Lalle, I.; De Be a dinis, B.; Na a in, S.; Mag ini, L.; Piccoli, A.; Codogno o, M.;
Cas ello, L.M.; A anzi, G.C.; e al. Use ulness o combining admission b ain na iu e ic pep ide (BNP)
plus hospi al discha ge bioelec ical impedance ec o analysis (BIVA) in p edic ing 90 days ca dio ascula
mo ali y in pa ien s wi h acu e hea ailu e. In e n. Eme g. Med. 2017,12, 445–451. [C ossRe ]
61.
Piccoli, A.; Codogno o, M.; Cianci, V.; Ve o e, G.; Zanino o, M.; Plebani, M.; Maisel, A.; Peacock, W.F.
Di e en ia ion o ca diac and nonca diac dyspnea using bioelec ical impedance ec o analysis (BIVA).
J. Ca d. Fail. 2012,18, 226–232. [C ossRe ]
62.
Da ies, S.J.; Da enpo , A. The ole o bioimpedance and bioma ke s in helping o aid clinical
decision-making o olume assessmen s in dialysis pa ien s. Kidney In . 2014,86, 489–496. [C ossRe ]
63.
Chamney, P.W.; Wabel, P.; Moissl, U.M.; Mülle , M.J.; Bosy-Wes phal, A.; Ko h, O.; Fulle , N.J. A whole-body
model o dis inguish excess luid om he hyd a ion o majo body issues. Am. J. Clin. Nu .
2007
,85, 80–89.
[C ossRe ]
64.
Wabel, P.; Moissl, U.; Chamney, P.; Ji ka, T.; Machek, P.; Ponce, P.; Tabo sky, P.; Te a, C.; Velasco, N.; Vlasak, J.;
e al. Towa ds imp o ed ca dio ascula managemen : The necessi y o combining blood p essu e and luid
o e load. Neph ol. Dial. T ansplan . 2008,23, 2965–2971. [C ossRe ]
65.
Oei, E.L.; Fan, S.L. P ac ical aspec s o olume con ol in ch onic kidney disease using whole body
bioimpedance. Blood Pu i . 2015,39, 32–36. [C ossRe ]
66.
O’Lone, E.L.; Visse , A.; Finney, H.; Fan, S.L. Clinical signi icance o mul i- equency bioimpedance
spec oscopy in pe i oneal dialysis pa ien s: Independen p edic o o pa ien su i al. Neph ol. Dial.
T ansplan . 2014,29, 1430–1437. [C ossRe ]
67.
Wabel, P.; Chamney, P.; Moissl, U.; Ji ka, T. Impo ance o whole-body bioimpedance spec oscopy o he
managemen o luid balance. Blood Pu i . 2009,7, 75–80. [C ossRe ]
68.
Wizemann, V.; Wabel, P.; Chamney, P.; Zaluska, W.; Moissl, U.; Rode, C.; Malecka-Masalska, T.; Ma celli, D.
The mo ali y isk o o e hyd a ion in haemodialysis pa ien s. Neph ol. Dial. T ansplan .
2009
,24, 1574–1579.
[C ossRe ]
69.
Tabino , M.; Elphick, E.; Dudson, M.; Kwok, C.S.; Lambie, M.; Da ies, S.J. Bioimpedance-de ined
o e hyd a ion p edic s su i al in end s age kidney ailu e (ESKF): Sys ema ic e iew and subg oup
me a-analysis. Sci. Rep. 2018,13, 4441. [C ossRe ]
70.
Hu , E.; Us a, M.; Toz, H.; Asci, G.; Wabel, P.; Kah ecioglu, S.; Kayikcioglu, M.; Demi ci, M.S.; Ozkahya, M.;
Duman, S.; e al. E ec o luid managemen guided by bioimpedance spec oscopy on ca dio ascula pa ame e s
in hemodialysis pa ien s: A andomized con olled ial. Am. J. Kidney Dis. 2013,61, 957–965. [C ossRe ]
71.
Ono iescu, M.; Hogas, S.; Vo oneanu, L.; Ape ii, M.; Nis o , I.; Kanbay, M.; Co ic, A.C. Bioimpedance-
guided luid managemen in main enance hemodialysis: A pilo andomized con olled ial. Am. J. Kidney
Dis. 2014,64, 111–118. [C ossRe ]
Nu ien s 2019,11, 809 21 o 22
72.
Van Biesen, W.; Williams, J.D.; Co ic, A.C.; Fan, S.; Claes, K.; Lichodziejewska-Niemie ko, M.; Ve ge , C.;
S eige , J.; Schode , V.; Wabel, P.; e al. Fluid s a us in pe i oneal dialysis pa ien s: The Eu opean Body
Composi ion Moni o ing (Eu oBCM) s udy coho . PLoS ONE 2011,6, e17148. [C ossRe ]
73.
Ronco, C.; Ve ge , C.; C epaldi, C.; Pham, J.; De Los Ríos, T.; Gauly, A.; Wabel, P.; Van Biesen, W.; IPOD-PD
S udy G oup. Baseline hyd a ion s a us in inciden pe i oneal dialysis pa ien s: The ini ia i e o pa ien
ou comes in dialysis (IPOD-PD s udy). Neph ol. Dial. T ansplan . 2015,30, 849–858. [C ossRe ]
74.
O gan, L.W.; B adham, G.B.; Go e, D.T.; Lozie , S.L. Segmen al bioelec ical impedance analysis: Theo y and
applica ion o a new echnique. J. Appl. Physiol. 1994,77, 98–112. [C ossRe ]
75.
Sipahi, S.; Hu , E.; Demi as, S.; Kocayigi , I.; Bozku , D.; Tame , A.; Gunduz, H.; Duman, S. Body
composi ion moni o measu emen echnique o he de ec ion o olume s a us in pe i oneal dialysis
pa ien s: The e ec o abdominal ullness. In . U ol. Neph ol. 2011,43, 1195–1199. [C ossRe ]
76.
Pa men ie , S.P.; Schi u schke, H.; Schmi , B.; Schewe, J.; He b ig, K.; Pis osch, F.; Passaue , J. In luence
o pe i oneal dialysis solu ion on measu emen s o luid s a us by bioimpedance spec oscopy. In . U ol.
Neph ol. 2013,45, 229–232. [C ossRe ]
77.
A oyo, D.; Panizo, N.; Abad, S.; Vega, A.; Rincón, A.; de José, A.P.; López-Gómez, J.M. In ape i oneal luid
o e es ima es hyd a ion s a us assessmen by bioimpedance spec oscopy. Pe i . Dial. In .
2015
,35, 85–89.
[C ossRe ]
78.
Oh, K.H.; Baek, S.H.; Joo, K.W.; Kim, D.K.; Kim, Y.S.; Kim, S.; Oh, Y.K.; Han, B.G.; Chang, J.H.; Chung, W.;
e al. Con ol o luid balance guided by body composi ion moni o ing in pa ien s on pe i oneal dialysis
(COMPASS) s udy. Does ou ine bioimpedance-guided luid managemen p o ide addi ional bene i o
non-anu ic pe i oneal dialysis pa ien s? Resul s om COMPASS clinical ial. Pe i . Dial. In .
2018
,
38, 131–138.
79.
Tabino , M.; Da ies, S.J. The use o bioimpedance spec oscopy o guide luid managemen in pa ien s
ecei ing dialysis. Cu . Opin. Neph ol. Hype ens. 2018,27, 406–412. [C ossRe ]
80. Zha, Y.; Qian, Q. P o ein nu i ion and malnu i ion in CKD and ESRD. Nu ien s 2017,9, 208. [C ossRe ]
81.
Cupis i, A.; B uno i, G.; Di Io io, B.R.; D’Alessand o, C.; Pas icci, F.; Cosola, C.; Bellizzi, V.; Bolasco, P.;
Capi anini, A.; Fan uzzi, A.L.; e al. Nu i ional ea men o ad anced CKD: Twen y consensus s a emen s.
J. Neph ol. 2018,31, 457–473. [C ossRe ]
82.
Bellizzi, V.; Scal i, L.; Te acciano, V.; De Nicola, L.; Minu olo, R.; Ma a, M.; Guida, B.; Ciancia uso, B.;
Con e, G.; Di Io io, B.R. Ea ly changes in bioelec ical es ima es o body composi ion in ch onic kidney
disease. J. Am. Soc. Neph ol. 2006,17, 1481–1487. [C ossRe ]
83.
Lukaski, H.C.; Kyle, U.G.; Kond up, J. Assessmen o adul malnu i ion and p ognosis wi h bioelec ical
impedance analysis: Phase angle and impedance a io. Cu . Opin. Clin. Nu . Me ab. Ca e
2017
,20, 330–339.
[C ossRe ]
84.
Piccoli, A.; Codogno o, M.; Piasen in, P.; Naso, A. Combined e alua ion o nu i ion and hyd a ion in
dialysis pa ien s wi h bioelec ical impedance ec o analysis (BIVA). Clin. Nu .
2014
,33, 673–677. [C ossRe ]
85.
Piccoli, A.; Pas o i, G.; Codogno o, M.; Paoli, A. Equi alence o in o ma ion om single equency .
bioimpedance spec oscopy in bodybuilde s. B . J. Nu . 2007,97, 182–192. [C ossRe ]
86.
Te uel-B iones, J.L.; Fe nández-Lucas, M.; Ruiz-Roso, G.; Sánchez-Ramí ez, H.; Ri e a-Go in, M.;
Gomis-Cou o, A.; Rod íguez-Mendiola, N.; Que eda, C. Analysis o conco dance be ween he bioelec ical
impedance ec o analysis and he bioelec ical impedance spec oscopy in haemodialysis pa ien s. Ne ologia
2012,32, 389–395.
87.
Gonzales-Co ea, C.H.; Caicedo-E aso, J.C. Bioelec ical impedance analysis (BIA): A p oposal o he
s anda diza ion o he classical me hod o adul s. J. Phys. Con . Se . 2012,407, 012018. [C ossRe ]
88.
Gen on, L.; He mann, F.R.; Spö i, A.; G a , C.E. Associa ion o mo ali y and phase angle measu ed by
di e en bioelec ical impedance analysis (BIA) de ices. Clin. Nu . 2018,37, 1066–1069. [C ossRe ]
89.
Sil a, A.M.; Ma ias, C.N.; Nunes, C.L.; San os, D.A.; Ma ini, E.; Lukaski, H.C.; Sa dinha, L.B. Lack o
ag eemen o
in i o
aw bioimpedance measu emen s ob ained om wo single and mul i- equency
bioelec ical impedance de ices. Eu . J. Clin. Nu . 2018,1, 22. [C ossRe ]
90.
Di Somma, S.; Lukaski, H.C.; Codogno o, M.; Peacock, W.F.; Fio ini, F.; Asp omon e, N.; Ronco, C.;
San a elli, S.; Lalle, I.; Au unno, A.; e al. Consensus pape on he use o BIVA (Bioele ical Impendance
Vec o Analysis) in medicine o he managemen o body hyd a ion. Eme g. Ca e J.
2011
,4, 6–14. [C ossRe ]

Nu ien s 2019,11, 809 22 o 22
91.
Co ic, A.; Ciumanghel, A.I.; Si iopol, D.; Kanbay, M.; Dumea, R.; Ga ilo ici, C.; Nis o , I. Value o
bioimpedance analysis es ima ed “d y weigh ” in main enance dialysis pa ien s: A sys ema ic e iew and
me a-analysis. In . U ol. Neph ol. 2017,49, 2231–2245. [C ossRe ]
92.
Sco land, G.; C uickshank, M.; Jacobsen, E.; Coope , D.; F ase , C.; Shimono ich, M.; Ma ks, A.; B azzelli, M.
Mul iple- equency bioimpedance de ices o luid managemen in people wi h ch onic kidney disease ecei ing
dialysis: A sys ema ic e iew and economic e alua ion. Heal h Technol. Assess. 2018,22, 1–138. [C ossRe ]
©
2019 by he au ho s. Licensee MDPI, Basel, Swi ze land. This a icle is an open access
a icle dis ibu ed unde he e ms and condi ions o he C ea i e Commons A ibu ion
(CC BY) license (h p://c ea i ecommons.o g/licenses/by/4.0/).