NDT Plus (2010) 3 [Suppl 2]: ii32–ii36
doi: 10.1093/nd plus/s q064
E ec o long- e m s e oid wi hd awal in enal ansplan ecipien s: a
e ospec i e coho s udy
Miguel Gonzalez-Molina
1
, Miguel Angel Gen il
2
, Dolo es Bu gos
1
, Me cedes Cabello
1
,
Ca men Cobelo
1
, Jesús Bus aman e
3
, Ped o E as i
4
, An onio F anco
5
and Domingo He nández
1
1
Di ision o Neph ology (Renal T ansplan Uni ), Hospi al Uni e si a io Ca los Haya, Málaga, Spain,
2
Di ision o Neph ology (Renal
T ansplan Uni ), Hospi al Vi gen del Rocío, Se illa, Spain,
3
Di ision o Neph ology (Renal T ansplan Uni ), Uni e si y Hospi al
(Se illa), Hospi al Clínico Uni e si a io, Valladolid, Spain,
4
Di ision o Neph ology (Renal T ansplan Uni ), Clínica Uni e si a ia de
Na a a, Pamplona, Spain and
5
Di ision o Neph ology (Renal T ansplan Uni ), Hospi al de Alican e, Alican e, Spain
Co espondence and o p in eques s o: Miguel Gonzalez-Molina; E-mail: [email p o ec ed]
Abs ac
Backg ound. S e oids a e la gely e ec i e o he immu-
nosupp essi e ea men in enal ansplan pa ien s, bu
cause se e e side e ec s. Whe he s e oid wi hd awal con-
e s long- e m bene icial e ec s emains unclea .
Me hods. Da a on 4481 cada e ic kidney ansplan eci-
pien s we e collec ed o es ima e he impac o s e oid
wi hd awal on kidney unc ion and g a and pa ien su -
i al using mul i a ia e Cox eg ession models.
Resul s. A o al o 923 pa ien s (20.6%) had s e oid ea -
men wi hd awn. This was mo e common in ecipien s om
younge dono s and in olde ecipien s, and in ecipien s wi h
a i s ansplan , hose who had p e- ansplan o de no o di-
abe es melli us and hose wi h ewe episodes o acu e ejec-
ion (AR) (22.4% s. 29.2%, P< 0.001). Cox mul i a ia e
analysis s a i ying by p opensi y sco es showed ha long-
e m s e oid he apy was associa ed wi h a 70% inc ease in
he isk o pa ien dea h. The epea ed measu es linea model
showed ha , al hough he abb e ia ed Modi ica ion o Die
in Renal Disease (aMDRD) alues changed o e ime (P=
0.002), his was independen o s e oid wi hd awal (P=
0.08). In addi ion, o he 772 (17.2%) ecipien s who de el-
oped de no o diabe es melli us, 204 (26.4%) ceased an idia-
be ic he apy, wi h mo e o hese among hose who ceased
s e oids (23% s. 33.3%, P= 0.003). Blood p essu e, choles-
e ol and iglyce ide alues we e all signi ican ly lowe in
he pa ien s who ceased s e oids.
Conclusions. S e oid wi hd awal in selec ed pa ien s had
no nega i e e ec o e ime on enal unc ion and g a
su i al, and i was associa ed wi h educed mo ali y.
Keywo ds: g a ; pa ien su i al; s e oid wi hd awal
In oduc ion
S e oids ha e p o ed o be e y e ec i e o immunosup-
p essi e ea men in enal ansplan pa ien s, bu long-
e m he apy causes side e ec s leading o inc eased mo -
bidi y, mo ali y and economic cos s [1]. Acco dingly, i is
o in e es o educe he dose o s e oids, wi hd aw hem
ea ly o e en use a s e oid- ee immunosupp essan p o o-
col in he mode n ansplan e a in o de o imp o e kidney
ansplan ou come [2].
In he e a o cyclospo ine (CsA), whe he o no accom-
panied by aza hiop ine, s e oids began o be wi hd awn,
wi h he esul s showing ha ea ly wi hd awal, black ace
and enal unc ion we e all isk ac o s o acu e ejec ion
(AR) and long- e m g a loss [3,4]. In ligh o hese da a,
he Eu opean Bes P ac ice Guidelines o Renal T ans-
plan a ion ecommended ha s e oid wi hd awal is sa e on-
ly in low- isk pa ien s and ha , a e wi hd awal, enal
unc ion should be moni o ed ca e ully because o he isk
o p og essi e wo sening [5].
The in oduc ion o mycophenola e mo e il (MMF) e-
duced he numbe o episodes o AR, and s e oid wi hd a-
wal in low- isk pa ien s ea ed wi h CsA and MMF did no
inc ease he incidence o AR, while enal unc ion e-
mained s able [6]. The associa ion o MMF and ac olimus
(TAC) u he lowe ed he incidence o AR and enabled
ea lie and sa e wi hd awal o s e oids. As an example, a
andomized s udy o s e oid wi hd awal 3 days a e ans-
plan a ion in pa ien s ea ed wi h MMF, TAC and induc ion
wi h daclizumab showed no signi ican di e ences in he
incidence o AR and enal unc ion a 1yea compa ed wi h
he con ol g oup [7]. In selec ed pa ien s [s able enal unc-
ion, no AR and panel- eac i e an ibody (PRA) <50%] ea-
ed wi h TAC, MMF and s e oids who we e andomized o
con inue iple he apy o wi hd awal o s e oids beginning
3 mon hs a e ansplan a ion, he incidence o AR was 6%
in pa ien s who ceased s e oids and 3% in hose who we e
main ained on s e oids a e 2 yea s o ollow-up [8].
The aim o his e ospec i e coho s udy was o ana-
lyse he e ec s o s e oid wi hd awal on enal unc ion
and g a and pa ien su i al in deceased dono kidney
ansplan a ion pe o med in Spain since 1990.
© The Au ho 2010. Published by Ox o d Uni e si y P ess on behal o ERA-EDTA. All igh s ese ed. This is an Open Access a icle dis ibu ed unde he e ms o he
C ea i e Commons A ibu ion Non-Comme cial License (h p://c ea i ecommons.o g/licenses/by-nc/2.5), which pe mi s un es ic ed non-comme cial use, dis ibu ion and
ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly ci ed.
Ma e ials and me hods
S udy design and pa ien popula ion
This e ospec i e coho s udy was ca ied ou in all ecipien s om 34 o
he 38 adul kidney ansplan cen es in Spain. This popula ion ep esen s
96% o all adul ecipien s (≥18 yea s) who ecei ed a p ima y o epea
deceased dono ansplan pe o med du ing he calenda yea s 1990,
1994, 1998 and 2002 and who su i ed a leas 1yea , wi h a ollow-up
un il 31 Decembe 2005. The quali y o he da a se was e i ied by an-
dom examina ion o sou ce documen s a each o he pa icipa ing ans-
plan cen es.
A o al o 4842 ecipien s, mos ly Caucasian, we e analysed, o whom
361 we e excluded as hey had no ecei ed an ini ial s e oid he apy (n=
145) o due o lack o ollow-up da a (n= 131). The p ima y s udy end
poin was o analyse he long- e m e ec o s e oid wi hd awal on kidney
unc ion and g a and pa ien su i al.
Ins i u ional e iew and pa ien p o ec ion
Medical eco d e iew was pe o med acco ding o he Spanish law on
clinical da a con iden iali y. This s udy was app o ed by he e hics com-
mi ee o he hospi al and was conduc ed acco ding o he p inciples de-
sc ibed in he Decla a ion o Helsinki.
Clinical a iables
The a iables analysed included he cause o dono dea h ( auma o
s oke), age and gende o he dono and he ecipien , body mass index
(BMI), i s o e- ansplan a ion, ime on dialysis, PRA a peak and a
ansplan a ion, human leucocy e an igen (HLA) misma ches, cold and
e-wa m ischaemia imes, delayed g a unc ion (DGF), AR (p e and
pos -s e oid wi hd awal), p o einu ia, hepa i is C i us (HCV), diabe es
melli us p e- and pos - ansplan a ion, he apy wi h s a ins, immunosup-
p ession (in en ion o ea ), kidney unc ion, and g a and pa ien su i al.
De ini ion o a iables
Delayed g a unc ion was de ined as he equi emen o dialysis du ing
he i s week a e ansplan a ion, a e uling ou accele a ed o hype -
acu e ejec ion, ascula h ombosis and u ina y ac obs uc ion. AR was
de ined by he need o ea men , wi h o wi hou biopsy con i ma ion.
G a ailu e was de ined as dea h o e u n o dialysis. The o al numbe
o HLA misma ches was calcula ed as he sum o he misma ches in he
A, B and DR loci. New-onse diabe es a e ansplan a ion (NODAT) was
de ined as he need o ea men wi h insulin o o al an idiabe ic agen s
a e ansplan a ion.
Immunosupp essi e he apy
The pa ien s in he 1990 and 1994 coho s we e ea ed wi h CsA and
p ednisone, wi h o wi hou aza hiop ine. Mos o he 1998 coho was
ea ed wi h CsA, MMF and p ednisone, and he 2002 coho was ea ed
wi h TAC, MMF and p ednisone. Induc ion he apy wi h polyclonal o
monoclonal an ibodies was ecei ed by 17.7% and 34% o he pa ien s,
espec i ely.
Renal unc ion
Renal allog a unc ion was calcula ed om he se um c ea inine mea-
su emen , using he abb e ia ed Modi ica ion o Die in Renal Disease
(aMDRD) equa ion [9].
S a is ical analysis
Desc ip i e esul s a e exp essed as he mean ± s anda d de ia ion o
con inuous a iables. Compa isons o con inuous a iables be ween s udy
pe iods we e made by he Mann–Whi ney U- es . The chi-squa e es , o
Pea son, and Fishe 's exac es when app op ia e, we e used o in e -
g oup compa isons o ca ego ical a iables. Kaplan–Meie su i al
cu es we e used o es ima e g a and pa ien su i al, and he log- ank
es o compa e su i al cu es. Uni a ia e and mul i a ia e Cox p opo -
ional haza ds eg ession models we e used o iden i y baseline isk ac-
o s o g a ailu e and pa ien dea h. The ollowing a iables we e
included in he model: ecipien and dono age and gende , cause o dono
dea h, DGF, AR, PRA (≤15% s. >15%), p o einu ia (inc ease be ween 3
and 12 mon hs), c ea inine a 3 mon hs and he c ea inine del a (inc ease
be ween 3 and 12 mon hs), diabe es melli us be o e ansplan a ion, NO-
DAT, e- ansplan , he apy wi h s a ins, HCV-posi i e ecipien , and mean
imes o ansplan a ion and s e oid wi hd awal. S e oid wi hd awal was
in oduced in he Cox eg ession model as a ime-dependen a iable.
When s e oid wi hd awal was signi ican in he uni a ia e Cox model, a
mul i a ia e Cox model was pe o med en e ing a p opensi y sco e as an
independen a iable. This sco e was de ined as he condi ional p obabil-
i y o s e oids wi hd awal du ing ollow-up, based on he cha ac e is ic o
he ecipien [10]. The es ima ed p opensi y sco es, ca ego ized in o quin-
iles, we e used o s a i y Cox eg ession analysis. A gene al linea model
o epea ed measu es was used o assess he e ec o enal unc ion, es-
ima ed by aMDRD, on su i al. P- alues o <0.05 we e conside ed o
indica e s a is ical signi icance.
The SPSS p og am ( e sion 15; Inc, Chicago, IL) was used as a da a-
base and o he desc ip i e s a is ical analysis.
Resul s
Table 1 shows he signi ican demog aphic and back-
g ound cha ac e is ics o pa ien s wi h o wi hou s e oid
wi hd awal. S e oids we e wi hd awn in 923 (20.6%) pa-
ien s a e a mean ime o 3.7 ± 2.8 yea s (23.2% du ing
he i s yea ). Table 2 shows he immunosupp essi e ea -
men and di e ences be ween he di e en coho s. Wi h
e ec om he in oduc ion o MMF, and la e TAC
(1998 and 2002 coho s), s e oid supp ession was ea lie
and mo e usual. No signi ican di e ences we e ound in
cold ischaemia ime, BMI, gende , PRA a he ime o
ansplan o he his o ical peak alue o in HLA mis-
ma ches. The incidence o AR a e s e oid wi hd awal
was 3%.
G a su i al
The Kaplan–Meie uncenso ed and dea h-censo ed g a
su i al cu es we e signi ican ly g ea e in he pa ien s
who had s e oids wi hd awn. Mean g a su i al was
13.6 s.12.6yea s(P< 0.001) and 14.3 s.13.8yea s
(P< 0.001), espec i ely. The mul i a ia e Cox p opo ion-
al eg ession analysis showed ha hose pa ien s who did
no cease s e oids had a g ea e isk o dea h-censo ed
Table 1. Signi ican demog aphic and backg ound cha ac e is ics o he
pa ien s wi h and wi hou s e oid wi hd awal
S e oid wi hd awal
PNo Yes
T ansplan s (n) 3.558 923
Dono age (yea s) 42.7 ± 16.8 39.5 ± 16.8 <0.001
Recipien age (yea s) 46.1 ± 13.2 48.0 ± 13.1 <0.001
WIT (min) 46.7 ± 19.3 39.4 ± 19.3 <0.001
AR (%) 29.2 22.4 <0.001
Dono dea h (% s oke) 51.4 44.2 <0.001
Re- ansplan (%) 13.5 8.1 <0.001
Diabe es melli us (%) 4.7 9.9 <0.001
New-onse diabe es melli us (%) 15.5 23.1 <0.001
MWT (yea s) 3.4 ± 3.8 2.9 ± 3.8 <0.001
WIT, wa m ischaemia ime; AR, acu e ejec ion; MWT, mean wai o
ansplan a ion.
E ec o long- e m s e oid wi hd awal ii33
g a loss [ ela i e isk (RR) 1.50, 95% con idence in e -
al, 1.21–1.85; P< 0.001].
The mul i a ia e Cox model o ac o s p edic ing he el-
a i e isk o uncenso ed g a loss by p opensi y sco es
showed ha non-wi hd awal o s e oids ep esen ed an in-
c eased isk o g a loss o 36% (Table 3) a e adjus ing
o o he con ounde a iables, including ansplan yea .
Howe e , when he model was censo ed o pa ien dea h,
he di e ences we e no signi ican .
Pa ien su i al
Kaplan–Meie pa ien su i al (Figu e 1) was signi ican ly
g ea e in he pa ien s who ceased s e oids (P< 0.001).
Dea hs occu ed in 8.5% s. 12.5% (P< 0.001), espec-
i ely. Non-wi hd awal o s e oids was associa ed wi h a
highe isk o dea h (70%) in mul i a ia e eg ession anal-
ysis (Table 4).
Special men ion should be made o he analysis o he
1990 coho a e 15 yea s o ollow-up. O he 740 pa-
ien s, 143 (19.3%) ceased s e oids, wi h no di e ences be-
ween he wo g oups acco ding o dono age, PRA, cold
ischaemia ime, gende , AR du ing he i s yea , HLA o
Table 3. Mul i a ia e Cox model (uncenso ed): ac o s p edic i e o he
isk o g a loss s a i ied by p opensi y sco es
Be a
S anda d
e o
RR
(e
Be a
) 95% CI P
Recipien age
a
0.45 0.09 1.58 1.32–1.89 <0.0001
HCV
b
0.33 0.07 1.39 1.19–1.63 <0.0001
Re- ansplan
c
0.30 0.10 1.35 1.10–1.64 0.002
Dono age
d
0.19 0.08 1.21 1.02–1.45 0.02
Acu e ejec ion
e
0.33 0.06 1.39 1.21–1.58 <0.0001
S a ins a 1 yea
0.21 0.08 1.23 1.04–1.47 0.01
Se um c ea inine a
3 mon hs
0.53 0.06 1.69 1.49–1.93 <0.0001
Del a o se um c ea inine 0.67 0.04 1.96 1.80–2.13 <0.0001
P o einu ia a 3 mon hs 0.15 0.02 1.16 1.10–1.23 <0.0001
Inc ease in p o einu ia
om 3 o 12 mon hs
0.22 0.02 1.24 1.17–1.31 <0.0001
Diabe es melli us
g
No pos - ansplan
diabe es
0.00 1
P e- ansplan diabe es 0.62 0.15 1.86 1.37–2.51 <0.0001
Pos - ansplan diabe es 0.48 0.12 1.62 1.26–2.08 0.0001
No s e oid wi hd awal
h
0.31 0.09 1.36 1.44–1.63 0.0006
Included in he model bu no in he able was ansplan yea (P> 0.05).
HCV, hepa i is C i us; DGF, delayed g a unc ion; Del a o se um c e-
a inine, di e ence be ween 12 and 3 mon hs; CI, con idence in e al.
a
Recipien s <60 yea s.
b
Fi s ansplan .
c
Hepa i is C i us nega i e.
d
Dono <60 yea s.
e
No acu e ejec ion.
S a in ea men .
g
No p e- ansplan diabe es melli us.
h
S e oid wi hd awal.
Table 2. Immunosupp essi e ea men in he di e en coho s and da a
ela ed wi h s e oid wi hd awal
Yea o ansplan a ion
1990 1994 1998 2002
T ansplan s (n) 740 1040 1465 1236
CsA (%) 97.27 78.87 74.53 17.39
Az (%) 58.11 68.02 20.13 0.56
TAC (%) 0.13 0.77 12.08 66.66
P ednisone (%) 100 100 100 100
MMF (%) 0.13 0.58 62.59 79.77
ALA (%) 26.87 17.73 25.80 34.38
SRL (%) 2.38 3.31
Follow-up (yea s) 15 11 8 3
TPSW (%) 15 11 25.4 16.3
Fi s yea (%) 15.4 9.7 22 44.8
MTW (yea s) 6.1 ± 3.8 5.0 ± 2.6 3.2 ± 2.0 1.6 ± 0.8
ARSW (%) 5.4 3.9 0 3
CsA, cyclospo ine; Az, aza hiop ine; TAC, ac olimus; MMF, mycophe-
nola e mo e il; ALA, an i-lymphocy e an ibodies; SRL, si olimus; TPSW,
o al pe cen o s e oid wi hd awal; MTW, mean ime o s e oid wi hd aw-
al; ARSW, acu e ejec ion pos -s e oid wi hd awal.
Fig. 1. Kaplan–Meie pa ien su i al wi h and wi hou s e oid
wi hd awal (log- ank es , P< 0.001).
Table 4. Cox mul i a ia e model: ac o s p edic i e o he isk o dea h
s a i ied by p opensi y sco es
Be a
S anda d
e o
RR
(e
Be a
) 95% CI P
Recipien age
a
1.17 0.09 3.23 2.68–3.90 <0.0001
HCV
b
0.32 0.10 1.38 1.13–1.70 0.0017
DGF
c
0.18 0.09 1.20 1.00–1.44 0.04
Se um c ea inine a
3 mon hs
0.21 0.09 1.23 1.02–1.48 0.026
Del a o se um c ea inine 0.25 0.08 1.29 1.10–1.51 0.0014
P o einu ia a 3 mon hs 0.12 0.60 1.13 1.01–1.27 0.034
Inc ease in p o einu ia
om 3 o 12 mon hs
0.14 0.04 1.15 1.05–1.26 0.0001
Diabe es melli us
d
No pos - ansplan diabe es 0.00 1
P e- ansplan diabe es 0.68 0.19 1.97 1.35–2.87 0.0004
Pos - ansplan diabe es 0.45 0.16 1.58 1.15–2.16 0.0044
No s e oid wi hd awal
e
0.53 0.12 1.70 1.33–2.18 <0.0001
Included in he model bu no in he able was ansplan yea (P> 0.05).
HCV, hepa i is C i us; DGF, delayed g a unc ion; Del a o se um c e-
a inine, di e ence be ween 12 and 3 mon hs; CI, con idence in e al.
a
Recipien s <60 yea s.
b
Hepa i is C i us nega i e.
c
No DGF.
d
No p e- ansplan diabe es melli us.
e
S e oid wi hd awal.
ii34 M. Gonzalez-Molina e al.
cause o dono dea h. Di e ences we e ound, howe e ,
be ween hose who did no and hose who did cease s e oid
he apy in ecipien age (42.1 ± 12.2 s. 45.8 ± 11.7 yea ,
P= 0.001), ime on dialysis (3.7 ± 3.8 s. 3.0 ± 3.0 yea s,
P= 0.04), e- ansplan (11.7% s. 4.2%, P= 0.008), p e-
ansplan diabe es melli us (1.2% s.9.5%,P< 0.001)
and NODAT (11.2% s. 23.8%, P< 0.001). Again, non-
wi hd awal o s e oids was associa ed wi h a highe isk
o uncenso ed (RR 1.65, 95% con idence in e al,
1.21–2.25; P= 0.001) and dea h-censo ed g a su i al
(RR 1.48, 95% con idence in e al, 1.03–2.13; P= 0.03)
as well as pa ien dea h (RR 1.66, 95% con idence in e -
al, 1.11–2.49; P= 0.01). The mean aMDRD a 1yea in
he non-wi hd awal and he wi hd awal g oups was simila
(52.4 ± 18.7 and 49.2 ± 19.3 mL/min/1.73 m
2
;P= 0.69,
espec i ely), emaining non-signi ican o e ime.
Renal unc ion
The aMDRD was signi ican ly g ea e a he ime o
wi hd awal in he pa ien s who ceased s e oids (excep
in he 1990 coho ) and emained so du ing he ollow-
up (P< 0.05). By applying a gene al linea model o
epea ed means showed ha aMDRD alues changed
o e ime (P= 0.002), bu independen ly o s e oid wi h-
d awal (P= 0.08).
Ca dio ascula isk ac o s
P e- ansplan diabe es melli us was mo e p e alen in pa-
ien s wi h s e oid wi hd awal (Table 1). A o al o 772
(17.2%) pa ien s de eloped NODAT. O hese, 204
(26.4%) s opped ea men wi h insulin o o al an idiabe ic
agen s du ing he pos - ansplan pe iod, and he highes
pe cen age belonged o pa ien s who discon inued s e oids
(23% s. 33.3%, P= 0.003).
To al choles e ol alues we e signi ican ly lowe in he
pa ien s who ceased s e oids om he i s pos - ansplan
yea (220.7 ± 46.4 s. 214.4 ± 44.5 mg/dL, P= 0.001) o
he 10 h yea (211.7 ± 41.1 s. 199.9 ± 34.1 mg/dL, P<
0.001). A simila si ua ion was seen wi h he iglyce ides
om he i s yea (152.2 ± 74.0 s. 138.7 ± 65.6 mg/dL, P
< 0.001) o he 12 h yea (148.2 ± 81.8 s. 118.8 ±
49.8 mg/dL, P= 0.008).
Finally, al hough he numbe o an ihype ensi e d ugs
was signi ican ly highe in pa ien s who did no cease s e -
oids, bo h sys olic and dias olic blood p essu es we e sim-
ila du ing long- e m ollow-up.
Discussion
This e ospec i e coho s udy ca ied ou in adul enal
ansplan uni s in Spain analysed he e y long- e m im-
pac o s e oid wi hd awal on enal unc ion and g a and
pa ien su i al. This is a ho ly deba ed opic in he ield o
enal ansplan a ion ha has been analysed by many in-
es iga o s [11,12], al hough mos ly ocused on he sho -
o medium- e m isk–bene i a io [8,13,14].
Du ing he e a o CsA, an ea ly and apid discon inua-
ion o s e oids was associa ed wi h a high incidence o AR
and wo sening enal unc ion [15]. Howe e , in a ou o
ea ly s e oid wi hd awal was he ac ha ce ain side e -
ec s o s e oids, once begun, p og essed despi e wi hd aw-
al. Wi h his in mind, many s udies o s e oid wi hd awal in
his pe iod ocused mainly on analysing which g oup o
pa ien s could bene i om s e oid supp ession and wi h
e ec om when [2,3], limi ing he implemen a ion o his
he apeu ic s a egy in ou ine clinical p ac ice.
Wi h he in oduc ion o TAC and MMF, s e oid wi h-
d awal in selec ed pa ien s did no signi ican ly inc ease
he isk o AR o in luence g a o pa ien su i al and
kidney unc ion. Consequen ly, wi hd awal began o ake
place ea lie and in mo e pa ien s, as seen in his s udy
[16,17].
In suppo o hese a gumen s, a p ospec i e, mul i-cen-
e s udy o 1110 enal ansplan pa ien s wi h no immuno-
logical isk and wi h good enal unc ion who discon inued
s e oids a e 6mon hs showed inc eases o g a (81.9% ±
1.8% s. 75.3% ± 1.2%, P= 0.0001) and pa ien su i al
(88.8% ± 1.5% s. 84.3% ± 1.0%; P= 0.0016), as well as a
educ ion in ca dio ascula isk ac o s [18].
Al hough in ou s udy s e oid wi hd awal was la e , un-
scheduled and unde a iable c i e ia, he esul s a e simi-
la o p e ious epo s. O no e was he highe mo ali y in
he pa ien s who did no cease s e oids in all he mul i a -
ia e models es ed and in all he coho s s udied. We con-
duc ed an analysis o p opensi y o s e oid wi hd awal in
o de o a oid selec ion bias wi h he elimina ion o s e -
oids based on clinical cha ac e is ics du ing ollow-up. In
addi ion, s e oid wi hd awal was en e ed in he Cox model
as a ime-dependen co a ia e. Thus, pa ien s wi h no s e-
oid wi hd awal showed a highe ela i e isk o uncen-
so ed g a loss in he Cox eg ession analysis a e
adjus ing o o he con ounding co a ia es, including p o-
pensi y sco e. Ne e heless, his was no obse ed when
dea h-censo ed g a analysis was assessed. A highe mo -
ali y in he pa ien s who con inued s e oids may explain
hese di e ences.
The analysis o he 1990 coho dese es special men-
ion. This coho o 740 pa ien s wi h a ollow-up o
15 yea s expe ienced signi ican ly highe mo ali y in
hose who con inued s e oids, despi e he ac ha he o he
g oup (who did cease s e oids) had a highe ecipien age
and g ea e p opo ion o diabe es melli us. No di e ences
we e ound in he selec ion a iables mos commonly used
o wi hd aw s e oids (good enal unc ion and absence o
lowe incidence o AR).
S e oid wi hd awal is a usual clinical p ac ice in Spain,
whe e selec ed pa ien s only ecei ed 5 mg pe day o
p ednisone. Al hough his dosage may con e a low isk
o ca dio ascula disease, i is possible ha s e oid sup-
p ession could op imize he ca dio ascula p o ile by e-
ducing isk ac o s o mo ali y such as blood p essu e
o insulin esis ance, especially in p edisposed indi iduals.
In ou s udy, elimina ion o s e oids was associa ed wi h a
be e con ol o glucose me abolism, lipid p o ile and
blood p essu e compa ed wi h he ecipien s who con in-
ued s e oids. P e ious s udies ha e demons a ed simila
indings [18–20].
Rega ding enal unc ion, s e oid wi hd awal did no
modi y long- e m g a unc ion. Al hough enal unc ion,
E ec o long- e m s e oid wi hd awal ii35
e alua ed by aMDRD, showed changes o e ime, his e -
ec was independen o s e oid wi hd awal. This was ob-
se ed in all he coho s analysed and con i ms ha his
he apeu ic s a egy may p ese e enal unc ion in he
long e m, e en in he p esence o o he isk ac o s.
In conclusion, his la ge e ospec i e s udy demons a es
ha , wi h e ec om he e a o CsA, s e oid wi hd awal in
enal ansplan ecipien s does no ha e a nega i e impac
on g a unc ion o su i al, and is associa ed, in he long
e m, wi h a signi ican educ ion in mo ali y.
Acknowledgemen s. The au ho s hank Jo di Cu o o he s a is ical
analysis and Ian Johns one o help wi h he English language e sion
o he manusc ip . The au ho s also wish o hank D . Daniel Se on o
his aluable con ibu ion in encou aging and o ganizing he p ojec .
Con lic o in e es s a emen . None decla ed.
Re e ences
1. Augus ine JJ, H icik DE. S e oid spa ing in kidney ansplan a ion:
changing pa adigms, imp o ing ou comes, and emaining ques ions.
Clin J Am Soc Neph ol 2006; 1: 1080–1089
2. S a a RJ, A mb us MJ, Oh CS e al. Wi hd awal o s e oid inmu-
nosupp ession in enal ansplan ecipien s. T ansplan a ion 1988;
45: 323–328
3. H icik DE, Whalen CC, Lau man J e al. Wi hd awal o s e oids a e
enal ansplan a ion—clinical p edic o s o ou come. T ansplan a ion
1992; 53: 41–45
4. Sinclai NR. Low-dose s e oid he apy in cyclospo ine- ea ed enal
ansplan ecipien s wi h well- unc ioning g a s. CMAJ 1992; 147:
645–657
5. Eu opean Bes P ac ice Guidelines o Renal T ansplan a ion. Sec ion
IV: long- e m managemen o he ansplan ecipien . IV.3.1 Long-
e m immunosupp ession. La e s e oid o cyclospo ine wi hd awal.
Neph ol Dial T ansplan 2002; 17: 19–20
6. G inyo JM, Gil-Ve ne S, Se on D e al. S e oid wi hd awal in my-
cophenola e mo e il- ea ed enal allog a ecipien s. T ansplan a ion
1997; 63: 1688–1690
7. Te Muelen CG, an Riemsdijk I, Hene RJ e al. S e oid-wi hd awal
a 3 days a e enal ansplan a ion wi h an i-IL-2 ecep o alpha he -
apy: a p ospec i e, andomized, mul icen e s udy. Am J T ansplan
2004; 4: 803–810
8. Sola E, Al e ez MJ, Cabello M e al. Low-dose and apid s e oid
wi hd awal in enal ansplan pa ien s ea ed wi h ac olimus and
mycophenola e mo e il. T ansplan P oc 2002; 34: 1689–1690
9. Le y AS, Bosch JP, Lewis JP e al. A mo e accu a e me hod o es i-
ma e glome ula il a ion a e om se um c ea inine: a new p edic o
equa ion. Modi ica ion o die in Renal Disease S udy G oup. Ann
In e n Med 1999; 130: 461–470
10. Heinze G, Mi e baue C, Regele H e al. Angio ensin-con e ing en-
zyme inhibi o o angio ensin II ype 1 ecep o an agonis he apy is
associa ed wi h p olonged pa ien and g a su i al a e enal ans-
plan a ion. J Am Soc Neph ol 2006; 17: 889–899
11. Kasiske BL, Chakke a HA, Louis TA e al. A me a-analysis o im-
munosupp ession wi hd awal ials in enal ansplan a ion. J Am Soc
Neph ol 2000; 11: 1910–1917
12. Pascual J, Que eda C, Zamo a J e al. S e oid wi hd awal in enal
ansplan pa ien s on iple he apy wi h calcineu in inhibi o and
mycophenola e mo e il: a me a-analysis o andomized, con olled
ials. T ansplan a ion 2004; 78: 1548–1556
13. Maio ca R, C is inelli L, B uno i G e al. P ospec i e con olled ial
o s e oid wi hd awal a e six mon hs in enal ansplan ecipien s
ea ed wi h cyclospo ine. T ansplan P oc 1988; 20: 121–123
14. Schulak JA, Maye s JT, Mo i z CE e al. A p ospec i e, andomized
ial o p ednisone e sus no p ednisone main enance he apy in cy-
clospo ine- ea ed and aza hiop ine- ea ed enal ansplan pa ien s.
T ansplan a ion 1990; 49: 327–332
15. H icik DE, O'Tool M, Schulak JA e al. S e oid- ee immunosupp es-
sion in cyclospo ine- ea ed enal ansplan ecipien s: a me a-analy-
sis. J Am Soc Neph ol 1993; 4: 1300–1305
16. Wloda zyk Z, Waleszewsky J, Pe en F e al. F eedom om ejec ion
and s able kidney unc ion a e excellen c i e ia o s e oid wi hd awal
in TAC- ea ed kidney ansplan ecipien s. Ann T ansplan 2002; 76:
28–31
17. Woodle ES, Vincen i F, Lo be MI e al. A mul icen e pilo s udy
o ea ly (4-day) s e oid cessa ion in enal ansplan ecipien s un-
de simulec , ac olimus and si olimus. Am J T ansplan 2005; 5:
157–166
18. Opelz G, Döle B, Laux G. Fo he collabo a i e ansplan s udy-
Long- e m p ospec i e s udy o s e oid wi hd awal in kidney and
hea ansplan ecipien s. Am J T ansplan 2005; 5: 720–728
19. Ojo AO, Hanson JA, Wol e RA e al. Long- e m su i al in enal ans-
plan ecipien s wi h g a unc ion. Kidney In 2000; 57: 307–313
20. H icik DE, Ba ucci MR, Meyes JT e al. The e ec s o s e oid wi h-
d awal on he lipop o ein p o iles o cyclospo ine- ea ed kidney and
kidney-panc eas ecipien s. T ansplan a ion 1992; 54: 868–871
Recei ed o publica ion: 18.12.09; Accep ed in e ised o m: 29.3.10
ii36 M. Gonzalez-Molina e al.