RESEARCH ARTICLE
La e e eb al side e ec s in long- e m
su i o s o i adia ed childhood b ain umo
Mi o-Pekka JussilaID
1,2
, Tiina Remes
3,4☯
, Julia An onen
3,4☯
, A ja Ha ila-Saa i
5
,
Jaakko Niinima
¨ki
1,2,6
, Ty i Pokka
3,4
, Pa
¨i i Koskenko a
7
, Anna Su ela
7
, Sanna Toi iainen-
Salo
8
, Pekka A ikoski
9
, Pekka Riikonen
9
, Mikko A ola
10
, Pa
¨i i La
¨h eenma
¨ki
11
,
Ki s i Si kia
¨
12
, Heikki Ran ala
3,4
, Ma ia Suo-Palosaa iID
1,2,6‡
, Ma ja OjaniemiID
2,3,4‡
*
1Resea ch Uni o Medical Imaging, Physics and Technology, Facul y o Medicine, Uni e si y o Oulu, Oulu,
Finland, 2Medical Resea ch Cen e , Uni e si y o Oulu, Oulu, Finland, 3Depa men o Pedia ics and
Adolescence, Oulu Uni e si y Hospi al, Oulu, Finland, 4PEDEGO Resea ch Uni , Oulu Uni e si y Hospi al
and Uni e si y o Oulu, Oulu, Finland, 5Depa men o Women’s and Child en’s Heal h, Uppsala Uni e si y,
Uppsala, Sweden, 6Depa men o Diagnos ic Radiology, Oulu Uni e si y Hospi al and Uni e si y o
Oulu, Oulu, Finland, 7Depa men o Clinical Radiology, Kuopio Uni e si y Hospi al, Kuopio, Finland,
8Depa men o Pedia ic Radiology, HUS Medical Imaging Cen e , Radiology, Uni e si y o Helsinki and
Helsinki Uni e si y Hospi al, Helsinki, Finland, 9Depa men o Pedia ics and Adolescence, Kuopio
Uni e si y Hospi al and Uni e si y o Eas e n Finland, Kuopio, Finland, 10 Depa men o Pedia ics, Tampe e
Uni e si y Hospi al and Uni e si y o Tampe e, Facul y o Medicine and Li e Sciences, Tampe e, Finland,
11 Depa men o Pedia ics and Adolescen Medicine, Tu ku Uni e si y Hospi al and Tu ku Uni e si y,
Tu ku, Finland, 12 Depa men o Pedia ics and Adolescence, Helsinki Uni e si y Hospi al, Helsinki, Finland
☯These au ho s con ibu ed equally o his wo k.
‡ These au ho s also con ibu ed equally o his wo k.
*ma ja.ojaniemi@oulu. i
Abs ac
Pu pose
Long- e m side e ec s o he ea men s a e common in su i o s o i adia ed pedia ic
b ain umo s. Ionizing adia ion in combina ion wi h su ge y and chemo he apy du ing child-
hood may educe e eb al heigh and bone mine al densi y (BMD), and cause g ow h ail-
u e. The aim o his s udy was o e alua e he la e consequences o umo ea men s on
e eb ae in su i o s o childhood b ain umo s.
Me hods
72 adul su i o s (mean age 27.8 yea s, s anda d de ia ion 6.7) o i adia ed childhood b ain
umo we e s udied by spinal magne ic esonance imaging (MRI) o e eb al abno mali ies
om he na ional coho o Finland. Pa ien s we e ea ed in i e uni e si y hospi als in Finland
be ween he yea s 1970 and 2008. Subjec heigh and weigh we e measu ed and body
mass index (BMI) was calcula ed. The mo phology and heigh /dep h a io o he e eb ae in
he middle o he kypho ic ho acic cu a u e (Th8) and lumba lo dosis (L3) we e examined.
Ve eb ae we e analyzed by Genan ’s semiquan a i e (SQ) me hod and spinal de o mi y
index (SDI) was calcula ed. BMD was measu ed by using dual X- ay abso p iome y.
Resul s
4.2% (3/72) o he pa ien s had undiagnosed asymp oma ic e eb al ac u e and 5.6%
(4/72) o pa ien s had adia ion-induced dec eased e eb al body heigh . Male pa ien s had
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0209193 Decembe 18, 2018 1 / 14
a1111111111
a1111111111
a1111111111
a1111111111
a1111111111
OPEN ACCESS
Ci a ion: Jussila M-P, Remes T, An onen J, Ha ila-
Saa i A, Niinima¨ki J, Pokka T, e al. (2018) La e
e eb al side e ec s in long- e m su i o s o
i adia ed childhood b ain umo . PLoS ONE 13
(12): e0209193. h ps://doi.o g/10.1371/jou nal.
pone.0209193
Edi o : Robe Daniel Blank, Medical College o
Wisconsin, UNITED STATES
Recei ed: Oc obe 11, 2018
Accep ed: Decembe 1, 2018
Published: Decembe 18, 2018
Copy igh : ©2018 Jussila e al. This is an open
access a icle dis ibu ed unde he e ms o he
C ea i e Commons A ibu ion License, which
pe mi s un es ic ed use, dis ibu ion, and
ep oduc ion in any medium, p o ided he o iginal
au ho and sou ce a e c edi ed.
Da a A ailabili y S a emen : All ele an da a a e
in he pape and i s Suppo ing In o ma ion iles.
Funding: This wo k was suppo ed by Special S a e
G an s o Heal h Resea ch in he Depa men o
Pedia ics and Adolescence, Oulu Uni e si y
Hospi al, Finland (h p://www.ppshp. i) (TR); Va¨ e
Founda ion o Pedia ic Cance Resea ch, Finland
(h p:// a eensaa io. i/) (M-PJ TR); he Pa¨i ikki and
Saka i Sohlbe g Founda ion, Finland (h p://www.
pss-saa io. i/english.h m) (TR); he A o and Lea
Ylppo¨Founda ion, Finland (h p://www.alys. i/)
la e e eb ae compa ed wi h emales. Pa ien age a he ime o i adia ion, BMI and i a-
dia ion a ea co ela ed o e eb al mo phology di e en ially in males and emales. BMD
had no associa ion wi h he e eb al shape. Pa ien s who had ecei ed c aniospinal i adia-
ion we e sho e han he gene al popula ion.
Conclusion
Childhood b ain umo su i o s had a high numbe o e eb al abno mali ies in young
adul hood. I adia ion was associa ed wi h abno mal e eb al mo phology and comp o-
mised inal heigh . Male gende may p edispose e eb ae o he side e ec s o i adia ion.
In oduc ion
Cen al ne ous sys em umo s a e he mos common solid umo s in child en [1,2], and hey
accoun o 26% o all pedia ic cance s [3]. The mo ali y a e o b ain umo s has dec eased
o e ecen decades due o mo e e ec i e cance ea men s [1], esul ing in a g owing adul
popula ion o childhood b ain umo su i o s. T ea men op ions o b ain umo s include
su gical ope a ions, chemo he apy and adia ion he apy [4].
The mos common la e e ec s o childhood b ain umo s and hei ea men s a e neu o-
psychological and endoc inological dis u bances [4]. In addi ion, o he known la e e ec s
include ca dio ascula diseases, educed e ili y, seconda y malignancies, speci ic o gan oxic-
i ies and delayed g ow h [1,4,5]. A b ain umo i sel , adia ion he apy and o he ea men s
may a ec bone heal h [5–9]. Radia ion he apy can di ec ly cause bone des uc ion [6,10],
and c aniospinal i adia ion has been shown o a ec pa ien g ow h [11,12]. The lumba e -
eb ae a e mo e p one o g ow h impai men caused by i adia ion han ce ical o ho acic
e eb ae [11]. The wide he i adia ed a ea o he spine is, he s onge he impac on he
pa ien heigh [13]. The signal in ensi y o he e eb ae has been shown o be he e ogeneous
and high on T1-weigh ed (T1W) images a e c aniospinal i adia ion, sugges ing a y
eplacemen o spinal bone ma ow [13].
This s udy was ca ied ou o e alua e he la e e ec s on e eb ae in su i o s o i adia ed
childhood b ain umo . Fo his, a na ional coho o i adia ed b ain umo su i o s was
ga he ed, and he e eb al bone heal h o he subjec s was sys ema ically analyzed.
Me hods
S udy popula ion
The s udy popula ion consis ed o pa ien s who had been ea ed o b ain umo s in child-
hood be ween he yea s 1970 and 2008. B ain umo s we e diagnosed be o e he age o 16
yea s. The ea men o he b ain umo s included su ge y, i adia ion and/o chemo he apy.
The he apies had ended a minimum o 5 yea s be o e his s udy s a ed. All subjec s we e a
leas 16 yea s o age and no known p og essi e diseases had been diagnosed a he ime o he
p esen s udy. Pa ien s we e ea ed a he Uni e si y Hospi als o Oulu, Kuopio, Tu ku, Tam-
pe e o Helsinki. Clinical examina ions we e done on 74 o he 127 in i ed subjec s and spinal
magne ic esonance imaging (MRI) on 72 o he 74 clinically examined subjec s (Fig 1).
Ve eb al abno mali ies in i adia ed childhood b ain umo su i o s
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0209193 Decembe 18, 2018 2 / 14
(TR); he Founda ion o Pedia ic Resea ch,
Finland (h p://www.las en au ien u kimussaa io. i/)
(TR); he Cance Socie y o Finland (h ps://www.
cance socie y. i/) (MO HR); he Thelma Ma¨kiky o¨
Founda ion, Finland (TR); he Cance Founda ion o
No he n Finland (h ps://www.pohjois-
suomensyopayhdis ys. i/) (TR); he Alma and K.A.
Snellman Founda ion, Finland (h ps://www.
snellmansaa io. i) (MO TR); and Founda ion o he
Ma¨ a Donne Founda ion, Finland (TR). The
unde s had no ole in s udy design, da a collec ion
and analysis, decision o publish, o p epa a ion o
he manusc ip .
Compe ing in e es s: The au ho s ha e decla ed
ha no compe ing in e es s exis .
Spinal MRI analyses
Spinal MRI was done o 72 subjec s du ing he yea s 2010–2015. MRI scans we e done wi h a
Siemens Magne om Esp ee 1.5T scanne a Oulu Uni e si y Hospi al, Siemens A an o 1.5T
scanne s a Helsinki, Kuopio and Tampe e Uni e si y Hospi als, and a Philips Ingenia 1.5T
scanne a Tu ku Uni e si y Hospi al. The spinal MRI p o ocol included con as -enhanced
T1-weigh ed (T1W) spin echo (SE) sagi al and T2-weigh ed (T2W) SE sagi al sequences.
Gadolinium con as agen (Do a em, 0.2 ml/kg, Gue be , F ance) was used. MRI scans we e
e alua ed on a esea ch PACS/DICOM iewing applica ion o diagnos ic adiology (neaView,
Neagen, Helsinki, Finland). The shape o he e eb ae was e alua ed and iden i ied o possi-
ble ac u es. Ve eb ae om he ou h ho acic o he ou h lumba e eb a we e analyzed
acco ding o Genan ’s isual semiquan i a i e (SQ) me hod [14,15]. A spinal de o mi y index
(SDI) was calcula ed by summing he g ades o e eb al de o mi ies [14]. P ima y and sec-
onda y umo s o he spine we e excluded. To analyze he mo phology o he ho acic and
lumba e eb ae mo e speci ically, he heigh and dep h o one ho acic e eb a (Th8) in he
middle o kypho ic ho acic cu a u e and one lumba e eb a (L3) in he middle o lo do ic
lumba cu a u e we e measu ed. The heigh o he Th8 and L3 e eb ae was measu ed om
he lowes pa o he e eb al body. The dep h (an e io -pos e io measu emen ) o he Th8
and L3 e eb al bodies was measu ed om he supe io and in e io end pla es. The heigh /
dep h a io o hese e eb ae was calcula ed.
Clinical examina ion
Clinical examina ion was done o 74 pa icipan s by a pedia ic neu ologis (au ho TR). A
ques ionnai e was used o ga he in o ma ion abou p e ious ac u es. The umo ea men ,
including co icos e oid medica ion, was s udied om he pa ien iles. Subjec heigh and
weigh we e measu ed and body mass index (BMI) was calcula ed. The s anda d de ia ion
sco e (SDS) o he heigh o each subjec was ob ained by using he new Finnish g ow h s an-
da d [16]. All subjec s excep one we e capable o mo ing wi hou any walking aid.
Densi ome y
Subjec s we e examined using dual X- ay abso p iome y (DXA). He e, he esul s om ou
p e iously published DXA analyses [9] we e compa ed wi h he spinal MRI indings. As
Fig 1. Flowcha o he s udy subjec s.
h ps://doi.o g/10.1371/jou nal.pone.0209193.g001
Ve eb al abno mali ies in i adia ed childhood b ain umo su i o s
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0209193 Decembe 18, 2018 3 / 14
desc ibed ea lie , he bone mine al con en and bone mine al densi y (BMD) o he lumba
spine and ou emo al si es ( emo al necks and o al hips) we e measu ed by using DXA wi h
Luna P odigy DXA bone densi ome y a Oulu, Luna P odigy Ad ance DXA bone densi-
ome y a Kuopio, Luna iDXA DXA densi ome y a Tampe e (Luna Co po a ion, Gene al
Elec ic Madison, WI, USA), Hologic Disco e y A DXA a Helsinki and Hologic QDR 4500C
DXA densi ome y a Tu ku (Hologic Inc., Bed o d, MA, USA). The esul s we e exp essed as
age- and gende -no malized z-sco es p o ided by he manu ac u e s. The In e na ional Soci-
e y o Clinical Densi ome y ecommends epo ing BMD in p emenopausal women and in
men <50 yea s o age as z-sco es (BMDZ) [17]. A z-sco e o -2.0 indica es a esul below he
expec ed ange o age and gende [17]. Thus, wo BMD g oups we e o med, z-sco e >-2.0
g oup and z-sco e �-2.0 g oup.
S a is ics
The analysis o a iance (ANOVA) was used o de e mine whe he he e we e any s a is ically
signi ican di e ences be ween he means o mo e han wo g oups. A non-pa ame ic K us-
kal-Wallis es was used when he a iables we e no no mally dis ibu ed. The associa ion o
umo ea men s (i adia ion, chemo he apy and o al dose o co icos e oids adminis e ed
du ing b ain umo ea men ), BMI and ea men age o e eb a heigh /dep h a io was
examined wi h mul i a ia e linea eg ession model wi h a s epwise a iable selec ion p oce-
du e o iden i y he se o he a iables ha bes p edic ed e eb al mo phology. Spea man
co ela ion coe icien was used when e alua ing co ela ion be ween co icos e oid dose o
he e eb a. Subjec s we e analyzed in h ee age g oups (0–6, 7–11 and 12–16 yea s) aking
in o accoun di e en g ow h pe iods a he ime o he adia ion he apy. Fo all es s, a signi -
icance le el o <0.05 was used. All da a we e analyzed using IBM SPSS s a is ic 25.0 so wa e
(A monk, NY: IBM Co p.).
E hics
W i en in o med consen was ob ained om all he pa icipan s included in his s udy and/o
hei legal gua dians. The Resea ch E hics Commi ees o he No he n Os obo hnia Hospi al
Dis ic , No he n Sa o Hospi al Dis ic , Sou hwes Finland Hospi al Dis ic , Pi kanmaa
Hospi al Dis ic and Helsinki and Uusimaa Hospi al Dis ic app o ed he s udy. The
esea ch was in acco dance wi h he p inciples o he Decla a ion o Helsinki.
Resul s
Pa ien cha ac e is ics
The coho has been desc ibed in de ail by Remes e al. [9]. The mos common umo s o he
pa ien coho a e shown in Table 1. Males accoun ed o 63.5% (n= 47) o he subjec s and
emales 36.5% (n= 27). Su gical ope a ions we e done in 82.4% o he cases. In 10 cases, a
biopsy only was aken, and in h ee cases, no su gical p ocedu es we e done. A o al umo
esec ion was done in 29 and pa ial esec ion in 32 o he cases. No p ima y o seconda y
umo s o he spine we e ound.
All pa ien s had ecei ed i adia ion: 40.5% (n= 30 o 74) o he pa ien s had ecei ed c a-
niospinal i adia ion, and c anial i adia ion only was done o 59.5% (n= 44 o 74) pa ien s.
O hose who had ecei ed c anial i adia ion (n= 44), whole b ain i adia ion was done o
6.8% (n= 3 o 44) o he pa ien s, local umo i adia ion was done o 88.6% (n = 39 o 44) o
he pa ien s and s e eo ac ic i adia ion was done o 4.5% (n= 2 o 44) o he pa ien s. Chemo-
he apy was used in 63.5% (n= 47 o 74) o he cases: 90% (n= 27 o 30) o he pa ien s ea ed
Ve eb al abno mali ies in i adia ed childhood b ain umo su i o s
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0209193 Decembe 18, 2018 4 / 14
wi h c aniospinal i adia ion and 45.5% (n= 20 o 44) o he pa ien s ea ed wi h c anial i a-
dia ion only had chemo he apy.
The e we e 65 pa ien s who ecei ed co icos e oid he apy. The median cumula i e
dose as p ednisolone equi alen was 2.3 g/m
2
(IQR = 0.9–7.0). The median dose was highe in
hose pa ien s who had ecei ed chemo he apy s pa ien s who had no [(3.7 g/m
2
, IQR = 1.5–
8.2 s 1.1g/m
2
, IQR = 0.6–1.5), p<0.001]. Co icos e oid doses a ied be ween age g oups
(p= 0.001). The median co icos e oid dose was highes in he younges age g oup (0–6 yea s)
(5.7 g/m
2
) and lowes in oldes age g oup (12–16 yea s) (1.0 g/m
2
).
Ve eb al ac u es and end-pla e i egula i ies
In 18 subjec s, SDI was g ea e han 0, indica ing e eb al de o mi y. In 11 o he 18 cases,
he e we e only bo de line de o med e eb ae (SDI = 0.5) ha we e no conside ed o be de i-
ni e ac u es. When conside ing only he e eb ae wi h Genan ’s g ade �1 [14], se en o he
18 subjec s had a leas one e eb a mee ing he c i e ia o os eopo o ic ac u e. Howe e ,
ou o hese subjec s had dec eased e eb al body heigh and end-pla e i egula i ies in se -
e al adjacen ho acic e eb ae no ep esen ing ac u es (Fig 2A) bu o he adia ion-
induced changes in he spine [13]. Thus, he e we e h ee de ini e cases o e eb al ac u e o
Genan ’s g ade �1, shown in he ep esen a i e image Fig 2B. The h ee subjec s wi h e e-
b al ac u es we e he oldes pa ien s in his s udy. De ails o hese h ee subjec s and hei
umo ea men s a e shown in Table 2. The heigh o he ce ical e eb ae was no mal. The e
we e no ac u es o mo phological changes in ce ical e eb ae o he s udy subjec s.
Ve eb al mo phology
To e alua e he changes o he e eb al mo phology, heigh /dep h a io was calcula ed om
he Th8 and L3 e eb ae. Females had g ea e mean in heigh /dep h a io compa ed wi h
males o bo h Th8 and L3 e eb ae. Mean alues o Th8 and L3 heigh /dep h a io a e p e-
sen ed in Table 3.
Table 1. Pa ien cha ac e is ics and umo ypes.
Pa ien gende , n(%) male 47 (63.5)
emale 27 (36.5)
Tumo ype, n(%) As ocy oma 24 (32.4)
Medulloblas oma 20 (27.0)
Ependymoma 8 (10.8)
Ge minoma 7 (9.5)
O he 15 (20.3)
Age a diagnosis, y , mean (SD) male 7.7 (4.3)
emale 8.2 (4.3)
Age a i adia ion, y , mean (SD) male 8.1 (4.1)
emale 8.7 (4.0)
C aniospinal i adia ion, n(%) yes 30 (40.5)
no 44 (59.5)
Chemo he apy, n(%) yes 47 (63.5)
no 27 (36.5)
Age a he s udy MRI, y , mean (SD) male 26.8 (6.1)
emale 29.4 (7.4)
y = yea s, MRI = magne ic esonance imaging, SD = s anda d de ia ion
h ps://doi.o g/10.1371/jou nal.pone.0209193. 001
Ve eb al abno mali ies in i adia ed childhood b ain umo su i o s
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0209193 Decembe 18, 2018 5 / 14

Fig 2. Rep esen a i e spinal magne ic esonance imaging (MRI) showing e eb al changes on T2-weigh ed
(T2W) sagi al images. Se e al adjacen la ho acic e eb ae wi h end-pla e i egula i ies a e demons a ed on T2W
image o a male aged 31 yea s (A, a owheads). T2W image o a emale aged 43 yea s demons a es sligh ly comp essed
Th6 e eb a, Genan ’s g ade 1 (B, a ow) and e eb al ac u e o Th8, Genan ’s g ade 2 (B, a owhead).
h ps://doi.o g/10.1371/jou nal.pone.0209193.g002
Table 2. Desc ip ion o he subjec s who had e eb al ac u es.
Pa ien
gende
Genan ’s
g ade
Pa ien age (y ) a he ime
o spinal MRI
Co icos e oid dose
g/m
2
Chemo he apy C aniospinal
i adia ion
Z-sco e �
-2.0
P e ious numbe o ac u es
in long bones
male 1 43 1.72 yes no no 1
male 1 41 7.98 no no yes 1
emale 2 43 2.28 yes yes no 4
y = yea , MRI = magne ic esonance imaging
h ps://doi.o g/10.1371/jou nal.pone.0209193. 002
Table 3. Mean alues o Th8 and L3 e eb ae measu emen s in males and emales.
MRI esul s All n= 72 Male n= 46 Female n= 26 p- alue
Mean (SD) Mean (SD) Mean (SD)
Th8 heigh /dep h a io % 67.3 (9.0) 65.7 (9.0) 70.2 (8.4) 0.041
L3 heigh /dep h a io % 75.3 (9.9) 73.3 (10.0) 78.9 (8.9) 0.019
SD = s anda d de ia ion, Th = ho acic, L = lumba
h ps://doi.o g/10.1371/jou nal.pone.0209193. 003
Ve eb al abno mali ies in i adia ed childhood b ain umo su i o s
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0209193 Decembe 18, 2018 6 / 14
Mo phology o he Th8 e eb a. Th8 heigh /dep h a io di e ed be ween he ea men
age g oups. A s a is ically signi ican di e ence was obse ed be ween he younges and oldes
age g oups, [mean 69.8 s 61.3, (mean di e ence (MD) 8.5, 95% con idence in e al (CI) 2.4 o
14.7), p= 0.009]. Th8 heigh /dep h a io was he smalles in he oldes ea men age g oup
(Fig 3A). BMI co ela ed nega i ely o he Th8 heigh /dep h a io in males (Fig 3B). Co icos e-
oid doses did no co ela e o he Th8 heigh /dep h a io ( = -0.006, p= 0.962).
To e alua e i he a ea o he i adia ion had an e ec on e eb al mo phology, Th8 heigh /
dep h a io was compa ed be ween pa ien s who had ecei ed c anial i adia ion and pa ien s
who had ecei ed c aniospinal i adia ion. No s a is ically signi ican di e ences we e ound
in Th8 heigh /dep h a io be ween hese g oups [mean 68.2 s 65.9, (MD = 2.3, 95% CI -2.0 o
6.6), p = 0.287].
In mul i a ia e analysis, BMI (adjus ed β= -0.377, 95% CI (-0.678 o -0.076), p= 0.015),
ea men age (adjus ed β= -0.782, 95% CI (-1.248 o -0.316), p= 0.001) and pa ien gende
(adjus ed β= -5.984, 95% CI (-9.978 o -1.990), p= 0.004) associa ed signi ican ly wi h he Th8
mo phology (R
2
adj = 0.233). When unning mul i a ia e eg ession analysis sepa a ely on
males and emales, he highe ea men age associa ed signi ican ly wi h lowe Th8 heigh /
dep h a io in emales (adjus ed β= -0.902, 95% CI (-1.740 o -0.064), p= 0.036) (R
2
adj =
0.147). In males, highe BMI (adjus ed β= -0.638, 95% CI (-1.054 o -0.222), p= 0.003) and
olde ea men age (adjus ed β= -0.688, 95% CI (-1.254 o -0.123), p= 0.018) associa ed wi h
lowe Th8 heigh /dep h a io (R
2
adj = 0.259).
Fig 3. Associa ion o he e eb al mo phology o he BMI, mode o i adia ion and ea men age. Associa ion o he Th8 e eb a mo phology
wi h ea men age (A) and BMI (B). Associa ion o he L3 e eb a mo phology wi h BMI (C) and mode o i adia ion (D).
h ps://doi.o g/10.1371/jou nal.pone.0209193.g003
Ve eb al abno mali ies in i adia ed childhood b ain umo su i o s
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0209193 Decembe 18, 2018 7 / 14
Mo phology o he L3 e eb a. No co ela ion be ween heigh /dep h a io o L3 e eb a
and pa ien ea men age was de ec ed ( = -0.130). Howe e , L3 heigh /dep h a ios we e
smalle in pa ien s who had ecei ed chemo he apy [mean 73.5 s 78.5, (MD = 5.0, 95%CI 0.3
o 9.8), p= 0.038]. When emales and males we e analyzed sepa a ely, no s a is ically signi i-
can e ec o chemo he apy was de ec ed in emales (p = 0.427), whe eas in males a signi ican
e ec was obse ed [mean 70.6 s 77.4, (MD 6.8, 95% CI 1.0 o 12.6), p= 0.023]. In males, BMI
co ela ed nega i ely o he L3 heigh /dep h a io (Fig 3C), and he e was a nega i e co ela ion
be ween co icos e oid dose and L3 heigh /dep h a io ( = -0.333, p= 0.024).
When compa ing he e ec s o he di e en a eas o i adia ion, L3 heigh /dep h a io was
smalle in pa ien s who had ecei ed c aniospinal i adia ion e sus hose who had ecei ed
local, ull c anial o s e eo ac ic i adia ion (71.0 s 78.3 (MD = 7.3, 95% CI 2.9 o 11.8),
p= 0.002) (Fig 3D). In emales, no di e ence be ween he L3 heigh /dep h a io alues in di -
e en i adia ion g oups was no ed (p = 0.284). Howe e , he e was a signi ican di e ence
be ween he g oups in males [mean 68.8 s 77.0, (MD 8.3, 95% CI 2.8 o 13.7), p= 0.004] (Fig
3D). The eby, c aniospinal i adia ion associa ed wi h lowe L3 heigh /dep h a io in males.
In mul i a ia e eg ession analysis, pa ien gende (adjus ed β= -4.857, 95% CI (-9.381 o
-0.333), p= 0.036) and c aniospinal i adia ion (adjus ed β= -6.829, 95% CI (-11.265 o
-2.393), p= 0.003) associa ed wi h L3 e eb a mo phology (R
2
adj = 0.166). When compa ing
sepa a ely emales and males, emales did no ha e any s a is ically signi ican p edic o s o L3
e eb a mo phology. In males, c aniospinal i adia ion (adjus ed β= -8.377, 95% CI (-13.685
o -3.069), p= 0.003) and highe BMI (adjus ed β= -0.475, 95% CI (-0.950 o 0.0), p= 0.050)
associa ed wi h lowe L3 heigh /dep h a io (R
2
adj = 0.210), espec i ely.
Bone mine al densi y
The esul s om ou p e iously published DXA measu emen s [9] we e compa ed wi h he
spinal MRI indings in his s udy. The BMD measu emen da a we e a ailable o 72 subjec s
and in 70 o hose, da a we e eco ded om all i e measu emen poin s including bo h emo-
al necks, hip bones and lumba spine (Fig 1). When he cases wi h he z-sco e below he
expec ed ange (z �-2.0) in a leas one o he measu ed a eas we e compa ed wi h he cases
wi h z-sco es o >-2.0, no s a is ically signi ican di e ence in Th8 o L3 heigh /dep h a io
was de ec ed (Table 4). Di e ences on he ea men s be ween hese g oups a e shown in
Table 4.
Final heigh
The e ec s o he di e en i adia ion a eas on inal heigh we e analyzed. Pa ien s who had
ecei ed c aniospinal i adia ion we e sho e han hose ha ing unde gone b ain i adia ion
only, al hough he di e ence was s a is ically signi ican only in males ( emales p= 0.610 and
males p= 0.047). I espec i e o i adia ion ype, subjec s in his s udy we e gene ally sho e
compa ed o he gene al Finnish popula ion, wi h he nega i e heigh SDS dis ibu ion in bo h
gende subg oups (Fig 4,Table 5). Radia ion a ea p edic ed heigh SDS in males so ha he
pa ien s who ecei ed c aniospinal i adia ion ended o be sho e (p = 0.062). The e was no
signi ican co ela ion be ween Th8 o L3 heigh /dep h a io and subjec heigh i espec i e o
gende .
Discussion
Radia ion he apy is an e ec i e and o en necessa y ea men o malignan umo s and
imp o es su i al [4]. Howe e , he long- e m su i o s o cance su e om se ious side
e ec s caused by ionizing adia ion o su ounding issues o he umo [4,5,9,18]. In he
Ve eb al abno mali ies in i adia ed childhood b ain umo su i o s
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0209193 Decembe 18, 2018 8 / 14
Table 4. Tumo ea men s, ea men age and e eb al heigh /dep h a ios in subjec s wi h z-sco e �-2.0 and z-sco e >-2.0.
z-sco e �-2.0 z-sco e >-2.0 p- alue
Females
C aniospinal i adia ion, n(%) yes 5 (62.5) 3 (37.5) 0.004
no 1 (5.6) 17 (94.4)
Co icos e oid dose, median (IQR) 6.6 (7.4) 2.1 (3.3) 0.033
Age a he i adia ion, y , mean (SD) 7.2 (4.5) 9.3 (3.9) 0.278
Th8 e eb a heigh /dep h a io, mean (SD) 74.1 (5.9) 68.6 (8.9) 0.179
L3 e eb a heigh /dep h a io, mean (SD) 81.4 (9.1) 78.9 (8.6) 0.547
Males
C aniospinal i adia ion, n(%) yes 3 (14.3) 18 (85.7) 0.188
no 8 (32.0) 17 (68.0)
Co icos e oid dose, median (IQR) 1.5 (6.7) 1.4 (4.9) 0.894
Age a he i adia ion, y , mean (SD) 5.5 (2.9) 8.8. (4.2) 0.017
Th8 e eb a heigh /dep h a io, mean (SD) 67.3 (8.2) 65.4 (9.3) 0.548
L3 e eb a heigh /dep h a io, mean (SD) 76.5 (13.2) 72.7 (8.4) 0.260
IQR = in e qua ile ange, SD = s anda d de ia ion, y = yea s, Th = ho acic, L = lumba
h ps://doi.o g/10.1371/jou nal.pone.0209193. 004
Fig 4. Compa ison o he heigh o he s udy subjec s wi h he gene al Finnish popula ion. A: Heigh (cm) by gende , adul male heigh e e ence
(solid line), adul emale heigh e e ence (do ed line). B: Heigh s anda d de ia ion sco e (SDS) by gende wi h 0 SDS (dashed line). C anial i adia ion
(solid squa e), c aniospinal i adia ion ( iangle).
h ps://doi.o g/10.1371/jou nal.pone.0209193.g004
Table 5. Heigh and heigh SDS o he subjec s by i adia ion ype.
C anial C aniospinal p- alue
Heigh (cm), mean/median (SD)
Female 157.5/160.6 (10.9) 159.4/159.9 (4.0) 0.610
Male 172.9/172.6 (7.1) 168.3/169.0 (8.2) 0.047
Heigh SDS (uni ), mean/median
(SD)
Female -1.9/-1.3 (2.0) -1.5/-1.4 (0.7) 0.571
Male -1.1/-1.3 (1.3) -1.9/-1.8 (1.3) 0.062
S anda d de ia ion (SD), S anda d de ia ion sco e (SDS)
h ps://doi.o g/10.1371/jou nal.pone.0209193. 005
Ve eb al abno mali ies in i adia ed childhood b ain umo su i o s
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0209193 Decembe 18, 2018 9 / 14