Developing a national reference range for paediatric bone density

LEABHARLANN CHOLAISTE NA TRIONOIDE, BAILE ATHA CLIATH

TRINITY COLLEGE LIBRARY DUBLIN

OUscoil Atha Cliath

The University of Dublin

Terms and Conditions of Use of Digitised Theses from Trinity College Library Dublin

Copyright statement

All material supplied by Trinity College Library is protected by copyright (under the Copyright and

Related Rights Act, 2000 as amended) and other relevant Intellectual Property Rights. By accessing

and using a Digitised Thesis from Trinity College Library you acknowledge that all Intellectual Property

Rights in any Works supplied are the sole and exclusive property of the copyright and/or other I PR

holder. Specific copyright holders may not be explicitly identified. Use of materials from other sources

within a thesis should not be construed as a claim over them.

A non-exclusive, non-transferable licence is hereby granted to those using or reproducing, in whole or in

part, the material for valid purposes, providing the copyright owners are acknowledged using the normal

conventions. Where specific permission to use material is required, this is identified and such

permission must be sought from the copyright holder or agency cited.

Liability statement

By using a Digitised Thesis, I accept that Trinity College Dublin bears no legal responsibility for the

accuracy, legality or comprehensiveness of materials contained within the thesis, and that Trinity

College Dublin accepts no liability for indirect, consequential, or incidental, damages or losses arising

from use of the thesis for whatever reason. Information located in a thesis may be subject to specific

use constraints, details of which may not be explicitly described. It is the responsibility of potential and

actual users to be aware of such constraints and to abide by them. By making use of material from a

digitised thesis, you accept these copyright and disclaimer provisions. Where it is brought to the

attention of Trinity College Library that there may be a breach of copyright or other restraint, it is the

policy to withdraw or take down access to a thesis while the issue is being resolved.

Access Agreement

By using a Digitised Thesis from Trinity College Library you are bound by the following Terms &

Conditions. Please read them carefully.

Developing a National Reference

Range for Paediatric Bone Density

Doctor in Medicine

2014

I declare that this work

a) has not been submitted as an exercise for a degree at this or any other

University,

b) is entirely my own work,

c) may be lent or copied by the Library upon request.

Aisling Laura Dobbyn Snow

TRINITY COLLEGE

SUMMARY

This study aimed to acquire normative dual energy x-ray absorptiom etry (DXA]

bone density data for Irish Caucasian children. A total of 162 healthy Irish

Caucasian children

(84 male, 78 female) aged betw een six and

16 years

underw ent anthropom etric m easurem ents and a single DXA examination as part

of the study.

101

participants

(47 male, 54 female] returned diet and lifestyle

questionnaires and 48

participants

(21 male,

27 female] performed self-

assessm ent of pubertal stage. In order to take account of bone size, areal bone

mineral density (aBMD] results were converted to a volumetric estimate,

corrected BMD (BMDcorr] using the Kroger m ethod of BMDcorr = BMC/Volume =

aBMD x

[4/ ( i t

x

Width]]. Participants were grouped by sex and age and the Cole

and Green (LMS] method was used to analyse aBMD and

results by

median M(AgeO, coefficient of variation S(Agei] and the Box-Cox pow er L(AgeO.

Height and weight w ere analysed by age and compared to the Irish reference

data. Body mass index (BMI] was also calculated, as weight (kg] divided by

height^ (m2]. Separate comparisons of aBMD and BMDcorr by Tanner stage were

performed for male and female participants using a paired Student’s t-test.

Dietary factors w ere analysed by sex and duration of daily exercise was analysed

by age range for males and for females and compared to aBMD and to BMDcorr

using a paired Student’s t-test.

LMS coefficients are presented in individual tables for males and females and can

be used to calculate the aBMD or

Z-score for an individual child using the

given for m ales an d for females. P ubertal stage, d ie t and lifestyle data are

p r e s e n te d in ta b u la te d an d c h a rt form.

As expected, aBMD, BMDcorr an d h eig h t all increase w ith age in both m ales and

females. Overall, th e female p a rtic ip a n ts in th e stu d y a r e significantly taller th an

th e Irish re fe re n c e sta n d a r d ; male p a rtic ip a n ts also te n d to be taller th a n the

re fe re n c e s ta n d a r d b u t n ot to a significant level. Analysis of b o n e d en sity in th o se

w h o exercise m o re th a n an d less th an one h o u r p e r day reveals a significant

inverse re la tio n sh ip b e tw e e n h ig h er levels of exercise an d both aBMD and

BMDcorr in females aged b e tw e e n 11 an d 16 years. It is also ev id e n t that, a fte r the

age of 12 years, th e female s tu d y p a rtic ip a n ts b ec a m e less involved in exercise

w h e r e a s m ales co n tin u ed to m aintain th e ir levels of activity.

ACKNOWLEDGEMENTS

P rofe ssor Edna Roche has g e n e ro u sly p ro v id e d m e with th e s u p p o r t and ex p e rt

insight n e c e ssa ry to en a b le m e to bring s tr u c tu r e an d p u rp o s e to this research.

P rofe ssor Myra O’Regan has p ro v id e d invaluable guidance an d e x p e rtise in the

form ation of a solid statistical m eth o d o lo g y for d ata analysis.

I w a s lucky enough to begin my radiology tra in in g in th e Adelaide and Meath

Hospital in co rp o ratin g th e National C hildren’s Hospital w h e r e P ro fe sso r William

Torreggiani ta u g h t m e th e g re a t im p o rta n c e of dedication to continuing

education and re searc h. Dr Samuel Hamilton w a s also on h an d to p rovide

en c o u rag e m e n t, heavy doses of reality and th o ro u g h tu to rin g in p r o p e r

g ra m m a r. Dr Martin Ryan and Dr Eric Colhoun both en c o u ra g e d my i n te re st in

th e s ubspec ia lty of p a e d ia tric radiology, for w hich 1 am very grateful.

Ms Inez Byrne, Ms Anne Stafford, Ms Liz Shields and Ms Darina O’Sullivan

expertly p e r fo rm e d th e DXA scans for th e study. Ms B e rn a d e tte Rowley provided

d edicated an d v ery m uch a p p r e c ia te d a d m in istra tiv e s u p p o r t th ro u g h o u t. Ms

Elaine O’Mullane, re s e a rc h n u rs e in t h e D e p a rtm e n t of Paediatrics in Trinity

College Dublin, w a s alw ays available to assist w ith practical details. Dr Alison

Hurley w a s alw ays r e a d y to advise an d help w ith practical planning. Ms Alison

Clapp assisted im m e a su ra b ly by lending h e r lite ra tu re sea rch expertise.

DEDICATION

To my p a re n ts, Joan Dobbyn an d David Snow, to w h o m I ow e endless

a p p reciatio n for so m a n y things including exceptional e n c o u ra g e m e n t ov er my

m a n y y ea rs of study.

To my s iste r Fiona, w h o is alw ays available w ith help and e n c o u ra g e m e n t and

in sp ires m e w ith h e r am az ing creativity, c o m m itm e n t to family and dedication to

h a r d work.

TABLE OF CONTENTS

DECLARATION OF ORIGINAL AUTH ORSH IP...2

SUMMARY... 3

ACKNOWLEDGEMENTS... 5

DEDICATION... 6

TABLE OF CONTENTS... 7

LIST OF TABLES... 1 0

LIST OF FIGURES... 11

CHAPTER 1 - INTRODUCTION...12

1.1 Bone H e a lth ...13

1.2 P aediatric DXA... 14

1.3 DXA and Bone S iz e ...15

1.3A Correction of BMD Results fo r Bone Size...17

1.4 DXA Cross C alibration... 18

1.5 Influences on Bone S tru c tu re and H e a l th ... 19

l.S.i Ethnicity...

19

1.5.a Nutrition...19

l.S.iii Body Composition, Birth W eight...21

l.S.iv Exercise...

23

1.5.V Exposure to Cigarette Sm oke...

25

1.6 DXA Reference R a n g e s ...25

CHAPTER 2 - METHODOLOGY...2 8

2.1 P a rtic ip a n ts ...28

2.2 C onsent, A uxiology...29

2.3 P u b ertal A s s e s s m e n t...30

2.4 Diet an d Lifestyle Q u e stio n n a ire ... 30

2.5 Bone M ineral D ensity A s s e s s m e n t... 31

2.6 BMD C orrection for Bone S ize...32

2.7 Statistical A nalysis of Bone D en sity ...32

CHAPTER 3 - RESULTS...3 4

3.1 S tudy P o p u la tio n ... 34

3.2 Bone D e n sito m e try ...35

3.3 B irth W eight, Height, W eight, BM I... 40

3.4 P u b ertal S ta g e ...43

3.5 D iet...44

3.6 E x ercise... 46

3.7 S ed en tary P a stim e s...49

3.8 C igarette Sm oke E x p o su re ... 51

CHAPTER 4 - DISCUSSION...52

4.1 F indings in o u r Study C o h o rt...53

4.1.i N u tritio n

...

53

4.1.H Body C om position...53

4.1.iii Pubertal S ta g e...

54

4.1.iv E xercise...55

4.1.V Cigarette Sm oke E xp o su re...56

4.2 DXA A nalysis...57

4.3 DXA Bone Density Reference R anges...59

4.4 Conclusion... 59

APPENDICES...6 1

Appendix 1. P a r e n t/G u a r d ia n Inform ation S h e e t...61

Appendix 2. P ubertal Self A sse ssm e n t F o r m s ... 65

Appendix 3. Diet an d Lifestyle Q u e s tio n n a ir e ... 67

A ppendix 4. Bone Density R e s u lts ...68

A ppendix 5. Height, Weigh and Body Mass Index R e s u l t s ...70

A ppendix 6. P ubertal Stage R esu lts...73

A ppendix 7. Dietary F a c t o r s ... 75

LIST OF TABLES

T a b le 1.

Study participants by age and se x ... 35

T a b le 2,

LMS data for aBMD in m ale s...37

T a b le 3.

LMS data for BMDcorr in m ales... 38

T a b le 4.

LMS data for aBMD in fem ales... 38

T ab le 5.

LMS data for BMDcorr in fem ales... 39

T ab le 6.

aBMD by T anner stage in m ales... 44

T a b le 7.

BMDcorr by Tanner stage in m a le s ...44

T a b le 8.

Mean aBMD and breastfeeding in males and fem ales

45

T a b le 9.

Mean BMDcorr and breastfeeding in males and fem ales

45

T ab le 10. aBMD and exercise in m a le s... 47

T a b le 11. BMDcorr and exercise in m ale s...47

T a b le 12. aBMD and exercise in fem ales...47

T a b le 13. BMDcorr and exercise in females ... 48

T a b le 14. Bone density results by age in m ales...68

T ab le 15. Bone density results by age in fem ales... 69

T a b le 16. Height and w eight by age in m a le s ... 70

T ab le 17. Height and w eight by age in fem ales... 71

T a b le 18. Mean body m ass index by age in m a le s... 72

T a b le 19. Mean body m ass index by age in fem ales...72

T a b le 20. Self-assessed pubertal stage in m ales...73

T a b le 21. Self-assessed pubertal stage in fem ales... 74

.36

.37

.39

.40

.41

.42

.42

.43

.46

.48

,49

.50

.50

.51

LIST OF FIGURES

Age by aBMD by g e n d e r ...

Age by BMD

by g e n d e r ...

aBMD percentiles for m a l e s ...

aBMD percentiles for f e m a le s ...

aBMD by birth w e ig h t by g e n d e r ...

CHAPTER 1

INTRODUCTION

1.1 B one health

Bone health is a com plex concept t h a t has m a n y in te rw o v e n influencing factors.

The s tre n g th of a b o n e is d e te r m in e d n o t only by th e a r c h ite c tu re of its physical

s tr u c tu r e b u t also by th e d ensity of th e s t r u c t u r e itself(l). In childhood th e

im m a tu r e b u t grow ing skeleto n is in a s ta te of flux w h e r e b o n e s h a p e an d

com position a re c o n sta n tly developing u n d e r a m ultifactorial influence(2). Many

of th e key d e te r m in a n ts of b o n e d e v e lo p m e n t re la te to ethnicity an d o th e r

in h e rite d variables b u t n u m e r o u s e n v iro n m e n ta l factors also play an im p o r ta n t

role. T hese include diet, exercise, vitam in D levels, p u b e rta l stage, w eight, body

com position and p re s e n c e o r a b se n ce of negative im pacts such as chronic ill

health, b one disease, d e trim e n ta l m edica tions an d h o r m o n e im b alan ce (3 ). The

influence of modifiable en v iro n m e n ta l factors is of p a rtic u la r im p o rta n c e in

childhood an d adolescence, d u ring w hich tim e it is possible to significantly

im p a c t b o n e d e v e lo p m e n t and, su b seque ntly, influence bone h e a lth in

a d u lth o o d (4 ].

a c cu ra te analytic t e c h n iq u e s a n d a r e co n sid e re d to be th e gold sta n d a r d in

evaluation of b o n e s tr u c tu r e a n d m ass h o w e v e r th e y re q u ire, a t m inim um , a

substan tial b one biopsy an d a r e clinically im practical, especially in th e p ae diatric

p o p ulation[6, 7]. The m o s t clinically re le v a n t op tio n s for analysis of b one density

are dual energy x-ray a b s o r p tio m e tr y [DXA), q u a n tita tiv e c o m p u te d to m o g ra p h y

[QCT), p e r ip h e ra l QCT (pQCT] and, to a le sse r degree, q u a n tita tiv e u ltra so u n d

(QUS)(1, 5, 8],

The s tru c tu re and m a ss of b o n e affect b o n e s tre n g th in different w ays(9], QCT

has bee n k n o w n for s o m e tim e n o w to be a non-invasive, a c c u ra te m e th o d of

asse ssin g both th e s tr u c t u r e a n d d e nsity of b o n e (1 0 ]. Until th e re c e n t ad v e n t of

im p ro v ed CT technology an d ad v a n c e d so ftw a re analytic algorithm s, th e use of

QCT had been limited d u e to ra d ia tio n dose concerns, p artic u la rly in the

pa e d ia tric p o p u l a t i o n ( 5 , 11). pQCT, w h e re b o n e analysis of an a na tom ic

p e r ip h e r y is the focus, initially e m e r g e d as a lo w e r-d o se CT alternative, albeit

w ith o u t th e ability to p ro v id e spinal data o r th e level of detail collected by

QCT(l). Both CT m e th o d s hav e an a d v a n ta g e o v e r DXA, in th a t th e y are able to

p rovide stru ctu ra l d etails a b o u t th e bone being analysed. This can be com bined

w ith o th e r m e a s u r e m e n ts such as regional m uscle s tre n g th to en a b le in tere stin g

m e th o d s of in te r p r e ta tio n such as th e functional a p p r o a c h desc rib ed by

Schoenau e t al(12). At p r e s e n t h ow ever, th e n u m b e r of c e n tr e s offering QCT

imaging is limited an d few p a e d ia tric re fere n ce d a ta s e ts have bee n published.

1.2 Paediatric DXA

h o r m o n e in a bid to stabilise o r im prove b o n e density. W hilst th e s e

m ed ica tio n s can be beneficial, th e y are asso c ia te d with clinically significant

side effect profiles and it is highly desira b le to limit th e ir use to patien ts

m o s t likely to benefit. Misdiagnosis of low b o n e d ensity on DXA can lead to

th e in acc u rate identification of patien ts in nee d of pharm acologic

in terve ntion.

Multiple factors c o n trib u te to the m isdiagnosis of low b o n e d e nsity in children. In

one stu d y of pa e d ia tric DXA results, up to 8 8 % of scans had one or m o re

in te r p re ta tiv e erro rs, th e m o st co m m o n of w hich w a s th e use of a T-score r a th e r

th an a Z-score (14). O ther e r r o r s included technical p r o b le m s with data

acquisition, statistical e r r o r s in in te rp re ta tio n an d failure to ac co u n t for ethnicity,

sex or b o n e size.

1.3 DXA and Bone Size

Correction of DXA re su lts for b o n e size o r h eig h t is n e c e s s a ry bec au se DXA

s c a n n e rs o btain d ata by projecting an x-ray beam th ro u g h th e p a tie n t to a

Many pa e d ia tric p a tie n ts u n d e rg o in g DXA e x a m ina tion have chronic conditions

th a t place th e m a t risk of having sm aller th a n ave rage b o n e s a n d / o r p o o r b one

health. In o r d e r to p ro v id e a c cu ra te pa e d ia tric DXA results, it is im p o r ta n t th a t

imaging is p e r fo rm e d on a vi^ell-maintained DXA scanner, t h a t ac co u n t is m a d e of

p a tie n t h eight an d t h a t re su lts a r e p u t in th e co ntext of a re le v a n t re fere n ce

range. It has been suggested t h a t deviations in b o d y an d b o n e size could im pact

aBMD to a clinically significant d egree in up to 17% of children u n d e rg o in g DXA

im aging(16).

Volum etric es tim a te s of aBMD re su lts a t t e m p t to tak e ac co u n t of v aria tions in

b o n e size and have bee n show^n to ca rry a h ig h e r coefficient of v a r ia n t to ash

analysis th an aBMD(7). in o r d e r to m aintain an acceptable level of accuracy in

th e use of DXA in children it is th e re fo re n e c e s s a ry to a d ju s t areal DXA d ata to

e stim a te a v o lu m etric result. Accordingly, a guideline d o c u m e n t p u b lished by the

In te rn a tio n a l Society for Clinical D e n sito m e try (ISCD] states th a t p ed iatric aBMD

re su lts should be c o rre c te d for height(17). The ISCD guidelines, p u b lished in

2007, do n o t specify th e m e th o d of c orrec tion t h a t should be used. Some DXA

s y ste m s p rovide an inbuilt option for height correc tion b u t th e ab se n ce of such

an option is com m on. DXA scans p e r fo rm e d a t th e Adelaide an d Meath Hospital

in c o rp o ra tin g th e National C hildren’s Hospital (AMNCH) are p e r fo rm e d on a GE

L unar Prodigy DXA system t h a t utilises th e m o s t re c e n t GE p a e d ia tric analysis

softw are. C o rre sp o n d e n c e w ith GE H ealthcare p rio r to this stu d y re vealed that,

w h ilst th e m achine u ses ad v a n c e d p ed iatric softw are, t h e r e is no available

1.3.i Correction o f BMD results fo r Bone Size

The literature reveals two principal methods for the adjustm ent of aBMD

results for bone size to produce a volumetric BMD (vBMD) estimate. The

method described by Kroger et al, in which the vertebral body is assumed to

be cylindrical in shape, provides a corrected BMD [BMD

corr

Kroger et al:

B M D corr = BMC/Volume = aBMD

x

[4/ ( t t

x

Width)]

The method described by Carter et al, who coined the term bone mineral

a pparent density (BMAD), uses bone area to estimate bone width(18, 19):

Carter et al:

BMAD = BMC/Volume = BMC/(Area)i-5

Both the Kroger and Carter methods provide densitom etry results in g/cm^

and both have been used to estimate vBMD in published studies. The Carter

method initially provided an estimate of bone volume at a time before DXA

scanners had the ability to automatically m easure vertebral width. Vertebral

width is now routinely provided as p a rt of the result data. The Kroger

method was chosen for use in this study as it requires p aram eters that are

readily accessible in GE Lunar Prodigy reports and is used by a num ber of

academic groups who have acquired and published normative paediatric

In o r d e r for BMDcorr to be clinically useful, it n e e d s to be r e p o rte d as a Z-

score in th e co ntext of a re fere n ce ra n g e th a t is re le v a n t to both th e p a tie n t

being asse sse d an d th e DXA system p e r fo rm in g th e d ata acquisition and

analysis.

1.4 DXA Cross Calibration

Detailed analysis of d e n s ito m e try re s u lts from different DXA system s suggests

t h a t th e p r im a r y cause for in te r-sy ste m v aria tio n is th a t th e y em ploy different

m athem atical e q u a tio n s in b o n e d e te c tio n [2 0 ). C om parison of DXA results

obtained on different s c a n n e rs is n o t s tra ig h tfo rw a rd an d re q u ir e s the

calculation an d im p le m e n ta tio n of com plex m a th em atica l 'cross calibration'

eq u a tio n s(2 1 -2 3 ]. Significant differences in Z-scores have even been

d e m o n s tra te d a m o n g re fere n ce d a ta b a s e s a c q u ired on the sa m e b ra n d of

sca n n e r[2 4 ). In th e case of so m e system s, for exa m p le GE DPX-L and GE Lunar

Prodigy, th e c o m p a riso n of d ata is m a d e m o re com plicated by funda m e ntal

differences in scan acquisition tec h n iq u e . Older scanners, such as th e DPX-L, use

a 'pencil b e a m ’ te c h n iq u e w hile t h e L u n ar Prodigy u ses a m o re adva n ced 'fan

b e a m ' m e th o d o f imaging. It is well d o c u m e n te d t h a t pencil bea m an d fan beam

DXA sy stem s p r o d u c e differing results, w h ic h can be clinically significant[25).

One s tu d y th a t eva lu a ted DXA d ata a c q u ir e d using GE DPX-L a n d GE Lunar

P rodigy DXA s y stem s in th e sa m e children re v e a le d t h a t lu m b a r spine BMD w as

1.6% hig h er on th e Prodigy th a n t h e DPX-L system , with p<0.0 0 01(26].

re le v a n t to th e b o d y region in question. This allows an estim atio n of th e over- or

u n d e r -e s tim a tio n of DXA p a r a m e te r s m e a s u r e d on one system b u t analysed

using a re fe re n c e ra n g e ac q u ired on a n o th e r (2 7 ). Cross calibration e q u a tio n s

th e re fo re differ d e p e n d in g on th e specific DXA s c a n n e rs concerned. The

im p le m e n ta tio n of cross calibration e q u a tio n s reduces, b u t d oes n ot eliminate,

th e v aria tio n in t h e s e re su lts(2 8 ).

1.5 Influences of Bone structure and Health

1.5.1

Ethnicity

Ethnic differences in BMC and aBMD w e re confirm ed in th e Bone Mineral Density

in Childhood Study (BMDCS)(29]. Bone size has been sh o w n to be th e p rim a ry

d e te r m in a n t of differences in BMC b e tw e e n ethnicities; a s e c o n d a ry effect is

m e d ia te d by extrinsic factors such as diet an d exercise[30].

l.S .ii Nutrition

The 2003 W orld Health O rganisation [WHO) r e p o r t on diet, n u tritio n and

be o th e rw is e healthy.

Vitamin D3 [cholecalciferol) is g e n e ra te d in th e skin w h e n it is exposed to

ultrav io let light. Vitamin D2 (e rgosterol], on th e o t h e r hand, is ingested from

d ie ta ry sources. Both cholecalciferol an d erg o ste ro l u n d e r g o initial hydroxylation

in th e liver, re sulting in m etabolically active 25-dihydroxycholecalciferol

[25[OH)D) com p o u n d s. Some 25(OH)D is co n v e rted in th e kidneys to calcitriol,

w hich has a re g u la to ry effect on th e level of calcium in th e blood an d has a

positive im p a c t on b o n e g ro w th an d rem odeling. It is 25(OH)D th a t is u sed as a

m e a s u r e of system ic Vitamin D; p lasm a 25(OH]D less th a n 5 0 n m o l/I is

co n sid e re d low in Ireland.

Low vitam in D levels, especially co m m o n in w in t e r an d s p rin g in Europe, m ay

e x a c e rb a te t h e d e trim e n ta l effect of hypocalcaem ia on b o n e m inera lisatio n (3 3 ).

Vitamin D deficiency h as bee n s h o w n to co rrelate w ith in cre ased body fat and

w ith re d u c e d height, b u t d oes no t a p p e a r to be directly linked to lo w er p e a k bone

m a ss[3 4 ]. T here a r e conflicting r e p o r ts of the in d irec t im p a c t of low circulating

plasm a 25[OH)D on size-adjusted BMC b u t m e e tin g th e criteria for n orm al

vitam in D statu s in childhood has b e e n sh o w n by so m e to positively affect b one

m a ss(3 5 ). Despite a p re viously held belief th a t low m a te r n a l vitam in D levels

m ay a d v e rsely im p a c t infant BMC, r e c e n t stu d ies have failed to s u p p o r t this

theory, w ith no d e m o n s tr a b le association d etec te d (3 6 -3 9 ).

The effect of fruit an d veg e ta b le c o n su m p tio n on BMC and BMD has bee n widely

studied. Overall, th e evidence su g g ests a positive association b e tw e e n

c o n su m p tio n of fruit and, to a le s s e r degree, vegetables, a n d b o n e m ineral values

hea lth in d e p e n d e n t of gender, exercise, height, weight, BMI an d bo d y fat; milk

p ro te in co n su m p tio n in childhood a p p e a r s to confer b enefit to th e b one

m in era lisatio n p ro c ess (41-43). Studies hav e failed to s h o w a d irec t positive

effect of b re a stfe e d in g o v e r an d above form ula feeding on b o n e d en sity la te r in

childhood(44).

l.S .iii Body Composition, Birth Weight

Body com position b ro a d ly refers to an in dividua l’s relative a m o u n ts of fat and

lean tissu e mass; it p ro v id es m o re detailed in form ation th a n bo d y m ass index,

which does no t differentiate b e tw e e n fat an d lean mass. As well as m easu rin g

b o n e d en sity p a r a m e te rs , DXA sc a n n e rs can p rovide body com position data. It is

im p o r ta n t to tak e into ac co u n t b o th lean b o d y m ass (which m ay have a positive

association w ith BMD) an d fat b o d y m ass (which m ay have a negative association

w ith BMD) w h e n evaluating b o n e hea lth in children(45). W hole b o d y DXA is one

m e th o d of estim atin g lean and fat mass. Although a d o lesc en ts w h o have higher

bo d y fat also te n d to have h ig h e r b o n e mass, t h e association a p p e a r s to be

m e d ia te d by th e ir conco m itan tly h ig h e r lean bo d y mass; it has b e e n s h o w n t h a t

fat m ass alone does n o t im p ro v e BMD in o v e rw e ig h t a d o le sc e n t boys(46, 47). In

girls w h o are o v e rw e ig h t o r obese, BMC an d BMD increase in p ro p o r tio n to the

in cre ased lean c o m p o n e n t of th e ir b o d y m ass r a t h e r th a n to th e fat

c o m p o n e n t(4 8 ). In fact, a lthough in c re a se d w eig h t is, overall, beneficial to bone

health, fat m ass (as o p p o se d to lean m a ss) m ay actually have a negative effect; in

tw o individuals of th e sa m e w e ig h t an d sex BMC has bee n found to be lo w e r in

in girls h a s a negative asso c ia tio n w ith vBMD in th e fe m u r an d tibia[51). It has

also b e e n s h o w n th a t ra ise d BMI in childhood can lead to early puberty, th u s

negatively affecting th e ac h ievable p e a k b o n e m ass(52).

Bone is c o m p o sed of a d e n se o u t e r rim of cortical b o n e an d a less d en se in n e r

core t h a t is re fe rr e d to as ‘ca n cello u s’ o r ‘tr a b e c u l a r ’ bone. T ra b e c u la r b o n e has a

la ttic e w o rk s tru c tu re , w ith a fine solid m a trix t h a t is s u r r o u n d e d by b one

m a rr o w . Bone m a r r o w v arie s in its com position; in childhood, hae m a to p o ie tic

m a r r o w p r e d o m in a te s but, as skeletal d e v e lo p m e n t p roc eeds, this is gradually

c o n v e rte d to m o re fatty m a rr o w . In a d d itio n to total body fat having an im p a c t on

b o n e density, th e a m o u n t of fat w ith in th e b o n e m a r r o w also a p p e a rs to have an

effect. The specific im p act o f m a r r o w fat on b o n e d en sity has bee n studied with

QCT analysis of m a r r o w com position. W h e re a s DXA ca n n o t differentiate the

relative p e rc e n ta g e s of tra b e c u la r an d cortical bone, QCT easily distinguishes

th e s e c o m p o n e n ts (5 3 ) . One s tu d y using QCT d e m o n s tr a te d th a t bone s tre n g th

w a s in c re a se d and m a r r o w fat w a s d e c re a s e d in female athletes, with the a u th o rs

suggesting th a t incre ased o s te o b la s t activity, trig g ere d by re d u c e d m a r r o w fat

levels, m ay play a key role in t h e e n h a n c e d b o n e s tre n g th conferre d by

exercise(54).

b e tw e e n u n co m p licated p r e m a tu r ity and low b o n e m a ss in adulthood,

p r e m a tu r ity asso c ia te d w ith v ery low b irth w e ig h t (VLBW] (birth w eig h t less

th a n l,5 0 0 g ) o r o th e r com plications has bee n sh o w n to be associated w ith

im p aired BMD a t th e p o in t of pea k b o n e m a ss in ea rly a d u lth o o d (6 0 ). In addition

to th e findings re g a rd in g b irth w e ig h t and la te r b o n e mass, low BMD in childhood

te n d s to p e r s is t o v e r m e d iu m -te rm follow u p (6 1 ). It has also b e e n sh o w n that,

w ith o u t b one-m odifying interventions, b o n e m ass tra c k s along percen tile curves

in adolesc ence (62, 63].

l.S.iv Exercise

Given th e m alleability of th e developing skeleton an d th e te n d e n c y of BMD to

follow pe rc e n tile curves over time, exercise d u rin g th e perio d of childhood bone

g ro w th is of p a rtic u la r benefit in maximizing b o n e s tre n g th and m inimizing

fractu re risk in ad u lth o o d (6 4 ]. Some in cre ase in BMC an d BMD continues even

b eyond th e age of p u b erty ; it has bee n s h o w n th a t exercise in y o ung m en can also

increase b o th BMC an d b o n e volu m e[6 5 ). However, th e tim e of p e a k effect of

exercise on b o n e s tr u c tu r e an d m in era lisatio n is in childhood and

adolesc en ce [6 6 ]. The perio d of maximal b enefit of exercise on b o n e hea lth in

girls h as bee n suggested to be T a n n e r stage I, before th e perio d of accelerated

skeletal g ro w th associated w ith p u b e r ty ( 6 7 ). Early m o d e ra te ly vigorous physical

activity has b e e n s h o w n in o th e r stu d ies to have a lo n g -term beneficial effect on

BMC in boys[68]. In fact, th e b enefit of w e ig h t b e a rin g exercise in childhood can

be seen well into adulth o o d , in th e form of m e a s u r e d s tru c tu ra l an d m ineral

b e n e fit in te rm s of b o n e h ea lth p a r a m e te r s into a d u lth o o d ev e n if partic ip a tio n in

physical activity ceases[71). Similarly, contin u atio n of childhood exercise

p a rtic ip a tio n from adolesc ence into early a d u lth o o d assists in m a intaining pea k

b o n e m a ss in females[72).

In a d d itio n to th e tim ing of exercise in childhood, t h e type, in tensity an d daily

d u ra tio n of physical activity a r e im p o rta n t d e te r m in a n ts of b o n e health. The type

of exercise chosen d e te r m in e s n o t only w h e th e r t h e r e will be a benefit to b one

h e a lth b u t also w hich bones, if any, will be affected. Aerobic b u t non -w e ig h t

b e a rin g s p o rts re d u c e BMl b u t do n o t te n d to co n fer an in cre ase in bone

m a ss(7 3 , 74). P opu la tio n -b ased exercise p r o g r a m m e s for school aged children

h av e b e e n sho w n to have a beneficial effect on b o n e m ass a n d b o n e size[75). In a

S w edish study, school aged boys parta k in g in th e re c o m m e n d e d 60 m in u te s of

school exercise p e r w e ek w e r e c o m p ared w ith boys p a r ta k in g in 40 m in u te s of

school exercise p e r day; th o se w ith th e higher exercise partic ip a tio n had

i n c re a se d lu m b a r BMC afte r tw o y ea rs(7 6 , 77). One stu d y d e te r m in e d th a t a

positive effect on b o n e m a ss w a s seen w ith just 28 m in u te s p e r day of vigorous

should be performed every day[81).

1.5.V

Sm oking

In addition to the more established non-hereditary influences on bone

health, there is limited evidence to support an association betw een passive

exposure to household cigarette smoke in early adulthood and reduced

prem enopausal bone mass in adult women(82]. There is little in the

literature th at addresses the impact of passive cigarette smoke on bone

health in children.

1.6 Reference Ranges

In adult women, significant differences betw een BMD z-scores calculated from

US and UK reference data suggest that the populations cannot be used

interchangeably for the purpose of calculating Z-scores(83). This may also apply

to paediatric DXA scanning. Analysis of Irish and UK height and weight reference

data by Hoey also revealed significant differences in the trajectory of the

percentile curves betw een the Irish and UK populations(84].

In the lead up to this research an initial audit was undertaken to determine

w h e th e r it would be worthwhile developing an Irish reference range for

pediatric aBMD and

The aOMD results of 66 children who were

of p a tie n ts w e r e included; th o s e b o rn small for gestational age an d had

failed to d e m o n s tr a te catch-up g ro w th [n=19), an d a g roup of p a tie n ts with

cystic fibrosis [CF] (n=47).

w a s calculated for each p a tie n t using the

Kroger m e th o d a n d su b s e q u e n tly co m p a re d to age- and sex-m atched

re fe re n c e d ata from th e N e th e rla n d s(8 6 ). The Dutch s tu d y w a s chosen

b e c a u se its c o h o r t closely r e s e m b le d th e p ro sp e c tiv e Irish c o h o rt of healthy

Caucasian children. In addition, th e stu d y p u b lish ed th e s ta n d a r d deviation

r e q u ir e d to allow calculation of a Z-score. The p a tie n ts ' original Z-score,

calculated from aBMD, an d t h e i r heig h t-a d ju sted Z-score, calculated from

B M D c o r r ,

w e r e c o m p a r e d by a p plying S tu d e n t's t-test. We found th a t th e Z-

scores for BMDcorr differed significantly from Z-scores for aBMD in both

g ro u p s studied. In the SGA g roup, th e m ean aBMD Z-score w a s -1.1 and the

m ean BMDcorr Z-score w a s 0.1 [p=0.000). In th e CF group, th e m ean aBMD Z-

score w a s -1.3 a n d the m ean

Z-score w a s -0.4 (p=0.002). Overall, the

m e a n aBMD Z-score w a s -1.1 a n d the m ean BMD

S C O r0 W 3 S

(p=0.000). With conve rsion o f aBMD to

t h r e e p a tie n ts w h o initially

had Z-scores th a t w e r e low en o u g h for th em to be co n sid ered ca ndida tes for

b is p h o s p h o n a te o r g ro w th h o r m o n e th e r a p y w e r e found to have BMDcorr Z-

scores above t h e th re s h o ld for tr e a tm e n t. This a u d it w as limited both by th e

a b se n c e of an Irish re fe re n c e ra n g e against w hich to co m p a re th e results

a n d by th e fact t h a t th e Dutch d a ta s e t w a s a c q u ired on a different DXA

s y stem [a GE s c a n n e r b u t of t h e DPXL r a th e r th a n L unar P rodigy subtype).

Despite its lim itations, t h e a u d it p ro v id ed som e s u p p o r t for th e

non-Irish, n o n -L u n a r Prodigy re fe re n c e d a ta s e t w as u s e d to eva lua te s ta n d a r d

deviations from th e m ean.

1.7 Research question

The aim of this s tu d y is to ac q u ire n o rm ativ e DXA d ata for Caucasian Irish

children sca n n ed on a GE L unar Prodigy DXA system . This is w ith a view to

p ro v iding a re le v a n t local re fe re n c e ra n g e th a t facilitates contextualisation of

p a e d ia tric DXA results, allows a d ju s tm e n t for b one size a n d m inim ises th e risk

th a t pa e d ia tric DXA re su lts will e r ro n e o u s ly lead to p h arm acologic in tervention

CHAPTER 2

METHODOLOGY

2.1 P a rticip a n ts

P articipants w e r e re c ru ite d from t h e Dublin a r e a th ro u g h th e National Children’s

Hospital [NCH] in Tallaght. The NCH is a 6 5-bed p a e d ia tric teaching hospital

associated w ith T rinity College Dublin a n d has Accident an d Emergency,

inpatient, o p erativ e an d o u tp a tie n t services; a p p r o x im a te ly 65,000 children

a tte n d th e hospital every year. A d v e rtise m e n ts w e r e placed in th e w aiting areas

of th e p a e d ia tric o u tp a tie n t clinic, o p e r a tin g ro o m and radiology d e p a r t m e n t in

An electronic a d v e r tis e m e n t visible to all staff m e m b e r s a t AMNCH w as also

placed on th e hospital in tr a n e t n o ticeb o a rd ; this w as r e n e w e d every 6 m o n th s

for th e first 18 m o n th s of th e study. T he m ajo rity of r e s p o n d e n ts contacted the

a d m in istra tiv e staff of th e pa e d ia tric d e p a r t m e n t of th e NCH in p e r s o n o r by

te lephone ; th e r e m a in d e r con tac ted th e p rinciple in v estig ato r by email. At least

th r e e a tte m p ts w e r e m a d e by th e p rincipa l in v estig ato r to c o n tac t each

re s p o n d e n t by telephone.

gestational age a t birth, b irth w eight, medical conditions, m edications, allergies

and fra c tu re history. P rospec tive p a rtic ip a n ts w e r e excluded if th e y w e r e not

Caucasian, h ad a diagnosed chronic medical c o ndition [including b one disease,

cystic fibrosis, in flam m ato ry bow el disease an d se v e re asth m a], used

m ed ica tio n s t h a t im p act b o n e hea lth (for e x a m p le oral steroids], had a significant

family h isto ry of p rim a ry o ste o p o ro sis o r had a h isto ry of fractu re (s] associated

w ith no or m inim al tr a u m a (for exam ple following a fall from s ta n d in g height].

Female p a r tic ip a n ts w h o had begun m e n s tr u a tio n w e re sch e d u le d for DXA

imaging b e tw e e n day one an d day ten of th e ir m e n s tru a l cycle. This w a s in

com pliance w ith th e local 'Ten-Day Rule’ re g u la tio n gov ern in g th e use of ionising

ra d ia tio n in females of childbearing potential.

2.2 Consent, Auxiology

Ethical ap p ro v al for th e stu d y w a s g ra n te d by th e joint AMNCH - St Jam es’s

Hospital (SJH] R esearch Ethics Com mittee. At th e tim e of th e ir child o r children's

scan each p a r e n t o r g u a rd ia n signed a form co n se n tin g to th e ir p articipation. The

s tu d y ra d io g ra p h e rs, w h o are tra in e d in t h e c o r re c t p r o c e d u re , p e r fo rm e d

Auxiological asse ssm e n t. Height w a s m e a s u r e d w ith a 'H a rp e n d e n ' s ta d io m e te r

using a s ta n d a r d iz e d te chnique , w ith t h e h e a d in th e F ra n k fu rt plane. W eight w as

a sse sse d using a self-zeroing Seca (H amburg, G erm any] electronic scales, w ith

th e p a r tic ip a n ts w e a rin g light in d o o r clothing a n d no shoes. Height an d w eig h t

w e re ana lyse d by age an d c o m p a r e d to t h e Irish re fe re n c e data. Body m ass index

2.3 Pubertal Assessment

Form al a s s e s s m e n t of p u b e rta l stage w a s offered to m ale p a rtic ip a n ts nine y ea rs

an d o lder an d to female p a rtic ip a n ts eight y e a rs and o lder w h o had n o t y e t had a

m e n s tru a l period. Due to v ery low ra te s of ac cep tan c e to u n d e rg o formal clinical

a sse ssm e n t, optional self-assessm en t of p u b e r ta l stage w a s offered as an

alternative; p u b e rta l se lf-assessm en t form s w e r e given to th e p a r e n t or g uardia n

of re le v a n t p a r tic ip a n ts for com pletion a t hom e. The sta n d a r d is e d m ale and

female forms asked p a rtic ip a n ts or th e i r p a r e n t o r g uardia n to asse ss w hich of a

serie s of s ta n d a r d is e d p h o to g ra p h s c o r re s p o n d in g to T a n n e r stages I to V m o st

closely re sem b le d th e ir p u b erta l stage (Appendix 2](87). The re la tionship of

T a n n e r stage to aBMD an d BMDcorr w a s evaluated.

2.4 Diet and Lifestyle Questionnaire

Diet an d lifestyle w e r e a sse sse d by postal q u estio n n aire . P a rticipants w e re asked

to detail th e ir dietary intake (dairy p ro d u c ts, vegan o r veg e ta rian diet and

d ie ta ry s u p p le m e n ts), physical activity levels, s e d e n ta r y habits, use of p o rtab le

digital devices an d e x p o s u r e to e n v iro n m e n ta l sm o k e ov er th e perio d leading up

to th e ir DXA a s s e ssm e n t. Completed form s w e r e re tu r n e d to th e principal

investig ato r in th e NCH radiology d e p a rtm e n t. Dietary factors w e r e ana lyse d by

2.5 Bone Mineral Density Assessment

Bone d e n s ito m e tr y w a s a s s e sse d using DXA imaging. DXA scanning w as

p e r fo rm e d by one of four pa e d ia tric DXA ra d io g r a p h e r s , ail of w h o m p erform

p a e d ia tric a n d a d u lt DXA scans on a w eek ly basis in AMNCH an d a r e tra in e d in

the use of th e h a r d w a r e an d softw are. All scans w e r e a c q u ired using a GE Lunar

Prodigy DXA s c a n n e r [GE Healthcare, Chalfont St Giles, United Kingdom]. The

s c a n n e r u n d e r g o e s daily quality a s s u ra n c e (QA] using a s ta n d a r d GE p h a n to m [a

p h a n to m is a precisely s ta n d a r d is e d model th a t can be u sed to t e s t th e accuracy

of a piece of imaging eq u ip m e n t). Weekly QA using th e specific lu m b a r spine

p h a n to m [L unar 1 8562) is also p e r fo rm e d a n d th e s c a n n e r is serviced regularly

according to th e GE service schedule. A single a n t e r o p o s te r io r [AP] DXA scan of

th e lu m b a r sp in e w a s p e r fo rm e d on each p artic ip a n t, w ith each DXA scan limited

to the AP p rojec tion of th e l u m b a r spine in o r d e r to m in im ise th e dose of

rad ia tio n im p a r te d to th e p a r tic ip a n ts while still p ro v id in g th e d ata re q u ir e d to

com plete t h e study.

The s ta n d a r d DXA b o n e d ata w a s ac q u ired for each p artic ip a n t, including BMC,

BMD an d v e r te b ra l w id th for each v e rte b ra l level from LI to L4. The DXA s c a n n e r

so ftw a re autom atically p r e s e n ts t h e d ata relative to t h e p r e - p r o g r a m m e d norm al

range, p ro v iding a Z-score base d on th e p a t i e n t ’s age a n d sex. The d ata w e re

t r a n s f e r r e d autom atically to th e p ictu re archiving an d com m u n icatio n system

[PACS) of th e h o spital radiology d e p a r t m e n t and s to r e d as a p e r m a n e n t p a r t of

th e p a r tic ip a n t’s medical record. Scans w e r e re p o r te d using th e s ta n d a r d

to have lov^ aBMD [a Z-score of <1 SD c o m p a r e d to th e age an d sex m atch e d

m e a n ] vi^ere re fe rr e d for clinical p a e d ia tric re v ie w in th e NCH.

2.6 BMD Correction for Bone Size

B M D

re sults w e r e ad ju ste d for b o n e size p ro d u c in g co rrec te d

B M D [ B M D c o r r ]

by

using th e m eth o d p u b lish ed by Kroger et al(18):

B M D c o r r

= BMC/Volume = aBMD x [ 4 /(tt

W idth)]

2.7 Statistical Analysis of Bone Density

P articipants w e r e g ro u p e d by sex an d age (in single years]. The LMS (Cole and

Green] m e th o d w a s u sed to d e te r m in e th e percen tile curves for aBMD an d for

B M D c o r r

for each g e n d e r using th e G eneralised Additive Models for Location, Scale

a n d Shape (GAMLSS] statistical m odelling package in th e R so ftw a re

language(88-90]. The d istrib u tio n of aBMD an d BMDcorr a t each g roup w a s

su m m a ris e d by th e m ed ian M(AgeO, coefficient of v ariation S(AgeO an d th e Box-

Cox p o w e r L(Agei)- The Box-Cox p o w e r w a s u sed to tra n s f o rm th e d ata to m ak e

th e m norm ally d is trib u te d w ithin a p a r tic u la r age group. A sim p le r m odel using

only m ean s an d v aria n c e s w a s also built, in w hich only th e M an d S c o m p o n e n ts

of th e LMS model w e r e used. W o rm plots w e r e u sed to check th e model.

th e re su lts b u t due to small sa m p le sizes, b o o ts tra p p in g of th e LMS m odel w as

n o t possible. B oo tstrap p in g of th e s im p le r MS m odel w a s found to be possible.

The re la tionship of both aBMD an d BMDcorrto b irth w e ig h t w a s analysed and

dep icte d using sc a tte r plots. S e p arate analysis of th e re la tio n sh ip of b o n e density

to b re a stfe e d in g w a s p e rfo rm e d . D uration of daily exercise w a s analysed by age

ra n g e for m ales a n d for females an d co m p a re d to aBMD an d BMDcorr using a

p a ire d S tu d e n t’s t-test.

In o r d e r to calculate th e s ta n d a r d deviation for aBMD o r BMDcorr in an individual

child th e re le v a n t LMS values for th e ir age and g e n d e r a r e used. The following

form ula calculates th e Z-score for a p a r tic u la r yj (in this case aBMD score) a t age

Agei:

CHAPTER 3

RESULTS

3.1 Study pop ulation

Over th e p e rio d of th e stu d y (January 2 0 1 0 to June 2012), a r e q u e s t for fu r th e r

inform ation a b o u t th e stu d y w as received from ap p ro x im a te ly 180 p a r e n ts o r

g uardians, m a n y of w h o m had m o re th an one child w ithin th e age range covered

by th e study. In total, th e p a r e n ts or g u a rd ia n s of a p p ro x im a te ly 260 children

b e tw e e n th e ages of 6 and 16 y e a rs re s p o n d e d to e ith e r th e p r in te d or electronic

a d v e rtise m e n t. The stu d y could n o t be d iscussed with the p a r e n ts or g u a rd ia n s of

a p p ro x im a te ly 40 children due to in co rrect c o n ta c t details being provided, the

r e s p o n d e n t no longer having in te re st in t h e stu d y o r the r e s p o n d e n t being

u nable to discuss th e stu d y a t the tim e of contact.

The total s tu d y p o p u latio n consisted of 162 h e a lth y Irish Caucasian children [84

male, 78 female] aged b e tw e e n six an d 16 y e a rs (Table 1). All p e r fo rm e d scans

Age (years)

G ender

Total

Fem ale (n)

Male (n)

6-6.9

7

8

15

7-7.9

8

6

14

8-8.9

9

4

13

9-9.9

8

11

19

10-10.9

6

9

15

11-11.9

10

7

17

12-12.9

10

9

19

13-13.9

9

11

20

14-14.9

6

9

15

15-15.9

1

7

8

16-16.9

4

3

7

Total

78

84

162

Table 1. Study p artic ip a n ts by age a n d sex.

3.2 Bone densitom etry

The aBMD of all female p a rtic ip a n ts w a s found to lie w ith in tw o sta n d a r d

deviations of th e m e a n for age w h e n e v a lu a te d using th e inbuilt DXA softw are; all

female p a rtic ip a n ts w e r e t h e re fo re included in th e statistical analysis. Two male

p a r tic ip a n ts (one aged six an d one aged 15 y e a rs] w e r e found to have an aBMD

less th a n or equal to tw o s ta n d a r d dev iatio n s belo w th e autom atically-calculated

m e a n for age; th ey w e r e u ltim ate ly included in th e re fe re n c e d ata as b o th had a

BMDcorr w ith th e n o rm al range. Due to t h e v ery small n u m b e r of male

p a r tic ip a n ts aged 16, th e s e p a rtic ip a n ts w e r e o m itted in th e analysis using th e

Cole and Green m ethod. T h re e fu r th e r p o in ts w e re also o m itted from th e m ale

LMS analysis; one low aBMD re su lt in a 14 y e a r old an d tw o low aBMD re su lts in

B o th aBMD a n d

i n c r e a s e d w i t h a g e in m a l e s a n d in fe m a le s ; t h e r e s u l t s

a r e p r e s e n t e d a s s c a t t e r p l o t s b y a g e a n d se x (F ig u re 1, 2). T h e LMS c o efficien ts

a r e p r e s e n t e d in in d iv id u a l ta b l e s fo r m a l e s (T a b le 2, 3) a n d f e m a l e s (T ab le 4, 5]

a n d c a n b e u tilis e d in c o n j u n c t i o n w i t h t h e f o r m u l a

z\ =

[yi/M(Agej)]L(Agei) - i

j

L(Agei)S(AgeO to c a lc u la te t h e aBMD o r

Z -sc o re fo r a n in d iv id u a l child.

T h e r e s u l t s o f t h e s i m p l e r MS m o d e l u s i n g o n ly m e a n s a n d v a r i a n c e s a r e

p r e s e n t e d a s p e r c e n t i l e c u r v e s fo r m a l e s (F ig u r e 3) a n d f e m a le s (F ig u r e 4); t h e s e

d e m o n s t r a t e g r e a t e r v a r i a t i o n for f e m a le s t h a n fo r m ales.

E

'

5

m

ra

Male

Female

10

12

A ge (y ears)

— V

14

•

.

I

16

• Male

Female

o

Q

5

10

I

12

14

16

Age (years)

Figure 2. Age by

b y gender.

aBMD for Males

Age (years)

L

M

S

6-6.9

4.4744

0.6412

0.1451

7-7.9

3.2363

0.6949

0.1405

8-8.9

2.1516

0.7333

0.1284

9-9.9

1.2635

0.7477

0.1143

10-10.9

0.5507

0.7501

0.1236

j

i i - i i . 9

0.0394

0.7636

0.1567

12-12.9

-0.2644

0.8096

0.1765

13-13.9

-0.4308

0.8996

0.1439

14-14.9

-0.5284

1.0209

0.0806

15-15.9

-0.6007

1.1497

0.0806

B M D c o r r

fo r M a les

Age (y ea rs)

L

M

S

6-6.9

1.0701

0.2739

0.4607

7-7.9

-0.0069

0.2772

0.4283

8-8.9

-0.7272

0.2799

0.3991

9-9.9

-0.9303

0.2813

0.3737

10-10.9

-0.6933

0.2824

0.3502

11-11.9

-0.2341

0.2851

0.3264

12-12.9

0.3035

0.2923

0.2996

13-13.9

0.9528

0.3061

0.2692

14-14.9

1.7363

0.326

0.2378

15-15.9

2.606

0.349

0.209

Table 3.

L M S

data for

B M D c o r r

in m ales.

aBMD for F em ales

Age (y ea rs)

L

M

S

6-6.9

-0.9908

0.656

0.088

7-7.9

-0.7609

0.698

0.1591

8-8.9

-0.531

0.74

0.2072

9-9.9

-0.3012

0.7821

0.184

10-10.9

-0.0713

0.8242

0.1502

11-11.9

0.1586

0.8664

0.1527

12-12.9

0.3885

0.9087

0.1654

13-13.9

0.6184

0.951

0.1383

14-14.9

0.8483

0.9934

0.0962

15-15.9

1.0782

1.0357

0.068

16-16.9

1.308

1.0781

0.0533

BMDcorr

for F em ales

Age (years)

L

M

S

6-6.9

-3.6708

0.2887

0.3157

7-7.9

-2.9743

0.297

0.3125

8-8.9

-2.2779

0.3052

0.3097

9-9.9

-1.5815

0.3134

0.3072

10-10.9

-0.8851

0.3217

0.3047

11-11.9

-0.1887

0.3299

0.3026

12-12.9

0.5077

0.3381

0.3007

13-13.9

1.2041

0.3463

0.299

14-14.9

1.9005

0.3546

0.2974

15-15.9

2.5969

0.3628

0.296

16-16.9

3.2933

0.371

0.2947

Table 5.

LMS

data for

BMDcorr

in fem ales.

<£)

97 5th

50th

25th

o

CO

o

CD

O

'3'

O

6

8

10

12

14

16

A ge (years)

97.5th

50th

2 5th

o

00

o

<D

d

6

10

14

16

Age (years)

Figure 4. aBMD percentiles fo r females.

3.3 Birth Weight, Height, Weight, BMI

B irth w e ig h t did n o t corre la te w ith aBMD o r

in la te r ch ild h o o d [F ig u re 5,

6). As expected, heig h t increased w ith age fo r b o th male and fem ale p a rtic ip a n ts

(Figure 7, 8). O verall the fem ale p a rtic ip a n ts in th e study c o h o rt w e re

s ig n ific a n tly ta lle r than the Iris h reference standard [p = 0 .0 0 0 3 5 ]; th e male

p a rtic ip a n ts w ere also ta lle r b u t th is was n o t sig n ific a n t [p = 1 .7 5 ]. Height, w e ig h t

• Male

* Female

♦

♦

*

• ♦

•

♦

*

•

♦

♦

. •

• •

•

-♦

•

•

♦♦

♦

♦ •

•

«

*

^

♦

♦

•

•

♦

•

•

.

* 1

•

•

«

•

♦

•

•

•

♦

4>

« •

•

•

•

♦

\--- 1---1--- 1--- 1---r ~

2 0

2.5

3 0

3 5

4 0

4

5

Birth weight (kilograms)

in CO o

♦ Male

• Female

Q

2

CO

•

V

•

• • •

<N

O

— I

---

--- 1

--- 1

---

r-2 0

2,5

3.0

3.5

4 0

Birth weight (l<liograms)

4 5

Figure 6. BMDcorr by birth weight by gender.

)0

16

0

17

0

180

!

______ ■ ■

JLX

i

i

.

11

.

-H

e

ig

h

t

(c

m

)

0

11

0

12

0

13

0

14

0

IE

L .

___

J

1

1

I

1

, 1

... !' T ? ! ? ! t I ' ! 1 1 : ■ I ! 1 • ’ 1

6

7

8

9

10

11

12

13

14

15.5

16.5

A ge (years)

Illllll

JlU

il

Age (years)

Figure 8. Height by age in females.

3.4 Pubertal Stage

P u b ertal self-a sse ssm e n t w a s c o m p leted by 27 females w h o w e r e 8 y ea rs or

o ld e r an d by 27 m ales w h o w e r e 9 y e a rs o r o lder a t th e tim e of th e ir DXA

exam ination. P u b ertal stage d ata is su m m a ris e d in A ppendix 6. Male aBMD and

B M D c o r r

re su lts for T a n n e r stages O-I an d 11+ a r e sh o w n b e lo w (Table 6, 7). All

T a n n e r Stage

N=

Mean aBMD

Cl for Mean

P=

(g/cm 2)

M ales >9 y e a rs

0-1

11

0.776

0.6836-0.8677

11+

16

0.853

0.7768-0.9295

0.19

Table 6. aBMD by T a n n e r stage in males. Cl = confidence interval.

T a n n e r Stage

N=

Mean BMDcorr

Cl for Mean

P=

(g/cm 2)

M ales

>9

y e a rs

O-I

11

0,281

0.2605-0.3011

11+

16

0.296

0.2790-0.3127

0.25

Table 7.

by T a n n e r stage in males. Cl = confidence interval.

3.5 Diet

T h e r e w a s no significant association b e tw e e n b re a stfe e d in g an d b one d e n sity in

m ales o r females [Table 8, 9). Dietary intake re su lts are s u m m a ris e d in Figure 9

N

a B M D

9 5 % Cl

P

Females

Breastfed

29

0.829

0.7673-0.8913

Not

25

0.809

0.7451-

0.66

breastfed

0.8737

Males

Breastfed

33

0.287

0.2738-0.2992

Not

14

0.288

0.2686-

0.89

breastfed

0.3076

Table 8. Mean aBMD in breastfed and n o n -b re a stfe d m a les

and fem ales. Cl

=

co n fid e n c e interval.

N

95%

Females

Breastfed

29

0.327

0.312-0.3413

Not

breastfed

25

0.321

0.305-0.3365

0.59

Males

Breastfed

33

0.287

0.2738-0.2992

Not

breastfed

14

0.288

0.2686-0.3076

0.89

n

I

I

I

I

I

■I ■■

■

l | l | III I

I I l l I

B re a s tfe d B re s tfe d > 3

m o n th s

C h e e s e

Milk

Y o g h u rt

Vit D

M ultivitam in

s u p p le m e n t

O % of male respondents

■ % of female respondents

Figure 9. S u m m a ry of dietary intak e re sults in male and female re s p o n d e n ts.

3.6 Exercise

A total of 54 females an d 47 m ales r e tu r n e d exercise data. For m ales of all ages

an d females in th e 6-10 y e a r age range t h e r e w a s no significant difference in

aBMD o r BMDcorr b e tw e e n th o s e w h o exercised less t h a n one h o u r p e r d ay and

th o se w h o exercised m o r e th a n on e h o u r p e r day [Table 1 0 ,1 1 ) . For females

aged 11-16 y e a rs t h e r e w a s a significant inverse re la tio n sh ip b e tw e e n both

aBMD an d BMDcorr an d exercise of one h o u r o r m o re p e r day (p=0.0009 and

Exercise

N=

Mean aBMD

Cl for Mean

P=

(g/cm 2)

All m a les

<1 h o u r/d ay

20

0.786

0.7181-0.8S42

>1 hour/day

27

0.783

0.7244-0.8415

0.94

Table 10. aBMD and e x er c ise in m ales. Cl = c o nfidence

interval.

Exercise

N=

Mean BMDcorr

Cl fo r Mean

P=

(g/cm 2)

All M ales

<1 h o u r/d ay

20

0.285

0.2685-0.3011

>1 hour/day

27

0.289

0.2746-0.3027

0.72

Table 11.

B M D co rr

and ex e r c ise in m ales. CF = confidence

interval.

Exercise

N=

Mean aBMD

(g/cm 2)

Cl for Mean

P=

All fem ales

<1 hou r/d ay

23

0.909

0.8492-0.9694

>1 hour/day

31

0.753

0.7015-0.8051

0.66

6-10.9 y e a rs

<1 hour/day

9

0.768

0.6854-0.8500

>1 hour/day

21

0.729

0.6750-0.7828

0.43

11-16.9 y e a rs

<1 hou r/d ay

14

1

0.9323-1.068

>1 hour/day

10

0.804

0.7240-0.8850

0.0009

Exercise

N=

Mean

Cl for Mean

P=

(g/cm2)

All females

<1 hour/day

23

0.342

0.3270-0.3572

>1 hour/day

31

0.31

0.2974-0.3234

0.0024

6-10.9 years

<1 hour/day

9

0.315

0.2904-0.3387

>1 hour/day

21

0.31

0.2942-0.3259

0.75

11-16.9 years

<1 hour/day

14

0.36

0.3426-0.3771

>1 hour/day

10

0.311

0.2908-0.3316

0.001

Table 13. B M D co rr an d exercise in females. Cl

= confidence interval.

Males c o ntinued to exercise in both th e h o m e an d school e n v iro n m e n ts

t h r o u g h o u t th e age ra n g e studied (Figure 10]. Females, on th e o th e r hand,

te n d e d to re d u c e th e ir daily exercise (b o th at h o m e and at school) after th e age of

12 y e a rs (Figure 11).

Age

Figure 10. Exercise at home and in school by age in males.

O

o

o

O)

o

0 0

o

o

<D

O

kO

o

o

CO

o

CN

O

O

Age

■ Females some exercise at home

□ Females some exercise in school

Figure 11. Exercise at home and in school by age in females.

3.7 Sedentary pastimes

Analysis o f tim e spent in sedentary pastimes revealed that the m ajority of both

males and females watched between one and three hours o f television per day

[Figure 12). Approxim ately h alf o f participants o f both sexes also played up to an

“cr

o

CD

O LO

O

O CO

o

CNJ

o

No TV

0-1 hrs per day

1-3 hrs per day

> 3 h rs p e rd a y

□ % o f male respondents

■ % of female respondents

Figure 12. D aily television w a tc h in g in male and fem ale respondents.

o

CD

O

<=> I

CM

No computer games

0-1 hrs per day

□ % of male respondents

■ % of female respondents

3.8 Cigarette Smoke Exposure

V e ry few respondents re p o rte d being exposed to smoi<e in the hom e and none

re p o rte d exposure to smoke in the car (Figure 14).

<D ["■

Smoker at home in the

Smoker at home outside

house

Smoker in car

□ % of male respondents

■ % o f female respondents

CHAPTER 4

DISCUSSION

Despite ad v a n c e s in QCT technology, DXA re m a in s a very im p o r ta n t tool in the

evaluation of children at risk of low b o n e density, n o t least be c a u se of its

ubiquity. Radiologists and clinicians w h o o v ersee p a e d ia tric DXA sca nning have a

re sp o n sib ility to o ptim ise p a tie n t care by m aking th e re su lts th e y g e n e ra te as

a c c u ra te an d as re le v a n t as possible. The use of DXA in th e p a e d ia tric setting

re q u ir e s a t h o ro u g h k n o w led g e of th e com plexity of b one d e v e lo p m e n t and b one

health. A ttention to detail is re q u ir e d in th e acquisition an d in te rp re ta tio n of DXA

data. P aediatric DXA re c o m m e n d a tio n s clearly a d d r e s s a n u m b e r of key areas,

including th e use of Z-scores r a t h e r th a n T-scores in children, the im p o rta n c e of