0021-972x/96/$03.00/0
Journal
of Clinical
Endocrinology
and Metabolism
Copyright
0 1996 by The Endocrine
Society
Vol. 81, No. 8
Printed
in U.S.A.
The Effect of Long-Term
Growth
Hormone
(GH)
Treatment
on Bone Mineral
Density
in Children
with
GH Deficiency.
Role of GH in the Attainment
of Peak
Bone Mass
GIUSEPPE
SAGGESE,
GIAMPIERO
BARSANTI
IGLI
BARONCELLI,
SILVANO
BERTELLONI,
AND STEFANIA
Endocrine
Pisa, Italy
Unit,
Chair
of Preventive
Pediatrics,
Department
of Pediatrics,
ABSTRACT
The effect of long-term GH treatment on bone mass was examined
in 32 children with GH deficiency (GHD) aged 7.2-16.3 yr by measuring radial (distal third, single-photon absorptiometry) and lumbar
(L2-L4, dual energy x-ray absorptiometry)
bone mineral density
(BMD) (group A). All patients were longitudinally
followed and received recombinant hGH therapy for a mean period of 48.2 + 13.2
months. BMD values were corrected for bone age and expressed as
Z-score in comparison with normative data. In addition, lumbar BMD
and lumbar BMD corrected for the estimated vertebral volumes were
assessed in 11 patients with GHD acred 16.0-18.7 vr at the time thev
reached their f&al height (group B)-and, in 17 subjects with familial
short stature aged 16.4-19.8 yr, as controls (group C) for patients of
group B. Patients of group B had received discontinuous treatment
with pituitary-derived
hGH and subsequently recombinant hGH (total duration of treatment 151.5 2 9.7 months). The off-treatment
period was 4.7 t 2.6 months. Before treatment, patients of group A
C
HILDREN with GH deficiency (GHD) have a reduced
bone mineral density (BMD) partly becauseof delayed
bone maturation (l-3). A reduced BMD alsohasbeenshown in
adult patients with childhood (4-7) or adult-onset GI-JD (8-10)
compared with healthy controls. On the other hand, GH treatment improved BMD in both children (2, 3) and adults with
childhood-onset GHD (6,11,12). Theseresultssuggestthat GH,
in addition to the effect on linear growth and skeletal maturation, is involved in the buildup and probably alsoin the maintenance of bone mass.However, there are no conclusive data
demonstrating that GH treatment determines the attainment of
peak bone mass(PBM) during childhood and upon reaching
early adulthood. Although there is no consensuson the timing
of PBM, defined as the highest level of bone massachieved as
a result of normal growth (13), most of the skeletal mass is
acquired during late adolescenceor young adulthood; the same
is true for cortical and trabecular bone (13-15). The buildup of
bone massis maximal during puberty in both sexes(14, 15);
indeed, approximately 37% of skeletal mass is accumulated
between pubertal stages2 and 5 (13).PBM isimportant because,
University
of Pisa,
showed significantly reduced (P < 0.001) radial and lumbar BMD
(- 1.7 2 0.4 Z-score and - 1.5 ? 0.5 Z-score, respectively) compared
with normative data. During treatment, radial and lumbar BMD
Z-scores improved significantly (P < 0.001); in the patients treated for
the longest time, the BMD was within 0.5 SD of age-matched mean
levels. In patients of group B, lumbar BMD and lumbar BMD corrected for the estimated vertebral volumes were significantly reduced
in comparison with subjects of group C (- 1.2 t 0.4 Z-score and - 1.0 &
0.4 Z-score, P < 0.01 and P < 0.03, respectively).
The results show that children with GHD have reduced BMD.
Optimal GH treatment improves BMD, whereas inappropriate treatment is a main cause of reduced BMD at time of final height. These
findings suggest an important role of GH therapy in the attainment
of peak bone mass in children with GHD. GH treatment should be
continued until the attainment of peak bone mass irrespective of the
height achieved. (J Clin Endocrinol Metab 81: 30773083, 1996)
together with the age-related loss later on, it is a main factor
determining the individual’s bone masslater in life; moreover,
the attainment of PBM seemsto be the best way to prevent the
development of osteoporosisand susceptibility to fractures (13).
In this regard, it has been reported recently that osteoporosis
clearly is more prevalent in adults with GHD (16,17). Thus, the
attainment and the maintenanceof PBM should be considered
primary goals of GH treatment in patients with GHD.
The aim of the study was to investigate the role of GH in
the attainment of PBM. We measured radial and lumbar
BMD in children with GHD before and longitudinally during
long-term GH replacement therapy, comparing the BMD
values with normative data. Furthermore, we assessedlumbar BMD and lumbar BMD corrected for the estimated bone
volumes in another group of GH-treated patients with GHD
at the time they reached their final height; BMD values of
these patients were compared with those we found in a
group of subjects with familial short stature.
Materials
and Methods
Patients
Received December 6, 1995. Revision received February 22, 1996.
Accepted March 1, 1996.
Address all correspondence and requests for reprints to: Giuseppe
Saggese, Endocrine Unit, Chair of Preventive Pediatrics, Department of
Pediatrics, University of Pisa, Via Roma 35, Pisa, Italy I-56125.
A total of 43 Caucasian patients (26 males and 17 females) aged
7.2-18.7 yr with isolated GHD were recruited from our Endocrine Unit
at the Department of Pediatrics of our university. At the start of GH
therapy, all patients were prepubertal and fulfilled the clinical and
diagnostic criteria for GHD: GH peaks less than 10 pg/L after two
3077
SAGGESE
1. Clinical
data
l/M
2/M
3/!?
4lM
5/M
6/M
7/l?
8/F
9/F
10/M
11/F
12/F
13/M
14/M
15/F
16/M
17/F
18/F
19/M
20/M
21/F
22lM
23/M
24lM
26/F
27/F
28/F
29/M
30/M
31/M
32lM
-t
with
GHD
receiving
CA
Case/sex
Mean
in children
SD
CA, chronologic
11.7
rhGH
JCE & M . 1996
Vol81.No8
Controls
provocative
pharmacologic
stimuli (levodopa
and insulin tolerance
test)
and reduced
spontaneous
GH secretion
for 24 h (mean GH concentrations < 3 pg/L)
(18). All patients had normal wt and length at birth, had
normal renal and liver function,
and did not take drugs known
to affect
bone or mineral metabolism.
There was no history of any other chronic
illness or bone disease. Karyotype,
examined
in all girls, was 46,Xx. At
the time of the present study, the patients
were subdivided
into two
groups. Group A consisted of32 children
(19 males and 13 females) aged
7.2-16.3 yr (11.7 t 2.7 vr) (mean 2 SD) receiving
recombinant
human GH
(rhGH)
treatment
(prepubertal
patients 0.6 If/ kg weekly,
pubertal
patients 0.9 W/kg
weekly,
SC at bedtime
6 times a week). The duration
of
rhGH
treatment
ranged
24-79
months
with a mean of 48.2 2 13.2
months. At the time of the study, no patient had attained
final height or
had wrist epiphyses
fused. Six patients (4 males and 2 females) had taken
part in a previous
study investigating
the effect of rhGH treatment
for
12 months
on bone and mineral
metabolism
(3). Clinical
data of the
patients
at the time of the study
are reported
in Table 1. Group
B
consisted of 11 patients (7 males and 4 females) aged 16.0-18.7 yr (17.3 -C
0.9 yr) who discontinued
the treatment
when they attained
their final
height. At the time of the study, the mean off-treatment
period was 4.7 t
2.6 months. In the past, these patients had received discontinuous
treatment with pituitary-derived
hGH (approximately
0.1 IU/kg,
im at bedtime 2-3 times a week) for 43.6 + 9.0 months;
subsequently,
they had
continuously
received rhGH for 107.3 % 4.5 months at the same dose and
regimen
as patients
of group A. The total duration
of GH treatment
ranged from 135-165 months (151.5 r+- 9.7 months).
Clinical data of these
patients
are reported
in Table 2. All pubertal
patients of group A (n =
16, 10 males and 6 females)
and all patients
of group
B developed
spontaneous
pubertal
maturation.
No side effects of the treatment
were observed.
No patient experienced a history of bone fractures
before or during the treatment.
Compliance with the treatment
was good in all patients of groups A and B.
TABLE
ET AL.
In order to reduce the risk of underestimating
BMD in patients
of
group B, we enrolled
as a control group in the study nine males aged
17.5-19.8
yr (18.5 2 0.8 yr) and eight females aged 16.4-18.2
vr (17.2 +
0.6 yr) who attended
our Endocrine
Unit for avhistory
of familial
short
stature (group C) (Table 2). Their final height ranged from the third to
the tenth-percentile;
however,
mean final height w& significantly
higher
in subjects of group C than in patients
of group B (males: mean height
difference
(D) = 7.3 cm, P < 0.001; females: mean height difference
(D) =
10.0 cm, P < 0.001). Body mass index (BMI) did not differbetween
group
B and group C.
Study
design
In patients of group A, radial and lumbar
BMD was measured
longitudinally
during treatment
approximately
every 12 months;
however,
lumbar
BMD was assessed in all but six oatients
at the start and at 12
months
of treatment
and, in all but three patients,
at 24 months
of
treatment.
BMD values were corrected
for bone age and compared
with
appropriate
sex- and age-reference
values by using our own values (19)
and those reported
by Del Rio ef al. (20) for radial and lumbar
BMD,
respectively.
The normative
data of Del Rio et al. were obtained
in a
Spanish population
of children
and adolescents
living approximately
at
the same latitude
as our patients
by using the same machinery
we
employed.
In patients
of group B, lumbar
BMD was measured
only after GH
treatment
was stopped because they had attained
their final height. In
addition,
lumbar
BMD was corrected
for the size of the vertebra
and
compared
with the mean BMD value of subjects of group C. No patient
of groups A and B or subject of group C performed
physical
activity,
with the exception
of daily normal
activities.
Informed
consent to perform
the study wasobtained
from the parents
treatment
of the
study
(group
A)
(yr)
g
(Yd
Height
(cm)
Height
(Z-score)
BMI
rhGH therapy
(months)
7.2
7.6
8.0
8.2
8.5
8.7
8.9
9.0
9.6
10.0
10.0
10.1
10.2
10.6
10.7
11.8
12.0
12.5
13.0
13.0
13.4
13.7
13.8
13.9
14.1
14.3
14.4
15.9
15.10
16.1
16.3
5.2
4.11
6.3
6.2
6.0
6.3
6.0
6.6
7.0
7.11
8.2
8.7
7.11
8.0
9.0
9.1
10.9
10.5
9.8
11.2
11.4
12.2
11.7
11.4
12.2
11.11
11.9
13.8
14.0
14.2
14.0
5.0
5.0
6.10
6.0
5.6
6.0
5.9
6.10
6.10
1.6
8.10
7.10
7.0
8.6
8.3
8.6
10.6
8.10
12.6
12.0
10.6
12.0
11.6
11.6
12.0
10.6
10.0
13.6
13.6
13.6
13.0
109.0
107.0
113.9
116.6
115.1
116.0
113.0
117.3
119.5
125.4
127.5
128.8
125.0
126.5
130.8
132.0
140.0
138.2
144.5
143.6
143.5
148.2
145.1
144.0
151.7
148.0
149.0
158.0
160.2
163.5
162.4
-2.3
-2.9
-1.9
-1.9
-2.4
-2.4
-2.7
-2.2
-2.3
-1.9
-1.4
-1.2
-2.1
-2.0
-1.6
-2.0
-1.2
-2.0
-1.1
-1.3
-2.0
-1.3
-1.7
-1.8
-1.2
-2.0
-1.8
-1.8
-1.5
-1.2
-1.5
16.6
15.9
15.8
16.0
15.9
15.7
14.4
13.9
14.8
15.4
16.1
16.2
16.2
13.9
15.4
15.5
15.3
15.6
16.9
16.6
16.7
15.8
16.6
15.5
16.6
17.8
18.0
18.0
18.5
19.0
18.4
35
38
36
52
43
49
51
50
47
30
33
60
55
48
33
54
42
65
24
32
58
36
36
56
54
44
50
74
65
70
79
9.6 2 2.8
9.3 2 2.7
2 2.7
age; SA, statural
BA
at time
age; BA, bone
age.
134.7
? 16.4
-1.8
2 0.5
16.3
k 1.3
48.2
2 13.2
GROWTH
TABLE
controls
2. Clinical
(group
C)
data
in patients
with
GHD
HORMONE
at time
of their
THERAPY
AND
BONE
final
(group
B) and in subjects
height
Group
CA
(yr)
l/F
2/F
3/F
4/F
5/M
6/M
7/M
8iM
9/M
10/M
11/M
Final
height
(cm)
BMI
-2.6
-3.1
-2.3
-2.9
-2.3
-2.2
-2.5
-2.6
-2.0
-2.3
-2.7
18.8
18.5
18.7
18.6
20.4
21.0
22.0
20.7
20.6
21.2
21.2
Males:
Females:
17.9
16.2
t 0.5
-+ 0.2
158.9
145.6
5 1.4
2 1.5
-2.4
-2.7
i 0.2
5 0.3
Males:
(n = 9)
Females:
(n = 8)
18.5
i
0.8
166.2
t 1.8”
21.3
2 0.6
17.2
k 0.6
155.6
k 2.2h
18.9
t 0.2
CA, chronologic
age; Pd, pituitary-derived.
a P < 0.001 and ’ P < 0.001 in comparison
with
male
PdhGH/rhGH
(months)
and female
with
familial
therapy
short
stature
Total duration
GH therapy
(months)
39 1108
35 I107
50 ill1
31/104
46 I101
49 /103
43 /105
59 /103
55 DO9
52 /113
27 1116
21.0 2 0.5
18.7 k 0.1
Group
3079
as
B
Height
(Z-score)
146.8
143.3
147.2
145.0
159.4
160.0
158.0
157.5
161.1
159.6
156.7
16.0
16.2
16.3
16.5
17.0
17.7
17.9
18.0
18.1
18.5
18.7
DENSITY
46.4
38.8
k 8.81107.2
-+ 7.1l107.5
Off-therapy
(months)
147
142
161
135
147
152
148
162
164
165
143
k 5.2
+ 2.5
154.4
146.3
1
3
2
4
1
5
6
8
6
7
9
t 8.4
i 9.5
6.0 + 2.4
2.5 f 1.1
C
patients,
respectively.
of each subject when the chronological
age of their child was lower than
18 and directly
from each subject whose chronological
age was higher
than 18. The study was approved
by the ethics committee
for human
investigation
of our department.
mineral
content
measurements
for bone size by using the formula:
BMD “OlmnC = bone mineral
content/v01
= BMD X [4/(n
X width]
(24).
Although
the volume
correction
is not anatomic,
BMD,,,,,,,,,
values
provide
an approximation
of the true bone density
(24).
Assessment
Statistical
of anthropometric
findings
Standing
height was measured
with a wall-mounted
stadiometer
by
one of us. To allow comparison
among
different
ages and genders,
height was expressed
as Z-score with respect to height SD according
to
the method
of Tanner et al. (27) by using the formula:
measured
individual
value - mean normal
value for age and gender/so
of normal
mean. Height
was considered
as final adult stature when growth
velocity during
the last year was less than 2 cm and wrist epiphyses
fused
(22). Bone age was evaluated
by using the Greulich
and Pyle method
(23). BMI was calculated
using the formula
wt (kg)/height
(m’).
Assessment
of bone mineral
analysis
The results are expressed
as mean f SD. In patients
of group A, the
values of radial and lumbar
BMD,,,,
are reported
after correction
for
bone age. Comparison
of the data was determined
with
the nonparametric Wilcoxon’s
(Mann-Whitney)
rank-sum
test by using a statistical
system (LabStat.
303, Sibioc, Milan)
adapted
for the IBM personal
computer. Linear regression
analysis by Pearson’s
formula
was performed
to determine
correlation
coefficients.
A P less than 0.05 was considered
significant.
density
Radial BMD (bone mineral
content normalized
for bone width,
expressed as g/cm’)
was measured
by single-photon
absorptiometry
(Norland Corp., model 2783, Atkinson,
WI) at the distal third (33% site) of the
nondominant
forearm,
a site that contains predominantly
cortical bone.
Lumbar
BMD,,,,
(bone mineral content corrected
by the vertebral
surface
area scanned, expressed as g/cm’)
was measured
by postero-anterior
dual
energy x-ray absorptiometry
(DEXA, Lunar DPX-L, Lunar Corp., Madison,
WI) in the lumbar spine at L2-L4 level, a site which provides
a measure of
integral (cortical plus trabecular)
bone. BMD value of each subject represented the means of two scans. The results were calculated
as Z-score by
using the same formula we employed
to calculate height Z-score. In patients
of group A, the correction
for bone age was performed
using bone age
instead of chronologic
age to compute
the Z-score. The coefficient
of precision irr vim was less than 1.5% for single-photon
absorptiometry
and 1.0%
for dual energy x-ray absorptiometry.
To minimize
the effect of bone size on L2-L4 BMD values,
we used
the model for correction
of BMD values for antero-posterior
depth to
obtain bone apparent
volumetric
density
(BMD,,,,,,,,
expressed
as
g/cm”)
proposed
by Kroger et al. (24). The mean volume
of L2-L4 was
approximated
as follows:
vol = m X (widthi2)’
X (A/ width),
where
width = mean width of L2-L4, and A = mean area of L2-L4 (24). Width
and area were provided
by the dual energy x-ray absorptiometry
software program.
The estimated
bone volume
was used to correct bone
BMD values in children
treatment (group A)
Results
with GHD receiving
rhGH
Before therapy, mean radial and lumbar BMD,,,, Z-scores
were significantly
reduced in comparison
with their normal
mean (P < 0.001); the mean reduction
in radial BMD was
slightly
but not significantly
lower
than that of lumbar
BMD,,,,
(-1.7 -C 0.4 Z-score and -1.5 i 0.5 Z-score, P = NS,
respectively)
(Fig. 1). A value of BMD below 2 sd of the
normal mean was found in four patients (two males and two
females) (12.5%) at radial site and in three patients
(two
males and one female) (11.5%) at lumbar site. Anyway,
all
patients had a BMD value at least 1 SD below the normal
mean at both radial and lumbar sites.
During
treatment,
mean radial and lumbar BMD,,,,
improved significantly
and, in the patients treated for the longest time, the BMD was within
0.5 SD of age-matched
mean
levels (Fig. 1). No patient showed a value of BMD below 2
SD of the normal
mean after 24 months and 12 months of
therapy at radial and lumbar site, respectively.
Mean height
SAGGESE
3080
-2,5
(32)
/
I
I
0
12
24
30-42
43-55
rhGH
treatment,
mo
I
I
FIG.
1. Mean radial BMD
Z-score
(Left panel)
receiving
rhGH
treatment
(group A). Between
P < O.Oi and **, P < O.OOl-VS. 6.
12
24
30-42
rhGH
I
-2,5
69-81
/
I
I
0
12
24
I
rhGH
and mean lumbar
brachets
is reported
43-55
treatment,
I
56-68
JCE
ET AL.
56-68
BMD,,,,
Z-score (right panel)
the number
of the examined
69-81
mo
2. Mean
percent
increment
in radial
and lumbar
BMD,,,,
(g/cm’)
and in height
(cm) with respect
to pretreatment
values
in
children
with GHD receiving
rhGH
treatment
(group
A). The mean
increment
in radial and lumbar
BMD,,,,
was significantly
(P < 0.001)
higher
than the mean increment
in height
only after 24 months
of
rhGH
treatment.
The mean increment
in radial
BMD did not differ
(P = NS) in comparison
with that in lumbar
BMD,,,,
at all the
time-points
of measurement.
FIG.
also improved
significantly
(before treatment
-3.2 2 0.7
Z-score, at time of the study -1.8 -C 0.5 Z-score, P < 0.001;
D = 1.4 Z-score). After 24 months of treatment,
the mean
percent increment in radial and lumbar BMD,,,, (g/cm’) was
significantly
higher (P < 0.001) than that in height (cm) with
respect to pretreatment
value
(Fig. 2).
Mean radial and lumbar
BMD,,,,
Z-score did not differ
between males and females either before or during treatment
(data not shown).
A significant
positive
relationship
between
radial BMD
(g/cm’)
and lumbar BMD,,,,
(g/cm’) was found before (r =
0.91, P < 0.001) and during treatment
(r = 0.88, P < 0.001).
Comparison
between BMD in patients with GHD at time of
their final height (group Bj and subjects with familial
short
stature as controls (group Cj
During treatment,
mean height improved
significantly
in
patients of group B (before treatment -3.3 t 0.4 Z-score, final
height -2.5 -C 0.3 Z-score, P < 0.001; D = 0.8 Z-score).
corrected
children.
I
30-42
43-55
treatment,
mo
& M . 1996
Vol81
. No 8
I
56-68
for bone age in children
0, P < 0.001 US. reference
I
69-81
with GHD
values;
*,
Mean values of lumbar
BMD,,,,
(g/cm*)
and lumbar
reduced in male and
BMD vo,umf (g/cm”) were significantly
female patients
of group B in comparison
with those of
male
and female subjects of group C (Fig. 3). Expressed
as
Z-score, mean values of lumbar
BMD,,,,
and lumbar
BMD “,,iulne were -1.2 3- 0.5 and -1 .O ? 0.4 in males and
-1.2 ? 0.3 and -0.9 + 0.2 in females, respectively
(P = NS
between
males and females for both BMD,,,,
and BMDvolume) in comparison
with the mean values of male and
female subjects of group C. Thus, in patients
of group B,
lumbar
BMD was still reduced
after correction
for bone
volume,
even though the degree of reduction
was slightly
less than that indicated
from BMD,,,,
measurement.
In
regard to individual
values, eight (five males and three
females)
(72.7%) and six (four males and two females)
(54.5%) patients of group B had a value at least 1 SD below
the mean of subject of group C for BMD,,,,
and BMD,,,,,,,
respectively;
of these, only one male of group
B had a
value
below
2
SD
(-2.1
SD)
with
respect
to the
BMDarea
mean
BMD,,,,
in males of group C.
Influence
on BMD
of height,
BMI,
and duration
of GH treatment
In patients of group A, a significant
positive relationship
between
radial BMD (g/cm’)
and height (cm) or BMI was
found during
treatment
(r = 0.84, P < 0.001 and r = 0.74,
P < 0.01, respectively).
Similar
results were found for
lumbar BMD,,,,
(r = 0.88, P < 0.001 and r = 0.67, P < 0.01,
respectively).
The changes in radial
or lumbar
BMD,,,,
(g/cm’)
were significantly
correlated
with the duration
of
treatment
(r = 0.76, P < 0.01 and r = 0.69, P < 0.01,
respectively).
In patients of group B, a significant
positive relationship
between lumbar BMD,,,,
(g/cm’)
and height (cm) (r = 0.98,
P < 0.001) or BMI (r = 0.93, P < 0.001) or the duration
of
treatment
(r = 0.67, P < 0.03) was found. There was no
significant
correlation
of lumbar
BMD,,,,,,
(g/ cm3) with
height or BMI (r = 0.23, P = NS and r = 0.18, P = NS,
respectively),
whereas
lumbar
BMD,,,,,,
(g/cm”)
signifi-
GROWTH
I,22
Males
HORMONE
THERAPY
AND BONE DENSITY
Females
“E
2 12
UJ
3081
Males
*
:
.
$j I,18
z
* I,16
ZJ
$114
1 s
4
.
-x-
t .
I
,
4-.
*
I,12
Group 6
FIG.
3. Mean
at the
time
*, P <
0.01.
and individual
of their final
Group C
Group B
i
0,275
.
3
§
0,25
Group C
values of lumbar
BMD area (g/cm’)
(left panel)
height
(group
B) and in subjects
with familial
cantly correlated with the duration of treatment (r = 0.63, P <
0.04).
Discussion
The results demonstrate the strong impact of GHD in
reducing BMD and the effectiveness of an optimal GH treatment in improving BMD in children with GHD.
Before treatment, patients of group A had reduced BMD
at both radial and lumbar sites, suggesting that GHD affects both cortical and trabecular bone. Appendicular cortical bone seemed slightly more sensitive to GHD than
axial integral cortical plus trabecular bone. Indeed, a
greater reduction in peripheral cortical bone than in vertebral trabecular bone, measured by quantitative computed tomography, also has been shown in young adult
patients with childhood-onset GHD (6). The evidence of
reduced baseline or circadian rhythm in serum osteocalcin
(2, 3, 25) and carboxyterminal propeptide of type I procollagen (3, 25, 26) concentrations suggested that a reduced bone formation is likely to be a main cause of
reduced BMD in children with GHD.
During treatment, patients of group A showed a significant increase of BMD Z-score, corrected for bone age
at both radial and lumbar sites; in particular, BMD was
within 0.5 SD of age-matched mean levels in the patients
treated for the longest time. These data suggest that GH
treatment stimulates bone mass accumulation and the progression towards PBM in children with GHD. Anyway, as
we had no untreated GHD controls, the increment in BMD
could also be due to a spontaneous progression towards
PBM. However, the reduced BMD we found in patients of
group B, in whom a suboptimal treatment was probably
performed (see below), seems to confirm the effect of GH
treatment in the attainment of PBM. Furthermore, the increase in radial and lumbar BMD,,,, was greater than
could be accounted for by the degree of increase in height,
suggesting a specific role of GH on the buildup of bone
massindependent of the increase in linear growth. On this
matter, a clear relationship between height gain and bone
mass accumulation has been shown before puberty,
whereas it became weak during puberty (15). These authors (15) also demonstrated that the increment in bone
mass as a function of height gain follows a loop pattern
;
8
.
-+i
Group B
I
Females
Group C
.
-t-
:
-T-
.
*
A
.
Group B
and lumbar
BMD volume (g/cm31 (right panel)
short stature
as controls
(group
C). 0,
A
in patients
.
Group C
with
GHD
P < 0.04; 3, P < 0.02;
when pubertal stages are taken into consideration; the
breaking point approximately corresponds to pubertal
stage 3 and pubertal stages 3-4 in males and females,
respectively. Thus, in the last phase of pubertal development, the increment in bone massis much greater than that
in height (14, 15). Therefore, the dissociation between
height gain and bone mass accumulation we observed
during follow-up may be related, at least in part, to the
spontaneous pubertal development that occurred in half of
our patients.
In patients of group A, the improved BMD at radial and
lumbar sites suggested an effect of GH treatment on both
cortical and trabecular bone. The slightly greater increase in
lumbar than in radial BMD we observed during treatment
probably was related to the much higher turnover of the axial
component of trabecular bone compared with that of appendicular cortical bone. Indeed, the annual turnover rate is
approximately 25% for trabecular bone, whereas it is approximately 23% for cortical bone (27). Other reports (5, 6,
11) seem to confirm this result. However, the age at development of GHD and the method or the site of BMD measurement could influence the response of cortical or trabecular bone to GH treatment.
Reduced BMD,,,, and BMD,,,,,,
were shown in patients
of group B, even though the degree of reduction in BMDwas less than that indicated by measurement of
volume
BM%w Such a result suggested that the reduced lumbar
BMDareawas related, at least in part, to the smaller vertebral
sizes in patients of group B in comparison with those of
subjects of group C. Anyway, the reduced BMD in patients
of group B suggested that GH therapy did not promote an
optimal bone mass accumulation; an inadequate GH treatment during childhood was the most likely causefor reduced
BMD at the time of their final height. This result may be
ascribed to the discontinuous treatment during the first years
of diseasecausedby the limited supplies of pituitary-derived
GH associatedwith the low frequency of GH administration
per week, as previously also suggestedby Kaufman et al. (4)
and De Boer et al. (7). According to this view, the mean height
Z-score of patients of group B was lower than that of patients
of group A, who continuously received GH treatment since
the first years of disease.Thus, these data seem to indicate
that the lack of continuity and the poor frequency of GH
SAGGESE ET AL.
administration may affect the positive effect of GH treatment
on bone mass accumulation in addition to linear growth.
Recent data (13-15) demonstrated that in normal individuals the timing of cessation of longitudinal bone growth
occurs l-7 yr earlier than the cessation of the rapid accumulation of bone mass at various skeletal sites. Thus, the
reduced BMD in patients of group B may be related, at least
in part, to the fact that they had not yet attained their PBM
at the time of discontinuation
of therapy. Furthermore,
a
delayed timing of PBM in patients with GHD, in comparison
with .normal individuals,
could also occur as a result of
growth failure. On the other hand, it must be considered that
BMD still may increase in our patients after the discontinuation of treatment for a persisting effect of previous GH
administration,
as reported by Holmes et al. (28). In the opinion of these authors, the increase in BMD they showed after
the cessation of GH therapy was not attributable to a further
spontaneous attainment of bone mass before PBM was
reached becauseof an increased forearm cortical BMD without a concomitant increased forearm bone width. On the
whole, our results indicate that an optimal replacement therapy since the diagnosis and an appropriate duration of treatment have a main role in the progression towards PBM in
children with GHD.
In adults, it has been demonstrated (29) that fracture risk
is significantly increased if BMD is 1 SD below the agepredicted normal mean value. However, even though many
patients of group A and B showed a BMD value below 1 SD
in comparison with the normal mean, no patient had a history of bone fractures, as observed by Shore et al. (1). Anyhow, a reduced BMD does not necessarily mean a decrease
in bone quality, which is considered an important factor for
fracture risk (30). On the contrary, an increased prevalence
of vertebral osteoporotic fractures has been reported in adult
patients with GHD, with 17% of patients having established
osteoporosis (16). In patients with adult-onset GHD, a higher
total fracture incidence rate (24.1%) in comparison with controls (11.8%) was recently found by Rosen et aI. (17). These
data seemto indicate that GHD probably affects bone quality
to a greater extent in adults than in children. Moreover, a
reduced effect of GH in the maintenance of PBM, combined
with the physiological bone lossthat occurs in adult life, may
be an additional factor in increasing the fracture risk in GHD
adult patients. On the other hand, the lack of history of bone
fractures in our patients may also suggest a positive effect of
GH treatment on bone quality independent of the degree of
reduction in BMD.
In conclusion, the results of our study confirm that children with GHD have a reduced BMD. Optimal GH replacement therapy can improve the BMD in children with GHD;
conversely, inadequate treatment during childhood seemsto
be a main cause of reduced BMD at the time of final height.
These findings suggest that GH treatment has an important
role in bone mass accumulation in children with GHD, so
that the attainment of PBM, as well aslinear growth, should
be considered a goal of treatment. Therefore, GH treatment
should be continued until the attainment of PBM irrespective
of the height achieved. However, we do not know whether
or when the patients with GHD will attain their PBM, as well
JCE & M . 1996
Vol81 . No 8
as the best dosage of GH to achieve it. Further studies are
needed to define these questions.
Acknowledgments
TheauthorsthankMrs. I?.Gerini,nurse
assistance
with
of our Endocrine
Unit,
for her
the study.
References
1. Shore RM, Chewy
RW, Mazess
RB, Rose PG, Bargman
GJ. 1980 Bone
mineral
status in growth
hormone
deficiency.
J Pediatr.
96:393-396.
2. Zamboni
G, Antoniazzi
F, Radetti
G, Musumeci
C, Tatb L. 1991 Effect of two
different
regimens
of recombinant
human
growth
hormone
therapy
on the
bone mineral
density
of patients
with growth
hormone
deficiency.
J l’ediatr.
119x483-485.
3. Saggese G, Baroncelli
GI, Bertelloni
S, Cinquanta
L, Di Nero G. 1993 Effects
of long-term
treatment
with growth hormone
onbone
and mineral
metabolism
in children
with growth
hormone
deficiency.
J Pediatr.
122:37-45.
4. Kaufman
J-M, Taelman
P, Vermeulen
A, Vandeweghe
M. 1992 Bone mineral
status in growth hormone-deficient
males with isolated
and multiple
pituitary
deficiencies
of childhood
onset. J Clin Endocrinol
Metab.
74:118-123.
5. Hyer SL, Rodin
DA, Tobias JH, Leiper
A, Nussey
SS. 1992 Growth
hormone
deficiency
during
puberty
reduces adult bone mineral
density.
Arch Dis Child.
671472-1474.
6. O’Halloran
DJ, Tsatsoulis
A, Whitehouse
RW, Holmes
SJ, Adams JE, Shalet
SM. 1993 Increased
bone density
after recombinant
human
growth
hormone
(GH) therapy
in adults with isolated
GH deficiency.
J Clin Endocrinol
Metab.
76:1344-1348.
7. De Boer H, Blok GJ, Van Lingen
A, Teule GJJ, Lips P, Van Der Veen EA.
1994 Consequences
of childhood-onset
growth
hormone
deficiency
for adult
bone mass. J Bone Miner
Res. 9:1319-1326.
8. Johansson
AG, Burman
P, Westermark
K, Ljunghall
S. 1992 The bone mineral
density
in acquired
growth
hormone
deficiency
correlates
with circulating
levels of insulin-like
growth
factor 1. J Intern Med. 232447-452.
9. Rosen T, Hansson
T, Granhed
H, Szucs J, Bengtsson
B-A. 1993 Reduced
bone
mineral
content
in adult patients
with growth
hormone
deficiency.
Acta Endocrinol
(Copenh).
129:201-206.
10. Holmes
SJ, Economou
G, Whitehouse
RW, Adams
JE, Shaiet
SM. 1994
Reduced
bone mineral
density
in patients
with adult onset growth
hormone
deficiency.
J Clin Endocrinol
Metab.
78:669-674.
11. Degerblad
M, Elgindy
N, Hall K, Sjoberg
HE, Thoren
M. 1992 Potent effect
of recombinant
growth
hormone
on bone mineral
density
and body cornposition in adults with panhypopituitarism.
Acta Endocrinol
(Copenh).
126:387393.
12. Vandeweghe
M, Taelman
I?, Kaufman
JM. 1993 Short- and long-term
effects
of growth
hormone
treatment
on bone turnover
and bone mineral
content
in
adult growth
hormone-deficient
males. Clin Endocrinol
(Oxf). 39~409-415.
13. Matkovic
V, Jelic T, Wardlaw
GM, et al. 1994 Timing of peak bone mass in
Caucasian
females
and its implication
for the prevention
of osteoporosis.
Inference from a cross-sectional
model. J Clin Invest. 93:799-808.
14. Bonjour
J-PH, Theintz
G, Buchs 8, Slosman
D, Rizzoli
R. 1991 Critical
years
and stages of puberty
for spinal and femoral
bone mass accumulation
during
adolescence.
J Clin Endocrinol
Metab.
73:555-563.
15. Theintz
G, Buchs B, Rizzoli
R, et al. 1992 Longitudinal
monitoring
of bone
mass accumulation
in healthy
adolescents:
evidence
for a marked
reduction
after 16 years of age at the levels of lumbar
spine and fermoral
neck in female
subjects. J Clin Endocrinol
Metab.
75:1060-1065.
16. Wuster
C, Slenczka
E, Ziegler
R. 1991 Increased
prevalence
of osteoporosis
and arteriosclerosis
in conventionally
substituted
anterior
pituitary
insufficiency: need for additional
growth
hormone
substitution?
Klin Wochenschr.
691769-773.
17. Rosen T, Wilhelmsen
L, Landin-Wilhelmsen
K, Lappas
G, Lindstedt
G,
Bengtsson
B-A. Increased
fracture
rate in growth
hormone
deficient
adults.
Program
of the 77th Annual
Meeting
of The Endocrine
Society,
Washington
DC, 1995, p 353.
18. Saggese G, Cesaretti
G. 1989 Criteria
for recognition
of the growth-inefficient
child who may respond
to treatment
with growth
hormone.
Am J Dis Child.
143:1287-1293.
19. Saggese G, Federico
G, Ghirri
P, Cipriani
J, Bertelloni
S, Baroncelli
GI. 1986
Bone mineral
content in pediatrics:
normal
values between
2 and 19 years. First
it&an
data. Minerva
Pediatr.
38~545551.
20. Del Rio L, Carrascosa
A, Pans F, Gusinys
M, Yeste D, Domenech
FM. 1994
Bone mineral
density
of the lumbar
spine in white
mediterranean
Spanish
children
and adolescents:
changes related to age, sex, and puberty.
Pediatr
Res.
35:362-366.
21. Tanner
JM, Whitehouse
RH, Takaishi
M. 1966 Standards
from birth to maturity for height,
weight, height velocity
and weight
velocity:
British
children
1965. Arch Dis Child.
41:454-471,
613-634.
22. Ranke
MB. 1995 Growth
hormone
therapy
in children:
when to stop? Horm
Res. 43:122-125.
GROWTH
HORMONE
THERAF’Y
23. Greulich
WW, Pyle SI. 1959 Radiographic
atlas of skeletal
development
of the
hand and wrist. 2nd ed. Stanford,
California:
Stanford
University
Press.
24. Kroger
H, Kotaniemi
A, Vainio
P, Alhava
E. 1992 Bone densitometry
of the
spine and femur in children
by dual-energy
x-ray absorptiometry.
Bone Miner.
17~75-85.
25. Saggese G, Baroncelli
GI, Bertelloni
S, Cinquanta
L, Di Nero G. 1994Twentyfour-hour
osteocalcin,
carboxyterminal
propeptide
of type I procollagen,
and
amino-terminal
propeptide
of type III procollagen
rhythms
in normal
and
growth-retarded
children.
Pediatr
Res. 35:409-415.
26. Carey DE, Goldberg
B, Ratzan
SK, Rubin
KR, Rowe DW. 1984 Radioimmunoassay
for type I procollagen
in growth hormone-deficient
children
before
and during
treatment
with growth
hormone.
Pediatr
Res. 19:8-11.
AND BONE
DENSITY
3083
27. Dempster
DW, Anderson
JJ. 1990 Pathogenesis
of osteoporosis.
Lancet.
341:
797-801.
28. Holmes
SJ, Whitehouse
RW, Economou
G, O’Halloran
DJ, Adams
JE, Shalet
SM. 1995 Further increase
in forearm cortical
bone mineral
content after discontinuation
of growth hormone
replacement.
Clin Endocrinol
(0x0.423-7.
29. Melton
LJ, Atkinson
EJ, O’Fallon
WM, Wahner
HW, Riggs BL. 1993 Longterm fracture
prediction
by bone mineral
assessed at different
skeletal
sites.
J Bone Miner Res. 8:1227-1233.
30. Schonau
E, Schiedermaier
U, Mokow
E, et al. 1996 Development
of muscle
and bone strength
during
childhood
and adolescence.
In: Schonau
E, ed. New
trends
and diagnostic
possibilities
in paediatric
osteology.
Amsterdam:
Elsevier
Scientific;
147-158.