Download Original Article Osteoporosis

Osteoporos Int (2002) 13:429–433
ß 2002 International Osteoporosis Foundation and National Osteoporosis Foundation
Osteoporosis
International
Original Article
Bone Mineral Density and Metabolism in Familial Dysautonomia
C. Maayan1, E. Bar-On2, A. J. Foldes3, B. Gesundheit1 and R. Dresner Pollak4
Departments of 1Pediatrics, 4Internal Medicine and the 3Jerusalem Osteoporosis Center, Hadassah University Hospital, Mount
Scopus, Hebrew University Hadassah Medical School, Jerusalem, and 2Pediatric Orthopedic Unit, Schneider Children’s Medical
Center of Israel, Rabin Medical Center Petach-Tikva, Israel
Abstract. Familial dysautonomia (FD) patients suffer
from multiple fractures and have reduced bone pain,
which defers the diagnosis. The pathogenesis of bone
fragility in FD is unknown. This study aimed to
characterize bone mineral metabolism and density in
FD. Seventy-nine FD patients aged 8 months to 48 years
(mean age 13.9 10.4 years, median 12.3) were studied.
Clinical data included weight, height, bone age, weekly
physical activity and history of fractures. Bone mineral
density (BMD) of the lumbar spine (n = 43), femoral
neck (n = 26), total hip (n = 22) and whole body (n = 15)
were determined by dual-energy X-ray absorptiometry.
Serum 25-hydroxyvitamin D3, osteocalcin, bone alkaline
phosphatase (B-ALP), parathyroid hormone and urinary
N-telopeptide cross-linked type 1 collagen (NTx) were
determined in 68 patients and age- and sex-matched
controls. Forty-two of 79 patients (53%) sustained 75
fractures. Twenty-four of 43 patients had a spine Z-score
5–2.0, and 13 of 26 had a femoral neck Z-score 5–2.0.
Mean femoral neck BMD Z-score was lower in patients
with fractures compared with those without (–2.5 0.9
vs –1.5 1.0, p = 0.01). Mean body mass index (BMI)
was 16 kg/m2 in prepubertal patients and 18.4 kg/m2 in
postpubertal patients. Bone age was significantly lower
than chronological age (75.5 vs 99.3 months in
prepubertal patients, p50.001; 151 vs 174 in postpubertal patients, p50.05). NTx and osteocalcin levels
were higher in FD patients compared with controls (400
338 vs 303 308, BCE/mM creatinine p50.02; 90 59.5 vs 61.8 36.9 ng/ml, p50.001, respectively). BALP was lower in FD patients compared with controls
Correspondence and offprint requests to: Channa Maayan, MD,
Department of Pediatrics, Hadassah University Hospital, Mount
Scopus, PO Box 24035, Jerusalem 91240, Israel. Tel: +972 2
5844510. Fax: +972 2 6510804.
(44.66 21.8 vs 55.36 36.6 ng/ml, p50.04). Mean
spine Z-score was significantly lower in physically
inactive compared with active patients (–3.00 1.70
vs –1.77 1.3, respectively, p = 0.05). We conclude that
fractures in FD patients are associated with reduced
BMD. FD patients have increased NTx and osteocalcin.
Contributing factors include reduced BMI, failure to
thrive and reduced physical activity. Preventive therapy
and early diagnosis are essential.
Keywords: Bone mineral density; Bone turnover
markers; Familial dysautonomia; Fractures; Osteopenia;
Osteoporosis
Introduction
Familial dysautonomia (FD) is an autosomal recessive
disease affecting the autonomic, sensory and motor
nervous systems. The disease occurs essentially among
Ashkenazi Jews, in whom the incidence is 1:3703 [1].
The FD gene was recently mapped to chromosome 9q31
[2]. The pathogenesis of the disease is unknown.
Neuropathologic findings include a decrease in mainly
unmyelinated neuronal fibers as well as paucity of
neurons in the spinal cord and in sensory and autonomic
ganglia [3]. The clinical manifestations result from
severe autonomic and sensory dysfunction involving
most of the organ systems. There is discoordination of
the gastrointestinal tract with misdirection of food and
fluids into the lungs, gastroesophageal reflux and
vomiting crises resulting in recurrent aspirations and
chronic lung disease. These patients suffer from labile
blood pressure with postural hypotension and dizziness
affecting their gait and stability. The main neurologic
430
deficits include: lack of overflow tears, impaired pain
sensation including bone pain, decreased corneal reflex,
decreased or diminished deep-tendon reflexes, developmental delay, convulsions, and lack of axon flare after
intradermal injection of histamine [4].
FD patients suffer from various skeletal disorders
including severe kyphoscoliosis, foot deformities, arthropathies and multiple fractures [5]. We have
previously reported the incidence of fractures to be
40% of FD patients with a mean of 1.5 fractures per
patient [5]. Since these patients have reduced bone pain,
their fractures are diagnosed at a late stage, thereby
complicating the treatment. The mechanisms of multiple
fractures in FD have not been elucidated.
The aim of the present study was to characterize bone
mineral metabolism and density in FD patients. To our
knowledge this is the first study on this subject in FD.
The results may increase our understanding of recurrent
fractures in FD and possibly in other diseases with
sensory autonomic neuropathies.
C. Maayan et al.
Osteoporosis Center, Jerusalem, using a Hologic QDR
4500 apparatus. The coefficients of variation of our
BMD measurements are 1.2% for the total hip, 1.8% for
the femoral neck and 1.2% for the lumbar spine. In the
adult patients (above 18 years of age) BMD was
measured at the lumbar spine (anteroposterior projection) and the proximal femur (femoral neck and ‘total’
hip regions). In patients under 18 years of age BMD was
measured at the lumbar spine and, in some of them, also
at the proximal femur and/or whole body (see Table 3).
The results of patients’ BMD measurements were
compared with control values matched for age and
gender, and were expressed as units of standard
deviations of the mean control value (Z-score). The
adoption of the Z-score compensated for any inherent
inconsistency derived from the employment of various
DXA models in the study. Control values were derived
from normal databases provided by the manufacturers.
In the case of children and adolescents studied by
Hologic machines, control values for femoral and whole
body BMD were derived from the published literature
[7,8].
Subjects and Methods
The study was carried out on 79 FD patients, consisting
of 42 males and 37 females. All patients had been
followed at the Familial Dysautonomia Center, Hadassah
University Hospital, Mount Scopus, Jerusalem, Israel,
since 1982. The diagnosis of FD was confirmed by FD
signs and symptoms [4], a positive histamine test [4] and
by genetic markers [6]. Patients’ ages ranged from 8
months to 48 years (mean 13.9 10.4, median 12.3
years).
The following clinical data were obtained: age,
weight, height, 50% weight and height ages (the age at
which the patient’s height and weight would equal the
50th percentile), use of medications known to affect
bone mineral metabolism, bone age, past history of
fractures and amount of weight-bearing activity. The
estimated weight-bearing physical activity was graded
according to the number of hours of walking per week;
grade 1, 8–10 h/week; grade 2, 4–7 h/week, grade 3,
below 4 h/week. Body mass index (BMI) was calculated
(kg/m2).
The Ethics Committee of Hadassah University
Hospital approved the study protocol. Informed consent
was obtained from patients or their guardians.
Bone Mineral Density
Bone mineral density (BMD) was measured by dualenergy X-ray absorptiometry (DXA). The measurements
were performed at various diagnostic centers using either
a Hologic apparatus (models QDR 4500 and QDR 1000,
Hologic, Waltham, MA) or Lunar apparatus (models
DPX and DPX-IQ, Lunar, Madison, WI). A third of
BMD measurements were performed at the Hadassah
Biochemical and Endocrine Measurements
Serum measurements included total calcium, phosphorus, total protein, albumin and thyroid stimulating
hormone (TSH). Arterial pH was also determined. 25hydroxyvitamin D3 (25(OH)vitD3) was measured by
RIA (IncStar, Stillwater, MN). Serum intact parathyroid
hormone (PTH) was measured by N-tact PTH SP IRMA
kit (DiaSorin, Stillwater, MN). Serum estradiol level was
previously measured in the female patients included in
this study [9].
Biochemical markers of bone turnover were studied in
68 FD patients and in healthy gender- and age-matched
controls. The serum tests for bone formation included
osteocalcin and bone-specific alkaline phosphatase (BALP). Serum total osteocalcin (OC) was measured with
a human-specific two-site immunoradiometric assay
(IRMA; ELSA-OSTEO, Cis Biointernational, Bagnols,
Ceze, France). The intra- and interassay coefficients of
variation (CVs) were below 4% and 6%, respectively,
and sensitivity was 0.4 ng/ml. Serum B-ALP (ng/ml)
was measured with a human-specific IRMA (Ostease,
Hybritech, San Diego, CA). The sensitivity of the assay
was 0.2% and the intra- and interassay CVs were 7% and
9% respectively.
Urinary N-telopeptide cross-linked collagen type I
(NTx), a marker of bone resorption, was measured in a
second morning void (Osteomark, Ostex International,
Seattle, WA). The intra- and interassay coefficients of
variation were less than 8%, and sensitivity was 25
nmol/l. The measurements were corrected by the urinary
creatinine concentration measured by a standard calorimetric method, and are expressed as bone collagen
equivalent (nmol BCE/mmol creatinine).
BMD and Metabolism in Familial Dysautonomia
431
Data Analysis
Table 2. Distribution of fractures and age of occurrence
Descriptive statistics are presented as mean standard
deviation of the mean (SD), unless otherwise specified.
The values of B-ALP, osteocalcin and NTx in FD
patients and controls were compared using a Wilcoxon
signed rank test. Data were analyzed for prepubertal and
postpubertal patients as well as for the whole group.
Bone age was compared with chronological age using a
paired t-test. FD patients with fractures were compared
with those without fractures by non-paired t-test.
Comparisons between the three groups of patients
divided by degree of physical activity were performed
with two-way ANOVA. p50.05 was defined as
statistically significant. All analyses were performed
using Statistix software package (Analytical Software,
Tallahassee, FL).
Location
Results
Clinical Characteristics
The clinical characteristics of the study population are
presented in Table 1. There were 57 (33 male, 24
female) prepubertal and 22 (9 male, 13 female)
postpubertal patients. The mean age at puberty was
15.5 2.2 years in females and 17.8 2.1 years in
males. Twenty percent of patients were treated with
Maalox (Rhône-Poulenc Rorer) and/or Gaviscon (R &
C). Thirty percent of the patients were treated with
anticonvulsants but none was on idantoin, and only one
patient was on chronic glucocorticoid therapy for
bronchial asthma. FD patients were small and short for
their age. The mean age compatible with their height and
weight at the 50th percentile was only 8 years. Mean
BMI in prepubertal patients was 16 kg/m2 and in
postpubertal patients was 18.4 kg/m2 (p = 0.001). Mean
BMI in prepubertal patients corresponded to the 10th
percentile for height and weight at their mean age of
13.9 years. Mean bone age determined in 25 patients was
76 months compared with 99 months of chronological
age (paired t-test, p50.001) in 20 prepubertal patients,
and 151 months compared with 174 months in 5
postpubertal patients (paired t-test, p50.05). There
were no gender differences with respect to mean
chronological age, BMI, or mean age compatible with
height and weight at 50th percentile. Mean BMI was
Table 1. Clinical data of 79 FD patients
Male/female
Age (years)
Weight (kg)
Height (m)
BMI (kg/m2)
% patients with one fracture
% patients with two fractures
% patients with three or more fractures
Weight-bearing physical activity (walking h/week)
Data are mean SD.
42/37
13.9 28.0 1.26 16.6 26.6
13.9
12.5
8.9 10.4
13.3
0.27
2.6
5.0
Radius and/or ulna
Tibia and/or fibula
Humerus
Calcaneus and/or metatarsus
Femur
Metacarpal and/or fingers
Clavicle
Patella
Mandible
No. of
fractures
23
16
13
7
7
5
2
1
1
Age (years)
Mean SD
Median
8.4
10.8
5.3
6.6
17.1
6.2
8.5
22
10
9
8.4
5
3
18
7
4.7
9.8
2.6
5.7
9.7
3.2
significantly lower in prepubertal patients (16 2.2 kg/
m2 vs 18.42.7 kg/m2, p50.01).
Forty-two of 79 patients (53%) had a total of 75
fractures (Table 1). The distribution of fractures is
presented in Table 2. Over 50% of fractures (39/75)
occurred at the radius, ulna, tibia and fibula. Fractures
occurred mostly in the upper limbs before puberty and in
the lower limbs after puberty (Table 2). The mean age of
occurrence of all fractures was 9.1 7.1 years with a
median age of 8 years. There was no difference in mean
fracture number with respect to gender or pubertal stage.
There were no significant differences in mean BMI
between pre- and postpubertal FD patients who sustained
a fracture and those who did not. Twenty-eight patients
were physically active at grade 1, 36 patients at grade 2
and 15 patients at grade 3. There were no statistically
significant differences between these three groups with
respect to gender, mean age, pubertal stage, BMI and
number of fractures.
Bone Mineral Density
Lumbar spine BMD measurements with appropriate Zscores were obtained in 43 randomly selected patients
(30 prepubertal, mean age 7.9 4 years; 13 postpubertal,
mean age 29.6 10.6 years). Z-scores for the femoral
neck, total hip and whole body BMD measurements
were available for 26 (14 prepubertal, mean age 9.43.2
years; and 12 postpubertal, mean age 30.1 10.9 years),
22 (13 prepubertal, mean age 9.4 3.4 years; 9
postpubertal, mean age 30.1 9.9 years) and 15
prepubertal patients, respectively.
Mean Z-scores for the lumbar spine, femoral neck,
total hip and whole body are presented in Table 3. In the
first three sites more than 90% of patients studied had a
negative Z-score, implying a BMD value that is below
the expected value of healthy subjects matched for age
and gender. Furthermore, at least 50% of the patients
studied had a Z-score 5–2.0, indicating a marked
reduction in BMD, equivalent to the 3rd percentile or
less of the age- and gender-matched expected value.
Mean whole body Z-score was reduced to a lesser extent,
with 33% of the patients having a Z-score 5–2.0. There
432
C. Maayan et al.
Table 3. BMD Z-Scores at the various measured sites
n
Lumbar spine
43
Femoral neck
26
Total proximal femur 22
Whole body
15
Z-score
(mean SD)
Patients
with
Z-score
50 (%)
Patients
with
Z-score
4–2.0 (%)
–2.0
–2.1
–2.2
–1.2
91
96
100
80
53
54
50
33
1.43
1.05
1.06
1.3
were no gender differences in mean BMD Z-scores of
the lumbar spine, total hip, femoral neck and total body.
Pre-pubertal patients had a significantly lower spinal
Z-score compared with postpubertal patients (–2.4 1.2
vs –1.2 1.6, p = 0.008). No such difference was found
at the femoral neck or total hip. Mean femoral neck
BMD Z-score was significantly lower in patients with
fractures compared with those without (–2.5 0.9 vs
71.5 1.0, p = 0.01). There were no significant
differences in spine, total hip, and whole body BMD Zscores between these two groups.
Mean spine Z-score was significantly lower in
physically inactive patients (grade 3) compared with
relatively active patients (grade 1) (–3.0 1.8 vs –1.8 1.3, respectively, p = 0.05).
Biochemical Measurements
Mean serum levels of calcium, phosphorus, TSH,
25(OH)vitD3, PTH, total protein, albumin and arterial
pH in FD patients were within normal limits (Table 4).
Nine percent of the patients had vitamin D deficiency
defined as 25(OH)vitD3 510 ng/ml.
Mean serum levels of B-ALP, osteocalcin and urinary
NTx in FD patients and controls are presented in Table
4. Serum osteocalcin and urinary NTx levels were
significantly higher in FD patients compared with ageand sex-matched controls. On the other hand, B-ALP
Table 4. Indices of bone mineral metabolism in FD patients and in
age- and sex-matched controls
Osteocalcin (ng/ml)
B-ALP (ng/ml)
NTx (nmol BCE/mmol Cr)
Total protein (g/l)
Albumin (g/l)
Ca (mM/l)
P (mM/l)
PTH (pmol/l)
25(OH)vitD3 (ng/ml)
Arterial pH
n
FD patients
68
68
68
67
66
68
68
54
56
68
90
44.7
400
73.2
40.6
2.4
1.4
5.1
25.4
7.4
Normal
range
59.5**
–
21.9*
–
338*
–
7.4
60–80
4.2
35–50
0.1
2.0–2.6
0.3
0.8–1.4
5.6
1.2–6.8
13
10–45
0.03
7.35–7.45
Controls
61.8 36.9
55.4 36.6
303 308
–
–
–
–
–
–
–
Data are mean SD.
*p50.05 FD patients versus controls; **p50.01 FD patients versus
controls.
levels were significantly lower in FD patients compared
with age- and gender-matched controls.
Serum osteocalcin was significantly higher in prepubertal FD patients who had sustained fractures
compared with those without a previous fracture (112
50 vs 84 32 ng/ml, p = 0.01). There were no
significant differences between patients with fractures
and those without in mean 25(OH)vitD3, PTH, B-ALP
and NTx. There were no gender differences in all
biochemical measurements except for serum osteocalcin
level which was significantly lower in females compared
with males (71.5 35.6 ng/ml vs 99.2 33.8 ng/ml,
p = 0.02 in females and males respectively).
Discussion
The unselected FD group of patients studied had a high
prevalence of fractures (450%) and significantly
reduced BMD. Markers of bone formation and resorption, osteocalcin and urinary NTx, were higher in FD
patients compared with age- and sex-matched controls.
Since FD patients have a higher threshold of bone pain,
the diagnosis of a fracture in these patients may be
delayed for months. Furthermore, osteoporosis may play
a role in the pathogenesis of their spinal deformities, and
has been shown to impair the fixation of spinal fusion
operations often performed in these patients [5].
Several factors can contribute to decreased BMD and
increased fracture incidence in FD. These patients
consume decreased amounts of dairy products due to
recurrent aspirations and the possible deleterious effect
of dairy products, which can aggravate bronchial
hypersensitivity and increase the risk of aspiration.
Phosphorus absorption might be impaired in these
patients by chronic antacid therapy, such as aluminum
hydroxide, as it combines to form insoluble compounds
of aluminum phosphate. Most FD patients consume
antacids regularly for the management of their gastroesophageal reflux. Even patients who have undergone
fundoplication and gastrostomy operations consume
antacids during FD crises or acute illness. Interestingly,
vitamin D deficiency was uncommon in our patients,
possibly due to high exposure to sun in Israel.
FD patients in our study had a significantly low BMI,
a known risk factor for decreased BMD. Low BMI in FD
patients is probably a marker of their chronic illness, and
results from decreased caloric intake and frequent
dysautonomic crises characterized by, among other
symptoms, nausea and vomiting [4]. Other indicators
of failure to thrive and chronic illness in our patients
include reduced bone age compared with their chronological age and the discrepancy between their mean
chronological age and the calculated mean age
compatible with their height and weight at the 50th
percentile. FD patients were shorter for their age. It is
noteworthy that the short stature of our patients may
induce a downward skew in their BMD readings. While
protein malnutrition was previously reported as a major
risk factor for osteoporosis in the pediatric age group
BMD and Metabolism in Familial Dysautonomia
[10], there was no evidence of protein malnutrition in
our patients, suggested by their normal serum albumin
level.
Weight-bearing physical activity plays a major role in
bone acquisition and the maintenance of bone mass.
Impaired mobility, hypotonia and decreased weightbearing are believed to account for increased osteoporosis risk in central nervous system and neuromuscular
diseases [11]. The mean estimated number of hours of
walking per week in our FD patients was only 8.9 5.
Furthermore, FD patients in our study who were the least
active had the lowest spine BMD Z-scores compared
with the more active group. Moreover, FD patients suffer
from ataxia, gait instability and are prone to recurrent
falls, further increasing their fracture risk. Thus, fracture
risk in FD patients increases not only due to decreased
bone strength but also due to their increased propensity
for falling.
Skeletal mineralization is affected by normal sexual
maturation in both females and males. We have
previously shown that menarch in FD females is delayed
(average age 15.5 years) compared with their unaffected
mothers (average age 13.6 years), although the sex
hormone profile was normal [9]. Delayed puberty with a
lower estrogen exposure score was previously shown to
be associated with decreased bone mass in adolescent
girls [12].
We speculate that extreme blood pressure fluctuations,
a typical symptom in FD patients, might affect bone
perfusion and may also contribute to their bone disease.
Bone disease due to vascular changes was previously
described in FD [5,13]. Significant alterations of blood
pressure can affect bone perfusion, and may account for
the clinical presentations of Legg-Calve-Perthes disease
in two of our patients and navicular bone necrosis in
another one. These patients had no other apparent cause
for their vascular bone disease. Thus, the role of bone
perfusion in the pathogenesis of bone fragility in FD
needs to be studied further.
We have previously demonstrated reduced plasma
calcitonin gene-related peptide (CGRP) in FD [13].
CGRP is a prominent neurotransmitter in C-fibers, which
are involved in thermo-, mechano- and nociception
perception, and FD patients were found to have reduced
C-fibers [3]. CGRP was shown to inhibit osteoclastic
bone resorption in vitro [14]. Furthermore, CGRP was
also found to have effects on osteoblasts, increasing
mineral formation [15]. We hypothesize that decreased
local CGRP levels in FD can explain their lack of bone
pain and adversely affect bone turnover.
In this study increased levels of osteocalcin and NTx
were found in FD patients compared with age- and sexmatched controls. B-ALP was lower in FD patients
compared with controls. Since patients and controls were
not matched for their pubertal stage, their higher bone
markers could be partially explained by delayed puberty.
In addition, low body weight accompanied by a low
muscular mass can result in a low urinary creatinine, and
433
thus artificially increased values of urinary creatininecorrected NTx [16]. In addition, the limitations of the
clinical utility of bone markers in the pediatric age group
should be taken into account.
We conclude that FD patients are among the pediatric
groups at risk for osteopenia and osteoporosis. Since the
time of diagnosis can be delayed due to lack of bone
pain, there should be awareness of falls and limb
swelling because of a possible fracture. Our findings in
FD patients may have clinical and therapeutic implications. Calcium, vitamin D supplement and a high-calorie
diet should be provided. Elevated bone markers
suggested increased bone turnover. Thus, antiresorptive
agents, such as the bisphosphonates, may be effective in
fracture prevention in these patients.
References
1. Maayan C, Kaplan E, Shachar S, Peleg O, Godfrey S. Incidence
of familial dysautonomia in Israel 1977–1981. Clin Genet
1987;32:106–8.
2. Slaugenhaupt SA, Blumenfeld A, Gill SP, et al. Tissue specific
expression of a splicing mutation in the IKBKAP gene causes
familial dysautonomia. Am J Hum Genet 2001;68:598–605.
3. Pearson J, Dancis J, Axelrod F, Grover N. The sural nerve in
familial dysautonomia. J Neuropathol Exp Neurol 1975;34:413–
24.
4. Axelrod FB. Familial dysautonomia. In: Roberston D, Biaggion I,
editors. Disorders of the autonomic nervous system, vol 5.
Newark, NJ: Harwood Academic Publishers, 1995:217–31.
5. Bar-On E, Floman Y, Sagiv S, et al. Orthopedic manifestations of
familial dysautonomia: a review of one hundred and thirty-six
patients. J Bone Joint Surg AM 2000;82:1563–70.
6. Blumenfeld A, Slaugenhaupt SA, Axelrod FB, et al. Localization
of the gene for familial dysautonomia on chromosome 9 and
definition of DNA markers for genetic diagnosis. Nature Genet
1993;4:160–4.
7. Faulkner RA, Bailey DA, Drinkwater TD, McKay HA, Arnold C,
Wilkinson AA. Bone densitometry in Canadian children 8–17
years of age. Calcif Tissue Int 1996;59:344–51.
8. Taylor A, Konrad PT, Norman ME, Harcke TH. Total body
mineral density in young children: influence of head bone mineral
density. J Bone Miner Res 1997;12:652–5.
9. Maayan C, Sela O, Axelrod F, Kidron D, Hochner-Celnikier D.
Gynecological aspects of female familial dysautonomia. Isr Med
Assoc J 2000;2:679–83.
10. Branca F, Robins SP, Ferro-Luzzi A, Golden MHN. Bone
turnover in malnourished children. Lancet 1992;340:1493–6.
11. Steelman J, Zeitler P. Osteoporosis in pediatrics. Pediatr Rev
2001;22:56–64.
12. Mitnick JS, Axelrod FB, Genieser NB, Becker M. Aseptic
necrosis in familial dysautonomia. Radiology 1982;142:89–91.
13. Maayan C, Becker Y, Gesundheit B, Girgis SI. Calcitonin gene
related peptide in familial dysautonomia. Neuropeptides
2001;35:189–95.
14. Zaidi M, Fuller K, Bevis PJR, et al. Calcitonin gene related
peptide inhibits osteoclastic bone resorption: a comparative study.
Calcif Tissue Int 1987;40:149–54.
15. Ballica R, Valentijn K, Khachatryan A, et al. Targeted expression
of calcitonin gene-related peptide osteoblasts increases bone
density in mice. J Bone Miner Res 1999;14:1067–74.
16. Szuec P, Seeman E, Delmas PD. Biochemical measurements of
bone turnover in children and adolescents. Osteoporos Int
2000;11:281–94.
Received for publication 21 May 2001
Accepted in revised form 27 November 2001