Acta Paediatr 81: 953-8. 1992
Lumbar bone mineral content measured by dual energy X-ray absorptiometry in newborns and infants Bernard L Salle', Pierre Braillon2, Francis H Glorieux3, Jacques Brunet', Eduardo Cavero' and Pierre J Meunier' Department of Neonatology', Edouard Herrior Hospital, Lyon. Department of Rheumatology and Bone Patholog?, INSERM Unit 234, Edouard Herriot Hospital. Lyon. France and Genetics Unit', Shriners Hospital, Departments of Surgery and Pedialrics, McGill University. Montreal. Quebec, Canada
Salle BL, Braillon P, Glorieux FH, Brunet J, Cavero E, Meunier PJ. Lumbar bone mineral content measured by dual energy X-ray absorptiometry in newborns and infants. Acta Pzediatr 1992; 81: 953-8. Stockholm. ISSN 0803-5253 Dual energy X-ray absorptiometry (DXA), a non-invasive method for measuring small amounts of mineral, was used to assess the bone mineral content (BMC) and bone mineral density (BMD) of the lumbar spine (5 vertebrae) in 57 newborns (on day 1-2) and 22 infants (1-24 months of age). A modified high-resolution program (Hologic) allowed us to assess BMC and BMD with a precision higher than 2.4% and 1.5%, respectively. In newborns, BMC and BMD correlated positively with birth weight, body area, length and gestational age: r=0.73,0.71,0.63 and 0.60, respectively, for BMC; and r=0.59, 0.58, 0.54 and 0.53, respectively, for BMD. In infants, both BMC and BMD were highly correlated with weight, age, length and body area over two years (r=0.94 or better in each instance). The data provide normal values for lumbar spine BMC and BMD in newborns (gestational age 31-40) and infants up to two years of age; DXA appears to be an excellent and safe tool for pediatric bone mineral measurements. 0Bone mineral content, bone mineral density, dual energy X-ray absorptiometry, infants, premature newborns
BL Salle, Department of Neonatology, Edouard Herriot Hospital, Place d'Arsonval6943 7, Lyon Cedex 03, France
Single photon densitometry of the radius has been used since 1979 for analysis of preterm cortical bone mineral content (BMC) (1-12); it was a potentially important tool for assessing BMC in preterm infants, because newborns weighing less than 1500 g are at risk of developing osteopenia and rickets. However, recent controversy has arisen on the use of rectilinear or modified linear sources because of the low precision of the method in newborn infants: 4-9% for the radius (3, 5, 7) and 3-4% (8, 11) for the humerus. The growth related changes in BMG and bone mineral density (BMD) of the lumbar spine in children have been studied using the dual X-ray absorptiometry (DXA) approach (13- 15). This technique was recently extended to the study of mineral content in the lumbar spine of newborns. Preliminary results validating the technique have been published elsewhere (16). In short, the precision in this particular application was found to be 2.4% and 1.5% for BMC and BMD measurements, respectively, after repositioning of the subject. The accuracy obtained by comparison of the measured BMC of excised bones and their ash weight (or calcium content), was found to be better than 90% for calcium content as low as 320 mg. Thus it appears that DXA is a non-invasive, rapid, accurate and highly reproducible method of assessment of the spine mineral content.
Furthermore, the very low irradiation ( < 3 mRem) sustained during the examination allows repeat analyses during longitudinal studies. In the present study, we have used DXA to establish normal values of BMC and BMD of the lumbar spine in normal newborns at birth (gestational age (GA) 32 to 40 weeks) and in normal infants aged 1 to 24 months.
Materials and methods Fifty-seven singleton newborns (28 girls and 29 boys) were studied within the first 48 h of life. They were appropriate for GA and born from healthy mothers after an uneventful pregnancy. Birth weight ranged from 1.800 to 4.080 kg, length from 38 to 5 1 cm, body surface area from 0.143 to 0.225 m2 and GA from 32 to 40 weeks. In each instance, the latter was assessed by the date of the last menses and by fetal ultrasonography performed between 10 and 20 weeks' gestation. Twenty-two normal infants (aged 1-24 months) were also studied. Weight varied from 3.880 to 11.700 kg, length from 52 to 89 cm and body area from 0.237 to 0.505 m2. Informed consent was obtained from all parents and
954
BL Salle et al.
ACTA P R D I A T R 81 (1992)
Fig. 1. BMC and BMD measured by the UHR program in the lumbar spine (five vertebrae), gestational age= 34 weeks, body weight =2200 g at one day of age.
the study was approved by the Ethics Committee of the Universitt Claude Bernard, Lyon. BMC and BMD were measured on a Hologic QDR 1000 densitometer (Hologic Inc., Waltham, MA) which uses an X-ray tube as the radiation source. During the analysis, subjects were supine with the legs in spontaneous semi-flexion. Some were sedated (Butobarbital suppository, 15 mg in one single dose). A special program for small bone measurements is available from the manufacturer, but the time necessary to scan the lumbar spine (8-10 min) was considered too long for the baby to remain still during the procedure. For this study, we used an "ultra high resolution" (UHR) version of the standard program for the lumbar spine scanning in adults (16). While this protocol did not alter the image resolution, it used a lower threshold detection and therefore correctly identified bone edges, which appeared as a dotted line surrounding the vertebrae (Fig. 1). Another advantage is the short scanning time (lumbar spine 2-3 min). These programs also calculate the value of ds, a parameter related to soft tissue thickness. All five lumber vertebrae were included in the scan of newborns and infants (Fig. 1). The irradiation dose was approximately 3 mRem, which is
/ oy2
y = 1.736~-0.958 r I 0.71
= 0.60
0 , 30
'
i 32
'
i ' i ' i 34
36
38
s i ' 1 40
0
I
I
I
42
GESTATIONAL AGE (Weeks)
-i
n m-
r-:
3
1
y = 0.682~+0.296 r = 0.73
0
' i ' i ' i ' i ' i ' i
0
40
y = 0.13ox -3.985 r10.63
45 50 LENGTH (cm)
0 55
Fig. 2. Correlation between lumbar spine bone mineral content (BMC) and gestational age, body area, weight or length in 57 normal newborns at birth. The equation of the regression line and its correlation coefficient are indicated. The upper and lower lines encompass the normal range (+2SD).
DXA measurement by BMC in newborns
ACTA P E D I A T R 81 (1992)
about 10% of the exposure for a standard chest X-ray (1 6). Precision was assessed in several instances by repeating the scan two or three times after repositioning of the subject (16). In newborns, the variation of soft tissue thickness in front of the spine had a mean value of 7.10 f0.15 cm. Separate measurements performed on excised bones immersed in a water bath showed a slight decrease of I % in BMC and 0.9% in BMD for every centimeter of increase in water thickness (16). The results were expressed in grams of hydroxyapatite for BMC and in grams of hydroxyapatite/cm2 for BMD. Length was assessed with a stadiometer and weight with an electronic scale. Body area was derived from weight and length measurements, using the normogram of Du Bois & Du Bois (17).
Statistical analyses
Linear correlation coefficients were calculated between BMC and BMD and the various clinical parameters (birth weight, gestational age, postnatal age, body area and length).
Results Newborns
There was no difference in BMC between boys and girls. Figures 2 and 3 show the relationship between BMC and BMD, and gestational age, body area, weight and length. The best correlation was found between BMC and weight (r = 0.73, p < 0.001), followed by body area (r=0.71,p
Lumbar bone mineral content measured by dual energy X-ray absorptiometry in newborns and infants Bernard L Salle', Pierre Braillon2, Francis H Glorieux3, Jacques Brunet', Eduardo Cavero' and Pierre J Meunier' Department of Neonatology', Edouard Herrior Hospital, Lyon. Department of Rheumatology and Bone Patholog?, INSERM Unit 234, Edouard Herriot Hospital. Lyon. France and Genetics Unit', Shriners Hospital, Departments of Surgery and Pedialrics, McGill University. Montreal. Quebec, Canada
Salle BL, Braillon P, Glorieux FH, Brunet J, Cavero E, Meunier PJ. Lumbar bone mineral content measured by dual energy X-ray absorptiometry in newborns and infants. Acta Pzediatr 1992; 81: 953-8. Stockholm. ISSN 0803-5253 Dual energy X-ray absorptiometry (DXA), a non-invasive method for measuring small amounts of mineral, was used to assess the bone mineral content (BMC) and bone mineral density (BMD) of the lumbar spine (5 vertebrae) in 57 newborns (on day 1-2) and 22 infants (1-24 months of age). A modified high-resolution program (Hologic) allowed us to assess BMC and BMD with a precision higher than 2.4% and 1.5%, respectively. In newborns, BMC and BMD correlated positively with birth weight, body area, length and gestational age: r=0.73,0.71,0.63 and 0.60, respectively, for BMC; and r=0.59, 0.58, 0.54 and 0.53, respectively, for BMD. In infants, both BMC and BMD were highly correlated with weight, age, length and body area over two years (r=0.94 or better in each instance). The data provide normal values for lumbar spine BMC and BMD in newborns (gestational age 31-40) and infants up to two years of age; DXA appears to be an excellent and safe tool for pediatric bone mineral measurements. 0Bone mineral content, bone mineral density, dual energy X-ray absorptiometry, infants, premature newborns
BL Salle, Department of Neonatology, Edouard Herriot Hospital, Place d'Arsonval6943 7, Lyon Cedex 03, France
Single photon densitometry of the radius has been used since 1979 for analysis of preterm cortical bone mineral content (BMC) (1-12); it was a potentially important tool for assessing BMC in preterm infants, because newborns weighing less than 1500 g are at risk of developing osteopenia and rickets. However, recent controversy has arisen on the use of rectilinear or modified linear sources because of the low precision of the method in newborn infants: 4-9% for the radius (3, 5, 7) and 3-4% (8, 11) for the humerus. The growth related changes in BMG and bone mineral density (BMD) of the lumbar spine in children have been studied using the dual X-ray absorptiometry (DXA) approach (13- 15). This technique was recently extended to the study of mineral content in the lumbar spine of newborns. Preliminary results validating the technique have been published elsewhere (16). In short, the precision in this particular application was found to be 2.4% and 1.5% for BMC and BMD measurements, respectively, after repositioning of the subject. The accuracy obtained by comparison of the measured BMC of excised bones and their ash weight (or calcium content), was found to be better than 90% for calcium content as low as 320 mg. Thus it appears that DXA is a non-invasive, rapid, accurate and highly reproducible method of assessment of the spine mineral content.
Furthermore, the very low irradiation ( < 3 mRem) sustained during the examination allows repeat analyses during longitudinal studies. In the present study, we have used DXA to establish normal values of BMC and BMD of the lumbar spine in normal newborns at birth (gestational age (GA) 32 to 40 weeks) and in normal infants aged 1 to 24 months.
Materials and methods Fifty-seven singleton newborns (28 girls and 29 boys) were studied within the first 48 h of life. They were appropriate for GA and born from healthy mothers after an uneventful pregnancy. Birth weight ranged from 1.800 to 4.080 kg, length from 38 to 5 1 cm, body surface area from 0.143 to 0.225 m2 and GA from 32 to 40 weeks. In each instance, the latter was assessed by the date of the last menses and by fetal ultrasonography performed between 10 and 20 weeks' gestation. Twenty-two normal infants (aged 1-24 months) were also studied. Weight varied from 3.880 to 11.700 kg, length from 52 to 89 cm and body area from 0.237 to 0.505 m2. Informed consent was obtained from all parents and
954
BL Salle et al.
ACTA P R D I A T R 81 (1992)
Fig. 1. BMC and BMD measured by the UHR program in the lumbar spine (five vertebrae), gestational age= 34 weeks, body weight =2200 g at one day of age.
the study was approved by the Ethics Committee of the Universitt Claude Bernard, Lyon. BMC and BMD were measured on a Hologic QDR 1000 densitometer (Hologic Inc., Waltham, MA) which uses an X-ray tube as the radiation source. During the analysis, subjects were supine with the legs in spontaneous semi-flexion. Some were sedated (Butobarbital suppository, 15 mg in one single dose). A special program for small bone measurements is available from the manufacturer, but the time necessary to scan the lumbar spine (8-10 min) was considered too long for the baby to remain still during the procedure. For this study, we used an "ultra high resolution" (UHR) version of the standard program for the lumbar spine scanning in adults (16). While this protocol did not alter the image resolution, it used a lower threshold detection and therefore correctly identified bone edges, which appeared as a dotted line surrounding the vertebrae (Fig. 1). Another advantage is the short scanning time (lumbar spine 2-3 min). These programs also calculate the value of ds, a parameter related to soft tissue thickness. All five lumber vertebrae were included in the scan of newborns and infants (Fig. 1). The irradiation dose was approximately 3 mRem, which is
/ oy2
y = 1.736~-0.958 r I 0.71
= 0.60
0 , 30
'
i 32
'
i ' i ' i 34
36
38
s i ' 1 40
0
I
I
I
42
GESTATIONAL AGE (Weeks)
-i
n m-
r-:
3
1
y = 0.682~+0.296 r = 0.73
0
' i ' i ' i ' i ' i ' i
0
40
y = 0.13ox -3.985 r10.63
45 50 LENGTH (cm)
0 55
Fig. 2. Correlation between lumbar spine bone mineral content (BMC) and gestational age, body area, weight or length in 57 normal newborns at birth. The equation of the regression line and its correlation coefficient are indicated. The upper and lower lines encompass the normal range (+2SD).
DXA measurement by BMC in newborns
ACTA P E D I A T R 81 (1992)
about 10% of the exposure for a standard chest X-ray (1 6). Precision was assessed in several instances by repeating the scan two or three times after repositioning of the subject (16). In newborns, the variation of soft tissue thickness in front of the spine had a mean value of 7.10 f0.15 cm. Separate measurements performed on excised bones immersed in a water bath showed a slight decrease of I % in BMC and 0.9% in BMD for every centimeter of increase in water thickness (16). The results were expressed in grams of hydroxyapatite for BMC and in grams of hydroxyapatite/cm2 for BMD. Length was assessed with a stadiometer and weight with an electronic scale. Body area was derived from weight and length measurements, using the normogram of Du Bois & Du Bois (17).
Statistical analyses
Linear correlation coefficients were calculated between BMC and BMD and the various clinical parameters (birth weight, gestational age, postnatal age, body area and length).
Results Newborns
There was no difference in BMC between boys and girls. Figures 2 and 3 show the relationship between BMC and BMD, and gestational age, body area, weight and length. The best correlation was found between BMC and weight (r = 0.73, p < 0.001), followed by body area (r=0.71,p
