- Email: [email protected]
Improving reporting of bone densitometry data in children—The Manchester experience
ABSTRACTS / Bone 40 (2007) S22–S89
b 5 ng/ml (n = 1; 3%), moderate deficiency 5–15 ng/ml, (n = 10; 30%), insufficiency 16–30 ng/ml (n = 12;40%), normal N30 ng/ ml, (n = 6; 20%). 1,25-OH2D: 34 pg/ml (range 4–67). Two patients (7%) were 1,25-OH2D deficient. No relationship between 25OHD, Ca++, Mg, CCr or years since Tx was observed. PTH: 58.8 pg/ml (range 12.4–255), 33% with hyperPTH. 25OHD correlated inversely with PTH (r = − 0.42, p = 0.03). In 21 out of 24 children with low 25OHD, who received dietary counselling and Vitamin D supplements, plasma levels increased from 13.2 (10.5–19.5) mg (25–75th) to 24.2 (17.1–35.9) ng/ml (p = 0.007), 7.7 ± 2.4 months later, whereas PTH decreased from 64.7 (43.8–99.6) to 51.9 (30.3– 96.2) pg/ml (p = 0.03). Correlation between 25OHD and PTH levels was not further detected. Only 2/21 children remained with moderate 25OHD deficiency, or 8/21 insufficiency, but 11/ 21 reached normal levels. Comments: The high prevalence of 25OHD deficiency in this series suggests Vitamin D should be carefully monitored in pediatric kidney Tx. Long-term relationships with excess cardiovascular risk and increased adverse immune function should be avoided through Vitamin D supplementation to this vulnerable population.
S23
used to test for gender and PS interactions, and the effect of FL, upon each measurement. The independent samples t-test was used to test for gender differences in unadjusted and adjusted data at each PS. Results: There was a significant interaction between gender and PS on all measurements (p b 0.001–0.003). Males had greater FL, BA, CA, SSI and CSMA at all PS (p b 0.001– 0.046), greater CT at PS4 (p = 0.012), greater BMC at PS2-5 (p b 0.001–0.041) and MA at PS1 and PS3–5 (p b 0.001– 0.013), in comparison to females. There were significant effects of FL upon all measured parameters (p b 0.001–0.017). After adjusting for FL, BA and MA were larger in males at PS1 and PS3–5 (p b 0.001–0.016) and CA, CC, SSI, CSMA were greater males, compared to females, at PS3–5 (p b 0.001–0.016). CT was greater in females at PS1, in comparison to males (p = 0.005). Conclusion: These data confirm there is a gender-specific pattern of development in bone geometry, strength and muscle of the midshaft radius and forearm during pubertal development. However, after taking into account body size, most parameters tend to be greater in males between mid to late puberty; it is feasible this may represent an adaptive response to greater muscle forces in males.
doi:10.1016/j.bone.2007.04.003 doi:10.1016/j.bone.2007.04.004
Gender differences in cortical bone geometry and strength of the midshaft radius in healthy children during pubertal development after adjustment for body size R.L. Ashby 1, M.Z. Mughal 2, J.E. Adams 1, K.A. Ward 1 1 Imaging Science and Biomedical Engineering, The University of Manchester, Manchester, United Kingdom 2 Saint Mary’s Hospital for Women and Children, Manchester, United Kingdom
Improving reporting of bone densitometry data in children—The Manchester experience R.L. Ashby 1, M.Z. Mughal 2, K.A. Ward 1, J.E. Adams 1 1 Imaging Science and Biomedical Engineering, The University of Manchester, Manchester, United Kingdom 2 Saint Mary’s Hospital for Women and Children, Manchester, United Kingdom
Background: Gender differences in bone size during growth are well studied. However, there are a lack of data describing gender differences in cortical bone geometry and strength during pubertal development. In addition, few studies have investigated gender differences in cortical parameters after adjustment for body size. Aim: The aim of this study was to investigate gender differences in cortical bone geometry, strength and crosssectional muscle area (CSMA) of the midshaft radius and forearm in males and females during pubertal development, and to assess gender differences after adjusting for forearm length (FL) at each pubertal stage (PS). Participants and methods: A total of 369 (198 M, age 12.3 ± 3.3, range 5–19 years) healthy children were studied. FL (mm) was measured. Each participant’s PS was assigned a grade of 1–5 according to the Tanner (1969) criteria. The Stratec XCT-2000 pQCT scanner was used to measure bone area (BA), cortical area (CA), medullary area (MA) and CSMA (mm2), cortical bone mineral content (BMC, mg/mm) and the strength strain index (SSI, mm3) of the 50% midshaft radius diaphysis (midshaft radius). Two-way ANOVA was
The reporting of bone densitometry data in children is currently problematic. There is a paucity of appropriate reference data, and some manufacturers do not provide any method of size-adjustment for DXA measurements. In light of these inadequacies, our aim was to improve the reporting and interpretation of paediatric bone densitometry reports in children. A total of 435 (235 male; age 12.0 ± 3.1, range 5–18 years) healthy Caucasian children from Greater Manchester (United Kingdom [UK]) were studied. The Hologic QDR Discovery DXA scanner was used to provide measurements of the lumbar spine (LS) and left femoral neck (FN); bone mineral apparent density (BMAD, g/cm3) was calculated. The Stratec XCT-2000 pQCT scanner measured total and trabecular volumetric BMD (vBMD, mg/mm3) of the 4% distal radius. The LMS method (Cole and Green, 1992) was used to generate gender-specific centile curves for LS and FN BMAD, total and trabecular vBMD at each yearly age group. Centile curves for LS BA for height and LS BMC for BA (Molgaard et al., 1997) were produced for each gender. A report is shown in example 1. The automatically generated report shows actual and z-scores (z); the child’s results are also
S24
ABSTRACTS / Bone 40 (2007) S22–S89
plotted on centile charts. These provide clinicians with an informative assimilation of paediatric bone densitometry and anthropometric data, which is easy to interpret. We are in the process of implementing data for assessing bone mineral content in relation to lean muscle mass in our reports, to further improve the diagnosis of bone diseases in children. Example 1: Bone densitometry report for a male of age 15.7 Height (176.5 cm) (z = 0.6), weight (69.8 kg) (z = 1.0), BMI (22.4 kg/m2) (z = 0.1)1 pQCT trabecular vBMD 84.8 mg/cm3 Reduced (z = − 6.2) pQCT total vBMD 212.0 mg/cm3 Reduced (z = − 3.2) DXA Lumbar spine BMAD 0.153 g/cm3 Reduced (z = − 3.5) DXA left femoral neck BMAD 0.236 g/cm3 Reduced (z = − 2.2)2 Molgaard (1997) calculations for LS DXA: Are bones short for age? NO (z = 0.6); Are bones narrow for height? NO (z = 0.3); Are bones undermineralised for their size? YES (z = − 2.2) Conclusion: Osteopaenia confirmed by pQCT, DXA femoral neck and lumbar spine. Based on Molgaard calculations, bones are normal for age and height, but undermineralised. 1 Calculated from (Cole et al., 1990; Freeman et al., 1990). 2 Calculated from Manchester normative reference data (Ward et al., 2007). doi:10.1016/j.bone.2007.04.005
The effect of socioeconomic status upon bone geometry and bone mineral density at different skeletal sites in healthy children R.L. Ashby 1, S.A. Roberts 2, M.Z. Mughal 3, J.E. Adams 1, K.A. Ward 1 1 Imaging Science and Biomedical Engineering, The University of Manchester, Manchester, United Kingdom 2 Medical Biostatistics, The University of Manchester, Manchester, United Kingdom 3 Saint Mary’s Hospital for Women and Children, Manchester, United Kingdom Background: Studies indicate children from socially deprived backgrounds have higher fracture rates in comparison to those from more affluent areas (Stark et al., 2002). However, the effect of socioeconomic status upon bone geometry and bone mineral density (BMD) is not well studied (Clark et al., 2005). Aim: The aim of this study was to assess the effect of socioeconomic status upon bone parameters of the whole body (WB), lumbar spine (LS), femoral neck (FN), distal and midshaft radius in a cohort of healthy children. Participants and methods: A total of 435 (235 male; age 12.0 ± 3.1, range 5–18 years) healthy Caucasian children from the Greater Manchester region (United Kingdom [UK]) were studied. Height (cm) and weight (kg) were measured; body
mass index (BMI [kg/m2]) was calculated. The Hologic QDR Discovery DXA scanner measured WB, LS and FN bone area (BA, cm2) and bone mineral content (BMC, g). Bone mineral apparent density (BMAD, g/cm3) of the LS and FN was calculated. The Stratec XCT-2000 pQCT scanner measured total and trabecular volumetric BMD (vBMD, mg/mm3) and BA (mm2) of the 4% distal radius and BA, cortical area, medullary area, cross-sectional muscle area (mm2), cortical BMC (mg/ mm), cortical vBMD and cortical thickness (mm) of the 50% midshaft radius. Standard deviation scores (SDS) for anthropometric measurements and all bone parameters were calculated. Participants postcodes were used to determine the Townsend Deprivation Score (TDS) as an indicator of socioeconomic status, using data from the UK 2001 Census. Results: The TDS was determined for the locality of 382 participants. The median TDS was − 0.38, close to the population average (interquartile range − 2.5 to 2.6). Fortythree percent of participants had TDS N 0; the distribution was skewed towards higher deprivation (more deprived) scores (range − 4.8 to 13.9). There was a significant positive association between the TDS and weight (r = 0.11, p = 0.036) and BMI (r = 0.10, p = 0.056) SDS. Distal radius BA increased with the TDS (r = 0.21, p b 0.001), as did trabecular vBMD (r = 0.09, p = 0.089), while total vBMD decreased (r = − 0.17, p b 0.001). There were no significant associations for any parameters at the WB, LS, FN or midshaft radius. Conclusion: These data indicate children from socially deprived backgrounds have greater BA and trabecular vBMD, but reduced total vBMD at the distal radius; this may be due to greater body weight and BMI. However, we cannot ascertain from these data whether the TDS and associated lifestyle factors also have an effect; this requires further investigation. doi:10.1016/j.bone.2007.04.006
Influence of body composition and weight-bearing physical activity in BMD of pre-pubertal children F. Baptista, I. Fragoso, F. Vieira, C. Barrigas, L. Soutello, P.J. Teixeira, H. Santa-Clara, P. Mil Homens, M.J. Valamatos, L.B. Sardinha Exercise and Health, Faculty of Human Movement, Technical University of Lisbon, Lisbon, Portugal The main purpose of this study was to analyze the influence of mechanical factors such as body composition and physical activity on bone mineral density (BMD) of pre-pubertal girls and boys, accounting for bone age, a common confounding variable. Of the 141 starting participants 117 completed all assessments and were included in this study, namely 53 girls (G) (chronological age, 8.5 ± 0.5 years; bone age 8.5 ± 1.3 years; height,131.8 ± 5.3 cm; weight, 30.2 ± 6.5 kg) and 64 boys (B) (chronological age, 8.5 ± 0.5 years; bone age, 8.9 ± 1.1 years; height,133.4 ± 7.0 cm; weight, 32.3 ± 7.3 kg). Evaluation of body weight, total lean and fat mass, and BMD of lumbar spine (L1–L4), femoral neck, and radius were performed with DXA.
b 5 ng/ml (n = 1; 3%), moderate deficiency 5–15 ng/ml, (n = 10; 30%), insufficiency 16–30 ng/ml (n = 12;40%), normal N30 ng/ ml, (n = 6; 20%). 1,25-OH2D: 34 pg/ml (range 4–67). Two patients (7%) were 1,25-OH2D deficient. No relationship between 25OHD, Ca++, Mg, CCr or years since Tx was observed. PTH: 58.8 pg/ml (range 12.4–255), 33% with hyperPTH. 25OHD correlated inversely with PTH (r = − 0.42, p = 0.03). In 21 out of 24 children with low 25OHD, who received dietary counselling and Vitamin D supplements, plasma levels increased from 13.2 (10.5–19.5) mg (25–75th) to 24.2 (17.1–35.9) ng/ml (p = 0.007), 7.7 ± 2.4 months later, whereas PTH decreased from 64.7 (43.8–99.6) to 51.9 (30.3– 96.2) pg/ml (p = 0.03). Correlation between 25OHD and PTH levels was not further detected. Only 2/21 children remained with moderate 25OHD deficiency, or 8/21 insufficiency, but 11/ 21 reached normal levels. Comments: The high prevalence of 25OHD deficiency in this series suggests Vitamin D should be carefully monitored in pediatric kidney Tx. Long-term relationships with excess cardiovascular risk and increased adverse immune function should be avoided through Vitamin D supplementation to this vulnerable population.
S23
used to test for gender and PS interactions, and the effect of FL, upon each measurement. The independent samples t-test was used to test for gender differences in unadjusted and adjusted data at each PS. Results: There was a significant interaction between gender and PS on all measurements (p b 0.001–0.003). Males had greater FL, BA, CA, SSI and CSMA at all PS (p b 0.001– 0.046), greater CT at PS4 (p = 0.012), greater BMC at PS2-5 (p b 0.001–0.041) and MA at PS1 and PS3–5 (p b 0.001– 0.013), in comparison to females. There were significant effects of FL upon all measured parameters (p b 0.001–0.017). After adjusting for FL, BA and MA were larger in males at PS1 and PS3–5 (p b 0.001–0.016) and CA, CC, SSI, CSMA were greater males, compared to females, at PS3–5 (p b 0.001–0.016). CT was greater in females at PS1, in comparison to males (p = 0.005). Conclusion: These data confirm there is a gender-specific pattern of development in bone geometry, strength and muscle of the midshaft radius and forearm during pubertal development. However, after taking into account body size, most parameters tend to be greater in males between mid to late puberty; it is feasible this may represent an adaptive response to greater muscle forces in males.
doi:10.1016/j.bone.2007.04.003 doi:10.1016/j.bone.2007.04.004
Gender differences in cortical bone geometry and strength of the midshaft radius in healthy children during pubertal development after adjustment for body size R.L. Ashby 1, M.Z. Mughal 2, J.E. Adams 1, K.A. Ward 1 1 Imaging Science and Biomedical Engineering, The University of Manchester, Manchester, United Kingdom 2 Saint Mary’s Hospital for Women and Children, Manchester, United Kingdom
Improving reporting of bone densitometry data in children—The Manchester experience R.L. Ashby 1, M.Z. Mughal 2, K.A. Ward 1, J.E. Adams 1 1 Imaging Science and Biomedical Engineering, The University of Manchester, Manchester, United Kingdom 2 Saint Mary’s Hospital for Women and Children, Manchester, United Kingdom
Background: Gender differences in bone size during growth are well studied. However, there are a lack of data describing gender differences in cortical bone geometry and strength during pubertal development. In addition, few studies have investigated gender differences in cortical parameters after adjustment for body size. Aim: The aim of this study was to investigate gender differences in cortical bone geometry, strength and crosssectional muscle area (CSMA) of the midshaft radius and forearm in males and females during pubertal development, and to assess gender differences after adjusting for forearm length (FL) at each pubertal stage (PS). Participants and methods: A total of 369 (198 M, age 12.3 ± 3.3, range 5–19 years) healthy children were studied. FL (mm) was measured. Each participant’s PS was assigned a grade of 1–5 according to the Tanner (1969) criteria. The Stratec XCT-2000 pQCT scanner was used to measure bone area (BA), cortical area (CA), medullary area (MA) and CSMA (mm2), cortical bone mineral content (BMC, mg/mm) and the strength strain index (SSI, mm3) of the 50% midshaft radius diaphysis (midshaft radius). Two-way ANOVA was
The reporting of bone densitometry data in children is currently problematic. There is a paucity of appropriate reference data, and some manufacturers do not provide any method of size-adjustment for DXA measurements. In light of these inadequacies, our aim was to improve the reporting and interpretation of paediatric bone densitometry reports in children. A total of 435 (235 male; age 12.0 ± 3.1, range 5–18 years) healthy Caucasian children from Greater Manchester (United Kingdom [UK]) were studied. The Hologic QDR Discovery DXA scanner was used to provide measurements of the lumbar spine (LS) and left femoral neck (FN); bone mineral apparent density (BMAD, g/cm3) was calculated. The Stratec XCT-2000 pQCT scanner measured total and trabecular volumetric BMD (vBMD, mg/mm3) of the 4% distal radius. The LMS method (Cole and Green, 1992) was used to generate gender-specific centile curves for LS and FN BMAD, total and trabecular vBMD at each yearly age group. Centile curves for LS BA for height and LS BMC for BA (Molgaard et al., 1997) were produced for each gender. A report is shown in example 1. The automatically generated report shows actual and z-scores (z); the child’s results are also
S24
ABSTRACTS / Bone 40 (2007) S22–S89
plotted on centile charts. These provide clinicians with an informative assimilation of paediatric bone densitometry and anthropometric data, which is easy to interpret. We are in the process of implementing data for assessing bone mineral content in relation to lean muscle mass in our reports, to further improve the diagnosis of bone diseases in children. Example 1: Bone densitometry report for a male of age 15.7 Height (176.5 cm) (z = 0.6), weight (69.8 kg) (z = 1.0), BMI (22.4 kg/m2) (z = 0.1)1 pQCT trabecular vBMD 84.8 mg/cm3 Reduced (z = − 6.2) pQCT total vBMD 212.0 mg/cm3 Reduced (z = − 3.2) DXA Lumbar spine BMAD 0.153 g/cm3 Reduced (z = − 3.5) DXA left femoral neck BMAD 0.236 g/cm3 Reduced (z = − 2.2)2 Molgaard (1997) calculations for LS DXA: Are bones short for age? NO (z = 0.6); Are bones narrow for height? NO (z = 0.3); Are bones undermineralised for their size? YES (z = − 2.2) Conclusion: Osteopaenia confirmed by pQCT, DXA femoral neck and lumbar spine. Based on Molgaard calculations, bones are normal for age and height, but undermineralised. 1 Calculated from (Cole et al., 1990; Freeman et al., 1990). 2 Calculated from Manchester normative reference data (Ward et al., 2007). doi:10.1016/j.bone.2007.04.005
The effect of socioeconomic status upon bone geometry and bone mineral density at different skeletal sites in healthy children R.L. Ashby 1, S.A. Roberts 2, M.Z. Mughal 3, J.E. Adams 1, K.A. Ward 1 1 Imaging Science and Biomedical Engineering, The University of Manchester, Manchester, United Kingdom 2 Medical Biostatistics, The University of Manchester, Manchester, United Kingdom 3 Saint Mary’s Hospital for Women and Children, Manchester, United Kingdom Background: Studies indicate children from socially deprived backgrounds have higher fracture rates in comparison to those from more affluent areas (Stark et al., 2002). However, the effect of socioeconomic status upon bone geometry and bone mineral density (BMD) is not well studied (Clark et al., 2005). Aim: The aim of this study was to assess the effect of socioeconomic status upon bone parameters of the whole body (WB), lumbar spine (LS), femoral neck (FN), distal and midshaft radius in a cohort of healthy children. Participants and methods: A total of 435 (235 male; age 12.0 ± 3.1, range 5–18 years) healthy Caucasian children from the Greater Manchester region (United Kingdom [UK]) were studied. Height (cm) and weight (kg) were measured; body
mass index (BMI [kg/m2]) was calculated. The Hologic QDR Discovery DXA scanner measured WB, LS and FN bone area (BA, cm2) and bone mineral content (BMC, g). Bone mineral apparent density (BMAD, g/cm3) of the LS and FN was calculated. The Stratec XCT-2000 pQCT scanner measured total and trabecular volumetric BMD (vBMD, mg/mm3) and BA (mm2) of the 4% distal radius and BA, cortical area, medullary area, cross-sectional muscle area (mm2), cortical BMC (mg/ mm), cortical vBMD and cortical thickness (mm) of the 50% midshaft radius. Standard deviation scores (SDS) for anthropometric measurements and all bone parameters were calculated. Participants postcodes were used to determine the Townsend Deprivation Score (TDS) as an indicator of socioeconomic status, using data from the UK 2001 Census. Results: The TDS was determined for the locality of 382 participants. The median TDS was − 0.38, close to the population average (interquartile range − 2.5 to 2.6). Fortythree percent of participants had TDS N 0; the distribution was skewed towards higher deprivation (more deprived) scores (range − 4.8 to 13.9). There was a significant positive association between the TDS and weight (r = 0.11, p = 0.036) and BMI (r = 0.10, p = 0.056) SDS. Distal radius BA increased with the TDS (r = 0.21, p b 0.001), as did trabecular vBMD (r = 0.09, p = 0.089), while total vBMD decreased (r = − 0.17, p b 0.001). There were no significant associations for any parameters at the WB, LS, FN or midshaft radius. Conclusion: These data indicate children from socially deprived backgrounds have greater BA and trabecular vBMD, but reduced total vBMD at the distal radius; this may be due to greater body weight and BMI. However, we cannot ascertain from these data whether the TDS and associated lifestyle factors also have an effect; this requires further investigation. doi:10.1016/j.bone.2007.04.006
Influence of body composition and weight-bearing physical activity in BMD of pre-pubertal children F. Baptista, I. Fragoso, F. Vieira, C. Barrigas, L. Soutello, P.J. Teixeira, H. Santa-Clara, P. Mil Homens, M.J. Valamatos, L.B. Sardinha Exercise and Health, Faculty of Human Movement, Technical University of Lisbon, Lisbon, Portugal The main purpose of this study was to analyze the influence of mechanical factors such as body composition and physical activity on bone mineral density (BMD) of pre-pubertal girls and boys, accounting for bone age, a common confounding variable. Of the 141 starting participants 117 completed all assessments and were included in this study, namely 53 girls (G) (chronological age, 8.5 ± 0.5 years; bone age 8.5 ± 1.3 years; height,131.8 ± 5.3 cm; weight, 30.2 ± 6.5 kg) and 64 boys (B) (chronological age, 8.5 ± 0.5 years; bone age, 8.9 ± 1.1 years; height,133.4 ± 7.0 cm; weight, 32.3 ± 7.3 kg). Evaluation of body weight, total lean and fat mass, and BMD of lumbar spine (L1–L4), femoral neck, and radius were performed with DXA.