Association of Gross Motor Function Classification System Level and School Attendance with Bone Mineral Density in Patients With Cerebral Palsy

ARTICLE IN PRESS Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health, vol. ■, no. ■, 1–6, 2016 © Copyright 2016 by The International Society for Clinical Densitometry 1094-6950/■:1–6/$36.00 http://dx.doi.org/10.1016/j.jocd.2016.09.002

Original Article

Association of Gross Motor Function Classification System Level and School Attendance with Bone Mineral Density in Patients With Cerebral Palsy Ki Jin Jung,1 Soon-Sun Kwon,2 Chin Youb Chung,3 Kyoung Min Lee,3 Ki Hyuk Sung,3 Byung Chae Cho,3 Myoung Ki Chung,4 Seung Jun Moon,3 Jaeyoung Kim,3 and Moon Seok Park*,3 1

Department of Orthopedic Surgery, Soonchunhyang University Hospital, Cheonan, South Korea; 2Department of Mathematics, College of Natural Science, Ajou University, Suwon, South Korea; 3Department of Orthopedic Surgery, Seoul National University Bundang Hospital, Kyungki, South Korea; and 4Department of Orthopedic Surgery, Gangwon National University Hospital, Gangwon, South Korea

Abstract The present study aimed to evaluate bone mineral density (BMD) in children and adolescents with cerebral palsy (CP) and to critically analyze the effects of a variety of factors, particularly the Gross Motor Function Classification System (GMFCS) level, the Caregiver Priorities and Child Health Index of Life with Disabilities questionnaire, and the Pediatric Outcomes Data Collection Instrument (PODCI), on BMD. Fifty patients with CP who underwent dual-energy X-ray absorptiometry were included. Collected data included the extent of involvement, muscle tone, demographic data, factors determined through chart review, and laboratory results. Factors associated with BMD in this group were analyzed by performing multiple regression analysis. The mean Z-scores in male and female patients were −3.252 ± 1.822 and −3.789 ± 1.764, respectively, in the proximal part of the femur and −2.219 ± 1.323 and −2.451 ± 1.434, respectively, in the lumbar spine. In multiple regression analysis, the GMFCS level and the average frequency of missed school in the PODCI were significant factors associated with both femur and lumbar spine BMD. Both the GMFCS level and school attendance were independently associated with BMD and should be considered for the prevention and management of osteoporosis in patients with CP. Key Words: Bone mineral density; cerebral palsy; GMFCS; school attendance.

Introduction

live births, making it one of the most common developmental disorders (1). In recent years, the incidence rate of CP has been increasing internationally due to increased survival of low birth weight infants. CP concurs with epilepsy and various disturbances in domains such as sensation, cognition, and communication (1,2). Patients with CP also have deficient bone growth and consequently an increased probability for nontraumatic fractures. In a reported series of nonambulatory children and young adults, nearly 20% had sustained a femur fracture at some point in their lives (3,4). Although these fractures are treated nonoperatively, hospitalization may be required, and malunion of the fracture is common (5,6). Repeated fractures decrease the quality of life for such children and add to their care

Cerebral palsy (CP) is defined as a group of nonprogressive motor impairment syndromes secondary to lesions or anomalies of the brain in its early developmental stages. The prevalence of CP is above 2.0 per 1000

Received 06/22/16; Revised 08/29/16; Accepted 09/8/16. Ki Jin Jung and Soon-Sun Kwon contributed equally to the writing of this article. *Address correspondence to: Moon Seok Park, MD, Department of Orthopedic Surgery, Seoul National University Bundang Hospital, 300 Gumi-Dong, Bundang-Gu, Sungnam, Kyungki 463707, South Korea. E-mail: [email protected]

1

ARTICLE IN PRESS 2 requirements. It has been demonstrated that the most significant problem that provoked medical expenses in a residential school for children with severe CP was treatment of fractures (7). There are multiple potential causes of deficient bone mineralization in children with CP. First, limited weightbearing ambulation during skeletal growth is a major factor. Physically handicapped CP patients do not have the opportunity to stand, which may inhibit normal mineralization. Another mechanical factor is the temporary immobilization that children with CP often experience after multiple operative procedures. Children who are managed with a hip spica cast are at considerable risk for subsequent fractures, because even normal children who have an uncomplicated fracture of the lower limb treated with 4 weeks of casting demonstrate a small, persistent deficit in bone density (3,8). Other factors may also play a role. Oral–motor dysfunction can make feeding difficult for many individuals with CP. Poor nutrition and low calcium intake are common in this population and could contribute to poor mineralization (9–11). Additionally, many children with CP take or have taken anticonvulsant medication, which may also result in diminished bone mineralization (11–13). Metabolic bone disease is associated with prematurity. Infants who are born prematurely with a low birth weight have lower than normal bone mineral density (BMD) at 1 yr of age. In normally growing children, the accrual of peak bone mass follows peak height velocity. However, children with CP demonstrate differences in linear growth and also frequently grow slowly. The impact of these altered growth patterns on skeletal development and bone density is significant. Not all these risk factors apply to all patients with CP. However, these patients are at increased risk for the development of osteoporosis through a variety of potential mechanisms. The purpose of the present study was to evaluate the BMD of children and adolescents with CP and to critically analyze the effects of a variety of factors on BMD, with a focus on Gross Motor Function Classification System (GMFCS) level, the Caregiver Priorities and Child Health Index of Life with Disabilities (CPCHILD) questionnaire, and the Pediatric Outcomes Data Collection Instrument (PODCI).

Materials and Methods This retrospective study was approved by the institutional review board at our institute and adhered to the tenets of the Declaration of Helsinki. Inclusion criteria were as follows: (1) consecutive patients <20 yr old with CP, who visited our clinic (a tertiary referral hospital for patients with CP) from July 2007 to February 2014, and (2) patients who underwent dual-energy X-ray absorptiometry (DXA). We excluded patients with recognized chromosomal abnormalities and genetic disorder, which influence the BMD. Patients with motion artifact on DXA or previous hip or scoliosis surgery were also excluded for

Jung et al. reliable assessment of BMD in the proximal femurs and lumbar spine. Finally, 50 patients were enrolled in the present study.

Measurements of BMD BMD in the proximal femurs and lumbar spine was measured with DXA (GE Lunar Prodigy; Lunar Corporation, Madison, WI) using standard scanning procedures. The scans were performed by specially trained technicians, according to the manufacturer-provided protocol. We endeavored to reduce the coefficient of variation to <1.0% for the femoral neck and lumbar spine measurements. BMD measurements (gram per square centimeter) were converted to age- and gender-normalized standard deviation (SD) scores (Z-scores). Osteopenia was defined as a calculated BMD Z-score less than −2 SD from the expected mean value for age and gender.

Previous Medical Assessment A chart review was conducted to determine the extent of involvement (quadriplegia, diplegia, or hemiplegia), muscle tone (spastic, dystonic, or mixed), and demographic data such as gender and age. Motor functional status related to CP was graded according to the GMFCS (14,15). Previous periods of immobilization, usually due to orthopedic surgery procedures, were categorized as none (patients who had never been managed with immobilization), brief (patients who had sustained soft tissue surgery or had been managed with cast immobilization for 4 weeks or less), and prolonged (patients who had undergone bony surgery or had been managed with immobilization for more than 4 weeks). Additionally, we checked the history of anticonvulsant medication and categorized patients according to whether or not they were using anticonvulsant medication at the time of evaluation. We assessed nutrition and oral motor dysfunction in several ways. The parameters of height and weight were measured using standardized instruments, and the body mass index was calculated as weight in kilogram divided by the square of the height in meter. Serum data were also used for analysis.The levels of protein, cholesterol, calcium, phosphate, creatinine, albumin, and alkaline phosphatase (ALP) were analyzed. Osteocalcin, 25hydroxyvitamin D, N-terminal telopeptide, and C-terminal telopeptide were also assessed. Osteocalcin and ALP, synthesized by osteoblasts during the process of bone formation, are released into the blood. Although ALP is produced by several organs including the liver, bile duct, kidneys, and bone, it is primarily associated with osteoblastic activity in bone metabolism (16). Creatinine is a breakdown product of creatine phosphate in muscle, which the body is known to produce at a constant rate and filter out via the kidneys. Consequently, the serum creatinine level was considered to reflect patients’ muscle mass. Similarly, N-terminal telopeptides are breakdown products of collagen resulting from an osteoclast-mediated resorption of the extracellular matrix of bone and can be used as a marker of bone resorption.

Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health

Volume ■, 2016

ARTICLE IN PRESS Association of GMFCS and School attendance with BMD in CP CPCHILD Questionnaire The CPCHILD questionnaire was developed to assess the health-related quality of life of children with severe developmental disabilities from the perspective of caregivers (17,18). It consists of 36 items divided into 7 domains: Personal Care/Activities of Daily Living (9 items); Positioning, Transferring, and Mobility (8 items); Comfort (7 items); Emotions (2 items); Communication and Social Interaction (7 items); Health (2 items); and Overall Quality of Life (1 item). We assessed the score of the above 7 domains and an additional 2 items (time of parent caregiving and academic ability of the parent).

PODCI Questionnaire The PODCI questionnaire was designed to assess functional health outcome in children and adolescents, and places a focus on musculoskeletal health (19,20). These parent and adolescent self-report forms have been used to measure improvements following orthopedic surgery in various diseases including CP. The items on the PODCI focus on function and quality of life, and in the clinical setting, the PODCI is frequently used to provide patients and parents with evidence of a likely clinical outcome. The PODCI consists of 6 subscales (Upper Extremity Function, Transfers and Basic Mobility, Sports and Physical Function, Comfort, Happiness, and Global Function). In addition, subsections addressing ability to use a fork and spoon and average frequency of missed school were evaluated to survey oral intake and sunlight exposure in daily life.

Data Analysis The femur and lumbar spine BMDs were compared according to the GMFCS level and the extent of involvement. The subjects were divided into 4 groups according to their GMFCS level (I–II, III, IV, and V). The continuous or categorical variables that demonstrated a significant correlation with the femur and the lumbar spine BMDs were examined separately. The variables that were significantly correlated with BMD and the conditions wherein the BMD significantly differed between cases where the condition was present and absent were included in the multiple regression analysis. The factors that significantly affected the BMDs of the femur and the lumbar spine were evaluated using multiple regression analysis and were categorized as associated factors with BMDs. A descriptive analysis of the entire dataset, including average and SDs or proportions, was performed. The data were assessed for normality using the Kolmogorov-Smirnov test. Correlations between the variables and BMDs of the femur and the lumbar spine were evaluated using the Pearson correlation coefficient or the Spearman correlation coefficient, depending on the normality of the data. BMDs were compared across GMFCS levels I–II, III, IV, and V using analysis of variance. A stepwise selection of the variables was adopted in the multiple regression analysis. The variables that were significantly correlated with

3

Table 1 Demographic Data and Clinical Features of Motor Impairment in 50 Patients With Cerebral Palsy Variable Mean age (years, mean ± SD) Male : female Extent of motor impairment Hemiplegia Diplegia Quadriplegia Types of motor impairment Spastic Dystonic GMFCS level I II III IV V

n (%) 10.68 ± 3.992 27:23 (54:46) 6 (12) 16 (32) 28 (56) 49 (98) 1 (2) 4 6 5 17 18

(8) (12) (10) (34) (36)

Abbr: GMFCS, Gross Motor Function Classification System; SD, standard deviation.

BMDs and the conditions wherein BMD significantly differed between cases where the condition was present and absent were included as independent variables, whereas the BMDs of the femur and the lumbar spine were the dependent variables in the multiple regression analysis. All statistical analyses were performed using SPSS 22.0 (SPSS Inc., Chicago, IL), and the threshold for statistical significance was set at p < 0.05.

Results The mean age of our patient population was 10.6 yr (range: 4.9–20.0 yr). From the chart review, we determined that the patient population as a whole contained slightly more male (27, 54%) than female (23, 46%) patients. The most common diagnosis was quadriplegia in 28 patients (56%), diplegia in 16 patients (32%), and hemiplegia in 6 patients (12%). Forty-nine (98%) patients had muscle tone characterized as spastic, and 1 patient (2%) had dystonic-type muscle tone (Table 1). BMDs in the proximal parts of the femur and the lumbar spine were measured. The mean BMD Z-scores for male and female children were −3.252 ± 1.822 and −3.789 ± 1.764, respectively, for the proximal part of the femur and −2.219 ± 1.323 and −2.451 ± 1.434, respectively, for the lumbar spine. Thus, the mean BMD was 2 or 3 SDs below that of normal children of the same age (Table 2). GMFCS level was highly correlated with BMD. For the proximal part of the femur, the mean Z-score was −1.525 ± 0.332 for children who were GMFCS level I and decreased serially to a minimum of −4.538 ± 1.22 for children classified as GMFCS level V (p < 0.0001). A similar

Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health

Volume ■, 2016

ARTICLE IN PRESS 4

Jung et al. Table 2 Bone Density Z-Scores: Univariate Analyses BMD Z-score (mean ± SD)

Variable Demographic variables 2.0–7.9 yr 8.0–13.9 yr 14.0–20.0 yr Gender Male Female Extent of impairment Hemiplegia Diplegia Quadriplegia GMFCS level I II III IV V

n

Femur

Table 3 Candidate Factors Significantly Affecting BMD of the Femur and the Lumbar Spine Femur

Lumbar spine Factor

10/12 21/27 9/9

−3.360 ± 1.623 −3.423 ± 1.844 −3.866 ± 2.000

−2.250 ± 1.283 −2.114 ± 1.217 −3.222 ± 1.231

21/26 19/22

−3.252 ± 1.822 −3.789 ± 1.764

−2.257 ± 1.234 −2.472 ± 1.355

6/6 13/16 21/26

−1.050 ± 1.111 −2.815 ± 1.162 −4.638 ± 1.296

−1.097 ± 1.811 −1.718 ± 0.855 −3.038 ± 0.932

4/4 6/6 5/5 12/16 13/17

−1.525 ± 1.332 −1.883 ± 1.387 −2.020 ± 0.944 −4.483 ± 1.383 −4.538 ± 1.220

−0.671 ± 1.284 −1.783 ± 1.312 −0.960 ± 0.673 −2.725 ± 0.949 −3.017 ± 0.990

Abbr: BMD, bone mineral density; GMFCS, gross motor function classification system; SD, standard deviation.

pattern was found with regard to the BMD of the lumbar spine: the mean Z-score was −0.671 ± 1.284 for children who were GMFCS level I and decreased to a minimum of −3.017 ± 0.99 for those classified as GMFCS level V (p < 0.0001). Correlation with the pattern of involvement revealed that the Z-scores of the femur and the lumbar spine were −4.638 ± 1.296 and −3.038 ± 0.932, respectively, for patients with quadriplegia, and increased serially for diplegic and hemiplegic groups (p < 0.0001; Table 3). In multiple regression analysis using stepwise selection, the GMFCS level and the average frequency of the missed school subsection in the PODCI were significant factors associated with the femur and lumbar spine BMDs (Table 4).

Discussion In recent years, concern about the bone health of patients with CP has increased as a result of greater attention to their quality of life and increasing concern over the social and economic costs of fracture (21). The purpose of the present study was to evaluate BMD in children and adolescents with CP and to investigate factors associated with BMD at the femur and the lumbar spine. We aimed to include as many clinical variables as possible to exclude or adjust for the effects of potential confounding factors. When the factors possibly correlated with femur and lumbar spine BMDs were evaluated, the GMFCS level, the body

Age Body mass index Antiepileptic drug Protein Cholesterol Creatinine Albumin Alkaline phosphatase Calcium Phosphorous Osteocalcin N-terminal telopeptide C-terminal telopeptide Vitamin D PODCI Upper extremity function Transfer, basic mobility Sports, physical function Comfort Happiness Global function Oral intake Absenta CPCHILD Personal care/ activities Positioning, mobility Comfort Emotions Communication Health Quality of life Time of parent caregiving Academic ability of parent Total

Lumbar spine

R

p

r

p

−0.154 0.364 0.007 0.228 0.229 0.247 0.236 0.255 0.249 0.250 0.416 0.215 −0.015 −0.182

0.342 0.022 0.969 0.157 0.153 0.124 0.141 0.112 0.120 0.118 0.060 0.322 0.946 0.729

−0.277 0.280 −0.141 0.197 0.203 0.020 0.081 0.147 0.155 0.197 0.473 0.069 −0.040 −0.132

0.056 0.055 0.409 0.179 0.164 0.892 0.581 0.316 0.290 0.179 0.019 0.739 0.849 0.754

0.639 <0.001

0.581 <0.001

0.807 <0.001

0.671 <0.001

0.752 <0.001

0.659 <0.001

0.401 0.025 0.296 0.083 0.336 0.064 0.354 0.036 0.801 <0.001 0.678 <0.001 −0.614 <0.001 −0.594 <0.001 −0.425 0.017 −0.515 <0.001 0.696 <0.001

0.693 <0.001

0.780 <0.001 0.634 <0.001 0.243 0.186 0.529 0.002 0.446 0.011 0.438 0.013 0.146 0.439

0.742 0.692 0.065 0.598 0.362 0.506 0.354

<0.001 <0.001 0.705 <0.001 0.03 <0.001 0.036

0.101

0.206

0.233

0.592

0.878 <0.001

0.588 <0.001

Note: Concerns that were statistically significant are expressed in bold (p < 0.05). Abbr: BMD, bone mineral density; CPCHILD, Caregiver Priorities and Child Health Index of Life with Disabilities; PODCI, Pediatric Outcomes Data Collection Instrument. a Refers to the average frequency of the missed school subsection.

Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health

Volume ■, 2016

ARTICLE IN PRESS Association of GMFCS and School attendance with BMD in CP

5

Table 4 Multiple Regression Analysis to Identify Significant Determinants of Femur and Spine BMDs Femur BMD Factor Intercept Age Gender (male) Female—ref. BMI GMFCS (I,II)—ref. Level III Level IV Level V Absenta

Spine BMD

B

t

p

B

t

p

−2.334 −0.105 0.407 — −0.005 — −0.715 −1.816 −2.490 −0.431

−1.29 −1.51 0.82 — −0.07 — −1.1 2.11 3.21 −2.22

0.208 0.146 0.423 — 0.943 — 0.282 0.046 0.004 0.037

−1.053 −0.061 −0.253

−0.97 −1.4 −0.83

0.340 0.172 0.411

−0.001

−0.02

0.982

−0.421 −1.287 −1.757 −0.418

−1.01 2.27 3.48 −3.7

0.324 0.032 0.001 0.001

Notes: Concerns that were statistically significant are expressed in bold (p < 0.05). This table means, for example, that femoral BMD was diminished to −2.490 at GMFCS level 5 compared to that at GMFCS level I. In the absence scale, femoral BMD was −0.431, diminished 2 times lower in the patients who were absent 2 times in a week than in the patients who were never absent in a week. Abbr: BMD, bone mineral density; BMI, body mass index; GMFCS, gross motor function classification system; ref., reference. a Refers to the average frequency of the missed school subsection.

mass index, several subscales of the PODCI questionnaire, several domains of the CPCHILD questionnaire, and osteocalcin were associated with the BMD of patients with CP. These results indicate the possible risk factors for low BMD and can be applied in the clinical setting. We observed covariance between many of these factors, and the results cannot be interpreted to indicate a direct causeand-effect relationship. Multiple regression analysis of these data is important to aid in the distinction between direct and indirect relationships to BMD. After multiple regression analysis, the GMFCS level and the average frequency of the missed school subsection of the PODCI were included as significantly associated factors. It is important to address the limitations of the study. First, it is a retrospective, cross-sectional study. It is difficult to obtain specific information about skeletal health in patients with CP by a retrospective chart review. However, we endeavored to identify factors associated with BMD using CPCHILD and PODCI chart reviews as well as laboratory results. Furthermore, we have been collecting longitudinal data on BMD and laboratory examinations. Future studies will demonstrate more precise information about BMD in patients with CP. Second, due to the retrospective nature of the study, the age and the GMFCS level of subjects were not evenly distributed.Therefore, most patients (70%) were classified as GMFCS level IV–V. We expect that balancing the numbers of subjects according to GMFCS level may provide more meaningful observations. In the future, population-based studies may be required to obtain more accurate information on the prevalence rates of low BMD in children with CP, as well as its associated factors. Third, there may be some debate regarding the decision not to use the distal femur BMD,

because the distal femur was reported to be the most common fracture site (22,23). However, the great majority of orthopedic surgery procedures are performed at the proximal femur and lumbar spine. Although measurement of the distal femur could be meaningful in some regards, we placed more value on the proximal femur and lumbar spine based on their ease of accessibility and comparison. Fourth, we did not consider bone architecture, which is important for fractures. Children with CP typically have smaller bones with a thinner cortex (24). Bone strength is determined not only by bone density but also by bone geometry, including size (25,26). Consequently, further investigation of the relation of bone architecture and fracture is needed to determine the risk of fracture. An earlier report found that age was predictive of low BMD and that Z-scores declined by −0.7 per year of age (27).A more recent study confirmed that the BMD Z-score declined with age. The mean BMD Z-score in subjects less than 11 yr old was −3.3 compared to −5.1 in subjects over 11 yr old (22). However, age was not a significant factor associated with femur and lumbar spine BMDs in our study. Gender was also not a significant factor associated with BMD. As previously reported, there is a strong correlation between BMD and the severity of CP determined by the GMFCS level. In the present study, the GMFCS level, along with the extent of neurological involvement, was the most important factor related to low BMD in patients with CP. Although the concept of load absence on the limbs was postulated as the most significant direct cause of low bone density, the present evidence suggests that many more factors are involved. We used the PODCI and CPCHILD questionnaires to assess the relationship between BMD and health-related quality of life and well-being that is not fully

Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health

Volume ■, 2016

ARTICLE IN PRESS 6 expressed by the GMFCS level. Most of the factors in the CPCHILD questionnaire and PODCI were explained by the GMFCS level. However, the average frequency of the missed school subsection in the PODCI was revealed to be significantly associated with the BMD of CP patients independently. The average frequency of the missed school subsection in the PODCI was designed to represent the sunlight exposure of patients with CP in present study. However, there are various other possible meanings to the association of this subsection with BMD, because regularly attending school or participating in outside activities increases not only sunlight exposure and physical activity but also communication and quality of life. This result demonstrated that participation in school can have various positive effects including improved bone health in patients with CP. And we emphasize that parents and caregivers must recognize the importance of osteoporosis treatment, especially for patients with GMFCS level IV or V because they have few chances of weight bearing. Treatment options should focus on ensuring the adequacy of dietary nutritional intake and on the use of bisphosphonate medications, and an additional active weight-bearing program is needed.

Acknowledgment This research was supported by the Ministry of Trade, Industry and Energy, Korea (Grant Nos. 10049711 and 10045220) and, by projects for Research and Development of Police science and Technology under Center for Research and Development of Police science and Technology and Korean National Police Agency funded by the Ministry of Science, ICT and Future Planning (Grant No. PA-C000001-2015-202).

References 1. Odding E, Roebroeck ME, Stam HJ. 2006 The epidemiology of cerebral palsy: incidence, impairments and risk factors. Disabil Rehabil 28(4):183–191. 2. Mutch L, Alberman E, Hagberg B, et al. 1992 Cerebral palsy epidemiology: where are we now and where are we going? Dev Med Child Neurol 34(6):547–551. 3. Sturm PF, Alman BA, Christie BL. 1993 Femur fractures in institutionalized patients after hip spica immobilization. J Pediatr Orthop 13(2):246–248. 4. Pritchett JW. 1990 Treated and untreated unstable hips in severe cerebral palsy. Dev Med Child Neurol 32(1):3–6. 5. Sherk HH, Cruz M, Stambaugh J. 1977 Vitamin D prophylaxis and the lowered incidence of fractures in anticonvulsant rickets and osteomalacia. Clin Orthop Relat Res 129:251–257. 6. Pritchett JW. 1983 The untreated unstable hip in severe cerebral palsy. Clin Orthop Relat Res 173:169–172. 7. Presedo A, Dabney KW, Miller F. 2007 Fractures in patients with cerebral palsy. J Pediatr Orthop 27(2):147–153. 8. Henderson RC. 1997 Bone density and other possible predictors of fracture risk in children and adolescents with spastic quadriplegia. Dev Med Child Neurol 39(4):224–227.

Jung et al. 9. Thommessen M, Kase BF, Riis G, Heiberg A. 1991 The impact of feeding problems on growth and energy intake in children with cerebral palsy. Eur J Clin Nutr 45(10):479– 487. 10. Sullivan PB, Lambert B, Rose M, et al. 2000 Prevalence and severity of feeding and nutritional problems in children with neurological impairment: Oxford Feeding Study. Dev Med Child Neurol 42(10):674–680. 11. Henderson RC, Lark RK, Gurka MJ, et al. 2002 Bone density and metabolism in children and adolescents with moderate to severe cerebral palsy. Pediatrics 110(1 Pt 1):e5. 12. Hahn TJ. 1976 Bone complications of anticonvulsants. Drugs 12(3):201–211. 13. Crosley CJ, Chee C, Berman PH. 1975 Rickets associated with long-term anticonvulsant therapy in a pediatric outpatient population. Pediatrics 56(1):52–57. 14. Palisano R, Rosenbaum P, Walter S, et al. 1997 Development and reliability of a system to classify gross motor function in children with cerebral palsy. Dev Med Child Neurol 39(4):214–223. 15. Woo SJ, Ahn J, Park MS, et al. 2011 Ocular findings in cerebral palsy patients undergoing orthopedic surgery. Optom Vis Sci 88(12):1520–1523. 16. Leung KS, Fung KP, Sher AH, et al. 1993 Plasma bonespecific alkaline phosphatase as an indicator of osteoblastic activity. J Bone Joint Surg Br 75(2):288–292. 17. Narayanan UG, Fehlings D, Weir S, et al. 2006 Initial development and validation of the Caregiver Priorities and Child Health Index of Life with Disabilities (CPCHILD). Dev Med Child Neurol 48(10):804–812. 18. Sung KH, Kwon SS, Narayanan UG, et al. 2015 Transcultural adaptation and validation of the Korean version of Caregiver Priorities & Child Health Index of Life with Disabilities (CPCHILD). Disabil Rehabil 37:620–624. 19. Kwon DG, Chung CY, Lee KM, et al. 2011 Transcultural adaptation and validation of the Korean version of the Pediatric Outcomes Data Collection Instrument (PODCI) in children and adolescents. J Pediatr Orthop 31(1):102–106. 20. Gates PE, Campbell SR. 2015 Effects of age, sex, and comorbidities on the pediatric outcomes data collection instrument (PODCI) in the general population. J Pediatr Orthop 35:203–209. 21. Stevenson RD, Conaway M, Barrington JW, et al. 2006 Fracture rate in children with cerebral palsy. Pediatr Rehabil 9(4):396–403. 22. Henderson RC, Berglund LM, May R, et al. 2010 The relationship between fractures and DXA measures of BMD in the distal femur of children and adolescents with cerebral palsy or muscular dystrophy. J Bone Miner Res 25(3):520– 526. 23. Harke HT. 1999 Pediatric bone densitometry: technical issues. Semin Musculoskelet Radiol 3(4):371–378. 24. Houlihan CM, Stevenson RD. 2009 Bone density in cerebral palsy. Phys Med Rehabil Clin N Am 20(3):493–508. 25. Bachrach LK. 2001 Acquisition of optimal bone mass in childhood and adolescence. Trends Endocrinol Metab 12(1):22– 28. 26. Ward KA, Adams JE, Hangartner TN. 2005 Recommendations for thresholds for cortical bone geometry and density measurement by peripheral quantitative computed tomography. Calcif Tissue Int 77(5):275–280. 27. Henderson RC, Kairalla J, Abbas A, Stevenson RD. 2004 Predicting low bone density in children and young adults with quadriplegic cerebral palsy. Dev Med Child Neurol 46(6):416– 419.

Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health

Volume ■, 2016