ISSN

2277 - 3282

e ISSN

2277 - 3290

Publisher

Journal of Science

EFFECTIVE PREDICTION OF OSTEOPOROSIS RISK IN CEREBRAL PALSY CHILDREN
Author / Afflication
Abeer Hegazy

Rehabilitation Center, Dammam Medical Complex, Dammam, Saudi Arabia.
Paula Karabelas

Independent Scientist, New Jersey, NJ, USA.
Abdulhafez Selim

Philadelphia College of Osteopathic Medicine (PCOM), Philadelphia, PA, USA.
Keywords
Osteoporosis ,BMD ,BMI ,GMFCS ,prediction ,
Abstract

Cerebral palsy (CP) children are mostly malnourished and growth retarded. In addition to their poor growth, children with moderate to severe CP often endure pathological fractures. The risk of an osteoporotic fracture is further complicated by the difficulty associated with bone mineral density (BMD) assessment in these children. Therefore, clinicians need a method that can easily and effectively predict the osteoporosis risk. We conducted a study to evaluate and satisfy this clinical need. The study included thirty patients, between 5-8 years of age. Case record forms were used to record full medical history. Information included age; sex; weight; height; Body Mass Index (BMI); mid arm circumference; head circumference; triceps skinfold thickness; Gross Motor Functional Classification System (GMFCS); presence of epilepsy; intake of antiepileptic drugs; presence of pseudobulbar palsy, presence of oromotor dysfunction; spasticity level; and socioeconomic status. The patients were appointed for Dual energy x-ray absorptiometry (DXA) scans and Bone Mineral Density (BMD) z-scores were collected. Data were analyzed using data mining, correlation, and Bayesian simulation approaches. Data mining outcomes suggested that BMI and GMFCS are the most important prediction factors. Furthermore, we tested the correlation between the BMI values, GMFCS, and the BMD z-scores. The results showed that there is a significant correlation between BMD z-scores and BMI values, as well as GMFCS (P-values <0.05 and 0.001, respectively). We then applied Bayesian simulation approach to predict the BMD z-scores using BMI and GMFCS. Predicted BMD zscores were then compared to the actual BMD z-scores. The comparison showed that BMI and GMFCS could provide a diagnostic tool with high predictability. These data proposed that managing physicians could use the patient’s BMI and GMFCS data to predict the osteoporosis risk in CP children.

Volume / Issue / Year

6 , 8 , 2016

Starting Page No / Endling Page No

420 - 425