METHOD We examined 56 children with spastic CP (10 diplegia, 12 hemiplegia and 34 quadriplegia) aged 4 to 12 years (35 males, 21 females) and 29 typically developing children. Children with CP were stratified into three groups based on Gross Motor Function Classification System (GMFCS) levels I to II (n=22), III (n=8), and IV to V (n=26). Growth and clinical variables, bone markers, distal femur and lumbar areal bone mineral density (BMDa), and calcaneal broadband ultrasound attenuation (BUA) were assessed.
RESULTSThe femur BMDa and calcaneal BUA values were lower in children in low GMFCS levels than in children in high GMFCS levels (p<0.05; femur BMDa: levels I-III, 0.6-0.7g ⁄ cm 2 ; levels IV-V, 0.5g ⁄ cm 2 ; calcaneal BUA: levels I-II, 39db ⁄ MHz; levels III-V, 20-21db ⁄ MHz). Lumbar BMDa and most bone markers did not differ significantly among CP and healthy groups. Regression analysis revealed that growth variables and GMFCS level were mainly associated with lower limb BMDa and BUA, and growth variables were mainly associated with lumbar BMDa (adjusted r 2 =0.48-0.56). None of the bone markers were associated with bone density.INTERPRETATION Bone densities vary and are associated with a number of factors in different skeletal regions in children with CP with a range of motor severities.