Bone Abstracts

Background: Scoliosis is lateral curvature of the spine, and adolescent idiopathic scoliosis (AIS) accounts for the majority of cases of scoliosis. One potential determinant of scoliosis that is of great interest is bone size and density. However, there are no published studies that have investigated determinants of scoliosis using a prospective cohort design making the establishment of cause and effect difficult.

Methods: This study was based on the Avon Longitudinal Study of Parents and Children (ALSPAC), a population-based cohort. Data on total body (minus head) bone area were collected by DXA in 7333 children aged 10 years. Children with scoliosis already present at aged 10 were excluded. Data was collected on the presence or absence of scoliosis at aged 15. Other potential confounding variables were also measured. At aged 15, peripheral quantitative CT (pQCT) was used to measure bone circumference and cortical thickness. Associations between DXA bone variables and risk of scoliosis developing over the following 5 years were examined by logistic regression. Cross-sectional analyses were also carried out between pQCT bone variables and presence of scoliosis at 15.

Results: Of 4022 children, 175 (4.4%) had developed scoliosis by aged 15. After adjustment for confounders, the OR for scoliosis at aged 15 per SD increase in bone size relative to body size at aged 10 was 0.61 (95% CI 0.42 to 0.89, P=0.009). Girls with scoliosis at 15 had smaller periosteal circumference (67.81.60 vs 68.86 mm, P<0.05) and reduced cortical thickness (5.02 vs 5.16 mm, P<0.01) compared to those without scoliosis.

Conclusions: Our results show that children with smaller bones at aged 10 are more likely to develop scoliosis. This is the first prospective study in this area and adds weight to the hypothesis that smaller bone size is associated with an increased risk of scoliosis.