Background: Adolescent Idiopathic Scoliosis is a 3-dimensional deformity of the spine affecting peri-pubertal adolescents (10-17-y) mostly. Although generalised osteopenia is well documented in AIS, the patho-physiology of AIS related osteopenia is uncertain.

Aim: We studied the association between pubertal-growth, BMD, bone-turnover, calcium intake (CA) and physical-activity (PA) in AIS and compared to those of healthy girls.

Methods: 894 girls (594 AIS & 300 healthy controls) aged 11–16-y entered the study. Anthropometric parameters, areal-BMD of the proximal-femur and volumetric-BMD of the distal-tibia were determined by Dual-x-ray-Absorptiometry and peripheral QCT respectively. Bone-turnover-markers: bone-alkaline-phosphatase (bALP) and deoxypyridonine (Dpd) were assayed. CA and weight-bearing PA were assessed by FFQ method.

Results: Weight of AIS at < 12-y and 13-y was significantly lower than controls (P< 0.05). From 13-y, corrected right and arm-span of AIS were significantly longer than the controls (P< 0.02). aBMD and vBMD were 6.7% and 8.4% respectively lower than the controls across the ages (P< 0.05). The disparity in BMD compared with controls increased with age. CA was not different between the AIS and controls (361 mg/d, IQR:230–532mg/d vs. 319 mg/d, IQR:220–494mg/d; P=0.063). Weight-bearing PA of AIS was significantly lower than those of controls (P< 0.02).

CA of AIS and controls reached < 40% of the Chinese calcium DRI (1000 mg/d). Both CA and weight-bearing PA were correlated with BMD in AIS (P< 0.04 & P=0.002 respectively). Both CA and PA were independent predictors on the variations of aBMDs (P< 0.03) and vBMDs (P< 0.04) in AIS after controlling for confounders in multivariate analysis. Regarding bone turn-over rate, bALP in AIS was 38.6% higher than the controls from 13-y onwards (P< 0.005) while Dpd of AIS was 30.4% lower than controls at age 15-y (P=0.003). Furthermore, bALP in AIS was negatively correlated with age-adjusted BMD (r=−0.34, P< 0.001) while the correlation was weaker in the controls (P=0.14, P< 0.002).

Conclusion: The correlation of calcium intake and physical activity with BMD occurred predominantly in AIS only and that these two factors were also independent determinants on BMD of AIS implying that calcium intake and physical activity were significant modulators on BMD in AIS. Significantly faster physical-growth, higher bone formation rate were associated with lower BMD. Osteopenia in AIS could be interplayed by abnormally faster pubertal-growth and bone-turnover. In fact, Calcium intake of AIS was too low to meet the calcium demand for bone-mineralization. A controlled calcium supplementation and programmed physical activity intervention trial is merited to confirm the effect of Calcium intake and physical activity on bone acquisition in AIS at peripubertal period.

Background: Low bone mass in patients with adolescent idiopathic-scoliosis has been well reported in cross-sectional studies. No large-scale longitudinal-study has been conducted to track bone-mineral-density (BMD) trajectory in peripubertal AIS with varying scoliosis-severity.

Aim: We evaluated the BMD trajectory and factors determining BMD accretion in AIS during peripubertal period.

Method: One hundred and ninety-six newly diagnosed AIS girls with Cobb-angle > 100 and 122 healthy girls, aged 12–15 years were followed-up for two years. Weight, height, leg length, menarche and Cobb-angle were determined. Areal lumbar-spinal BMD (LSBMD) and femoral-neck BMD (FNBMD), and volumetric distal-tibial BMD (TiBMD) were evaluated by dual-energy-x-ray-absorptiometry and peripheral QCT respectively. BMD growth-models were fitted by multilevel modelling (mixed longitudinal design).

Results: At baseline, 93% participants were pre-menarchial or within three years of menarche. Average Cobb-angles at baseline and subsequent follow-ups were 260, 230 and 260 respectively. TiBMD of AIS (moderate- and severe-severity) was significantly lower than the controls from 13–16 years (ANOVA, P< 0.05). Posthoc-test showed that TiBMD of severe-AIS was lower than moderate-AIS at 15–16 years (P< 0.05). LSBMD accrual was significantly lower among AIS (moderate- and severe-severity) than the controls from age 13–17 years (ANOVA, P< 0.05). FNBMD of AIS (moderate- and severe-severity) was lower than the controls at 15 years (ANOVA, P< 0.05). BMD trajectories of individuals differed inter-personally and intra-personally over time and that BMD growth followed a curvilinear pattern. The rates of BMD accretion reduced with retarded growth across peripubertal-period. Weight and height were significant time-varying predictors on BMD growth. BMD of AIS was persistently lower than the healthy girls throughout the study (P< 0.05).

Conclusions: This large-scale longitudinal study in AIS girls with moderate to severe-curve-severity showed for the first time that both the volumetric and areal BMD were persistently lower when compared to the age-matched healthy girls throughout 12–17 years. AIS with more severe curve-severity were found to have much lower BMD throughout the peripubertal period. Promotion of a higher bone-mass is important for AIS to modify scoliosis-progression and to achieve peak bone mass in order to reduce the risk of osteoporosis later in life.

We undertook a comparative study of magnetic resonance imaging (MRI) vertebral morphometry of thoracic vertebrae of girls with adolescent idiopathic thoracic scoliosis (AIS) and age and gender-matched normal subjects, in order to investigate abnormal differential growth of the anterior and posterior elements of the thoracic vertebrae in patients with scoliosis. Previous studies have suggested that disproportionate growth of the anterior and posterior columns may contribute to the development of AIS. Whole spine MRI was undertaken on 83 girls with AIS between the age of 12 and 14 years, and Cobb’s angles of between 20° and 90°, and 22 age-matched controls. Multiple measurements of each thoracic vertebra were obtained from the best sagittal and axial MRI cuts. Compared with the controls, the scoliotic spines had longer vertebral bodies between T1 and T12 in the anterior column and shorter pedicles with a larger interpedicular distance in the posterior column. The differential growth between the anterior and the posterior elements of each thoracic vertebra in the patients with AIS was significantly different from that in the controls (p < 0.01). There was also a significant positive correlation between the scoliosis severity score and the ratio of differential growth between the anterior and posterior columns for each thoracic vertebra (p < 0.01). Compared with age-matched controls, the longitudinal growth of the vertebral bodies in patients with AIS is disproportionate and faster and mainly occurs by endochondral ossification. In contrast, the circumferential growth by membranous ossification is slower in both the vertebral bodies and pedicles.