OR03-6 Tibial and Radial Bone Structure as Assessed by HRpQCT May Explain Differences in Peripheral Skeletal Integrity and Fracture Risk Across the Weight Spectrum That Cannot Be Explained by Areal BMD Alone

No study has investigated tibial or radial bone structure or estimated strength by high-resolution peripheral quantitative CT (HRpQCT) in women across the weight spectrum. We studied 139 women, 17-46yo: 1) anorexia nervosa (AN n=74), 2) lean controls (HC n=33) and 3) overweight/obese (OB n=32). All HC and OB were eumenorrheic. Areal BMD (aBMD) and appendicular lean mass (ALM) were measured by DXA; cortical and trabecular volumetric BMD (vBMD), cortical thickness, cortical porosity, failure load (finite element analysis) and trabecular plate parameters (individual trabecular segmentation) at the tibia and radius by HRpQCT. Mean age was similar among groups. Mean BMI was 17.7 ± 2.1 vs 22.6 ± 1.4 vs 35.2 ± 3.2 kg/m(2) (p<0.0001). Mean aBMD Z-scores were lowest in AN, intermediate in HC and highest in OB (PA spine -1.6 ± 1.3 vs -0.5 ± 0.8 vs 0.5 ± 0.8; total hip -1.0 ± 1.1 vs 0.2 ± 0.9 vs 1.0 ± 0.9; radius -0.6 ± 1.0 vs 0.1 ± 0.9 vs 0.7 ± 0.9; p<0.0001). At the tibia and radius, mean trabecular vBMD and failure load were lowest in AN, intermediate in HC and highest in OB, while mean trabecular plate bone volume fraction, plate number density and plate-plate junction density were lower in AN (p<0.05) and similar in HC and OB. Mean cortical vBMD and cortical porosity were similar among groups, but cortical thickness was lower in AN vs OB (p<0.01). There was a quadratic relationship between BMI and trabecular vBMD, failure load and all plate parameters (R=0.40-0.67, p<0.0001), such that the higher the BMI, the smaller the increase in these parameters. In contrast, ALM had a consistent positive linear relationship with these parameters (R=0.37-0.78, p<0.0001). IGF-1 levels were positively associated with tibial and radial estimated strength and trabecular structure (R=0.21-0.37, p<0.05). Duration of amenorrhea in AN was negatively associated with radial estimated strength and trabecular structure (R= -0.25- -0.37, p<0.01). Despite similar mean aBMD at all sites, those with a history of fracture (n=57) had lower mean tibial and radial trabecular vBMD and plate-plate junction density; lower mean tibial cortical vBMD, plate bone volume fraction and plate number; and higher mean tibial cortical porosity compared to those without (n=72) (p<0.05). Conclusions: Trabecular bone structure and estimated strength were generally higher with greater weight, but the incremental increase was smaller as BMI increased into the obese range. Cortical bone parameters generally did not increase with greater weight. This suggests that the skeleton in women with obesity may not be able to fully adapt to an increased weight load. Muscle mass and IGF-1 levels were positive, and duration of amenorrhea a negative, determinant of bone structure and estimated strength. Across the weight spectrum, tibial and radial bone structure by HRpQCT may explain differences in peripheral skeletal integrity and fracture risk that cannot be explained by aBMD alone.