Document Type


Date of Degree

Fall 2015

Degree Name

PhD (Doctor of Philosophy)

Degree In


First Advisor

Torner, James C

First Committee Member

Burns, Trudy

Second Committee Member

Janz, Kathleen

Third Committee Member

Eichenberger Gilmore, Julie

Fourth Committee Member

Levy, Steven

Fifth Committee Member

Schlechte, Janet


The objective of this research was to evaluate the relationships between greater adiposity and bone development during adolescence. Bone was evaluated from age 11 to 17 years in the Iowa Bone Development Study using peripheral quantitative computed tomography (pQCT). Body composition (fat and lean mass) was estimated by dual energy x-ray absorptiometry (DXA).

The first research aim evaluated the associations between greater overall adiposity and subsequent maturation and bone strength in 135 girls and 123 boys. Greater adiposity was defined according to age 8 Body Mass Index (BMI) to categorize participants as overweight (OW) or healthy-weight (HW). Maturation was defined as the age of peak height velocity (PHV). Bone strength was assessed at the radius and tibia (bone strength index, BSI, and strength-strain index, SSI). Differences in bone strength between OW and HW were evaluated with sex-specific multi-level regression models to account for individual growth and correlation between repeated measurements. Analyses were adjusted for centered age (measurement visit age - grand mean age of cohort), change in fat mass, and limb length in Model 1, with additional adjustment for lean mass in Model 2. Analyses were repeated using biological age (visit age - age PHV). BMI was positively associated with age of maturation in girls and boys (p< 0.05). HW versus OW girls had significantly lower BSI and SSI at the radius and tibia (p< 0.05) in Model 1. Results remained significant except for radial BSI in Model 2. HW versus OW boys had significantly lower BSI and SSI (all p< 0.05) at the tibia, but not radius, in Model 1. Significant differences were sustained in Model 2. Analyses were repeated using biological age, which yielded similar results for boys, but reduced parameter estimates were observed in girls, with only tibial SSI significant in Model 2 (p< 0.05). These findings support a stronger role for greater adiposity in the occurrence of earlier maturation and greater bone strength in girls than boys while greater lean mass appeared to play a greater role in boys.

The second research aim evaluated associations between abdominal adiposity and bone in 132 girls and 122 boys. Visceral adipose tissue area (VAT, cm2) and subcutaneous adipose tissue area (SAT, cm2) were estimated from DXA scans. Sex-specific analyses evaluated the fat-bone relationship with growth models using biological age as the time variable adjusted for limb length and lean mass. There were no significant associations between bone parameters and VAT or SAT in girls. In boys, greater VAT was associated with lower trabecular bone density (tBMD) and BSI (all p< 0.05) at the tibia, but not radius. Greater VAT and SAT were associated with smaller cortical bone size and thickness (all p< 0.01) at the radius, but not tibia. Analyses limited to overweight participants showed VAT was negatively associated with periosteal circumference at the radius and tibia, cortical bone thickness at the tibia and SSI (all p< 0.05) at the radius in girls. In boys, the results were relatively unchanged for VAT, while SAT was only significantly associated with lower tBMD (p< 0.05) at the tibia. These results suggest the bone-fat relationship may vary depending on adiposity and bone site.

The third research aim evaluated the longitudinal association between intramuscular fat and cortical bone at the tibia from age 11 to 17 years in 153 girls and 143 boys. Muscle density (MD) was used to estimate intramuscular fat (IMF). Lower MD indicates greater IMF. The relationships between muscle density and cortical bone parameters were modeled using multi-level regression models adjusted for biological age, limb length and muscle cross-sectional area measured by pQCT. In the adjusted multi-level regression models, MD was positively associated with cortical bone parameters, but only reached statistical significance for BMD, bone mineral content (BMC), bone cross-sectional area, cortical thickness and SSI in girls, while only SSI was significant in boys (all p< 0.05). These results suggest that greater fat content within muscle may be harmful to weight-bearing cortical bone during adolescence.

In conclusion, findings from the first aim suggest there are sex- and site-specific differences in the relationship between adiposity and bone during adolescence. Findings from the second and third aims indicate these differences could be explained, in part, by the existence of specific fat depots (abdominal more so than intramuscular fat) that could be harmful to bone and that may be more apparent in boys due to a sex-specific fat distribution pattern that favors accumulation of abdominal rather than peripheral fat.


bone development, bone strength, childhood obesity, osteoporosis, pQCT


xiv, 186 pages


Includes bibliographical references (pages 167-186).


Copyright © 2015 Natalie Ann Glass