Gymnastic Training and Bone Mass in Prepubescent Females: Magnitude and Volume Effects of Impact Loading

Abstract

Nineteen elite (E) gymnasts (>15hours/week gymnastic training), 14 high recreation (HR) gymnasts (8-15hours/week), and 15 low recreation (LR) gymnasts (1-7. 9 hours/week) were investigated to determine the effects of varying volumes of gymnastic training on bone mineral density (BMD) in prepubescent girls. Two normoactive control groups were additionally investigated to determine whether there was a magnitude effect of mechanical loading on BMD: 16 controls (C) and 15 height-and weight-matched controls (M). The mother of each daughter was measured in order to control and investigate the familial component of bone mass. Areal bone mineral density at the left proximal femur, lumbar spine (LS), and whole body (WB) and % body fat were measured by dual energy x-ray absorptiometry (DXA), and volumetric BMD was measured at the distal radius by peripheral QCT (pQCT). DXA BMD measures were corrected for bone size and expressed as bone mineral apparent density (BMAD). The HR group was significantly younger (8.68 ± 0.844 y, mean± SD) than both the E (10.02 ± 0.776 y) and C (9.96 ± 0.898 y) groups. The C group was significantly heavier (38.88 ± 4.868 kg) than the E (27.15 ± 2.819 kg), HR. (25.44 ± 3.564 kg), LR (32.98 ± 5.786 kg), and M (26.95 ± 3.301 kg) groups. Additionally, the LR group was significantly heavier than all other groups, with the exception of the C group. Femoral neck (FN) BMD was only significantly different between the E (0.706 ± 0.051 g•cm⁻²) and LR (0.649 ± 0.069 g•cm⁻²) groups. FNBMAD was only greater in the E (0.232 ± 0.048 g•cm⁻³) group compared to the C (0.191 ± 0.052 g•cm⁻³) group. LSBMAD and WBBMAD were significantly greater in both E (0.233 ± 0.019 and 0.100 ± 0.008 g•cm⁻³) and HR (0.239 ± 0.038 and 0.100 ± 0.006 g•cm⁻³) groups when compared to the LR (0.212 ± 0.022 and 0.090 ± 0.008 g•cm⁻³) and C (0.219 ± 0.020 and 0.085 ± 0.004 g•cm⁻³) groups, respectively. Total radial and cortical radial BMD was greater in both E (360.50 ± 51.569 and 484.28 ± 70.179 mg•cm⁻³) and HR (373.10 ± 45.318 and 480.66 ± 46.720 mg•cm⁻³) groups compared to the C (296.61 ± 29.677 and 426.144 ± 37.652 mg•cm⁻³) and M (306.42 ± 24.430 and 414.571 ± 25.194 mg•cm⁻³) groups, respectively. Radial trabecular BMD was greater in both E (211. 19 ± 38. 202 mg•cm⁻³) and HR (212. 61 ± 44.299 mg•cm⁻³) groups compared to the LR (175.89 ± 29.191 mg•cm⁻³), C (162.68 ± 27.304 mg•cm⁻³), and M (171.05 ± 30.639 mg•cm⁻³) groups. There were no significant differences for any bone measure among the groups of mothers. Mother-daughter correlations were relatively weak, and often insignificant, for BMD measures (r = 0.10-0.37), but strong for radial morphometric measures (r = 0.43-0.55). Radial trabecular BMD (r = 0.37; p<0.01) was more significantly correlated with gymnastic training volume (hours/week) than radial cortical BMD (0.30; p<0.05). These results suggest that there is a volume of training effect on BMD and a magnitude effect of mechanical loading on BMD. It appears that trabecular bone at the distal radius may adapt more rapidly or be more sensitive than cortical BMD to the strains imposed by impact loading. Additionally, it appears that, during prepubescence in females, bone morphometric properties may be more genetically regulated than bone mineralization.