Twelve-month resistance and impact exercise program or risedronate provides a relative benefit to hip bone structure in postmenopausal women: results from a randomized controlled trial.

Abstract

Exercise and bisphosphonates reduce fracture risk, but their impact on estimates of bone strength remains uncertain. This study compared changes in tibial bone strength using peripheral quantitative computed tomography (pQCT) and hip structure analysis (HSA) outcomes from dual-energy X-ray absorptiometry (DXA) scans in postmenopausal women with low bone mass assigned to 12 months of exercise, risedronate, or control. In this RCT, 276 postmenopausal women within 6 years of menopause were randomly assigned to three groups: exercise (92), risedronate (91), or control (93). Exercise included weighted jogging and progressive resistance exercises; risedronate treatment was 150 mg monthly; all groups received calcium and vitamin D. pQCT and DXA images were obtained at baseline and 6 and 12 months and compared between groups over time. Participants had a mean (± SD) age of 54.5 (± 3.2) years with an average of 36.7 (± 40.7) months postmenopause. No significant differences were found between groups for the change in pQCT outcomes (volumetric bone mineral density, area, and strength estimates). At 12 months, mean percent differences (95% CI) in HSA measures between exercise and controls were as follows: intertrochanteric, cross-sectional area 2.25% (0.28, 4.12) (p = .03), cross-sectional moment of inertia (CSMI) 5.67% (1.47, 9.87) (p < .01), and section modulus (SM) 4.38% (1.02, 7.74) (p = .01), and narrow neck, average cortical thickness 2.37% (-0.08, 4.83) (p = .031). Mean percent differences (95% CI) in HSA measures between risedronate and control were as follows: intertrochanteric, CSMI 4.28% (-0.24, 8.81) (p = .03) and SM 3.35% (-0.21, 6.91) (p = .03), and shaft, subperiosteal width 0.82% (0.05, 1.58) (p = .047), CSMI 2.53% (0.88, 4.18) (p = .004), and SM 1.57% (0.34, 2.8) (p = .008). Exercise maintained neck-shaft angle compared to both control 1.27% (0.13, 2.41) (p = .04) and risedronate 1.31% (0.23, 2.39) (p = .03). All other differences for changes in HSA outcomes over time were not significantly different between the exercise and risedronate groups. Exercise and bisphosphonates may influence structural and strength estimates at the hip, but not at peripheral sites (tibia). Neither exercise nor bisphosphonates were found to be superior in improving estimates of hip bone strength.