Effects of physical activity on evolution of proximal femur structure in a younger elderly population

• Published in: Bone (New York, N.Y.) N.Y.), 2007-02, Vol.40 (2), p.506-515
• Main Authors: Kaptoge, S, Jakes, R.W, Dalzell, N, Wareham, N, Khaw, K.T, Loveridge, N, Beck, T.J, Reeve, J
• Format: Article
• Language: English
• Subjects: Aged; Aging; Biological and medical sciences; Body Weight; Centre of mass; Diseases of the osteoarticular system; Exercise; Female; Femur - physiology; Femur Neck - physiology; Fracture resistance; Hip geometry; Hip structural analysis; Humans; Injuries of the limb. Injuries of the spine; Male; Medical sciences; Middle Aged; Orthopedics; Osteoporosis; Osteoporosis. Osteomalacia. Paget disease; Physical activity; Prospective Studies; Traumas. Diseases due to physical agents
• ISSN: 8756-3282; 1873-2763
• DOI: 10.1016/j.bone.2006.09.021

Abstract Introduction: For a fixed weight, a wider bone of standardised length is stiffer. But moving the cortices away from the centre of mass risks creating structural (elastic) instability, and hip fractures have been postulated to occur as a consequence of buckling of the thinned supero-lateral femoral neck cortex during a fall. We hypothesised that stereotyped physical activity (e.g., walking) may help conserve bending resistance (section modulus, Z ) through redistribution of bone tissue, but it might be at the expense of supero-lateral cortical stability. Methods: Hip structural analysis (HSA) software applied to DXA scans was used to derive measurements of section modulus and distances of a cross-section's centre of mass from the supero-lateral cortical margin (lateral distance, in cm). DXA scans were obtained on 1361 men and women in the EPIC-Norfolk population-based prospective cohort study. Up to 4 repeat DXA scans were done in 8 years of follow-up. Weight, height and activities of daily living were assessed on each occasion. A detailed physical activity and lifestyle questionnaire was administered at baseline. The lateral distance was measured on three narrow cross-sections with good precision: narrow neck (NN, coefficient of variation 2.6%), intertrochanter (IT) and shaft (S). A linear mixed model was used to assess associations with predictors. Results: Ageing was associated with medial shifting of the centre of mass, so that lateral distance increased. Both greater weight and height were associated with greater lateral distance ( P < 0.0001). Among physical activity-related variables, walking/cycling for > 1 h/day ( P = 0.025), weekly time spent on moderate impact activity ( P = 0.003), forced expiratory volume in 1 s (NN and IT, P < 0.026) and lifetime physical activity (IT, P < 0.0001) were associated with higher lateral distance. However, after adjusting for these variables, activities of daily living scores (NN, P < 0.0001) and weekly time spent on low impact hip flexing activities were associated with shorter lateral distance ( P = 0.001). Greater baseline lateral distance was significantly associated with increased risk of subsequent hip fracture ( n = 26) in females ( P < 0.05, all regions) independently of age, height and bone mineral content. Conclusion: The age-related shift medially of the centre of mass of the femoral neck and trochanter may have adverse effects on fracture resistance in the event of a fall, so compromising the b