RESUMO
Multiple studies have observed a relationship of bone mineral density (BMD) measured by Dual energy X-ray absorptiometry (DXA) and mortality. However, areal BMD (aBMD) measured by DXA is an integrated measure of trabecular and cortical bone and does not measure the geometry of bone. Peripheral Quantitative Computed Tomography (pQCT) provides greater insights on bone structure, geometry and strength. To examine whether higher bone phenotypes and muscle density as measured by pQCT are associated with a lower all-cause mortality, we studied 245 men and 254 women (all age >60) recruited in the Mobility and Independent Living among Elders Study in rural south India. Cox proportional hazards models estimated hazard ratios (HR [95% Confidence Intervals]). After an average follow-up of 5.3 years, 73 men and 50 women died. Among men, trabecular volumetric bone mineral density (vBMD) of radius (HR per SD increase in parameter = 0.59 [0.43, 0.81]) and tibia (0.60[0.45, 0.81]), cortical vBMD of radius (0.61, [0.47, 0.79]) and tibia (0.62, [0.49, 0.79]), cortical thickness of radius (0.55, [0.42, 0.7]) and tibia (0.60, [0.47, 0.77]), polar strength strain index (SSIp) of tibia (0.73 [0.54, 0.98]), endosteal circumference of radius (1.63, [1.25, 2.12]) and tibia (1.54, [1.19, 1.98]) were associated with all-cause mortality. Muscle density (0.67, [0.51, 0.87]) was associated with lower mortality in men. Among women cortical vBMD of radius (0.64, [0.47, 0.87]) and tibia (0.60 [0.45, 0.79]), cortical thickness of radius (0.54, [0.37, 0.79]) and tibia (0.43, [0.30, 0.61]), SSIp of radius (0.70 [0.48,1.01]) and tibia (0.58 [0.37, 0.90]) and endosteal circumference of radius (1.33 [0.97, 1.82]) and tibia (1.83, [1.37, 2.45]) were associated with all-cause mortality. Among men, gait speed mediated the association of muscle density and mortality but there was no mediation for any bone parameters. Conclusion: pQCT bone measures and muscle density were independently associated with mortality among rural south Indian elders.
