Associations between age-related changes in bone microstructure and strength and dietary acid load in a cohort of community-dwelling, healthy men and postmenopausal women

Abstract

The importance of dietary acid load (DAL) in the pathogenesis of osteoporosis is still debated. Age-related changes in bone microstructure and strength in relation to DAL remain largely unexplored. We investigated the associations between changes in areal and volumetric bone mineral density (BMD), bone microstructure and strength, fracture risk, and DAL in a prospective cohort of 65-y-old healthy men and postmenopausal women. Potential renal acid load (PRAL; mEq/d) was calculated as a DAL proxy to characterize participants' diet as alkaline (Alk-D; PRAL < -5), neutral (Neut-D; -5≤PRAL≤5), or acidic (Acid-D; PRAL >5). We measured areal BMD (aBMD) by DXA, and distal radius and tibia bone microstructure using high-resolution peripheral quantitative computed tomography, at baseline (n=853) and after 6.1±1.4 y (n=708). Bone strength was estimated using finite element analyses at baseline and after 3.0±0.5 y (n=613). Prevalent and incident fractures were recorded. The majority of the participants (59%) had an Alk-D, while 23% had a Neut-D, and 18% an Acid-D. Baseline aBMD and bone microstructure and strength did differ or were slightly better in women or men with an Acid-D versus those consuming an Alk-D or Neut-D. Indeed, women with an Acid-D had higher trabecular number (P=0.010 vs. Alk-D; P=0.001 vs. Neut-D), while men had higher hip and radius aBMD (P=0.008 and 0.024 vs. Neut-D, respectively) and radius strength (P=0.026 vs. Neut-D). Over the follow-up, women in the Acid-D group experienced lower cortical and endocortical bone loss at the radius than did the Alk-D and Neut-D groups (cortical thickness, P=0.008 and<0.001; trabecular area, P=0.001 and<0.001, respectively). No association between fractures and PRAL was observed. These null or favourable associations between baseline values or changes in aBMD, bone microstructure and strength, and DAL in this cohort of 65-y-old healthy individuals do not support adverse DAL-mediated effects on bone. This trial was registered at http://www.isrctn.com as ISRCTN11865958.