Because kidney dysfunction reduces the ability to excrete dietary acid excess, we hypothesized that underlying kidney function may have confounded the mixed studies linking dietary acid load with the risk of osteoporosis and fractures in the community. In a relatively large survey of elderly men and women, we report that dietary acid load did neither associate with DEXA-estimated bone mineral density nor with fracture risk. Underlying kidney function did not modify these null findings. Our results do not support the dietary acid-base hypothesis of bone loss.
Impaired renal function reduces the ability to excrete dietary acid excess. We here investigate the association between dietary acid load and bone mineral density (BMD), osteoporosis, and fracture risk by renal function status.
An observational study was conducted in 861 community-dwelling 70-year-old men and women (49 % men) with complete dietary data from the Prospective Investigation of the Vasculature in Uppsala Seniors (PIVUS). The exposure was dietary acid load as estimated from 7-day food records by the net endogenous acid production (NEAP) and potential renal acid load (PRAL) algorithms. Renal function assessed by cystatin C estimated glomerular filtration rate was reduced in 21 % of the individuals. Study outcomes were BMD and osteoporosis state (assessed by DEXA) and time to fracture (median follow-up of 9.2 years).
In cross-section, dietary acid load had no significant associations with BMD or with the diagnosis of osteoporosis. During follow-up, 131 fractures were validated. Neither NEAP (adjusted hazard ratios (HR) (95 % confidence interval (CI)), 1.01 (0.85-1.21), per 1 SD increment) nor PRAL (adjusted HR (95 % CI), 1.07 (0.88-1.30), per 1 SD increment) associated with fracture risk. Further multivariate adjustment for kidney function or stratification by the presence of kidney disease did not modify these null associations.
The hypothesis that dietary acid load associates with reduced BMD or increased fracture risk was not supported by this study in community-dwelling elderly individuals. Renal function did not influence on this null finding.
2015. Vol. 26, no 2, 563-570 p.