| The modern diet is purported to induce osteoporosis due to diet acid-ash, net acid excretion, altered acid-base balance and finally bone demineralization.;Objectives. The literature was systematically reviewed. Associations between urine measures of dietary acid load (pH, sodium, potassium, calcium, magnesium, phosphate, chloride, and sulfate, measured, and organic acids, calculated, in fasting morning urine samples (n = 795) were examined with the bone outcomes ((changes in bone mineral density (BMD) (femoral neck, lumbar spine, and total hip) over 5 years) and fractures over 7 years). Stability of the urine measures were assessed over 5-years.;Methods. A systematic review was undertaken based on Hill's criteria of causation. Fasting urine samples and DEXA BMD were measured at baseline and at 5-years among the Canadian Multicentre Osteoporosis cohort Study. Multiple linear (BMD) and logistic (fractures) regression analysis were undertaken controlling for potential confounders eg: age, gender, family history of osteoporosis, physical activity, smoking, calcium intake, vitamin D status, hormonal status, medications, renal function, urine creatinine, change of body mass index. Intraclass correlation coefficients (ICC) described the stability of the urine measures.;Results. The limitations of the literature included: primary outcome was urinary calcium, which may be confounded by absorption, and results of cohort studies are inconsistent. The internal consistency was dubious as a consequence of the conflicting roles of phosphate, sodium, and milk, and there is no mechanism described that would function to demineralize bone at physiological pH.;There was no association between urine pH and either the change of BMD or the incidence of fractures over 7 years. Among 11 measured associations between urine measures of dietary acid load and BMD at different sites and fragility fractures, only low urine potassium was associated with increased BMD in the lumbar spine over 5-years, percent explained variance = 4%. Urine potassium was not associated with change at the other bone sites. The ICCs for the urine measures of diet acid load were 0.08 to 0.64.;Conclusion. The finding of only one weak association between urine potassium and the change of spine BMD, and the systematic review, provided little support for the acid-ash hypothesis. It is unknown if this limited evidence was due to the absence of an association or the limitation of fasting urine samples to be assessed for diet acid load. |
