Abstract
INTRODUCTION: Nephrolithiasis is associated with an increased fracture risk, but predictors of bone mineral density (BMD) in stone formers (SFs) remain poorly defined. METHODS: We conducted a retrospective analysis in the Bern Kidney Stone Registry (BKSR), an observational cohort of kidney SFs. Inclusion criteria were age ≥18 years and ≥1 past stone episode. Participants with non-calcium (Ca)-containing kidney stones, a history of primary hyperparathyroidism or antiresorptive or anabolic bone treatment were excluded. Multivariable linear regression analyses were used to assess the association of blood and 24-hours urine parameters and stone composition with BMD at the lumbar spine and femoral neck. RESULTS: In the analysis, 504 participants were included, mean age was 46 years, and 76% were male. In multivariable analyses, fasting (β: -0.031; P = 0.042), postload (β: -0.059; P = 0.0028) and Δ postload - fasting (β: -0.053; P = 0.0029) urine Ca-to-creatinine ratios after 1 week of a sodium- and Ca- restricted diet and Ca oxalate dihydrate stone content (β: -0.042; P = 0.011) were negatively associated with z scores at the lumbar spine. At the femoral neck, alkaline phosphatase (β: -0.035; P = 0.0034) and parathyroid hormone (PTH) (β: -0.035; P = 0.0026) were negatively associated with z scores, whereas 24-hours urine Ca (β: 0.033; P = 0.0085), magnesium (β: 0.043; P = 3.5 × 10(-4)), and potassium (β: 0.032; P = 0.012) correlated positively with z scores at the femoral neck. CONCLUSION: Our study reveals distinct predictors of BMD in SFs. Commonly available clinical parameters, such as kidney stone composition results, can be used to identify SFs at risk for low BMD.