Standard Calcium Supplementation May Increase Kidney Stone Risk: A Study in Women with Postmenopausal Osteoporosis
MetadataShow full item record
INTRODUCTION: The US prevalence of kidney stones has increased from 2.6% in 1972 to 8.4% in 2010.The majority of stones contain calcium (Ca) with hypercalciuria (hCa) highly associated with stone formation. Postmenopausal osteoporosis (pmo) is a common problem affecting 30% of postmenopausal women in the US. Supplementation with Ca and vitamin D (D) is recommended for most older women. PURPOSE: To determine the percentage of patients (pts) who become hypercalciuric while receiving Ca and D supplementation for pmo and to identify biochemical predictors for higher risk of developing hCa. METHODS: 84 ambulatory women aged ≥55 yrs in a RT comparing alendronate (ALN) and sustained-release sodium fluoride (SRF) for the tx of pmo received standard Ca and D supplements. 24-hr urinary Ca (uCa) and deoxypyridinoline and serum D, PTH and bone-specific alk. phos. were measured at 0 and 12 mos. We determined the percentage of pts who became hypercalciuric (uCa >250 mg/24h) during tx. to identify predictors of hCa. Changes in biochemical variables were assessed with mixed model repeated measures analysis. Logistic regression analysis was used to assess predictors of elevated uCa and construct receiver operating characteristic (ROC) curves. RESULTS: 42 pts were randomized to ALN and 42 to SRF. 67 pts completed ≥ one yr. 90% (27/30) of the ALN group and 92% (33/36) of the SRF group had normal uCa excretion at baseline. Patients with normal uCa at baseline experienced significant increases in uCa in the first year (ALN p=0.01, SRF p<0.0001). However, baseline hypercalciurics experienced no significant increase in uCa from baseline after Ca and D supplementation. In all, 13% (4/30) of ALN pts became hypercalciuric (p= 0.41) vs. 28% (10/36) in the SRF group (p=0.002). The best-fit multi-variable model determined baseline uCa (p=0.02) and D (p=0.03) were strong predictors of hCa at 12 mo. and produced a favorable ROC curve (0.90). Baseline uCa was a consistently strong predictor of hCa and a simple logistic regression analysis generated a ROC curve (0.84) which determined that 180 mg/d uCa at baseline was a strong predictive cut-point for detection of pts at higher risk of hCa with treatment. CONCLUSION: 21% of patients became hypercalciuric on recommended doses of Ca and D. Current Ca and D supplementation practice may have significant public health consequences by contributing to the growing incidence of nephrolithiasis. Practice guidelines should consider assessing baseline 24-hr uCa in all pts and 12 mo. 24-hr uCa in pts with baseline uCa ≥of 180 mg/24h.