Urine calcium and urine sodium concentrations are not related, after adjustment for urine magnesium.

BACKGROUND AND OBJECTIVES: Urine calcium correlates with urine sodium. The aims of this study were to investigate whether the urine sodium-calcium relationship persists into old age and whether it holds after adjustment for urine magnesium. DESIGN: Cross-sectional descriptive analysis. PATIENTS: Residents of two aged care institutions (median age 84 years) who were not taking diuretics, calcium or vitamin D supplements. MEASUREMENTS: Early morning urine calcium, sodium and magnesium, plasma creatinine and serum 25-hydroxyvitamin D and parathyroid hormone. RESULTS: Urine calcium correlated with urine sodium (r = 0.29, P < 0.01) and with urine magnesium (r = 0.56, P < 0.001). After adjustment for urine magnesium, the relationship between urine sodium and urine calcium was no longer significant. Forty-five percent of the interindividual variation in urine calcium was explained by a linear model on the basis of urine magnesium and plasma creatinine. CONCLUSION: The data indicate that a correlation between urine sodium and calcium persists in very old age. However, this correlation no longer holds after adjustment for urine magnesium. Further studies examining urine calcium excretion should also consider urine magnesium.

Comparison of a radioimmunoassay and a microbiological assay for measurement of serum vancomycin concentrations.

A newly developed radioimmunoassay was used to measure the concentration of vancomycin in 137 specimens of serum from patients being treated with this antibiotic. Of these sera, 84 were also analyzed with a microbiological assay technique for vancomycin. Duplicate determinations were done with each of the techniques. Individual values and averaged values for both methods were used for statistical analyses. The correlation coefficients between all possible combinations of radioimmunoassay and microbiological assay results for the 84 sera were greater than or equal to 0.99 (P less than 0.01). Values for the regression coefficients of radioimmunoassay results on microbiological assay results ranged from 0.98 +/- 0.01 to 1.03 +/- 0.01. The mean percent deviation of radioimmunoassay versus microbiological assay results was -1.56 +/- 0.60. A one-way analysis of variance demonstrated that the use of different standard curves for each batch of specimens assayed by microbiological assay did not significantly influence the results (P = 0.07). The microbiological assay and the radioimmunoassay for measurement of serum vancomycin levels yielded essentially identical results.