High calcium concentration and calcium oxalate crystals cause significant inaccuracies in the measurement of urinary osteopontin by enzyme linked immunosorbent assay.

Strong evidence that osteopontin (OPN) is a determinant of urolithiasis has prompted studies comparing the protein's urinary excretion in healthy subjects and stone formers. However, reported mean urinary values have varied widely, from <1 microg/mL to more than 20 times that value. Since OPN binds to CaOx crystals, the presence of crystals in urine may cause underestimation of its urinary levels. Using a commercial ELISA, we measured urinary OPN levels in the presence of endogenous or exogenous CaOx monohydrate (COM) and dihydrate (COD) crystals. OPN concentrations decreased in the presence of endogenous and exogenous CaOx crystals, but never below 2 microg/mL. Increasing the urinary calcium concentration decreased detectable OPN levels, possibly as a result of changes in the three-dimensional conformation of the protein. Because calcium concentration and the formation of CaOx crystals cannot be controlled in urine, the use of urinary OPN levels as a biomarker for any human pathology must be seriously questioned, but particularly for the investigation of stone formers in whom hypercalciuria and crystalluria are more common than in healthy subjects.