Comparison of serum total calcium, albumin-corrected total calcium, and ionized calcium in 1213 patients with suspected calcium disorders.

The correlations between serum ionized calcium, serum total calcium, total calcium corrected for albumin and calculated ionized calcium were investigated in a prospective multicentre investigation of 1213 patients suspected of having calcium metabolic disease. Diagnostic discordance between serum total calcium and measured ionized calcium was found in 31% of the patients. With the calculation of albumin-corrected total calcium or calculated ionized calcium the discordance decreased to 17.9%. The diagnostic discordance which could be ascribed to the analytical imprecision (CV = 1.5%) amounted to only 6.7%. Although we found highly significant correlations between the parameters, a considerable scatter around the regression line made prediction of ionized calcium from albumin-corrected total calcium unreliable in many patients.