["Corrected" calcium: calcium status underestimation in non-hypoalbuminemic patients and in hypercalcemic patients].

It is often reminded that if the ionized calcium is not measured, the interpretation of total calcemia should consider serum albumin. Two formulas are usually employed: ["Corrected" Ca (mmol/L) = Ca measured (mmol/L) + 0.020 or 0.025 (40 - albumin (g/L))]. This adjustment formula arises from works of Payne published in 1973. In a control population, we established the median values of calcium, albumin and ionized calcium (corrected to pH 7.40), respectively 2.34 mmol/L, 45.7 g/L and 1.23 mmol/L with our laboratory's methods (albumin - bromocresol green and Ca - ortho-cresolphtalein on a Modular analyser, Roche Diagnostics; ionized calcium with ion-selective electrode, Radiometer SA). Based on this, we retrospectively compared for 71 patients who do not belong to the control population the "corrected calcium" resulting from the two formulas and the measured calcemia to the ionized calcium corrected at pH 7,40. This comparison shows that in our laboratory, the two formulas lead to a rising underestimation of the calcium for albumin values greater than 40 g/L, reaching -0,20 mmol/L for albumin values above 44 g/L. The use of this formulas may also mask an hypercalcemia, indeed half of our patients' hypercalcemia (ionised Ca ((pH 7,40)) > 1,29 mmol/L) is not found. These results agree with Payne's recommendations for the use of his adjustment formula: the clinically justified adjustment of a low calcemia due to an hypoalbuminemia should not be extended to other situations, particularly when albumin is increased.