Clinically significant effect of protein concentration on ion-selective electrode measurements of ionised calcium.

Others, using an Orion SS-20 ionised calcium analyser, noted that the ionised calcium concentration of a native serum sample was 8% greater than that of its ultrafiltrate. The experiments described here, using a Nova 2 ionised calcium analyser, confirmed a positive protein interference which was greater for human albumin than for IgG. Uncharged dextran showed no positive interference but dextran sulphate, which is highly charged and binds calcium, showed a large effect. Thus the interference is related to macromolecular charge. Dialysis experiments with normal and pathological human serum samples indicated that the ionised calcium of diffusible plasma water was overestimated by an average of 9.6% at an albumin concentration of 40 g/l and by 4.8% at 20 g/l. It is concluded that the measurement of ionised calcium with existing analysers can be clinically misleading in patients with abnormal plasma proteins. Hypocalcaemia is likely to be overdiagnosed and hypercalcaemia underdiagnosed in the presence of hypoalbuminaemia.