Sodium measurements in the presence of paraproteins by four direct ISE methods and flame photometry compared.

Sodium was determined by flame photometry and by direct potentiometry using four different systems (Ektachem DT60, Corning 614, Orion 1020 and Nova-1) in serum samples from patients with normal proteins and paraproteins. Flame photometric and direct potentiometric differences (ISE-flame), showed the expected relationship with total protein whether or not the sample contained paraprotein, an exception being the Ektachem DT60 on normal protein. With the exception of the Ektachem DT60 with samples with normal protein levels, correlation between different ISE-flame estimates was good. Differences between direct potentiometric measurements are probably more attributable to efforts made by manufacturers to relate them to flame photometric measurements in 'normal samples', than to the volume occupied by increased protein content. The apparent lack of influence of protein on ISE-flame for the Ektachem DT60 was shown to relate to the original external reference electrode salt bridge used. A second study showed that changing this fluid to an equi-transferrant high salt content caesium chloride based reference fluid partially unmasked the protein effect.