Article Text
Statistics from Altmetric.com
Sir,In their review of hyponatraemia, Gill and Leese1 identify pseudohyponatraemia as a methodological problem which can lead to a spuriously low sodium. However, their assertion that this is not a problem when an ion-selective electrode (ISE) is used for sodium analysis needs clarification.
Two types of methods using ISEs are currently in use within clinical laboratories. In direct ISE methods the specimen is brought to the electrode surface without dilution and inindirect ISE methods the specimen is diluted with buffer prior to electrode contact. As the problem of pseudohyponatraemia is related to the dilution of specimen, the more commonly encountered indirect ISE methods will suffer the same error as the older flame photometer2 technique.
This problem is explained by the solvent exclusion effect. Sodium is only distributed in the aqueous phase of a plasma specimen and if the non-aqueous fraction increases, for whatever reason, then a dilutional effect will result. The degree of error will be proportional to the volume occupied by the non-aqueous portion of the specimen, and both lipids and protein will contribute.
A question we quite often encounter in the laboratory relates to the magnitude of likely error. When we had access to a direct reading ISE which produced comparable results to our routine indirect reading ISE (mean difference +0.9 mmol/l, range − to +5 mmol/l,n=19, p=0.11 ) we briefly investigated the problem. Four specimens with visible lipaemia gave a mean difference of − 4.8 mmol/l, range − 4 to − 5 mmol/l. Five specimens with raised total protein (greater than 80 g/l) gave a mean difference of − 6.8 mmol/l, range − 3 to − 15 mmol/l. The degree of error observed is in broad agreement with a correction table which has been published.3
In summary, indirect reading ISE devices are in common usage and are subject to pseudohyponatraemia. When pseudohyponatraemia is suspected, discussion with the local laboratory and access to a direct reading ISE (sometimes an integral part of blood gas instrumentation) will be of value.