Drug-induced hyponatraemia

Editor, – The useful review of drug-induced hyponatraemia (Aust Prescr 2003;26:114-7), states that 'blood glucose concentrations above 20 mmol/L can spuriously reduce the serum sodium concentration when measured by flame photometry'. This mistakenly implies that hyperglycaemia produces method-dependent pseudohyponatraemia of the type seen with marked hypertriglyceridaemia.

The hyponatraemia of marked hyperglycaemia is dilutional, from osmotic movement of water from the intracellular space, and is independent of method.¹ The measured serum sodium concentration is analytically valid, but needs to be corrected before relating the value to the normal reference interval. A useful correction, derived from a formula originally given in metric units, is to add a third to a half of the glucose excess in mmol/L to the measured serum sodium concentration.²

Apparent hyponatraemia is a reassuring finding in severely hyperglycaemic dehydrated patients, as the serum sodium concentration, when corrected as above, is often close to normal. An apparently normal serum sodium concentration without correction implies hypernatraemia and indicates a water deficit much larger than the salt deficit. Severe hypernatraemia can then be anticipated during resuscitation with isotonic sodium chloride, especially if hyperglycaemia is rapidly corrected.

J. R. Stockigt
Professor of Medicine, Monash University
Department of Endocrinology and Diabetes
Alfred Hospital
Prahran, Vic.

Editor, – I found the article on hyponatraemia (Aust Prescr 2003;26:114-7) interesting. I have had several elderly patients who have developed severe hyponatraemia while on tramadol which has been corrected on its cessation. Tramadol is thought to inhibit reuptake of serotonin which causes increased serotonin levels and presumably causes hyponatraemia due to a similar mechanism to the selective serotonin reuptake inhibitors. Tramadol was not included in Table 3 'Drugs commonly associated with hyponatraemia' and it is not listed in the product information as an adverse effect. I would appreciate the authors' comments.

Robin Hunter
Rehabilitation Physician
Brighton, Vic.

Dr S. Fourlanos and Dr P. Greenberg, the authors of the article, comment:

Professor Stockigt correctly indicates a factual error in our paper. We agree with him that the hyponatraemia associated with marked hyperglycaemia is dilutional, not method-dependent.

We note Dr Hunter's comment with interest. We were unaware of any association between tramadol and hyponatraemia, however the Adverse Drug Reactions Advisory Committee has received 12 reports.