To the Editor.—
Paraproteinemias, bromism, and severe hypermagnesemia are uncommon but life-threatening disorders that unequivocally and often strikingly reduce the anion gap.1,2 Hyponatremia, hypoalbuminemia, and random laboratory error occur far more frequently and also lower the anion gap.2,3 Most physicians would comfortably predict that any patient with a persistently low anion gap most likely has either a low serum sodium or low serum albumin concentration.Goldstein et al (1980;243:1737) reviewed 67,740 sets of electrolytes—ie, they assayed more than 0.25 million ions and proved that hyponatremia, hypoalbuminemia, and something less than perfect results in routine electrolyte analyses were more commonly responsible for a low anion gap than were myeloma, bromism, and hypermagnesemia. No one denies that the latter three diseases are uncommon, but by the same token, no one can deny that these uncommon diseases usually lower the anion gap. Contrary to the title of the authors' work, there