TO THE EDITOR:
A frequently encountered problem in clinical practice is a patient who presents with acidosis and hyperglycemia. It has been my experience that the correct calculation of the anion gap in the face of hyperglycemia is often confusing. An example would best serve to illustrate the point. Assume a patient who is admitted with new-onset diabetes mellitus and has the following blood test results: glucose level, 700 mg/dL; sodium level, 128 mEq/L; chloride level, 97 mEq/L; and bicarbonate level, 21 mEq/L. The anion gap in this patient is [Na] ?([Bicarbonate] + [Cl]) = 128 ?(97 + 21) = 10, a value within normal limits; the patient has a mild non-anion gap acidosis. However, physicians often correct the sodium level in the face of hyperglycemia by adding 1.6 mEq/L to the sodium concentration for each 100-mg/dL increment in glucose levels above 100 mg/dL. This correction does not apply to the calculation of the anion gap in patients with acidosis and hyperglycemia because the water moving from the intracellular compartment to the extracellular compartment as a result of the hyperglycemia equally dilutes all electrolytes, including the chloride and bicarbonate. If in this case the sodium level is "corrected" for the hyperglycemia, it will be calculated as 138 mEq/L and lead to a falsely elevated calculated anion gap of 20. Thus, the patient’s condition would be erroneously diagnosed as severe anion gap acidosis, most probably diabetic ketoacidosis.
Tomer, Y. Annals of Internal Medicine 129:9 p753