Plasma Acid-Base Patterns in Diabetic Ketoacidosis

In a study of the types of plasma acid-base patterns present at 196 admissions for diabetic ketoacidosis we found no relation between the initial level of serum total carbon dioxide and the plasma anion gap; instead, there was a broad spectrum of acid-base patterns, ranging from pure anion-gap acidosis to pure hyperchloremic acidosis. Although the degree of renal dysfunction on admission, which reflected the magnitude of volume depletion, was independent of the severity of metabolic acidosis, it was responsible for the variable retention of plasma ketones: the more severe the volume depletion on admission, the greater the ketone retention and the less prominent the hyperchloremic acidosis. Recovery from acidosis was significantly slower in patients admitted with pure hyperchloremic acidosis. After therapy, hyperchloremia developed in most patients at four to eight hours after admission, because of the retention of chloride in excess of sodium and the excretion of ketones by the kidney. (N Engl J Med. 1982; 307:1603–10.) IN diabetic acidosis, protons from the ketoacids that are produced titrate bicarbonate and other buffers while the ketone anions accumulate in plasma. This process leads to an increase in the unmeasured anions in plasma and is responsible for an increased plasma anion gap. 1 2 3 According to this classic concept, confirmed in several studies in which mean values were reported, the increment in the plasma anion gap in this condition should be approximately equal to the deficit in plasma bicarbonate. 4 5 6 7 8 9 10 11 However, when we evaluated individual patients admitted to the hospital because of diabetic ketoacidosis, we often failed to observe this expected equivalence . . .