NAGMA in DKA patients from ketone excretion

The patient presents with diabetic ketoacidosis (DKA), evidenced by metabolic acidosis with increased anion gap (AG), increased serum glucose level, and increased β-hydroxybutyrate level. With volume and insulin infusions, the AG closes, but the patient now has a normal AG (NAG) metabolic acidosis, also called a nongap metabolic acidosis, owing to loss of bicarbonate bound to ketoacid anions in the urine. This progression is common in the resuscitation for DKA, especially when 0.9% saline is used as the principal fluid for volume resuscitation.

Insulin deficiency in DKA results in lipolysis and generation of the ketoacids acetoacetic acid and β-hydroxybutyric acid, as well as acetone. The AG is increased on an equimolar basis with the presence of ketoacids in the serum. The protons from these ketoacids also titrate bicarbonate on an equimolar basis so that the delta AG-delta bicarbonate ratio is 1:1. Insulin infusion results in the functional metabolism of ketoacids to bicarbonate. If all of these ketoacid ions remained within the bloodstream, then the bicarbonate lost during DKA would be completely regenerated. Because the ketoacid anions are filtered but not reabsorbed, large amounts of them are lost in the urine, so the serum bicarbonate cannot be completely repleted by the metabolism of the ketoacid ions that remain. Even when all the ketones have been metabolized to bicarbonate and the AG has closed, the bicarbonate level remains low, with the result being continued acidosis. If ketone clearance was incomplete, the AG would persist; the normal AG in this case rules out incomplete ketone clearance. 

The patient is pregnant, as evidenced by her increased β-human chorionic gonadotropin, consistent with the first trimester. Although pregnancy-related stress can predispose a patient to ketoacidosis, ketone shifts across the placenta have not been described as a source of normal AG acidosis.

In both diabetic and alcoholic ketoacidosis, the predominant ketone body is β-hydroxybutyrate, in ratios as high as 10:1 compared with acetoacetate. Acetoacetate and β-hydroxybutyrate are, in fact, interconvertible during the course of DKA, but the conversion does not affect either acidosis or the AG. 

Several small randomized trials have suggested that normal AG acidosis may be avoided if a balanced salt solution containing acetate or lactate, rather than 0.9% saline, is used in resuscitation. Normal saline contributes to a minor extent to the hyperchloremia associated with normal AG acidosis. Acetate and lactate are both metabolized to bicarbonate and contribute to restoring the bicarbonate deficit. A retrospective subgroup analysis of patients with DKA from the SMART and SALT-ED trials found that DKA resolved approximately 4 h sooner in patients randomly assigned to a balanced salt solution compared with results in those receiving 0.9% saline.1234

Footnotes

  1. SEEK Questionnaires ↩

  2. Laffel L. Ketone bodies: a review of physiology, pathophysiology and application of monitoring to diabetes. Diabetes Metab Res Rev. 1999;15(6):412-426. PubMed ↩

  3. Oh MS, Carroll HJ, Uribarri J. Mechanism of normochloremic and hyperchloremic acidosis in diabetic ketoacidosis. Nephron. 1990;54(1):1-6. PubMed ↩

  4. Ramanan M, Attokaran A, Murray L, et al; SCOPE-DKA Collaborators and Queensland Critical Care Research Network (QCCRN). Sodium chloride or Plasmalyte-148 evaluation in severe diabetic ketoacidosis (SCOPE-DKA): a cluster, crossover, randomized, controlled trial. Intensive Care Med. 2021;47(11):1248-1257. PubMed ↩