Hyperglycemic Crises in Adult Patients With Diabetes: Response to Yasuda et al.

Diabetic ketoacidosis (DKA) is a state of severe diabetic decompensation characterized by uncontrolled hyperglycemia, ketosis, and acidemia. Until recently, the most widely used diagnostic criteria for DKA were a blood glucose >250 mg/dl, an arterial pH <7.30, serum bicarbonate <15 mEq/l, and moderate degree of ketonemia and/or ketonuria (1). Accumulation of ketoacids usually results in an increased anion gap metabolic acidosis. Although these criteria for DKA served well for research purposes, they were somewhat restrictive for clinical practice. For example, the majority of patients admitted with the diagnosis of DKA are alert and present with a mild to moderate degree of metabolic acidosis (bicarbonate 10 to 18 mEq/l and pH 7.10 to <7.40) compared with patients with severe DKA (bicarbonate <10 mEq/l and pH <7.00) who frequently present with mental obtundation and coma. In many patients with DKA, admission plasma glucose does not correlate with the severity of metabolic acidosis but β-hydroxybutytrate and bicarbonate do correlate well (2). These facts lead to more recent diagnostic criteria outlined in the American Diabetes Association Technical Review and Position Statement on the management of hyperglycemic crises in adult patients with diabetes (1,3) aiming to provide health care workers with more clinically appropriate diagnostic guidelines, as well as guidance on patient disposition and choice of therapy.

Yasuda et al. (4) make an important point about the complexity of acidosis in DKA. The presence of other acid-base disturbances (metabolic alkalosis, hyperchloremic acidosis, and lactic acidosis) and differences in respiratory compensation in patients with DKA introduces variability in pH, bicarbonate, and anion gap that has been acknowledged (2). However, it is desirable to have objective diagnostic and severity criteria. In our view, it is not practical to simply exclude these tests in patients with double and triple acid-base disturbances because there are no quantitative criteria to take their place. Serum β-hydroxybutyrate avoids some of the pitfalls with the less-specific pH, bicarbonate, and anion gap tests, but it is not available except in a few specialized centers. We recommend that clinicians use the criteria provided, mindful of their finite sensitivity and specificity, and be aware of the confounding effects of other acid-based disturbances and variable respiratory compensation.