Stewart (physicochemical) approach versus conventional anion gap approach for resolution of metabolic acidosis in diabetic ketoacidosis

Abstract

Diabetic ketoacidosis (DKA) frequently requires emergency admission. The anion gap approach is conventionally used for the diagnosis and documenting the resolution of acidosis during treatment. However, it fails to detect hyperchloremic acidosis during the resolution and may result in the prolongation of treatment. To determine the role of the Stewart approach of acid-base disorder during DKA management for the prediction of an earlier resolution. A prospective comparative study was conducted between January 2017 and December 2017 at a single academic hospital in north India. Patients aged above 12 years with a diagnosis of DKA were randomly divided into two groups—the conventional group and the Stewart group, according to the approach used for DKA resolution. The primary outcome was the time duration required for resolution. The secondary outcomes were the therapeutic requirement of intravenous fluid, insulin, and potassium, Acute Physiology and Chronic Health Evaluation II (APACHE II) score at the time of resolution, and hospital stay. Forty-four DKA patients were equally distributed in the two groups with comparable baseline parameters. The Stewart group had early resolution of DKA (mean, 32.4±17.5 h versus 41.7±19.6 h; p value <0.001) at similar APACHE II scores. The duration of hospital stay was reduced but was not statistically significant (mean, 5.6±3.2 days versus 7.0±3.8 days; p value 0.16). The therapeutic requirement of fluid, insulin, and potassium was similar in groups. The Stewart approach may be a better alternative to the conventional anion gap approach for guiding the resolution of DKA.