RESUMEN
[K+] < 3.5 mmol/L is reported to occur in approximately 4 % of patients with diabetic ketoacidosis (DKA.) Therefore, the American Diabetes Association (ADA) and Joint British Diabetes Societies (JBDS) recommend the assessment of [K+] before the initiation of insulin treatment to avoid the precipitation of morbid hypokalemia. The purpose of this study was to assess the incidence of hypokalemia in patients presenting to the emergency department (ED) with DKA. This was a multicenter retrospective, cross-sectional study at EDs with a combined annual adult census of 155,000. Adult patients diagnosed with DKA in the ED, or who were admitted from the ED and subsequently diagnosed with DKA as determined from the hospital electronic database between January 2008 and December 2008, were included for analysis if they had the following initial laboratory values: (1) serum glucose >13.9 mmol/L (250 mg/dL), (2) serum bicarbonate <18 mmol/L (18 mEq/L) or anion gap >15, and (3) evidence of ketonaemia or ketonuria. 537 patients were diagnosed with DKA in the ED at the participating institutions during the reference period. The median [K+] was 4.9 mmol/L (IQR 4.3, 5.5). There were a total of seven patients with an initial 3.3 < [K+] < 3.5 mmol/L, but none with a [K+] < 3.3 mmol/L. Thus, no patients in our study sample required potassium supplementation before the initiation of insulin treatment. The incidence of hypokalemia in our sample of patients with DKA was much less than previously reported, with no cases requiring potassium supplementation before insulin administration.
