Current protocols which include the administration of a single dextrose dose concomitantly with insulin are inadequate as hypoglycemia commonly occurs 60 min after insulin administration and may persist for up to two hours post-insulin administration. To prevent delayed hypoglycemic events, our institution revised our adult acute hyperkalemia order set to include hypoglycemic preventative measures not currently described in the literature.

The primary purpose of this retrospective study was to determine if the new adult acute hyperkalemia order set resulted in lower rates of hypoglycemia (glucose <70 mg/dL) compared to the old order set in patients with impaired renal clearance and lower pre-insulin glucose values. In addition to reducing the IV regular insulin dose from 10 to 5 units, the new order set recommends patients receive a 250 mL dextrose 10% solution over two hours in addition to a 50 mL dextrose 50% IV push concomitantly with IV regular insulin if their pre-insulin glucose is ≤250 mg/dL. Patients were included if they were adults, received IV regular insulin from the order set within six hours of presenting to the ED, had a pre-insulin potassium >5.5 mmol/L, had a pre-insulin glucose ≤250 mg/dL, and had impaired renal clearance [creatinine clearance (CrCl) < 30 mL/min or dialysis dependent].

100 patients were included in each arm. The median pre-insulin potassium levels were 6.4 mmol/L and 6.3 mmol/L in the old and new groups, respectively (p = 0.133). The median pre-insulin glucose levels were 120 mg/dL and 107.5 mg/dL in the old and new groups, respectively (p = 0.013). Twenty (20%) patients in the old group developed hypoglycemia, whereas six (6%) patients in the new group developed hypoglycemia (p = 0.003). There was no significant difference between the two groups in number of patients who achieved a post-insulin potassium level ≤ 5.5 mmol/L.

Our study found that our approach of additionally administering a 250 mL dextrose 10% solution upon therapy initiation is associated with significantly lower rates of hypoglycemia. Our findings indicate that hypoglycemia rates can be significantly reduced in vulnerable populations if additional preventative measures are employed.