Improving the Safety of Insulin Treatment for Hyperkalemia

Document Type

Conference Proceeding

Publication Date


Publication Title

Ann Emerg Med


Study Objectives: This study’s objective was to determine key clinical predictors of hypoglycemia following insulin therapy for the emergent management of hyperkalemia and derive a safe insulin treatment strategy. Methods: We performed a retrospective observational study across 4 hospitals that was inclusive of consecutive ED and hospitalized adults that received insulin for the management of hyperkalemia over a 12-month period. We excluded patients treated in the setting of cardiac arrest. The primary outcome was hypoglycemia (glucose < 70 mg/dL) following treatment. We performed multivariate logistic regression to determine clinical predictors of hypoglycemia. We then tested the pre-treatment glucose-insulin ratio (glucose divided by planned weight-based insulin dose) and tested its ability to alert clinicians to the risk of treatment related hypoglycemia. Results: The study included 1,307 patients, of whom 507 were in the ED and 800 were inpatient. The mean pre-treatment K was 6.1 (SD 0.76). Hypoglycemic events occurred 238 times (18.4%). Patients in the ED had the highest rate of hypoglycemic events (23.3%) compared to hospitalized patients (15.3%, p<0.001). Patient’s with a pre-treatment glucose < 100 mg/dl had 31.0% the rate of hypoglycemia compared to those with higher glucose levels (13.7%, p<0.001). Adjusting for multiple clinical covariates, male sex (OR 1.4, 95% CI 1.1 - 2.0), insulin dosing > 0.1 units/kg (OR 1.5, 95% CI 1.1 - 2.1), and pre-treatment glucose < 100 mg/dl (OR 2.4, 95% CI 1.8 - 3.2) were significant predictors of hypoglycemia. The median glucose-insulin ratio was 1310 [IQR 915, 2025]. A ratio < 1310 identified 71.9% of hypoglycemic events (OR 3.0, 95% CI 2.2 - 4.1). Conclusion: Male sex, insulin dose, and pre-treatment glucose are predictors of hypoglycemia in the treatment of hyperkalemia. Attention to the simple ratio of the pre-treatment glucose to weight-based insulin dose identifies most patients at risk of hypoglycemia and can alert clinicians to adjust glucose and insulin administration proactively.





First Page


Last Page


This document is currently not available here.