MON-167 Hypoglycemia Prevention: A Quality Improvement Project to Reduce Inpatient Hypoglycemia Rate Following Treatment of Hyperkalemia

Background: The use of fixed doses of intravenous insulin and dextrose for hyperkalemia treatment is well-established. Iatrogenic hypoglycemia as a result of treatment although common, can have negative consequences. Literature reveals various practices to minimize hypoglycemia risk. Aim: Our aim was to minimize hypoglycemia risk by modifying the original hyperkalemia protocol that included Point of Care(POC) glucose at 0 and 60 minutes after insulin/dextrose, and educating providers on factors that increase hypoglycemia susceptibility, with the goal to reduce hypoglycemia by a percent-decrease of 10 within 3 months. Methods: We revised the hyperkalemia protocol to include hourly blood glucose monitoring at 0, 60, 120, and 180 minutes, allowing for additional dextrose administration. We educated providers and nurses about hyperkalemia management options including increased caution when treating patients with K<6.0, low baseline POC glucose, and low EGFR. We completed a 3-month pre-intervention retrospective analysis, and a 3-month post-intervention analysis and characterized patients based on the above factors. Fisher’s exact test was used to compare categorical variables and Wilcoxon rank sums was used for continuous variables. Results: Pre-intervention, 8.93% of treatments resulted in hypoglycemia, versus 7.55% post-intervention (p=0.807), which is a 15.5 percent-decrease post-intervention. The hourly-monitoring intervention revealed that 6 of 8 hypoglycemic events occurred after 120 minutes. Median EGFR amongst hypoglycemic patients was 20.8 pre-intervention, and 19.4 post-intervention (p=0.311). Among hypoglycemics, median baseline (time 0) glucose was 102 mg/dL pre-intervention versus 99 mg/dL post-intervention(p=0.563). Patients who did not become hypoglycemic had median glucose of 136 mg/dL pre-intervention and 138 mg/dL post-intervention(p=0.929). Conclusion: The decrease in hypoglycemia rate, after hourly monitoring and provider education, while not statistically significant, is clinically meaningful. Our study demonstrates that most hypoglycemic events occur in patients with low baseline POC glucose, low EGFR, and can occur beyond one hour after hyperkalemia treatment. With this change, patients are systematically treated in a controlled setting, instead of rescued when symptomatic. We recommend providers be cognizant of these factors and use hourly-monitoring for both prediction and prevention of hypoglycemia. References: (1) Apel J, et al. Hypoglycemia in the treatment of hyperkalemia with insulin in patients with end-stage renal disease.&nbsp;Clinical Kidney Journal. 7.3 (2014)