Lower vs. Standard Insulin-Dextrose Doses for Treating Mild to Moderate Hyperkalemia in the Emergency Department

Overview

This clinical trial aims to evaluate the safety and efficacy of a lower dose insulin compared to the conventional dose for treating mild to moderate hyperkalemia, a condition characterized by elevated serum potassium levels below 6.4 mmol/L. The study focuses on adult patients in the emergency department. The primary objective is to determine whether the efficacy of lower doses of insulin and dextrose is non-inferior to conventional doses in reducing serum potassium levels in patients with mild to moderate hyperkalemia. The research compares two regimens: 5 units of insulin with 25 mL of dextrose 50% versus 10 units of insulin with 50 mL of dextrose 50%, assessing the mean reduction in serum potassium levels, the incidence of hypoglycemia, and the risk of extravasation injury. Participants will be monitored in the emergency department for 6 hours. Serum potassium levels will be measured at 1, 2, 4, and 6 hours post-intervention, while blood glucose levels will be monitored at 30, 60, 90, 120, 240, 300, and 360 minutes post-intervention.

The primary research question that guided the design of this study arose from the consideration of optimizing insulin and dextrose dosing in the management of mild to moderate hyperkalaemia, specifically in patients with serum potassium levels below 6.4 mmol/L. This focus acknowledges that severe hyperkalaemia necessitates higher insulin doses, often accompanied by a corresponding increase in dextrose, to achieve a rapid and significant reduction in serum potassium levels. However, the necessity of administering the full conventional dose of insulin and dextrose in moderate hyperkalaemia is still unclear. Investigators hypothesize that a lower dose regimen may achieve similar efficacy in reducing serum potassium levels while potentially minimizing the risks associated with higher doses, such as hypoglycaemia and extravasation injury. Previous studies have reported reductions in potassium ranging from 0.6-1.17 mmol/L following administration of lower dose of insulin. In this study, investigators aim to assess whether lower doses of insulin and dextrose are non-inferior to the standard dosing regimen. The non-inferiority margin has been defined as a reduction in serum potassium by 0.5 mmol/L for both treatment groups. This margin was carefully selected because a 0.5 mmol/L decrease in potassium is clinically significant, shifting the condition from moderate to mild hyperkalaemia, thereby reducing the risk of life-threatening arrhythmias. This threshold prioritises patient safety while potentially minimising adverse events, particularly hypoglycaemia, commonly linked to higher insulin doses. Furthermore, if the lower doses are shown to be non-inferior, adverse events such as hypoglycaemia and extravasation could be reduced, enhancing the safety profile of hyperkalaemia management. This improvement represents a key secondary objective of the study and underscores the broader potential of dose optimisation. Research specifically addressing optimal insulin dosing for hyperkalaemia is limited. Given limited research on optimal insulin dosing for hyperkalaemia, these findings could inform clinical guidelines, encouraging flexible, evidence-based dosing strategies in emergency and critical care settings To determine the required sample size for this non-inferiority study comparing insulin dosages for treating moderate hyperkalaemia in the emergency department, investigators utilized a sample size calculation based on the method by Julious, 2004 and the following parameters. Investigators aimed to detect whether the mean difference between the standard treatment and the experimental treatment does not exceed the non-inferiority limit of 0.5 mmol/L, with 80% power and a significance level of 0.05 for a two-sided test. Estimating a standard deviation of 0.68 mmol/L based on prior studies, and using the formula for comparing two means, 23 participants per arm are needed. Including a 10% dropout rate, the total sample size required for the study is 50 participants, with 25 allocated to each treatment group.