Efficacy of standard- vs reduced-dose insulin for treatment of hyperkalemia: A quasi-experiment.

ABSTRACT

PURPOSE: Hyperkalemia more commonly affects patients with a glomerular filtration rate of less than 60 mL/min. Using intravenous (IV) insulin to shift potassium intracellularly may cause hypoglycemia, requiring additional treatment or longer hospitalization. Literature on insulin dosing in this context is limited, with one previous study indicating that 5 units of IV insulin might be as effective and result in less hypoglycemia than the standard dose of 10 units of IV insulin. The hyperkalemia treatment pathway at our institution was revised in May 2018 to include a reduced-dose option (5 units of insulin) for patients with end-stage renal disease. This study aimed to compare the prevalence of hypoglycemia between patients who received standard-dose vs reduced-dose IV insulin. METHODS: This single-center, retrospective, quasi-experimental study evaluated the impact of revision of the hyperkalemia treatment pathway by assessing rates of hypoglycemia during the 6 months before and after implementation of the revised pathway. The primary endpoint was prevalence of hypoglycemia, defined as a blood glucose level of less than or equal to 70 mg/dL. RESULTS: There was no statistically significant difference in the occurrence of hypoglycemia when comparing the pre- and postimplementation groups (36 [17.7%] patients vs 34 [18.7%] patients; P = 0.7924). The postimplementation group had a statistically significant lower reduction in potassium levels after treatment than the preimplementation group (mean [interquartile range], -0.9 [-1.3, -0.5] mEq/L vs -0.6 [-1.2, -0.2] mEq/L; P = 0.0095). Baseline potassium levels were similar between the groups. CONCLUSION: Administration of reduced-dose IV insulin for treatment of hyperkalemia was significantly less effective in lowering serum potassium levels and did not decrease prevalence of hypoglycemia. When accounting for potential confounders, the only variable that was associated with hypoglycemia was pretreatment glucose level.