Evaluation of Insulin Dosing Strategies for Hyperkalemia Management at an Academic Medical Center

PURPOSE

While intravenous (IV) insulin is often administered at a fixed dose of 10 units for acute hyperkalemia, optimal dosing for minimizing hypoglycemia while effectively reversing hyperkalemia has not been established. The purpose of this analysis was to evaluate the effect of insulin dosing strategies on hypoglycemia in patients with hyperkalemia.

METHODS

Adult patients presenting to an academic medical center who received IV insulin for hyperkalemia between 2016 and 2020 were retrospectively identified. Patients treated with 10 units of insulin (fixed) were compared to those who received < 10 units (reduced). The primary outcome was the incidence of hypoglycemia (blood glucose < 70 mg/dL) within 12 hours of insulin administration. Secondary outcomes included the incidence of severe hypoglycemia (blood glucose < 40 mg/dL) and change in potassium. Multivariable analyses were used to assess for risk factors for hypoglycemia and severe hypoglycemia.

FINDINGS

Of the 2576 patients included, 305 (11.8%) received reduced dosing and 2271 (88.2%) received fixed dosing. Hypoglycemia occurred in 16.7% of the reduced group and 15.9% of the fixed group (P = 0.70). Severe hypoglycemia occurred in 2.3% of the reduced group and 2.5% of the fixed group (P = 0.86). Median potassium reduction from baseline to first check post-insulin was less with reduced dosing (-0.6 mEq/L vs -0.8 mEq/L, P < 0.001). On multivariable regression analysis, greater weight-based insulin dose and ED location were significant predictors for hypoglycemia and severe hypoglycemia. Location in the intensive care unit was associated with a decreased risk of hypoglycemia. Higher pre-insulin glucose was protective for hypoglycemia and severe hypoglycemia.

IMPLICATIONS

The incidence of hypoglycemia was similar among both groups. Greater weight-based insulin dose was a significant risk factor for hypoglycemia, while higher baseline glucose levels were associated with a decreased risk, indicating that patient-specific insulin dosing for hyperkalemia may be warranted.

Evaluation of Goal Phenytoin Levels After an Initial Intravenous Loading Dose at an Academic Medical Center

BACKGROUND

Phenytoin intravenous loading doses are administered in status epilepticus to rapidly achieve therapeutic levels. Accurately assessing phenytoin levels after the initial load can be challenging because of its complex pharmacokinetic profile and nonstandardized weight-based loading doses.

OBJECTIVES

The objectives of this analysis were to determine the incidence of patients achieving goal phenytoin levels after the initial loading dose and characterize factors that contribute to achieving the goal level.

METHODS

This single-center, retrospective cohort analysis was approved by our institutional review board and included adult patients who received a phenytoin load from May 2016 to March 2021. Patients were excluded if no total phenytoin level was drawn within 24 hours of the load, if the maintenance dose was given before the first level was drawn, or if the patient was on phenytoin before the load. The major endpoint was the percentage of patients achieving a corrected goal phenytoin level of ≥10 mcg/mL after the initial load. Multivariate regression was used to determine predictors of achieving the goal phenytoin level.

RESULTS

Of the 152 patients included, 139 patients (91.4%) achieved a corrected goal level after the first load. Patients at goal received a significantly higher median weight-based loading dose (19.1 mg/kg [15.0-20.0] vs 12.6 mg/kg [10.1-15.0], P < 0.01). The multivariate analysis identified weight-based dosing as a statistically significant predictor of achieving the corrected goal level (odds ratio, 1.30; 95% CI, 1.12-1.53; P < 0.01).

CONCLUSION AND RELEVANCE

Most patients achieved a corrected goal phenytoin level after the initial load. A higher median weight-based loading dose was shown to be a predictor of achieving the goal level and should be encouraged for rapid seizure termination. Future studies are warranted to confirm patient-specific factors that affect rapid achievement of the goal phenytoin level.

Evaluation of Dexmedetomidine as an Adjunct to Phenobarbital for Alcohol Withdrawal in Critically Ill Patients

INTRODUCTION

Dexmedetomidine (DEX) is commonly used with benzodiazepines for the management of alcohol withdrawal syndrome (AWS), but limited data exist regarding its use with phenobarbital (PHB). This analysis evaluated the utility of DEX in addition to PHB for AWS in adult patients admitted to the intensive care unit (ICU).

METHODS

This was a single-center, retrospective cohort analysis of critically ill adult patients who received PHB plus either DEX or different adjunctive therapies (NO-DEX) for AWS between 2018 and 2021. Patients were excluded if they had underlying altered mental status or seizure disorder unrelated to AWS or received PHB at outside hospitals. Coarsened exact matching (CEM) was performed to match patients on baseline characteristics in a 1:1 ratio. The primary outcome was ICU length of stay (LOS). A multivariate linear regression analysis was performed to assess the effects of DEX on ICU LOS when accounting for confounders. Secondary outcomes included days with delirium and incidence of mechanical ventilation after PHB administration.

RESULTS

Of the 606 encounters evaluated, 197 met criteria for inclusion. After CEM, 56 encounters remained in each group for analysis. The median ICU LOS was 97.2 [50.1:139.5] hours for the DEX group and 47.5 [28.8:88.1] hours for the NO-DEX group (P = .002). The multivariate linear regression analysis showed the use of DEX (P = .008) was independently associated with an increased ICU LOS by 49.8 h. The DEX group had higher rates of total delirium days (208 vs 143 days, P < .001) and a higher incidence of mechanical ventilation after PHB administration (32% vs 9%, P < .001).

CONCLUSION

This analysis suggests the use of adjunctive DEX with PHB for AWS was associated with a prolonged ICU LOS. Additional studies are needed to further understand the role of adjunctive DEX in the treatment of AWS in critically ill patients.

Intravenous levothyroxine stewardship program at a tertiary academic medical center

PURPOSE

Based on the pharmacokinetic profile of levothyroxine, a 3-day hold guideline for adult patients ordered for intravenous (IV) levothyroxine was implemented at a tertiary academic medical center. The purpose of this study was to evaluate the impact of the implementation of an IV levothyroxine hold guideline.

METHODS

This single-center, retrospective analysis identified patients ordered for IV levothyroxine during a 13-week period before and after implementation of the guideline. The primary outcome was guideline adherence, defined as full implementation of the 3-day hold. Secondary outcomes included the number of IV levothyroxine administrations avoided in the post-guideline group, extrapolated yearly cost avoidance (EYCA) after guideline implementation, reasons for guideline non-adherence, and number of safety reports involving IV levothyroxine.

RESULTS

A total of 166 and 134 patients met inclusion criteria for the pre- and post-guideline groups, respectively. Guideline adherence was observed in 94 (70.1%) patients, resulting in 276 vials saved in the 13-week post-guideline period, which translated to an EYCA of $139,877. Forty orders (29.9%) were non-adherent to the guideline, with the most common reason stated as nil per os (NPO). No difference in safety outcomes was seen between the pre- and post-guideline groups, as evidenced by 1 safety report in each group.

CONCLUSION

We observed a high rate of adherence to an IV levothyroxine hold guideline. This was associated with a substantial cost savings over the study period with no increase in reported safety events. To our knowledge, this is the first published report of an inpatient IV levothyroxine 3-day hold guideline.