A Retrospective Study of Patiromer as Adjunct to Insulin Therapy for Acute Hyperkalemia in the Emergency Department

Introduction

Hyperkalemia is a prevalent and potentially life-threatening electrolyte imbalance. In the United States, the estimated prevalence of hyperkalemia, defined as a serum potassium concentration exceeding 5.5 mmol/L, is approximately 3.6% within the emergency department (ED).¹ Notably, a higher prevalence is reported among patients with conditions such as heart failure, those recently initiated on renin-angiotensin system inhibitors and aldosterone antagonists, and individuals with chronic kidney disease or acute kidney injury.^2,3

Prior studies have demonstrated that the administration of patiromer monotherapy leads to rapid reduction of serum potassium levels in patients not receiving concomitant insulin.^4–6 This effect initiates within the initial 0–6 hours (mean draw time of 2.9 hours) and endures for at least 24 hours.⁷ This observed reduction in potassium levels is of particular interest in the context of the ED where patients frequently present with acute episodic hyperkalemia.³ Among the available treatment options, injectable insulin stands out as a commonly employed approach due to its ability to rapidly induce intracellular potassium shifts. However, a limitation of this method is the diminishing effect observed after approximately 4–6 hours.⁸ Consequently, healthcare practitioners often find it necessary to administer supplementary doses of potassium-lowering agents.

Given the pharmacokinetics of patiromer, it presents itself as a promising candidate for co-administration with insulin. This combined approach holds the potential to sustain the trajectory of potassium reduction beyond the initial 4-hour window typically achieved with insulin alone. The potential advantage of this combined strategy is maintaining a more prolonged reduction in potassium levels and reducing the reliance on frequent re-dosing of insulin, which has been associated with higher risk of hypoglycemia and death.⁹ Such benefits hold particular significance within the resource-constrained environment of the ED.

An open-label, randomized pilot study (REDUCE), showed that patiromer plus standard of care (insulin and albuterol) significantly reduced serum potassium levels to a greater extent than standard of care alone at 2 hours, however there were no significant differences at 4 or 6 hours.¹⁰ A larger, double-blind, randomized controlled trial, Patiromer Utility as an Adjunct Treatment in Patients Needing Urgent Hyperkalemia Management (PLATINUM), is currently underway to further investigate the advantages of this combined approach in the ED setting.¹¹ In the interim, there remains a gap in evidence on the effectiveness of co-administering patiromer with insulin in an acute care setting that this observational study aims to address. Additionally, this study provides a unique opportunity to derive valuable insights from well-designed investigations rooted in local, real-world evidence within a single health system and benchmark these findings against the clinical trial results.

The objective of our study is to assess whether co-administering patiromer with insulin for the management of episodic hyperkalemia in the ED enhances potassium reduction during the 4–12 hours following administration. Furthermore, we aim to determine whether this combined approach reduces the need for adjunct interventions (such as insulin, dialysis, potassium binders, and albuterol) following the initial treatment, in comparison to using insulin alone as the primary treatment strategy.

Methods

Study Design

This retrospective, observational cohort study was conducted utilizing electronic health record (EHR) data and the target trial emulation framework.¹² The study followed the Reporting of Studies Conducted Using Observational Routinely Collected Health Data (RECORD) reporting guideline, an extension of the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.¹³

Study Population

The study included patients that presented to the ED for any reason between July 2018 and July 2023, were found to have hyperkalemia, defined as a serum potassium level exceeding 5.0 mEq/L, and treated with at least one intravenous dose of regular insulin (5–12 units) for hyperkalemia. Individuals with missing baseline data, chronic use of potassium binders (defined as an outpatient prescription in the preceding 6 months), other forms of insulin therapy (eg, long acting or infusion) administered in the ED, or co-administration of potassium binders other than patiromer were excluded.

Intervention and Control Groups

Patients who were administered patiromer within one hour before or after their insulin dose were categorized as the intervention group. Patients without an order for patiromer within the EHR were allocated to the control group. Matching was performed at a 1:1 ratio within each group, employing coarsened exact matching based on baseline serum potassium elevation, insulin dose, dialysis status, age, estimated glomerular filtration rate (eGFR), and prior exposure to renin-angiotensin-aldosterone system inhibitors (RAASi).

Data Collection and Primary Outcome

Potassium levels were collected from time zero (one hour after insulin administration) and up to 24 hours later. The primary outcome, mean change in potassium levels from baseline to the latest value within the 4–12 hour interval, was compared between the intervention and control groups.

Secondary Outcomes

Multiple secondary outcomes were evaluated. Adjunct interventions for potassium lowering within the 12 hours following insulin administration were compared between groups, which included supplementary intravenous insulin doses, nebulized albuterol, subsequent administrations of potassium binders, and hemodialysis sessions. To align our findings with the primary endpoint of the PLATINUM trial, we calculated a group-level net clinical benefit marker, defined as the mean difference in the number of interventions minus the change in serum potassium. We also assessed the incidence of hypokalemia (defined as potassium levels <3.5 mEq/L) as our safety endpoint. Two pre-specificized subgroup analyses were conducted—one restricted to patients with baseline potassium levels >5.8 mEq/L (consistent with the inclusion criteria of the PLATINUM trial), and another restricted to patients with an eGFR >30 mL/min.

Statistical Analysis

Continuous measures, such as potassium reduction and the total count of additional interventions, were assessed using t-tests. Categorical variables were analyzed using the Chi-squared test. Analyses were performed using R version 4.2.1 (R Foundation for Statistical Computing, Vienna, Austria).

Ethical Considerations

In accordance with the principles of the Declaration of Helsinki, this study was approved by the Albert Einstein College of Medicine Institutional Review Board, with a waiver of informed consent due to retrospective nature of the chart review. All data were stored and analyzed on an encrypted platform to maintain security and patient confidentiality.

Results

Of the 2689 eligible patients treated with insulin for hyperkalemia during the study period, the final analysis included 133 patients treated with patiromer plus insulin matched with 133 patients treated with insulin alone. Pre-specified baseline variables were similar between the two groups (see Table 1). Participants had a mean age of 71 years (SD 14 years); 43% were female, 31% self-identified as “Black or African-American”, and 38% self-identified as “Spanish/Hispanic/Latino”. Approximately 34% of patients exhibited an eGFR of <15 mL/min, and 14% were diagnosed with dialysis- dependent end-stage kidney disease (ESKD). The majority (55%) of patients concurrently received intravenous calcium with their insulin treatment, while 38% were co-administered albuterol at baseline.

Table 1 Baseline Demographics

No significant changes were observed in potassium levels from baseline (mean levels 6.2 mEq/L in each group) to the 4–12 hour time frame in the overall cohort (patiromer group: −0.90 mEq/L vs insulin-only group: −0.98 mEq/L, p = 0.51, Figure 1). Notably, 40% of patients did not undergo follow-up potassium testing during this interval. The number of additional potassium-lowering interventions did not exhibit significant differences between the groups (1.08 vs 0.80 per person, p = 0.09). The calculated net clinical benefit of potassium reduction was −0.25 in favor of the patiromer plus insulin arm; however, this difference did not reach statistical significance. Furthermore, no significant changes were noted in secondary and safety outcomes (Table 2).

Table 2 Treatment Outcomes

Figure 1 Potassium Trends Over 12 hours Post-Treatment.

Subgroup analyses (Table 3 and Table 4) were similar to those of primary endpoints. Effects were more pronounced within the subgroup of patients with an eGFR > 30 mL/min. In this subgroup, patients receiving insulin plus patiromer showed a numerical decrease in potassium-lowering interventions (0.63 vs 1.12, p = 0.057) and a statistically significant reduction in additional potassium binder doses (0.02 vs 0.18 doses per patient, p = 0.009) compared to those receiving insulin alone. The net clinical benefit calculation was even more pronounced in both subgroups (−0.31 in eGFR > 30 and −0.38 in potassium > 5.8 mEq/L), favoring patiromer; however, statistical significance was not achieved in either subgroup.

Table 3 Subgroup Analysis of Patients with Baseline Potassium > 5.8 mEq/L

Table 4 Subgroup Analysis of Patients with Baseline eGFR > 30 mL/min

Discussion

Patiromer is an orally administered ion-exchange resin that binds excess potassium predominantly in the colon where potassium concentration is the highest. Initially studied to manage chronic hyperkalemia,¹⁴ it has recently garnered attention as a potential tool for addressing acute potassium elevations, especially in ED or hospital settings. Although the underlying causes of acute hyperkalemia in acute settings may vary due to clinical etiology and comorbidity, patiromer offers a promising means of sustainably and reliably lowering serum potassium levels, affording clinicians the time and opportunity to address the root causes of hyperkalemia.

In this study, we found that concurrent administration of patiromer did not significantly sustain the reduction of serum potassium levels nor reduced the need for additional interventions in a cohort of ED patients with hyperkalemia when compared to insulin administration alone. In the subgroup of patients with eGFR >30 mL/min, those in the patiromer group required fewer additional potassium-lowering interventions. While this finding did not reach statistical significance, likely due to limited power in the subgroup analysis, it suggests a trend worth noting. This may be because patients with lower renal function were more likely to undergo dialysis, reducing potassium levels independently of pharmacological treatments. Future research should focus on patients who are unlikely to require hemodialysis, as patiromer appears to be most effective for episodic hyperkalemia in this population. In resource-constrained healthcare systems, the burden of additional testing and therapies for acute hyperkalemia can strain already busy ED clinical care. Treating hyperkalemia with insulin is challenging due to the potential for patient harm from misadministration, dosing errors, and treatment-induced hypoglycemia. Therefore, there is a need to find opportunities for improving treatment safety and mitigating the need for additional testing in the management of hyperkalemia.

This study has noteworthy limitations due to the challenges of assessing the comparative effectiveness of potassium lowering in observational study designs. We expected to observe a difference in potassium reduction in the 4–12 hour period, which did not account for additional potassium-lowering interventions if the initial reduction was deemed insufficient. This underscores the importance of testing alternative effectiveness measures, such as the net clinical benefit employed in the PLATINUM trial. Additionally, our study was unable to differentiate ESKD from acute kidney injury based on eGFR data alone. While we used diagnoses to identify patients with ESKD, the discrepancy between eGFR and diagnosis-based estimates suggests some measurement error. We also did not include hypoglycemia as a potential adverse event since it would necessitate building a separate confounding model and matching process apart from the primary endpoint.

In summary, this observational study suggests that the addition of patiromer to insulin does not significantly improve potassium reduction in the overall cohort of patients in the ED with episodic hyperkalemia. However, the observed trend of reduced need for additional potassium-lowering interventions in patients with adequate renal function warrants further investigation.