Assessment of Patiromer Monotherapy for Hyperkalemia in an Acute Care Setting.

Importance: Hyperkalemia is a common electrolyte disorder in hospitalized patients; however, the clinical usefulness of administering patiromer for reduction of serum potassium levels in this setting is unknown. Objective: To evaluate the outcomes associated with patiromer as monotherapy in patients with acute hyperkalemia in an acute care setting. Design, Setting, and Participants: This cohort study used electronic health record data from adult patients treated with patiromer for acute hyperkalemia in emergency departments, inpatient units, and intensive care units at an urban, academic medical center in the Bronx, New York, between January 30, 2018, and December 30, 2019. Data analysis was conducted between June 2020 and February 2021. Exposures: A single dose of oral patiromer (8.4 g, 16.8 g, or 25.2 g). Main Outcomes and Measure: The primary outcome was the mean absolute reduction in serum potassium level from baseline at 3 distinct time intervals after patiromer administration: 0 to 6 hours, greater than 6 to 12 hours, and greater than 12 to 24 hours. Key secondary outcomes were the incidence of hypokalemia and potassium reduction stratified by baseline potassium level and care setting.
Results: Among 881 encounters of patiromer treatment, the mean (SD) age of patients was 67.4 (14.4) years; 463 encounters (52.6%) were for male patients, and most (338 [38.4%]) were for patients who identified as non-Hispanic Black. The mean (SD) baseline serum potassium level was 5.60 (0.35) mEq/L (to convert to mmol/L, multiply by 1.0), and within the first 6 hours after patiromer administration, the mean (SD) potassium reduction was 0.50 (0.56) mEq/L (P < .001). Both absolute and relative potassium reduction from baseline varied across baseline hyperkalemia severity but not by care setting. The lowest dose of patiromer (8.4 g) was used in 721 encounters (81.8%), and in 725 encounters (82.3%), no further doses of a potassium binder were required. Hypokalemia was noted in 2 encounters (0.2%) at 24 hours after patiromer administration. Conclusions and Relevance: In this cohort study of patients with acute, non-life-threatening hyperkalemia, a single dose of patiromer was associated with a significant decrease in serum potassium levels and a low incidence of hypokalemia. These findings suggest that patiromer monotherapy may be useful in an institutional setting for managing elevated potassium levels and minimizing the risk of hypokalemia associated with other potassium control measures.

Introduction

Hyperkalemia is a common and potentially life-threatening electrolyte disturbance with a reported incidence of 1.1% to 10.0% in hospitalized patients.[1,2,3] Non–life-threatening hyperkalemia episodes occurring in acute care settings are frequently treated with sodium polystyrene sulfonate (SPS), a cation-exchange resin that eliminates total body potassium through the gastrointestinal tract.[4] Despite a long history of SPS use, efficacy data are limited[5] and complicated by the variable onset of action (2-6 hours) and duration of action (6-24 hours) as well as an elevated risk of hypokalemia owing to overcorrection.[6,7,8,9] In addition, SPS may be associated with adverse drug events, including hypernatremia, colonic necrosis, and rare instances of bowel perforation, and with increased risk of hospitalization for serious gastrointestinal events.[10,11] The approval of SPS by the US Food and Drug Administration preceded the era of evidence-based medicine; however, it remains the preferred treatment modality to lower total body potassium levels in institutional settings.

Patiromer is a novel, sodium-free, nonabsorbed polymer agent that significantly reduces potassium within 7 hours of the first dose.[12] Furthermore, the absolute reduction of potassium is more pronounced in patients with significant elevation of potassium levels at baseline, suggesting that rapid and significant potassium level reductions may occur in patients with acute hyperkalemia.[13] Patiromer exchanges potassium for calcium in the colon,[13] where potassium is most abundant. Numerous clinical trials[12,13,14,15,16] have established its usefulness for the treatment of chronic hyperkalemia in patients with diabetes, heart failure, or chronic kidney disease and/or who are taking concurrent renin-angiotensin-aldosterone system inhibitors. The tolerability, safety, and efficacy of patiromer[12,13,14,15,16] renders it an attractive alternative to SPS; however, its use for acute non–life-threatening hyperkalemia is limited by a scarcity of evidence. An initial pilot study[17] in emergency departments suggested that patiromer may be associated with short-term reductions in potassium levels, but the applicability was limited by a small sample size. The current study aimed to address this knowledge gap by evaluating real-world patiromer use and analyzing outcomes associated with its use as monotherapy for non–life-threatening hyperkalemia at institutional settings in a racially and ethnically diverse urban population.

Methods

Study Design and Data Source

This retrospective cohort study used deidentified electronic health record data from Montefiore Medical Center, a large academic medical center in the Bronx, New York. Data were collected using the health system’s proprietary business intelligence platform, which captures encounter-level laboratory values, medication orders, and patient demographic characteristics. Hospital and emergency department encounters from January 30, 2018, to December 30, 2019, that contained at least 1 medication order for oral patiromer were eligible for inclusion in the study. This study was approved by the institutional review board of the Albert Einstein College of Medicine, which waived the need for informed consent for use of these retrospective data owing to minimal risk associated with this study. The study followed the Reporting of Studies Conducted Using Observational Routinely Collected Health Data (RECORD) reporting guideline,[18] an extension of the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

Study Population

Patients aged 18 years or older who received at least 1 dose of patiromer (8.4 g, 16.8 g, or 25.2 g) for the treatment of an acute non–life-threatening hyperkalemia episode, defined as a serum potassium level greater than 5.0 mEq/L (to convert to mmol/L, multiply by 1.0), in any institutional setting (emergency department, inpatient unit, or intensive care unit) were included. Race and ethnicity data were collected to characterize the cohort, were captured directly from the electronic health record, were based on patient self-identification, and were categorized as Asian, Hispanic, non-Hispanic Black, non-Hispanic White, other (non-Hispanic patients indicating their race as multiple selected, American Indian or Alaska Native, or other), or unknown or declined. Time 0 was set to the first patiromer administration ordered as a 1-time dose during the encounter to further ensure that the treatment was intended for short-term lowering of the potassium level. The analysis included patients with multiple, unique emergency department or hospital encounters. Patients who received a potassium binder for chronic hyperkalemia, indicated by evidence of previous therapy with patiromer, SPS, or sodium zirconium cyclosilicate documented within the electronic health record in the year before the hospital admission or emergency department visit, were excluded. To avoid influence of potassium-shifting agents or agents that lower the total body potassium level, patients who received regular insulin in combination with dextrose 3 hours before or after patiromer administration or patients who received patiromer coadministered with furosemide or albuterol were excluded, as were patients who had received another dose within 3 hours of any potassium binder (SPS, sodium zirconium, patiromer cyclosilicate, or patiromer). Pending dialysis orders at the time of patiromer administration or a diagnosis of dialysis dependence also precluded patients from analysis to minimize the influence of dialysis on study outcomes (eFigure 1 in the Supplement).

Study End Points

The primary end point was the mean absolute reduction in serum potassium level from baseline at 3 distinct time intervals after patiromer administration: 0 to 6 hours, greater than 6 to 12 hours, and greater than 12 to 24 hours. These time intervals were prespecified to facilitate analysis in the absence of a standardized, institutional hyperkalemia protocol. Reduction of serum potassium levels was also assessed using relative (percentage) reduction owing to the hypothesized mechanism of more prominent lowering with higher baseline elevation. Outcomes were compared across prespecified subgroups based on care setting, patiromer dose, and baseline hyperkalemia severity stratified as mild (potassium level, 5.1-5.5 mEq/L), moderate (potassium level, 5.6-6.4 mEq/L), or severe (potassium level, ≥6.5 mEq/L). Subsequent use of any potassium binder was captured as well as standard-of-care hyperkalemia control measures (eg, calcium, albuterol, regular insulin coadministered with dextrose, and dialysis). Time to a second potassium binder dose and time to insulin dose were assessed. Incidence of hypokalemia, defined as a serum potassium level less than 3.5 mEq/L, and incidence of hypomagnesemia, defined as a serum magnesium level of 1.7 mg/dL or less (to convert to mmol/L, multiply by 0.4114), at 24 hours were recorded. Potassium laboratory test orders and their performance times after the initial patiromer administration were evaluated at intervals of 0 to 6 hours, greater than 6 to 12 hours, and greater than 12 to 24 hours. The time to the earliest recheck and the presence of repeated laboratory tests in each time interval were evaluated to provide insight on potassium monitoring patterns in real-world clinical practice without a treatment protocol.

Statistical Analysis

Data analysis was conducted between June 2020 and February 2021. For baseline characteristics, continuous variables were summarized as means and SDs and nonparametric distributions as medians and IQRs. Categorical variables were described as numbers and percentages. Based on a previous trial of the effectiveness of patiromer,[12] it was estimated that a total of 35 patients were needed to provide 95% power to detect a mean 24-hour change of −0.3 mEq/L in the serum potassium level from baseline using a 1-group, 2-sided t test with a significance level of 0.05. End points comparing the mean change in serum potassium concentration from baseline to the prespecified time intervals were analyzed using the paired t test. For the primary end points of potassium level reduction at each time interval, we used a Bonferroni-corrected P value of less than 0.016 (0.05 divided by 3 comparisons) to signify statistical significance. All other analyses were considered exploratory, and therefore, no specific P value adjustments for multiple comparisons were performed. All statistical analyses were conducted using JMP Pro, version 15 (SAS Institute Inc), and R, version 4.0.1 (R Project for Statistical Computing).

We conducted a post hoc sensitivity analysis to evaluate the potential association of adjunct potassium control measures (other binders, furosemide, albuterol, insulin, and dialysis) administered throughout the entire 24-hour study period with the overall potassium level trends. First, we created a separate cohort, which excluded any potassium laboratory results with adjunct measures of potassium control administered in the 24 hours preceding the time of blood sample obtainment. We then compared the potassium level trends in the original cohort with those in the new cohort with the additional exclusions.

Results

A total of 881 unique encounters of patiromer administration in an acute care setting were identified between January 30, 2018, and December 30, 2019. Baseline characteristics of patients are provided in Table 1. The mean (SD) age of patients was 67.4 (14.4) years, and 463 encounters (52.6%) were for male patients. In 15 encounters (1.7%), patients identified as Asian; 334 (37.9%), Hispanic/Latinx; 338 (38.4%), non-Hispanic Black; 114 (12.9%), non-Hispanic White; and 80 (9.1%), other or unknown race and ethnicity. In 370 encounters (42.0%), patients were treated with renin-angiotensin-aldosterone system inhibitors before their hyperkalemia episode, which included angiotensin-converting enzyme inhibitors (258 encounters [29.3%]), angiotensin-II receptor blockers (141 [16.0%]), and aldosterone antagonists (81 [9.2%]). Common comorbidities included heart failure with reduced ejection fraction (161 encounters [18.3%]) and moderate to severe chronic kidney disease (665 [75.5%]). Patiromer was primarily administered in inpatient units (679 encounters [77.0%]), followed by emergency departments (152 [17.3%]) and intensive care units (50 [5.7%]), and the mean (SD) length of stay before patiromer administration was 5.4 (8.4) days.

Table 1.

Baseline Demographic and Clinical Characteristics of Patients Treated With Patiromer for Acute Hyperkalemia

Patient or clinical characteristic	Encounters (N = 881)a
Age, mean (SD), y	67.4 (14.4)
Sex
Male	463 (52.6)
Female	418 (47.4)
Race and ethnicity
Asian	15 (1.7)
Hispanic/Latinx	334 (37.9)
Non-Hispanic
Black	338 (38.4)
White	114 (12.9)
Other or unknownb	80 (9.1)
Baseline serum potassium level, mean (SD), mEq/L	5.60 (0.35)
CKD stage (GFR, mL/min/1.73 m2)
1 (≥90)	44 (5.0)
2 (60-89)	168 (19.1)
3 (30-59)	318 (36.1)
4 (15-29)	210 (23.8)
5 (<15)	137 (15.6)
Unknown	4 (0.4)
Heart failure with reduced ejection fraction	161 (18.3)
Previous RAAS inhibitor usec
Any RAAS inhibitor	370 (42.0)
ACE inhibitor	258 (29.3)
ARB	141 (16.0)
Aldosterone antagonist	81 (9.2)
Length of stay, mean (SD), d	5.4 (8.4)
Care setting
ED	152 (17.3)
ICU	50 (5.7)
Inpatient	679 (77.0)
Hyperkalemia severity (serum potassium level, mEq/L)
Mild (5.1-5.5)	450 (51.0)
Moderate (5.6-6.4)	412 (46.8)
Severe (≥6.5)	19 (2.2)
Initial patiromer dose, g
8.4	721 (81.8)
16.8	154 (17.5)
25.2	6 (0.7)

Abbreviations: ACE, angiotensin converting enzyme; ARB, angiotensin II receptor blocker; CKD, chronic kidney disease; ED, emergency department; GFR, glomerular filtration rate; ICU, intensive care unit; RAAS, renin-angiotensin-aldosterone system.

SI conversion factor: To convert to millimoles per liter, multiply by 1.0.

Data are presented as number (percentage) of encounters unless otherwise indicated.

Other includes non-Hispanic patients indicating their race and ethnicity as multiple selected, American Indian or Alaska Native, some other race and ethnicity, or unknown or declined.

Patients may have received multiple RAAS inhibitors.

Evaluation of Treatment Outcomes

The mean (SD) baseline serum potassium concentration before administration of patiromer was 5.6 (0.35) mEq/L, and most patients had mild hyperkalemia (potassium level, 5.1-5.5 mEq/L) or moderate hyperkalemia (potassium level, 5.6-6.4 mEq/L). Patients had severe hyperkalemia (potassium level, ≥6.5 mEq/L) in only 19 encounters (2.2%). The lowest dose of patiromer (8.4 g) was administered in 721 encounters (81.8%). The mean (SD) time from the recording of baseline serum potassium levels in the electronic health record to patiromer administration was 6.7 (4.0) hours.

The results of the primary end point analysis are presented in Figure 1. The mean (SD) reduction in serum potassium levels was 0.50 (0.56) mEq/L at 0 to 6 hours, 0.46 (0.60) mEq/L at greater than 6 to 12 hours, and 0.52 (0.64) mEq/L at greater than 12 to 24 hours (P < .001 for all compared with baseline). This represents a mean (SD) relative reduction from baseline of 8.5% (9.5%), 7.9% (10.1%), and 9.0% (11.2%) for each respective time interval (eFigure 2 in the Supplement).

Figure 1.

Absolute Serum Potassium Level Reduction Across All Time Intervals

Reductions across all time intervals were significantly different from baseline (P < .001). The mean (SD) times of obtaining a blood sample after patiromer administration within each prespecified interval were 2.9 (1.8) hours at 0 to 6 hours, 8.9 (1.7) hours at greater than 6 to 12 hours, and 17.2 (3.1) hours at greater than 12 to 24 hours. Whiskers indicate SEs.

Both absolute and relative potassium reduction from baseline varied across baseline potassium level categories (Figure 2 and eFigure 3 in the Supplement) but not by care setting (eFigures 4 and 5 in the Supplement). In addition, there appeared to be a trend toward a greater relative reduction in potassium levels associated with patiromer doses of 16.8 g or more compared with 8.4 g at greater than 12 to 24 hours (eFigure 6 in the Supplement).

Figure 2.

Absolute Serum Potassium Level Reduction by Hyperkalemia Severity

Whiskers indicate SEs.

In the 24 hours after the initial administration of patiromer, in 725 encounters (82.3%), patients received no further doses of potassium binders, whereas in 137 encounters (15.5%), patients received 1 additional dose and in 19 encounters (2.2%), patients received 2 or more additional doses (Table 2). Of those receiving additional binder doses, the mean (SD) time to administration was 14.8 (5.7) hours. In addition, in 76 encounters (8.6%), patients received at least 1 dose of insulin in the 24 hours after patiromer administration, which was administered within a mean (SD) of 13.5 (6.7) hours from the initial patiromer dose. Other subsequent standard-of-care measures for hyperkalemia included furosemide treatment (77 encounters [8.7%]), hemodialysis (8 [0.9%]), and intravenous calcium treatment (91 [10.3%]). In the sensitivity analysis, which excluded any laboratory results obtained after adjunct potassium control measures, the overall trends in potassium levels after patiromer administration remained largely unchanged (eFigures 7-9 in the Supplement).

Table 2.

Evaluation of Outcomes Associated With Patiromer, Including Potassium Measurement Trends

Outcome	Encounters (N = 881)a
Subsequent potassium binder doses, No.
0	725 (82.3)
1	137 (15.5)
2	13 (1.5)
3	6 (0.7)
Subsequent binder dose
Patiromer, g
8.4	137 (15.6)
16.8	1 (0.1)
Sodium polystyrene sulfonate, 15 g	8 (0.9)
Sodium zirconium cyclosilicate, g
5	1 (0.1)
10	9 (1.0)
Time to second potassium binder dose, mean (SD), h	14.8 (5.7)
Subsequent use of standard-of-care therapyb
Intravenous calcium	91 (10.3)
Furosemide	77 (8.7)
Insulin	76 (8.6)
Dialysis	8 (0.9)
Albuterol	0 (0)
Time to first insulin dose, mean (SD), h	13.5 (6.7)
Hypokalemia at 24 h after first dose of patiromer	2 (0.2)
Hypomagnesemia at 24 h after first dose of patiromer	68 (10.0)
Repeated potassium level monitoring after patiromer administration
Yes	782 (88.8)
No	99 (11.2)
Time to first repeated potassium level measurement, mean (SD), h	9.7 (6.5)
Repeated potassium level monitoring by prespecified time interval, hc
0-6	272 (30.9)
>6 to 12	277 (31.4)
>12 to 24	539 (66.1)
Time to repeated potassium level monitoring by prespecified time interval, mean (SD), h
0-6	2.9 (1.8)
>6 to 12	8.9 (1.7)
>12 to 24	17.2 (3.1)

Data are presented as number (percentage) of patients unless otherwise indicated.

Patients may have received multiple standard-of-care therapies.

Patients may have had multiple repeated potassium level measurements.

In 2 encounters (0.2%), patients experienced a hypokalemia episode (serum potassium level, <3.5 mEq/L), and in 68 encounters (10.0%), patients developed hypomagnesemia (serum magnesium level, <1.7 mg/dL) after patiromer administration. Both episodes of hypokalemia were mild and asymptomatic, with a serum potassium concentration between 3.2 and 3.4 mEq/L.

Evaluation of Practice Context

After patiromer administration, in 749 encounters (85.0%), patients had at least 1 repeated measurement of serum potassium level within 24 hours, with a mean (SD) time to repeated potassium level measurement of 9.7 (6.5) hours. Within the prespecified time intervals, repeated potassium level monitoring occurred in 272 encounters (30.9%) at 0 to 6 hours, 277 (31.4%) at greater than 6 to 12 hours, and 539 (66.1%) at greater than 12 to 24 hours. The mean (SD) potassium level measurement times from baseline were 2.9 (1.8) hours at 0 to 6 hours, 8.9 (1.7) hours at greater than 6 to 12 hours, and 17.2 (3.1) hours at greater than 12 to 24 hours. A detailed evaluation of potassium level measurement trends by hyperkalemia severity is reported in the eTable in the Supplement.

Discussion

To our knowledge, this cohort study is the first to use real-world clinical data from a large representative sample of a racially and ethnically diverse patient population to evaluate outcomes associated with patiromer administration in an acute care setting. Whereas the usefulness of potassium binders is well established for chronic hyperkalemia,[12,14,15] they are not currently approved for use as an emergency treatment for life-threatening hyperkalemia because of the delayed onset of action. A lack of evidence remains regarding the effectiveness of potassium binders for non–life-threatening acute hyperkalemia because, to our knowledge, there have been no prospective studies evaluating this practice. Although historically, SPS has been used for inpatient management of hyperkalemia, evidence-based guidelines for inpatient treatment of hyperkalemia are lacking.[9,19,20] A small pilot study of patiromer as an adjunct to the standard of care in patients with end-stage kidney disease who had serum potassium levels of 6.0 mEq/L or higher in the emergency department revealed significantly lower potassium values compared with baseline at 2 hours (0.52 mEq/L), 4 hours (0.69 mEq/L), and 6 hours (0.61 mEq/L) after patiromer administration in the group receiving patiromer plus the standard of care but only at 4 hours (0.60 mEq/L) in the group receiving the standard of care only.[17] Our study expands these findings to the inpatient and intensive care settings, and to our knowledge, it is the first to isolate the outcomes associated with patiromer from those associated with adjunct interventions to lower serum potassium levels. In addition, our study included a high representation of non-Hispanic Black and Hispanic/Latinx patients (totaling 76.3% of the study population), who are frequently underrepresented in clinical trials.

In our study, patients treated with patiromer experienced sustained reductions in their serum potassium level (mean [SD] reduction, 0.50 [0.56] mEq/L, an 8.5% decrease from baseline) within the first 6 hours of administration (mean [SD] blood sample obtainment time, 2.9 [1.8] hours) in all care settings, suggesting that its onset of action may be faster than previously reported.[11] Furthermore, the association of patiromer with reduced serum potassium levels persisted across all baseline potassium ranges, with infrequent use of additional potassium control measures (such as insulin or repeated potassium binder doses) within the initial 24 hours after patiromer administration. The incidence of hypokalemia was low in our study, with patients experiencing mild hypokalemia (serum potassium level, <3.5 mEq/L) in 0.2% of encounters compared with approximately 4.7% in clinical trial data.[12,13,14,15,16,21] The incidence of hypomagnesemia (10.0%) was similar to that in clinical trial data (9.0%),[12,13,14,15,16,21] although our study’s cutoff was 1.7 mg/dL vs 1.4 mg/dL and we did not evaluate serum magnesium levels at baseline or a change in magnesium levels. Of interest, recently published rates of decreased potassium and magnesium levels from a 4-year global pharmacovigilance database of adverse drug events in patients who were prescribed patiromer in clinical practice were 0.45% and 0.16%, respectively.[22] The variance in these rates across trial data, ambulatory surveillance, and institutional settings likely reflects different hyperkalemia causes, patient populations, and monitoring patterns.

Limitations

This study has limitations. First, the study did not have a control group; therefore, whether observed reductions in potassium levels were associated with patiromer alone could not be determined. Nevertheless, we took care to exclude instances of coadministration with other potassium-lowering agents, including insulin, furosemide, albuterol, and other potassium binders. Furthermore, the intent of our study was to evaluate outcomes associated with patiromer administration in a real-world clinical setting and to quantify the potassium level reduction associated with this agent. In addition, the absence of a protocol for treatment and monitoring of hyperkalemia required us to exclude observations from our analysis in each time interval if a laboratory blood sample was not obtained. Whereas this may have influenced the overall potassium-lowering estimates (eg, patients with higher baseline potassium levels may have been more likely to receive earlier or more frequent monitoring), we evaluated potassium level reduction in both absolute and relative terms as well as stratified by baseline potassium levels, which provided findings consistent with the overall end points.

Our analysis of the clinical practice context indicated wide variability and inconsistent follow-up monitoring. Further research should explore implementation of protocols for treatment, including novel potassium binders, for non–life-threatening hyperkalemia in an acute care setting.

Conclusions

In this cohort study of patients with acute, non–life-threatening hyperkalemia, a single dose of patiromer was associated with a significant decrease in serum potassium level and a low incidence of hypokalemia. The findings suggest that episodes of non–life-threatening hyperkalemia may be treated with patiromer monotherapy in an acute care setting to help minimize the risk of hypokalemia associated with other potassium control measures.