Sodium Zirconium Cyclosilicate for Management of Hyperkalemia During Spironolactone Optimization in Patients With Heart Failure
Original Research
Central Illustration
Abstract
Background
Mineralocorticoid receptor antagonists (MRA) improve outcomes in patients with heart failure and reduced ejection fraction (HFrEF) but are underused in clinical practice. Observational data suggest that hyperkalemia is the leading obstacle for the suboptimal use of MRA.
Objectives
This study sought to evaluate the effects of sodium zirconium cyclosilicate (SZC) in optimizing use of spironolactone among participants with HFrEF and hyperkalemia.
Methods
REALIZE-K (Study to Assess Efficacy and Safety of SZC for the Management of High Potassium in Patients With Symptomatic HFrEF Receiving Spironolactone) was a prospective, double-blind, randomized- withdrawal trial in participants with HFrEF (NYHA functional class II-IV; left ventricular ejection fraction ≤40%), optimal guideline-directed therapy (except MRA), and prevalent or incident MRA-induced hyperkalemia. During open-label run-in, participants underwent spironolactone titration (target: 50 mg/day); those with hyperkalemia started SZC. Participants with normokalemia (potassium: 3.5-5.0 mEq/L) on SZC and spironolactone ≥25 mg/day were randomized to continued SZC or placebo for 6 months. The primary endpoint was optimal treatment response (normokalemia on spironolactone ≥25 mg/day without rescue therapy for hyperkalemia [months 1-6]). The 5 secondary endpoints were tested hierarchically. Exploratory endpoints included a composite of adjudicated cardiovascular death or worsening heart failure (HF) events (hospitalizations and urgent visits).
Results
Overall, 203 participants were randomized (SZC: 102; placebo: 101). Higher percentage of SZC- vs placebo-treated participants had optimal response (71% vs 36%; OR: 4.45; 95% CI: 2.89-6.86; P < 0.001). SZC (vs placebo) improved the first 4 secondary endpoints: normokalemia on randomization dose of spironolactone and without rescue therapy (58% vs 23%; OR: 4.58; 95% CI: 2.78-7.55; P < 0.001); receiving spironolactone ≥25 mg/day (81% vs 50%; OR: 4.33; 95% CI: 2.50-7.52; P < 0.001); time to hyperkalemia (HR: 0.51; 95% CI: 0.37-0.71; P < 0.001); and time to decrease/discontinuation of spironolactone due to hyperkalemia (HR: 0.37; 95% CI: 0.17-0.73; P = 0.006). There was no between-group difference in Kansas City Cardiomyopathy Questionnaire–Clinical Summary Score at 6 months (−1.01 points; 95% CI: −6.64 to 4.63; P = 0.72). Adverse events (64% vs 63%) and serious adverse events (23% vs 22%) were balanced between SZC and placebo, respectively. Composite of cardiovascular (CV) death or worsening HF occurred in 11 (11%) participants in the SZC group (1 with CV death, 10 with HF events) and 3 (3%) participants in the placebo group (1 with CV death, 2 with HF events; log-rank nominal P = 0.034).
Conclusions
In participants with HFrEF and hyperkalemia, SZC led to large improvements in the percentage of participants with normokalemia while on optimal spironolactone dose, and reduced risk of hyperkalemia and down-titration/discontinuation of spironolactone. Although underpowered for clinical outcomes, more participants had HF events with SZC than placebo, which should be factored into the clinical decision making. (Study to Assess Efficacy and Safety of SZC for the Management of High Potassium in Patients With Symptomatic HFrEF Receiving Spironolactone; NCT04676646)
Introduction
Mineralocorticoid receptor antagonists (MRA) are a cornerstone of guideline-directed medical therapy (GDMT) for patients with heart failure and reduced ejection fraction (HFrEF), with well-demonstrated effects on reducing morbidity and mortality.1,2 Despite compelling evidence for efficacy, registry data from the United States and Europe demonstrate continuing gaps in use of MRA in clinical practice, with >50% of eligible patients not receiving these agents.3 Observational data strongly suggest that either prevalent hyperkalemia or the perceived risk of hyperkalemia remain the leading obstacle in the use of MRA,4 and many clinicians pursue down-titration and discontinuation of MRA as a common tool of managing hyperkalemia in real-world practice.5 Moreover, few patients with a prior history of hyperkalemia are ever re-exposed to MRA due to safety concerns.6
The emergence of novel potassium (K+) binders, which have favorable gastrointestinal tolerability profiles that make them suitable for long-term use, has introduced the concept of enabling optimization of renin-angiotensin-aldosterone system inhibitors (RAASi), including MRA, long term in patients with HFrEF who had experienced hyperkalemia or are considered to be at high risk for hyperkalemia.7 Sodium zirconium cyclosilicate (SZC) is an orally administered, inorganic crystal that has high affinity for the K+ ion and exchanges K+ for hydrogen and sodium.8-10 Prior trials have demonstrated that SZC is effective at rapidly lowering K+ and maintaining normal K+ levels long term in patients with hyperkalemia.9,11,12
Only a small proportion of patients in prior SZC trials had a prior history of heart failure (HF), and those who did were not well characterized in terms of their ejection fraction or other clinical parameters.8,9 A prior trial in patients with HFrEF that was designed to evaluate the effects of SZC on RAASi optimization was stopped early due to the COVID-19 pandemic and was inconclusive.13 Therefore, whether SZC can effectively optimize the use of MRA in patients with HFrEF remains unknown. Accordingly, we designed the REALIZE-K (Study to Assess Efficacy and Safety of SZC for the Management of High Potassium in Patients with Symptomatic HFrEF Receiving Spironolactone) trial to determine the efficacy and safety of SZC in optimizing MRA in participants with symptomatic HFrEF and hyperkalemia.14
Methods
Study design
The study design and baseline characteristics of REALIZE-K have been previously published.14 Briefly, the study was a prospective, double-blind, placebo-controlled, randomized-withdrawal trial of SZC to enable optimal dosing of spironolactone in participants with HFrEF and prevalent hyperkalemia or at high risk for development of incident hyperkalemia. The study was performed across 85 sites in 8 countries across Europe and North and Latin America. The trial was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice guidelines. The study protocol was reviewed by ethics committees or Institutional Review Boards at each participating site before enrollment of the first participant (details in Supplemental Table 1) and all participants signed written informed consent for participation.
Participants
Eligible participants were adults with an established HF diagnosis (≥3 months duration), left ventricular ejection fraction ≤40%, and NYHA functional class II to IV symptoms who were receiving treatment with an angiotensin-converting enzyme inhibitor, angiotensin receptor blocker, or angiotensin receptor-neprilysin inhibitor (ARNi), and a beta-adrenergic receptor blocker (unless contraindicated) at stable doses for ≥4 weeks. Participation was permitted for both those untreated with a MRA and those receiving spironolactone or eplerenone <25 mg once daily provided that they had evidence of either prevalent hyperkalemia (defined as serum K+ 5.1-5.9 mEq/L and estimated glomerular filtration rate [eGFR] ≥30 mL/min/1.73 m2 at study enrollment; Cohort 1) or were at high risk for hyperkalemia (defined as either a history of serum K+ >5.0 mEq/L within the prior 36 months and eGFR ≥30 mL/min/1.73 m2 at screening, or serum K+ 4.5-5.0 mEq/L with eGFR 30-60 mL/min/1.72 m2, or age >75 years at screening; Cohort 2). Key exclusion criteria included inpatient hospitalization with unstable HF (patients hospitalized due to HF could be enrolled if HF was hemodynamically stable), symptomatic hypotension or systolic blood pressure <90 mm Hg, end-stage kidney disease requiring hemodialysis or anticipated to require dialysis within 3 months, >30% decrease in eGFR in the 30 days before enrollment, prior intolerance of MRA due to gynecomastia, or any use of K+-binding resins, K+ supplements, or K+-sparing diuretics (other than spironolactone or eplerenone) within 7 days of enrollment.
Randomization and study procedures
Participants meeting all eligibility criteria entered an open-label run-in period (4-6 weeks) for optimization of spironolactone dosing and normalization of serum K+ with use of SZC orally as needed. Those previously treated with eplerenone were converted to equivalent doses of spironolactone before entering the run-in period. For those with prevalent hyperkalemia in Cohort 1, SZC was initiated at 10 g 3 times daily orally for up to 48 hours to correct K+ to the normal range (3.5-5.0 mEq/L) and then reduced to a maintenance dose of 10 g once daily. After normalization of serum K+, spironolactone was initiated or uptitrated per protocol to a target dose of 50 mg/day with further adjustment of SZC dosing (5 g every other day up to 15 g once daily orally) as needed to maintain K+ in the normal range.
Titration of spironolactone to a target dose of 50 mg was based on the American College of Cardiology/American Heart Association/Heart Failure Society of America1 and European Society of Cardiology2 Heart Failure Clinical Practice Guidelines (as an example, the European Society of Cardiology Heart Failure Guidelines state that the “starting dose [of spironolactone] is 25 mg/day, target dose 50 mg/day”).2
For those at risk for hyperkalemia in Cohort 2, spironolactone was initiated and up-titrated toward the target dose at the start of the run-in period; those developing hyperkalemia within the first 4 weeks received SZC 10 g 3 times daily orally for up to 48 hours to correct K+ to the normal range and then transitioned to the maintenance dose of 10 g once daily, with further dose adjustment as guided by the protocol to maintain normokalemia. Participants in this cohort who failed to become hyperkalemic within 4 weeks (and therefore did not meet criteria for initiation of SZC) were discontinued from the study.
Participants continuing on SZC and spironolactone ≥25 mg/day (which is consistent with the dose recommended by international HF guidelines)1,2 and maintaining serum K+ in the normal range (3.5-5.0 mEq/L) at the end of the run-in period entered a double-blind, placebo-controlled, randomized-withdrawal phase in which they were randomly assigned 1:1 to continued treatment with the same doses of spironolactone and SZC orally or to continued treatment with spironolactone and matching placebo orally for 6 months. All participants entering the randomized-withdrawal treatment period were centrally assigned to double-blind randomized study intervention using an interactive response technology/randomization and trial supply management system. Randomization was performed using randomly permuted blocks and randomization codes were computer generated and loaded into the IxRS database. Randomization was stratified according to the presence of hyperkalemia at screening (Cohort 1 vs Cohort 2).
The frequency of laboratory surveillance of serum K+ and creatinine as well as directions for dose adjustments of SZC and rescue therapy of hyperkalemia during study follow-up were governed by the protocol. After discontinuation of SZC at the end of the randomized phase, further follow-up was conducted for an additional week to ensure stable serum K+.
Outcomes
The primary endpoint was the occurrence of optimal treatment response, defined as the composite of serum K+ in the normal range (3.5-5.0 mEq/L) while on a spironolactone dose of ≥25 mg/day (which is consistent with the dose recommended by international HF guidelines)1,2 and without the need for rescue therapy for hyperkalemia since the prior study visit (months 1-6 during the randomized phase). The primary and key secondary endpoints were tested in a hierarchical fashion to ensure strong control for type 1 error with an alpha of 0.05. The key secondary endpoints were: occurrence of normokalemia on the randomized dose of spironolactone and without rescue therapy for hyperkalemia since the prior study visit (months 1-6); occurrence of spironolactone ≥25 mg/day dose (months 1-6); time to first hyperkalemia episode (serum K+ >5.0 mEq/L) during follow-up; time to first decrease or discontinuation of spironolactone dose due to hyperkalemia during follow-up; and between-group difference in the change from baseline to 6 months in the Kansas City Cardiomyopathy Questionnaire-Clinical Summary Score (KCCQ-CSS). Exploratory endpoints included time to first occurrence of cardiovascular (CV) death or worsening HF events (HF hospitalizations or urgent visits as adjudicated by a centralized clinical events committee blinded to treatment assignment), N-terminal pro–B-type natriuretic peptide (NT-proBNP) at 6 months postrandomization, and daily loop diuretic dose (furosemide equivalent) at 6 months postrandomization. Safety and tolerability were monitored through reporting of adverse events (AEs) and serious adverse events (SAEs), as well as protocol-directed monitoring of vital signs, physical examination, and laboratory tests. Due to the potential for sodium retention based on the mechanism of action of SZC, edema was prospectively assessed in all participants using a standardized questionnaire at weeks 5, 17, and 25.
Statistical analysis
Details of the statistical methods have been previously published.14 A randomized sample of 166 participants was estimated to provide 90% power to detect a treatment difference of 20% between SZC and placebo with regard to the primary endpoint, assuming target response rates of 0.7 for the SZC group and 0.5 for the placebo group, a correlation of 0.55 between scheduled visits, a death rate of 10 per 100 patient-years, and a 2-sided type 1 error of 0.05 based on a generalized estimating equation (GEE) model.
All efficacy analyses were conducted in the Full Analysis Set, comprising randomized participants who had ≥1 postrandomization central laboratory serum K+ measurement available. The main analysis of the primary endpoint and its components was performed using a GEE model using the data from postrandomization study visits (from 4 weeks to 6 months), assuming a binomial distribution with an exchangeable correlation structure to account for repeated measures. Sensitivity analyses were conducted varying the K+ threshold used to define normokalemia (K+ 3.5-5.5 mEq/L rather than 3.5-5.0 mEq/L) and excluding participants receiving SZC 15 g (in accordance with the labeled indication outside of the United States). Subgroup analyses evaluated the primary endpoint by stratifying participants according to demographic and clinically relevant baseline characteristics, using a GEE model with a binomial family and a log link, a dependent variable of response per visit, fixed independent variables of randomized treatment, participant recruitment country, a per visit indicator variable, open-label period cohort, subgroup, and randomized treatment interaction with subgroup. Each participant was treated as a cluster.
A similar statistical approach as the primary endpoint was used for the first 2 secondary endpoints in the hierarchical sequence. For the secondary endpoints of time to hyperkalemic event and time to spironolactone dose down-titration or discontinuation, time-to-event analyses were presented graphically using Kaplan-Meier curves for each treatment group and the HR (SZC arm/placebo arm), adjusted for the stratification factor (Cohort 1 vs Cohort 2 at study entry) and its 2-sided 95% CI were estimated using the Cox regression model with treatment and recruitment country as fixed factors. The proportionality of hazards was assessed using diagnostics based on weighted Schoenfeld residuals. A sensitivity analysis was conducted for the secondary endpoint of time to first hyperkalemic event, varying the K+ threshold used to define hyperkalemia (K+ ≥5.5 mEq/L rather than >5.0 mEq/L). Between-group differences in postrandomization KCCQ-CSS (at 6 months) were compared using an analysis of covariance (ANCOVA) model including terms for treatment group, baseline score, stratification factor, and recruitment country as independent variables.
Exploratory analyses included time to first adjudicated event of CV death or worsening HF (defined as HF-related hospitalizations or urgent visits) with between-group comparison performed using log-rank test, treatment ratio of log-transformed NT-proBNP at 6 months, and loop diuretic dose at 6 months postrandomization. Treatment ratio of log-transformed NT-proBNP was estimated using an ANCOVA model, with log-transformed NT-proBNP at 6 months as the dependent variable, and treatment group and baseline log-transformed NT-proBNP as independent variables. The mean daily furosemide equivalent loop diuretic dose during the randomized-withdrawal period was summarized by treatment group using descriptive statistics. Additionally, daily loop diuretic dose at 6 months postrandomization was analyzed using an ANCOVA model with the daily furosemide equivalent dose at the 6-month visit as the dependent variable, and treatment group and baseline dose as independent variables. Additional post hoc exploratory analyses were also conducted to examine adjudicated events of worsening HF when stratifying participants based on the baseline NT-proBNP levels, and 6-month postrandomization values for body weight and systolic blood pressure (using an ANCOVA model with the relevant outcome at the 6-month visit as the dependent variable, and treatment group and baseline value of that outcome as independent variables).
On-treatment safety events were analyzed using the Safety Set Randomized data (comprising randomized participants who received ≥1 dose of SZC or placebo postrandomization) and reported summarizing the numbers and percentages of participants with an event.
All statistical analyses were conducted in SAS v9.4. Exploratory endpoint analyses were not controlled for multiple comparisons, and a P value of <0.05 was considered statistically significant.
Role of the funding source
The study was designed by the academic steering committee in collaboration with the sponsor (AstraZeneca). The results of the primary and key secondary efficacy endpoints in the testing hierarchy were validated by a sponsor-independent academic statistical group at Saint Luke’s Mid America Heart Institute who had full access to all relevant data sets.
Results
From April 5, 2021 to December 29, 2023, 447 participants were enrolled (Figure 1). Overall, 366 entered the open-label treatment phase (Figure 1). Of these, 95 participants entered the open-label run-in phase with prevalent hyperkalemia and eGFR ≥30 mL/min/1.72 m2. The most common reason for not proceeding to randomization in this group was persistent hyperkalemia at the end of the open-label phase (n = 11; 12%). The remaining participants (n = 271) who entered the open-label run-in phase were at risk for hyperkalemia; the 2 most common reasons for not proceeding to randomization were absence of hyperkalemia (n = 72; 27%) and persistent hyperkalemia (n = 33; 12%). The baseline characteristics of participants who entered the open-label run-in phase have been previously published.14 In total, 203 participants were randomized and were included in the Full Analysis Set (SZC, n = 102; placebo, n = 101). One participant was randomized in error (in the SZC group) and, therefore, 202 participants received treatment. A total of 171 participants completed treatment (SZC, n = 88 [86%]; placebo, n = 83 [82%]), and 32 prematurely discontinued treatment (SZC, n = 14 [14%]; placebo, n = 18 [18%]). Overall, 175 participants completed the study and 28 discontinued from the study prematurely; 4 participants withdrew consent to remain in the study and 1 was lost to follow-up. The median duration of follow-up was 179 days (Q1-Q3: 170-185 days) for SZC and 177 days (Q1-Q3: 170-184 days) for placebo.
Participant Disposition
aDefined as serum potassium 5.1-5.9 mEq/L. bDefined as either a history of serum potassium >5.0 mEq/L within the prior 36 months and eGFR ≥30 mL/min/1.73 m2, or serum potassium 4.5-5.0 mEq/L with eGFR between 30 and 60 mL/min/1.72 m2 or age >75 years. cOne participant was randomized in error (in the SZC group) and, therefore, 202 participants received treatment. eGFR = estimated glomerular filtration rate; SZC = sodium zirconium cyclosilicate.
Baseline characteristics of the randomized cohort
In the Full Analysis Set, participants were mostly male (n = 151; 74%), White (n = 185; 91%), and in NYHA functional class II (n = 170; 84%) at the time of enrollment (Table 1). Median left ventricular ejection fraction was 33% (Q1-Q3: 28%-37%). The majority (n = 201; 99%) of participants were receiving angiotensin-converting enzyme inhibitor/angiotensin receptor blocker/ARNi, 130 (64%) were receiving sacubitril/valsartan, 143 (71%) were receiving a sodium-glucose cotransporter 2 (SGLT2) inhibitor, 194 (96%) were receiving a beta-blocker, 115 (57%) were receiving loop diuretics, and 106 (53%) were receiving a low-dose MRA. Despite randomization, there were differences between the SZC and placebo groups for several baseline characteristics. Participants in the SZC vs placebo group were older (median: 73 vs 69 years) and had lower eGFR (median: 48 vs 60 mL/min/1.73 m2). Participants in the SZC group also had higher NT-proBNP (median: 1,268 vs 910 pg/mL) and were more often treated with loop diuretics (Table 1). In total, the majority of participants (n = 158; 78%) were receiving 50 mg/day dose of spironolactone at randomization (Table 1).
| SZC (n = 102)a | Placebo (n = 101) | |
|---|---|---|
| Age, y | 73 (67-79) | 69 (63-76) |
| Male | 76 (75) | 75 (74) |
| Race | ||
| Black or African American | 5 (5) | 5 (5) |
| Other | 6 (6) | 2 (2) |
| White | 91 (89) | 94 (93) |
| Region | ||
| Latin America | 31 (30) | 19 (19) |
| North America | 13 (13) | 24 (24) |
| Europe | 58 (57) | 58 (57) |
| History of type 2 diabetes | 27 (27) | 25 (25) |
| History of atrial fibrillation | 36 (35) | 36 (36) |
| Previous HF hospitalization | 48 (47) | 51 (51) |
| BMI, kg/m2 | 28 ± 5 | 28 ± 5 |
| eGFR, mL/min/1.73 m2 | 48 (42-61) | 60 (45-74) |
| eGFR, ≤60 mL/min/1.73 m2 | 73 (73) | 53 (54) |
| UACR, mg/g | 84 ± 280 | 128 ± 306 |
| UACR | ||
| <30 mg/g | 44 (43) | 38 (38) |
| 30 to <300 mg/g | 20 (20) | 26 (26) |
| ≥300 mg/g | 0 | 5 (5) |
| Serum K+, mEq/L | 5.0 ± 0.5 | 5.0 ± 0.5 |
| NYHA functional class | ||
| II | 85 (83) | 85 (84) |
| III/IV | 17 (17) | 16 (16) |
| LVEF, % | 33 (28-37) | 33 (27-37) |
| NT-proBNP, pg/mL | 1,268 (523-3,725) | 910 (378-2,858) |
| CRT-P or CRT-D | 10 (10) | 6 (6) |
| ICD | 24 (24) | 34 (34) |
| HF therapy | ||
| ACEI | 27 (27) | 26 (26) |
| ARB | 11 (11) | 9 (9) |
| ARNi (sacubitril/valsartan) | 63 (62) | 67 (66) |
| ACEI/ARB/ARNi | 101 (99) | 100 (99) |
| Beta-blockers | 96 (95) | 98 (97) |
| SGLT2 inhibitor | 73 (72) | 70 (69) |
| Low-dose MRA | 44 (44) | 62 (61) |
| Loop diuretics | 66 (65) | 49 (48) |
| Any diuretic | 74 (73) | 69 (68) |
| Spironolactone dose/d at randomization | ||
| 12.5 mg | 1 (1) | 0 |
| 25 mg | 24 (24) | 15 (15) |
| 37.5 mg | 2 (2) | 3 (3) |
| 50 mg | 75 (74) | 83 (82) |
Primary and secondary endpoints
Use of SZC led to a greater occurrence of response for the primary endpoint (defined as maintenance of serum K+ in the normal range [3.5-5.0 mEq/L] while on a spironolactone dose of ≥25 mg/day and without the need for rescue therapy for hyperkalemia since the prior study visit) compared with placebo (OR: 4.45; 95% CI: 2.89-6.86; P < 0.001; estimated percentages 71% vs 36%. (Figure 2, Table 2); the results were consistently favorable for SZC vs placebo for each of the postrandomization visits (Supplemental Figure 1). In 18 prespecified subgroup analyses stratifying participants according to demographic and clinically relevant baseline characteristics, there were significant interactions in the following subgroups: participants who were older (vs younger), those with (vs without) diabetes, and those with lower (vs higher) eGFR (Figure 3). For all of these interactions, the heterogeneity was in the magnitude (and not the directionality) of the treatment effect.
Primary Endpoint: Occurrence of Normokalemia (K+ 3.5-5.0 mEq/L) on a Spironolactone Dose of ≥25 mg/day Without the Need for Rescue Therapy for Hyperkalemia Since Prior Study Visit (Months 1-6 After Randomization)
The analysis was performed using a GEE model with a binomial family and a log link, a dependent variable of response per visit, fixed independent variables of randomized treatment, participant recruitment country, a per visit indicator variable, and open-label period cohort. Each participant was treated as a cluster. The P value reflects the 2-sided z-test. GEE = generalized estimating equation.
| na | Estimate (95% CI) | P Value | ||
|---|---|---|---|---|
| Primary endpoint | ||||
| Percentage of participants with normokalemia,b spironolactone ≥25 mg/d and without rescue therapy since prior visit, months 1-6 after randomization | ||||
| SZC | 102 | 71% | OR: 4.45 (2.89-6.86) | <0.001 |
| Placebo | 101 | 36% | ||
| Secondary endpoints | ||||
| Percentage of participants with normokalemia,b on the same dose of spironolactone as randomization and without rescue therapy since prior visit, months 1-6 after randomization | ||||
| SZC | 102 | 58% | OR: 4.58 (2.78-7.55) | <0.001 |
| Placebo | 101 | 23% | ||
| Percentage of participants on spironolactone ≥25 mg/d, months 1-6 after randomization | ||||
| SZC | 102 | 81% | OR: 4.33 (2.50-7.52) | <0.001 |
| Placebo | 101 | 50% | ||
| Time to first hyperkalemia episode (serum potassium >5.0 mEq/L) | ||||
| SZC | 102 | - | HR: 0.51 (0.37-0.71) | <0.001 |
| Placebo | 101 | - | ||
| Time to first decrease or discontinuation of spironolactone dose due to hyperkalemia | ||||
| SZC | 102 | - | HR: 0.37 (0.17-0.73) | 0.006 |
| Placebo | 101 | - | ||
| Change in KCCQ-CSS at 6 mo | ||||
| SZC | 72 | Between group mean difference −1.01 points | 95% CI: −6.64 to 4.63 | 0.724 |
| Placebo | 59 | |||
Effects of SZC vs Placebo on the Primary Endpoint, Stratified by Prespecified Participant Subgroups
The analysis was performed using a GEE model with a binomial family and a log link, a dependent variable of response per visit, fixed independent variables of randomized treatment, participant recruitment country, a per visit indicator variable, open-label period cohort, subgroup, and randomized treatment interaction with subgroup. Each participant was treated as a cluster. ACEI = angiotensin-converting enzyme inhibitor; ARB = angiotensin II receptor blockers; ARNi = angiotensin receptor-neprilysin inhibitor; LVEF = left ventricular ejection fraction; MRA = mineralocorticoid receptor antagonist; NT-proBNP = N-terminal pro-B-type natriuretic peptide; SGLT2 = sodium-glucose cotransporter 2; sK+ = serum potassium; other abbreviations as in Figures 1 and 2.
SZC (vs placebo) also improved the first 4 of the 5 hierarchically tested secondary endpoints: the occurrence of normokalemia on the randomized dose of spironolactone and without rescue therapy for hyperkalemia (OR: 4.58; 95% CI: 2.78-7.55; P < 0.001; estimated percentages 58% vs 23%); the occurrence of spironolactone ≥25 mg/day dose (OR: 4.33; 95% CI: 2.50-7.52; P < 0.001; estimated percentages 81% vs 50%); time to first hyperkalemia episode (serum K+ >5.0 mEq/L; HR: 0.51; 95% CI: 0.37-0.71; P < 0.001) (Figure 4A); and time to first decrease or discontinuation of spironolactone dose due to hyperkalemia (HR: 0.37; 95% CI: 0.17-0.73; P = 0.006) (Figure 4B). Overall, 13 (13%) of participants in the SZC group and 26 (26%) in the placebo group had a dose decrease or discontinuation of spironolactone due to hyperkalemia. Of these, for SZC and placebo 5 (5%) and 17 (17%) participants, respectively, had a decrease in spironolactone dose, whereas 5 (5%) and 8 (8%) participants had spironolactone therapy interrupted or withdrawn due to hyperkalemia (an additional 2 participants in the SZC group and 0 in the placebo group had spironolactone discontinuation due to withdrawal from the study). There was no between-group statistically significant difference in KCCQ-CSS at 6 months for SZC vs placebo (mean treatment difference: −1.01 points; 95% CI: −6.64 to 4.63; P = 0.72).
Time-to-First Event Endpoints
Time-to-first event endpoints: (A) time to first hyperkalemia and (B) time to first decrease or discontinuation of spironolactone dose due to hyperkalemia. Kaplan-Meier survival curves are truncated at 180 days postrandomization. Analysis of HR was performed using a Cox regression model including randomized treatment group and participant recruitment country, adjusted for the stratification factor (hyperkalemia vs normokalemia at study entry). Placebo group was used as reference level in the Cox model. Abbreviations as in Figure 1.
In sensitivity analyses, the primary endpoint results were consistent when participants receiving the 15 g dose of SZC (or matching placebo) were excluded and when normokalemia was defined as serum K+ 3.5-5.5 mEq/L instead of 3.5-5.0 mEq/L (Supplemental Table 2, Supplemental Figures 2 and 3). In the sensitivity analysis for the secondary endpoint of time to first hyperkalemia episode, which used a more conservative K+ threshold to define hyperkalemia (K+ ≥5.5 mEq/L rather than >5.0 mEq/L), the results were also consistent with the main analysis (HR: 0.55; 95% CI: 0.35-0.86; P = 0.01).
Select Exploratory Endpoints
In the analysis of the composite of adjudicated CV death or worsening HF, 11 (11%) participants in the SZC group and 3 (3%) in the placebo group had an event (log-rank nominal P = 0.034) (Supplemental Figure 4). There was no between-group difference in CV death (1 event in each treatment group) and more participants had an adjudicated HF event (hospitalization or urgent visit) with SZC than placebo (n = 10 [10%] vs n = 2 [2%], respectively).
NT-proBNP at 6 months postrandomization was higher with SZC vs placebo, but this difference was not statistically significant (estimated treatment ratio: 1.26; 95% CI: 0.99- 1.61; nominal P = 0.061). There was also no statistically significant difference in the daily dose of loop diuretic (furosemide equivalents) between SZC and placebo at 6 months postrandomization (estimated mean difference: 0.53 mg/day; 95% CI: −9.70 to 10.75 mg/day; P = 0.92). In post hoc exploratory analysis, there were no significant differences between SZC and placebo in body weight (−0.33 kg; 95% CI: −1.88 to 1.21 kg; nominal P = 0.67) or systolic blood pressure (2.40 mm Hg; 95% CI: −2.61 to 7.42; nominal P = 0.34) at 6 months.
In a post hoc exploratory analysis, among participants with NT-proBNP levels >4,000 pg/mL, adjudicated HF events occurred in 7 of 24 participants in the SZC group and in 1 of 16 participants in the placebo group; whereas among participants with NT-proBNP levels ≤4,000 pg/mL, adjudicated HF events occurred in 3 of 75 participants in the SZC group and 1 of 80 participants in the placebo group (Supplemental Table 3).
Safety
Treatment-emergent AEs and SAEs (overall and leading to discontinuation) were overall balanced between the SZC and placebo groups (Table 3). There were differences in the percentages of participants with SAEs of cardiac failure (n = 12 [12%] [14 events] vs n = 4 [4%] [4 events]) and AEs of peripheral edema (n = 6 [6%] vs n = 2 [2%]), in the SZC and placebo groups, respectively. Peripheral edema events collected using the dedicated edema-specific case report forms were reported in 22 (22%) SZC-treated and 16 (16%) placebo-treated participants (Table 3). Hypokalemia events occurred in 7 (7%) participants in the SZC group (including 1 SAE); there were no events of hypokalemia in the placebo group.
| SZC (n = 101) | Placebo (n = 101) | |
|---|---|---|
| AEs | ||
| Any AE | 65 (64) | 64 (63) |
| Any SAEs | 23 (23) | 22 (22) |
| AEs leading to discontinuation | 6 (6) | 6 (6) |
| SAEs leading to discontinuation | 3 (3) | 2 (2) |
| Peripheral edema AEs | 6 (6) | 2 (2) |
| Hypokalemia | 7 (7) | 0 (0) |
| AE with outcome of death | 1 (1) | 2 (2) |
| Cardiac failure SAEs | 12 (12)a | 4 (4) |
| CV death | 1 (1) | 1 (1) |
| Edema questionnaire datab | ||
| Peripheral edema | 22 (22) | 16 (16) |
| Location of edema | ||
| Pedal | 13 (13) | 10 (10) |
| Ankle | 17 (17) | 13 (13) |
| Pretibial | 11 (11) | 5 (5) |
| Thigh | 1 (1) | 0 (0) |
Discussion
In this double-blind, placebo-controlled, randomized-withdrawal trial, the use of SZC (vs placebo) led to more participants achieving the primary endpoint of normokalemia while on ≥25 mg/day dose of spironolactone and not requiring rescue therapy for hyperkalemia (Central Illustration). The use of SZC also improved 4 of 5 hierarchically tested secondary endpoints, including time to first hyperkalemic event and time to first discontinuation or down-titration of spironolactone. The treatment effects for all of these efficacy endpoints were clinically meaningful. There was no difference in HF-related symptoms and physical limitations between the treatment groups. Although the overall numbers of AEs and SAEs were similar in the SZC and placebo groups, we observed a difference in adjudicated events of HF, more of which occurred in the SZC than in the placebo groups.
Key Findings
Key findings of the REALIZE-K trial. CV = cardiovascular; REALIZE-K = Randomized-Withdrawal Trial Evaluating Sodium Zirconium Cyclosilicate for the Management of Hyperkalemia in Patients With Symptomatic Heart Failure With Reduced Ejection Fraction and Receiving Spironolactone; SZC = sodium zirconium cyclosilicate.
Despite international clinical practice guidelines providing the highest level of recommendations for MRA in patients with HFrEF due to their benefits in reducing HF-related mortality and morbidity,1,2 the underuse of MRA in this patient group remains a common and recalcitrant issue.3 Both the real and perceived risk of hyperkalemia remains the leading reason for suboptimal use of MRA, and commonly results in down-titration and discontinuation of treatment4,5—a practice, that has been associated with worse HF outcomes in observational studies.5 Accordingly, there is a need for additional treatment approaches to enable optimization of MRA therapy in patients with HFrEF, an issue that has been highlighted by the clinical community and professional societies.15 One such solution is the “enablement strategy” using novel K+ binders, which have a more favorable gastrointestinal tolerability profile that allows long-term therapy.7
Two such novel K+ binders (patiromer and SZC) are now available in clinical practice in many countries.8,9,16 Two prior trials of patiromer have suggested that its use can improve optimization of spironolactone dose in patients with HF (including HFrEF), but these were limited by their design features, such as use of patiromer for RAASi optimization among participants who did not have hyperkalemia at baseline (and thus may achieve optimal use of MRA even without the use of a K+ binder).17,18 Prior trials of SZC included relatively small proportions of participants with HF, and those who were included were not well characterized in terms of their clinical parameters of HF.8,9 The only prior dedicated SZC trial in patients with HFrEF was PRIORITIZE-HF (Potassium Reduction Initiative to Optimize RAAS Inhibition Therapy With Sodium Zirconium Cyclosilicate in Heart Failure), however, that study differed from REALIZE-K in terms of the participant population and the key objectives, was stopped early due to the COVID-19 pandemic, and was therefore inconclusive.13
REALIZE-K is the first successfully completed trial that was specifically designed to determine the efficacy and safety of SZC in optimizing MRA use in participants with HFrEF and hyperkalemia, and it had several key design features that differed from PRIORITIZE-HF and were implemented to optimize its clinical relevance. First, unlike in other prior K+ binder trials in patients with HF, only those participants who had hyperkalemia at baseline, or developed it during spironolactone therapy optimization, were treated with the K+ binder—an approach that reflects the use of K+ binders in real-world clinical practice. Second, both the titration of spironolactone and SZC (or placebo) during open-label run-in and randomization phases of the study were strictly governed by the study protocol, which minimized large variations typically observed in routine clinical practice. Of note, in the open-label run-in phase during initiation and titration of spironolactone, only 27% of participants at risk of hyperkalemia did not become hyperkalemic, emphasizing the unmet clinical need for K+-lowering agents. Third, all HF events were adjudicated by an independent blinded events committee, and assessments of edema were protocol-mandated and routinely measured using a dedicated case report form. Finally, participants in the study received some of the highest rates of guideline-directed medical therapy for HFrEF, notably exemplified by the majority of participants receiving sacubitril/valsartan and SGLT2 inhibitors.
The REALIZE-K trial results demonstrate the efficacy of SZC in optimizing the use of MRA in high-risk patients with HFrEF who have prevalent or incident hyperkalemia. Both the primary endpoint and 4 of the 5 secondary endpoints were not only statistically improved by SZC vs placebo, but these effects were also clinically meaningful and remained robust in sensitivity analyses, which used a more conservative threshold for hyperkalemia favored by some clinicians (ie, ≥5.5 mEq/L rather than ≥5.0 mEq/L). Furthermore, the rates of hyperkalemia were high in the placebo group, and the effects of SZC on optimizing MRA while maintaining normokalemia were pronounced despite the high rates of ARNi, SGLT2 inhibitor, and loop diuretic use at baseline—strategies that have been associated with lower risk of developing hyperkalemia.
Despite considerably higher rates of optimal MRA use, we observed differences in investigator-reported edema events and adjudicated HF hospitalizations and urgent visits between the treatment groups. Although the number of HF events was small, more of these occurred with SZC than placebo. There are several potential explanations for this finding. Despite randomization, several chance differences in baseline characteristics were noted, suggesting more advanced HF in the SZC group, including older age, lower eGFR, more loop diuretic use, and higher NT-proBNP levels, which may have, in part, contributed to this observation. However, it is also possible that some patients with HF, including those with a more tenuous volume status and higher NT-proBNP, may be susceptible to fluid overload when exposed to SZC, given that it exchanges K+ for hydrogen and sodium, which may result in increased sodium absorption. From a clinical standpoint, the findings from the REALIZE-K trial suggest that a balance between the efficacy (higher use of MRA) and safety (potential for fluid retention and HF events) of SZC therapy should be considered as a part of the clinical decision making.
Study Limitations
First, REALIZE-K was not powered for clinical events (and, therefore, had a high likelihood of type 1 error for the exploratory endpoint of adjudicated HF events) and had a relatively short duration of follow-up. To give a clear answer on HF outcomes with the MRA optimization using SZC enablement strategy, a sufficiently powered and large outcome trial would have been required. Given the relatively modest sample size, the statistical power for subgroup analyses was limited. Similar to other trials of patients with HFrEF,18-20 the proportion of females and under-represented minorities was small, limiting generalizability; this lack of diversity is an issue that has been recognized for CV clinical trials.21 As previously noted, and as a function of the modest trial sample size, chance differences in several key baseline characteristics were noted between the treatment groups, which could have played a role in the safety findings.
Conclusions
In participants with symptomatic HFrEF and hyperkalemia, SZC led to large improvements in optimizing the use and dose of MRA and K+ levels, and reductions in the risk of hyperkalemia and down-titration or discontinuation of spironolactone. Although underpowered for clinical outcomes, more participants had adjudicated HF events with SZC than placebo despite the more optimal use of spironolactone, which should be factored into the clinical decision making.
Funding Support and Author Disclosures
The REALIZE-K trial was funded by AstraZeneca. Dr Kosiborod will become an AstraZeneca BioPharmaceuticals Research and Development Employee as Senior Vice President for Late Stage Cardiovascular, Renal and Metabolic Disease, effective January 6, 2025; this role entails development oversight for several compounds, including SZC. During conduction of the trial, Dr Kosiborod acted as the Primary Investigator from an independent academic research organization; he was not an employee of AstraZeneca during the design or conduction of the REALIZE-K trial or at the time of manuscript submission; has received research grants (payment to institution) from AstraZeneca, Boehringer Ingelheim, and Pfizer; has received consultant/advisory board fees (payment to institution) from 35Pharma, Alnylam, Amgen, Applied Therapeutics, Arrowhead Pharmaceuticals, AstraZeneca, Bayer, Boehringer Ingelheim, Corcept Therapeutics, Cytokinetics, Dexcom, Eli Lilly, Esperion Therapeutics, Imbria Pharmaceuticals, Janssen, Lexicon Pharmaceuticals, Merck (Diabetes and Cardiovascular), Novo Nordisk, Pfizer, Pharmacosmos, Regeneron, Roche, Sanofi, scPharmaceuticals, Structure Therapeutics, Vifor Pharma, and Youngene Therapeutics; has received other research support (payment to institution) from AstraZeneca and Vifor Pharma; has received honoraria (payment to institution) from AstraZeneca, Boehringer Ingelheim, and Novo Nordisk; and owns stock options in Artera Health and Saghmos Therapeutics. Dr Cherney has received honoraria from Boehringer Ingelheim-Lilly, Merck, AstraZeneca, Sanofi, Mitsubishi-Tanabe, Abbvie, Janssen, AMGEN, Bayer, Prometic, Bristol Myers Squibb, Maze, Gilead, CSL-Behring, Otsuka, Novartis, Youngene, Lexicon, Inversago, GlaxoSmithKline, and Novo Nordisk; and has received operational funding for clinical trials from Boehringer Ingelheim-Lilly, Merck, Janssen, Sanofi, AstraZeneca, CSL-Behring, Novo Nordisk, and Bayer. Dr Desai has received institutional research grants from Abbott, Alnylam, AstraZeneca, Bayer, DevPro Biopharma, Novartis, Pfizer; and has received personal consulting fees from Abbott, Alnylam, AstraZeneca, Avidity Biopharma, Axon Therapeutics, Bayer, Biofourmis, Boston Scientific, Endotronix, GlaxoSmithKline, Medpace, Medtronic, Merck, New Amsterdam, Novartis, Parexel, Porter Health, Regeneron, River2Renal, Roche, scPharmaceuticals, Verily, and Zydus. Dr Testani has received research funding in the form of grants to institutions from AstraZeneca and/or consulting for AstraZeneca. Dr Verma holds a Tier 1 Canada Research Chair in Cardiovascular Surgery; and has received grants and/or research support and/or speaking honoraria from Amarin, Amgen, AstraZeneca, Bayer, Boehringer Ingelheim, Canadian Medical and Surgical Knowledge Translation Research Group, Eli Lilly, HLS Therapeutics, Humber River Health, Janssen, Merck, Novartis, Novo Nordisk, Pfizer, PhaseBio, S&L Solutions Event Management Inc, Sanofi, and Sun Pharmaceuticals; and is the President of the Canadian Medical and Surgical Knowledge Translation Research Group, a federally incorporated not-for-profit physician organization. Dr Chinnakondepalli has received research funding in the form of grants to institutions from AstraZeneca and/or consulting for AstraZeneca. Dr Dolling is an employee of Fortrea, which received a consultancy fee for the conduct of the current study from AstraZeneca. Dr Patel has received research funding in the form of grants to institutions from AstraZeneca and/or consulting for AstraZeneca. Mr Dahl, Mr Eudicone, Ms Friberg, and Dr Ouwens are employees of and hold or may hold stock in AstraZeneca. Dr Antunes has received research funding in the form of grants to institutions from AstraZeneca and/or consulting for AstraZeneca. Dr Connelly is supported by the Keenan Chair in research leadership; and has received research funding in the form of grants and/or consulting for AstraZeneca. Dr Kuthi has received lecture fees from Boehringer, Novartis, and Richter (outside of the submitted work). Dr Lala has received research funding in the form of grants to institutions from AstraZeneca and/or consulting for AstraZeneca. Dr Lorenzo has received honoraria for lectures from AstraZeneca, Bayer, Novartis, and Viatris (outside of the submitted work). Dr Marcos has received honoraria for lectures or advisory boards from AstraZeneca, Boehringer Ingelheim, Bayer, Novartis, Novo Nordisk, Rovi, and Vifor CSL (outside of the submitted work). Dr Nuñez has received honoraria for lectures or advisory boards from Alleviant, AstraZeneca, Boehringer Ingelheim, Bayer, Novartis, Novo Nordisk, Pfizer, Roche, Rovi, and Vifor CSL (outside of the submitted work). Dr Merkely has received lecture fees from Abbott, AstraZeneca, Biotronik, Boehringer Ingelheim, CSL-Behring, Daiichi-Sankyo, Medtronik, and Novartis. Dr Squire has received speaker fees from PharmaCosmos; and his research department has received funding for research from Novartis, AstraZeneca, Boehringer Ingelheim, PharmaCosmos, the British Heart Foundation, and the National Institute for Health Research. Dr Wranicz has received research funding in the form of grants to institutions from AstraZeneca and/or consulting for AstraZeneca. Dr Petrie has received research grants or contracts from AstraZeneca, Boehringer Ingelheim, Boston Scientific, Medtronic, Novartis, Novo Nordisk, Pharmacosmos, Roche, SQ Innovations, and 3R LifeSciences; has received consulting fees or honoraria from Abbvie, Abott, Akero, Applied Therapeutics, Amgen, AnaCardio, AstraZeneca, Bayer, Biosensors, Boehringer Ingelheim, Cardiorentis, Corteria, Corvia, Eli Lilly, FIRE 1, Foundry, Horizon Therapeutics, LIB Therapeutics, Moderna, New Amsterdam, Novartis, Novo Nordisk, Pharmacosmos, Regeneron, Reprieve, Siemens, Takeda, Teikoku, Vifor, and 3R Lifesciences; and has participated on Data and Safety Monitoring Boards from AstraZeneca, Moderna, and Teikoku. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
Abbreviations and Acronyms
| AE | adverse event |
| ARNi | angiotensin receptor-neprilysin inhibitor |
| CV | cardiovascular |
| eGFR | estimated glomerular filtration rate |
| GEE | generalized estimating equation |
| HF | heart failure |
| HFrEF | heart failure and reduced ejection fraction |
| K+ | potassium |
| KCCQ-CSS | Kansas City Cardiomyopathy Questionnaire-Clinical Summary Score |
| MRA | mineralocorticoid receptor antagonist |
| NT-proBNP | N-terminal pro-B-type natriuretic peptide |
| RAASi | renin-angiotensin-aldosterone system inhibitor |
| SAEs | serious adverse events |
| SGLT2 | sodium-glucose cotransporter 2 |
| SZC | sodium zirconium cyclosilicate |
References
1. "2022 AHA/ACC/HFSA Guideline for the Management of Heart Failure: a report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines". J Am Coll Cardiol . 2022;79:e263-e421. https://doi.org/10.1016/j.jacc.2021.12.012.
2. "2021 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure". Eur Heart J . 2021;42:3599-3726. https://doi.org/10.1093/eurheartj/ehab368.
3. "Variation in use and dosing escalation of renin angiotensin system, mineralocorticoid receptor antagonist, angiotensin receptor neprilysin inhibitor and beta-blocker therapies in heart failure and reduced ejection fraction: association of comorbidities". Am Heart J . 2021;235:82-96. https://doi.org/10.1016/j.ahj.2021.01.017.
4. "Hyperkalemia as a cause of undertreatment with mineralcorticoid receptor antagonists for patients with newly onset of heart failure with reduced ejection fraction". Eur Heart J . 2021;42:1: 792. https://doi.org/10.1093/eurheartj/ehab724.0792.
5. "Stopping mineralocorticoid receptor antagonists after hyperkalaemia: trial emulation in data from routine care". Eur J Heart Fail . 2021;23:1698-1707. https://doi.org/10.1002/ejhf.2287.
6. "Incidence, predictors and clinical management of hyperkalaemia in new users of mineralocorticoid receptor antagonists". Eur J Heart Fail . 2018;20:1217-1226. https://doi.org/10.1002/ejhf.1199.
7. "Novel potassium binders to optimize RAASi therapy in heart failure: a systematic review and meta-analysis". Eur J Intern Med . 2024;119:109-117. https://doi.org/10.1016/j.ejim.2023.08.022.
8. "Sodium zirconium cyclosilicate in hyperkalemia". N Engl J Med . 2014;372:222-231. https://doi.org/10.1056/NEJMoa1411487.
9. "Effect of sodium zirconium cyclosilicate on potassium lowering for 28 days among outpatients with hyperkalemia: the HARMONIZE randomized clinical trial". JAMA . 2014;312:2223-2233. https://doi.org/10.1001/jama.2014.15688.
10. "Characterization of structure and function of ZS-9, a K+ selective ion trap". PLoS One . 2014;9:e114686 https://doi.org/10.1371/journal.pone.0114686.
11. "Sodium zirconium cyclosilicate among individuals with hyperkalemia: a 12-month phase 3 study". Clin J Am Soc Nephrol . 2019;14:798-809. https://doi.org/10.2215/CJN.12651018.
12. "Long-term safety and efficacy of sodium zirconium cyclosilicate for hyperkalaemia in patients with mild/moderate versus severe/end-stage chronic kidney disease: comparative results from an open-label, phase 3 study". Nephrol Dial Transplant . 2021;36:137-150. https://doi.org/10.1093/ndt/gfz285.
13. "Potassium reduction with sodium zirconium cyclosilicate in patients with heart failure". ESC Heart Fail . 2023;10:1066-1076. https://doi.org/10.1002/ehf2.14268.
14. "Sodium zirconium cyclosilicate in patients with HFrEF and hyperkalemia: REALIZE-K rationale, design, and baseline characteristics". JACC Heart Fail . 2024;12:1707-1716. https://doi.org/10.1016/j.jchf.2024.05.003.
15. "Expert consensus document on the management of hyperkalaemia in patients with cardiovascular disease treated with renin angiotensin aldosterone system inhibitors: coordinated by the Working Group on Cardiovascular Pharmacotherapy of the European Society of Cardiology". Eur Heart J Cardiovasc Pharmacother . 2018;4:180-188. https://doi.org/10.1093/ehjcvp/pvy015.
16. "Patiromer in patients with kidney disease and hyperkalemia receiving RAAS inhibitors". N Engl J Med . 2015;372:211-221. https://doi.org/10.1056/NEJMoa1410853.
17. "Hyperkalemia in hospitalized patients: causes, adequacy of treatment, and results of an attempt to improve physician compliance with published therapy guidelines". Arch Intern Med . 1998;158:917-924.
18. "Patiromer for the management of hyperkalemia in heart failure with reduced ejection fraction: the DIAMOND trial". Eur Heart J . 2022;43:4362-4373. https://doi.org/10.1093/eurheartj/ehac401.
19. "Angiotensin-neprilysin inhibition versus enalapril in heart failure". N Engl J Med . 2014;371:993-1004. https://doi.org/10.1056/NEJMoa1409077.
20. "Dapagliflozin in patients with heart failure and reduced ejection fraction". N Engl J Med . 2019;381:1995-2008. https://doi.org/10.1056/NEJMoa1911303.
21. "Representativeness in randomised clinical trials supporting acute coronary syndrome guidelines". Eur Heart J Qual Care Clin Outcomes . 2023;9:796-805. https://doi.org/10.1093/ehjqcco/qcad007.
Appendix
Supplemental MaterialFootnotes
Listen to this manuscript's audio summary by Editor Emeritus Dr Valentin Fuster on www.jacc.org/journal/jacc.
The authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the Author Center.
