Increasing the Potassium Level in Patients at High Risk for Ventricular Arrhythmias (POTCAST)

Authors, Journal, Affiliations, Type, DOI

• Authors: Christian Jøns, Chaoqun Zheng, Ulrik C.G. Winsløw, Elisabeth M. Danielsen, Tharsika Sakthivel, Emil A. Frandsen, Hillah Saffi, Sadjedeh S. Vakilzadeh-Hashemi, Ketil J. Haugan, Niels E. Bruun, Kasper K. Iversen, Helle S. Bosselmann, Niels Risum, Henning Bundgaard; for the POTCAST Study Group
• Journal: New England Journal of Medicine (NEJM)
• Affiliations: Rigshospitalet (Heart Center, Copenhagen); Zealand University Hospital (Roskilde); Herlev-Gentofte Hospital — all University of Copenhagen, Denmark
• Type: Investigator-initiated, multicenter, open-label, event-driven, randomized superiority trial (RCT)
• DOI: 10.1056/NEJMoa2509542

Overview

The POTCAST trial enrolled 1,200 ICD patients (baseline potassium ≤4.3 mmol/L) in Denmark and randomized them to a strategy of actively targeting high-normal potassium (4.5–5.0 mmol/L) via potassium supplementation, mineralocorticoid receptor antagonist (MRA), dietary guidance, and diuretic reduction versus standard care. After a median 39.6 months, the primary composite endpoint (sustained VT, appropriate ICD therapy, unplanned hospitalization for arrhythmia or heart failure, or all-cause death) occurred significantly less in the high-normal potassium group (22.7% vs 29.2%; HR 0.76; P=0.01; NNT=12.3). The benefit was consistent across all prespecified subgroups and was present even in those who did not reach the target potassium range, suggesting the mechanism relates to avoidance of low-normal potassium rather than achieving high-normal levels per se. This is the first large RCT to directly test a potassium-targeting strategy for arrhythmia prevention.

Keywords

Potassium, ventricular arrhythmias, ICD, mineralocorticoid receptor antagonist, potassium supplementation, hypokalemia, hyperkalemia, sudden cardiac death, heart failure

Key Takeaways

Background and Rationale

• A U-shaped relationship exists between plasma potassium and mortality; potassium in the high-normal range (4.5–5.0 mmol/L) is associated with lower mortality compared with low-normal levels (3.5–4.0 mmol/L)
• Hypokalemia and low-normal potassium increase arrhythmia susceptibility through known electrophysiological mechanisms (increased IKr current reduction, membrane hyperpolarization prolonging repolarization)
• MRAs (spironolactone, eplerenone, finerenone) reduce mortality in HF trials (RALES, EPHESUS, EMPHASIS-HF, FINEARTS-HF) and increase serum potassium by 0.1–0.3 mmol/L; POTCAST tests whether the potassium rise, rather than non-potassium MRA effects, mediates some of this benefit
• Guidelines have not addressed the potential benefit of targeting high-normal potassium in at-risk patients

Study Design

• Multicenter, open-label, event-driven, randomized superiority trial — 3 Danish ICD implant centres
• Population: Adults with ICD or CRT-D; baseline K ≤4.3 mmol/L; eGFR ≥30 mL/min/1.73m²
• Intervention: Potassium chloride tablets (max 4.5 g/day) + MRA (spironolactone max 100 mg/day; eplerenone max 50 mg/day) + dietary guidance + reduction of potassium-losing diuretics; titrated to K 4.5–5.0 mmol/L; biweekly blood tests during adjustment then 6-monthly
• Control: Standard care only
• Primary endpoint: Composite of documented sustained VT >125 bpm lasting >30s, any appropriate ICD therapy, unplanned hospitalization >24h for arrhythmia or heart failure leading to treatment change, or all-cause death (time-to-first-event)
• N=1,200 (600 per arm); enrolled March 2019 – September 2024; trial ended March 31, 2025

Participant Characteristics

• Mean age 62.7±12.0 years; 19.8% women
• ICD for primary prevention in 43.3%
• 64.6% with history of heart failure; 50% ischemic heart disease; ~50% cardiomyopathy or primary arrhythmia disorder
• Mean baseline potassium 4.01±0.24 mmol/L
• Median follow-up 39.6 months (IQR 26.4–49.3)

Intervention Adherence and Potassium Achieved

• 95.3% (572/600) completed the drug-adjustment process
• 74 of 600 started on MRA at baseline; 181 new MRA starts → 399 total on MRA at end of adjustment
• Mean potassium increased 4.01 → 4.36 mmol/L after median 85-day adjustment period in the intervention group (vs 4.01 → 4.05 in standard care)
• Only 249 of 600 (41.5%) reached the target range of 4.5–5.0 mmol/L — main barriers: maximum tolerated dose (25.5%), patient refusal of more medication (12.4%), protocol maximum dose ceiling (9.1%)
• 75.2% remained adherent to trial regimen until death or trial end

Primary Endpoint Results

• Primary composite: 136 (22.7%; 7.3/100 py) vs 175 (29.2%; 9.6/100 py); HR 0.76 (95% CI 0.61–0.95; P=0.01)
• Absolute difference at 5 years: 7.7 percentage points; NNT = 12.3 (95% CI 2.0–14.0)
• Consistent across all prespecified subgroups: age, sex, ischemic vs non-ischemic, HF vs no HF, baseline potassium, primary vs secondary prevention ICD

Secondary Endpoint Results

• Appropriate ICD therapy or documented VT: 15.3% vs 20.3%; HR 0.75 (95% CI 0.57–0.98)
• All-cause death: 5.7% vs 6.8%; HR 0.85 (95% CI 0.54–1.34) — not significant
• Hospitalization for cardiac arrhythmia: 6.7% vs 10.7%; HR 0.63 (95% CI 0.42–0.93)
• Hospitalization for heart failure: 3.5% vs 5.5%; HR 0.64 (95% CI 0.37–1.11) — not significant
• Inappropriate ICD therapy: not separately reported as significant

Subgroup Analysis — Mechanistic Insights

• Benefit was similar whether or not patients received MRAs: MRA group HR 0.75 (0.58–0.97); non-MRA group HR 0.77 (0.56–1.00)
• Benefit present even in those who did NOT reach target K range (HR 0.70; 0.53–0.92) and those who did (HR 0.84; 0.63–1.12, NS for target-range achievers alone)
• Interpretation: effect likely driven by avoidance of low-normal potassium episodes rather than requiring achievement of high-normal range

Safety

• No trial-regimen-related deaths
• Mean plasma creatinine difference at trial end: +4.11 μmol/L (P=0.002) — statistically significant but clinically small
• Hyperkalemia hospitalizations: 7 (1.2%) intervention vs 2 (0.3%) standard care; hypokalemia: 1 vs 4; kidney failure: 9 vs 6 — overall electrolyte/renal hospitalization similar (HR 1.75; 95% CI 0.80–3.82; NS)
• Total unplanned hospitalization or death: 29.5% vs 33.2%; HR 0.88 (P=0.22) — NS

Limitations of the Document

• ICD-only population: Efficacy cannot be inferred for high-risk patients without ICD
• Single country (Denmark): Patient demographics, background GDMT, and potassium intake patterns may differ from other health systems; representativeness concerns noted (Table S6 in supplementary)
• Open-label design: Risk of performance and detection bias, mitigated by external independent endpoint adjudication
• Target range reached in <50%: Underdosing could mean true effect size is larger than observed (intentionally conservative — intention-to-treat)
• Dietary adherence unmeasured: Dietary potassium guidance given but not quantified or verified
• Small disease-specific subgroups: No adjustment for multiplicity; disease-specific conclusions unreliable
• eGFR <30 excluded: Applicability to advanced CKD and dialysis patients unknown
• European-descent majority: Potential ethnic group differences in potassium handling

Key Concepts Mentioned

• concepts/Potassium-and-Ventricular-Arrhythmias — central thesis of the trial
• concepts/Cardiac-Action-Potential — electrophysiological basis of potassium effects
• concepts/Electrical-Storm — extreme manifestation of ventricular arrhythmia burden

Key Entities Mentioned

• entities/ICD — trial population exclusively ICD patients; appropriate ICD therapy as primary endpoint component
• entities/Mineralocorticoid-Receptor-Antagonists — one arm of the potassium-raising intervention; MRA-mediated potassium rise implicated as partial mechanism behind prior HF landmark trials

Wiki Pages Updated

• wiki/sources/potcast-nejm-2025.md — created (this file)
• wiki/concepts/Potassium-and-Ventricular-Arrhythmias.md — created
• wiki/entities/ICD.md — POTCAST section added; connections and sources updated
• wiki/sourceindex.md — entry added
• wiki/wikiindex.md — entries added