Stopping mineralocorticoid receptor antagonists after hyperkalemia: trial emulation in data from routine care.

Abstract

AIMS\nWhether to continue or stop mineralocorticoid receptor antagonists (MRA) after an episode of hyperkalemia is a challenge in clinical practice. While stopping MRA may prevent recurrent hyperkalemias, it deprives patients of their cardioprotection. We here assessed the association between stopping vs continuing MRA therapy after hyperkalemia and the subsequent the risks of adverse health events.\n\n\nMETHODS AND RESULTS\nObservational study from the Stockholm Creatinine Measurements (SCREAM) project 2006-2018. We identified patients initiating MRA and surviving a first-detected episode of hyperkalemia (plasma potassium>5.0\u2009mmol/L). Using target trial emulation methods, we assessed the association between stopping vs continuing MRA within 6 months after hyperkalemia and subsequent outcomes. The primary outcome was the composite of hospital admission with heart failure, stroke, myocardial infarction or death. The secondary outcome was occurrence of another hyperkalemia event. Among 24,100 patients initiating MRA, we identified 5,672 who developed hyperkalemia. Median age was 76\u2009years, 44% were women and 74% had a history of heart failure. Following hyperkalemia, 1652 (30%) discontinued treatment. Compared with continuing MRA, stopping therapy was associated with a lower 2-year risk of recurrent hyperkalemia (HR 0.77; 95% CI 0.73-0.81), but a higher risk of the primary outcome (HR 1.07; 95% CI 1.03-1.12). Similar results were observed in patients with heart failure, after censoring when the treatment decision was changed, and across pre-specified subgroups.\n\n\nCONCLUSIONS\nStopping MRA after an episode of hyperkalemia was associated with reduced risk for recurrent hyperkalemia, but higher risk of death or cardiovascular events. Recurrent hyperkalemia was common in either strategy.