LAUSR

Background The incidence of dyskalemias and associated outcomes in acute myocardial infarction (AMI) are unknown in real‐world settings and likely differ from the controlled environment of randomized controlled trials. Methods We examined consecutive survivors of an AMI during 2006‐2011 in SWEDEHEART registry and with plasma potassium at discharge (exposure). Study outcomes were 1‐year risk of hyperkalemia (potassium >5.0 mmol/L), hypokalemia (potassium <3.5 mmol/L), and others (1‐year risk of death, new myocardial infarction, heart failure, and de novo atrial fibrillation). Covariates included demographics, comorbidities, hospital procedures, and medications. Results We included 4,861 patients (65% male, age 71.4 ± 12.6 years) with mean discharge potassium of 4.0 ± 0.4 mmol/L. Within 1 year, 784 (16.1%) new hyperkalemic and 991 (20.4%) new hypokalemic events occurred. Discharge potassium and kidney dysfunction were independent predictors of their occurrence. Compared with discharge potassium of 4.0 to <4.5 mmol/L, the adjusted risk of incident hyperkalemia was 1.71 (95% confidence interval 1.41‐2.06) for potassium of 4.5‐5.0 mmol/L and 2.38 (1.69‐3.35) for potassium of >5.0 mmol/L; the adjusted risk of incident hypokalemia was 1.43 for potassium of 3.5 to <4.0 mmol/L (1.23‐1.66) and 3.12 (2.58‐3.77) for potassium of <3.5 mmol/L. A U‐shaped association was observed between discharge potassium and the risk of death (n = 718), with increased hazards for potassium <3.5 and >4.5 mmol/L. No association was found between discharge potassium and the risk of new myocardial infarction, heart failure, or de novo atrial fibrillation. Conclusions Among real‐world AMI survivors, both hyperkalemia and hypokalemia are frequent. Discharge potassium and kidney function strongly predicted their occurrence, as well as the 1‐year risk of death.