Effect of Dynamic Potassium Change on In-Hospital Mortality, Ventricular Arrhythmias, and Long-Term Mortality in STEMI

We evaluated the effect of serum potassium (K) deviation on in-hospital and long-term clinical outcomes in patients with ST-segment elevation myocardial infarction who were normokalemic at admission. A total of 2773 patients with an admission serum K level of 3.5 to 4.5 mEq/L were retrospectively analyzed. The patients were categorized into 3 groups according to their K deviation: normokalemia-to-hypokalemia, normokalemia-to-normokalemia, and normokalemia-to-hyperkalemia. In-hospital mortality, long-term mortality, and ventricular arrhythmias rates were compared among the groups. In a hierarchical multivariable regression analysis, the in-hospital mortality risk was higher in normokalemia-to-hypokalemia (odds ratio [OR] 3.03; 95% confidence interval [CI], 1.72-6.82) and normokalemia-to-hyperkalemia groups (OR 2.81; 95% CI, 1.93-4.48) compared with the normokalemia-to-normokalemia group. In a Cox regression analysis, long-term mortality risk was also higher in normokalemia-to-hypokalemia (hazard ratio [HR] 3.78; 95% CI, 2.07-7.17) and normokalemia-to-hyperkalemia groups (HR, 2.97; 95% CI, 2.10-4.19) compared with the normokalemia-to-normokalemia group. Ventricular arrhythmia risk was also higher in normokalemia-to-hypokalemia group (OR 2.98; 95% CI, 1.41-5.75) compared with normokalemia-to-normokalemia group. The current study showed an increased in-hospital ventricular arrhythmia and mortality and long-term mortality rates with the deviation of serum K levels from normal ranges.