Abstract
Potassium levels (K, mEq/L) fluctuate in patients with acute myocardial infarction (AMI). Potassium was reported to be associated with prognosis in patients with AMI; however, studies evaluating the prognostic value of K fluctuations in this setting are scarce. We retrospectively analyzed patients with AMI hospitalized in a tertiary medical center, through 2002 to 2012. Patients on chronic dialysis or mechanical ventilation were excluded. Based on all K values during hospitalization, minimal, maximal, and fluctuation (gap between 2 consecutive K) were recorded. Primary outcome was inhospital all-cause mortality. Overall, 10 032 patients were studied (age 68.1 ± 14.3 years, 65.4% males, 44.2% ST-segment elevation MI), of which 507 (3.7%) died in hospital. Potassium decreased during the first 2 to 3 days (P for trend <.001), followed by stabilization (P for trend = .807). Potassium in the extreme categories (<3.8 and ≥4.7) and absolute fluctuations >0.1 mEq/L were more common among nonsurvivors than survivors (P < .001 each). In a multivariate analysis, combinations of minimal K <3.8 with maximal K ≥4.7 (odds ratio [OR] = 18.1), K ≥4.4 with fluctuation ≥0.1 (OR = 1.74), or <−0.1 (OR = 2.6) and minimal K after the first 2 admission days (OR = 2.07) were associated with increased risk of mortality (P < .001 each). Potassium fluctuations, peak and nadir K, and its timing independently predict inhospital mortality in patients with AMI.
Get full access to this article
View all access options for this article.
