BACKGROUND: This study aimed to investigate the association between postoperative nadir serum potassium and 28-day mortality in patients undergoing surgical aortic valve replacement (SAVR) using the Medical Information Mart for Intensive Care (MIMIC-IV) database. METHODS: A retrospective cohort study included 2,877 patients undergoing SAVR from MIMIC-IV. The independent variable was postoperative nadir serum potassium, and the primary outcome was 28-day all-cause mortality. Logistic regression models adjusted for sex, age, BMI, use of beta-blockers, antidiabetic drugs, diuretics, comorbidities (COPD, heart failure, peripheral vascular disease), and indexes of kidney (BUN, Creatinine). Restricted cubic spline (RCS) regression assessed the potential nonlinear relationship, and Kaplan-Meier survival curves compared mortality risk across different serum potassium levels. RESULTS: A U-shaped relationship was observed between postoperative nadir potassium and 28-day all-cause mortality. Higher nadir potassium levels were significantly associated with reduced mortality (odds ratio OR = 0.40, 95% confidence interval CI: 0.24-0.69, p < 0.001). The optimal threshold was 3.55 mmol/L; patients below this level had a markedly increased risk of death. CONCLUSION: Lower postoperative nadir serum potassium is independently associated with increased 28-day mortality in SAVR patients. Maintaining postoperative nadir serum potassium no lower than 3.5 mmol/L may be associated with reduced short-term mortality, although this threshold should not be interpreted as a target level to be constantly maintained. The nadir value represents only the lowest single measurement during ICU stay and does not reflect overall potassium homeostasis.
Zhang et al. (Thu,) studied this question.