Prevalence of hyperkalemia and its effects on mortality in hospitalized patients: a single center 10-year retrospective analysis

Abstract

Objective: To investigate the prevalence of hyperkalemia in hospitalized patients, and analyze the effects of different serum potassium levels and change rates of serum potassium on the mortality of hospitalized patients. Methods: The clinical data of 944 446 hospitalized patients in Sichuan Provincial People's Hospital from January 2009 to December 2018 were retrospectively analyzed. Hyperkalemia is defined as serum potassium ≥ 5.5 mmol/L. The effects of serum potassium level and its change rate on hospitalized mortality were analyzed. Results: There were 15 771 patients with hyperkalemia, and the prevalence of hyperkalemia was 1.7% (15 771/944 446). However, the discharge diagnosis rate was only 11.0% (1 735/15 771), and the missed diagnosis rate was 89.0% (14 036/15 771). Cox regression analysis showed that serum potassium<3.5 mmol/L (HR=1.338, 95%CI: 1.164-1.537, P<0.001) or ≥ 6.5 mmol/L (HR=1.421, 95%CI: 1.158-1.744, P=0.001) increased the risk of hospitalized mortality compared with patients with normal serum potassium. Compared with the increased rate of serum potassium by 0.01-0.10 mmol/d, patients who reached the peak of serum potassium at admission (HR=1.251, 95%CI: 1.077-1.453, P=0.003), increased rate of serum potassium by 0.11-0.51 mmol/d (HR=1.499, 95%CI: 1.315-1.709, P<0.001) or >0.51 mmol/d (HR=2.431, 95%CI: 2.105-2.807, P<0.001) increased the risk of mortality. Of patients with hyperkalemia, those who did not repeat the serum potassium test had a higher risk of mortality (HR=1.656, 95%CI: 1.434-1.914, P<0.001). Conclusions: The prevalence of hyperkalemia in hospitalized patients was 1.7%, and the missed diagnosis rate was high at discharge. Patients who had hypokalemia at admission, severe hyperkalemia, rapid increased serum potassium, or failed to repeat serum potassium test during hospitalization, had higher risk of mortality.