Estimates of the incidence of hyperkalemia in patients with chronic kidney disease (CKD) vary widely. Our objective was to estimate hyperkalemia incidence in patients with CKD from routine clinical care, including by level of kidney damage or function and among important patient subgroups. Retrospective cohort study. 1,771,900 patients with stage 1-4 CKD identified from the US Optum De-Identified electronic health records database. Impaired kidney damage or function level at baseline based on urinary albumin-creatinine ratio (UACR) and estimated glomerular filtration rate (eGFR), respectively, and selected patient subgroups. Hyperkalemia: 2 elevated serum potassium values (≥5.5mmol/L) from the inpatient setting (2-24 hours apart) or outpatient setting (maximum 7 days apart), or 1 elevated serum potassium value plus pharmacotherapy initiation or hyperkalemia diagnosis (maximum 3 days apart). Incidence rates of hyperkalemia were calculated. Estimates were stratified by UACR and eGFR level at baseline and patient subgroups. Over a mean follow-up of 3.9 years, the incidence of hyperkalemia was 3.37 events/100 person-years (95% confidence intervals, 3.36-3.38). Higher incidence rates were observed with increased UACR and lower eGFR. Highest rates were observed with UACR≥3,500 (up to 19.1/100 person-years) irrespective of decreased eGFR level. High rates also occurred in patients with type 2 diabetes mellitus (T2DM, 5.43/100 person-years), heart failure (8.7/100 person-years), and those prescribed steroidal mineralocorticoid receptor antagonists (sMRAs, 7.7/100 person-years). Potential misclassification of variables from possible medical coding errors; potential data incompleteness issues if patients received care at institutions not included in Optum. Hyperkalemia is a frequent occurrence in CKD, particularly in patients with T2DM, heart failure, or prescribed sMRAs, indicating the need for regular serum potassium and UACR monitoring in this patient population to help mitigate risk.
Read full abstract