Predictors of hyperkalemia in pediatric patients on dialysis: international prospective observational study.

Abstract

Background: Hyperkalemia is an important issue in kidney failure. The aim of the study was to investigate the predictors of hyperkalemia in children receiving maintenance dialysis.

Methods: This was an international prospective cross-sectional observational study involving patients < 18 years receiving chronic hemodialysis or peritoneal dialysis. Hyperkalemia was defined as serum potassium (sK⁺) ≥ 5 mEq/L based on the Pediatric Renal Nutrition Taskforce recommendations. We recorded age, dialysis vintage, urine output (24-h urine collection); dietary K⁺, energy, protein and sodium intake (three-day diaries); office blood pressure (BP) in children < 5 years and 24-h ABPM in older patients; biochemistry (creatinine, urea, sodium, bicarbonate, hemoglobin, phosphate, albumin) and antihypertensive drugs.

Results: Forty-one patients were enrolled (10 peritoneal dialysis, 31 hemodialysis), median age 13.3 (IQR 10.6-15.8) years; 15 of them (36.6%) showed hyperkalemia, and median sK⁺ was 4.7 (4.4-5.0) mEq/L. Renin-angiotensin-aldosterone system inhibitors (RAASi) were prescribed in 9/15 patients with hyperkalemia (60%) and 7/26 (26.9%) without hyperkalemia (p = 0.04). Patients with hyperkalemia were older and had higher urea and creatinine than those with normal sK⁺. A backward stepwise multivariable model showed that the only predictors of hyperkalemia were age (b = 0.53, p = 0.01), urea (b = 0.02, p = 0.03) and treatment with RAASi (b = 2.75, p = 0.021).

Conclusions: While higher age, higher urea levels and treatment with RAASi independently predicted the occurrence of hyperkalemia, K⁺ intake was not associated with sK⁺ in children on dialysis. This emphasizes the importance of considering non-dietary causes of hyperkalemia and considering the bioavailability of K⁺ more than the total dietary K⁺ intake.