Potassium responses to sodium zirconium cyclosilicate in hyperkalemic hemodialysis patients: post-hoc analysis of DIALIZE

Abstract

Background: Sodium zirconium cyclosilicate (SZC) is an effective and well-tolerated treatment for hyperkalemia in maintenance hemodialysis patients. In post-hoc analyses of the phase 3b DIALIZE study, we examined the spectrum of potassium responses to SZC.

Methods: Post-hoc analyses with SZC and placebo included: the number of long interdialytic interval (LIDI) visits during the 4-week evaluation period where patients attained pre-dialysis serum potassium (sK⁺) concentrations of 4.0-5.0 and 4.0-5.5 mmol/L; potassium gradient (the difference between pre-dialysis sK⁺ and dialysate potassium) at days 36, 43, 50, and 57, and change from baseline to the end of treatment (EOT) using categories of potassium gradient (1 to < 2, 2 to < 3, 3 to < 4, and ≥ 4 mmol/L).

Results: A greater proportion of patients achieved the ranges of pre-dialysis sK⁺ concentration with SZC versus placebo for ≥1, ≥ 2, ≥ 3, and 4 LIDI visits over 4 weeks; 23.7 and 48.5% of patients in the SZC group achieved pre-dialysis sK⁺ concentrations of 4.0-5.0 and 4.0-5.5 mmol/L, respectively, at all 4 LIDI visits. Baseline mean potassium gradient was similar with SZC and placebo. At day 57, mean (standard deviation) potassium gradient was 2.78 (0.08) mmol/L with SZC and 3.52 (0.08) mmol/L with placebo; mean difference (95% confidence interval) was - 0.74 mmol/L (- 0.97 to - 0.52). A greater reduction in potassium gradient category from baseline towards lower-risk categories at EOT was observed with SZC versus placebo.

Conclusions: These analyses expand our knowledge of the spectrum of potassium responses with SZC in hyperkalemic hemodialysis patients.

Trial registration: NCT03303521 .

Keywords: Chronic kidney disease; Hemodialysis; Hyperkalemia; Potassium; Sodium zirconium cyclosilicate.

Fig. 1

Proportion of patients achieving pre-dialysis serum potassium ranges of a 4.0–5.0 mmol/L and b 4.0–5.5 mmol/L for ≥1, ≥ 2, ≥ 3, and 4 LIDI visits during the 4-week evaluation period. Includes pre-dialysis serum potassium values obtained at the LIDI visits in the evaluation period (days 36, 43, 50, and 57). No imputation of missing data was conducted. Patients receiving rescue therapy were included in the analysis. LIDI, long interdialytic interval; SZC, sodium zirconium cyclosilicate

Fig. 2

Mean serum potassium to dialysate potassium gradient (mmol/L) during the 4-week evaluation period. Baseline (visit 1, day −7) mean (SD) potassium gradient was comparable between the treatment arms: SZC 3.78 (0.59) mmol/L versus placebo 3.73 (0.64) mmol/L. Error bars represent 95% CIs. All estimates and 95% CIs in the 4-week evaluation period are obtained from a linear model with gradient as response and treatment as the single covariate. The mean for each treatment group is the least-squares mean from this model. The model was fitted for each visit separately. CI, confidence interval; SD, standard deviation; SZC, sodium zirconium cyclosilicate

Fig. 3

Change in serum potassium to dialysate potassium gradient categories from baseline to EOT with SZC and placebo. Missing data are not shown. Dialysate potassium gradient data at EOT are missing for the following baseline categories: baseline gradient 2–< 3 mmol/L: SZC n = 2 (16.7%), placebo n = 2 (12.5%); baseline gradient 3–< 4 mmol/L: SZC n = 7 (15.6%), placebo n = 3 (6.0%); baseline gradient ≥4 mmol/L: SZC n = 5 (13.2%), placebo n = 5 (15.6%). EOT, end of treatment; SZC, sodium zirconium cyclosilicate