Efficacy of Potassium Supplementation in Hypokalemic Patients Receiving Peritoneal Dialysis: A Randomized Controlled Trial

Rationale & Objective Hypokalemia is a common electrolyte abnormality in patients on peritoneal dialysis (PD) and has been associated with increased risks of peritonitis and death. Whether correction of hypokalemia improves these outcomes is unknown. Study Design Multicenter, open-label, prospective, randomized controlled trial. Setting & Participants Adult (aged ≥18 years) PD patients with hypokalemia (defined as at least 3 values or an average value <3.5 mEq/L in the past 6 months). Randomization was stratified according to center and residual urine output (≤100 or >100 mL/day). Interventions Random assignment to either protocol-based potassium supplementation (titratable dose of oral potassium chloride to maintain serum potassium of 4-5 mEq/L) or conventional potassium supplementation (reactive supplementation when serum potassium is <3.5 mEq/L) over 52 weeks. Treatment groups were compared using intention-to-treat analyses implemented using Cox proportional hazards regression. Outcome The primary outcome was time from randomization to first peritonitis episode (any organism). Secondary outcomes were all-cause mortality, cardiovascular mortality, hospitalization, and conversion to hemodialysis. Results A total of 167 patients with time-averaged serum potassium concentrations of 3.33 ± 0.28 mEq/L were enrolled from 6 PD centers: 85 were assigned to receive protocol-based treatment, and 82 were assigned to conventional treatment. The median follow-up time was 401 (IQR, 315-417) days. During the study period, serum potassium levels in the protocol-based treatment group increased to 4.36 ± 0.70 mEq/L compared with 3.57 ± 0.65 mEq/L in the group treated conventionally (mean difference, 0.66 [95% CI, 0.53-0.79] mEq/L; P < 0.001). The median time to first peritonitis episode was significantly longer in the protocol-based group (223 [IQR, 147-247] vs 133 [IQR, 41-197] days, P = 0.03). Compared with conventional treatment, the protocol-based group had a significantly lower hazard of peritonitis (HR, 0.47 [95% CI, 0.24-0.93]) but did not differ significantly with respect to any of the secondary outcomes. Asymptomatic hyperkalemia (>6 mEq/L) without characteristic electrocardiographic changes occurred in 3 patients (4%) in the protocol-based treatment group. Limitations Not double-masked. Conclusions Compared with reactive potassium supplementation when the serum potassium level falls below 3.5 mEq/L, protocol-based oral potassium treatment to maintain a serum potassium concentration in the range of 4-5 mEq/L may reduce the risk of peritonitis in patients receiving PD who have hypokalemia. Trial Registration Registered at the Thai Clinical Trials Registry with study number TCTR20190725004. Hypokalemia is a common electrolyte abnormality in patients on peritoneal dialysis (PD) and has been associated with increased risks of peritonitis and death. Whether correction of hypokalemia improves these outcomes is unknown. Multicenter, open-label, prospective, randomized controlled trial. Adult (aged ≥18 years) PD patients with hypokalemia (defined as at least 3 values or an average value <3.5 mEq/L in the past 6 months). Randomization was stratified according to center and residual urine output (≤100 or >100 mL/day). Random assignment to either protocol-based potassium supplementation (titratable dose of oral potassium chloride to maintain serum potassium of 4-5 mEq/L) or conventional potassium supplementation (reactive supplementation when serum potassium is <3.5 mEq/L) over 52 weeks. Treatment groups were compared using intention-to-treat analyses implemented using Cox proportional hazards regression. The primary outcome was time from randomization to first peritonitis episode (any organism). Secondary outcomes were all-cause mortality, cardiovascular mortality, hospitalization, and conversion to hemodialysis. A total of 167 patients with time-averaged serum potassium concentrations of 3.33 ± 0.28 mEq/L were enrolled from 6 PD centers: 85 were assigned to receive protocol-based treatment, and 82 were assigned to conventional treatment. The median follow-up time was 401 (IQR, 315-417) days. During the study period, serum potassium levels in the protocol-based treatment group increased to 4.36 ± 0.70 mEq/L compared with 3.57 ± 0.65 mEq/L in the group treated conventionally (mean difference, 0.66 [95% CI, 0.53-0.79] mEq/L; P < 0.001). The median time to first peritonitis episode was significantly longer in the protocol-based group (223 [IQR, 147-247] vs 133 [IQR, 41-197] days, P = 0.03). Compared with conventional treatment, the protocol-based group had a significantly lower hazard of peritonitis (HR, 0.47 [95% CI, 0.24-0.93]) but did not differ significantly with respect to any of the secondary outcomes. Asymptomatic hyperkalemia (>6 mEq/L) without characteristic electrocardiographic changes occurred in 3 patients (4%) in the protocol-based treatment group. Not double-masked. Compared with reactive potassium supplementation when the serum potassium level falls below 3.5 mEq/L, protocol-based oral potassium treatment to maintain a serum potassium concentration in the range of 4-5 mEq/L may reduce the risk of peritonitis in patients receiving PD who have hypokalemia.