BACKGROUND: Continuous dialysis could provide benefit by constant potassium and phosphate. This study investigates the suitability of potassium and phosphate sorbents for incorporation in an extracorporeal capacity and regenerability testing. METHODS: Capacity testing was uraemic plasma. Regenerability was tested for potassium sorbents, with based on cationic exchange for sodium, with 0.1 M and 1.0 M NaCl. To phosphate sorbents, with adsorption based on anionic exchange, 0.1 M and NaHCO3 and NaOH were used. Subsequently, sodium polystyrene sulphonate (RES-A) and iron oxide hydroxide (FeOOH) beads were cartridge for testing in bovine blood using a recirculating blood dialysis circuit separated by a high-flux dialyzer (dynamic setup). was tested to assess whether this could limit calcium and magnesium RESULTS: In the batch-binding assays, zirconium phosphate most potently potassium (0.44 +/- 0.05 mmol/g) and RES-A was the best regenerable sorbent (92.9 +/- 5.7% with 0.1 M NaCl). Zirconium oxide hydroxide (ZIR- most potently adsorbed phosphate (0.23 +/- 0.05 mmol/g) and the sevelamer carbonate was the best regenerable sorbent (85.7 +/- 5.2% with NaHCO3). In the dynamic setup, a potassium adsorption of 10.72 +/- 2.06 h was achieved using 111 g of RES-A and a phosphate adsorption of 4.73 mmol in 3 h using 55 g of FeOOH. Calcium and magnesium preloading was reduce the net adsorption in 3 h from 3.57 +/- 0.91 to -0.29 +/- 1.85 +/- 0.05 to -0.31 +/- 0.18 mmol, respectively. CONCLUSION: RES-A and suitable, regenerizable sorbents for potassium and phosphate removal in regeneration. Use of zirconium carbonate and ZIR-hydr may further phosphate adsorption, but may compromise sorbent regenerability. Use of amines for phosphate adsorption may enhance sorbent regenerability. magnesium preloading considerably reduced net adsorption of these ions.