Phosphate Kinetic Modeling in Patients Treated With Hemodialysis or Hemodiafiltration: A Prospective, Multicenter,...

Studies suggest increased phosphate removal when using hemodiafiltration (HDF) compared with hemodialysis (HD), but a complete analytic comparison has not been reported.

We analyzed data from a 6-month prospective, multicenter, cross-sectional study that enrolled patients treated with high-flux HD or HDF and in whom residual kidney phosphate clearances (KrPhos) were measured. Modeling data and dietary survey data were available from 115 patients (59 treated with HD and 56 treated with HDF).

Predialysis (midweek) serum phosphate values averaged 4.37 ± 1.00 and 4.60 ± 1.18 mg/dL in the HD and HDF groups (p = NS). Mean prescribed phosphate binder equivalent dose (PBED) (including zero values) was 3.33 ± 2.94 g/day in HD and 2.64 ± 2.68 g/day in HDF (p = 0.19). Mean modeled phosphate ingestion was similar in HD and HDF (911 ± 231 vs. 911 ± 300 mg/day, p = NS), but phosphate ingestion by dietary survey was higher in HDF vs. HD (1119 ± 520 vs. 801 ± 420 mg/day, p < 0.001). Mean predialysis serum phosphate values in patients with residual kidney function (defined as KrPhosWater > 1.0 mL/min) and anuric patients were similar (4.43 ± 0.73 vs. 4.50 ± 1.20 mg/dL, respectively), whereas mean prescribed PBED was lower in patients with KrPhosWater > 1.0 mL/min (1.83 ± 2.02 vs. 3.40 ± 2.96 g/day, p < 0.01).

Predialysis serum phosphate is not always lower in patients treated with HDF compared with HD, and this can possibly be explained by a trend to a lower prescribed PBED and/or by a higher dietary phosphate intake.