Introduction: The haemodialysis (HD) dose, as expressed by Kt/V urea, is currently routinely estimated with the second generation Daugirdas (D2) equation (Daugirdas in J Am Soc Nephrol 4:1205-1213, 1993). This equation, initially devised for a thrice-weekly schedule, was modified to be used for all dialysis schedules (Daugirdas et al. in Nephrol Dial Transplant 28:2156-2160, 2013), by adopting a variable factor that adjusts for the urea generation (GFAC) over the preceding inter-dialysis interval (PIDI, days). This factor was set at 0.008 for the mid-week session of the standard thrice-weekly HD schedule. In theory, by setting PIDI = 7, one could get GFAC = 0.0025, to be used in patients on the once-weekly (1HD/wk) schedule, but actually this has never been tested. Moreover, GFAC was derived not taking into account the residual kidney urea clearance (Kru). Aim of the present study was to provide a specific value of GFAC for patients on a once-weekly hemodialysis schedule.
Subjects and methods: The equation to predict GFAC (GFAC-1) in the 1HD/wk schedule was established in a group of 80 historical Italian patients (group 1) and validated in a group of 100 historical Spanish patients (group 2), by comparing the Kt/V computed using GFAC-1 (Kt/VGFAC-1) with the reference Kt/V (Kt/VSS) values, as computed with the web-based Solute-Solver software (SS) (Daugirdas et al. in Am J Kidney Dis 54:798-809, 2009). Three more sets of Kt/V (Kt/V0.008, Kt/V0.0025 and Kt/V0.0035) values were computed using the GFAC of the original D2 equation (0.008), the GFAC predicted by PIDI/7 (0.0025) and the mean observed GFAC-1 (0.0035), respectively. They were compared with the reference Kt/VSS values.
Results: The predicting equation obtained from group 1 was: GFAC-1 = 0.0022 + 0.0105 × Kru/V (R2 = 0.93). Mean Kt/VSS in the group 2 was 1.54 ± 0.29 SD (N = 500 HD sessions). The mean percent differences for Kt/V0.008, Kt/V0.0025, Kt/VGFAC-1, and Kt/V0.0035 were 5.1 ± 1.0%, - 1.4 ± 0.7%, 0.0 ± 0.3%, - 0.3 ± 0.7%, respectively. No statistically significant difference was found between Kt/V values, except for Kt/V0.008.
Conclusion: A linear relationship was found between GFAC and Kru/V in patients on the 1HD/wk schedule. Such a relationship is able to improve the "second generation Daugirdas equation" for an accurate estimate of the single pool Kt/V in this setting. However, a simple replacement in the D2 equation of 0.008 with the mean observed GFAC (0.0035) could suffice in the clinical practice.
Keywords: Incremental haemodialysis; Kidney urea clearance; Kt/V; Once-weekly haemodialysis; Urea kinetic modelling.