Adequacy in hemodialysis: New concepts derived from online urea monitoring

Abstract

Background. The estimation of urea kinetic parameters [urea generation rate (G(u)), normalized protein catabolic rate (NPCR), and dialysis dose (Kt/V)] is routinely performed during a single hemodialysis session as a representative sample of a stable series. To ascertain whether the stability assumption is tenable and to estimate the variability of urea kinetic parameters, a number of stable patients on regular dialysis treatment in their usual clinical setting were followed. Methods. Thirteen stable patients on regular hemodialysis aged 61 ± 11 were monitored from 5 to 24 weeks (median, 10 weeks) with the urea biosensor system in double-chamber hemodiafiltration. Body weight did not change appreciably. Residual renal function was negligible or absent. Weekly averaged urea concentration (TACW), G(u), NPCR, and Kt/Veq were calculated, and their serial patterns and interrelationships were evaluated through graphical analysis and linear regression. Results. In six patients, the urea pool was substantially unchanged, but variability of G(u) and Kt/V was comparable to that of the other groups. In three patients, body urea pool increased. G(u) went from 4.78 ± 0.44 to 5.40 ± 0.65 mg/min, and Kt/V went from 1.25 ± 0.25 to 1.34 ± 0.31. In four patients, body urea pool decreased; G(u) went from to 6.55 ± 1.91 to 5.85 ± 2.26 mg/min, and Kt/V did not change appreciably. Parameters might change in a nearly linear trend or occasionally as abrupt or oscillating phases. G(u) was the main factor involved, and the only one affecting four of the seven unstable patients. Kt/V was never solely involved. Conclusions. Our data indicate that the estimation of urea kinetic parameters is often affected by a non-negligible degree of variability, which can be ascribed to the variability of the dialytic dose delivered and, above all, to the daily changes of diet protein assumption.