Improvement of decisions regarding hemodialysis patients by urea kinetic modeling

The Correlation between Protein Catabolic Rate and Kt/V in Hemodialysis is not a Mathematical Artifact

To test whether the demonstrated correlation between Kt/V urea and normalised protein catabolic rate (NPCR) is caused by a mathematical artifact, we examined 12 adult and 8 paediatric haemodialysis patients. Kt/V was determined by the classical method. As an equivalent of NPCR we measured normalised urea generation rate (NUGR) by a one-week collection of dialysate and urine using total body water estimated by bioelectrical impedance for normalisation. The correlation between Kt/V and NUGR was 0.58 (all), 0.55 (adults), and 0.62 (children). Since the determination of Kt/V and NUGR are not interdependent, we conclude that the correlation between Kt/V and protein catabolism is not artifactual.

Comparison of approximate and iterative methods to calculate the efficiency of haemodialysis as KT/V

Dialysis dose is given by the index KT/V, where K is the dialyser urea clearance, T the duration of dialysis and V the notional urea distribution volume. KT/V may be calculated using the single pool model and an iterative calculation which requires a computer program. Three estimates of KT/V, each based on simple correction of the log ratio in (C0/C1), where C0, C1 denote the pre- and post-dialysis blood urea concentrations in mg/ml, were developed and compared with the values of KT/V given by iteration. The first estimate was given by adding two terms to the log ratio to account for the effect of ultrafiltration and urea generation while the second and third estimates were given by increasing the log ratio by 12.5% and 15% respectively. All estimates gave good agreement when compared with the iterative technique by Bland-Altman analysis for 27 patients on standard 4 hour dialysis treatment using 4 different dialysers (0.71<KT/V<1.62). The mean difference in all cases was less than 0.025 and the differences were not clinically significant. The use of these approximate formulae to estimate dialysis efficiency avoids the need for complex calculation or additional blood samples.