Comparative Survival Analysis of Urea Kinetic Based Indices

Abstract

Background

Although urea kinetic modeling indices for measuring dialysis dose are recommended by world expert groups, it is not quite clear whether some of these are superior in predicting the outcome over others. This prospective, single-center study was carried out with the aim to compare predictive value of different indices and methods of measuring dialysis dose.

Methods

The analysis included 93 anuric patients having been on hemodialysis for at least 2 years who were followed-up for 75-months. The dialysis dose was measured by Kt/V (formal UKM, 3 and 2 urea samples), Kt/V (Daugirdas), Kt/V (Lowrie), eKt/V (Daugirdas), URR and TAC urea.

Results

Correlations between dialysis indices and survival time were significant for all indices (p<0.01) except for TAC. All indices, except for TAC urea, were significant predictors of mortality (multivariate Cox regression analysis; p<0.01) and differences of significant levels among these colinear parameters were small.

Conclusion

All examined indices except for TAC urea were highly predictive of patient mortality. Daugirdas and Lowrie simplified Kt/V indices are as predictive of all-cause mortality as more complex formal UKM methods in long-term patients on a 3×4h/week schedule.