Résumé
Background. Urea kinetic modelling-based methods are widely used to assess dialysis efficacy. However, they require blood sampling and are susceptible to a number of errors, mainly from the calculated parameters (particularly V). Spent dialysate determinations have been used and have been shown to be reliable and simple to use. In this study, we associated dialysate-based and clearance determinations along with Kt/V to estimate blood urea levels. Methods. Urea kinetic modelling, continuous sampling of spent dialysate and ionic dialysance were determined in 18 stable dialysis patients during 126 dialysis sessions. Mean blood urea levels were estimated as follows: mean urea level = spent dialysate − urea mass/(dialysance * T). Blood urea levels before and after dialysis were calculated based on the same determinations and extended formulae. Results. Estimated mean urea level was significantly correlated with measured mean blood urea level (R2 = 0.957; P < 0.0001), and Bland and Altman analysis showed significant agreement between estimated and measured levels. Estimated and measured blood urea levels were also correlated before and after dialysis (R2 = 0.972 , P < 0.0001 and R2 = 0.903 , P < 0.0001, respectively), with good agreement for both blood urea before and after dialysis and their respective estimates. Conclusions. Blood urea levels may be reliably estimated from the total mass of urea removed in the dialysate and the dialysance measured during dialysis. Coupling both measurements allows a precise monitoring of dialysis efficacy and a specific evaluation of the patient’s urea metabolism status. Technical dysfunctions and patient variations may be easily identified using this approach without blood sampling.