Urea distribution volume and ionic dialysance
Pozzoni, P.; Pozzi, M.; Di Filippo, S.
Giornale Italiano di Nefrologia Organo Ufficiale Della Societa Italiana di Nefrologia 21(Suppl 30): S226-S230
2004
ISSN/ISBN: 0393-5590 PMID: 15750991 Document Number: 583698
Direct dialysis quantification (DDQ) represents the gold standard for determining urea distribution volume (V) in hemodialysis (HD) patients, but is impractical for routine use because it requires equilibrated post-dialysis plasma water urea concentration. The "formal" single-pool, variable volume, urea kinetic model (SPVV-UKM) is easier to use, needing a blood sample drawn immediately after the dialytic session, but to obtain a V value consistent with the DDQ method, it requires a correct estimate of dialyzer urea clearance (Kd), actually often overestimated. Ionic dialysance (ID) accurately estimates the "effective" urea clearance (Keff), namely Kd corrected for total recirculation. Using ID as an input parameter to SPVV-UKM, correct V values are expected when end-dialysis plasma water urea concentrations are determined in a blood sample drawn after the blood pump speed has been reduced to 50 ml/min for 2 min (Upwt2'). We aimed to compare the V values determined by SPVV-UKM, ID and Upwt2' (VID), those determined by "formal" SPVV-UKM (VKd) and those determined by the anthropometric method proposed by Watson (VA), with the V values determined by the gold standard DDQ method (VDDQ). Thirty-one anuric patients on chronic thrice-weekly HD were studied in 31 dialysis sessions (one per patient). VDDQ = 26.5 +/- 5.3 L; VID = 26.2 +/- 5.1 L; VKd = 32.1 +/- 5.7 L;. VA = 33.2 +/- 5.8 L. The mean (VID - VDDQ) difference was -0.2 +/- 1.3 L, not statistically significant (95% confidence interval (95% CI) -0.7 to 0.2 L; p=0.302); the mean (VKd - VDDQ) difference was 5.6 +/- 2.3 L, statistically significant (95% CI 4.7 to 6.4 L; p<0.001); the mean (VA - VDDQ) difference was 6.7 +/- 2.7 L, statistically significant (95% CI 5.7 to 7.7 L; p<0.001). ID use as an input parameter to SPVV-UKM allows adequate V determinations and, at the same time, circumvents the problem of delayed post-dialysis blood samples. On the other hand, the use of "formal" SPVV-UKM or of anthropometric equations leads to a significant overestimation in urea distribution volume.