Effects of postdialysis urea rebound on the quantification of pediatric hemodialysis

Urea rebound (UR) causes single pool urea kinetic modeling (UKM), which is based on end-dialysis urea instead of its equilibrated value (Ceq), to erroneously quantify hemodialysis (HD) treatment. We estimated the impact of postdialysis UR on the results of formal variable volume single pool (VVSP) UKM [Kt/V, urea distribution volume (V), urea generation rate (G), normalized protein catabolic rate (nPCR), and urea reduction ratio (URR)] in children on chronic HD. Thirty-eight standard pediatric HD sessions in 15 stable patients (9 female, 6 male) aged 14.5 +/- (SD) 3.28 years were investigated. The HD sessions lasted 3.75 +/- 0.43 h. The single pool urea clearance was 4.84 +/- 1.25 ml/min/kg. All HD sessions were evaluated by VVSP and URR (%) with postdialysis urea taken at the end of HD and with Ceq taken 60 min after the end of HD, incorporating double pool effects and representing true double pool values. The anthropometric V was calculated by Cheek and Mellits formulae for children. VVSP significantly overestimated Kt/V by 0.26 +/- 0.18 U (1.68 +/- 0.36 vs. 1.42 +/- 0.30, p < 0.0001), i.e., 19. 05 +/- 13.07%, G/V (0.20 +/- 0.04 vs. 0.18 +/- 0.04, p < 0.0001), nPCR (1.26 +/- 0.23 vs. 1.18 +/- 0.22 g/kg/day, p < 0.0001), and URR (73.92 +/- 6.49 vs. 69.22 +/- 7.06, p < 0.0001). VVSP significantly underestimated kinetic V in comparison to anthropometric V (18.74 +/- 4.04 vs. 20.76 +/- 4.43 liters or expressed as V/body weight: 58 +/- 8 vs. 65 +/- 9%, p < 0.05), while double pool kinetic V was more accurate (21.45 +/- 4.34 liters, V/body weight: 64 +/- 6%, p > 0.05). We conclude that UR has a significant effect on all results of UKM even after standard pediatric HD, and the degree of this efffect is documented. We suggest an increase of the minimum required prescribed single pool Kt/V in children and reduction of any delivered single pool Kt/V by approxiamtely 0.26 Kt/V U. Overestimation of nPCR by approximately 0.08 g/kg/day and underestimation of V by 8.5% should be kept in mind.