Background: The interplay of correct solute mass balances, such as that of sodium (Na+) and potassium (K+) (respectively, Na+MB and K+MB) with adequate ultrafiltration volumes (V(UF)), is crucial in order to achieve haemodynamic stability during haemodialysis (HD). The GENIUS single-pass batch dialysis system (Fresenius Medical Care, Germany) consists of a closed dialysate tank of 90 L; it offers the unique opportunity of effecting mass balances of any solute in a very precise way.
Methods: The present study has a crossover design: 11 stable anuric HD patients underwent two bicarbonate HD sessions, one of 4 h and the other of 8 h in a random sequence, always at the same interdialytic interval, at least 1 week apart. The GENIUS system and high-flux FX80 dialysers (Fresenius Medical Care, Germany) were used. The volume of blood and dialysate processed, V(UF) and dialysate Na+ and K+ concentrations were prescribed to be the same. Plasma water Na+ and K+ trends during dialysis as well as Na+MBs and K+MBs were determined. At the same time, systolic blood pressure (SBP) and diastolic blood pressure (DBP), mean arterial pressure (MAP) and heart rate trends during dialysis were analysed. Plasma volume (PV) changes were computed from plasma total protein concentrations and their trends analysed.
Results: Plasma water Na+ and K+ levels were not significantly different when comparing the start and the end of the sessions of the two treatments. Both the increase of plasma water Na+ levels and the decrease of plasma water K+ levels in the first 4 h were significantly slower during the 8-h sessions when compared with the 4-h ones (P < 0.048 and P < 0.006, respectively). Dialysis sessions were uneventful. SBP decreased significantly during the 4-h sessions, whereas it remained stable during the 8-h ones (P < 0.0001 and P = NS, respectively). Statistically significantly lower intradialysis decreases of SBP (-4.5 ± 16.2 vs -20.0 ± 15.0 mmHg, P < 0.02) and MAP (-1.4 ± 11.7 vs -8.6 ± 11.0 mmHg, P < 0.04) were achieved in the 8-h sessions with respect to the 4-h sessions, in spite of no significant difference for mean V(UF) (2.9 ± 0.9 vs 2.9 ± 0.8 L; P = NS) and mean Na+MBs (-298.1 ± 142.2 vs -286.2 ± 150.7 mmol; P = NS). The decrease of PV levels in the first 4 h was significantly slower during the 8-h sessions when compared with the 4-h ones (P < 0.0001). PV decrease was significantly higher at the end of the 4-h HD sessions than at the end of the 8-h HD sessions (P < 0.043).
Conclusions: The present highly controlled experiments using a crossover design and precise Na+MB and K+MB controls showed that better haemodynamic stability was achieved in the 8-h sessions with respect to the 4-h sessions, in spite of no difference for mean V(UF) and Na+MBs. Thus, other pathophysiological mechanisms, namely, a better PV preservation, must be advocated in order to explain the better haemodynamic stability peculiar to long-hour slow-flow nocturnal HD treatments.