The on-line measurement during hemodialysis of ionic dialysance provides an estimation of urea clearance with a good and already proven correlation. Some discrepancies remain controversial, and the influence of the dialyzer membrane is still being debated. Eighty-eight measurements of ionic dialysance (ID) were performed with a Diascan module (Hospal R&D, Int., Lyon, France), 51 with cellulosic membranes, and 37 with synthetic membranes, chosen according to their surface charges. The ID was compared to the urea clearance (UK) measured from the blood (n=16) and dialysate (n=88) sides. The ID is closely correlated (r=0.91) but significantly (p < 0.01) lower than the UK by 5% (ID/UK=0.95+/-0.06). The correlation is improved by a semilogarithmic regression analysis (r=0.93). Regarding the influence of the membrane charge, a slight difference is only evidenced for UK < 180 ml/min whereby ID is closer to the urea clearance for the charged membranes (ID/UK=0.98+/-0.05 for charged membranes versus 0.95+/-0.05 for noncharged membranes, p < 0.05). The discrepancy between ID and UK could be related with the difference in the blood distribution volume of urea and that of electrolytes. The good correlation provides the major argument for ID being used as a monitoring parameter of the delivered dialysis dose. Having integrated the discrepancy between ID and UK, prescription can be guided by ID for delivering the adequate normalized dialysis dose as defined by Kt/V.