Forecasting correct sodium balance in hemodiafiltration procedures involving infusions

Four patients, stable on acetate hemodialysis (AHD), were switched to acetate-free biofiltration (AFB) which differs from AHD and bicarbonate hemodialysis (BHD) in that the dialysate contains no buffer, which is given intravenously as a hypertonic (1/6 M) Na bicarbonate solution. Within the 1st month the patients developed thirst and hypertension attributed to a positive Na balance. The aim of this investigation was to check this (1) by a study based on the predictable changes induced in the body compartments of 13 patients by the infusion and ultrafiltration (UF) of a hypertonic solution and (2) by direct determination and calculation of 28 Na mass balances in BHD and AFB. The theoretical model indicated that infusion of 4.87 liters of a 166.7 mEq/l Na bicarbonate solution and UF of the same amount caused a positive balance of 233 mosm of Na. The Na mass balances showed a relationship between Na transmembrane gradient and loss or gain of Na in both methods (p less than 0.0001). The slopes of the regression lines were not significantly different but there was a highly significant difference between the y axis intercepts (p less than 0.0001), which indicates that the same Na transmembrane gradient that gives no net change of Na in BHD, induces a net gain of 240 mosm (120 mEq of Na) in AFB and that to obtain the same Na balance dialysate Na should be reduced by about 8 mEq/l in AFB. These data are the same as the theoretical forecast which could be extended to all hemodiafiltration methods in which solutions of any tonicity have to be infused, in order to correctly predict the Na balance.