MO656: Validity of the Hydrogen Ion Mobilization Model During Haemodialysis with Time-Dependent Dialysate Bicarbonate Concentrations

Abstract

Abstract BACKGROUND AND AIMS The hydrogen ion (H+) mobilization model has been previously shown to describe blood bicarbonate (HCO3) kinetics during haemodialysis (HD), but this model has only been evaluated when the dialysate HCO3 concentration ([HCO3]) was constant throughout the treatment [1]. To further assess the H+ mobilization model, we tested its ability to describe blood HCO3 kinetics during HD treatments with a time-dependent dialysate [HCO3]. METHOD The H+ mobilization model describes the time dependence of blood [HCO3] assuming HCO3 is distributed in the extracellular fluid volume and removed from that space by mobilization of H+ from buffers and other sources as previously described [1]. HCO3 transfer across the haemodialyzer during HD is assumed governed primarily by diffusion and the difference between dialysate and blood [HCO3]. In this work, we evaluated this model by comparing its predictions with data from a recent clinical study where blood [HCO3] was measured in 11 chronic, thrice-weekly HD patients (5 male, 6 female) during 4-h treatments with i) constant dialysate [HCO3] of 35 mEq/L; ii) dialysate [HCO3] of 35 mEq/L for the first 2-h and 30 mEq/L for the second 2-h; and iii) dialysate [HCO3] of 30 mEq/L for the first 2-h and 35 mEq/L for the second 2-h. Treatment interventions were fixed during a given week, and arterial blood samples were obtained at the start of treatment and every hour during the first, second and third treatments of the week. Blood acid-base composition was determined using a blood gas analyzer; [HCO3] was calculated from measured pH and partial pressure of carbon dioxide using the Henderson–Hasselbalch equation. The H+ mobilization model was used to determine the H+ mobilization parameter (mH+) by comparing the time dependence of the measured blood [HCO3] with that predicted by the model using nonlinear regression; a total of 66 HD treatments were monitored. Determination of similar mH+ values for each patient, independent of the time-dependent dialysate [HCO3], would demonstrate model validity. RESULTS Example fits of model predictions to blood HCO3 kinetic data from one patient for all three treatment types are shown in the Figure. For all patients, mH+ values during treatments A, B and C were 139 ± 68 (mean ± standard deviation) mL/min, 178 ± 96 mL/min and 156 ± 55 mL/min; they were not statistically different from each other (P = 0.43). The sum of the squared differences between the measured [HCO3] and that predicted by the model (mEq2/L2) was similar during treatments A, B and C (1.48 versus 1.85 versus 1.61), indicating a similar degree of model fit to the data. CONCLUSION This work supports the validity of the H+ mobilization model to describe intradialytic blood HCO3 kinetics with a time-dependent dialysate [HCO3]. Further work is needed to validate the model during post-dialytic and interdialytic periods.