DATA-ACQUISITION SYSTEM FOR DIALYSIS MACHINES - A MODEL FOR MEMBRANE HYDRAULIC PERMEABILITY

Citation
F. Vaussenat et al., DATA-ACQUISITION SYSTEM FOR DIALYSIS MACHINES - A MODEL FOR MEMBRANE HYDRAULIC PERMEABILITY, ASAIO journal, 43(6), 1997, pp. 910-915
Citations number
10
Categorie Soggetti
Engineering, Biomedical
Journal title
ISSN journal
10582916
Volume
43
Issue
6
Year of publication
1997
Pages
910 - 915
Database
ISI
SICI code
1058-2916(1997)43:6<910:DSFDM->2.0.ZU;2-9
Abstract
Membrane permeability is a key determinant of dialyzer performance; in vivo, membrane hydraulic permeability is affected by the formation of a protein cake on its surface, reducing ultrafiltration and convectiv e fluxes. The purpose of this work was to evaluate the real hydraulic permeability of high flux polysulfone membrane under conditions of hem odiafiltration, and to consequently develop a mathematical model to es timate ultrafiltration Kuf and protein adsorption Kc coefficients. The DIB08 data acquisition system adapted to the Fresenius 2008E dialysis machine (Fresenius, Bad Homburg, Germany) allowed the recording of us eful information for dialysis quantification, which was then processed by a bedside computer. The system was able to evaluate Kuf(t) profile , by calculation from the transmembrane pressure over time (TMP(t)) an d ultrafiltration rate (Quf): Kuf(t) = Quf/TMP (t). Subsequent modelin g of Kuf involved the determination of two key parameters: Kufhd (dial yzer permeability during diffusion only) (in mL/h/mmHg), and Kc(protei n adsorption coefficient) (in mL/h/mmHg(2)). The model chosen was the following: Kuf(t) = Kuf(0) x (1 - (Kc/Kuf(0)) x In(t + 1)) where Kuf(0 ) represents the initial Kuf obtained at the beginning of the session. Thirty-one sessions were evaluated by real kinetic analysis, from whi ch the mathematical model was derived. It included 27 postdilutional o n-line hemodiafiltration and four hemodialysis sessions performed in f our patients with nonreused HF80s dialyzers. For the analysis, three s ubgroups were defined: Group 1, first session of the week (Monday or T uesday); Group 2, second session of the week (Wednesday or Thursday); and Group 3, third session of the week (Friday or Saturday). Results o f Kuf and Kc obtained by real kinetic analysis are presented. The midw eek session was associated with a higher membrane hydraulic permeabili ty, most likely relative to lesser ultrafiltration rates and an associ ated relative decrease in membrane protein coating, represented by Kc. The described data acquisition system allowed the assessment of real time membrane hydraulic permeability and the subsequent development of a mathematical model to estimate this fundamental parameter as it fun ctions to hemodialyzer performance.