NDT Advance Access published online on December 21, 2007
Nephrology Dialysis Transplantation, doi:10.1093/ndt/gfm916
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Differences in Solute Removal by Two High-Flux Membranes of Nominally Similar Synthetic Polymers
1 Department of Medicine, University of Louisville, Louisville, KY, USA 2 Department of Renal Medicine, University Hospital Birmingham, QEMC, Birmingham, UK
Correspondence and offprint requests to: Richard A. Ward, Kidney Disease Program, University of Louisville, 615 S. Preston Street, Louisville, KY 40202-1718, USA. Tel: +1-502-852-5757; Fax: +1-502-852-7643; E-mail: richard.ward{at}louisville.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReference |
|---|
Background. Membranes fabricated from nominally similar polymers may be markedly different in chemical composition, morphology and geometry. To examine the relative importance of these factors to dialyzer performance, the removal of small and large uraemic toxins was determined for dialyzers containing ‘polysulfone’ membranes of different composition and morphology, with and without fibre undulations.
Methods. Total removal and instantaneous clearances of urea, phosphorus, β2-microglobulin, leptin, angiogenin, complement factor D and immunoglobulin 
light chain were determined in randomized cross-over studies. Total solute removal was assessed from the pre- to post-dialysis change in plasma concentration and the total amount of solute recovered in the dialysate. Trapping of solute at the membrane was determined as the difference between solute lost from plasma water and solute recovered in the dialysate.
Results. Total removal of urea and phosphorus was independent of the membrane composition and structure. Large molecule removal differed significantly between the two membranes, particularly for β2-microglobulin. The importance of trapping at the membrane as a mechanism of β2-microglobulin removal also differed significantly between the two membranes, with trapping being less important for the membrane with the greatest β2-microglobulin removal. As molecular size increased, the contribution of trapping at the membrane to solute removal increased and the difference between the two membranes decreased.
Conclusions. High-flux membranes fabricated from nominally similar polymers may differ significantly in their ability to remove low molecular weight protein uraemic toxins.
Keywords: adsorption; clearance; haemodialysis; low molecular weight proteins; membrane
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReference |
|---|
The current standard for adequate dialysis is based on the removal of urea [1]. Urea removal by haemodialysis depends mostly on the blood and dialysate flow rates and treatment time; the choice of dialyzer has a much smaller impact with dialyzers currently in clinical use. However, recent reports serve as a reminder that other molecules with potentially different kinetics of removal contribute importantly to long-term morbidity and mortality in haemodialysis patients. Several studies have shown that accumulation of β2-microglobulin (11.8 kDa) contributes to the development of haemodialysis-associated amyloid disease [2,3], while a retrospective analysis of the USRDS database suggested that increased middle molecule clearance by dialyzers is associated with a reduction in mortality [4]. Such observations were a motivation for initiating the NIH HEMO study, which examined the effect of increased large molecule clearance on morbidity and mortality in a randomized, prospective, multicentre study [5]. While the HEMO study did not find a significant impact of high-flux membranes on all cause mortality, on post hoc analysis the risk of cardiovascular death was significantly reduced in those patients treated with high-flux membranes compared to those treated with low-flux membranes [5] and an increase in pre-dialysis β2-microglobulin concentration was associated with a significant increase in relative risk of death after adjustment for residual renal function and dialysis vintage [6].
As the size of uraemic toxins increases, the choice of dialyzer becomes a more important determinant of their removal. The mass transfer properties of a dialyzer depend on the membrane surface area and permeability and the resistance to mass transfer in the boundary layers that form on the blood and dialysate sides of the membrane. The apparent permeability of a membrane to a solute reflects the relative contributions of three different modes of solute removal, diffusion and convection of the solute across the membrane and trapping of the solute at the membrane surface [7]. Which of these mechanisms predominates for a given membrane depends on the membrane composition and morphology. Boundary layer resistance depends on the thickness of the boundary layer, which is a function of the flow pattern in the blood and dialysate compartments. Apart from increasing the blood flow rate, it is difficult to modify the boundary layer thickness on the blood side of the membrane. The boundary layer thickness on the dialysate side of the membrane may, however, be reduced by altering the flow pattern in the dialysate compartment. Increases in the dialysate flow rate [8] and the addition of undulations to the fibres or inclusion of spacer yarns in the fibre bundle [9] have all been used to improve dialysate flow distribution and clearance.
The relative importance of these different strategies to increasing large solute removal remains unclear. The objective of this study was to compare and contrast the effect of differences in membrane composition and morphology on small and large solute removal.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReference |
|---|
Study design
Two separate studies were performed with dialyzers containing two different ‘polysulfone (PS)’ membranes, one based on polyarylethersulfone (PAES) and the other based on PS. Each study followed a randomized, cross-over design. Subjects underwent six treatments in the first study, three treatments with a Polyflux 17S dialyzer (PAES membrane, Gambro Renal Products, Lakewood, CO, USA) and three treatments with a F70NR dialyzer (PS membrane, Fresenius Medical Care, Lexington, MA, USA). Subjects underwent nine treatments in the second study, three treatments with a Polyflux 170H dialyzer (PAES membrane, Gambro), three treatments with a Polyflux 210H dialyzer (PAES membrane, Gambro) and three treatments with an Optiflux F200NR dialyzer (PS membrane, Fresenius). Each dialyzer was used for three consecutive treatments and the order in which the dialyzers were used was random. A new dialyzer was used for each study treatment; there was no dialyzer reuse. Details of the dialyzers are presented in Table 1.
|
In both studies, dialyzer performance was evaluated during the second and third treatments with each dialyzer using the nominal blood and dialysate flow rates of 400 mL/min and 800 mL/min, respectively. Actual blood flow rates were verified by using an ultrasonic flow meter (model HD01, Transonic, Ithica, NY, USA) and by blood pump calibration. The second treatment with each dialyzer was used to assess overall solute removal by measuring the pre- to post-dialysis change in solute concentration and the total amount of solute appearing in the dialysate. The third treatment with each dialyzer was used to assess instantaneous solute removal by measuring clearances during the first hour of dialysis. The instantaneous clearances were measured at a net ultrafiltration rate of zero. Urea and phosphorus were used as representative small solutes, while large solutes included β2-microglobulin (11.8 kDa), angiogenin (14.4 kDa), leptin (16 kDa), immunoglobulin
light chain (22.5), complement factor D (24 kDa) and albumin (
66 kDa). Treatments were performed using model 2008H dialysis delivery systems (Fresenius) in the first study and Phoenix dialysis delivery systems (Gambro) in the second study. Anticoagulation was achieved with unfractionated heparin administered as an initial loading dose followed by a constant infusion. Ultrafiltration rates were set according to the clinical needs of the patient.
Patients
Twelve patients were enrolled in each study. For inclusion in the study, patients were required to be older than 18 years of age and have a stable haemodialysis prescription. They must have been dialyzing through a native fistula or Gore–Tex graft capable of providing a blood flow of 400 mL/min and have had a fluid removal requirement of no more than 3 L per treatment. Exclusion criteria included non-compliance with their dialysis prescription, a haematocrit <28% or an active infection. Patients enrolled in the first study were routinely treated with dialyzers containing high-flux PS membrane (F80, Fresenius), while patients enrolled in the second study were routinely treated with dialyzers containing the PAES membrane (Polyflux 21R, Gambro). The Human Studies Committee of the University of Louisville reviewed the study protocols and all patients provided informed consent before participating in the study. Details of the patients participating in the two studies are given in Table 2.
|
Measurements of solute removal
Measurement of solute removal over a complete dialysis (overall solute removal). The change in solute concentration over the entire treatment was determined from pre- and post-dialysis blood samples. Pre-dialysis blood samples were drawn from the access needle immediately following needle insertion. Post-dialysis blood samples were drawn from the arterial blood line 20 s after reducing the blood flow to 80 mL/min to mitigate any access recirculation. An integrated sample of the effluent dialysate was collected to determine the total amount of a solute appearing in the dialysate [10]. A pump was used to sample the effluent dialysate continuously at a rate of 1 mL/min from a sample cell placed in the outlet dialysate line immediately before it entered the drain. At the end of dialysis, the integrated sample of effluent dialysate was mixed and subsamples were stored at –70°C until analysis.
The pre- to post-dialysis change in solute concentration was calculated from
|
|
where CPre and CPost are the pre- and post-dialysis solute concentrations, respectively. The post-dialysis concentrations of β2-microglobulin, angiogenin, leptin, complement factor D and immunoglobulin light chain were corrected for haemoconcentration according to Bergström and Wehle [11] before calculating the pre- to post-dialysis change in concentration. The total amount of solute recovered in the dialysate (mD) was calculated from
|
|
where CDS is the concentration in the dialysate sample, QD is the inlet dialysate flow rate, TD is the treatment duration and UF is the total ultrafiltrate volume for the treatment.
An overall clearance of β2-microglobulin, Kβ2m, was calculated from the change in serum concentration from pre- to post-dialysis as described previously [12] using the following equation:
|
|
where Cpost is not corrected for haemoconcentration and Vβ2m is the post-dialysis extracellular fluid volume calculated as one-third of total body water estimated according to Watson and colleagues [13].
The methods of Daugirdas and Schneditz [14] were used to calculate eKt/Vurea, while Kt/V for β2-microglobulin was calculated from Kβ2m, Vβ2m and TD [6].
Measurement of instantaneous solute removal.
Solute clearances were determined by standard methods. Briefly, blood and dialysate flow rates were set at the desired values and the net ultrafiltration rate was set to zero, using the delivery system displays. After allowing 15 min for a steady state to be established, blood samples were drawn from the inlet and outlet bloodlines of the dialyzer and a dialysate sample obtained from the outlet line of the dialyzer using a specially inserted sample port. Plasma was obtained by centrifugation. Before plasma separation, a well-mixed aliquot was taken from the inlet blood sample and used to determine haematocrit by the micro-haematocrit method. Plasma and dialysate samples were stored at –70°C until analysis.
Instantaneous dialyzer clearances (K) were calculated using the following standard equation [15]:
|
|
where QBE is the effective blood flow rate, QUF is the instantaneous ultrafiltration rate, CPi and CPo are the plasma concentrations at the inlet and outlet of the dialyzer, respectively, and CDo is the outlet dialysate concentration. QBE was assumed to be the blood water flow rate for urea and the plasma water flow rate for β2-microglobulin, angiogenin, leptin and immunoglobulin light chain. The blood water flow rate (QBW) and the plasma water flow rate (QPW) were calculated from
|
|
where QB is the blood flow rate, Hct is the fractional haematocrit and the constants 0.93 and 0.86 correct for the protein content of plasma water and the fractional water content of red blood cells, respectively. Phosphorus clearances were calculated using the dialysate concentration and flow rate to avoid the uncertainty associated with any plasma water—red blood cell disequilibrium that developed as blood transited the dialyzer.
The product of mass transfer coefficient and membrane area for urea, KoA, was determined from the instantaneous clearance and the blood water and dialysate flow rates according to Michaels [15]
|
|
The mass transfer coefficient, Ko, was calculated by dividing KoA by the nominal membrane surface area of the dialyzer.
The instantaneous rate of trapping of proteins at the dialyzer membrane was calculated as the difference between the amount of protein lost from the plasma water and the amount recovered in the effluent dialysate:
|
|
Analytical methods
Urea concentrations were determined by standard clinical laboratory methods. Phosphorus concentrations were measured by the method of Chen et al. [16] adapted for a microplate spectrophotometer. β2-microglobulin concentrations were measured by an enzyme immunoassay (R&D Systems, Minneapolis, MN, USA) (study 1) or an immunometric assay (Immulite, Diagnostic Products Corporation, Los Angeles, CA, USA) (study 2). Concentrations of angiogenin and leptin were measured by enzyme immunoassay (R&D Systems). Concentrations of immunoglobulin light chain were measured by immunoassay (Freelite, The Binding Site, Birmingham, UK). Complement factor D concentrations were measured by ELISA [17]. Dialysate albumin concentrations were measured by enzyme immunoassay (Alpco, Windham, NH, USA).
Statistical methods
Differences in solute removal between dialyzers were assessed by analysis of variance (SPSS for Windows, version 12, SPSS, Chicago, IL, USA). Where significant differences were found (P < 0.05), differences between individual pairs of dialyzers were determined using the Student–Newman–Keuls correction for multiple comparisons. Data were presented as mean ± SEM for n observations.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReference |
|---|
Within each study, there were no differences between dialyzers with respect to treatment duration, actual blood flow rate, blood water flow rate, plasma water flow rate or ultrafiltration volume (Table 3). However, the ultrafiltration volume for the treatments used to determine total solute removal in the second study was significantly greater than that in the first study (3.1 ± 0.2 L and 4.4 ± 0.2 L in the first and second studies, respectively, P = 0.05). The total heparin dose for a treatment averaged 9100 ± 1200 IU and did not differ between dialyzers in either study or between studies. A urea mass balance was used to evaluate the quality of the data used to calculate instantaneous clearances. Overall, the mean absolute urea mass balance error was 5 ± 2% and did not differ between dialyzers in either study.
|
Small molecule removal was unaffected by membrane composition.
Measures of the removal of urea by the different dialyzers are presented in Tables 4 and 5. Within each study, there were no differences between dialyzers regarding overall urea removal, as assessed by the pre- to post-dialysis reduction in the concentration of urea, the amount of urea recovered in the dialysate or eKt/Vurea. Within each study, instantaneous clearances of urea did not differ between dialyzers with comparable membrane surface areas; however, in the second study the instantaneous clearance of urea was significantly less for the Polyflux 170H, which has a membrane surface area of 1.7 m2, than for the Optiflux F200NR and Polyflux 210H, which have membrane surface areas of 2.0 m2 and 2.1 m2, respectively (Table 5). A similar result was found for KoA (Table 5). Fibre undulations have been proposed as means of improving dialysate flow distribution and, thereby, the clearance of small solutes [9]. KoA provides a patient- and flow rate-independent index of solute removal for a dialyzer. Comparison of urea KoA for the dialyzers that contained fibres without undulations (study 1) with urea KoA for the dialyzers that contained fibres with undulations (Study 2) showed significantly greater values for the Optiflux F200NR and Polyflux 210H dialyzers containing fibres with undulations than for the F70NR and Polyflux 17S dialyzers containing fibres without undulations (Tables 4 and 5). However, the Optiflux F200NR and Polyflux 210H contain more membrane surface area than the F70NR and Polyflux 17S, and when differences in the membrane surface area were eliminated by dividing KoA by the nominal membrane surface area, the only dialyzer with a significantly greater urea Ko was Optiflux F200NR.
|
|
In general, the pattern of phosphorus removal was similar to that observed for urea, with no differences being observed between the dialyzers within each study (Tables 6 and 7). More phosphorus was recovered in the dialysate with dialyzers containing fibres with undulations (Optiflux F200NR, Polyflux 170H and Polyflux 210H) than with dialyzers containing membranes without undulations (F70NR and Polyflux 17S). However, this difference may be the result of slightly higher pre-dialysis phosphorus concentrations and longer treatment times in the second study than in the first study. A trend for dialyzers containing fibres with undulations to have higher instantaneous clearances of phosphorus was observed, although the only significant difference was between Optiflux F200NR and Polyflux 17S and that difference could result from a difference in membrane surface area.
|
|
Large molecule removal depends on membrane composition and morphology.
Pre-dialysis serum β2-microglobulin concentrations were significantly higher in the first study than in the second study (Tables 8 and 9), possibly reflecting differences in the dialyzer used for the patients’ routine treatments. Tables 8 and 9 present four different measures of total β2-microglobulin removal, the pre- to post-dialysis reduction in concentration, total β2-microglobulin recovered in the dialysate, the overall clearance calculated according to Leypoldt and colleagues [12] and Kt/Vβ2m. In both studies, total removal of β2-microglobulin was significantly greater with the dialyzers containing PAES membranes than with the dialyzers containing PS membranes by all four measures. Instantaneous clearances of β2-microglobulin determined during the first hour of dialysis showed the same pattern observed for total removal, with higher clearances being obtained for the PAES membrane than for the PS membrane.
|
|
Similar to β2-microglobulin, the total removal of other large molecules differed between the PAES membrane and the PS membrane in both studies. In study 1, significantly more angiogenin, leptin and complement factor D was recovered in the dialysate with the PAES membrane than with the PS membrane (Table 10), while in the second study significantly more immunoglobulin
light chain was recovered in the dialysate with the PAES membrane than with the PS membrane (Table 11), even though the difference in pre- to post-dialysis reduction in concentration between dialyzers containing PS membrane and PAES membrane reached significance only for complement factor D and immunoglobulin light chains. As molecular weight increased, instantaneous clearances decreased and the difference between the two membranes was not significant for angiogenin, leptin and complement factor D (Table 10). The change in concentration from the inlet to the outlet of the dialyzer was too small to allow a meaningful estimation of the clearance of immunoglobulin light chain.
|
|
The mechanism of large molecule removal depends on membrane composition and morphology.
Removal of small proteins, such as β2-microglobulin, occurs by convection through the membrane, trapping at the membrane surface and to a lesser extent by diffusion across the membrane [7]. The contribution of trapping at the membrane surface to solute removal was determined by calculating the difference between the amount of solute lost from plasma water and the amount of solute appearing in the dialysate at an instant. The data in Tables 891011 show that trapping at the membrane contributed to large molecule removal for both the PS and PAES membranes. The contribution of trapping to solute removal was significantly greater for the PS membrane than for the PAES membrane for β2-microglobulin. As molecular weight increased, the contribution of trapping to solute removal increased and the difference between the two membranes decreased so that the contribution of trapping at the membrane to the removal of angiogenin, leptin and complement factor D was not significantly different between the PS and PAES membranes. Again, the changes in concentration of immunoglobulin
light chains within the dialyzer at an instant were too small to allow a reliable estimate of trapping to be calculated. The amount of albumin in the dialysate was less than the limit of detection (46 mg/4 h) for the dialyzers containing the PS membrane. Measurable albumin was found in the dialysate for the dialyzers containing the PAES membrane; however, this amount never exceeded 500 mg/4 h, even with an ultrafiltration rate as high as 20 mL/min. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReference |
|---|
High-flux haemodialysis remains the most common therapeutic approach to eliminating low molecular weight proteins, which are believed to comprise an important class of uraemic toxins. The results of our studies show that high-flux membranes of nominally similar materials can differ significantly in their ability to remove low molecular weight proteins, most likely because of differences in composition and morphology. In contrast, we show that removal of small solutes is relatively independent of the membrane material, but may be marginally improved by strategies designed to enhance flow distribution through the dialysate compartment of the dialyzer.
Membranes based on synthetic polymers are in widespread use for haemodialysis therapy. The most commonly used membranes are frequently described as being fabricated from ‘PS’. However, there are significant differences between different ‘PS’ membranes. The PAES membranes used in these studies have an asymmetric three-layer structure whereas the PS membrane has an inner skin layer and uniform sponge support structure. The chemical composition of the PAES membrane also differs from that of the PS membrane. The PAES membrane is fabricated using polyarylethersulfone, rather than PS, and also contains some polyamide, which is included to enhance endotoxin retention by the membrane. Both membranes contain some amount of polyvinylpyrrolidone to impart hydrophilicity to the membrane and make it permeable to water. Differences in chemical composition, such as the amount of polyvinylpyrrolidone, and in the precipitation process used in the manufacture of the membrane may influence the final morphology and physicochemical properties of the membrane.
Greater amounts of all five low molecular weight proteins considered in our studies were recovered in the dialysate when treatments were performed with dialyzers containing the PAES membrane than when dialyzers containing the PS membrane were used (Tables 891011). In addition, the PAES membrane provided a significantly greater reduction in the plasma concentration of β2-microglobulin from pre- to post-dialysis and significantly greater clearances of β2-microglobulin than did the PS membrane (Tables 8 and 9). A similar finding was obtained for the pre- to post-dialysis reduction in plasma concentration of complement factor D (Table 10) and immunoglobulin light chain (Table 11).
As well as demonstrating differences in the removal of low molecular weight proteins between the PAES and PS membranes, the data in Table 89Table 10 show differences in the mechanism of that removal. It has long been known that, in addition to removal by diffusion and convection through the membrane, adsorption plays a major role in the removal of β2-microglobulin by some synthetic membranes, such as a copolymer of acrylonitrile and sodium methallyl sulfonate (AN69) and polymethylmethacrylate [7]. Proteins may also adsorb to synthetic membranes based on PS because of the hydrophobic nature of the polymer [7]. In addition, ultrafiltration may carry large solutes into the pores of the membrane where they become trapped by steric hindrance. The data in Table 89Table 10 show that trapping of solute at the membrane contributed to the removal of low molecular weight proteins with both the PS and PAES membranes based on the difference between the rate of solute loss from plasma water and the rate of solute appearance in the dialysate. The importance of trapping in the membrane pores may depend on the degree of filtration and back-filtration that occurs in the dialyzer. The amount of filtration and back-filtration depends on the net ultrafiltration rate and the inner diameter of the fibres, with a smaller inner diameter promoting more filtration and higher clearances of large solutes [18]. In practice, it is difficult to separate adsorption to the membrane surface from trapping in the pores. However, it seems unlikely that the differences in large solute removal between the two membranes can be attributed to the differences in membrane geometry or operational parameters. Apart from the Polyflux 170H in the second study, within each study the dialyzers had comparable membrane surface areas and net ultrafiltration rates were not different. Moreover, while the PS membrane had a slightly smaller inner diameter than the PAES membrane (200 µm versus 215 µm), in general it provided lower clearances of large solutes than the PAES membrane.
Trapping of β2-microglobulin at the membrane was a significantly more important mechanism of removal with the PS membrane than with the PAES membrane. This finding is consistent with the observation of Hoenich and colleagues that the PAES membrane had a significantly higher sieving coefficient for β2-microglobulin than did the PS membrane [19]. The relative contribution of trapping at the membrane to total solute removal increased with increasing molecular weight. Indeed, so little solute was detected in the dialysate that trapping at the membrane accounted for essentially all of the removal of leptin, complement factor D and immunoglobulin light chain for the PS membrane and for the majority of the removal of immunoglobulin light chain for the PAES membrane. The latter observation is consistent with a previous report from Hutchison and colleagues [20].
The PAES and PS membranes did not differ significantly, however, in their ability to decrease the concentrations of angiogenin or leptin from pre- to post-dialysis (Table 7). These observations suggest that factors other than the membrane may influence the removal of low molecular weight proteins. Among these factors are likely to be the properties of the protein, including molecular size, charge, hydrophobicity and any post-translational modifications associated with uraemia that may affect the rigidity of the protein. Collectively, these properties may influence the interaction of the protein with the membrane material and affect the ability of the protein to pass through the membrane by diffusion or convection or become trapped in the membrane structure.
Whether the differences between the two membranes in β2-microglobulin removal by trapping arise from the differences in the chemical composition or the morphology of the membrane, or both, remains to be determined. One possible explanation for the lower level of trapping at the membrane seen with the PAES membrane than with the PS membrane is that the physicochemical heterogeneity provided by the hydrophilic–hydrophobic microdomains present at the surface of the PAES membrane impedes the formation of stable hydrophobic interactions between proteins and the membrane surface [21]. The importance of the nature of the surface rather than the cross-sectional structure of the membrane in determining the trapping and clearance of low molecular weight proteins is supported by observations with other membranes based on PS. Mandolfo and colleagues [22] reported similar reductions in the pre- to post-dialysis concentrations and similar instantaneous clearances for β2-microglobulin and leptin with the Asahi APS PS membrane to those obtained in this study and a previous study [23] with the PAES membrane. The Asahi APS PS membrane is reported by its manufacturer to have a cross-sectional structure similar to that of the PS membrane, but to have a hydrophilic gel layer on the membrane surface that inhibits protein adsorption.
In contrast to our findings for low molecular weight proteins, the clearance of urea and phosphorus differed little between the PAES and PS membranes after adjusting for membrane surface area. Strategies designed to improve flow distribution through the dialysate compartment, such as higher dialysate flow rates [8], the inclusion of spacer yarns [9] and the addition of undulations to the fibre [9], have been shown to improve small solute clearance. While we found higher instantaneous urea clearances for the dialyzers with fibre undulations than for those without fibre undulations, much of this difference could be accounted for by differences in the membrane surface area. Moreover, any ability of fibre undulations to increase the instantaneous clearance of urea had little impact on overall urea removal, as assessed by the pre- to post-dialysis reduction in concentration, eKt/V and total urea recovered in the dialysate (Tables 4 and 5). Total urea removal during a dialysis treatment depends not only on the dialyzer clearance, but also on the treatment time and the pre-dialysis urea concentration and in our studies pre-dialysis urea concentrations were lower for those treatments performed with the dialyzers having the higher urea clearances. In contrast to urea, significantly more phosphorus was recovered in the dialysate with dialyzers containing fibres with undulations, even though there was only a trend for fibre undulations to increase the instantaneous clearance of phosphorus (Tables 6 and 7). However, the difference in total phosphorus removal may be a consequence of slightly higher serum phosphorus concentrations and longer treatment times in the second study, rather than the presence of fibre undulations. Overall, incorporation of fibre undulations appeared to make little difference to small solute removal with either membrane.
From these studies, we conclude that dialyzers containing membranes fabricated from nominally similar polymers may differ significantly in their ability to remove larger uraemic toxins, such as low molecular weight proteins.
|
Acknowledgments |
|---|
The authors acknowledge the expert technical assistance of Karen Brinkley and Justin Utz. This work was supported by grants from Gambro Renal Products.
Conflict of interest statement. The results presented in this paper have not been published previously in whole or part, except in abstract format.
|
Reference |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReference |
|---|
- NKF-DOQI. Clinical Practice Guidelines for Hemodialysis Adequacy (1997) New York: National Kidney Foundation.
- Küchle C, Fricke H, Held E, et al. High-flux hemodialysis postpones clinical manifestation of dialysis-related amyloidosis. Am J Nephrol (1996) 16:484–488.[Web of Science][Medline]
- Koda Y, Nishi S-I, Miyazaki S, et al. Switch from conventional to high-flux membrane reduces the risk of carpal tunnel syndrome and mortality of hemodialysis patients. Kidney Int (1997) 52:1096–1101.[Web of Science][Medline]
- Leypoldt JK, Cheung AK, Carroll CE, et al. Effect of dialysis membranes and middle molecule removal on chronic hemodialysis patient survival. Am J Kidney Dis (1999) 33:349–355.[Web of Science][Medline]
- Eknoyan G, Beck GJ, Cheung AK, et al. Effect of dialysis dose and membrane flux in maintenance hemodialysis. N Engl J Med (2002) 347:2010–2019.
[Abstract/Free Full Text] - Cheung AK, Rocco MV, Yan G, et al. Serum β-2 microglobulin levels predict mortality in dialysis patients: results of the HEMO study. J Am Soc Nephrol (2006) 17:546–555.
[Abstract/Free Full Text] - Klinke B, Röckel A, Abdelhamid S, et al. Transmembrane transport and adsorption of beta-2-microglobulin during hemodialysis using polysulfone, polyacrylonitrile, polymethylmethacrylate and cuprammonium rayon membranes. Int J Artif Organs (1989) 12:697–702.[Web of Science][Medline]
- Ouseph R, Ward RA. Increasing dialysate flow rate increases dialyzer urea mass transfer-area coefficients during clinical use. Am J Kidney Dis (2001) 37:316–320.[Web of Science][Medline]
- Ronco C, Brendolan A, Crepaldi C, et al. Blood and dialysate flow distributions in hollow-fiber hemodialyzers analyzed by computerized helical scanning technique. J Am Soc Nephrol (2002) 13:S53–S61.
[Abstract/Free Full Text] - Ing TS, Yu AW, Wong FKM, et al. Collection of a representative fraction of total spent dialysate. Am J Kidney Dis (1995) 25:810–812.[Web of Science][Medline]
- Bergström J, Wehle B. No change in corrected β2-microglobulin concentration after cuprophane haemodialysis. Lancet (1987) 1:628–629.[Medline]
- Leypoldt JK, Cheung AK, Deeter RB. Single compartment models for evaluating β2-microglobulin clearance during hemodialysis. ASAIO J (1997) 43:904–909.[Web of Science][Medline]
- Watson PE, Watson ID, Batt RD. Total body water volumes for adult males and females estimated from simple anthropometric measurements. Am J Clin Nutr (1980) 33:27–39.
[Abstract/Free Full Text] - Daugirdas JT, Schneditz D. Overestimation of hemodialysis dose depends on dialysis efficiency by regional blood flow but not by conventional two pool urea kinetic analysis. ASAIO J (1995) 41:M719–M724.[Medline]
- Michaels AS. Operating parameters and performance criteria for hemodialyzers and other membrane-separation devices. Trans Am Soc Artif Int Organs (1966) 12:387–392.[Web of Science][Medline]
- Chen PS, Toribara TY, Warner H. Microdetermination of phosphorus. Anal Chem (1956) 28:1756–1758.
- Oppermann M, Baumgarten H, Brandt E, et al. Quantitation of components of the alternative pathway of complement (APC) by enzyme-linked immunosorbent assays. J Immunol Methods (1990) 133:181–190.[CrossRef][Web of Science][Medline]
- Ronco C, Brendolan A, Lupi A, et al. Effects of a reduced inner diameter of hollow fibers in hemodialyzers. Kidney Int (2000) 58:809–817.[CrossRef][Web of Science][Medline]
- Hoenich NA, Stamp S, Roberts SJ. A microdomain-structured synthetic high-flux hollow-fiber membrane for renal replacement therapy. ASAIO J (2000) 46:70–75.[CrossRef][Web of Science][Medline]
- Hutchison CA, Cockwell P, Reid S, et al. Efficient removal of immunoglobulin free light chains by hemodialysis for multiple myeloma: in vitro and in vivo studies. J Am Soc Nephrol (2007) 18:886–895.
[Abstract/Free Full Text] - Ronco C, Crepaldi C, Brendolan A, et al. Evolution of synthetic membranes for blood purification: the case of the Polyflux family. Nephrol Dial Transplant (2003) 18(Suppl_7):vii10–vii20.[Medline]
- Mandolfo S, Borlandelli S, Imbasciati E. Leptin and β2-microglobulin kinetics with three different dialysis modalities. Int J Artif Organs (2006) 29:949–955.[Web of Science][Medline]
- Wiesholzer M, Harm F, Hauser A-C, et al. Inappropriately high plasma leptin levels in obese haemodialysis patients can be reduced by high flux haemodialysis and haemodiafiltration. Clin Sci (1998) 94:431–435.[Web of Science][Medline]
Accepted in revised form: 5.12.07
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  E. L. Penne, N. C. van der Weerd, M. L. Bots, M. A. van den Dorpel, M. P. C. Grooteman, R. Levesque, M. J. Nube, P. M. ter Wee, P. J. Blankestijn, and On behalf of the CONTRAST investigators Patient- and treatment-related determinants of convective volume in post-dilution haemodiafiltration in clinical practice Nephrol. Dial. Transplant., November 1, 2009; 24(11): 3493 - 3499. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Meert, S. Eloot, M.-A. Waterloos, M. Van Landschoot, A. Dhondt, G. Glorieux, I. Ledebo, and R. Vanholder Effective removal of protein-bound uraemic solutes by different convective strategies: a prospective trial Nephrol. Dial. Transplant., February 1, 2009; 24(2): 562 - 570. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||