Dialysis membrane-dependent removal of middle molecules during hemodiafiltration: the beta2-microglobulin/albumin relationship.

AIM: Current hemodialysis therapy modalities such as online hemodiafiltration (HDF) attempt to enhance solute removal over a wide molecular weight range through a combination of diffusion and convection. While the effects of variations of treatment modalities and conditions have been studied reasonably well, few studies have examined the efficacy of HDF to remove middle molecules in relation to the dialyzer and membrane characteristics. In this investigation, diverse high-flux dialyzers, covering a wide range of membrane permeabilities, were compared under identical in vivo conditions to assess their ability to eliminate larger uremic retention solutes (using beta2-microglobulin as a surrogate of middle molecules) without simultaneously causing excessive leakage of useful proteins such as albumin. PATIENTS AND METHODS: In a prospective, crossover study, 3 ESRD patients were treated with 8 different brands of high-flux dialyzers at 4 different ultrafiltration (UF)/substitution flow rates (QS: 0, 30, 60, 90 ml/min) in post-dilution HDF mode. Thus, each patient underwent 32 treatment sessions, with a total of 96 treatment sessions conducted during the entire clinical study. Albumin and beta2-microglobulin levels were measured in both, dialysate and blood. Both, albumin and beta2-microglobulin elimination was dependent upon the permeability of the dialysis membrane as well as on the ultrafiltration/substitution flow rates applied. RESULTS: At the maximum UF rate of 90 ml/min, the total albumin loss (measured in the dialysate) ranged from 300 mg/4 h (for the FLX-15 GWS dialyzers) to 7,000 mg/4 h (for the BS-1.3U dialyzers). Up to 50% reduction of albumin occurred within the first 30 minutes of the dialysis treatment, and the leakage of albumin increased exponentially with increasing UF rates as well as increasing transmembrane pressure (TMP). The various dialyzers could be classified according to their UFR-dependent beta2-m reduction rates (RR), into low (< 50%; FLX-15 GWS, CT 150G), medium (50-70%; Polyflux 14 S, BLS 814SD, H4) and high (> 70%; BS-1.3U, APS 650, FX 60) removers of middle molecules. One dialyzer type (CT 150G) showed extremely low beta2-m RR and relatively high albumin losses. Most membranes, however, showed either low albumin leakage coupled with low beta2-m removal, or high beta2-m RR but at the expense of considerable albumin leakage. Only 2 membrane types approached the desired balance between high to medium beta2-m RR while simultaneously restricting the albumin leakage especially at higher filtration/substitution rates. CONCLUSION: Our investigations demonstrate that not all dialysis membranes classified as "high-flux" are comparable in their ability to specifically and efficiently remove middle molecules, or curtail the unwanted excessive leakage of essential proteins from the patient's blood. Thus, the selection of appropriate high-flux dialyzers for specific patient requirements should be based more upon clinical evaluations and analyses rather than on product specifications alone.

Verification of the chemical composition and specifications of haemodialysis membranes by NMR and GPC-FTIR-coupled spectroscopy.

The chemical composition of a dialysis membrane is decisive towards determining its physical and biochemical properties--two fundamental determinants of the success of therapy offered to patients suffering from chronic renal failure. From the vast variety of synthetic polymers available, only a few are suitable for the manufacture of dialysis membranes that have to conform to the diverse demands of modern haemodialysis and related therapies. Recently, a membrane labelled as polyamide (Polyamide S) has caused some confusion to end-users in that the product specification for the membrane is given as 'polyarylethersulfone' or simply as Polyamide S membrane. As the chemical and physical properties of these two polymer types are distinctly different, it is unclear whether the functional characteristics of Polyamide S are to be attributed to polyamide, polyarylethersulfone, or, to both polymers. We therefore undertook investigations to ascertain the exact chemical nature of the Polyamide S membrane using a series of chemical analytical tools and an appropriate polyamide reference. The analytical techniques were conventional gel permeation chromatography (GPC), GPC-FTIR coupled spectroscopy using dimethyl acetamide and hexafluoroisopropanol as solvents and nuclear magnetic resonance spectroscopy. Glass transition temperature measurements and quantitative elemental analysis were also carried out. None of the analytical techniques used showed any traces of polyamide in Polyamide S; no aliphatic or aromatic polyamide chemical entities were detected in any of the samples tested. The Polyamide S dialysis membrane thus comprises, solely, of polyarylethersulfone, which is also known as polyethersulfone.

Dialysis membranes today.

In recent years, hemodialytic therapies have evolved from the simple, diffusion-dependent removal of small molecular weight substances from blood to advanced therapy modalities involving the convective removal of larger uremic sloutes. The clinical benefits of removal of substances such as beta2-microglobulin (beta2-m) have been reported by several authors: elimination of large-molecular weight "uremic toxins" is now widely accepted as being beneficial to the overall quality of life of patients. This trend would not have been possible without parallel technical developments, especially that of new membranes having more open pore structures resulting in higher sieving coefficients and increased hydraulic permeability. Not all polymer types are suitable for the manufacture of high-flux membranes required for convective therapies in which large fluid volumes are exchanged. Amongst the more important criteria are: the selected polymer must be able to undergo steam sterilisation, have high endotoxin retention capabilities, be versatile for the fabrication of a range of hydraulic permeabilities and, of course, have high blood compatibility. The aim of this paper is, firstly, to review the major membrane development phases over the last quarter of a century. Secondly, the suitability of current membrane materials to meet the aforementioned requirements will be examined. Thirdly, in view of the recent, rapid proliferation of polysulfone-based membranes, dialysis membranes of the polysulfone 'family' are placed under scrutiny; membranes of this class represent a significant portion of the product portfolio of dialyser manufacturers today, yet, few end-users are able to distinguish between the salient features of the respective products because of a combination of confusing membrane nomenclature, classification, tradenames and product claims.

Surface topography and surface elemental composition analysis of Helixone, a new high-flux polysulfone dialysis membrane.

Modern dialysis membranes need to fulfil two basic requirements. Firstly, the membrane structure, defined in terms of the size, structure and distribution of the pores at the inner separating layer of the membrane must be such that uraemic solutes of a defined molecular-weight range are selectively removed. Secondly, the physical and chemical properties of the blood-contacting surface must be such that minimal blood-material interactions take place that could either affect the functioning of the membrane, or, cause adverse reactions for the patient. A new polysulfone dialysis membrane, Helixone, has been developed specifically for the elimination of larger uraemic toxins using convective therapy modalities such as haemodiafiltration. The membrane is characterised by the nanoscale modulation of the innermost surface structures that lead to significantly increased sieving coefficients for molecules such as beta2-microglobulin, while maintaining the extremely low albumin removal property of the high-flux Fresenius Polysulfone membrane. A recent publication (Ronco C, Bowry SK. Nanoscale modulation of the pore dimensions, size distribution and structure of a new polysulfone-based high-flux dialysis membrane. Int J Artif Organs 2001; 24: 726-35) described the characterisation of the membrane of Helixone in terms of the membrane wall structure- and permeation-related parameters. In this paper, we describe the analysis of membrane surface parameters that influence the biocompatibility as well as the functioning of a membrane. The degree of roughness and the type of chemical groups of a blood-contacting surface are two of the main determinants of the biocompatibility characteristics of a membrane. The surface elemental composition of Helixone was determined using electron spectroscopy for elemental analysis (ESCA) while the surface topography of the membrane was evaluated using atomic force microscopy (AFM). The analysis showed that Helixone has an improved, smoother blood-contacting surface and retains the essential surface chemistry, and therefore the acknowledged biocompatibility profile, of the Fresenius Polysulfone membrane.

Synthetically modified cellulose (SMC): a cellulosic hemodialysis membrane with minimized complement activation.

More dialysis treatments have been performed with cellulose based membranes than with any other material. As unmodified cellulose membranes activate the complement system, much effort has been directed toward the development of noncomplement activating cellulose membranes. One successful approach was the substitution of -OH groups in the cellobiose units of the cellulose molecule with tertiary amino groups, which resulted in a membrane called Hemophan. Synthetically modified cellulose (SMC) is a new hemodialysis membrane made by specific chemical modification whereby aromatic benzyl groups are covalently introduced into the cellulosic structure by ether bonds, creating hydrophobic domains within the overall hydrophilic cellulose surface: basic research investigations have shown that a characteristic hydrophobic-hydrophilic balance of surfaces is a prerequisite for improved hemocompatibility. Several cellulose modifications with aliphatic and aromatic groups were performed to achieve a membrane with the desired hemocompatibility profile; SMC, having hydrophobic benzyl groups, causes minimal activation of blood complement, coagulation, and cell activation systems. In vitro experiments with blood showed that C5a generation for SMC was reduced by 94% relative to Cuprophan (compared with 96% for polysulphone, a synthetic hemodialysis membrane). Activation of coagulation (formation of the thrombin-antithrombin III complex [TAT]) in a clinical study showed that SMC caused 16 ng/ml TAT generation compared with 36 ng/ml for polysulphone. SMC, a low-flux cellulosic dialysis membrane, thus combines the typically high diffusive performance characteristics of cellulosic membranes with excellent hemocompatibility, matching synthetic dialysis membranes.

A method for measuring capillary bleeding time in non-anaesthetized dogs: prolongation of the bleeding time by acetylsalicylic acid.

A reproducible standardized method for measuring capillary bleeding time (CBT) in dogs is presented. The skin was punctured at the toe from a forelimb of non-anaesthetized dogs parallel to the edge of the horny skin of the pad. Before the procedure, a cuff of a sphygmometer was placed above the antibrachium and a hyperaemie agent was applied to the shaved area. The normal CBT in non-anaesthetized dogs was 2.25 +/- 0.76 min (x +/- SD). After injection of acetylsalicylic acid (ASA) (20 mg/kg body weight), CBT was prolonged up to 25 min. Platelet aggregation in response to 10 micrograms collagen/ml was decreased in parallel after treatment with ASA.

Determination of contact phase activation by the measurement of the activity of supernatant and membrane surface-adsorbed factor XII (FXII): its relevance as a useful parameter for the in vitro assessment of haemodialysis membranes.

We investigated hemodialysis membrane biocompatibility with respect to contact phase activation by determination of FXII-like activity (FXIIA) on the membrane surface and in the supernatant phase, during plasma contact with various hemodialysis membranes using an in vitro incubation test cell. The results were compared to the influence of these membranes on the activation of purified FXII. A time course for the generation of activated FXII using purified FXII solution at physiologic concentrations on two similar negatively charged polymers was performed. The membranes assessed were regenerated cellulose (Cuprophan; Akzo Faser AG, Germany), modified cellulosic (Hemophan; Akzo Faser AG), acrylonitrile-sodium methallyl copolymer-based membrane AN69S (Hospal, France), and SPAN, a new polyacrylonitrile-based copolymer (akzo Nobel AG). The plasma FXIIA at the membranes surface was significantly different between the membranes, while the supernatant phase FXIIA exhibited no significant differences. In contrast, activation of purified FXII in a plasma-free system with respect to supernatant activity indicated significant differences between the materials. A similar finding for the membrane-bound factor XIIA was also observed when purified factor XII was used. The membrane-bound FXIIA values observed in the plasma system containing heparin were significantly greater than in citrated plasma. This demonstrated the strong influence of heparin and the interaction of other plasma components to the membrane surface on the activation of contact phase of coagulation.

In vitro contact phase activation with haemodialysis membranes: role of pharmaceutical agents.

Contact phase activation was investigated in vitro using flat sheet type of haemodialysis membranes, Cuprophan (Akzo, Faser, Germany) and AN69S (Hospal, France), and a negatively charged polyamide Ultipor NR 14225 membrane as a control. The investigation focussed on the determination of factor XII-like activity (FXIIA) as an indicator of contact phase activation in the supernatant phase and at the membrane surface after plasma-membrane contact using an incubation test cell. The findings were compared with the observations from a plasma-free system utilizing purified unactivated factor XII. The plasma FXIIA bound to the membrane surface was significantly different between the membranes, while the supernatant phase FXIIA exhibited no significant differences. In contrast, the plasma-free system exhibited significant differences in the supernatant FXIIA and membrane-bound FXIIA for all the materials used and the magnitude of the activity was significantly greater for negatively charged materials. This finding demonstrated the strong influence of the interaction of other plasma constituents on the membrane surface and as such the binding and subsequent activation of factor XII may be altered possibly due to competitive binding and steric hindrance. On the addition of anticoagulants such as heparin, low-molecular-weight heparin, citrate and hirudin, no significant differences were observed in plasma supernatant phase FXIIA. However, each anticoagulant appears to have a distinct influence on the magnitude of plasma membrane-bound FXIIA. On the addition of aprotinin (a kallikrein inhibitor), no significant differences were observed in the plasma supernatant FXIIA. In contrast, aprotinin appears to significantly reduce membrane-bound FXIIA on Cuprophan and polyamide NR, but significantly increase the magnitude of the membrane-bound FXIIA on AN69S.

The scientific basis for selection of measures of thrombogenicity.

The interaction of blood with materials is central in the thrombogenic response obtained in extracorporeal circulation. A number of processes are involved in determining whether there is generation of thrombin and the formation of an obstructive thrombus: these include activation of blood coagulation, platelet activation, trauma to the blood, blood flow, and use of anticoagulants (e.g. heparin). The way in which these processes produce their thrombogenic responses and interact is described to provide a basis for the understanding of the methodological approaches currently used to investigate thrombogenicity. It is concluded that much available methodology has limitations and no one indicator of the thrombogenic activity of a material or system (such as extracorporeal circulation) can be wholeheartedly recommended for all applications. Nevertheless, depending upon the circumstances of the particular investigation the assays to coagulation activation markers, thrombin-antithrombin III complex, prothrombin fragment F1+2 and fibrinopeptide A are emerging as valuable aids in the assessment of thrombogenicity. The information gained by use of these assays will require additional and complementary information from other assays of perhaps platelet function/activation, level of anticoagulant, and other coagulation markers.

Evidence of increased association of fibrinogen with platelets caused by sodium citrate.

Washed platelet suspensions are almost always prepared from blood anticoagulated with sodium citrate. Because citrate has been reported to affect platelet function, we examined the involvement of citrate on the platelet-fibrinogen interactions. More iodine 125-labeled fibrinogen was bound on washed platelets from citrate-anticoagulated blood (CP) than on washed platelets from non-anticoagulated blood (PNB) obtained by the rapid gel filtration of native blood from the same donor. Scatchard analyses of the equilibrium binding data gave linear plots for PNB indicating a single class of binding sites with 16,480 +/- 2,800 fibrinogen molecules bound per platelet (Kd 5.6 x 10(-7) mol/L); CP gave curvilinear plots that, when resolved into two components assuming two classes of binding sites, gave a high-affinity site (4,950 +/- 970 molecules per platelet; Kd 1.29 x 10(-7) mol/L) and a low-affinity site (25,660 +/- 4,600 molecules per platelet; Kd 1.02 x 10(-6) mol/L). When blood from one donor was collected into 10 mmol/L and 20 mmol/L citrate, increased binding of 125I-fibrinogen was observed on platelets exposed to 20 mmol/L citrate. Exposure of PNB to varying citrate concentrations showed enhanced fibrinogen binding with increasing citrate concentrations; differences in the Kd values between non-citrate-treated PNB and citrate-treated PNB were apparent at about 7.5 mmol/L of citrate. The effects of citrate in increasing the association of fibrinogen with platelets were not caused by variations in the pH; although fibrinogen binding was diminished at low pH of the citrate used, more binding was observed in the presence of citrate than with buffer of the same pH. The effects of citrate appear to be on the platelet fibrinogen receptor because nonspecific binding was not affected by the citrate. Inasmuch as no carbon 14-labeled citric acid binding to platelets was observed, citrate may affect the platelet-fibrinogen interactions without binding to the platelets. We conclude that platelet exposure to citrate increases the fibrinogen binding and may lead to the appearance of curved Scatchard plots.

Measurement of platelet loss in the blood compatibility assessment of biomaterials.

In the assessment of the in vitro blood compatibility of biomaterials, platelet loss is often attributed solely to platelet adhesion and consideration is not given to platelets lost in platelet aggregate formation. In order to distinguish between those platelets lost to adhesion and those lost to aggregate formation, the Wu and Hoak method for the quantification of circulating platelet aggregates in patients has been modified to establish a new test procedure. This procedure, which measures both platelet adhesion (PA) in the absence of platelets lost to aggregate formation and also the tendency of a material to induce aggregate formation, has been used to evaluate the influence of a range of polyamides and a hydrogel. The evaluation demonstrated the ability of polymers to induce readily platelet aggregates during in vitro blood-material contact. The sensitivity of the aggregate measurement was exemplified by the polyamides, where PA was similar for materials of different porosity but platelet aggregate formation increased significantly with porosity. The importance of considering platelets lost to aggregate formation was emphasized with the hydrogel, where PA was low.

Aggregation of washed platelets from non-anticoagulated human blood is not reversible.

Most of the knowledge acquired on platelet function and biochemistry has been obtained from platelets prepared from blood anticoagulated with sodium citrate. Using washed platelets from human blood (PNB) to which no anticoagulants were added, we report on responses not observed with platelets prepared from citrate-anticoagulated blood. Native blood was passed rapidly (within 5 min of venepuncture) through a Sephadex G-25/G-50 column to remove divalent ions and thus prevent coagulation. Platelets were separated from the gel-filtered blood by differential centrifugation. Responses of PNB to thrombin, collagen, calcium ionophore, ristocetin, release of 14C-5hydroxytryptamine and beta-thromboglobulin, and generation of thromboxane A2 were similar to those observed for citrated platelets. Comparison of PNB with thrombin-treated platelets, which demonstrate an increase of platelet factor 3 activity, a reduction of the adenylate energy charge and an impairment of clot retraction, indicated the absence of platelet activation. Unlike citrated platelets, however, aggregation of PNB in response to ADP was irreversible in the presence of Ca2+ and fibrinogen, even at concentrations as low as 0.2 microM ADP, with aggregation taking up to 25 times longer to reach the same extent of aggregation as for citrated platelets. PNB did not aggregate to epinephrine even in the presence of Ca2+ and fibrinogen. Sodium citrate impaired ADP-induced aggregation and clot retraction of PNB. Thus citrate affects platelet function and may cause changes resulting in the unphysiological behaviour and responses of platelets.

Preparation of washed platelets from non-anticoagulated human blood.

Sodium citrate is almost always used to anticoagulate blood for the preparation of washed platelet suspensions. Several adverse effects of citrate on platelet functional responses have been reported. We investigated the extent of activation of platelets and plasmatic coagulation during the preparation of washed platelets from human blood to which no anticoagulant was added. Washed platelets from native blood (PNB) were prepared by passing freshly-drawn human blood rapidly through a mixed Sephadex G-25/G-50 column to remove divalent cations. Gel-filtered blood (GFB), diluted by the elution medium containing 0.35% albumin and 2U/ml apyrase, was obtained within 5 minutes of venepuncture. Using CaCl2, the GFB was found to contain a mean of 1.65 X 10 mM calcium. Fibrinopeptide A measurements indicated less activation of plasmatic coagulation in GFB than in citrated blood. Measurements of beta-thromboglobulin and platelet factor 4 during the various stages of the preparation of PNB showed no platelet activation. No platelet aggregates, as measured by the platelet aggregate ratio method, were observed in GFB. Transmission electron microscopy showed intact discoid platelets similar to those in platelet-rich plasma. The reduced activation of platelets and of plasmatic coagulation was due to the more effective removal of calcium ions from the blood by gel filtration than chelation by citrate. PNB may thus provide a model for studying the requirements for calcium in platelet function in the absence of any citrate-platelet interactions.

A modification of the Wu and Hoak method for the determination of platelet aggregates and platelet adhesion.

The Wu and Hoak method for determining circulating platelet aggregates has poor reproducibility; problems have been reported with the composition of the buffer systems, haemolysis, the effects of blood collection technique and a divergence of the platelet aggregate ratio in blood for healthy donors from the theoretical value of 1. Our investigations suggest that the original technique is highly operator-dependent, especially the collection of blood and the method of counting platelets after centrifugation. We describe an improved modification of the Wu and Hoak technique; a new buffer system has been developed and the proportion of blood in the buffered EDTA and buffered EDTA-formalin solutions has been altered to obtain platelet rich plasma. The platelet aggregate ratio (PAR) by this modified method for healthy donors in two different studies was 0.97 +/- 0.02 and 0.98 +/- 0.01 respectively. Finally, the principle of Wu and Hoak was used to measure accurately platelet adhesion, without the role of platelet-platelet interactions (aggregation). Platelet adhesion and aggregation were then used to evaluate the thrombogenicity of various artificial surfaces, including silicone rubber and polytetrafluoroethylene (PTFE) vascular grafts.

Utilization of the platelet release reaction in the blood compatibility assessment of polymers.

The release reaction is directly associated with platelet adhesion and aggregation, which are primary events leading to thrombus formation following contact of blood with artificial surfaces. This investigation examined the release reaction from the alpha granules of platelets after blood-polymer interaction, and utilized the measurement of beta-thromboglobulin (BTG), a platelet-specific protein, in the assessment of the in vitro blood compatibility of polymers. A radioimmunoassay was used, to determine the release of BTG following contact of blood with tubes of siliconized glass and polypropylene and flat sheets of poly(vinyl chloride) and silicone rubber. Polypropylene tubes caused less release of BTG than those of siliconized glass and silicone rubber induced less BTG release than poly(vinyl chloride). The investigation indicates a role for BTG measurement in blood compatibility assessment.