Search

All available dialysis membranes prevent endotoxin (Et) from mixing with the blood under clinical conditions. However, maintenance dialysis patients are at risk of amyloidosis attributed to Et. This suggests that Et may affect the blood even if it does not mix with the blood. The objective of the present study is to evaluate the activity of Et trapped by membranes. We made mini modules out of hollow fibers using three different types of membranes and filtered Et solution. The lumen of the hollow fibers was then filled with limulus amebocyte lysate (LAL) for 15 min at 310 K. Et activity was then determined by measuring absorbance of the LAL reagent. The surfaces of test membranes were studied using an atomic force microscope. With polyester polymer alloy (PEPA), no Et leakage or Et activity was detected in the hollow fibers under any conditions. With polysulfone (PS) and polyether sulfone (PES), no Et leakage was detected under clinical conditions, but Et activity was detected in the hollow fibers. These results show that Et trapped by the inner skin may affect the blood, even if Et does not mix with the blood. Therefore, Et should be trapped by an outer skin or the part somewhat far from the blood which does not contact with the blood directly.

Using the derived mass transfer correlations for hollow fibers, hollow fiber arrangements were optimized for an artificial gill that uses an oxygen carrier solution. FC-40, a perfluorocarbon (PFC), was used as the oxygen carrier solution. In the oxygen uptake module, a hollow fiber arrangement with parallel coiled hollow fibers is preferred. The optimum outside diameter of the hollow fibers and the transverse pitch between them are 300 and 500 μm, respectively. In the oxygen release module, a hollow fiber arrangement of straight parallel hollow fibers is preferred. The optimum outside diameter of the hollow fibers and transverse pitch between them are 300 and 500 μm, respectively. In the case of humans, the scaling up was estimated from the oxygen transfer rates using these optimum hollow fiber arrangements. The required total membrane surface area is 50.8 m 2 , the total delivered pumping energy is 124 W, and the oxygen partial pressure in inspiration is 17.8 kPa. Importantly, the total membrane surface area required was significantly reduced using the modules with an optimum hollow fiber arrangement in comparison with that using connected membrane oxygenators as a gas exchanger. The optimization of hollow fiber arrangements in an artificial gill significantly enhances oxygen transfer from water to air.

An artificial gill has been developed that transfers oxygen from water to air, using oxo-molybdenum(IV)5,10,15,20-tetramesitylporphyrin (Mo IV O(tmp)) dissolved in o -xylene as an oxygen carrier solution and the energy of visible light. The oxygen partial pressure in the oxygen carrier solution is changed by photo-irradiation to enhance both the oxygen uptake from water and the oxygen release to air. The ratio of the oxygen mass transfer coefficient of the oxygen carrier solution to that of water is 0.746 for oxygen uptake and 0.654 for oxygen release. In designing a large-scale artificial gill for supplying oxygen to a closed space underwater such as submerged vessel, the required membrane surface area, the seawater flow rate and the reservoir tank volume were 123 m 2 , 0.00533 m 3 s −1 , and 5.06 m 3 , respectively. These values increased as the oxygen partial pressure of seawater decreased. However, the high partial pressure of oxygen required for human respiration (20.0 kPa) can be provided in a closed space even from seawater with an oxygen partial pressure as low as 10.0 kPa. This newly developed artificial gill may be useful for deep sea activities, such as underwater exploration, marine research and underwater habitation.

Polysulfone (PS) dialysis membranes hydrophilized by blending polyvinylpyrrolidone (PVP) are well known to have excellent biocompatibility in clinical use. The objective of the present study is thus to clarify how PVP improves biocompatibility of PS membranes and furthermore to develop a patient-friendly PS dialysis membrane with higher biocompatibility. Biocompatibility based on both lactate dehydrogenase (LDH) activity and amount of protein adsorption was greatly different among four commercially available PS hollow-fiber dialysis membranes. PVP present on the inner surface of the hollow fiber was quantitatively determined by X-ray photoelectron spectroscopy (XPS), demonstrating the amount of PVP to be varying for each membrane. Structure parameters such as surface roughness, three-dimensional surface area and polymer particle diameter, indications of the physicochemical properties of the membranes, were measured on the observed inner surface images in both wet and dry conditions by atomic force microscopy (AFM) to account for dependence of biocompatibility on these structure parameters. The higher regularity polymer particle structure has in the wet condition, the lower wet/dry ratio surface roughness has and the larger wet/dry ratio polymer particle diameter has, that is, the more greatly the polymer particles swell by wetting, the higher biocompatibility is achieved by “cushion effect”.

Morphology and solute diffusive permeability of thin layer of molecularly imprinted polymer (MIP) change in the presence of templates, which is termed as “gate effect”. To optimize morphological changes induced by the gate effect, the flexibility, density, and the amount of specific binding sites for templates of the MIP-grafted layer must be tightly controlled during radical polymerization. Living radical polymerization with “iniferter” ( ini tiator-trans fer agent- ter minator) is useful tool for controlling degree of polymerization by reaction time. In this work, photoactive iniferter (benzyl dietyldithiocarbamate) was immobilized on a cellulose membrane via a silane coupler. This treated membrane was grafted with theophylline-imprinted copolymer of methacrylic acid and ethylene glycol dimethacrylate by ultraviolet irradiation. The relationship between the amount of graft copolymer, the degree of the gate effect and the time of UV irradiation was studied. The amount of grafted copolymer increase by repeating polymerization cycle. In addition, the variation of the diffusive permeability by template clearly differs to that by analogue. Therefore, these MIP membranes can discriminate two alkaloids by difference of diffusive permeability. The variation by the template or the analogue and the selectivity of the permeability depended on irradiation time. Those results indicate that synthesized MIP has a “living nature” and the gate effect is feasible to control by irradiation time. Living radical polymerization is a promising method to build sophisticated architecture of MIP membranes possessing self-controllable permeability by gate effect.

We have reported a novel method of visualizing endotoxin (Et) distribution inside an Et-blocking filtration membrane using both fluorescence-labeled Et and a confocal laser scanning fluorescence microscope (CLSFM) in our previous paper [J. Membr. Sci. 210 (2002) 45]. The objective of the present study is to clarify Et-blocking mechanism of dialysis membranes. Six kinds of dialysis membranes with varying materials (hydrophilic and hydrophobic) and varying structures (pore diameter, skin layer location and thickness, and water content) were evaluated by CLSFM together with other techniques such as atomic force microscopy (AFM). Physicochemical property of a membrane material affects Et-adsorbing efficiency, and further membrane structure affects Et-plugging efficiency. Rejected Et distribution in the membranes with varying materials and structures is successfully visualized using fluorescence-labeled Et by CLSFM. Et adsorption on the membranes occurs first, followed by the narrowing of their pores, and afterward pore plugging is continued. Adsorption plays a vital role in Et-blocking. Double skin layer structure is valid for preventing of Et contamination than only inner skin layer structure because the double skin layer structure blocks Et more farther from blood-side surfaces than the only inner skin layer structure.

An artificial gill was developed using a concentrated hemoglobin solution containing inositol hexaphosphate (IHP), as the oxygen carrier solution, with the oxygen affinity controlled by temperature. Oxygen dissolved in sea water is first taken up from water to the oxygen carrier solution at 293 K. The oxygen carrier solution is then heated and the oxygen is released from the oxygen carrier solution to expired air at 310 K. The enhancement factors of the oxygen carrier solution that indicate its performance were obtained from the oxygen uptake rate and oxygen release rate. The values were approximately 3 at oxygen uptake and 16 at oxygen release. The scale-up for a human being at rest was estimated using these values, and the optimal operating condition was determined. The required membrane surface area for a human being is 63.8 m 2 . The oxygen partial pressure of inspiration is 20.7 kPa, adequate for the respiration. These indicate the feasibility of a compact and portable artificial gill device.

Since the finding of β 2 -microglobulin as a causal substance in the carpal tunnel syndrome of chronic hemodialysis patients, removal of β 2 -microglobulin has been performed using highly permeable dialysis membranes with larger pores. Such large-pore membranes tend to allow endotoxins (Et), harmful substances contained in dialysate, to enter blood. At present, as a countermeasure, Et-blocking filtration membranes are used to remove Et from dialysate. However, Et removal mechanism by these membranes has not been clarified yet. The objective of this study is thus to visualize distribution of fluorescence-labeled Et trapped inside the polyester–polymer alloy (PEPA) membrane, a widely used Et-blocking filtration membrane using a confocal laser scanning fluorescence microscope (CLSFM). Et were observed mainly in the outer skin layer of the hollow fiber, while some in the void and inner skin layers. No Et were present inside the hollow fiber. In conclusion, we succeeded in visualization of Et distribution inside the Et-blocking filtration membrane using CLSFM. This novel visualization technique may allow evaluation of distribution of Et trapped inside various kinds of Et-blocking filtration membranes.

Determining pore size distribution is important for characterization of a dialysis membrane. However, conventional microscopic techniques cannot present a sufficient image for determining pore size distribution. In the present study, tapping mode atomic force microscopy (TMAFM) has been shown to be a powerful tool for observing and evaluating the small surface pores of a hollow-fiber dialysis membrane. Sample fixing technique described below and a highly sharpened probe have made it possible to observe small pores on a soft and undulant surface of a dialysis membrane. This is the first time that clear TMAFM images of surface pores of a hollow-fiber dialysis membrane at such high resolution have been presented. Pore diameter was determined by image analysis. Average pore diameter of APS-150 (Asahi-kasei, Japan) determined by TMAFM was compared with those by field emission scanning electron microscopy (FESEM) and by the Hagen–Poiseuille equation. The average pore diameter of APS-150 determined by TMAFM was slightly higher than that by FESEM. The average pore diameter determined by the Hagen–Poiseuille equation was intermediate between values for that of inside and outside surfaces determined by TMAFM.