Currently, the foreign surfaces of various extracorporeal circulation devices are coated with a biocompatible polymer coating agent (BPA), which creates a hydrophilic blood-contacting layer to reduce thrombogenicity, while the membranes in hemodialyzers are not. We aimed to clarify other side effects of BPA-coated membranes by examining the diffusion performance in in vitro experiments.
We used a polyethersulfone membrane (sieving coefficient of albumin is ≤0.01) coated with BPA product, SEC-1TM (Toyobo), in a hemodialyzer. To estimate the diffusion rates of a wide range of molecules, 2 L of saline containing vancomycin, lysozyme, and albumin were recirculated in the circuit configured with a hemodialyzer, and dialyzed continuously using water. The concentrations of sodium, vancomycin, lysozyme, and albumin were measured every 5 minutes for 30 minutes and compared in experiments with BPA-coated (n = 4) and BPA-noncoated (n = 4) membranes.
The removal rates of sodium and vancomycin after 5 minutes of dialysis (n = 24) were significantly higher in BPA-coated than noncoated membranes, while those of lysozyme and albumin were not significantly different. The removal rates of sodium and vancomycin after 30 minutes of dialysis (n = 4) were significantly higher, and those of lysozyme were significantly lower in BPA-coated than noncoated membranes, while those of albumin were not significantly different.
The preliminary study suggests that BPA-coated membranes enhanced the diffusion rate of molecules with low and middle molecular weight without affecting the sieving coefficient of albumin. Thus, BPA coating can enhance the dialysis performance of membranes.
Post author correction
Article Type: ORIGINAL RESEARCH ARTICLE
AuthorsMasashi Tagaya, Saki Nagoshi, Morihiro Matsuda, Shunsuke Takahashi, Shinya Okano, Kazunobu Hara
- • Accepted on 23/06/2017
- • Available online on 02/08/2017
This article is available as full text PDF.
Hemodialysis serves to compensate for impaired renal function in patients with end-stage renal disease. A hemodialyzer consists of a membrane that corrects the levels of blood components mainly through diffusion (1-2-3-4). In order to correct blood component imbalances, patients undergo hemodialysis three times weekly for 3 to 4 hours per session (5, 6). The blood is therefore exposed to the circuit – including the hemodialyzer – for a long duration, which activates the whole coagulation cascade; hence, the administration of anticoagulants is required (7-8-9).
Currently, extracorporeal circulation circuits used in cardiopulmonary bypass surgeries are coated with biocompatible polymers to prevent adverse effects due to contact between the blood components and foreign surface of the circuits, although a similar effect has not been observed in hemodialysis. This biocompatible polymer coating agent (BPA) is composed of a hydrophobic backbone adherent to the surface, and a hydrophilic blood-contacting layer. The hydrophilic layer swells when blood contacts it, creating a water-filled layer that maintains protein conformation and prevents surface activation (10), and many studies have reported a reduction in the circuit’s thrombogenecity, or reduction in the heparin dose needed to achieve adequate anticoagulation by BPA coating (11-12-13-14-15). In our previous study, a hemoconcentrator, which has similar function as the hemodialyzer in cardiopulmonary bypass circuit, was probatively coated with BPA and investigated as regards hydrokinetics. BPA coating resulted in an enhancement of the filtration rate (16). However, the filtration rate enhancement leads to the possibility that the sieving coefficient of substances that are targeted to be dialyzed are altered, and those were not examined.
If BPA-coated membranes within a hemodialyzer have strong biocompatibility, being able to prevent thrombogenicity with coating of all surfaces of the hemodialysis circuit, anticoagulant-free dialysis may be possible, although investigations for markers of coagulation will be required. However, before examining the effect on anticoagulation, the performance shift of diffusion should be clarified first. In the present study, we aimed to clarify the diffusion performance of BPA-coated membranes in in vitro experiments to progress the investigation of BPA-coated membrane for clinical application.
Materials and methods
Experiments were performed to estimate diffusion rate alterations of a wide range molecules in a BPA-coated membrane by comparing diffusion rates between a coated and noncoated membrane under identical conditions.
Eight polymer membrane columns (Hemo Crystal; Mera), which had the same lot number, were employed as hemodialyzers in this study. Hemo Crystal, with 1.1 m2 of membranes, is used as hemoconcentrator in cardiopulmonary bypass surgery, but can function as a high-flux hemodialyzer because sieving coefficients of beta-2-microglobulin and albumin are 0.6 and ≤0.01, respectively (17). Four of these were coated with a BPA product (SEC-1TM; Toyobo): a coating agent composed of a copolymer comprised of hydrophobic alkyl-acrylate, hydrophilic polyethylene-glycol-acrylate, and water repellent silicon (silicone-methacrylate).
To estimate the variance of diffusion rates among a wide range of molecules, categorized as low (≤0.5 kDa), middle (0.5-5 kDa), and large molecular weight (≥0.5 kDa), in a BPA-coated membrane, we employed the following substances as the target solutes: sodium (23 Da), vancomycin (1485.87 Da) (18), lysozyme (14 kDa), and albumin (approximately 65 kDa).
Two liters of the controlled solution, composed of saline, 120 mg of vancomycin (vancomycin hydrochloride; Kobayashikako), 120 mg of lysozyme (lysozyme chloride refined from egg white; Nacalai Tesque), and 25 g of albumin (Albuminar®-25; CSL Behring), were prepared as the experimental solution. The experimental circuit was configured with a hemodialyzer, a roller pump, polyvinyl chloride tubes, a sampling port, and inlet and outlet chambers. The controlled solution was maintained at approximately 36°C, with continuous agitation, and was conveyed to the experimental circuit, which was primed with saline, and recirculated at 300 mL/min for several minutes to mix the saline-primed experimental circuit with the controlled solution in the experimental circuit (
Experimental circuit. Circuit pathway of controlled solution is shown in brown. Reverse osmosis (RO) water in the plastic reservoir was delivered to the hemodialyzer (shown in blue) by a roller pump, flowed outside the membranes, and discharged from the hemodialyzer (shown in green) by the same roller pump in order to equalize both delivered and discharged RO water. The sampling port was configured prior to the inlet chamber.
Similar experiments were performed with 4 BPA-coated hemodialyzers and 4 noncoated hemodialyzers. The circuit and the controlled solution were changed for each experiment.
Diffusion rate alterations were assessed by comparing the ratio of the concentrations of sodium, vancomycin, lysozyme, and albumin after to before 5 minutes and 30 minutes of dialysis (Ct/Ct-5; C30/C0, respectively) between BPA-coated and noncoated hemodialyzers. All variables were compared using Student’s t-test by Microsoft Excel software. The significance level was set at α = 0.05.
The percentage of the ratios of the remaining concentrations of sodium, vancomycin, lysozyme, and albumin after 5 minutes of dialysis (100*Ct/Ct-5; n = 24) were averaged in both BPA-coated and noncoated hemodialyzers, as shown
Changes in the concentration of sodium, vancomycin, lysozyme, and albumin with dialysis duration. The horizontal axis indicates dialysis duration (min), and the vertical axis indicates the percentages of remaining concentrations of each solute. The
NC = noncoated; BPA = BPA-coated; * p<0.05; N.S. = not significant.
Remaining ratios of sodium, vancomycin, lysozyme, and albumin, after 5 minutes of dialysis
|Target solutes||BPA-Coated (n = 24)||Noncoated (n = 24)||p value|
|Ratios (Ct/Ct-5) were averaged for 6 time points in the 4 BPA-coated and the 4 noncoated hemodialyzers (n = 24), represented as means ± standard deviations.|
|Sodium (%)||83.8 ± 0.61||86.3 ± 0.53||<0.05|
|Vancomycin (%)||89.3 ± 3.78||91.2 ± 3.09||<0.05|
|Lysozyme (%)||94.0 ± 4.48||91.5 ± 5.90||0.22|
|Albumin (%)||98.5 ± 4.48||99.4 ± 5.90||0.54|
To estimate diffusion rate alterations of a wide range of molecules in BPA-coated membranes, we compared diffusion rates of sodium, vancomycin, lysozyme, and albumin between both BPA-coated and noncoated membranes, in which the sieving coefficient of albumin is ≤0.01, under identical conditions. In accordance with these comparative results, BPA coating resulted in increased sodium and vancomycin removal, and no differences in albumin removal, indicating that the diffusion rates of molecules with low and middle molecular weight were specifically enhanced without affecting the sieving coefficient of albumin by BPA coating; that is to say, diffusion rates of the molecules that can easily pass through the membrane pore, are solely enhanced by BPA coating. This change is expected to be favorable for membrane technology.
We previously reported that BPA coating of membranes can enhance the filtration rate without physical alteration of the membranes, and concluded that the decrement of drag resulting from water-filled layers on BPA contributed to this enhancement (10, 16, 20). Similarly, the following laws can explain the reason why decrement of drag results in molecular transfer enhancement. Based on Fick’s laws of diffusion, diffusion flux (J) is directly proportional to concentration (C) and is inversely proportional to distance (x). Furthermore, based on the Stokes-Einstein equation, the molecular diffusion coefficient (D) is directly proportional to temperature (T) and mobility (B). The mobility is inversely proportional to viscosity (μ) and molecular radius (a). Those two laws indicate that the diffusion flux is increased proportionally with the decrement of viscosity.
Moreover, according to Newton’s law of friction, frictional stress (
Restating the expression of Newton’s law of friction differently, the reduction of friction stress induces the decrement of viscosity. The diffusion flux is therefore increased by the greater diffusion coefficient resulting from the decrement of viscosity. Considering these interpretations, the water-filled layer on BPA, which decreases drag in the fluid pathway, can reduce the frictional stress, leading to a higher diffusion performance for low- and middle-molecular-weight molecules, such as sodium and vancomycin. Similarly, larger molecular weight reduces molecular mobility, counteracting the effect of reduced frictional stress; thus, diffusion performance is not enhanced in molecules with large molecular weight such as albumin. Considering the above factors, enhanced diffusion rates in BPA-coated membrane can be attributed to drag decrement due to the water-filled layers on BPA.
Polymer membranes, which are called high-flux membranes, have been developed, and are currently widely used. The main focus of high-flux membrane development has been to sharpen the molecular weight cutoff of the membrane to maximize removal of low-molecular-weight proteins while minimizing the removal of albumin (21). Considering this concept, our finding that BPA coating can enhance the removal of low- and middle-molecular-weight solutes and while not affecting the sieving coefficient of albumin, is advantageous and can be considered a new improvement in membrane technologies.
On the other hand, the removal of lysozyme, which is a low-molecular-weight protein, was not maximized, which is amenable to further investigations. An explanation might be that noncoated membranes adsorb lysozyme, leading to deceptive enhancement of removal in noncoated membranes (22); thus, there is a possibility that essential diffusion rate enhancement occurred but was not observed in BPA-coated membranes. Furthermore, assessment of protein transfer is difficult since protein may undergo posttranslational modification with albumin (23). Therefore, for better clarification of the findings in the present study, additional investigations regarding low-molecular-weight proteins are needed.
In the field of cardiopulmonary bypass surgeries, polymers such as poly-2-methoxyethylacrylate (10, 12, 13) and 2-methacryloyloxyethyl phosphoryl choline (24-25-26), are used as BPA rather than the acrylic-monomers-copolymer that is used in this study. All of these BPAs can prevent thrombogenicity by an identical mechanism: the hydrophilic layer swells when blood contacts it, creating a water-filled layer that maintains protein conformation and prevents surface activation (10). Therefore, since all of these BPAs are expected to have an efficacy similar to that observed in this study, further studies with other BPAs could be justified. Furthermore, there are many reports about the favorable effect of BPAs on the reduction of cell adhesion, hemolysis,protein adhesion, and inflammatory markers (13, 24-25-26). Moreover, several reports on noncoated hemodialyzers showed that the exposure to polymer membranes has detrimental effects in terms of impaired platelet function, increased oxidative stress, and inflammation (27-28-29-30). BPA-coated membrane is therefore expected to improve hemodialysis biocompatibility, although further investigations are required.
Our investigation of BPA-coated membrane was based on the possibility of performing hemodialysis without anticoagulant administration. However, before investigating the relation to anticoagulation, if BPA coating proves to adversely affect the function of hemopurification, no further investigations would be pursued; thus, we first tried to assess the function of hemopurification in the present study. The present findings not only allow for the elimination of conceivable adverse effects by BPA coating, but also show that diffusion performance could be higher with the BPA-coated membrane. Therefore, further investigations for BPA-coated membranes are justified. We hope that our finding might support the clinical application of BPA-coated membranes.
Vancomycin and lysozyme are adsorbed by several polymers (18, 22). There is a possibility that our results may have been affected by the adsorption, which is a major limitation of this study. Therefore, future studies with complete elimination of efficacy of adsorption should be conducted to confirm our findings.
The results of this preliminary study suggest that the use of a hemodialyzer with a BPA-coated membrane, in which the sieving coefficient of albumin is ≤0.01, can enhance the removal of low- and middle-molecular-weight solutes without affecting the sieving coefficient of albumin. Those aspects may occur secondary to the drag decrement of the fluid pathway induced by water-filled layers of BPA.
We are grateful for the valuable advice provided by Hironori Izutani, Professor of Cardiovascular and Thoracic Surgery at Ehime University Graduate School of Medicine, Ehime, Japan; by Masuhide Yakehiro, Professor of Physiology at Hiroshima International University, Hiroshima, Japan; and by Naoko Mitsui, Pharmacist at National Hospital Organization Kure Medical Center and Chugoku Cancer Center, Hiroshima, Japan. We are also grateful for the analytical assistance provided by Yasuhiro Oota of LSI Medience corp., Hiroshima, Japan.
- Tagaya, Masashi [PubMed] [Google Scholar] 1, * Corresponding Author (firstname.lastname@example.org)
- Nagoshi, Saki [PubMed] [Google Scholar] 2
- Matsuda, Morihiro [PubMed] [Google Scholar] 3
- Takahashi, Shunsuke [PubMed] [Google Scholar] 4
- Okano, Shinya [PubMed] [Google Scholar] 1
- Hara, Kazunobu [PubMed] [Google Scholar] 1
Department of Medical Engineering, National Hospital Organization Kure Medical Center and Chugoku Cancer Center, Hiroshima - Japan
Department of Clinical Laboratory, National Hospital Organization Kure Medical Center and Chugoku Cancer Center, Hiroshima - Japan
Division of Preventive Medicine, Institution for Clinical Research, National Hospital Organization Kure Medical Center and Chugoku Cancer Center, Hiroshima - Japan
Department of Nephrology, National Hospital Organization Kure Medical Center and Chugoku Cancer Center, Hiroshima - Japan