State-of-the-art lung support systems are limited to short time application because of a lack of long term hemocompatibility and protein absorption on the membrane surfaces. In a highly interdisciplinary project at RWTH Aachen University a biohybrid lung assist system with endothelialised gas exchange flat membranes is developed to improve long term compatibility of oxygenators. To increase the gas exchange performance of flat membranes hollows are imprinted in the membrane surfaces. This approach is based on the research of B. J. Bell-house et al. , who discovered this effect, now known as Bellhouse-effect, around 1960. In this paper a processes to manufacture membrane assemblies for oxygenation with imprinted hollows on the flat membrane surfaces is reviewed.
In cases of acute respiratory failure oxygenators are already used to assume the function of the human lung for a limited time. In membrane oxygenators blood- and gas-phase are separated by semipermeable membranes . The duration of use of membrane oxygenators is limited because proteins and blood clots settle on the membrane surfaces over time. To improve long term hemocompatibility of lung-support-systems oxygenators with endothelialised membrane surfaces are developed at the RWTH Aachen University. Typically hollow fibre membranes are used for oxygenation because they have a higher surface area and thereby can reach higher gas exchange rates than flat membranes. For endothelialisation hollow fibre membranes are not well suited because there is a lot of friction between the individual fibres, so that cells can be easily rubbed off the fibre surfaces. A method to improve gas exchange performance of flat membranes was discovered by B.J. Bellhouse et al.  around 1960. It was found, that by using furrowed membranes the gas exchange rates of oxygenators with flat membranes can be increased. The rates of transfer using furrowed membranes can be varied by the strength of the secondary flows .
Aim of this project is to develop a technique to manufacture innovative membrane assemblies with hollows on the membrane surfaces, which can be used for oxygenation.
2 Material and methods
The gas exchange performance of oxygenators with flat membranes is not efficiently used. This is because of the laminar flow of blood on flat surfaces, whereby the blood close to the membranes contains more oxygen then the faster blood in the centre of the channel  as displayed in Figure 1.
2.1 Hollows for Bellhouse-effect
B.J. Bellhouse et al.  described, that hollows in membranes used for oxygenation can help to enhance oxygen transfer by causing vortices in the otherwise laminar flow. Blood from the wall region is mixed with blood from the centre of the channel because of the induced secondary flows. An almost homogeneous mixture could be received by selection of a furrow wavelength of about 2 mm, furrow depth of 0.5 mm and a channel spacing of 0.4 mm.
2.2 Production of furrowed membranes
To be able to produce membrane assemblies with furrowed membrane surfaces two main issues have to be solved. The first step consists in developing a process for hollow imprinting and in a second step a suitable textile for use as spacer in the membrane assembly has to be found. Hollow parameters defined by Bellhouse et al. are used as a reference for imprinting hollows for the project.
2.2.1 Techniques for hollow imprinting
For the membrane assemblies a textile spacer structure is combined with commercially available flat membranes produced by Sarstedt AG & Co., Nümbrecht. After placing the spacer fabric between two flat membranes, the membranes are welded together, what results in a completely leak-tight connection. To simplify the process no additional form but the textile structure is used to provide the design of the desired hollows. The imprinting itself is done by vacuum aspiration using the setup displayed in Figure 3. A cannula, connected to a vacuum generator, is inserted between the two welded membranes. Thereby a vacuum of approximately 0.35 bar is produced between the membranes. Because of the applied vacuum the membranes are sucked into the pores of textile structure. As a result hollows emerge on the membrane surfaces. The subsequently necessary heat-setting is processed in a water bath.
Best results could be achieved with a heat-treatment for 30 seconds at 70 °C. Another successful set up for heat-setting is the application of hot air of about 75 °C for 5 minutes.
2.2.2 Selection of textile spacer structure
Suitable textiles for implementation as textile spacer structures for oxygenation have to meet two main requirements. At first they have to allow a homogeneous gas flow between the membranes. On top of that they have to provide a suitable structure to define the form of hollows. Woven fabrics as well as warp-knitted fabrics can be produced with different pore sizes , so that they are particularly qualified for the application as spacer between the flat membranes. Two textiles, pictured on Figure 4, are selected as examples to evaluate which pores can be produced by which textile structure. The first textile is a plain woven wire mesh with pores of 1.5 mm × 1.5 mm in size. The second textile is a warp-knitted polyester (PES) fabric with pores about 3.6 mm × 5.8 mm. Both textiles were not expected to result in the required hollow size. They can be used as a first part result on the way to determine the required hollow size according to Bellhouse et al. .
After imprinting hollows into the membrane surfaces using the selected textiles as well as the defined process, the resulting hollows were filled up with silicon, receiving a negative form. The negatives were inspected under μ-Computer Tomography (μ-CT). As it can be seen in Figure 5 only the warp-knitted fabric leads to well defined round hollows. The hollows produced with the woven mesh have an irregular shape because of the yarn interlacing that is characteristic for woven fabrics.
Measuring the hollows produced with the warp-knitted spacer fabric indicated that the received hollow size corresponds almost directly to the textile pore size. The measured hollow size was 3.6 mm x 5.8 mm which is the same like the earlier defined textile pore size. A hollow depth of 0.35 mm was measured, which is slightly beyond the intended depth of 0.5 mm defined by Bellhouse.
Using the selected textiles and the described process to imprint hollows on flat membrane surfaces, we were able to produce membrane assemblies with furrowed surfaces. Warp-knitted fabrics are better suited for application as part in membrane assemblies for oxygenation, because the received hollows are round like it is defined by Bell-house et al. . It could be shown that the pore size corresponds directly to the selected pore size of the spacer fabric. This leads to the conclusion that by using a spacer textile with a pore size of 2 mm x 2 mm, the perfect pore size referring to Bellhouse et al. could be received. Further testing is needed to validate this theory. On top of that it is necessary to determine the degree of increase the imprinted hollows have on the gas transfer rates of oxygenators.
As a consequence membrane assemblies with flat membranes can be produced, that can be used for endothelialisation, what may lead to an improvement in long term hemocompatibility of lung support devices in the future.
The present work has been done within the EndOxy project, which is funded by the Interdisziplinäres Zentrum für Klinische Forschung (IZKF) at Uniklinikum Aachen.
Conflict of interest: Authors state no conflict of interest. Material and Methods: Informed consent: Informed consent has been obtained from all individuals included in this study. Ethical approval: The research related to human use has been complied with all the relevant national regulations, institutional policies and in accordance the tenets of the Helsinki Declaration, and has been approved by the authors’ institutional review board or equivalent committee.
 Bellhouse, B.J.; Bellhouse, F.H.; Curl, C.M.; Macmillan, T.I.; Gunning, A.J.; Spratt, E.H; MacMurray, S.B; Nelemens, J.M.: A high eflceny in membrane oxygenator and pulsatile pumping system and its application to animal trials; Vol. XIX Tras. Amer. Soc. Artif. Int. Organs; 197310.1097/00002480-197301900-00014Search in Google Scholar
 Stickel, F.: Aktueller Stand von Oxygenatorkonstruktionen; Kardiotechnik Vol.2, 2008.Search in Google Scholar
 Dorrington, K.L.; Ralph, M.E.; Bellhouse, B.J; Sykes, M.K.: Oxygen and CO2 Transfer of a Polypropylene Dimpled Membrane Lung with Variable Secondary Flows;J. Biomed. End. Vol.7, 198510.1016/0141-5425(85)90036-6Search in Google Scholar
 Dierickx, P.: Blood flow and gas transport in artificial lungs: in numeroand in vitro analysis, Gent University, Gent, dissertation, 2001Search in Google Scholar
© 2015 by Walter de Gruyter GmbH, Berlin/Boston
This article is distributed under the terms of the Creative Commons Attribution Non-Commercial License, which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.