As the low clearance rate of plasmaseparation limits its use in the treatment of patients suffering from liver failure, sepsis or MOF, we intend to develop strategies for a plasmaseparation unit which increases plasmafiltration rates. Our first question focused on whether commercially available plasmaseparation filters, and in particular their membranes, are suitable for the inversion of blood and plasma compartments. This experimental study was performed using in vitro systems. Commercially available plasmafilters PF2000N (Gambro) and Plasmaflo (Asahi) were compared in both their normal operating mode with blood flow through the capillary lumen, and in the inverse mode. Inverse mode means that blood flows through the outer space of the capillaries while plasma was obtained from the lumen. Heparinised porcine blood (5 I.U./ml) was used in a heated, recirculating in vitro circuit. Our main results were that the normal use of both filter types Plasmaflo and PF2000N enabled maximal blood flows (Qb) of 200 ml/min and filtration rates (Qf) of 25-40 ml/min. Operating the filters in the inverse mode enabled Qb up to 500 ml/min and Qf up to 100 ml/min. Hemolysis, platelet counts and coagulation parameters did not differ significantly regardless of whether the normal or inverse mode was used. The tested plasmafiltration membranes appear to be suitable for use in inverse mode. Although in our experiments, hemocompatibility tests did not indicate severe problems induced by the module geometry, the development of a module specially constructed for blood flow outside of the hollow fibers appears to be necessary in order to minimise shunts and low perfusion areas.

Culture media are frequently used in the evaluation of metabolical functions of hepatocytes in hybrid liver support systems (hLSS). However, media compositions differ substantially from those of plasma. Therefore, our study was designed to investigate whether current in vitro studies with medium are suitable to assess the metabolical competence of hLSS-cultures during clinical application as well as to explore whether the cell nutrition with medium provides a suitable modus operandi for stand by cultivation. Paired bioreactor cultures were perfused with either Williams' Medium E (MPB) or human plasma (PPB). About 6x108 primary pig hepatocytes (>97% viability) were cultured in three laboratory scale bioreactors designed according to Gerlach's bioreactor-concept. Different perfusion protocols were initiated after a standardised period allowing for cell attachment and reorganisation in aggregates. Whereas patterns of enzyme release were similar in both protocols the metabolical behaviour was different between MPB (anabolic state) and PPB (catabolic state). Furthermore, compared to MPB the lidocaine-MEGX-tests for PPB demonstrated lower MEGX-concentrations and a different reaction pattern. We conclude that the nutrition of hepatocytes with medium during the stand by period itself might influence the cell function and subsequently the efficacy of the hLSS-treatment during clinical application.

Pigs are widely used as models for a variety of human diseases, because many of their physiological functions closely resemble those of humans. However, information on instrumentation techniques is still scarce. In particular, experiments in conscious pigs focused on extracorporeal circuits are connected to a variety of methodical problems with respect to the handling of the animals. Usually, pigs are placed in restraint-slings during the application of an extracorporeal system. However, this method of restraint may lead to excessive mental distress even in trained animals. The latter might influence the results and certainly affects principles of animal welfare. Our own experiences with instrumented, conscious, but unrestrained dogs encouraged us to modify methods used for the fixation of in-dwelling central venous catheters in dogs with special regard to the species specific behaviour and phenotype of pigs. A cord retractable leash (CRL) was used for maintaining a safe distance between the animal and the outer ends of the catheters. To prevent dehiscences of the required fixation sutures a new catheter bag (CB) was designed to counteract tension forces caused by the CRL's spring-mechanism. The combination of both the CRL and CB enabled us to conduct safe experiments with conscious, unrestrained pigs. We alleviated the mental distress these animals were exposed to in comparison to former methods based on restraint of the animals.