MEMBRANELESS DIALYSIS IN A MICROFLUIDIC CONFIGURATION

Abstract

It has been shown to be possible to bring two miscible fluids together, allow diffusional interchange of species between them, and separate the fluids with little or no mixing. This procedure, visualized to occur on small ‘microfluidic’ chips, has been under development for analytical purposes but has recently been proposed as a means of blood purification. Such a system might: (i) enhance biocompatibility; (ii) allow much longer operating times without performance decrement, because there is no membrane to foul (iii) allow multi-step, on-chip processing that is unattractive to realize at macro-scale. These systems raise, however, serious questions: (i) how to achieve volume transport, (ii) how to keep flows in balance over longer times (to avoid mixing), (iii) how to approximate countercurrent flow, (iv) how to avoid excessive protein leakage, and (v) how to design well-behaved inlets and outlets. We report, here, that useful liquid-liquid contact areas behave stably and exchange marker molecules, on prototypes assembled in our laboratory. We show how these prototypes follow predictions of flow and diffusion theory. We describe functional inlets and outlets. We describe photolithographic and silicone casting methods for forming prototypes and demonstrate a ‘flowing sandwich’ (dialysate-blood surrogate-dialysate, see figure) that would eliminate all contact of blood with anything other than dialysate in the transport region of an exchange device.FigureThe challenge of scaling these systems to a level of clinical utility has no precedent in microfluidic research. We thus discuss a smaller system that would support in vitro organogenesis, without transport decrement, as a step toward a wearable clinical dialyzer. Finally, we describe how the technology might be configured to support an adult ESRD patient who uses the device 85% of the time (143 hr/wk).