Conventionally, blood plasma extraction is accomplished by macroscale centrifugation in hospitals.
Trained personnel are needed to operate the centrifuges which cannot be transported to the point of care (POC). The effort to reduce the complexity of this process and make it feasible for POC applications has been a long-term goal for many research groups in the field. To contribute to this ongoing research, we propose a microfluidic cross-flow filtration device for blood plasma separation. Such a filtration device is intended to be employed as a modular component for POC biosensors, acting as a sample processing module allowing whole blood analysis for small, disposable POC diagnostic systems 7, 41. To find the most promising technology, it is key to review literature on the subject.
The most recent review paper from 2016 by Mielczarek et al. deals with the question “Microfluidic blood plasma separation for medical diagnostics: Is it worth it?” 30. It is a detailed review of research conducted on on-chip blood plasma separation from 2013 to 2016. It found that there is no ideal solution for on-chip blood plasma separation. Every technique has to deal with trade-offs.
Mielczarek et al. states that the most important factors for a successful application of microfluidic plasma separation can be summed up in three main points:1. Avoiding blood dilution2. Making easy to fabricate and cost-effective designs3. Short work flows Our proposal intends to make improvements on state of the art plasma separation technology regarding all the mentioned points. In addition, this review has identified that a comprehensive analysis of plasma quality was often neglected in prior research papers 30. As advancement on this issue, a collaborative investigation with biologists from the Australian centre for Blood Diseases is intended to comprehensively analyse blood plasma yield and its biomarker content with regard to its further usability for biosensors.
Two generally different approaches for separation of blood plasma can be found. Active strategies use an external force field to drive separation, while passive strategies use internal, hydrodynamic forces or channel geometries 6. Active systems are for example dielectrophoresis, magnetic or acoustic separation 22. Passive systems are sedimentation, cell deviation and the strategy, that is used for this project: microfiltration 22. When looking at filtration technology in specific, two primary techniques can be differentiated: Dead-end and cross-flow filtration. In dead-end filtration, the liquid flow is perpendicular to the filter, which leads to the build-up of a cake layer on top of it. The obvious drawback in the application of dead end filtration is therefore rapid clogging of the filter membrane. In cross-flow filtration the liquid flow is in parallel to the filter, so that a cake layer can be partly removed by shear forces between the liquid and the filter.
Multiple research groups investigated the use of cross flow filtration for blood plasma separation. They found, that in order to avoid membrane clogging, additional blood dilution was necessary 18, 46. A close look at the pros and cons of filter technology for blood plasma separation reveals significant results in purity and cost-effectiveness as well as sufficient yields on the microliter scale 6, 30, which makes the technology worth investigating.
As important drawback of filter technology, clogging is mentioned which leads to short lifespans as well as relatively low yields on the millilitre scale. Until now this was mostly avoided through blood dilution 4, 18, 46. Blood dilution is itself rather problematic in POC application as it requires additional fluid handling steps, and reduces the concentration of target biomarkers. To address the issue of membrane clogging, Cheng et al. proposed an on-chip cross-flow filtration device with high throughput using micro pumps and buffer solutions for separation of red and white blood cells. The device measures 50×50 mm. The micropumps are used to pump blood over a cross-flow filter.
The membrane is then flushed by applying reverse mode after a filtering cycle. This showed that clogs were effectively removed 6. This technique of back flushing to unclog the membrane is therefore adapted for this project by use of microfluidic pumps. More research on cross-flow filtering for blood plasma separation in particular, was conducted by Chen et al. who used cross-flow filtration on diluted blood and found that higher separation efficiency can be achieved using a membrane with a smaller pore size of around 1 µm, increasing the dilution of blood and/or by injecting blood at a lower flow rate of about 0.
02 5. The common membrane size for this project is therefore also set to around 1 µm. Within the framework of this project, the objective is to extract a minimum of 1 µl of blood plasma from an undiluted whole blood sample volume in the order of 100µl.
The device is realised with microfluidic channels and a filtration membrane. It will be used as a disposable, modular component for a photonic biosensor. Plasma separation is achieved via cross flow filtering with a membrane pore size of about 1 µm. This setup should lead to sufficient yields, while keeping the complexity low. Blood flow across the filtration membrane is realised by pneumatic micropumps consisting of microfluidic valves fabricated using an injection moulding technique described by Szydzik et al. 42.
Clogging of the membrane is avoided using micropumps which pump low amounts of plasma back into the whole blood and thereby remove potential clots. This technique was first introduced by Cheng et al. 6.