Phosphatidylethanol (PEth) is a group of phospholipids formed in cell membranes following alcohol consumption. PEth measurement in whole blood samples is established as a specific alcohol biomarker with clinical and medico-legal applications. This study further evaluated the usefulness of dried blood spot (DBS) samples collected on filter paper for PEth measurement. Specimens used were surplus volumes of venous whole blood sent for routine LC–MS/MS quantification of PEth 16:0/18:1, the major PEth homolog. DBS samples were prepared by pipetting blood on Whatman 903 Protein Saver Cards and onto a volumetric DBS device (Capitainer). The imprecision (CV) of the DBS sample amount based on area and weight measurements of spot punches were 23–28%. Investigation of the relationship between blood hematocrit and PEth concentration yielded a linear, positive correlation, and at around 1.0–1.5 μmol/L PEth 16:0/18:1, the PEth concentration increased by ~ 0.1 μmol/L for every 5% increase in hematocrit. There was a close agreement between the PEth concentrations obtained with whole blood samples and the corresponding results using Whatman 903 (PEthDBS = 1.026 PEthWB + 0.013) and volumetric device (PEthDBS = 1.045 PEthWB + 0.016) DBS samples. The CV of PEth quantification in DBS samples at concentrations ≥ 0.05 μmol/L were ≤ 15%. The present results further confirmed the usefulness of DBS samples for PEth measurement.
Whole-blood microsampling provides many benefits such as remote, patient-centric, and minimally invasive sampling. However, blood plasma, and not whole blood, is the prevailing matrix in clinical laboratory investigations. The challenge with plasma microsampling is to extract plasma volumes large enough to reliably detect low-concentration analytes from a small finger prick sample. Here we introduce a passive plasma filtration device that provides a high extraction yield of 65%, filtering 18 mu L of plasma from 50 mu L of undiluted human whole blood (hematocrit 45%) within less than 10 min. The enabling design element is a wedge-shaped connection between the blood filter and the hydrophilic bottom surface of a capillary channel. Using finger prick and venous blood samples from more than 10 healthy volunteers, we examined the filtration kinetics of the device over a hematocrit range of 35-55% and showed that 73 +/- 8% of the total protein content was successfully recovered after filtration. The presented plasma filtration device tackles a major challenge toward patient-centric blood microsampling by providing high-yield plasma filtration, potentially allowing reliable detection of low-concentration analytes from a blood microsample.
Obtaining plasma from a blood sample and preparing it for subsequent analysis is currently a laborious process involving experienced health-care professionals and centrifugation. We circumvent this by utilizing capillary forces and microfluidic engineering to develop an autonomous plasma sampling device that filters and stores an exact amount of plasma as a dried plasma spot (DPS) from a whole blood sample in less than 6 min. We tested 24 prototype devices with whole blood from 10 volunteers, various input volumes (40-80 mu L), and different hematocrit levels (39-45%). The resulting mean plasma volume, assessed gravimetrically, was 11.6 mu L with a relative standard deviation similar to manual pipetting (3.0% vs 1.4%). LC-MS/MS analysis of caffeine concentrations in the generated DPS (12 duplicates) showed a strong correlation (R-2 = 0.99) to, but no equivalence with, concentrations prepared from corresponding plasma obtained by centrifugation. The presented autonomous DPS device may enable patient-centric plasma sampling through minimally invasive finger-pricking and allow generatation of volume-defined DPS for quantitative blood analysis.
Lateral flow assays is an example of a successful microfluidic platform relying on passive fluid transport, making them suitable for patient-centric and point-of-care applications. Flow control and valving in capillary driven devices typically rely on design-imprinted functions and operations which can be a limiting factor. This thesis explores dissolvable polymer valves in capillary driven microfluidic systems, a novel type of valves with a timing function. The dissolvable valve technology was used to develop autonomous operations in lamination-based polymer microfluidic systems such as sequential reagent delivery, reagent release and volume-metering, and further utilizes this technology in the Dried Blood Spot (DBS) and Dried Plasma Spot applications described below. Lamination technology is suitable for the integration of the water-dissolvable polymer layers and allows upscaling at a relatively low cost. Advances in the development of LC-MS/MS systems enable the quantification of analytes in microliter-sized blood samples such as DBS. This makes DBS sampling a minimally invasive alternative to venous blood sampling with logistical and ethical advantages for users and health care providers. Unknown sample volume, spot inhomogeneity and hematocrit-related issues have been an obstacle for a wider acceptance of DBS sampling technology. To address these issues, a novel blood-sampling device, the microfluidic DBS card, has been developed within this thesis. The device function is based on capillary driven volume-metering and allows accurate and user independent collection of microliter-sized DBS, directly from a finger-prick. The microfluidic DBS card could help to eliminate some of the issues related to DBS sampling and contribute to a wider acceptance of the technology. Usability and reliability have been considered during the development to enable testing of the microfludic DBS card in a pre-clinical setting. For many analytes and biomarkers, conventional blood sample analysis is performed on plasma or serum samples. This thesis further discusses the use of capillary driven plasma separation based on commercially available asymmetric filtration membranes and capillary driven flow in microchannels. A novel concept for hematocrit and input-volume-independent collection of a 11.6~µl plasma sample from a single drop of blood is demonstrated. The plasma sample is automatically transferred to a sample collection pad forming a Dried Plasma Spot. This could be the next generation of dried sample matrix, enabling an accurate quantification of analytes in Dried Plasma Spots.
This work presents passive time delay valves for micro channels in two different working modes for microfluidic on chip timing. The delay valve designs are compatible with conventional lamination techniques for microfluidics and allow to pre-program advanced sequential operations independent from the geometries of the microfluidic system. The time delay of the dissolvable valves ranges from 1.2 s up to 36 s per valve resulting in a time range from 1.2 seconds up to 11 minutes for 19 serial valves.
Background: Inhomogeneous sample distribution in DBS is a problem for accurate quantitative analysis of DBS, and has often been explained by chromatographic effects. Results: We present a model describing formation of inhomogeneous DBS during drying of the spot caused by higher evaporation rates of water at the edge as compared with the center. Color intensity analysis shows that the relative humidity and DBS card position affect the homogeneity of DBS. Conclusion: The so-called coffee-stain effect' explains the typical distribution pattern of analytes with higher concentrations measured along the edge of DBS as compared with the center. The driving mechanism and potential influencing factors should be considered when addressing the inhomogeneity of DBS in the future.
Blood plasma samples are widely used in clinical analysis but easy-to-use sampling methods for defined volumes are lacking. We introduce the first capillary driven microfluidic device that separates a specific volume of plasma from a blood sample of unknown volume. The input to the device is a small amount of whole blood in the range of 30-60 μl which results in a 4 μl isolated plasma sample within 3 minutes, available for subsequent processing and/or analysis, as demonstrated by collecting the sample in a paper substrate.
This work presents a disposable chip for metering and transferring an exactly defined liquid volume into a paper matrix by capillary filling and emptying of a microchannel together with self-actuated dissolvable valves. Once a liquid droplet of 20-50 μl is applied to the chip, a volume of 1 μl is automatically metered, separated from the applied volume and subsequently transferred into conventional Whatman 903 paper used in Dried Blood Spot (DBS) sampling. The presented concept allows accurate volume metering for lateral flow devices and is here designed to the specific purpose of metering blood spots for DBS analysis. The material costs for each chip are below 0.04 €.
This work presents a disposable chip for metering and transferring an exactly defined liquid volume into a paper matrix using self-actuated dissolvable valves. Once a liquid droplet of 20-50 μl is applied to the chip, a volume of 1 μl is automatically metered, separated from the applied volume and subsequently transferred into conventional Whatman 903 paper used in Dried Blood Spot (DBS) sampling.
Background: DBS samples collected from a fingerprick typically vary in volume and homogeneity and hence make an accurate quantitative analysis of DBS samples difficult. Results: We report a prototype which first defines a precise liquid volume and subsequently stores it to a conventional DBS matrix. Liquid volumes of 2.2 mu l +/- 7.1% (n = 21) for deionized water and 6.1 mu l +/- 8.8% (n = 15) for whole blood have been successfully metered and stored in DBS paper. Conclusion: The new method of collecting a defined volume of blood by DBS sampling has the potential to reduce assay bias for the quantitative evaluation of DBS samples while maintaining the simplicity of conventional DBS sampling.
Multi-step lab-on-chip assays typically require multiple manual pipetting steps of different reagents and a sophisticated design to run the assay, which is generally unfavorable for point-of-care applications. This work presents the use of dissolvable polyvinylalcohol films for (1) reagent storage and release together with (2) timed valving in capillary driven microfluidics. This allows four different volumes to be split up from a single liquid which is applied to the inlet of the chip. PVA captured reagents are released to each volume forming four different solutions which are separately released in a timed sequence to a common target zone. The presented chip thereby enables a single liquid-triggered multi-reagent sequence with the potential to be used for advanced point of care diagnostics.
Dried blood spot (DBS) sampling is a promising method for collection of microliter blood samples. However, hematocrit-related bias in combination with subpunch analysis can result in inaccurate quantification of analytes in DBS samples. In this study we use a microfluidic DBS card, designed to automatically collect fixed volume DBS samples irrespective of the blood hematocrit, to measure caffeine concentration in normal finger prick samples obtained from 44 human individuals. Caffeine levels originating from blood drops of unknown volume collected on the volumetric microfluidic DBS card were compared to volume-controlled pipetted DBS samples from the same finger prick. Hematocrit independence and volumetric sampling performances were also verified on caffeine-spiked blood samples in vitro, using both LC-MS/MS and gravimetric methods, on hematocrits from 26 to 62%. The gravimetric measurements show an excellent metering performance of the microfluidic DBS card, with a mean blood sample volume of 14.25 μL ± 3.0% (n = 51). A measured mean bias below 2.9% compared to normal hematocrit (47%) demonstrates that there is no significant hematocrit-induced bias. LC-MS/MS measurements confirm low CV and hematocrit independence of the sampling system and exhibit no substantial mean bias compared to pipetted DBS. Tests with 44 individuals demonstrated applicability of the microfluidic DBS card for direct finger prick blood sampling, and measured caffeine concentrations show a good agreement with measurements of pipetted DBS. The presented concept demonstrates a good volumetric performance which can help to improve the accuracy of DBS analysis by analyzing a whole spot, equivalent to a defined volume of liquid blood.