Abstract
In a biomedical diagnosis with a limited sample volume and low concentration, droplet-based microfluidics, also called digital microfluidics, becomes a very attractive approach. Previously, our group developed a magnetic-beads-based digital microfluidic immunoassay with a bead number of around 100, requiring less than 1 μL of sample volume to achieve a pg/mL level limit of detection (LOD). However, the bead number in each measurement was not the same, causing an unstable coefficient of variation (CV) in the calibration curve. Here, we investigated whether a fixed number of beads in this bead-based digital microfluidic immunoassay could provide more stable results. First, the bead screening chips were developed to extract exactly 100, 49, and 25 magnetic beads with diameters of less than 6 μm. Then, four calibration curves were established. One calibration curve was constructed by using varying bead numbers (50-160) in the process. The other three calibration curves used a fixed number of beads, (100, 49, and 25). The results indicated that the CVs for a fixed number of beads were evidently smaller than the CVs for varying bead numbers, especially in the range of 1 pg/mL to 100 pg/mL, where the CVs for 100 beads were less than 10%. Furthermore, the calculated LOD, based on the composite calibration curves, could be reduced by three orders, from 3.0 pg/mL (for the unfixed bead number) to 0.0287 pg/mL (for 100 beads). However, when the bead numbers were too high (more than 500) or too low (25 or fewer), the bead manipulation for aggregation became more difficult in the magnetic-beads-based digital microfluidic immunoassay chip.