A multiplexed three-dimensional paper-based electrochemical impedance device for simultaneous label-free affinity sensing of total and glycated haemoglobin: The potential of using a specific single-frequency value for analysis

A novel three-dimensional paper-based electrochemical impedance device (3D-PEID) is first introduced for measuring multiple diabetes markers. Herein, a simple 3D-PEID composed of a dual screen-printed electrode on wax-patterned paper coupled with a multilayer of magnetic paper was fabricated for label-free electrochemical detection. The results clearly demonstrated in a step-wise manner that the haptoglobin (Hp)-modified and 3-aminophenylboronic acid (APBA)-modified eggshell membranes (ESMs) were highly responsive to a clinically relevant range of total (0.5–20 g dL⁻¹; r² = 0.989) and glycated haemoglobin (HbA1c) (2.3%–14%; r² = 0.997) levels with detection limits (S/N = 3) of 0.08 g dL⁻¹ and 0.21%, respectively. The optimal binding frequencies of total haemoglobin and HbA1c to their specific recognition elements were 5.18 Hz and 9.99 Hz, respectively. The within-run coefficients of variation (CV) were 1.84%, 2.18%, 1.72%, and 2.01%, whereas the run-to-run CVs were 2.11%, 2.41%, 2.08%, and 2.21%, when assaying two levels of haemoglobin and HbA1c, respectively. The CVs for the haemoglobin and HbA1c levels measured on ten independently fabricated paper-based sheets were 1.96% and 2.10%, respectively. These results demonstrated that our proposed system achieved excellent precision for the simultaneous detection of total haemoglobin and HbA1c, with an acceptable reproducibility of fabrication. The long-term stability of the Hp-modified eggshell membrane (ESM) was 98.84% over a shelf-life of 4 weeks, enabling the possibility of storage or long-distance transport to remote regions, particularly in resource-limited settings; however, for the APBA-modified ESM, the stability was 92.35% over a one-week period. Compared with the commercial automated method, the results demonstrated excellent agreement between the techniques (p-value < 0.05), thus permitting the potential application of 3D-PEID for the monitoring of the glycaemic status in diabetic patients.