Engineering a bijel-templated material-integrated cannula for subcutaneous insulin delivery: Design, fabrication, in...

Continuous subcutaneous insulin infusion (CSII) improves glycemic control in people with type 1 diabetes. However, CSII reliability depends on insulin infusion sets (IISs), whose wear time is limited by device-related flow failures (e.g., occlusions) and tissue responses at the insertion site (i.e., foreign body response). This study reports the design, fabrication, and in vivo evaluation of an IIS that integrates a bijel-templated material (BTM) at the tip of the cannula, with the goal of improving subcutaneous drug delivery through a high-permeability porous network hypothesized to promote tissue integration. The BTM was produced by templating a particle-jammed emulsion gel formed by spinodal decomposition, then laser-shaped into a stepped geometry, mechanically reinforced, and integrated into fluoropolymer tubing. In addition, a custom boat-shaped needle was designed to guide the placement of the BTM-filled cannula. The device was evaluated in a pilot feasibility and failure-analysis study in a diabetic swine glucose clamp model with pharmacokinetic (PK) and pharmacodynamic (PD) testing, inline pressure monitoring, and end-of-wear microCT and histology to evaluate failure modes. The initial devices (Embodiment 1) exhibited seal leakage, elevated pressures, and periods of weak or non-physiological PK/PD responses, motivating a redesign (Embodiment 2) with a shorter intra-cannula BTM and reinforced junction seals. In the two animals in which it was evaluated, Embodiment 2 produced more consistent PK/PD responses than in the two animals for Embodiment 1, with mid-wear values that fell within the range of commercial IIS controls; these comparisons are descriptive and non-powered. The investigational devices did not improve flow reliability, exhibiting higher maximum inline pressures and more frequent pump-detected occlusions than controls. MicroCT and histology identified mechanical vulnerabilities, including kink-related flow restrictions, insertion tract backflow, and BTM-cannula separation. The findings motivate next-generation designs where the cannula transitions directly into a porous microarchitecture, removing delamination-prone interfaces, reducing the insertion profile, and better preserving porous outflow during wear.