Objective: Develop a test methodology for evaluating the reason for infusion set (IS) failure in persons with diabetes using insulin pump therapy (Continuous Sub-cutaneous Insulin Infusion, CSII). Insulin stability (physical, chemical, and microbiological) and/or the loss of preservatives in the infusate are thought to play significant roles in shortening IS wear time. In-vitro and in-vivo methods were used to assess the cause(s) site inflammation and the loss of preservatives.

Method: Fast-acting insulin (insulin aspart and lispro) were pumped through different infusion sets (>4) under simulated-use conditions. The pumped insulin and non-pumped insulin controls were characterized for physical, chemical, and microbiological properties. The pumped insulin, combined with fibrinogen (an initiator for material foreign body responses), were tested in murine macrophage cell culture for inflammatory responses. The corresponding infusion sets were tested in vivo in a diabetic porcine model to determine IS wear-time.

Result: Pumped insulins in ≥8 insulin/IS combinations met corresponding criteria in USP insulin monographs, indicating no significant chemical changes. Lower preservative content was found to increase insulin aggregates in-vitro/in-vivo. Pro-inflammatory cytokines (MIP-1α, and MCP-1) increased significantly in the inflammatory in-vitro model. The IS wear-time was shortened from 4.9±0.3 days (control) to 1.7±0.3 days (p<0.0001) with lower preservative content (<50% per label claim).

Conclusion: Preliminary findings from in-vitro and in-vivo test model development indicate that macrophage number, the inflammatory response, and device wear time were significantly impacted by loss of preservative and trace aggregates/particles. The new test methodology can further our understanding as to how to improve IS wear time.


S. Chattaraj: Employee; Self; Medtronic. G. Zhang: None. E. Anselmo: Employee; Self; Medtronic. J.C. Fusselman: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.