Insulin induces a progressive increase in the resistance of subcutaneous tissue to fluid flow: Implications for insulin pump therapy

The DERMIS Study: Methodologies, Results, and Implications for the Future

Ongoing innovation in diabetes technologies has led to the development of advanced tools such as automated insulin delivery (AID) systems that adjust insulin delivery in response to current and predicted glucose levels, residual insulin action, and other inputs (eg, meal and exercise announcements). However, infusion sets continue to be the "Achilles heel" of accurate and precise insulin delivery and continued device use. A recent study by Kalus et al (DERMIS Study) revealed higher vessel density and signals of inflammation by optical coherence tomography (OCT), in addition to increased inflammation, fat necrosis, fibrosis, and eosinophilic infiltration by histopathology. Although the study provided a comprehensive description of what was happening, the results raise important questions that require additional research. On February 29, 2024, the Leona M. and Harry B. Helmsley Charitable Trust sponsored a conference to begin addressing these issues. This article summarizes the DERMIS study findings and testing methodologies discussed at the conference and proposes the next steps for developing insulin infusion sets that reduce the variability in insulin delivery and extend wear.

Development of a Real-time Force-based Algorithm for Infusion Failure Detection

BACKGROUND

Continuous subcutaneous insulin infusion (CSII) is a common treatment option for people with diabetes (PWD), but insulin infusion failures pose a significant challenge, leading to hyperglycemia, diabetes burnout, and increased hospitalizations. Current CSII pumps' occlusion alarm systems are limited in detecting infusion failures; therefore, a more effective detection method is needed.

METHODS

We conducted five preclinical animal studies to collect data on infusion failures, utilizing both insulin and non-insulin boluses. Data were captured using in-line pressure and flow rate sensors, with additional force data from CSII pumps' onboard sensors in one study. A novel classifier model was developed using this dataset, aimed at detecting different types of infusion failures through direct utilization of force sensor data. Performance was compared against various occlusion alarm thresholds from commercially available CSII pumps.

RESULTS

The testing dataset included 251 boluses. The Bagging classifier model showed the highest performance metrics among the models tested, exhibiting high accuracy (96%), sensitivity (94%), and specificity (98%), with lower false-positive and false-negative rate compared with traditional occlusion alarm pressure thresholds.

CONCLUSIONS

Our study developed a novel non-threshold classifier that outperforms current occlusion alarm systems in CSII pumps in detecting infusion failures. This advancement has the potential to reduce the risk of hyperglycemia and hospitalizations due to undetected infusion failures, offering a more reliable and effective CSII therapy for PWD. Further studies involving human participants are recommended to validate these findings and assess the classifier's performance in a real-world setting.

Therapeutic Insulin Analogue Concentrations at Infusion Sites Enhanced the Pro-Inflammatory Response and Apoptosis in an In Vitro Macrophage-Material Interaction Model

Continuous subcutaneous insulin infusion for Type 1 diabetes relies upon insulin infusion sets (IIS) to reliably deliver insulin to a subcutaneous depot, where it is absorbed into systemic circulation. However, IIS are plagued by short wear times and high failure rates, due in part to inconsistent insulin absorption that can arise over time. While emerging evidence suggests that the local inflammatory response to the IIS cannula may impact both wear times and unreliable insulin adsorption, the mechanisms are poorly understood. Here, we investigated the effects of local infused insulin concentrations on the biomaterial host response to better understand the underlying factors that limit the IIS performance. We first modeled the insulin concentration for a constant basal infusion rate to select a relevant insulin concentration range of 0.1-10 U/mL within the infusion site. We then examined the influence of a commercial insulin analogue (Humulin-N) using an in vitro macrophage-material model, which uses adsorbed fibroblast lysate (containing damage-associated molecular patterns) to activate macrophages and recapitulates macrophage responses on implanted biomaterials. RAW-Blue macrophages cultured on lysate-adsorbed surfaces had increased nuclear factor-κB (NF-κB) and activating protein 1 (AP-1) activity and intracellular reactive oxygen species (ROS) accumulation compared to control surfaces. Humulin-N concentration (0.5-10 U/mL) enhanced the NF-κB/AP-1 activity and ROS accumulation in macrophages on lysate-adsorbed surfaces. However, Humulin-N had no effect on NF-κB/AP-1 or ROS in the absence of the inflammatory stimulus. Additionally, high insulin concentrations arising from therapeutic doses induced macrophage apoptosis with and without adsorbed lysate. This study contributes to emerging evidence that infused insulin affects the tissue response to IIS.

Performance of a Continuous Subcutaneous Insulin Infusion (CSII) Pump With Acoustic Volume and Flow Sensing in Simulated High-Consequence Situations

Goal: An insulin pump's failure to deliver insulin in the right amount at the right time is a preventable cause of hospitalization. We evaluated key performance metrics of a novel insulin pump that prevents "silent insulin non-delivery" caused by blockage, delivery of air and site leakage. This is accomplished via an acoustic sensor that measures the volume of insulin delivered with each pulse in real-time. Methods: We tested long and short-term flow accuracy, occlusion-detection time and pressure, and air management of the new device (ND) versus 3 U.S. commercial insulin pumps (CIPs) using standardized methods. Results: The ND outperformed CIPs on long-term basal flow rate error. Occlusion detection was 5 to 22.5 times faster depending on the basal rate and resulted in significantly lower (2 to 5x) pressures at time of occlusion. With air included in the drug reservoir, the tested CIPs can infuse air without detection, while the ND prevented air delivery without interruption. Conclusions: Bench tests of the ND versus 3 commercially available pumps showed improved occlusion detection and air management without flow performance tradeoffs. Additionally, the lower delivery pressure measured at time of occlusion suggests a substantially lower potential for site leakage at both basal and bolus rates. These enhancements combine to decrease the likelihood of silent insulin non-delivery.

New ex vivo method to objectively assess insulin spatial subcutaneous dispersion through time during pump basal-rate based administration

Glycemic variability remains frequent in patients with type 1 diabetes treated with insulin pumps. Heterogeneous spreads of insulin infused by pump in the subcutaneous (SC) tissue are suspected but were barely studied. We propose a new real-time ex-vivo method built by combining high-precision imaging with simultaneous pressure measurements, to obtain a real-time follow-up of insulin subcutaneous propagation. Human skin explants from post-bariatric surgery are imaged in a micro-computed tomography scanner, with optimised parameters to reach one 3D image every 5 min during 3 h of 1UI/h infusion. Pressure inside the tubing is recorded. A new index of dispersion (IoD) is introduced and computed upon the segmented 3D insulin depot per time-step. Infusions were hypodermal in 58.3% among 24 assays, others being intradermal or extradermal. Several minor bubbles and one occlusion were observed. IoD increases with time for all injections. Inter-assay variability is the smallest for hypodermal infusions. Pressure elevations were observed, synchronised with air bubbles arrivals in the tissue. Results encourage the use of this method to compare infusion parameters such as pump model, basal rate, catheter characteristics, infusion site characteristics or patient phenotype.