Detailed Analysis of Insulin Absorption Variability and the Tissue Response to Continuous Subcutaneous Insulin Infusion Catheter Implantation in Swine

Abstract

Background: Worldwide, ∼1 million people manage their type 1 diabetes with an insulin pump and a continuous subcutaneous insulin infusion (CSII) catheter. Patients routinely insert a new catheter every 2-3 days due to increasing variability of insulin absorption over time. Catheter insertion and maintenance damage capillaries, lymphatics, cells, and connective tissue leading to an acute inflammatory response.

Methods: We compared an investigational CSII catheter (IC) and a commercial CSII catheter (CC) regarding insulin absorption pharmacokinetics (PK) and tissue inflammation. The two different catheter designs were implanted into the subcutaneous tissue of six swine for 5 days. Insulin boluses were given on days 1, 3, and 5 of wear-time to assess PK. Tissue around catheters was excised and stained to visualize inflammation and morphological changes of adjacent tissue.

Results: Insulin absorption was better when infused through a CC with highest C_max and fastest t_max values on day 5 of catheter wear-time. Both catheter types produced high intra- and intersubject day-to-day insulin absorption variability. The IC caused significantly more tissue disruption and lead to irregular changes in tissue morphology. Both catheter types were surrounded by a layer of inflammatory tissue that varied in composition, thickness, and density over time. A catheter that was manually inserted by pushing a sharp tip through the skin caused more trauma and variability than a 90° Teflon cannula with automated insertion.

Conclusions: Insulin absorption variability could be attributed to the layer of inflammatory tissue, which may function as a mechanical barrier to insulin flow into adjacent vascular tissue. The impact of the acute inflammatory tissue response on insulin absorption has to be considered in future catheter designs. A catheter that was manually inserted by pushing a sharp tip through the skin caused more trauma and variability than a 90° Teflon cannula with automated insertion.

Keywords: Adipose tissue inflammation; CSII catheter; CSII therapy; Inflammatory response; Insulin absorption variability; Insulin pharmacokinetics.

FIG. 1.

Examples of subcutaneous AT histology surrounding two CCs (A, B) and two ICs (C, D) stained (from left to right) with hematoxylin and eosin, Masson's trichrome, and reticulin III. CSII cannulas were removed after tissue fixation leaving a void (V) that previously contained damaged adipose cells, capillaries, lymphatics, connective tissues, extracellular matrix, and thrombus (T). Subcutaneous tissue was pulled downward by cannula insertion leading to regions of compressed connective tissue and adipose cells. Dense areas of dark red cells were produced by hemorrhage, thrombosis, and fibrin deposition (F). Neutrophils, macrophages, fibroblasts, and other immune cells migrated into the damaged tissue to form regions of acute IT that nearly surrounded the implanted cannula. This layer of inflammatory tissue varied in thickness, composition, and continuity at various locations along the cannula shaft. Damaged cells and connective tissue were more pronounced at the cannula's distal end. Reticular fibers (R) were disrupted near the cannula tip. Areas of adipose cell necrosis (N) would have contained free fatty acid, triglyceride, cytoplasm, and organelles. Fibroblasts deposited FC throughout the wound (light pink or light blue). This layer of inflammatory cells, FC, and fibrin nearly surrounded the insertion channel (V) of each specimen. Dermal cells (D) commonly migrated downward around the cannula shaft. The ICs produced two voids in the tissue specimens due to the curved shape of the cannula. The CCs produced a single void due to the straight 6 mm Teflon cannula. AT, adipose tissue; CCs, commercial CSII catheters; CSII, continuous subcutaneous insulin infusion; FC, fresh collagen; ICs, investigational CSII catheter; IT, inflammatory tissue.

FIG. 2.

Insulin concentration versus time curves (A, C) and area under the insulin absorption curve (AUC, B, D) for CC (commercial) and IC (investigational) CSII catheters. (A) CC and (C) IC show average plasma insulin concentration curves on days 1, 3, and 5 of catheter wear-time. Dotted lines represent the average ± SD for the three PK studies. (B) CC and (D) IC show area under the insulin absorption curve (AUC) over 2.5 h. Dotted line represents the average AUC values for six swine on days 1, 3, and 5 (data from swine 6 on day 1 were excluded because we were unable to obtain blood samples). AUC, areas under the curve; PK, pharmacokinetics.

FIG. 3.

Individual plasma insulin concentration curves for six swine on days 1, 3, and 5 starting at the time of a 5 U insulin bolus infusion (t = 0) using (A) CCs and (B) ICs. (See Supplementary Figs. S16 and S17 for corresponding CSII catheter photos and tissue histology photos.)