Preservation of native extracellular matrix structure correlates with onset temperature in eight commercial collagen wound dressings

摘要

Introduction: Approximately 70-80% of the extracellular matrix (ECM) of the skin is composed of collagen, primarily types Ⅰ & Ⅲ. Following injury, damaged collagen attracts inflammatory cells, succeeded by fibroblasts and keratinocytes to the dermal wound, aiding in debridement, angiogenesis and re-epithelialization, respectively. Given its prominent role in wound healing, collagen has been widely used in commercial tissue engineering and chronic wound care products. In particular, collagen dressings in wound care vary in the preservation of native ECM, ranging from intact ECM to gelatin. The thermal and structural properties of eight commercially available collagen wound care dressings were explored using differential scanning calorimetry (DSC), scanning electron microscopy (SEM) and transmission electron microscopy (TEM). DSC has historically been used to demonstrate the effects of greater tissue processing in lowering the onset of melting temperature (T_0) of regenerative surgical scaffolds, but its use in comparing wound dressings is novel. Materials and Methods: DSC. Samples of CollaSorb (CS; N=45; Hartmann), Endoform Dermal Template (EDT; N=45; Hollister Incorporated), Promogran, Promogran Prisma (Pr and PrPr; N=45 each; Systagenix), Puracol and Puracol Plus (P & PP; N=45 each; Medline), Fibracol Plus (FP; N=15; Systagenix), and Skin Temp Ⅱ (ST; N=15; Human Biosciences) were hydrated in phosphate buffered saline, hermetically sealed in crucibles, and heated. In a preliminary survey, a scan from 2-125°C was performed on the first N=5 from each group. The remaining samples in each group of dressings were scanned from 5-85'C. All DSC scans were performed at 1 °C/min. Thermograms were analyzed using linear method to determine the T_0 of each sample. SEM. Cross-sections and both sides of dry samples of each collagen dressing were imaged at 250,500, & 1000X. TEM. EDT and Pr were hydrated, serially dehydrated and fixed then embedded in Spurr resin prior to imaging at 3524,65231 & 12.114X. Data Analysis: A nested ANOVA was performed to compare average T_0 across dressings. Tuke/s post hoc test and an F-test were used to identify significant differences between dressings and within lots of each dressing, respectively. Results: Overall mean T_0 ± std dev for the groups were CS: 43.26±1.3; EDT: 58.68±1.7; Pr 35.64±1.3; PrPr: 35.05±1.0; P: 41.35±2.8; PP: 41.64±3.7; FP: 36.18±0.4; & ST: 36.33±0.7°C, with EDT differing significantly from the other dressings (p＜0.001). SEM showed preservation of native ECM sidedness in EDT, while all other dressings appeared homogeneous across surfaces. TEM confirmed the presence of native collagen banding periodicity of ～65nm in EDT, which was lacking in Pr. Discussion: Collectively these DSC and SEM data demonstrate greater preservation of native ECM properties in EDT in comparison to the other dressings studied. TEM data confirmed these differences between EDT and Pr. While the impact of these thermal and structural differences on clinical performance are unknown, the retention of native ECM properties has been hypothesized to correlate to the observation of broader in vitro inhibition of proteases by EDT over Pr.