In vitro fibroblast populated collagen lattices are not good models of in vivo clinical wound healing.

Abstract

In chronic wounds, the healing process is prolonged and incomplete, proceeding in an uncoordinated manner, and resulting in poor anatomical and functional outcome. There have been numerous attempts to discover models that mimic human wound healing processes. The fibroblast populated collagen lattice is one such model that has been proposed. This study evaluated whether the fibroblast populated collagen lattice can be a model of chronic wound healing using the pressure ulcer as a paradigm. Fibroblast cultures of wound biopsies and wound volume measurements were obtained serially during a four arm blinded, placebo-controlled sequential cytokine clinical trial of pressure ulcers. Fibroblasts obtained from study patients were added to collagen lattices and contraction was determined daily for 10 days. Collagen gel-area measurements were converted to reflect percentage of gel contraction. These data of both edge and base wound biopsies on days 0, 10, and 36 were categorized into treatment groups and one-way analysis of variance showed no significant differences in contraction among these groups. When considering all fibroblast populated collagen lattices, there was significantly greater contraction at days 10 and 36 for cells from both edge and base biopsies compared to day 0 (p < 0.05). The Spearman Rank Correlation test comparing all patients with fibroblast populated collagen lattice results from fibroblasts obtained at the edge or base of the wound at days 0, 10, and 36 and clinical pressure ulcer healing on day 36 showed no correlation. This lack of correlation not only persisted for each of the four treatment arms but also for responder status based on decrease in wound volume over the 35 day trial period. In conclusion, chronic wound healing is a complex process that is not modeled by in vitro fibroblast populated collagen lattices.