Acellular dermal matrix allografts to achieve increased attached gingiva. Part 2. A histological comparative study

In part 1 of this study, we compared the clinical efficacy of freeze-dried acellular dermal matrix (ADM) allograft in 6 patients with autogenous free gingival graft (FGG) in 6 patients for increasing the width of attached gingiva in the mandibular anterior area. The purpose of the present study was to histologically compare the microstructure of ADM and FGG treated sites from the same group.

Biopsies were harvested from all 12 patients at 6 months postsurgery. The biopsies included the grafted sites with adjacent alveolar mucosa and gingiva propria and also donor palatal mucosa saved at the time of surgery. The 5 microm thick, neutral buffered formalin fixed, paraffin-embedded tissue sections were stained with hematoxylin and eosin (H&E), Masson's trichrome, and Verhoeff-van Gieson stains in order to investigate the density of collagen and elastic fibers. Additional sections were stained with periodic acid-Schiff (PAS) and Papanicolaou's stain to identify the presence of glycogen granules in the epithelial layer and to highlight the keratin layer respectively.

The unique appearance of ADM-derived tissue did not parallel any known oral mucosa. The connective tissue portion contained dense to extremely dense collagen fibers along with scattered elastic fibers. The demarcations between the ADM graft and the coronal gingiva as well as the apical alveolar mucosa were usually not very defined. A moderate to thin epithelial layer, with heterogeneous expression of keratinization and flat epithelium-connective tissue interface, covered the lamina propria. Both the thickness of the epithelium and the degree of keratinization decreased in apical direction, being mostly para- or orthokeratinized in the area close to gingiva and non-keratinized adjacent to the alveolar mucosa. In the FGG-treated sites, the density of collagen fibers was less than in ADM-derived tissue, palatal mucosa, and gingiva. Elastic fibers were very sparse, comparable to gingiva, but much less than in ADM-derived tissue. The epithelium was moderate, somewhat thinner but the shape of the rete ridges resembled that of palatal mucosa. Similar to gingiva epithelium, the epithelium of the FGG-treated area was relatively uniform in both thickness and keratinization, mostly para-keratinized with a well defined border to the non-keratinized alveolar mucosa. Underneath the FGG-alveolar mucosa junction, a scar band composed of extremely dense collagen fibers consistently existed.

The results of this 6-month histological evaluation suggest that: 1) the resultant tissue types of ADM grafts were similar to "scar" tissue; 2) the non-vital dermal matrix of ADM allograft lacked the capability of directing cyto-differentiation of the covering epithelium; 3) autogenous FGG-derived tissue was neither identical to donor palatal mucosa nor to adjacent gingiva propria; 4) the connective tissue of donor palatal mucosa only partially contributed to the differentiation of the epithelium covering the FGG-treated area; and 5) the epithelium/connective tissue microenvironment surrounding the recipient site influenced the epithelial differentiation of the graft; this may play a more critical role in ADM grafting than in the grafting of autogenous FGG.