Twelve young adult mongrel dogs, 35 to 50 kg each, were used in this study. The dogs were acclimated for 10 days before any procedures were performed. Dogs were provided with regular food and water and were exercised during the course of this study, but were placed on a modified soft diet for 2 weeks after surgical procedures were performed.

Insertion of Membranes.
The protocol design and procedures for this study were reviewed and approved by the Louisiana State University Medical Center Institutional Animal Care and Use Committee. All procedures were performed under sterile conditions in an operating room using intramuscular pentobarbital (30 mg/kg) for general anesthesia supplemented with local anesthesia in the form of 2% xylocaine with 1:100,000 epinephrine before any of the surgical procedures. Individual split-thickness flap pouched tunnels 3 mm deep were made over the lateral part of the palate at the first and third rugae posterior to the canine within the connective tissue to provide a good blood supply. Three different membranes (BG, A-P, and A-H) were used, and each dog received a pair of one of the membrane types on opposite sides of their mouth. The tunnel opening was sutured to obtain primary closure.
All dogs received buprenorphine (0.1 mg) im q12 h for pain and 2,000,000 units of bicillin im for 7 days postoperatively. The dogs were evaluated for healing and discomfort every day postoperatively until the palate was healed enough to return to the regular diet.

Biopsies.
According to a presurgically determined treatment code, the dogs were randomly assigned a procedure number, indicating material use and healing times. The treatment code was developed to ensure a balance in materials used, side of jaw treated, and healing times. The following was performed at 1 and 3 or 2 and 4 months, depending on to which group the dog was assigned. Dogs were reanesthetized as above. A biopsy of each tissue-containing rugae was performed using a 4-mmdiameter soft tissue punch (Miltrex Instrument, Lake Success, NY) to obtain a full-thickness cylindrical tissue sample from the surface tissue to the bone. Collagen powder (Avitene, Davol, Cranston, RI) was adapted to the wound for hemostasis. The dogs were evaluated daily postoperatively for healing and placed on a soft diet until the palate was healed enough to return to the regular diet.

Histologic Preparation.
The soft tissue samples were placed in individual vials containing fresh 10% neutral buffered formalin. Sections were mounted so that apical-occlusal cross sections were cut at 6 to 8 mm. A total of 192 sections of 48 biopsies were included in the analysis. Two samples of each biopsy were stained with silver stain (for elastic fiber identification), and two samples of each biopsy were stained with hematoxylin and eosin (for general morphology). Elastic fiber identification aided in distinguishing the membrane from surrounding tissue because those fibers are found in the membranes and very few are found in the palate of the dog.

Histologic Analysis.
Slides were evaluated at 310 magnification using a light microscope (Microstar, American Optical, Buffalo, NY). Two calibrated investigators independently evaluated each slide, and the scores were later compared. If the two investigators were discrepant, they reviewed the slide together to come to a consensus on the rating. The biopsies were compared with a normal dog palate and unused membrane samples. The membrane in each specimen was given a score from 0 to 4 for membrane condition (0 5 intact, easily identifiable; 1 5 slight degradation (,35%); 2 5 moderate degradation (35%–65%); 3 5 severe degradation (.65%); and 4 5 absent, not identifiable). The membrane area in each specimen was evaluated separately for blood vessel penetration, elastic fiber presence, and inflammation. The scoring system used for each of these evaluations was 0 5 none, 1 5 slight, 2 5 moderate, and 3 5 abundant.