The aim of the present study was to examine the influence of cross-linking on collagen membranes used for guided bone regeneration of calvarial defects in rats. In 48 Wistar rats, divided equally into 4 groups, 1 control and 3 experimental, standardized transosseous circular calvarial defects were made midparietally. In the control group, the defect was only covered by the soft tissue flap while in the 3 experimental groups, 3 differently cross-linked collagen membranes were interposed between the osseous defect and the overlying flap before suturing. The healing was assessed at 10, 20, and 30 days after surgery. The results showed that augmenting the degree of collagen cross-linking diminished the membrane resorption rate. Compared to the sham-operated sites, the membrane protected defects showed significantly more bone regeneration (on average 4 times more) as attested by histology and measured by histomorphometric analysis. Although the bone gain seemed to augment with increasing degrees of cross-linking, the results within the 3 experimental groups were not statistically different. Since longer healing periods might have been necessary to substantiate results within experimental groups, a study is currently undertaken to evaluate this aspect. This study demonstrated the efficacy of collagen membranes in guiding bone regeneration, as well as the importance of the type and degree of cross-linking.

To retard collagen membrane enzymatic degradation and to increase its mechanical strength, the diphenylphosphorylazide (DPPA) technique has been demonstrated to achieve natural cross-links between peptide chains of collagen without leaving any foreign product in the cross-linked molecule. In the present prospective clinical trial, the potential of a DPPA-cross-linked type I bovine collagen membrane was evaluated in the healing of 15 buccal soft tissue recessions in 15 patients according to the biological concept of guided tissue regeneration. The recession decreased from 3.7 mm (SD 1.4) at baseline to 0.8 mm (SD 1.2) at 2 years postsurgery, corresponding to a mean root coverage of 82.2% (P <0.0001). Concurrently, the clinical attachment level decreased from 5.4 mm (SD 1.6) at baseline to 1.9 mm (SD 1.2) 2 years postsurgery, corresponding to an average clinical attachment gain of 3.5 mm (SD 1.3) (P <0.0001). The 2-year postsurgical width of the keratinized tissue was not significantly different from baseline values. More than half (53%) of the treated sites showed complete root coverage and about two-thirds (73%) of the total cases showed a 75% to 100% disappearance of the mucogingival defect. The present investigation demonstrated that the use of DPPA-cross-linked collagen membranes in the treatment of human buccal soft tissue recessions results in predictable amounts of root coverage and clinical attachment gain. Long-term randomization controlled clinical trials of this material are needed to fully evaluate its potential for treating periodontal recession.

In the present study, the potential of a diphenylphosphorylazide-crosslinked type I bovine collagen membrane was evaluated in the healing of mandibular bone defects applying the biological concept of guided bone regeneration. The experiment was carried out on 25 Wistar rats. After exposing the mandibular ramus bilaterally, 5 mm diameter full-thickness circular bone defects were surgically created. While the defect on one side was covered by the membrane (experimental), the defect on the other side was left uncovered (control) before closure of the overlying soft tissues. The rats were sacrificed in groups of 5 after 7, 15, 30, 90, and 180 days of healing. Although at early stages of healing similar amounts of bone formation were observed in the experimental and control defects, after 1 month of healing, most of the experimental defects were completely closed with new bone, while in the control defects, only limited amounts of new bone were observed at the rims and in the lingual aspect of the lesions. In the 90- and 180-day animals, all experimental defects were completely closed, while in the control defects, no statistically significant increase in bone regeneration was observed. The increase in percentage of bone regeneration in the experimental defects was statistically significant between the 15-day specimens as compared with the 7-day specimens (P < 0.01) and likewise between 30-day and 15-day specimens (P < 0.001). It can be concluded that a DPPA-crosslinked collagen membrane yields biocompatibility, ad hoc mechanical hindrance, and handling characteristics suitable for guided bone regeneration applications in this experimental model.