RESUMO
BACKGROUND: The aim of the present study was to evaluate defect width and two different definitions of defect depth as prognostic factors of periodontal healing in infrabony defects treated by regenerative therapy 6 and 24 months after surgery. METHODS: In 32 patients with moderate to advanced periodontitis, 50 infrabony defects were treated by the guided tissue regeneration (GTR) technique using non-resorbable or bioabsorbable barriers. Clinical parameters were assessed, and 50 triplets of standardized radiographs were taken before surgery and 6 and 24 months after surgery. Using a computer-assisted analysis, the distances cemento-enamel junction (CEJ) to alveolar crest (AC), CEJ to bony defect (BD), horizontal projection of the most coronal extension of the bony wall to the root surface to BD, width, and angle of the bony defects were measured. Depth of the bony defect was 1) calculated as CEJ-BD minus CEJ-AC (INFRA1) and 2) measured as horizontal projection of the most coronal extension of the bony wall to the root surface to BD (INFRA2). RESULTS: Whereas statistically significant vertical clinical attachment level gains (CAL-V: 3.36 +/- 1.59 mm/ 3.41 +/- 1.72 mm; P < 0.001) could be found both 6 and 24 months after surgery, bony fill (0.70 +/- 2.52 mm; P = 0.056/1.21 +/- 2.55 mm; P < 0.005) was significant 24 months post-surgically only. In a multilevel regression analysis, CAL-V gain was predicted by bioabsorbable membrane (P = 0.005), baseline probing depths (PD) (P < 0.001), and actual smoking (P < 0.05). Bony fill could be predicted by baseline depth of the infrabony component as determined by INFRA2 (P < 0.05), angulation of bony defect (P < 0.005), and gingival index at baseline (P < 0.001). In narrow (< 37 degrees) and deep (> or = 4 mm) infrabony defects, bony fill was more pronounced than in wide and shallow defects (P < 0.001). CONCLUSIONS: Improvement achieved by GTR in infrabony defects can be maintained up to 24 months after surgery. Narrow and deep infrabony defects respond radiographically and are to some extent clinically more favorable to GTR therapy than are wide and shallow defects. The infrabony component of bony defects, as determined by the distance from the most coronal extension of the lateral bony wall to BD (INFRA2), is a better predictor of bony fill than that determined by AC-BD (INFRA1).
