Evidence for healing of class II and class III furcations 24 months after guided tissue regeneration therapy: Digital subtraction and clinical measurements

Background: The aim of this study was to compare bone regeneration as assessed by subtraction radiography with clinical attachment measurements 24 months after regenerative surgery. Methods: In 24 patients with advanced periodontitis, 34 Class II and 10 Class III furcations were treated by guided tissue regeneration using expanded polytetrafluoroethylene (n = 18) or bioabsorbable barriers (n = 26). Clinical parameters were assessed and 40 pairs of standardized radiographs were taken before and 24 months after surgery. Bony gain within furcations was assessed using subtraction analysis. Results: In Class II furcations, statistically significant (P <0.001) vertical (CAL-V: 1.06 +/- 1.56 mm) and horizontal (CAL-H: 2.05 +/- 1.24 mm) attachment gain was observed 24 months postsurgically. Within Class III defects CAL-V gain was assessed 1.78 +/- 1.43 mm (P <0.005). CAL-H gain in buccal and lingual Class II furcations was explained by the relative density gain, baseline CAL-H, gingival index (GI) reduction, and individual patients (P <0.001) whereas, multiple regression analysis failed to show a relation between CAL-H gain and any of the subtraction parameters for interproximal furcations. Vertical attachment gain in interproximal sites was explained by net bony area gain (subtraction), PD and GI reduction, baseline CAL-V, defect type, smoking, and individual patients (P <0.001). Conclusions: There was a statistically significant correlation between clinical and radiographic improvements within furcation defects. However, this correlation was modulated by other factors such as baseline CAL-V or CAL-H, PD and GI reduction, or smoking. The relative density gain seemed to sufficiently describe bony changes related to clinical changes within buccal and lingual furcations, whereas bone area gain suitably described bony changes in interproximal Class II and III furcations.