Remodelling of periodontal tissues adjacent to sites treated according to the principles of guided tissue regeneration (GTR)

Abstract

The aim of the present study was to assess the remodelling of alveolar bone adjacent to periodontal sites following therapy according to the principles of guided tissue regeneration (GTR) using computer‐assisted densitometric image analysis (CADIA), and to compare the radiographic results to traditional clinical parameters. As required for digital subtraction analyses, periodically reproducible radiographs were obtained using a modification of the Rinn System and individual acrylic bite blocks for periodical identical radiographs. Ideally, a digital subtraction image from a site where absolutely no change in density had occurred would show a perfect cancellation of the structures. An average grey level value of 128 (the middle of the digitizer grey level range set by software) would show up at each pixel. Areas with grey levels < 128 in the subtraction image would indicate loss in density and grey levels > 128 would indicate increase in density. Within the subtraction images, areas were defined using the cursor to draw “regions of interest” (ROI) projected on the bony defect exposed to GTR covering the crestal bone as well as the region of potential “bonefill”. The mean, median, the standard deviation and range of the grey levels of pixels within a particular ROI were calculated. Similarly sized ROI were drawn in bone areas not exposed to the GTR procedure serving as controls. The differences in the mean grey levels of all pixels within a particular ROI between the baseline, 3 and 12 months images were calculated for documentation of gain or loss in density. From 14 patients, standardized radiographs were available from baseline, 3 months and 12 months postsurgically, depicting one infraosseous defect before and after treatment according to the principles of GTR. The densitometric changes observed in these defects were compared to the clinically assessed changes measured at the site with the deepest baseline pocket depth. A mean clinical attachment gain of 2.36 mm after 3 and 3.22 mm after 12 months was measured. This was associated with a mean reduction in the PPD amounting to 3.36 mm and 3.79 mm. respectively. The changes in the level of the FGM were rather small considering the deep original mean PPD of 7.07 mm. Over the first months, a mean recession of 1.14 mm was observed which was followed by a coronal displacement of 0.43 mm. With respect to the remodelling of the alveolar bone adjacent to the defects assessed by means of CADIA, the most pronounced changes occurred when comparing the baseline to the 12 months radiographs. In the resulting subtraction images, the mean grey level was elevated over 128 when ROIs were evaluated projected on the test sites. The mean increase in density was not only indicated by the differences in the mean grey levels but also by their high standard deviations in the test ROI in such subtraction images. When ROIs projected on alveolar bone sites distant from the GTR procedures were evaluated over time, the mean grey levels were constant near 128 and the standard deviations were smaller compared to the test sites. The data obtained from the control sites represent the density‐changes measured due to physiologic remodelling and/or changes due to the sum of the methodological errors. When correlations between the changes in density and the traditional periodontal parameters were calculated, rather low correlation coefficients were obtained, stressing the different characters of information provided by the radiographic and by the clinical probing diagnostic procedures. In summary, the soft periodontal tissues demonstrated a fast repopulation of the previously diseased root surface which was followed by a slow delayed reorganisation of the soft tissue structures. The alveolar bone adjacent to sites exposed to GTR demonstrated a slow consolidation of the tissues represented by mineralization occurring delayed, compared to the changes in the probing attachment levels.