Periodontal disease is a chronic inflammatory condition characterized not only by clinical attachment loss and alveolar bone destruction but also by alterations in trabecular bone microarchitecture. Conventional radiographic evaluation primarily focuses on macroscopic bone loss and may fail to detect early microstructural changes. Fractal analysis enables quantitative assessment of trabecular bone complexity and may provide additional diagnostic information from routinely acquired dental radiographs. This study aimed to evaluate trabecular bone microarchitectural changes associated with different stages of periodontal disease using fractal dimension (FD) analysis on panoramic radiographs (PRs). A total of 205 male individuals aged 20–40 years were included and allocated into five groups: periodontally healthy, gingivitis, mild periodontitis, moderate periodontitis, and severe periodontitis. Standardized digital PRs were obtained using the same imaging unit and exposure parameters. Twelve predefined bilateral regions of interest (ROIs) were selected from standardized maxillary and mandibular anatomical sites. FD values were calculated using the box-counting method implemented in ImageJ (National Institutes of Health, Bethesda, MD, USA) following standardized image preprocessing steps. Intergroup comparisons were performed using the Kruskal–Wallis test with Bonferroni-adjusted post hoc analysis ( p < 0.05). Significant differences in FD values were observed among periodontal groups in multiple ROIs, particularly in posterior mandibular and condylar regions ( p < 0.05), with a progressive decrease corresponding to increasing periodontal disease severity. No significant differences were detected in selected anterior regions. Periodontal disease is associated with significant alterations in trabecular bone microarchitecture detectable on PRs. Fractal analysis may serve as a valuable adjunctive tool for quantitative assessment of periodontal bone changes and may enhance the interpretative value of routine panoramic imaging.
Laloğlu et al. (Mon,) studied this question.