Suitability of machine learning models for prediction of clinically defined Stage III/IV periodontitis from questionnaires and demographic data in Danish cohorts.

AIM: To evaluate if, and to what extent, machine learning models can capture clinically defined Stage III/IV periodontitis from self-report questionnaires and demographic data. MATERIALS AND METHODS: Self-reported measures of periodontitis, demographic data and clinically established Stage III/IV periodontitis status were extracted from two Danish population-based cohorts (The Copenhagen Aging and Midlife Biobank [CAMB] and The Danish Health Examination Survey [DANHES]) and used to develop cross-validated machine learning models for the prediction of clinically established Stage III/IV periodontitis. Models were trained using 10-fold cross-validations repeated three times on the CAMB dataset (n = 1476), and the resulting models were validated in the DANHES dataset (n = 3585). RESULTS: The prevalence of Stage III/IV periodontitis was 23.2% (n = 342) in the CAMB dataset and 9.3% (n = 335) in the DANHES dataset. For the prediction of clinically established Stage III/IV periodontitis in the CAMB cohort, models reached area under the receiver operating characteristics (AUROCs) of 0.67-0.69, sensitivities of 0.58-0.64 and specificities of 0.71-0.80. In the DANHES cohort, models derived from the CAMB cohort achieved AUROCs of 0.64-0.70, sensitivities of 0.44-0.63 and specificities of 0.75-0.84. CONCLUSIONS: Applying cross-validated machine learning algorithms to demographic data and self-reported measures of periodontitis resulted in models with modest capabilities for the prediction of Stage III/IV periodontitis in two Danish cohorts.

Aggressive periodontitis in a 16-year-old Ghanaian adolescent, the original source of Actinobacillus actinomycetemcomitans strain HK1651 - a 10-year follow up.

The highly leukotoxic JP2 clone of Actinobacillus actinomycetemcomitans is strongly associated with periodontitis in adolescents. Availability of the DNA sequence of the complete genome of A. actinomycetemcomitans strain HK1651, a representative strain of the JP2 clone (http://www.genome.ou.edu/act.html), has provided new possibilities in basic research regarding the understanding of the pathogenesis of A. actinomycetemcomitans in periodontitis. This case report describes the periodontal treatment of the original source of A. actinomycetemcomitans HK1651, a 16-year-old Ghanaian adolescent girl with aggressive periodontitis. The bacterial examination involved polymerase chain reaction analysis for presence of JP2 and non-JP2 types of A. actinomycetemcomitans. The treatment, including periodontal surgery supplemented by antibiotics, arrested the progression of periodontitis for more than 10 years. Initially, infection by A. actinomycetemcomitans, including the JP2 clone, was detected at various locations in the oral cavity and was not limited to the periodontal pockets. Post-therapy, the JP2 clone of A. actinomycetemcomitans disappeared, while the non-JP2 types of A. actinomycetemcomitans remained a part of the oral microflora.

Comparison of ready-to-use EMDOGAIN-gel and EMDOGAIN in patients with chronic adult periodontitis.

OBJECTIVES: The aim of this multicenter trial was to compare the clinical and radiographical outcome of a ready-to-use Emdogain-gel (test) with the marketed Emdogain (control). METHODS: Subjects with bilateral infrabony defects > or =4 mm deep and > or =2 mm wide according to radiographs were selected. 88 subjects with probing pocket depth (PPD) > or =6 mm > or =1 month after supervised oral hygiene and scaling participated. At baseline plaque index, bleeding on probing, PPD and probing attachment level were recorded and reproducible radiographs for computer-based bone level measurements were taken. In each subject, 1 tooth was randomly treated with the test and 1 tooth with the control gel. Examinations were repeated 8 and 16 months post-operatively. RESULTS: After 16 months, the mean test PPD was 4.1 mm and the mean control PPD 4.2 mm. The mean gain of attachment was 2.7 mm for test and 2.9 mm for the control sites, and the radiographic measurements demonstrated a mean gain of 1 mm for both test and control sites. CONCLUSION: This series of cases demonstrated a statistically significant reduction of pocket depths and gain of attachment and bone after 8 and 16 months with no difference between the 2 preparations.

Dissolution of type I collagen fibrils by gingival fibroblasts isolated from patients of various periodontitis categories.

The classification of periodontitis in various disease categories, including juvenile periodontitis, rapidly progressive adult periodontitis and slowly progressive adult periodontitis is based mainly on differences in disease progression and age group susceptibility. Because dissolution of collagen fibers is an integral part of periodontal attachment loss, we investigated whether the clinical differences among these periodontitis/control groups are reflected in the collagen-degrading activity of gingival fibroblasts isolated from affected tissues. All fibroblast strains isolated from the 4 groups (n = 48) displayed cell-associated collagenolytic activity when seeded in contact with a reconstituted film of type I collagen fibrils. Cells from the control group (n = 14) dissolved the collagen fibril film twice as fast as those from each of the 3 disease groups (juvenile periodontitis (n = 13), rapidly progressive adult periodontitis (n = 7), and slowly progressive adult periodontitis (n = 14)). Both interleukin-1 beta and phorbolester accelerated the rate of dissolution 2-4-fold, but even after cytokine or phorbolester stimulation control cells were still considerably more effective in dissolving the collagen fibrils than cells from the disease groups. The observation made in this study, that dissolution of collagen fibrils by gingival fibroblasts from periodontally diseased individuals is significantly slower than by cells from healthy control subjects, challenges disease paradigms based on a direct relationship between collagenolytic potential and disease activity.

Factors affecting IL-1-mediated collagen metabolism by fibroblasts and the pathogenesis of periodontal disease: a review of the literature.

Fibroblasts have been studied extensively for their contribution to connective tissue destruction in diseases where the metabolism of extracellular matrix components plays an essential part in their pathogenesis. A considerable dissolution, especially of collagen fibrils, is a well-known characteristic of the periodontal ligament and the gingival connective tissue in microbial-induced periodontal disease. Fibroblasts, responsible for the assembly of the extracellular matrix, are capable of responding directly to oral microbial challenges or indirectly, following activation of the host immune response, and can alter the composition of connective tissue in several ways: synthesis of inflammatory mediators, their receptors and antagonists; fibroblast proliferation; collagen synthesis; phagocytosis of collagen fibrils; and synthesis of proteolytic enzymes, including matrix metalloproteinases and their corresponding inhibitors. The contributions of these cellular fibroblastic properties to the pathogenesis of periodontal disease are reviewed in the context of the cytokine, interleukin-1, as the inflammatory regulator.