Periodontal disease is one of the most widespread medical conditions and the leading cause of tooth loss, imposing high economic costs and an increasing burden worldwide as life expectancy increases. Changes in the subgingival bacterial community during the progression of periodontal disease can affect the entire oral ecosystem, and bacteria in saliva can reflect the degree of bacterial imbalance in the oral cavity.

Ko Y, et al. J Periodontal Implant Sci. 2020 Jun;50(3):171-182. https://doi.org/10.5051/jpis.2020.50.3.171

The purpose of this study was to determine whether subgingival microbial changes according to the severity of periodontal disease and following the non-surgical periodontal treatment of periodontitis are reflected in the saliva microbiota. Subgingival and saliva samples were collected from 7 periodontally healthy controls, 14 patients with gingivitis, 12 with moderate periodontitis, and 18 with severe periodontitis. Among subjects who received treatment, seven moderate and seven severe periodontitis patients were selected for post-treatment microbial analysis, and their samples were analyzed at baseline and 6 months after treatment. The V3 and V4 regions of the 16S rRNA gene were sequenced, and correlations of the relative abundance of phyla and health- or periodontitis-dominant species between subgingival plaque and saliva samples were analyzed using Spearman signed-rank tests. Alpha diversity was higher in saliva than subgingival plaque samples, and beta diversity was significantly different between the two samples. However, levels of phyla and most health- or periodontitis-dominant species in salivary microbiota were significantly correlated with those in subgingival plaque. The improvement in clinical parameters following treatment was accompanied by a microbial shift not only in subgingival plaque but also in saliva. The abundance of 2 phyla including Bacteroidetes, 6 genera including Porphyromonas and Treponema , and 11 species including Porphyromonas gingivalis , Tannerella forsythia , and Filifactor alocis was significantly reduced in saliva following treatment. These results indicate that the salivary microbiota can reflect changes in the subgingival microbiota, suggesting that saliva can be used as a diagnostic tool to monitor the periodontal health status of individuals. IMPORTANCE: The salivary microbiota has attracted increasing attention as a promising method for monitoring periodontal disease. With regard to the pathogenesis of periodontal disease, however, subgingival plaque microbiota is the dominant etiological factor. Although it has been established that periodontopathogenic bacteria exist in saliva and their distribution differs, depending on the severity of the disease, it is necessary to analyze the extent to which the salivary microbiota reflects the subgingival microbiota. This study explored whether subgingival microbial changes according to the severity of periodontal disease and following the non-surgical periodontal treatment of periodontitis are reflected in the saliva microbiota and concluded that the salivary microbiota can reflect changes in the subgingival microbiota. Saliva can be used as a diagnostic tool to monitor the periodontal health status of individuals.

This study aimed to investigate the performance and reliability of data-driven models employing correlational feature analysis and clinical validation for predicting periodontal disease. The 7th Korea National Health and Nutrition Examination Survey (n = 10,654) was used for correlation analysis to identify significant risk factors for periodontitis. Periodontal prediction models were developed with the selected factors and database, followed by internal validation with 5-fold cross-validation and 1000 bootstrap resampling. External validation was conducted with clinical data (n = 120) collected through self-reported questionnaires, clinical periodontal parameters, and radiographic image analysis. Predictive performance was assessed for logistics regression, support vector machine, random forest, XGBoost, and neural network algorithms using the area under the receiver operating characteristic curves (AUC) and other performance metrics. Correlation analysis identified 16 features from over 1000 potential risk factors for periodontitis. The best data-driven model (XGBoost) showed AUC values of 0.823 and 0.796 for internal and external validations, respectively. Modeling with clinical data revealed those same measures to be 0.836 and 0.649, respectively. In addition, the data-driven model could predict other clinical periodontal parameters including severe bone loss (AUC = 0.813), gingival bleeding (AUC = 0.694), and tooth loss (AUC = 0.734). A patient case study about prognostic predictions revealed that the probability of periodontitis can be reduced by 6.0 % (stop smoking) and 0.6 % (stop drinking) on average. Data-driven models for predicting periodontitis and other periodontal parameters were developed from 16 risk factors, demonstrating enhanced prediction performance and reproducibility in internal–external validations.

Odontogenic Ameloblast-Associated Protein (ODAM) in gingival crevicular fluid (GCF) can provide evidence of the detachment of junctional epithelium from the tooth surface by periodontitis. This study sought to investigate the ability of ODAM to reflect the severity of periodontitis at a site-specific level; thus whether there was a relationship between clinical diagnostic parameters and the value of ODAM in GCF was analyzed. Eight periodontitis patients with various severities were enrolled, and the clinical parameters and samples of GCF were obtained from 44 to 60 sites of each subject. The ODAM concentration was quantified by enzyme-linked immunosorbent assay. Correlation analyses between clinical parameters and ODAM values and unadjusted and adjusted (linear) mixed model analyses were performed. The accuracy of ODAM to reflect sites having a probing depth (PD) ≥ 5 mm and a positive bleeding on probing (BOP) was evaluated by receiver-operating characteristic analysis. A total of 424 GCF samples were collected. The mean ODAM concentration from each patient varied from 0.2 to 1.52 ng/ml. Correlations between PD or clinical attachment level (CAL) and ODAM values were found (p < 0.0001). An adjusted linear mixed model showed that PD or CAL were associated with ODAM values (p < 0.05). The area under the curve of ODAM, which reflected sites with PD ≥ 5 mm and positive BOP, was 0.661 (p < 0.0001). This result shows the possibility of GCF ODAM as a site-specific biomarker for periodontal tissue destruction.