Refine
Year of publication
- 2019 (4) (remove)
Document Type
- Article (4)
Language
- English (4) (remove)
Has Fulltext
- yes (4)
Is part of the Bibliography
- no (4)
Keywords
- periodontitis (4) (remove)
Institute
Publisher
The aim of the present study was to construct a biological age score reflecting one’s physiologic capability and aging condition with respect to tooth loss over 10 y. From the follow-up to the population-based Study of Health in Pomerania (i.e., SHIP-2), 2,049 participants were studied for their baseline biomarker measures 10 y before (i.e., in SHIP-0). Metabolic and periodontal data were regressed onto chronological age to construct a score designated as “biological age.” For either sex separately, the impact of this individualized score was used to predict tooth loss in the follow-up cohort in comparison with each participant’s chronological age. Outcome data after 10 y with respect to tooth loss, periodontitis, obesity, and inflammation were shown to be better for biologically younger subjects than as expected by their chronological age, whereas for the older subjects, data were worse. Especially for tooth loss, a striking increase was observed in subjects whose biological age at baseline appeared to be higher than their chronological age. Biological age produced significantly better tooth loss predictions than chronological age (P < 0.001). Areas under receiver operating characteristic curves for tooth loss of ≥3 teeth in men during follow-up were 0.811 and 0.745 for biological and chronological age, respectively. For women, these figures were 0.788 and 0.724. For total tooth loss, areas under the curve were 0.890 and 0.749 in men and 0.872 and 0.752 in women. Biological age combines various measures into a single score and allows identifying individuals at increased risk of tooth loss.
Evidence is limited regarding whether periodontal treatment improves hemoglobin A1c (HbA1c) among people with prediabetes and periodontal disease, and it is unknown whether improvement of metabolic status persists >3 mo. In an exploratory post hoc analysis of the multicenter randomized controlled trial “Antibiotika und Parodontitis” (Antibiotics and Periodontitis)—a prospective, stratified, double-blind study—we assessed whether nonsurgical periodontal treatment with or without an adjunctive systemic antibiotic treatment affects HbA1c and high-sensitivity C-reactive protein (hsCRP) levels among periodontitis patients with normal HbA1c (≤5.7%, n = 218), prediabetes (5.7% < HbA1c < 6.5%, n = 101), or unknown diabetes (HbA1c ≥ 6.5%, n = 8) over a period of 27.5 mo. Nonsurgical periodontal treatment reduced mean pocket probing depth by >1 mm in both groups. In the normal HbA1c group, HbA1c values remained unchanged at 5.0% (95% CI, 4.9% to 6.1%) during the observation period. Among periodontitis patients with prediabetes, HbA1c decreased from 5.9% (95% CI, 5.9% to 6.0%) to 5.4% (95% CI, 5.3% to 5.5%) at 15.5 mo and increased to 5.6% (95% CI, 5.4% to 5.7%) after 27.5 mo. At 27.5 mo, 46% of periodontitis patients with prediabetes had normal HbA1c levels, whereas 47.9% remained unchanged and 6.3% progressed to diabetes. Median hsCRP values were reduced in the normal HbA1c and prediabetes groups from 1.2 and 1.4 mg/L to 0.7 and 0.7 mg/L, respectively. Nonsurgical periodontal treatment may improve blood glucose values among periodontitis patients with prediabetes (ClinicalTrials.gov NCT00707369).
Prediction models learn patterns from available data (training) and are then validated on new data (testing). Prediction modeling is increasingly common in dental research. We aimed to evaluate how different model development and validation steps affect the predictive performance of tooth loss prediction models of patients with periodontitis. Two independent cohorts (627 patients, 11,651 teeth) were followed over a mean ± SD 18.2 ± 5.6 y (Kiel cohort) and 6.6 ± 2.9 y (Greifswald cohort). Tooth loss and 10 patient- and tooth-level predictors were recorded. The impact of different model development and validation steps was evaluated: 1) model complexity (logistic regression, recursive partitioning, random forest, extreme gradient boosting), 2) sample size (full data set or 10%, 25%, or 75% of cases dropped at random), 3) prediction periods (maximum 10, 15, or 20 y or uncensored), and 4) validation schemes (internal or external by centers/time). Tooth loss was generally a rare event (880 teeth were lost). All models showed limited sensitivity but high specificity. Patients’ age and tooth loss at baseline as well as probing pocket depths showed high variable importance. More complex models (random forest, extreme gradient boosting) had no consistent advantages over simpler ones (logistic regression, recursive partitioning). Internal validation (in sample) overestimated the predictive power (area under the curve up to 0.90), while external validation (out of sample) found lower areas under the curve (range 0.62 to 0.82). Reducing the sample size decreased the predictive power, particularly for more complex models. Censoring the prediction period had only limited impact. When the model was trained in one period and tested in another, model outcomes were similar to the base case, indicating temporal validation as a valid option. No model showed higher accuracy than the no-information rate. In conclusion, none of the developed models would be useful in a clinical setting, despite high accuracy. During modeling, rigorous development and external validation should be applied and reported accordingly.
Periodontitis is one of the most prevalent oral diseases worldwide and is caused by multifactorial interactions between host and oral bacteria. Altered cellular metabolism of host and microbes releases a number of intermediary end products known as metabolites. There is an increasing interest in identifying metabolites from oral fluids such as saliva to widen the understanding of the complex pathogenesis of periodontitis. It is believed that some metabolites might serve as indicators toward early detection and screening of periodontitis and perhaps even for monitoring its prognosis in the future. Because contemporary periodontal screening methods are deficient, there is an urgent need for novel approaches in periodontal screening procedures. To this end, we associated oral parameters (clinical attachment level, periodontal probing depth, supragingival plaque, supragingival calculus, number of missing teeth, and removable denture) with a large set of salivary metabolites (n = 284) obtained by mass spectrometry among a subsample (n = 909) of nondiabetic participants from the Study of Health in Pomerania (SHIP-Trend-0). Linear regression analyses were performed in age-stratified groups and adjusted for potential confounders. A multifaceted image of associated metabolites (n = 107) was revealed with considerable differences according to age groups. In the young (20 to 39 y) and middle-aged (40 to 59 y) groups, metabolites were predominantly associated with periodontal variables, whereas among the older subjects (≥60 y), tooth loss was strongly associated with metabolite levels. Metabolites associated with periodontal variables were clearly linked to tissue destruction, host defense mechanisms, and bacterial metabolism. Across all age groups, the bacterial metabolite phenylacetate was significantly associated with periodontal variables. Our results revealed alterations of the salivary metabolome in association with age and oral health status. Among our comprehensive panel of metabolites, periodontitis was significantly associated with the bacterial metabolite phenylacetate, a promising substance for further biomarker research.