Risk factors for tooth loss and progression of periodontitis in patients undergoing periodontal maintenance therapy

Convolutional neural networks combined with classification algorithms for the diagnosis of periodontitis

OBJECTIVES

We aim to develop a deep learning model based on a convolutional neural network (CNN) combined with a classification algorithm (CA) to assist dentists in quickly and accurately diagnosing the stage of periodontitis.

MATERIALS AND METHODS

Periapical radiographs (PERs) and clinical data were collected. The CNNs including Alexnet, VGG16, and ResNet18 were trained on PER to establish the PER-CNN models for no periodontal bone loss (PBL) and PBL. The CAs including random forest (RF), support vector machine (SVM), naive Bayes (NB), logistic regression (LR), and k-nearest neighbor (KNN) were added to the PER-CNN model for control, stage I, stage II and stage III/IV periodontitis. Heat map was produced using a gradient-weighted class activation mapping method to visualize the regions of interest of the PER-Alexnet model. Clustering analysis was performed based on the ten PER-CNN scores and the clinical characteristics.

RESULTS

The accuracy of the PER-Alexnet and PER-VGG16 models with the higher performance was 0.872 and 0.853, respectively. The accuracy of the PER-Alexnet + RF model with the highest performance for control, stage I, stage II and stage III/IV was 0.968, 0.960, 0.835 and 0.842, respectively. Heat map showed that the regions of interest predicted by the model were periodontitis bone lesions. We found that age and smoking were significantly related to periodontitis based on the PER-Alexnet scores.

CONCLUSION

The PER-Alexnet + RF model has reached high performance for whole-case periodontal diagnosis. The CNN models combined with CA can assist dentists in quickly and accurately diagnosing the stage of periodontitis.

Periodontal Regeneration of Vital Poor Prognosis Teeth with Attachment Loss Involving the Root Apex: Two Cases with up to 5 Years Follow-Up

Teeth with attachment loss involving the root apex are severely compromised and have a poor periodontal prognosis. In cases where periodontal regeneration is possible, current guidelines suggest that endodontic treatment is performed first. However, root canal treatment increases the overall treatment time and costs, has risks of endodontic complications, and could predispose teeth to mechanical failure. In this case report, two patients diagnosed with periodontitis stage III/IV grade C, no history of smoking or diabetes, and attachment loss involving the root apex of a tooth, were treated with guided tissue regeneration. These two cases are unique because successful periodontal regeneration was carried out without endodontic treatment, and the vitality of these teeth was maintained longitudinally. This report presents the management that led to this clinical outcome, and important guidelines for case selection are identified. Within the limitations of this study, vital teeth with radiographic bone loss involving the apex may be treated successfully with periodontal regeneration and remain vital at least in the short- to medium-term.

Influence of mobility on the long-term risk of tooth extraction/loss in periodontitis patients. A systematic review and meta-analysis

The aim of this systematic review (SR) was to assess whether tooth mobility (TM) increases the risk of tooth extraction/loss. The protocol was registered in PROSPERO database (CRD42023485425). The focused PECO questions were as follows: (1) "In patients with periodontitis, undergoing periodontal treatment, are teeth affected by mobility at higher risk of being extracted/lost compared to non-mobile teeth, with a minimum follow-up of 10 years?" and (2) "In these patients, does varying degrees of tooth mobility increase the risk of tooth extraction/loss, with a minimum follow-up of 10 years?". Results were reported according to PRISMA statement. Electronic and manual searches were conducted to identify longitudinal studies. The different assessments of tooth mobility were pooled into three groups: TM0: Undetectable tooth mobility, TM1: Horizontal/Mesio-distal mobility ≤1 mm, TM2: Horizontal/Mesio-distal mobility >1 mm or vertical tooth mobility. Tooth loss was the primary outcome. Various meta-analyses were conducted, including subgroup analyses considering different follow-up lengths and the timing of TM assessment, along with sensitivity analyses. A trial sequential analysis was also performed. Eleven studies were included (1883 patients). The mean follow-up range was 10-25 years. The weighted total of included teeth, based on the sample size, was 18 918, with a total of 1604 (8.47%) extracted/lost teeth. The overall rate of tooth extraction/loss increased with increasing mobility: TM0 was associated with a 5.85% rate (866/14822), TM1 with the 11.8% (384/3255), TM2 with the 40.3% (339/841). Mobile teeth (TM1/TM2) were at an increased risk for tooth extraction/loss, compared to TM0 (HR: 2.85; [95% CI 1.88-4.32]; p < .00001). TM1 had a higher risk than TM0 (HR: 1.96; [95% CI 1.09-3.53]; p < .00001). TM2 had a higher risk than TM1 (HR: 2.85; [95% CI 2.19-3.70]; p < .00001) and TM0 (HR: 7.12; [95% CI 3.27-15.51]; p < .00001). The results of the tests for subgroup differences were not significant. Sensitivity meta-analyses yielded consistent results with other meta-analyses. Within the limits of the quality of the studies included in the meta-analyses, mobile teeth were at higher risk of being extracted/lost in the long-term and higher degrees of TM significantly influenced clinicians' decision to extract a tooth. However, most teeth can be retained in the long-term and thus TM should not be considered a reason for extraction or a risk factor for tooth loss, regardless of the degree of TM.

Influence of residual pockets on periodontal tooth loss: A retrospective analysis

BACKGROUND

Individuals enrolled in supportive periodontal therapy (SPT) can still present with tooth loss due to periodontitis (TLP). There is limited evidence on the influence of residual pockets (RPc) and a defined "threshold" at which a patient's profile is set to be at high risk for TLP in the literature. Therefore, this study aimed to assess the influence of RPc on TLP and determine the prognostic performance of RPc compared to the staging and grading of periodontitis on TLP risk.

METHODS

Clinical data from 168 patients (3869 teeth) treated for periodontitis and receiving SPT for at least 10 years were evaluated in this retrospective study. TLP and the percentage of sites with RPc ≥ 5 mm or ≥6 mm per patient were collected. The prognostic performance of RPc was compared to the staging and grading of the disease on TLP using a multilevel Cox proportional hazard regression model.

RESULTS

Over a median follow-up of 25 years, 13.7% of teeth were lost, 4.6% of which were due to periodontitis. Most patients with TLP had ≥1 site with RPc ≥5 mm (90.8%) or ≥6 mm (77.6%). Multivariate multilevel Cox regression revealed that patients with >15% of sites with RPc ≥5 mm had a hazard ratio of 2.34, and grade C had a hazard ratio of 4.6 for TLP compared to RPc ≤4 mm/grade A. Grading exhibited the best discrimination and model fit.

CONCLUSION

Patients with RPc ≥5 mm at >15% of the sites are at risk for tooth loss. Grading and RPc ≥5 mm displayed very good predictive capability of TLP.

Prevalence of stable and successfully treated periodontitis subjects and incidence of subsequent tooth loss within supportive periodontal care: A systematic review with meta-analyses

AIM

To identify (i) the prevalence of meeting the endpoints of 'stable periodontitis' (probing pocket depth [PPD] ≤ 4 mm, bleeding on probing [BoP] < 10%, no BoP at 4 mm sites), 'endpoints of therapy' (no PPD > 4 mm with BoP, no PPD ≥ 6 mm), 'controlled periodontitis' (≤4 sites with PPD ≥ 5 mm), 'PPD < 5 mm' and 'PPD < 6 mm' at the start of supportive periodontal care [SPC]) and (ii) the incidence of tooth loss in relation to not meeting these endpoints within a minimum of 5 years of SPC.

MATERIALS AND METHODS

Systematic electronic and manual searches were conducted to identify studies where subjects, upon completion of active periodontal therapy, entered into SPC. Duplicate screening was performed to find relevant articles. Corresponding authors were contacted to confirm inclusion and retrieve required clinical data for further analyses to assess the prevalence of reaching endpoints and incidence of subsequent tooth loss, if available, within at least 5 years of SPC. Meta-analyses were carried out to evaluate risk ratios for tooth loss in relation to not reaching the various endpoints.

RESULTS

Fifteen studies including 12,884 patients and 323,111 teeth were retrieved. Achievement of endpoints at baseline SPC was rare (1.35%, 11.00% and 34.62%, respectively, for 'stable periodontitis', 'endpoints of therapy' and 'controlled periodontitis'). Less than a third of the 1190 subjects with 5 years of SPC data lost teeth-a total of 3.14% of all teeth were lost. Statistically significant associations with tooth loss, at the subject-level, were found for not achieving 'controlled periodontitis' (relative risk [RR] = 2.57), PPD < 5 mm (RR = 1.59) and PPD < 6 mm (RR = 1.98).

CONCLUSIONS

An overwhelming majority of subjects and teeth do not achieve the proposed endpoints for periodontal stability, yet most periodontal patients preserve most of their teeth during an average of 10-13 years in SPC.

Advances in hydrogels for the treatment of periodontitis

Periodontitis is the second most prevalent oral disease and can cause serious harm to human health. Hydrogels are excellent biomaterials that can be used for periodontitis as drug delivery platforms to achieve inflammation control through high drug delivery efficiency and sustained drug release and as tissue scaffolds to achieve tissue remodelling through encapsulated cell wrapping and effective mass transfer. In this review, we summarize the latest advances in the treatment of periodontitis with hydrogels. The pathogenic mechanisms of periodontitis are introduced first, followed by the recent progress of hydrogels in controlling inflammation and tissue reconstruction, in which the specific performance of hydrogels is discussed in detail. Finally, the challenges and limitations of hydrogels for clinical applications in periodontitis are discussed and possible directions for development are proposed. This review aims to provide a reference for the design and fabrication of hydrogels for the treatment of periodontitis.