Aim: Comparison of the clinical efficacy (digitally volumetric, aesthetic, patient‐centred outcomes) of tunnel technique (TUN) with subepithelial connective tissue graft (CTG) versus coronally advanced flap (CAF) with enamel matrix derivate (EMD) 5 years after gingival recession therapy.Materials and methods: In 18 patients contributing 36 RT1 recessions, study models were collected at baseline and follow‐ups. Optical scans assessed recessions computer‐assisted [recession depth, recession reduction (RECred), complete root coverage (CRC), percentage of root coverage (RC), pointwise (pTHK) and mean areal (aTHK) marginal soft tissue thickness]. Root coverage aesthetic Score (RES) was used for aesthetic evaluation and visual analogue scales for patient‐centred data collection applied.Results: Sixty months after surgery, 50.0% (TUN+CTG) and 0.0% (CAF+EMD) of sites showed CRC (p = 0.0118), 82.2% (TUN+CTG) and 32.0% (CAF+EMD) achieved RC, respectively (p = 0.0023). CTG achieved significantly better RECred (TUN+CTG: 1.75±0.74 mm; CAF+EMD: 0.50 ± 0.39 mm; p = 0.0009) and aTHK (TUN+CTG: 0.95 ± 0.41 mm; CAF+EMD: 0.26 ± 0.28 mm; p = 0.0013). RES showed superior outcomes (p = 0.0533) for TUN+CTG (6.86 ± 2.31) compared to CAF+EMD (4.63 ± 1.99). The study failed to find significant differences related to patient‐centred outcomes (TUN+CTG: 8.30 ± 2.21; CAF+EMD: 7.50 ± 1.51; p = 0.1136).Conclusions: Five years after treatment, CTG resulted in better clinical and aesthetic outcomes than CAF+EMD. Increased THK was associated with improved outcomes for RECred and RC.

Aim: To develop and validate models based on logistic regression and artificial intelligence for prognostic prediction of molar survival in periodontally affected patients.Materials and Methods: Clinical and radiographic data from four different centres across four continents (two in Europe, one in the United States, and one in China) including 515 patients and 3157 molars were collected and used to train and test different types of machine‐learning algorithms for their prognostic ability of molar loss over 10 years. The following models were trained: logistic regression, support vector machine, K‐nearest neighbours, decision tree, random forest, artificial neural network, gradient boosting, and naive Bayes. In addition, different models were aggregated by means of the ensembled stacking method. The primary outcome of the study was related to the prediction of overall molar loss (MLO) in patients after active periodontal treatment.Results: The general performance in the external validation settings (aggregating three cohorts) revealed that the ensembled model, which combined neural network and logistic regression, showed the best performance among the different models for the prediction of MLO with an area under the curve (AUC) = 0.726. The neural network model showed the best AUC of 0.724 for the prediction of periodontitis‐related molar loss. In addition, the ensembled model showed the best calibration performance.Conclusions: Through a multi‐centre collaboration, both prognostic models for the prediction of molar loss were developed and externally validated. The ensembled model showed the best performance in terms of both discrimination and validation, and it is made freely available to clinicians for widespread use in clinical practice.

Background and Objective: Long‐term tooth retention is the ultimate goal of periodontal therapy. Aim of this study was to evaluate tooth loss (TL) during 10 years of supportive periodontal therapy (SPT) in periodontal compromised patients and to identify factors influencing TL on patient level.Material and Methods: Patients were re‐examined 120 ± 12 months after active periodontal therapy. TL and risk factors [smoking, initial diagnosis, SPT adherence, interleukin‐1 polymorphism, cardiovascular diseases, age at baseline, bleeding on probing (BOP), change of practitioner, insurance status, number of SPT, marital and educational status] influencing TL on patient level were assessed.Results: One‐hundred patients (52 female, mean age 65.6 ± 11 years) lost 121 of 2428 teeth (1.21 teeth/patient; 0.12 teeth/patient/y) during 10 years of SPT. Forty‐two of these were lost for periodontal reasons (0.42 teeth/patient; 0.04 teeth/patient/y). Significantly more teeth were lost due to other reasons (P < .001). Smoking, baseline severity of periodontitis, non‐adherent SPT, positive interleukin‐1 polymorphism, marital and educational status, private insurance, older age at baseline and BOP, small number of SPT were identified as patient‐related risk factors for TL (P < .05).Conclusion: During 120 ± 12 months of SPT, only a small number of teeth was lost in periodontally compromised patients showing the positive effect of a well‐established periodontal treatment concept. The remaining risk for TL should be considered using risk‐adopted SPT allocation.