Identifying prognostic factors for tooth survival in endo-perio lesions

Retrospective cohort study conducted at the Periodontal Clinic of Tokyo Medical and Dental University Hospital (April 2015–March 2021).

Inclusion criteria:

• Age ≥20 years.
• Teeth diagnosed with EPL, presenting with a deep periodontal pocket extending to or approaching the root apex with corresponding radiographic alveolar bone loss.
• Baseline probing depth (PD) ≥5mm.
• Minimum follow-up period of six months after root-canal filling (RCF).

Exclusion criteria:

• Smoking.
• Unavailable baseline periodontal records or radiographs.
• Transplanted teeth.
• Teeth with suspected root damage (cracks, fractures, or external root resorption).

EPL classification: EPLs were classified in patients with or without periodontitis and graded (1–3) based on the number and extent of deep periodontal pockets, according to the 2017 World Workshop Classification.

Radiographic assessment:

• Bone loss was calculated as the mean of mesial and distal measurements and expressed relative to root length.
• Periapical radiographs were evaluated by two calibrated periodontists using the Periapical Index (PAI) and the I–J–U subclassification:
  •  I, unilateral intrabony defect without periapical radiolucency.
  • J, unilateral intrabony defect with periapical radiolucency.
  • U, bilateral intrabony defects with periapical radiolucency (κ = 0.83).

Baseline clinical parameters: Probing depth, bleeding on probing, tooth mobility, furcation involvement, periodontal diagnosis, and history of endodontic treatment were evaluated.

Treatment:

• Endodontic treatment was performed first.
• Non-surgical periodontal therapy was initiated after about three months.
• Following re-evaluation, periodontal surgery was performed when indicated.
• Supportive periodontal care was provided every three to six months after active therapy.

Statistical analysis:

• Tooth survival was assessed using Kaplan–Meier analysis.
• Multivariable Cox proportional hazards regression was used to identify prognostic factors for tooth survival.
• Predictive accuracy was evaluated using time-dependent receiver operating characteristic (ROC) curves.
• Statistical significance was set at p < 0.05.

 

A total of 187 teeth with EPL lesions were included in the analysis, with a mean observation time of 3.1 ± 2.1 years (maximum 8.4 years).

Baseline characteristics:

• The average PD at the deepest site of EPL teeth was 8.7 ± 2.2mm. Of EPL-affected teeth, 80.7% had not been endodontically treated before. Among included teeth, 34 (18.2%) were I-shaped, 104 (55.6%) were J-shaped, and 49 (26.2%) were U-shaped.
• Seventy of the 187 teeth were extracted during the observation time. The estimated one- and five-year tooth survival rates were respectively 83.8% and 58.9%. The main reason for tooth extraction was disease progression (97.1%).
• Tooth survival rate: EPL grade (p = 0.006), PAI (p = 0.001), PD (p < 0.001), tooth mobility (p < 0.001), and I–J–U subclassification (p < 0.001) were significantly associated with the tooth survival rate in the stratified Kaplan Meier analysis.

Radiographic bone loss morphology:
 

• The estimated five-year survival rates for I-, J-, and U-shaped lesions were respectively 93.0%, 72.3%, and 13.4% (p < 0.001).
• J-shape (HR 4.44; 95% CI: 1.02–19.31; p = 0.047) and U-shape (HR, 28.06; 95% CI: 6.03–130.51; p < 0.001) were associated with a significantly higher risk of extraction compared with I-shape in the multivariate Cox regression model including the I–J–U classification.
• The model including the I–J–U subclassification showed a better fit (AIC 586.5 vs. 617.9) and a higher five-year predictive accuracy (AUC 0.85 vs. 0.77) compared with the model without it.