Patient population
A total of 151 AgP patients were enrolled in the periodontal Department of Peking University School and Hospital of Stomatology between 2001 and 2015. The diagnosis of AgP was based on the following clinical and radiographic criteria proposed by the 1999 International World Workshop for a Classification of Periodontal Diseases and Conditions(Lang N, 1999).
The inclusion criteria were as follows:
1) ≤35 years old;
2) at least 20 remaining teeth in the mouth;
3) At least 6 teeth have a probing depth ≥5 mm and attachment loss of ≥3 mm on adjacent surfaces;
4) Non-smokers.
These patients were diagnosed with stage III/IV grade C periodontitis according to the 2018 classification.
Before diagnosis and treatment, full-mouth periapical radiographs of all subjects were taken by radiology technicians. The study was approved by the Ethics Committee of PKUSS (approval number: PKUSSIIT02305).
Clinical examinations
All clinical data, including the following indicators, were obtained through repeated measurements by two experienced clinical physicians (Li Xu and Li Zhang): probing depth (PD), bleeding index (BI), and tooth mobility (TM). The average PD (APD) was determined by calculating the mean value of six measurement points for PD. Similarly, the average BI (ABI) was obtained by averaging measurements taken on both the buccal and lingual sides for the BI. The degree of tooth mobility is assessed using handheld dental examination forceps, and it is determined based on the direction and magnitude: if there is solely buccolingual movement, it is classified as degree 1; if there is movement in both the buccolingual and mesiodistal directions, it is classified as degree 2; and if vertical mobility occurs, it is classified as degree 3(Newman et al., 2015). Furthermore, considering the magnitude of tooth displacement, adjustments are made according to the principle of severity: tooth displacement less than 1 mm corresponds to degree 1, tooth displacement between 1-2 mm corresponds to degree 2, and tooth displacement exceeding 2 mm corresponds to degree 3(Lindhe et al., 2015).
Radiographic analyses
All periapical radiographs were confirmed to have an appropriate exposure time and suitable projection angle by experienced radiological technologists and clinical doctors, and the inclusion criteria for periapical radiographs were as follows:
(1) Radiographs should be accurately captured, ensuring precise positioning of the teeth at the center of the image and complete visibility of the entire tooth structure within the image boundaries;
(2) The image exhibits distinct clarity, with evident contrast in the photograph;
(3) The projection angle is appropriate, with the vertical angle accurately aligned and the length of the teeth closely approximating their actual measurements;
(4) The interproximal structure of the teeth appears well defined and intact, exhibiting correct horizontal angles, no apparent overlap with adjacent structures, and an absence of caries, fillings, restorations or rotations;
(5) No abnormalities were detected in the periodontal ligament space (widening or disappearance), periodontal-pulp complex lesions, or residual roots.
All included radiographic images were scanned for digital documentation (UMAX Powerlook1000 manual control, 600 dpi) and measured using GeoGebra (Classic 5.0.735.0-d, International GeoGebra Institute, Linz, Austria) by one researcher (Jia-Ming Li).
Measurement of periapical radiographs
1. Confirm point: Set point on the mesial and distal marginal ridge of crown (A, B); enamel-dentin junctions (C, D) situated mesial and distal; root apex point (G); and alveolar crest points (E, F) positioned mesial and distal to the target teeth. The reference for alveolar ridge point confirmation is based on the research conducted by Schulte et al (Schulte et al., 1992).
2. Confirm the tooth axis: Researchers set the tooth axis by dividing the crown into two parts, based on the area below the marginal ridge of the crown and approximately two-thirds of the root crown surface, both mesial and distal. The varying degrees of curvature of the apical third can be disregarded when determining the tooth axis; thus, it is not necessary for the apical point to intersect with or pass through the tooth axis.
3. Measure crown-to-root ratio (CRR): Connect the mesial and distal central CEJs (C, D), and draw parallel lines passing through the mesial and distal marginal ridge points (A, B), respectively, intersecting the long axis of the tooth at two points (H, I). The midpoint between these two points was selected as the crown point (J). A line parallel to the CEJ passing through the root apex point (G), intersecting the long axis of the tooth at one point considered a hypothetical root apex point (K), was drawn. The JL along the CEJ was measured as the crown length from J to K, which represents the root length. The ratio of crown length to root length is defined as JL/KL.
4. The parameter of root width (PRW) was measured according to the methodology proposed by Xu Li et al(Xu et al., 2009). According to the research of Liu et al., root width can be categorized into two types, namely, normal and cone roots, based on a threshold ratio of 0.37 between the surface area of the root collar and that below the ridge top(Liu et al., 2022);
5. Measure the extent of alveolar bone resorption: Construct parallel lines to the CEJ passing through points E and F at varying distances from the alveolar crest. These lines intersect the long axis of the tooth at two distinct points (R and Q). The linear distance between these points (RK) and the hypothetical root apex (QK) was measured. The average bone loss ratio (ABLR) was 1-.
6. Bone level crown-root ratio (B-CRR): B-CRR= (CRR+ ABLR)/(1- ABLR). This indicator can be utilized to estimate the remaining volume of periodontal membrane tissue and the lever arm magnitude during tooth loading, thereby partially reflecting the resistance torque generated by teeth when subjected to external forces.
The measurement method used is depicted in Fig. 1.
Statistical methods
The statistical significance tests in this study were conducted using two-tailed tests, with a criterion of P<0.05 to determine significance. Different statistical analysis methods were chosen based on the type of data (categorical or continuous). The statistical analysis was performed using R 4.2.3 software [GUI 1.79 (8198 High Sierra build), S. Urbanek & H.-J. Bibiko, © R Foundation for Statistical Computing, 2021]. The analysis methods included t tests, chi-square tests, nonparametric tests, and logistic regression analysis.
For continuous data, normality testing was initially conducted to assess distribution characteristics. Normally distributed continuous data are presented as the mean ± standard deviation (x̄±s), while nonnormally distributed continuous data are presented as the median and interquartile range [M(P_25~P_75)]. Categorical data are expressed as rates or proportions.
Paired t tests or analysis of variance (ANOVA) methods were employed for normally distributed continuous data; however, for nonnormally distributed variables such as the crown–root ratio and reference values for root width, either transformation techniques to achieve a normal distribution or nonparametric tests were utilized instead. Chi-square tests or similar approaches were used for categorical data.
Multiple factor analysis was performed to identify factors influencing tooth mobility. Tooth mobility served as the dependent variable (with control=I), which had more than two levels and passed the parallelism test with a result of P<0.05; therefore, an unordered multinomial logistic regression analysis was applied.
A backwards selection method was implemented to select variables and establish the optimal model that provides odds ratios (ORs) along with their corresponding 95% confidence intervals for relevant variables.