Abstract
Objective
To investigate the anatomical characteristics of the apical region of immature mandibular second premolars using CBCT.
Method
CBCT images of 100 mandibular second premolars at Nolla stage 8 and 100 at Nolla stage 9 were included. Using the CBCT built-in software, the buccolingual (A-B) and mesiodistal (C-D) diameters of the apical foramen, as well as the buccolingual (A1-B1), mesiodistal (C1-D1), and superoinferior (E-F) diameters of the physiologic apical radiolucency, were measured. Data were analyzed using SPSS 26.0.
Results
The values for Nolla stage 8 A-B, C-D, A1-B1, C1-D1, and E-F were 3.32 ± 0.38 mm, 2.45 ± 0.28 mm, 4.28 ± 0.43 mm, 3.29 ± 0.40 mm, and 3.04 ± 0.43 mm, respectively. The values for Nolla stage 9 were 2.24 ± 0.44 mm, 1.37 ± 0.23 mm, 3.11 ± 0.52 mm, 2.20 ± 0.29 mm, and 1.85 ± 0.30 mm respectively. The differences were all significant (P < 0.05). In the Nolla stage 8 group, the values of A-B, C-D, A1-B1, C1-D1, and E-F were significantly lower in females than in males (all P < 0.05). At Nolla stage 9, males had slightly higher values than females, the sex difference in C-D was statistically significant (P < 0.05). The values of A-B and C-D were significantly positively correlated with those of A1-B1, C1-D1, and E-F (P < 0.001). In Nolla stage 8, the apical foramen was predominantly flared (85%), thick lamina dura accounted for 62%, lamina dura density was mainly heterogeneous (28%) or moderate (53%), and the lamina dura was mainly rough (67%) or mildly rough (25%). In contrast, Nolla stage 9 predominantly showed parallel apical foramen (91%), thin lamina dura (84%), homogeneous lamina dura density (78%), and smooth lamina dura (72%).
Conclusion
The apical foramen and physiologic apical radiolucency of the second premolar in the mandible of Nolla stage 9 are smaller than those in Nolla stage 8. During these stages, the apical foramen transforms from a trumpet-shaped to a parallel shape, and the lamina dura becomes more mature.
Clinical significance
This study provides normative reference data on the apical anatomy of immature mandibular second premolars, which may serve as a baseline for future diagnostic accuracy studies or for comparison with pathological conditions in the same tooth type.
Supplementary Information
The online version contains supplementary material available at https://doi.org/10.1186/s12903-026-09221-6.
Keywords: Cone-beam computed tomography, Mandibular second premolar, Apical foramen, Physiologic apical radiolucency, Measurement
Introduction
Tooth development is a continuous and complex process. In immature permanent teeth, the developing root comprises three structures: the dental papilla, Hertwig’s epithelial root sheath (HERS), and the dental follicle [1, 2], collectively referred to as the developing root apex complex (DRAC) [3]. The mandibular second premolar typically erupts between 10 and 12 years of age [4]. At this time, the DRAC has not yet fully differentiated into periodontal tissues, on radiographic images, the corresponding area appears as a physiologic apical radiolucency. Root development takes approximately 2–3 years to complete [4, 5]. According to the Nolla staging system [6], tooth development is divided into 10 stages. During the window of 10–13 years, the mandibular second premolar is usually at Nolla stage 8 or 9. Nolla stage 8: approximately two-thirds of the root length has formed relative to the crown height, the root canal is relatively wide, and the apex appears divergent. Nolla stage 9: root formation is nearly complete, but the apex remains open. These teeth, with incompletely developed morphology and structure, are clinically defined as immature permanent teeth. Distinguishing between these two stages has important clinical significance because such teeth lack a natural apical constriction and may require specialized endodontic treatment protocols.
The anatomical characteristics of the apical region directly affect both diagnostic accuracy and treatment outcomes in immature teeth. When periapical lesions occur in an immature mandibular second premolar, damage to the DRAC can lead to arrested root development, with the apical foramen taking on a flared appearance, and the lamina dura may be partially or completely destroyed. Compared with the physiologic apical radiolucency and normal lamina dura morphology at the corresponding developmental stage, the radiographic image formed by periapical periodontitis shows characteristic differences [7]. Furthermore, the diameter and shape of the apical foramen are key factors influencing the accuracy of root canal working length measurement [8]. The larger the apical foramen, the more significant the measurement error of electronic apex locators becomes; measurement errors for flared or open apices are higher than those for parallel or convergent apices [9]. Thus, establishing normative data on apical foramen diameter and physiologic radiolucency size for each developmental stage is essential for accurate periapical diagnosis and reliable endodontic procedures.
Several radiographic modalities have been used to assess tooth development and periapical radiolucency in immature teeth. Panoramic radiography is commonly employed for developmental staging due to its low radiation dose and wide coverage; studies have reported its utility in evaluating Nolla stages of mandibular premolars [10, 11]. However, panoramic images are limited by magnification, distortion, and superimposition of anatomical structures, which compromise accurate measurement of the apical foramen and periapical radiolucency. In contrast, cone-beam computed tomography (CBCT) provides three-dimensional, high-resolution images without superimposition, allowing precise assessment of apical morphology and periapical status. Previous studies have used CBCT to evaluate root development and periapical radiolucency in immature teeth [12, 13], demonstrating its superiority over two-dimensional modalities for quantitative measurements. Nevertheless, the use of CBCT in children must be justified carefully, following the ALARA (As Low As Reasonably Achievable) and ALADA (As Low As Diagnostically Acceptable) principles [14]. A detailed discussion of the risk-benefit balance is provided in the Discussion section.
From an epidemiologic perspective, the need for normative data on Chinese children is particularly pressing. According to the Fourth National Oral Health Epidemiological Survey, the caries prevalence of permanent teeth among 12-year-old children in China is 34.5% [15]. Central cusp deformity is a major predisposing factor for caries and subsequent pulpitis in mandibular second premolars [16]; in children aged 10–13 years (Nolla stages 8–9), this extra occlusal bump is highly susceptible to wear or fracture, which can lead to caries and potentially progress to pulpitis or periapical periodontitis [16]. In comparison, the mandibular first premolar has a caries prevalence rate of 2.35%–3.92%, whereas the mandibular second premolar has a prevalence of 6.81%–9.74%, which is approximately two to three times higher [15, 17]. Therefore, mastering the imaging characteristics of the normal immature mandibular second premolar apex is the basis for correct early clinical diagnosis of periapical periodontitis and evaluation of treatment effects.
According to a PubMed search of the relevant literature, most currently available datasets on anatomical measurements of immature teeth are based on mixed dentition or non-Chinese populations. To date, no systematic study has specifically focused on the normal radiographic anatomical characteristics of the apical region of the mandibular second premolar at Nolla stages 8 and 9 in Chinese children. In the present study, we used CBCT with integrated NNT software to measure apical anatomy at these two developmental stages. Subgroup analyses were performed according to sex and side, and the sizes of the apical foramen and physiologic apical radiolucency were evaluated. These measurements provide normative reference data that can be used in future diagnostic accuracy studies or AI training models. Specifically, we tested two hypotheses: (1) within the same Nolla stage, sex and side do not influence apical measurements; and (2) significant differences exist between Nolla stages 8 and 9, thereby providing a quantitative reference for developmental assessment.
Participants and methods
Study participants
This study was carried out after being approved by the Ethics Committee of Jinan Stomatological Hospital. (Approval number: JNSKQYY-2025-020). The image data of CBCT subjects examined at Jinan Stomatological Hospital from June 2025 to December 2025 were selected. According to the Nolla tooth development stage criterion, the mandibular second premolars at Nolla stage 8 and Nolla stage 9 were analyzed.
Inclusion criteria
(1) Aged 10 to 13 years, which coincides with the peak eruption period of the mandibular second premolar, when the apical foramen remains unclosed in most cases; (2) Mandibular second premolars at Nolla stage 8 or stage 9 according to the Nolla system (comprises 10 stages); Specifically, Nolla stage 8: approximately two-thirds of the root is formed, with a wide root canal and diverging root ends; Nolla stage 9: root formation is nearly complete, but the apex remains open. (3) Normal eruption; (4) Normal maxillomandibular occlusal relationship: all included subjects presented with Class I molar relationships or mild spacing without significant malocclusion; (5) Clear images with a voxel size of 0.125 mm, free of motion artifacts and metal artifacts, meeting the requirements for measurement and evaluation.
Exclusion criteria
(1) Developmental abnormalities such as microdontia, fused teeth, geminated teeth, dilacerated teeth, or double root canals; (2) The presence of other lesions in the periapical region, such as alveolar bone sclerosis, bone islands, or bone defects; (3) Concurrent conditions affecting tooth development, such as developmental delay or rickets; (4) Unclear staging of Nolla stage 8 or Nolla stage 9.
This study is a cross-sectional observational study employing a two-group independent sample comparison design. With a two-tailed α level of 0.05, the test power (1 - β) is set at 0.80. According to the formula for calculating the sample size for comparing the means of two independent samples: n = 2(Z1−α/2 + Z1−β)²σ²/δ². To conduct the preliminary experiment, 20 samples at Nolla stage 8 and 20 samples at Nolla stage 9 were selected. In each stage, 10 teeth were normal, and 10 had non-carious dens evaginatus. Based on the measurement results, the mean difference (δ) between the two stages was set at 0.3 mm, the pooled standard deviation (σ) was 0.4 mm, Z₁₋α/₂ = 1.96, and Z₁₋β = 0.84. The measurement differences between teeth with dens evaginatus and normal teeth were not statistically significant. The required sample size was calculated to be approximately 28 cases per group. To ensure that the sample size met the requirements of this study, two research assistants (not involved in the subsequent assessments) screened 455 eligible patients from a database of 13,620 patients according to the inclusion and exclusion criteria. Using computer-based coding, a stratified sampling method was employed to select 100 mandibular second premolars at Nolla stage 8 and 100 at Nolla stage 9 for measurement and analysis (Fig. 1). The guardians of all the patients included in the study signed the informed consent form.
Fig. 1.
Study sample selection flowchart
Inspection and measurement methods
The examination was performed using a NEW TOM 5G Version FP CBCT scanner (QR SRL, Italy). The scanning parameters were: 110 kV, 8.34–16.41 mA, FOV: 10 cm × 10 cm, exposure time 3.6–5.4 s, and voxel size: 0.125 mm.
The relevant indicators of the scanned image data were measured using the NNT software that comes with the NEW TOM CBCT. The multiplanar reconstruction (MPR) function was used to adjust the horizontal line to be parallel to the occlusal plane, and the sagittal line to be parallel to the long axis of the teeth. Locate the maximum cross-sectional area of the physiologic apical radiolucency on the buccolingual section image. Measure the most apical buccolingual dimension of the apical foramen of the immature mandibular second premolar (A-B dimension: at the level of the apical foramen, from the lingual inner wall to the buccal inner wall of the apical foramen) and the buccolingual dimension of the largest cross-section of the physiologic apical radiolucency (A1-B1 dimension: at the maximal buccolingual horizontal plane, from the lingual inner wall to the buccal inner wall of the lamina dura) (Fig. 2a). In the mesiodistal section, locate the largest cross-section of the physiologic apical radiolucency, and measure the most apical mesiodistal dimension of the apical foramen (C-D dimension: at the level of the apical foramen, from the mesial inner wall to the distal inner wall of the apical foramen), the mesiodistal dimension of the largest cross-section of the periapical radiolucency (at the maximal mesiodistal horizontal plane, from the mesial inner wall to the distal inner wall of the lamina dura), and the superior-inferior dimension (E-F dimension: at the maximal vertical cross-section, from the coronal inner wall of the lamina dura to the level of the apical foramen) (Fig. 2b).
Fig. 2.
Image a: Buccolingual section, L = lingual, B = buccal, A-B: Buccolingual diameter of the apical foramen; A1-B1: Buccolingual diameter of the physiologic apical radiolucency; Image b: Mesiodistal section, M = mesial, D = distal, C-D: Mesiodistal diameter of the apical foramen; Mesiodistal diameter of the physiologic apical radiolucency; E-F: Superior-inferior diameter of the physiologic apical radiolucency
The apical canal morphology in the periapical region can be classified into two types: flared and parallel (Fig. 3). The flared apex shows an apical canal outline that diverges outward at the apex, whereas the parallel apex has a canal wall that runs vertically and remains parallel.
Fig. 3.
Nolla stage 8, buccolingual (a) and mesiodistal (b) views: The apical foramen has a flared appearance (white arrow). Nolla stage 9, buccolingual (c) and mesiodistal (d) views: The apical foramen is parallel-walled (white arrow)
Use the built-in linear measurement tool of the CBCT to select the thickest part of the lamina dura and measure its thickness (Fig. 4). Grading criteria: thin: ≤1 mm; thick: >1 mm.
Fig. 4.
Refer to the scale: image a: lamina dura thickness > 1 mm (white arrow); image b: lamina dura thickness 1 mm (white arrow)
Use the built-in grayscale measurement tool of the CBCT to measure the density of the lamina dura, and evaluate density uniformity by the difference between the maximum and minimum grayscale values (Fig. 5). Significantly non-uniform: difference ≥ 300 Hu; moderately non-uniform: 100–299 Hu; uniform: difference < 100 Hu.
Fig. 5.
Evaluate density uniformity by the difference between the maximum and minimum grayscale values, image a: significantly non-uniform, grayscale values difference > 300 Hu. Image b: moderately non-uniform; 100–299 Hu. Image c: uniform: difference < 100 Hu
Evaluate smoothness by the edge fluctuation amplitude (EFA) of the lamina dura contour, defined as the vertical distance between the high and low points on the outer surface within a 2 mm interval. Grading criteria (Fig. 6). significantly rough (EFA > 1 mm), mildly rough (EFA = 0.51–1 mm), smooth (EFA < 0.5 mm).
Fig. 6.
image a: lamina dura contour is significantly rough, image b: mildly rough, image c: smooth
All measurements and morphological assessments were independently performed by two trained oral and maxillofacial radiologists with 6 and 10 years of experience in dental medicine, respectively. All images were evaluated under identical conditions of brightness, contrast, and magnification (2×). Each parameter was measured twice, and the mean value was recorded in the statistical table. In cases where the two radiologists disagreed on any morphological assessment, the final decision was made by a third senior investigator from the research team (with 15 years of experience) as an adjudicator.
Statistical methods
Statistical analysis: Data analysis was conducted using the SPSS 26.0 software package. Continuous variables that followed a normal distribution were represented by (
), and independent-sample t-tests were employed. For comparisons among multiple subgroups, the Games-Howell test was employed for post-hoc multiple comparisons. Categorical variables were analyzed using the χ2 test. Correlation analysis was performed using Pearson correlation analysis, and differences were considered statistically significant when P < 0.05.The consistency test was performed using SPSSAU. The two researchers were blinded to each other’s results and to the Nolla stage. Repeated measurements (10 stage-8 and 10 stage-9 teeth) were conducted one week apart to minimize recall bias. For continuous variables, the Bland-Altman method and the intraclass correlation coefficient (ICC) were used. For categorical variables (thickness, morphology, density, and smoothness), inter- and intra-observer agreement were assessed using Cohen’s kappa (binary: apical foramen morphology, lamina dura thickness) or linear-weighted kappa (ordinal: density and smoothness of lamina dura).
Results
Sample characteristics
In this study, 100 mandibular second premolars of Nolla stage 8 and 100 mandibular second premolars of Nolla stage 9 were selected from the medical imaging system of Jinan Stomatological Hospital. All the patients were of Han ethnicity in China. There were 50 male and 50 female patients in each group. For each of the two stages (Nolla stage 8 and Nolla stage 9), 50 teeth on the left side and 50 teeth on the right side were included. There were no statistically significant differences in sex or side distribution between the two groups (P = 1.000), and they were comparable. The results showed that the average age of Nolla stage 9 was significantly higher than that of Nolla stage 8, the difference was statistically significant (P < 0.001). The stratified analysis revealed that the average age of females in Nolla stage 8 was significantly lower than that of males, and the difference was statistically significant (P < 0.001); the average age of females in Nolla stage 9 was slightly lower than that of males, but the difference was not statistically significant (P > 0.05). As shown in Table 1.
Table 1.
Comparison of ages in Nolla stage 8 and Nolla stage 9 [(years)]
| Characteristic | Total | Male | Female | 95% Confidence Interval | P value |
|---|---|---|---|---|---|
| Average value | 11.54 ± 0.89 | 11.71 ± 0.82 | 11.37 ± 0.93 | −1.136~−0.715 | 0.007 |
| Nolla stage 8 | 11.08 ± 0.63 | 11.31 ± 0.60 | 10.84 ± 0.57 | −0.711~−0.243 | 0.000 |
| Nolla stage 9 | 12.00 ± 0.86* | 12.10 ± 0.79* | 11.90 ± 0.92* | −0.343 ~ 0.343 | 0.246 |
* Compared with Nolla stage 8, P < 0.001
Standard consistency test
The Bland-Altman analysis showed that the mean difference in the measurement results between Researcher 1 and Researcher 2 was 0.033, indicating that there was no significant systematic bias in the measurement standards used by the two researchers. The 95% consistency limit is -0.514 to 0.580. 96% (96/100) of the data points (more than 95%) are distributed within this limit, and the scattered points are randomly distributed on both sides of the mean line, without showing any obvious proportion deviation. The ICC was 0.954, indicating a high degree of consistency and reliability between the two examiners (Fig. 7).
Fig. 7.
The Bland-Altman plots measured by Researcher 1 and Researcher 2: Bland-Altman plot illustrating the inter-examiner agreement. The solid line indicates the mean difference (− 0.03 mm), and the dashed lines represent the 95% limits of agreement (− 0.21 to 0.15 mm)
Excellent inter-researchers agreement was observed for all morphological parameters (kappa range: 0.875–0.959, all P < 0.001), with percentage agreements ranging from 92.0% to 98.0%. As shown in Table 2.
Table 2.
Inter-researchers agreement for morphological parameters
| Parameter | Agreement(%) | Kappa | 95% Confidence Interval | P-value |
|---|---|---|---|---|
| Apical foramen morphology | 96.0 | 0.875 | 0.761–0.989 | 0.000 |
| Lamina dura thickness | 98.0 | 0.959 | 0.917–1.000 | 0.000 |
| Lamina dura density* | 92.0 | 0.881 | 0.827–0.935 | 0.000 |
| Lamina dura smoothness* | 94.0 | 0.909 | 0.862–0.956 | 0.000 |
*Weighted kappa (linear weights) was used for the ordinal parameters (Density and Smoothness)
Measurement results
Multiple linear regression analyses were performed separately for Nolla stages 8 and 9 to assess the potential confounding effects of age and sex on the five root measurements (A-B, A1-B1, C-D, C1-D1, E-F). Within Nolla stage 8, age showed no significant effect on any measurement, with β coefficients ranging from − 0.009 (C-D) to 0.035 (E-F) and all P-values > 0.25 (minimum P = 0.252 for E-F). Sex also had no significant influence, with β coefficients from − 0.058 (A1-B1) to 0.064 (E-F) and all P-values > 0.35 (minimum P = 0.354 for E-F). Within Nolla stage 9, age was not significantly associated with any measurement (β: − 0.044 to 0.008, all P > 0.26; minimum P = 0.263 for E-F). Similarly, sex yielded no significant effects (β: − 0.047 to 0.036, all P > 0.46; minimum P = 0.461 for C-D and A-B). The results were consistent between left and right sides. Therefore, age and sex do not act as confounding factors when analyzing data within the same Nolla stage. As shown in Table 3.
Table 3.
Results of multiple linear regression analysis for the effect of age and sex
| Stage | Measurement | Age | Sex | ||
|---|---|---|---|---|---|
| β (95% Confidence Interval) | P-value | β (95% Confidence Interval) | P-value | ||
| Nolla stage 8 | A-B | 0.017 (-0.038 ~ 0.072) | 0.543 | -0.046 (-0.169 ~ 0.077) | 0.461 |
| A1-B1 | 0.028 (-0.035 ~ 0.091) | 0.382 | -0.058 (-0.193 ~ 0.077) | 0.398 | |
| C-D | -0.009 (-0.058 ~ 0.040) | 0.718 | 0.039 (-0.085 ~ 0.163) | 0.537 | |
| C1-D1 | 0.015 (-0.043 ~ 0.073) | 0.613 | -0.027 (-0.151 ~ 0.097) | 0.665 | |
| E-F | 0.035 (-0.025 ~ 0.095) | 0.252 | 0.064 (-0.072 ~ 0.200) | 0.354 | |
| Nolla stage 9 | A-B | -0.028 (-0.097 ~ 0.041) | 0.425 | 0.027 (-0.094 ~ 0.148) | 0.662 |
| A1-B1 | -0.019 (-0.088 ~ 0.050) | 0.587 | 0.012 (-0.116 ~ 0.140) | 0.854 | |
| C-D | -0.036 (-0.105 ~ 0.033) | 0.306 | -0.047 (-0.172 ~ 0.078) | 0.461 | |
| C1-D1 | 0.008 (-0.061 ~ 0.077) | 0.820 | -0.018 (-0.141 ~ 0.105) | 0.772 | |
| E-F | -0.044 (-0.121 ~ 0.033) | 0.263 | 0.036 (-0.092 ~ 0.164) | 0.583 | |
* β coefficients represent the change in measurement (mm) per 1-year increase in age, or for sex (male as reference, female = 1). Results shown are representative for side 1; side 2 showed nearly identical estimates. All P-values are non-significant (P > 0.05)
The data of each group all conform to the normal distribution. The statistical results show that the apical foramen and the physiologic apical radiolucencies in Nolla stage 9 were significantly smaller than those in Nolla stage 8. The differences in all indicators were statistically significant (P < 0.001), as shown in Table 4.
Table 4.
Comparison of Measurement Parameter between Nolla 8 and Nolla stage 9[(
)mm]
| Parameter | Nolla stage 8 | Nolla stage 9 | 95%(Confidence Interval) | P value |
|---|---|---|---|---|
| A-B | 3.32 ± 0.38 | 2.24 ± 0.44 | 0.966 ~ 1.194 | 0.000 |
| C-D | 2.45 ± 0.28 | 1.37 ± 0.23 | 1.008 ~ 1.152 | 0.000 |
| A1-B1 | 4.28 ± 0.43 | 3.11 ± 0.52 | 1.038 ~ 1.304 | 0.000 |
| C1-D1 | 3.29 ± 0.40 | 2.20 ± 0.29 | 0.990 ~ 1.186 | 0.000 |
| E-F | 3.04 ± 0.43 | 1.85 ± 0.30 | 1.091 ~ 1.297 | 0.000 |
Stratified analysis: The measurements for the apical foramen and physiologic apical radiolucency on the left and right sides are shown in Table 5. There were no statistically significant differences in these measurements between the left and right sides for either Nolla stage 8 or Nolla stage 9 (P > 0.05).
Table 5.
Comparison of measurement parameters between the left and right sides at Nolla stage 8 and Nolla stage 9 [(
) mm]
| Parameter | Nolla stage 8 | Nolla stage 9 | ||||||
|---|---|---|---|---|---|---|---|---|
| Right side | Left side | 95% (Confidence Interval) |
P value | Right side | Left side | 95% (Confidence Interval) |
P value | |
| A-B | 3.31 ± 0.40 | 3.32 ± 0.36 | -0.213 ~ 0.185 | 0.845 | 2.27 ± 0.49 | 2.21 ± 0.42 | -0.168 ~ 0.292 | 0.482 |
| C-D | 2.44 ± 0.29 | 2.47 ± 0.27 | -0.187 ~ 0.111 | 0.507 | 1.38 ± 0.24 | 1.37 ± 0.22 | -0.106 ~ 0.134 | 0.761 |
| A1-B1 | 4.28 ± 0.46 | 4.27 ± 0.41 | -0.215 ~ 0.239 | 0.890 | 3.12 ± 0.54 | 3.09 ± 0.50 | -0.243 ~ 0.303 | 0.775 |
| C1-D1 | 3.33 ± 0.42 | 3.25 ± 0.37 | -0.126 ~ 0.294 | 0.298 | 2.18 ± 0.32 | 2.22 ± 0.26 | -0.205 ~ 0.101 | 0.376 |
| E-F | 3.08 ± 0.49 | 3.00 ± 0.37 | -0.142 ~ 0.310 | 0.333 | 1.88 ± 0.34 | 1.80 ± 0.24 | -0.075 ~ 0.234 | 0.181 |
Sex-based comparison is shown in Table 6. At Nolla stage 8, all measured parameters were significantly smaller in females than in males (P < 0.05). At Nolla stage 9, females showed smaller apical foramen and physiologic apical radiolucency, with significant differences for C-D (P < 0.05).
Table 6.
Comparison of measurement parameters between sexes at Nolla stage 8 and Nolla stage 9 [(
) mm]
| Parameter | Nolla stage 8 | Nolla stage 9 | ||||||
|---|---|---|---|---|---|---|---|---|
| Males (n = 50) |
Females (n = 50) |
95% (Confidence Interval) | P value | Males (n = 50) |
Females (n = 50) |
95% (Confidence Interval) | P value | |
| A-B | 3.47 ± 0.33 | 3.16 ± 0.36 | -0.498~-0.138 | 0.000 | 2.33 ± 0.46 | 2.15 ± 0.43 | -0.399~0.0515 | 0.189 |
| C-D | 2.57 ± 0.27 | 2.34 ± 0.25 | -0.370~-0.098 | 0.001 | 1.44 ± 0.25 | 1.31 ± 0.18 | -0.245~-0.015 | 0.021 |
| A1-B1 | 4.51 ± 0.35 | 4.04 ± 0.37 | -0.655~-0.273 | 0.000 | 3.21 ± 0.48 | 3.00 ± 0.53 | − 0.477 ~ 0.057 | 0.176 |
| C1-D1 | 3.44 ± 0.35 | 3.14 ± 0.40 | -0.496~-0.105 | 0.000 | 2.26 ± 0.28 | 2.14 ± 0.30 | − 0.270~0.0030 | 0.165 |
| E-F | 3.18 ± 0.37 | 2.90 ± 0.45 | -0.495~-0.065 | 0.005 | 1.90 ± 0.31 | 1.79 ± 0.27 | -0.264 ~ 0.041 | 0.231 |
The Pearson correlation analysis is shown in Table 7. The size of the physiologic apical radiolucency is significantly correlated with the size of the apical foramen. The diameters of A1-B1, C1-D1, and E-F are significantly positively correlated with the diameter of A-B (P < 0.001); the diameters of A1-B1, C1-D1, and E-F are significantly positively correlated with the diameter of C-D (P < 0.001).
Table 7.
Pearson Correlation Analysis of the Apical Foramen and physiologic apical radiolucency
| Variable | A1-B1 | C1-D1 | E-F | |
|---|---|---|---|---|
| A-B | r value | 0.965 | 0.848 | 0.724 |
| P value | 0.000 | 0.000 | 0.000 | |
| C-D | r value | 0.879 | 0.928 | 0.820 |
| P value | 0.000 | 0.000 | 0.000 |
Anatomical morphology of the root apex region
Apical foramen morphology: the flared type predominated in Nolla stage 8 (85%), whereas the parallel type was more common in Nolla stage 9 (91%). Regarding lamina dura thickness, the proportion of thick walls was 62% in Nolla stage 8 and decreased to 16% in Nolla stage 9, while thin walls increased from 38% to 84%. For lamina dura density, Nolla stage 8 was mainly characterized by heterogeneous (28%) and moderate density (53%), with homogeneous density accounting for only 19%; in contrast, Nolla stage 9 showed a predominance of homogeneous density (78%), while heterogeneous (7%) and moderate (15%) densities were much less common. As for lamina dura smoothness, Nolla stage 8 was primarily characterized by rough (67%) and mildly rough surfaces (25%), with only 8% showing smooth surfaces; in Nolla stage 9, smooth surfaces predominated (72%), whereas rough (11%) and mildly rough surfaces (17%) were less frequent. All observed variables were statistically significant (P < 0.001), as shown in Table 8.
Table 8.
Comparison of apical region anatomical morphology of the mandibular second premolar between Nolla stage 8 and Nolla stage 9 (%)
| Parameter | Nolla8(n = 100) | Nolla9(n = 100) | χ2值 | P值 | |
|---|---|---|---|---|---|
| Apical foramen | Flared | 85 | 9 | 115.937 | 0.000 |
| Parallel | 15 | 91 | |||
| Lamina dura thickness | Thick | 62 | 16 | 44.472 | 0.000 |
| Thin | 38 | 84 | |||
| Lamina dura density | Heterogeneous | 28 | 7 | 66.722 | 0.000 |
| Moderate | 53 | 15 | |||
| Homogeneous | 19 | 78 | |||
| Lamina dura smoothness | Rough | 67 | 11 | 92.929 | 0.000 |
| Mildly rough | 25 | 17 | |||
| Smooth | 8 | 72 |
Discussion
According to the guidelines of the European Commission [18], CBCT is not recommended as a standard method for diagnosing periapical lesions. We fully acknowledge that CBCT should not be routinely used solely for developmental assessment in children, owing to concerns regarding radiation exposure. In the present study, all CBCT scans were obtained for legitimate clinical indications, rather than for research purposes alone. Informed consent was obtained from each patient’s guardian, and the study protocol received approval from the institutional ethics committee. Nevertheless, CBCT is indicated when conventional radiography shows no pathological changes despite contradictory positive clinical signs, when the diagnosis remains uncertain, or in cases involving complex dental and pulpal disorders [19]. The low-dose and high-resolution advantages of CBCT have been widely recognized. Compared with periapical radiographs, CBCT provides three-dimensional anatomical information without overlapping interference and can clearly delineate fine structures such as the apical foramen, apical constriction, and anatomical root apex. It enables accurate and reproducible linear as well as three-dimensional quantitative measurements, substantially improving the precision of assessing apical morphological parameters, the extent of periapical lesions, and the distance between the root apex and adjacent critical anatomical structures [20, 21]. Therefore, while we strongly emphasize that CBCT should not be used as a screening tool for developmental assessment, its use in our study was justified by the clinical need for detailed anatomical evaluation in patients with existing indications. The reference values reported here are intended solely for the interpretation of CBCT images already obtained for valid clinical reasons and not to encourage unnecessary imaging.
Based on the reports of Wang [7] and Takeshita et al. [22], this study applied CBCT MPR technology to reliably measure the size of the apical foramen and the periapical radiolucent area of the mandibular second premolar, and described the imaging characteristics of the apical foramen and the surrounding lamina dura. By employing a single-ethnicity Han Chinese cohort, the present study minimizes inter-ethnic variability, which may offer a more reliable baseline reference for East Asian adolescents compared with mixed-population data.
Kim et al. [23] reported that females generally show earlier and faster apical foramen closure than males. Valerie [24] reported that the effect of sex on tooth development is greater than that of race/genetic background, with females developing earlier. This study did not directly measure the rate of closure, the results showed that the mean age of individuals at Nolla stage 9 was older than that of those at Nolla stage 8. Stratified comparative analysis revealed that at Nolla stage 8, the mean age of females was significantly younger than that of males, indicating that our findings are consistent with the trend reported by Kim and Valerie et al. [23, 24]. To address the need for normative reference data in this research field, the dimensional reference values derived from our measurements are explicitly presented below. For the five indices (A-B, C-D, A1-B1, C1-D1, and E-F) at Nolla stage 8, the reference ranges are (3.32 ± 0.38) mm, (2.45 ± 0.28) mm, (4.28 ± 0.43) mm, (3.29 ± 0.40) mm, and (3.04 ± 0.43) mm, respectively. At Nolla stage 9, the corresponding reference values are (2.24 ± 0.44) mm, (1.37 ± 0.23) mm, (3.11 ± 0.52) mm, (2.20 ± 0.29) mm, and (1.85 ± 0.30) mm, respectively. The apical foramen and physiologic apical radiolucency at Nolla stage 9 were significantly smaller than those at stage 8. Furthermore, at the Nolla stage 8, male values were (3.47 ± 0.33) mm, (2.57 ± 0.27) mm, (4.51 ± 0.35) mm, (3.44 ± 0.35) mm, and (3.18 ± 0.37) mm for A-B, C-D, A1-B1, C1-D1, and E-F, respectively, while female values were (3.16 ± 0.36) mm, (2.34 ± 0.25) mm, (4.04 ± 0.37) mm, (3.14 ± 0.40) mm, and (2.90 ± 0.45) mm, respectively. Females exhibited significantly smaller apical dimensions than males at this stage (all P < 0.05). At Nolla stage 9, however, only the C-D index showed a statistically significant difference between sexes ((1.44 ± 0.25) mm vs. (1.31 ± 0.18) mm, P = 0.021), suggesting a sex-dependent developmental tempo rather than a permanent morphological difference.
Zhang [25] and Ong et al. [26] reported in studies on populations from different regions that tooth maturity is correlated with physical development. The fact that permanent tooth development in females generally precedes that in males may explain why females have smaller apical foramen and physiologic apical radiolucency areas. This finding is biologically plausible, as female skeletal and dental maturation typically precedes that of males by approximately 2.5–14 months during early adolescence [27, 28]. Wang et al. [7]. measured the apical anatomical morphology of immature maxillary central incisors and reported that at Nolla stage 9, the dimensions of both the apical foramen and the physiologic apical radiolucency were significantly smaller than those at Nolla stage 8. Female patients also exhibited smaller diameters than males, although this sex difference diminished at Nolla stage 9. A significant correlation was found between the size of the apical foramen and the size of the physiologic apical radiolucency. The results of the present study are generally consistent with the anatomical measurements of the apical region of maxillary central incisors reported by Wang [7], suggesting that the physiologic apical radiolucency gradually decreases in size as the apical foramen closes. Consequently, at the same chronological age and Nolla stage, females are likely further along the root-closure continuum, resulting in a narrower apical foramen and smaller radiolucent area. The disappearance of sex differences by Nolla stage 9 indicates that the apical anatomy converges as root development nears completion, a pattern consistent with the “catch-up” phenomenon observed in other craniofacial growth parameters [7, 27]. In addition, Leung et al. [29] reported that among 12-year-old children in southern China, females had smaller tooth sizes than males, with the exception of the left maxillary and mandibular lateral incisors and the bilateral mandibular central incisors. Ozge [30] measured 4,800 teeth in a Turkish population and found that male tooth dimensions were larger than female tooth dimensions, with the buccolingual diameter showing stronger sexual dimorphism than the mesiodistal diameter. This suggests that the smaller apical foramen and physiologic apical radiolucency in females may also be related to the characteristic of smaller tooth volume in females compared to males.
Nelson [4] pointed out that in the early stage, the localized radiolucency in the apical region has no lamina dura, and when the periodontal tissues are fully developed, this physiologic apical radiolucency disappears accordingly. Wang et al. [7]. reported that the apical region of maxillary central incisors at Nolla stage 8 was more likely to have a broad and blurred lamina dura, and that with the development of the apical foramen, the lamina dura in the apical region tended to become clear and sharp. The present study provides quantitative morphological evidence further characterizing this process. At Nolla stage 8, the apical foramen was predominantly flared (85%), indicating an open apex, whereas by Nolla stage 9 (root near completion, apex still open), the foramen was predominantly parallel (91%), suggesting final shaping of the root apex. Concurrently, the lamina dura in the apical region exhibited significant dynamic changes. To facilitate interpretation of lamina dura morphology, we previously established quantitative thresholds (see Methods): thin thickness ≤ 1 mm; density non-uniformity defined as a grayscale difference ≥ 300 HU; and significant roughness defined as EFA > 1 mm. These thresholds allowed consistent comparisons between groups. At Nolla stage 8, the lamina dura was characterized by thick walls (62%), rough surfaces (67%), and heterogeneous (28%) or moderate (53%) density, surrounding the physiologic apical radiolucency with an irregular, ill-defined margin; at Nolla stage 9, the lamina dura shifted to thin walls (84%), smooth surfaces (72%), and homogeneous density (78%), with a sharp, clear margin. These findings indicate that as the root apex transitions from a flared to a parallel morphology, the apical lamina dura undergoes a dynamic remodeling process from thick to thin, rough to smooth, and heterogeneous to homogeneous. By Nolla stage 9, the lamina dura in the apical region approaches morphological maturity. However, the interactive mechanisms between root apex morphogenesis and periapical lamina dura remodeling require further investigation.
From a clinical perspective, under the conditions of this study, significant statistical changes were observed in apical foramen morphology, lamina dura thickness, smoothness, and density during the transition from Nolla stage 8 to stage 9, confirming that apical closure is a quantifiable and active developmental event. This transitional period (approximately 10–12 years of age) can be considered an important window for clinical monitoring of the mandibular second premolar. The reference values established in this study may serve as a benchmark for detecting developmental deviations, such as asymmetric closure or persistently enlarged apical foramen. Regarding female patients at Nolla stage 8, it should be noted that a relatively smaller apical width may fall within the normal range, and misinterpretation as pathological narrowing or premature closure should be avoided. Furthermore, quantitative assessment of the lamina dura based on grayscale uniformity, thickness, and marginal fluctuation amplitude may help differentiate normal physiologic apical radiolucency from early periapical pathology. When the lamina dura appears uniform, smooth, and thin (characteristic of Nolla stage 9), it suggests a higher likelihood of physiological development.
Importantly, the present study does not claim diagnostic accuracy for pathological conditions. Rather, it establishes physiological reference ranges for the most vulnerable developmental stages of a high-risk tooth. Future longitudinal studies are needed to determine whether individuals with measurements at the extremes of these ranges are at increased risk of developing apical disease. Until then, our data should be used cautiously as an adjunct to clinical judgment, not as a standalone diagnostic test.
Conclusion
This study described the CBCT morphological characteristics of the apical foramen and lamina dura of immature mandibular second premolars at Nolla stages 8 and 9, and measured the anatomical parameters of the apical region, which may serve as a baseline for future diagnostic accuracy studies or for comparison with pathological conditions in the same tooth type. As this was a single-center study, there may be limitations due to sampling bias. Therefore, these findings should be generalized with caution to other racial groups or populations with different tooth development patterns. Large-scale, multicenter, prospective cohort studies are recommended to obtain more precise data.
Supplementary Information
Acknowledgements
The authors would like to thank Central Laboratory, Jinan Key Laboratory of Oral Tissue Regeneration, Jinan Stomatological Hospital, Jinan, China, and the oral and maxillofacial radiologists who performed the analysis for their expertise.
Abbreviations
- HERS
Hertwig’s Epithelial Root Sheath
- DRAC
Eveloping Root Apical Complex
- CBCT
Cone-beam Computed Tomography
- ALARA
As Low As Reasonably Achievable
- ALADA
As Low As Diagnostically Acceptable
- MPR
Multiplanar Reconstruction
- ICC
Intraclass Correlation Coefficient
Authors’ contributions
JX and JH design the research plan. JH, XM and MF contributed to the evaluation of CBCT images. JH, XM and JX contributed to interpretation and analyses of data. JH, JX, LW participated in manuscript writing and revision. All authors read and approved the final manuscript.
Data availability
The datasets generated and/or analyzed during the current study are available from the corresponding author upon reasonable request.
Declarations
Ethics approval and consent to participate
This study was performed in line with the principles of the Declaration of Helsinki. Ethical approval was obtained from the Ethics Committee of Jinan Stomatological Hospital (approval number: JNSKQYY-2025-020). Written informed consent was obtained from the parents or legal guardians of any participant included in the study.
Consent for publication
Not Applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Jiang W, Xu L, Wang M, et al. Research Progress on Independent Root Apical Development of Immature Permanent Teeth . J Stomatology. 2022;42(12):1149–52. 10.13591/j.cnki.kqyx.2022.12.019. [DOI] [Google Scholar]
- 2.Zhang S, Yang Y, Jia S, et al. Exosome-like vesicles derived from Hertwig’s epithelial root sheath cells promote the regeneration of dentin-pulp tissue. Theranostics. 2020;10(13):5914–31. 10.7150/thno.43156. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Jin Y. Xu Lin. Developing apical complex is the growth regulating center of tooth root and periodontium development. West China journal of stomatology,2009,27(5):469 – 72, 478. 10.1007/978-1-4020-9623-5_5 [DOI] [PubMed]
- 4.Nelson SJ. Wheeler’s Dental Anatomy, Physiology and Occlusion. 10th ed. Elsevier; 2014.
- 5.Jayakumar J, Corrigendum. Development and eruption of human teeth in the Chinese population: a comprehensive dental atlas. Front Dent Med. 2024;5:1504277. 10.3389/fdmed.2024.1504277. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Nolla CM. The development of permanent teeth. J Dent Child. 1960;27(4):254–66. [Google Scholar]
- 7.Wang LL, Qin XR, Pang YZ, et al. Measurement of the apical anatomy of immature maxillary central incisors using cone-beam computed tomography. BMC Oral Health. 2023;23(1):835. 10.1186/s12903-023-03569-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Golkar S, Khademi A, Saatchi A, et al. Factors influencing the accuracy of electronic apex locators: A scoping review. Dent Res J. 2025;27(22):26. 10.4103/drj.drj_634_23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Shilpa B, Benjamin R, Shankargouda P, et al. Effect of the voxel size on the accuracy of endodontic length measurements using cone-beam computed tomography: A systematic review conducted according to the PRISMA guidelines and Cochrane Handbook for Systematic Reviews of Interventions. Dent Med Probl. 2025;62(1):135–44. 10.17219/dmp/172845. [DOI] [PubMed] [Google Scholar]
- 10.Sevcihan Günen, Yılmaz A, Harorlı Münevver. Evaluation of the relationship between the Demirjian and Nolla methods and the pubertal growth spurt stage predicted by skeletal maturation indicators in Turkish children aged 10–15: investigation study. Acta Odontol Scand. 2018;77(2):107–13. 10.1080/00016357.2018.1510137. [DOI] [PubMed] [Google Scholar]
- 11.Rahim AAA, Davies JA, Liversidge H. Reliability and limitations of permanent tooth staging techniques. Forensic Sci Int. 2023;346:111654. 10.1016/j.forsciint.2023.111654. [DOI] [PubMed] [Google Scholar]
- 12.Shetty H, Shetty S, Kakade A, et al. Cone-beam computed tomographic and histological investigation of regenerative endodontic procedure in an immature mandibular second premolar with chronic apical abscess. J Invest Clin Dent. 2018;9(4):8. 10.1111/jicd.12352. [DOI] [PubMed] [Google Scholar]
- 13.Elheeny AAH, Tony GE. Two-Dimensional Radiographs and Cone-beam Computed Tomography Assessment of Concentrated Growth Factor and Platelet-Rich Fibrin Scaffolds in Regenerative Endodontic Treatment of Immature Incisors with Periapical Radiolucency: A Randomized Clinical Trial. J Endod. 2024;50(6):792–806. 10.1016/j.joen.2024.01.017. [DOI] [PubMed] [Google Scholar]
- 14.Benavides E, Krecioch JR, O’Brien KK, et al. American Dental Association and American Academy of Oral and Maxillofacial Radiology patient selection for dental radiography and cone-beam computed tomography: Clinical recommendations. J Am Dent Association. 2026;157(1):20–e355. 10.1016/j.adaj.2025.10.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Wang X. Fourth National Oral Health Epidemiological Survey Report [M]. People’s Medical Publishing House; 2018.
- 16.Sun J. Clinical Imaging Analysis of Malformed Central Cusp Treatment for Pulp Infection [Master’s thesis]. West China School of Stomatology, Sichuan University, 2021.
- 17.Hao CF, Hao YD, Lou XM, et al. Secular trends of dental caries and association with nutritional status: a retrospective analysis of 16,199 Chinese students from three successive national surveys from 2010 to 2019. Front public health. 2024;12:1379767. 10.3389/fpubh.2024.1379767. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.European Commission. Directorate-General for Energy. Cone beam CT for dental and maxillofacial radiology: evidence-based guidelines. Publications Office 2012. https://data.europa.eu/doi/10.2768/21874.
- 19.Chan F, Brown L, Parashos P. CBCT in contemporary endodontics. Aust Dent J. 2023;68(1):39–55. 10.1111/adj.12995. [DOI] [PubMed] [Google Scholar]
- 20.Zhang CC, Liu YJ, Yang WD, et al. Morphological changes of the root apex in anterior teeth with periapical periodontitis: an in-vivo study. BMC Oral Health. 2022;22(1):31. 10.1186/s12903-022-02062-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Soldado-Cañadas P, Muñoz-Garduño C, Sanz-Serrano D, et al. The Clinical Impact of the Combination of CBCT and Periapical Radiographs on Endodontic Diagnosis and Treatment Planning-A Retrospective Cohort Study. Int Endod J. 2026;59(2):190–202. 10.1111/iej.70050. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Takeshita Y, Iwanaga J, Asaumi J. Radiologic Signs on Panoramic Radiography and CBCT[M]. Cham: Springer; 2025. [Google Scholar]
- 23.Kim SJ, Kim H, Song JS, et al. Root Growth and Apical Foramen Closure Velocity of Maxillary Permanent Central Incisor in Korean Children. J Korean Acad Pediatr Dentistry. 2024;51(4):432–41. 10.5933/JKAPD.2024.51.4.432. [DOI] [Google Scholar]
- 24.Valerie Sgheiza. Sex and ancestry patterning of residual correlations in human dental development: Cooperative genetic interaction and phenotypic plasticity. Am J Biol Anthropol. 2024;184(1):e24908. 10.1002/ajpa.24908. [DOI] [PubMed] [Google Scholar]
- 25.Zhang YQ, Li H, Wu HH, et al. Timing of permanent tooth emergence and its association with physical growth among children aged from four to seven years in nine cities of China. Chin J Pediatr. 2020;58(3):206–12. [DOI] [PubMed] [Google Scholar]
- 26.Ong SH, Kim H, Song JS, et al. Correlation between dental and skeletal maturity in Korean children based on dental maturity percentile: a retrospective study. BMC Oral Health. 2024;24(1):377. 10.1186/s12903-024-04015-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Vandana S, Muthu MS, Akila G. Global variations in eruption chronology of permanent teeth: A systematic review and meta-analysis. Am J Hum biology: official J Hum Biology Council. 2024;36(8):e24060. 10.1002/ajhb.24060. [DOI] [PubMed] [Google Scholar]
- 28.Demirjian A, Levesque GY. Sexual differences in dental development and prediction of emergence. J Dent Res. 1980;59(7):1110–22. 10.1177/00220345800590070301. [DOI] [PubMed] [Google Scholar]
- 29.Leung EMY, Yang Y, Khambay B et al. A Comparative Analysis of Tooth Size Discrepancy between Male and Female Subjects Presenting with a Class I Malocclusion. Sci World J, 2018(1):7641908. 10.1155/2018/7641908. [DOI] [PMC free article] [PubMed]
- 30.Ozge Uslu-Akcam, Yıldız R. Mesiodistal and buccolingual crown diameters of permanent teeth. BMC Oral Health. 2025;25(1):870. 10.1186/s12903-025-06188-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets generated and/or analyzed during the current study are available from the corresponding author upon reasonable request.