This was a retrospective cross-sectional study conducted at the Laboratory for Innovation in Oral Radiology and Imaging (LIRIO), within the Department of Stomatology, Public Health, and Forensic Dentistry (DESCOL). The present research was approved by the Research Ethics Committee of the Ribeirão Preto School of Dentistry (FORP-USP), under protocol number 6.980.465 (CAAE 80541124.5.0000.5419).

The null hypothesis of the study is that there is no difference in dental follicle characteristics and detection rate of the gubernacular canal across different regions of the dental arches. The primary outcome of the study is the absolute and relative frequency of dental follicle characteristics and the detection rate of the gubernacular canal in supernumerary teeth from different regions of the dental arches.

Sample size calculation

The sample size calculation was based on the study by Nishina et al., 2024 [3], considering the proportions found for the detection of the gubernacular canal in supernumerary teeth that either shared or did not share a dental follicle with a normal-count tooth (79.5% vs. 86.8%). The calculation was performed using the “pwr.2p.test” function in R, which determines the parameters necessary to achieve a desired statistical power. The effect size was set at 0.2, with a significance level of 5% and a test power of 80% for a two-tailed test, resulting in a sample of 408 supernumerary teeth.

Sample selection

CBCT images were obtained from the image database of the Radiology Clinic at FORP-USP. Accordingly, the sample comprised CBCT scans of patients presenting intraosseous supernumerary teeth that met the inclusion and exclusion criteria described below. For sample selection, the Radiology and Radioprotection Section (SCRAD) of FORP-USP was requested to select CBCT scans of patients referred to the Oral Radiology service for evaluation of supernumerary teeth, encompassing all ages. The CBCT scans of the database were acquired using two CBCT devices, OP300 (Instrumentarium, Tuusula, Finland) and Eagle 3D (Dabi Atlante, Ribeirão Preto, Brazil). Both devices operated under variable acquisition protocols of kilovoltage (85 or 90 kV), tube current (2 to 13 mA), voxel size (85 to 320 μm), and field of view (FOV 5 × 5 to 23 × 16 cm). The scans were exported by SCRAD radiology technicians, who were not part of the research team, using an image acquisition software (Cliniview, Instrumentarium, Tuusula, Finland) with data anonymization. The only accompanying data were a random number corresponding to each scan, and the patient’s sex and age, ensuring that patient identification was impossible and maintaining confidentiality and privacy. All scans were exported in DICOM format and stored on an external hard drive.

After the initial selection by SCRAD, the final sample was determined by applying the following inclusion and exclusion criteria.

Inclusion criteria

CBCT scans showing intraosseous supernumerary teeth in formation, in the process of eruption, or impacted were included.

Exclusion criteria

Teeth in advanced eruption stages, where the follicular space is so close to the alveolar ridge that detection of the gubernacular canal is impossible. Fully erupted supernumerary teeth were excluded, as the evaluation of the gubernacular canal requires the presence of an intraosseous follicular space. The scans with motion artifacts preventing proper evaluation, and syndromic patients with systemic alterations affecting bone were excluded.

Sample evaluation

For each scan, the patient’s sex and age were recorded, along with the dental group of the supernumerary teeth according to the dental arch (maxilla or mandible) and their region (incisors, canines, premolars, or molars). Once classified, the teeth were evaluated as follows [adapted from Gaêta-Araujo et al., 2019 [4]:

Stage of development: crypt (no crown calcification visible), crown (from the beginning of enamel formation to the cement-enamel junction), up to half of the root (from the beginning of root formation until the root reaches the same length as the crown), beyond half of the root (root length exceeding crown length until the apex remains open), closed apex (tooth fully formed), or indeterminate (when extreme alterations in morphology did not allow assessment of the stage of development).

Morphology: indeterminate (when in crypt stage, morphology cannot be identified), normal (corresponding to teeth of that region), altered (microdontia, macrodontia, or other shape anomalies), or compound odontoma (three or more denticles within the same dental follicle).

Tooth inclination: indeterminate (when in crypt stage, inclination cannot be determined), vertical (long axis parallel to the long axis of adjacent teeth), inclined (long axis angled relative to adjacent teeth), horizontal (long axis perpendicular to adjacent teeth), or inverted (long axis exceeding 90° relative to adjacent teeth).

Relationship between the supernumerary tooth follicle and adjacent intraosseous teeth: isolated (follicle boundaries can be determined and no proximity to adjacent tooth follicles), in contact (follicle boundaries can be determined but touching adjacent tooth follicles), or fused (contact with adjacent tooth follicles with indistinguishable boundaries in the contact region).

Number of adjacent teeth: for teeth in contact or fused with adjacent teeth, the number of adjacent teeth in these conditions was recorded.

Type of adjacent teeth: for teeth in contact or fused with adjacent teeth, they were categorized as normal-count teeth, supernumerary teeth, or both.

Gubernacular canal: absent or present (when a hypodense band originating from the dental follicle and extending toward a bone surface was identified).

The selected scans were then evaluated using OnDemand3D software (Cybermed Inc., Seoul, Republic of Korea) on an MDRC-22 medical imaging monitor (BARCO, Kortrijk, Belgium), in a quiet environment with low ambient lighting suitable for image analysis. The evaluation was performed dynamically, navigating through all slices corresponding to the tooth under analysis. To begin the tooth assessment, the tomographic slices were aligned with the long axis of the tooth in the axial, sagittal, and coronal planes.

Initially, 200 cases were evaluated by two examiners, both dentists with experience in oral radiology (CBCT interpretation). After individual evaluations, the findings were verified jointly by both evaluators. Subsequently, 100 cases were analyzed by an oral and maxillofacial radiologist with more than 10 years of CBCT analysis experience and compared with the previous evaluations to refine calibration. Once these two calibration stages were completed, the entire sample was evaluated by a single examiner, including re-evaluation of the initial 200 cases.

Relationship between the dental follicle, gubernacular canal, and developmental theories

Based on the analysis of the relationship between the supernumerary tooth follicle and adjacent intraosseous tooth follicles, along with the presence or absence of the gubernacular canal (Fig. 1), hypotheses were formulated regarding the main developmental theories of supernumerary teeth: the dichotomy theory and the dental lamina hyperactivity theory. These hypotheses consider that teeth presenting a gubernacular canal originate from dental lamina and thus suggests development according to the hyperactivity theory. Conversely, the absence of the gubernacular canal may be suggestive of the follicular dichotomy. Accordingly, teeth with isolated follicles without a gubernacular canal suggest complete dichotomy, whereas teeth with isolated follicles and with a gubernacular canal could suggest the dental lamina hyperactivity theory. Supernumerary teeth with follicles in contact but lacking a gubernacular canal are suggestive of partial dichotomy, while teeth with follicles in contact and a gubernacular canal suggest dental lamina hyperactivity.

Regarding fused follicles, those without a gubernacular canal could be related to partial dichotomy, whereas fused follicles with a gubernacular canal could originate from dental lamina hyperactivity if the canal within the follicle belonged to that tooth, or from partial dichotomy if the canal belonged to another tooth. To determine which tooth in a fused follicle the gubernacular canal belonged to, the tooth type (normal-count or supernumerary) and its proximity to the canal were taken into account (in cases with more than one supernumerary tooth).

Fig. 1The alternative text for this image may have been generated using AI.

CBCT images with the classification of tooth follicles considering their relationship with adjacent teeth (isolated, in contact or fused) and the presence (indicated by the yellow arrows) or absence of the gubernacular canal. A Supernumerary tooth in lower canine region with an isolated dental follicle and absence of the gubernacular canal; B Supernumerary tooth in lower premolar region with the follicle in contact with the adjacent tooth and absence of the gubernacular canal. C Supernumerary tooth in lower premolars region with fusion between the dental follicles of adjacent teeth and absence of the gubernacular canal. D Supernumerary tooth in lower premolars region with an isolated follicle and presence of the gubernacular canal. E Supernumerary tooth in upper canine region with the follicle in contact with the adjacent tooth and presence of the gubernacular canal. F Supernumerary tooth in upper central incisor region with fusion between the dental follicles of adjacent teeth and presence of the gubernacular canal

Data analysis

Data analysis was performed using SPSS version 25.0 (IBM SPSS Statistics, Armonk, NY, USA) and Prism version 8.0 (GraphPad, La Jolla, CA, USA), with a significance level set at 5%. The chi-square test was used to compare tooth characteristics (region, development stage, morphology, and angulation) between dental arches, as well as to evaluate differences in dental follicle characteristics and gubernacular canal detection rates across regions of the dental arches. Additionally, it was applied to assess the association between the presence of the gubernacular canal and dental follicle characteristics across regions and dental arches, including their possible relationship with developmental theories.