Evaluation of Alveolar Bone Loss With ECR Using Cone‐Beam Computed Tomography: A Cross‐Sectional Study Authors: Ana Paula de Medeiros Silva, Thalles Gabriel Germano Lima, Ana Beatriz Leitão de Albuquerque Wanderley, Monikelly do Carmo Chagas Nascimento, Daniela Siqueira Lopes, Hannah Wesley Published: 2026 ABSTRACT AbstractAlveolar bone loss (ABL) and external cervical resorption (ECR) are destructive processes that can coexist and compromise periodontal prognosis. Cone‐beam computed tomography (CBCT), enables assessment of resorption severity and bone loss patterns, overcoming the limitations of two‐dimensional radiography. This study aimed to evaluate ABL and its correlation with ECR using CBCT. This analytical, cross‐sectional, observational study used images to database spanning January 2019 to April 2025, acquired with a CBCT scanner. A total of 9208 examinations were screened. After applying the inclusion criteria, 64 patients were enrolled, comprising 75 teeth diagnosed with ECR. Sociodemographic and clinical data, ECR characteristics, and ABL patterns and classification were analyzed. The findings demonstrated a statistically significant correlation between ABL and ECR according to tooth groups, with stronger associations observed in premolars and molars (10.7%), subcrestal lesion height (41.3%), and circumferential distribution 90° to 180° (44.0%). ABL was associated with age group and lesion circumferential distribution (p < 0.05). The horizontal pattern of bone loss was the most prevalent (54.7%), with significant associations between severity and age group (p = 0.022) and lesion height (p = 0.007). An association was also observed between lesion height and periodontal condition (p = 0.043). In conclusion, ECR performed a significant correlation with ABL in the studied population (p < 0.05). Three‐dimensional characterization of ECR lesions and their relationship with bone loss patterns and periodontal status underscores the critical role of imaging in the early detection of potentially progressive lesions.Keywords: alveolar bone loss, cone-beam computed tomography, diagnosis, root resorption 1. Introduction Dysbiotic bacterial communities and the host immune response, which are frequently associated with poor oral hygiene in clinical practice, contribute to a chronic inflammatory process that results in a reduction of the alveolar crest and subsequent alveolar bone loss (ABL). The severe condition is also the most prevalent among populations and is classified by pattern as horizontal or vertical based on the number of remaining bony walls [1, 2]. In addition, another clinical disorder is the external cervical resorption (ECR), which may lead to progressive tooth destruction and affect the patient’s quality of life. Usually begins in the cervical third of the root and extends toward the root canal as well as vertically toward the root structure [3]. Damage to the periodontal ligament and subepithelial cementum is rarely associated, thereby activating clastic cells and initiating a resorptive cascade [4, 5]. The resorptive process is dynamic and may involve the periodontal tissues, potentially altering alveolar bone. However, diagnosis remains challenging and may be influenced by periodontal conditions [4]. Early‐stage or interproximal lesions are often difficult to detect, whereas cavitated cervical defects on the buccal and lingual surfaces are more readily identified by direct visualization or periodontal probing [6]. Imaging systems provide valuable complementary support to clinical decision‐making and treatment. However, the assessment of ECR using two‐dimensional imaging is limited by geometric distortion, anatomical superimposition, and insufficient characterization of dental anatomy [7]. Cone‐beam computed tomography (CBCT) provides three‐dimensional (3D) visualization, reducing distortion and anatomical overlap, thereby enhancing diagnostic accuracy and improving treatment planning [8]. CBCT, such as PreXion 3D (PreXion Corporation, Japan), improves the assessment of probing depth, defect pattern (vertical and/or horizontal), classification of ABL, and the detection of ECR [1, 2, 9]. The specificity of CBCT depends on acquisition parameters, such as exposure settings, voxel configuration, detector sensitivity, and tomographic reconstruction algorithms, as well as on examiner experience [3]. Features include continuous X‐ray emission and an integrated motion‐artifact system, enhancing image quality and diagnostic reliability [8, 10]. Distance between the cementoenamel junction (CEJ) and the alveolar bone crest, as observed in postorthodontic bone dehiscences and periodontitis, is associated with a morphological pattern characterized by cervical resorptive processes, suggesting an overlap of etiological factors and clinical manifestations [11]. However, a scarcity of studies correlating ABL and ECR within a single three‐dimensional approach limits the understanding of the relationship between bone destruction and cervical resorption and the identification of ABL patterns as potential predictive markers of ECR. The present study aimed to evaluate ABL and its correlation with ECR using CBCT images. The hypothesis tested was that ECR severity is associated with the presence of adjacent ABL. 2. Materials and Methods 2.1. Study Design and Ethical Approval This analytical cross‐sectional observational study was conducted in accordance with the Declaration of Helsinki and approved by the Research Ethics Committee of the University of Pernambuco (approval number 7.129.157), in accordance with Resolution No. 466/2012 of the National Health Council/Ministry of Health. The study was reported following the STROBE guidelines for observational studies [12]. 2.2. Sample Selection A total of 9208 CBCT image records, obtained between January 2019 and April 2025 using the PreXion 3D system (PreXion Corporation, Japan), under the following operating parameters: 90 kVp, 4 mA, exposure time of 33.5 s, voxel size of 0.100 mm3, and field of view (FOV) of 50 mm × 50 mm. Images were analyzed using PreXion 3D Viewer software (PreXion Corporation, Japão. Patient identification codes (IDs), examination dates, and sociodemographic variables (age, sex, dental group, affected region, reason for examination request, and referring professional specialty) were recorded in a Microsoft Excel spreadsheet. To identify cases of ECR, digital radiology reports were retrospectively searched using the terms “resorption,” “cervical,” and “hypodense.” Subsequently, all CBCT images were screened for ABL. This process initially yielded 421 tomographic reports containing all three search terms. After removing duplicates and applying eligibility criteria, 217 cases were retained. Further DICOM file evaluation led to the exclusion of 153 cases due to caries‐related features, resulting in a final sample of 64 patients and 75 ECR‐affected teeth (Figure 1). Study phases of sample selection. 2.2.1. Inclusion Criteria CBCT image records: •Adult patients of both genders presenting one or more permanent teeth with ECR, regardless of the presence or absence of ABL. Adult patients of both genders presenting one or more permanent teeth with ECR, regardless of the presence or absence of ABL. •Teeth with a clearly identifiable CEJ on at least one tooth surface in the three‐dimensional CBCT images. Teeth with a clearly identifiable CEJ on at least one tooth surface in the three‐dimensional CBCT images. •Permanent teeth fully developed. Permanent teeth fully developed. 2.2.2. Exclusion Criteria CBCT image records: •Artifacts and/or lesions in the region of interest impair the evaluation of the structures. Artifacts and/or lesions in the region of interest impair the evaluation of the structures. •Impacted or partially erupted teeth. Impacted or partially erupted teeth. •Teeth located at the periphery of the FOV and with extensive restorations. Teeth located at the periphery of the FOV and with extensive restorations. 2.2.3. Examiner Training and Calibration Three calibrated examiners: an endodontist, a periodontist, and an experienced radiologist, carried out the assessments. A representative subset corresponding to 10% of the sample was evaluated [3, 12]. Inter‐ and intra‐examiner agreement was assessed using the Kappa statistic, with a significance level set at 5% (α = 0.05) [6]. Reliability was additionally estimated using Krippendorff’s alpha [3], yielding values of 0.907 in the first assessment and 0.815 in the second, indicating almost perfect agreement. Following calibration, all CBCTs included in the study were assessed by one calibrated examiner. 2.3. ECR Analysis Hypodense images were observed in the three orthogonal planes, originating from the cervical third of the root, and were analyzed [7]. Cases presenting hypodense or hyperdense images with reparative mineralized tissue, extending to the middle and apical thirds and circumferentially involving the root canal, were included [9]. Lesions were differentiated from dental caries by excluding rounded, focal hypodense areas with a mushroom‐shaped appearance, characterized by a narrow pathway through enamel and a broader area of demineralized dentin [13]. Teeth were classified according to the Patel classification, allowing assessment of lesion height, circumferential distribution, and relationship with the pulp [14, 15]. The apicocoronal extent of the lesion relative to the alveolar crest was categorized as supracrestal or subcrestal. The circumferential extent was classified as A (<90°), B (90°–180°), C (180°–270°), or D (>270°). The anatomical involvement was categorized as confined to dentin (D) or associated with pulpal involvement (P). In addition, the number of entry portals, their location (buccal, lingual, palatal, mesial, or distal), and the size of each entry portal (≤1 or >1 mm) were assessed (Figure 2). Measurements of external cervical resorption (ECR): coronal section—(A) lesion height; axial section—(B) circumferential distribution, (C) proximity to the root canal, (D) location of the entry portal, and (E) size of the entry portal. 2.4. ABL Analysis ABL was assessed by measuring the distance between the cemento‐enamel junction (CEJ) and the alveolar bone crest using the PreXion 3D Viewer software (PreXion Corporation, Japan). Mesial and distal surfaces and buccal and lingual surfaces were evaluated in the cross‐sectional view [16]. The most coronal level at which the periodontal ligament space width remained within normal limits (<2 mm) was the endpoint [17]. ABL was assessed in coronal sections and categorized by pattern (horizontal or vertical) and extent of root involvement as slight (≤15%), moderate (16%–30%), or severe (>30%) [1, 2]. ECR was considered associated with ABL when both defects involved the same root surface. This classification further required evidence of communication between the ECR and the periodontal space, indicated by thickening of the periodontal ligament observed across all tomographic sections (Figure 3). Measurements of cementoenamel junction (CEJ) to alveolar crest to determine alveolar bone loss (ABL). (A) Coronal section: a–b, CEJ to root apex; d–c, CEJ to mesial alveolar crest; e–f, CEJ to distal alveolar crest. (B) Sagittal section: a–b, CEJ to root apex; d–c, CEJ to buccal alveolar crest; e–f, CEJ to palatal alveolar crest. 2.5. Statistical Analysis Categorical variables are presented as frequencies and percentages, whereas continuous variables are expressed as mean ± standard deviation. Associations between ECR and ABL variables were examined using nonparametric tests (Mann–Whitney and Kruskal–Wallis) or categorical tests (Pearson’s chi‐square or Fisher’s exact test), according to the distribution and nature of the variables. Statistical significance was established at p < 0.05. To assess factors associated with ABL directly related to ECR, a linear mixed‐effects model with a binomial distribution and logit link function was executed. Age group, sex, tooth group, lesion height, circumferential lesion distribution, and lesion proximity to the root canal were included as fixed effects. As the 75 teeth evaluated belonged to 64 patients, patient was included as a random effect, with a random intercept to account for the potential correlation between teeth from the same patient. Associations were expressed as adjusted odds ratios (ORs) with corresponding 95% confidence intervals (95% CIs). 3. Results 3.1. Characteristics of the Study Participants In the analyzed subpopulation, ECR was observed in patients with a mean age of 46.1 years (±17.6), with a higher prevalence between 40 and 50 years, followed by 60 years or older (p ≤ 0.05). The sample was similar between genders, with 53.1% males and 46.9% females (p ≥ 0.05). The maxilla was the most frequently involved anatomical region. Maxillary incisors were the most affected teeth (90%), followed by maxillary canines (62.5%) and maxillary molars (52.3%). These findings were included in the association analysis with the other variables (p ≤ 0.05), as shown in Table 1. Table 1Characteristics of patients, analyzed region, and tooth group.Variable n (%) p‐ValueTotal64 (100.0)—Results by patientAge group— p a = 0.024 ∗ Up to 2913 (20.3)— 30 or older51 (79.7)— Total64 (100.0)—Gender— p a = 0.133 Male34 (53.1)— Female30 (46.9)— Total64 (100.0)—By anatomical region analyzedResultsAnatomical region requested— p a < 0.001 ∗ Maxilla49 (65.3)— Mandible19 (25.3)— Both7 (9.3)— Total75 (100.0)—Tooth group— p a = 0.004 ∗ Incisors30 (40.0)— Canines16 (21.3)— Premolars8 (10.7)— Molars21 (28.0)— Total75 (100.0)— aChi‐square test for comparison of proportions in a single sample. ∗Significant difference at the 5.0% level. Characteristics of patients, analyzed region, and tooth group. aChi‐square test for comparison of proportions in a single sample. ∗Significant difference at the 5.0% level. Endodontic evaluation accounted for the highest proportion of referrals for tomographic examinations in which ECR lesions were identified. However, ECR was not the primary diagnostic suspicion in all cases. Medical specialties, Stomatology, and Prosthodontics accounted for lower frequencies (1.6%, respectively) (Figures 4 and 5). Frequency of the reason for referral. Percentage frequency of the specialty. 3.2. Characteristics and Classification of ECR Regarding lesion height, subcrestal involvement accounted for the highest frequency (41.3%), whereas the lowest frequency was observed when the lesion extended to the apical third (6.7%). Regarding circumferential distribution, the greatest involvement occurred between 90° and 180°. Furthermore, concerning the proximity of the lesion to the root canal, dentin‐confined involvement was the most frequent finding, accounting for 64% of cases (p ≤ 0.05). The most frequent location of the entry portal was the palatal/lingual surface (36%), measuring less than 1 mm. Other locations had lower proportions, without significant differences among them (Table 2 and Figure 6). According to Table 3, type 2Bd was the most prevalent Patel classification among teeth with ECR (14.7%), followed by type 2Ad (12%), with observed differences being statistically significant (p ≤ 0.05) (Figure 7). Table 2Assessment of ECR characteristics.Variable n (%) p‐ValueTotal75 (100.0)—Lesion height— p a < 0.001 ∗ Supracrestal21 (28.0)— Subcrestal31 (41.3)— Extending to the middle third18 (24.0)— Extending to the apical third5 (6.7)—Circumferential distribution of the lesion— p a < 0.001 ∗ <9019 (25.3)— 90–18033 (44.0)— 180–2708 (10.7)— >27015 (20.0)—Proximity of the lesion to the root canal— p a) = 0.015 ∗ Within dentin48 (64.0)— With pulpal involvement27 (36.0)—Location of the portal of entry on the surface——Bucal— p a < 0.001 ∗ Yes19 (25.3)— No56 (74.7)—Palatal/lingual— p a = 0.015 ∗ Yes27 (36.0)— No48 (64.0)—Mesial— p a < 0.001 ∗ Yes19 (25.3)— No56 (74.7)—Distal— p a = 0.004 ∗ Yes25 (33.3)— No50 (66.7)—Lesion size at the portal of entry——Bucal— p a = 0.008 ∗ <1 mm12 (63.2)— >1 mm6 (31.6)— No reported1 (5.3)— Total19 (100.0)—Palatal/lingual— p a = 0.564 <1 mm15 (55.6)— >1 mm12 (44.4)— Total27 (100.0)—Mesial— p a = 0.004 ∗ <1 mm13 (68.4)— >1 mm4 (21.1)— No reported2 (10.5)— Total19 (100.0)—Distal— p a = 0.072 <1 mm17 (68.0)— >1 mm8 (32.0)— Total25 (100.0)— aChi‐square test for the comparison of proportions in a single sample. ∗Significant difference at the 5.0% level. Assessment of ECR characteristics. aChi‐square test for the comparison of proportions in a single sample. ∗Significant difference at the 5.0% level. Representative images of features associated with external cervical resorption (ECR). Arrows indicate (1–5): (1) subcrestal lesion level; (2) circumferential distribution ranging from 90° to 180°; (3) lesion confined to dentin; (4) entry portal located on the palatal surface; and (5) entry portal measuring <1 mm. Table 3Distribution of ECR classification.Variable n (%) p‐ValueTotal75 (100.0)—Patel classification— p a < 0.001 ∗ 1BD8 (10.7)— 1BP4 (5.3)— 1CD1 (1.3)— 1DP1 (1.3)— 2AD9 (12)— 2AP1 (1.3)— 2BD11 (14.7)— 2BP6 (8.0)— 2CD2 (2.7)— 2DP2 (2.7)— 3AD2 (2.7)— 3BD2 (2.7)— 3BP2 (2.7)— 3CD3 (4.0)— 3CP1 (1.3)— 3DP8 (10.7)— 4CD1 (1.3)— 4DP2 (2.7)— aChi‐square test for comparison of proportions in a single sample. ∗Significant difference at the 5.0% level. Distribution of ECR classification. aChi‐square test for comparison of proportions in a single sample. ∗Significant difference at the 5.0% level. Image illustrating category 2Bd of the Patel classification, as one of the findings of the study. 3.3. ABL Assessment In the evaluation of ABL in teeth with ECR, a predominance of the horizontal pattern (54.7%) and of cases classified as severe (60%) was observed. A higher prevalence was also noted in findings where bone loss was directly associated with ECR, as well as a hightest periodontal space communication (p ≤ 0.05) (Table 4 and Figure 8). Table 4Assessment of alveolar bone loss (ABL) data of the analyzed teeth.Variable n (%) p‐ValueTotal75 (100.0)—Pattern of bone loss— p a < 0.001 ∗ Vertical25 (33.3)— Horizontal41 (54.7)— No loss9 (12.0)—Bone loss classification— p a < 0.001 ∗ Mild12 (16.0)— Moderate18 (24.0)— Severe45 (60.0)—Bone loss directly related to external cervical resorption— p a < 0.001 ∗ Yes59 (78.7)— Not16 (21.3)—Communication of the resorption with the periodontal space— p a < 0.001 ∗ Yes52 (69.3)— Not23 (30.7)— aChi‐square test for the comparison of proportions in a single sample. ∗Significant difference at the 5.0% level. Assessment of alveolar bone loss (ABL) data of the analyzed teeth. aChi‐square test for the comparison of proportions in a single sample. ∗Significant difference at the 5.0% level. Illustrative image where the arrows indicate, from 1 to 4: (1) horizontal pattern of bone loss; (2) bone loss classified as severe; (3) bone loss directly associated with external cervical resorption; and (4) communication between the resorptive area and the periodontal space. 3.4. ABL and ECR Correlation A significant association between ABL directly related to ECR and lesion morphology (height and circumferential distribution) (Fisher’s exact test showed p < 0.05). Pearson’s chi‐square test showed no significant association with lesion proximity to the root canal (p > 0.05) (Table 5). Table 5Assessment of the occurrence of ABL directly related to ECR according to age group, gender, tooth group, lesion height, circumferential distribution, and proximity of the lesion to the root canal.VariableAlveolar bone loss directly associated with external cervical resorption (ECR)Total p‐ValueOR (95% CI)YesNo n (%) n (%) n (%)Overall group59 (78.7)16 (21.3)75 (100.0)——Age group—— — p a = 0.013 ∗ — Up to 29 years8 (53.3)7 (46.7)15 (100.0)—(1.0) 30 years or older51 (85.0)9 (15.0)60 (100.0)—5.0 (1.4–17.1)Gender——— p b = 0.825— Male35 (79.5)9 (20.5)44 (100.0)—1.1 (0.4–3.5) Female24 (77.4)7 (22.6)31 (100.0)—1.00Tooth group——— p a = 0.015 ∗ — Incisors/canines32 (69.6)14 (30.4)46 (100.0)—1.0 Premolars/molars27 (93.1)2 (6.9)29 (100.0)—5.9 (1.2–28.3)Lesion height in relation to crown/root——— p a = 0.043 ∗ — Supracrestal18 (85.7)3 (14.3)21 (100.0)—3.9 (0.9–17.0) Subcrestral27 (87.1)4 (12.9)31 (100.0)—4.3 (1.1–16.6) Extending to the apical third14 (60.9)9 (39.1)23 (100.0)—1.0Circumferential distribution of the lesion——— p a = 0.012 ∗ — Up to 18045 (86.5)7 (13.5)52 (100.0)—4.1 (1.3–13.1) >18014 (60.9)9 (39.1)23 (100.0)—1.0Proximity of the lesion to the root canal——— p b = 0.888— Confined to dentin38 (79.2)10 (20.8)48 (100.0)—1.1 (0.3–3.4) With pulpal involvement21 (77.8)6 (22.2)27 (100.0)—1.0 aFisher’s exact test. bPearson’s chi‐square test. ∗Significant difference at the 5.0% level. Assessment of the occurrence of ABL directly related to ECR according to age group, gender, tooth group, lesion height, circumferential distribution, and proximity of the lesion to the root canal. aFisher’s exact test. bPearson’s chi‐square test. ∗Significant difference at the 5.0% level. Subcrestal lesions were associated with a higher bias of ABL (OR = 4.3) compared with lesions extending to the apical third. Similarly, ECR lesions with circumferential involvement of 90° to 180° showed a higher odds of ABL (OR = 4.1) than those with involvement >270°. However, no significant association was observed between lesion proximity to the root canal and ABL (p > 0.05) (Table 5). Findings concern ABL and ECR. The Kruskal–Wallis test showed no significant association with the ABL pattern (p > 0.05), but a significant association with lesion height (p < 0.05). No significant association was observed between circumferential distribution and ABL (p > 0.05). The Mann–Whitney U test also showed no significant association between lesion next root canal and ABL (p > 0.05) (Tables 6 and 7 and Figures 9 and 10). Table 6Evaluation of the ABL pattern according to age group, gender, tooth group, lesion height, circumferential distribution, and proximity to the root canal.VariablePattern of bone lossTotal p‐ValueVerticalHorizontalBone loss n (%) n (%) n (%) n (%)Total group25 (33.3)41 (54.7)9 (12.0)75 (100.0)—Age group———— p a = 0.095 Up to 294 (26.7)6 (40.0)5 (33.3)15 (100.0)— 30 or older21 (35)35 (58.3)4 (6.7)60 (100.0)—Gender———— p a = 0.893 Male14 (31.8)25 (56.8)5 (11.4)44 (100.0)— Female11 (35.5)16 (51.6)4 (12.9)31 (100.0)—Tooth group———— p b = 0.463 Incisors8 (26.7)16 (53.3)6 (20.0)30 (100.0)— Canines6 (37.5)8 (50.0)2 (12.5)16 (100.0)— Premolars4 (50.0)3 (37.5)1 (12.5)8 (100.0)— Molars7 (33.3)14 (66.7)—21 (100.0)—Lesion height———— p b = 0.531 Supracrestal5 (23.8)15 (71.4)1 (4.8)21 (100.0)— Subcrestral10 (32.3)20 (64.5)1 (3.2)31 (100.0)— Extending to the middle third9 (50.0)4 (22.2)5 (27.8)18 (100.0)— Extending to the apical third1 (20.0)2 (40.0)2 (40.0)5 (100.0)—Circumferential distribution of the lesion———— p b = 0.413 <906 (31.6)12 (63.2)1 (5.3)19 (100.0)— 90–18012 (36.4)20 (60.6)1 (3.0)33 (100.0)— 181–2702 (25.0)4 (50.0)2 (25.0)8 (100.0)— >2705 (33.3)5 (33.3)5 (33.3)15 (100.0)—Proximity of the lesion to the root canal———— p a = 0.793 Within dentin14 (29.2)30 (62.5)4 (8.3)48 (100.0)— With pulpal involvement11 (40.7)11 (40.7)5 (18.5)27 (100.0)— aMann–Whitney test. bKruskal–Wallis test. Evaluation of the ABL pattern according to age group, gender, tooth group, lesion height, circumferential distribution, and proximity to the root canal. aMann–Whitney test. bKruskal–Wallis test. Table 7Evaluation of ABL classification according to age group, gender, tooth group, lesion height, circumferential distribution, and proximity to the root canal.VariableClassification of alveolar bone lossTotal p‐ValueMildModerateSevere n (%) n (%) n (%) n (%)Total group12 (16.0)18 (24.0)45 (60.0)75 (100.0)—Age group———— p a = 0.022 ∗ Up to 294 (26.7)6 (40.0)5 (33.3)15 (100.0)— 30 or older8 (13.3)12 (20.0)40 (66.7)60 (100.0)—Gender———— p a = 0.610 Male7 (15.9)12 (27.3)25 (56.8)44 (100.0)— Female5 (16.1)6 (19.4)20 (64.5)31 (100.0)—Tooth group———— p b = 0.388 Incisors7 (23.3)6 (20.0)17 (56.7)30 (100.0)— Canines3 (18.8)5 (31.3)8 (50.0)16 (100.0)— Premolars2 (25.0)1 (12.5)5 (62.5)8 (100.0)— Molars—6 (28.6)15 (71.4)21 (100.0)—Lesion height———— p b = 0.007 ∗ Supracrestala 2 (9.5)6 (28.6)13 (61.9)21 (100.0)— Subcrestrala 1 (3.2)10 (32.3)20 (64.5)31 (100.0)— Extending to the middle thirda 5 (27.8)1 (5.6)12 (66.7)18 (100.0)— Extending to the apical thirda 4 (80.0)1 (20.0)–5 (100.0)—Circumferential distribution of the lesion———— p b = 0.106 <902 (10.5)7 (36.8)10 (52.6)19 (100.0)— 90–1802 (6.1)7 (21.2)24 (72.7)33 (100.0)— 181–2703 (37.5)2 (25.0)3 (37.5)8 (100.0)— >2705 (33.3)2 (13.3)8 (53.3)15 (100.0)—Proximity of the lesion to the root canal———— p a = 0.758 Within dentin8 (16.7)12 (25.0)28 (58.3)48 (100.0)— Within pulpal involvement4 (14.8)6 (22.2)17 (63.0)27 (100.0)— Note: If all letters in parentheses are different, this indicates a significant difference between the corresponding results. aMann–Whitney test. bKruskal–Wallis test with Conover’s multiple comparisons. ∗Significant difference at the 5.0% level. Evaluation of ABL classification according to age group, gender, tooth group, lesion height, circumferential distribution, and proximity to the root canal. Note: If all letters in parentheses are different, this indicates a significant difference between the corresponding results. aMann–Whitney test. bKruskal–Wallis test with Conover’s multiple comparisons. ∗Significant difference at the 5.0% level. Illustrative image of the relationship between bone loss and external cervical resorption at the subcrestal level. Illustrative image of the relationship between severe bone loss and subcrestal lesion height. A significant association was found with ECR and the periodontal space lesion height (p < 0.05), but not with circumferential distribution (p > 0.05). Subcrestal ECR had the highest frequency. Twenty percent of lesions that reached the apical third communicated with the periodontal space. Lesions involving 90° to 180° were more likely to communicate (OR = 3.3), but this was not statistically significant (p > 0.05). Lesion proximity to the root canal showed no significant association (p > 0.05), despite a higher frequency when the pulp was involved (Table 8 and Figure 11). Table 8Evaluation of the communication between ECR and the periodontal condition according to age group, gender, tooth group, lesion height, circumferential distribution, and proximity to the root canal.VariableCommunication of the resorption with the periodontal spaceTotal p‐ValueOR (95% CI)YesNo n (%) n (%) n (%)Total group52 (69.3)23 (30.7)75 (100.0)——Age group——— p a = 0.369— Up to 2912 (80.0)3 (20.0)15 (100.0)—2.0 (0.5–7.9) 30 or older40 (66.7)20 (33.3)60 (100.0)—1.0Gender——— p b = 0.797— Male30 (68.2)14 (31.8)44 (100.0)—1.0 Female22 (71.0)9 (29.0)31 (100.0)—1.1 (0.4–3.1)Tooth group——— p a = 0.378— Incisors18 (60.0)12 (40.0)30 (100.0)—1.0 Canines12 (75.0)4 (25.0)16 (100.0)—2.0 (0.5–7.7) Premolars5 (62.5)3 (37.5)8 (100.0)—3.8 (0.9–15.8) Molars17 (81.0)4 (19.0)21 (100.0)—2.8 (0.8–10.5)Lesion height——— p a = 0.024 ∗ — Supracrestal13 (61.9)8 (38.1)21 (100.0)— ∗∗ Subcrestral26 (83.9)5 (16.1)31 (100.0)— ∗∗ Extending to the middle third12 (66.7)6 (33.3)18 (100.0)— ∗∗ Extending to the apical third1 (20.0)4 (80.0)5 (100.0)— ∗∗ Circumferential distribution of the lesion——— p a = 0.201— <9011 (57.9)8 (42.1)19 (100.0)—1.00 90–18027 (81.8)6 (18.2)33 (100.0)—3.3 (0.9–11.6) 181–2705 (62.5)3 (37.5)8 (100.0)—1.2 (0.2–6.6) >2709 (60.0)6 (40.0)15 (100.0)—1.1 (0.3–4.3)Proximity of the lesion to the root canal——— p b = 0.234— Within dentin31 (64.6)17 (35.4)48 (100.0)—1.0 With pulpal involvement21 (77.8)6 (22.2)27 (100.0)—1.9 (0.7–5.7) aFisher’s exact test. bPearson’s chi‐square test. ∗Significant difference at the 5.0% level. ∗∗Not calculated due to very low frequency and consequently very wide intervals. Evaluation of the communication between ECR and the periodontal condition according to age group, gender, tooth group, lesion height, circumferential distribution, and proximity to the root canal. aFisher’s exact test. bPearson’s chi‐square test. ∗Significant difference at the 5.0% level. ∗∗Not calculated due to very low frequency and consequently very wide intervals. Illustrative image of the communication between subcrestal external cervical resorption and the periodontal ligament. None of the variables included in the generalized linear mixed‐effects model showed a statistically significant association with the presence of ABL directly related to ECR (Table 9). The significant associations observed in the bivariate analysis for age group, tooth group, lesion height, and circumferential lesion distribution were no longer statistically significant in the generalized linear mixed‐effects model. Table 9Mixed‐effects logistic regression model for factors associated with alveolar bone loss directly related to external cervical resorption.VariableOR95% CI p‐ValueAge group——0.118 Up to 291.001.00— 30 or older3.030.71–20.77—Gender——0.819 Male1.730.25–5.61— Female1.001.00—Tooth group——0.213 Incisors/canines1.001.00— Premolars/molars3.090.50–21.92—Lesion height——0.639 Supracrestal2.010.15–7.46— Subcrestal2.980.33–19.65— Apical1.001.00—Circumferential distribution of the lesion——0.520 ≤180°2.570.27–13.07— >180°1.001.00—Proximity of the lesion to the root canal——0.601 With pulpal involvement1.001.00— Within dentin1.250.12–3.50— Note: Patient was included as a random intercept.Abbreviations: CI, confidence interval; OR, odds ratio. ∗Significant difference at the 5.0% level. Mixed‐effects logistic regression model for factors associated with alveolar bone loss directly related to external cervical resorption. Note: Patient was included as a random intercept. Abbreviations: CI, confidence interval; OR, odds ratio. ∗Significant difference at the 5.0% level. 4. Discussion The hypothesis regarding the coronal‐apical extent and circumferential involvement of ECR was confirmed by the significant association between periodontal space lesion height and greater ABL in this cross‐sectional study. These findings indicate that more extensive ECR lesions are associated with greater adjacent ABL, supporting a relationship between lesion severity and periodontal involvement. The sample comprised adults who underwent CBCT examination in a Brazilian clinical setting, with a predominance of males, and the findings were possibly associated with less frequent dental attendance and a higher incidence of dental trauma [18]. This pattern can also be observed in age distribution, particularly among middle‐aged individuals [6]. Factors such as dental trauma, previous orthodontic treatment, and periodontal disease tend to accumulate over time. This trend can, to some extent, explain the higher prevalence of ECR observed in this group. CBCT enables quantification of the periodontal destructive potential of resorptive lesions through its high spatial resolution and multiplanar reconstruction capabilities. The use of a reduced FOV (50 mm × 50 mm) facilitated the detection of subcrestal and subgingival lesions, whereas previous studies employing larger acquisition fields reported lower detection of these characteristics, potentially underestimating early‐stage lesions [8, 10]. A previous Brazilian study using the same CBCT system but a longer acquisition time reported less detailed characterization of the dentogingival interface [10]. Consistent with these observations, the present analysis showed a predominance of lesions with subcrestal height and circumferential involvement ranging from 90° to 180°. These findings suggest that the quality of the DICOM dataset improved the identification of morphological features, which may partly explain the variability reported in previous studies [3, 14]. Involvement of the palatal/lingual surface was observed, contrasting with a previous ex vivo tomographic study that reported a predominance of the buccal surface, identified as the portal of entry [19]. This discrepancy may be related to differences in the tomographic device, FOV, and image resolution, particularly in radiographically challenging areas such as the palatal/lingual surface. However, concluding that the PreXion CBCT provides superior diagnostic performance would be premature. Assessment of ABL patterns in cases of ECR showed a predominantly horizontal distribution, consistent with epidemiological findings in adult populations [9]. This pattern has been primarily attributed to periodontitis [9]. However, other authors suggested that local predisposing factors, such as dental trauma and orthodontic treatment, can also contribute to ECR development [5, 20]. Previous periodontal destruction may act as a risk marker as alterations in the periodontal microenvironment and inflammatory remodeling can predispose cervical root regions to resorption. Considering the demographic profile of the study’s representative sample, a higher frequency of severe ABL was observed in teeth with resorptive lesions than in studies evaluating isolated bone loss, in which severity has been primarily associated with systemic factors, oral hygiene, and socioeconomic conditions [21]. In our study, the presence of resorptive lesions was associated with increased ABL, suggesting the involvement of an additional aggravating factor. This supports the multifactorial nature of periodontal breakdown. Although accounting for multiple affected teeth within the same patient did not reveal significant associations with ABL in the mixed‐effects model, in contrast to some associations observed in the bivariate analysis, the wide CIs indicate considerable uncertainty in the estimates. This approach allowed a more detailed assessment of factors associated with ABL directly related to ECR, considering multiple lesion characteristics and different affected teeth within the same patient. To our knowledge, this aspect has received limited investigation in previous ECR studies [3, 6, 14]. Thus, these findings provide a basis for further investigation and characterization of the relationship between ECR and patterns of ABL. The assessment of ABL in teeth with ECR revealed a predominance on the palatal/lingual surface. This study is the first to evaluate the spatial relationship between ECR and the distribution of ABL, whereas previous studies have primarily described the morphological characteristics of the resorptive lesion [3, 10, 15]. Including ABL as a periodontal parameter expands the understanding of the biological interaction between ECR and the periodontal supporting tissues, providing relevant information. The predominance of lesions with subcrestal extent observed in the present sample was consistent with previous reports on ECR, which demonstrated a higher frequency of lesions in the middle third of the root [14, 22]. This finding suggests that, at the time of diagnosis of ECR, periodontal involvement may already be present, even in the absence of pulp involvement. Furthermore, angular measurements indicated that the crown–apical extent correlated more consistently with the severity of ABL than the circumferential distribution, which showed variable associations across tooth surfaces. This variability supports the multifactorial nature of periodontal involvement and alveolar bone changes, which depend on local anatomical and functional factors, and suggests that two‐dimensional imaging modalities may underestimate lesion remodeling and periodontal involvement [18]. CBCT has been established as a reliable method for diagnosing and characterizing alveolar bone resorption, as it facilitates the assessment of circumferential lesion spread and the presence of multiple portals of entry, which are often undetectable on two‐dimensional Imaging [3, 18, 23]. In our study, communication with the periodontal width was significantly associated with the subcrestal height of the lesions, suggesting that periodontal involvement occurs at a more advanced stage. A despite this, CBCT demonstrated higher sensitivity and specificity for early detection and therapeutic decision‐making. To the best of our knowledge, it has some limitations. Due to its retrospective design, access to detailed clinical information, including variables such as previous trauma and symptomatology. Although the large image records (n = 9208 patients) and the systematic assessment of three‐dimensional variables, including ABL, represent a novel approach in the literature, they enhance the robustness of the findings and partially mitigate this limitation. Although CBCT provides high‐resolution three‐dimensional imaging, enabling early detection and accurate lesion characterization, it underestimates the extent of ABL at the microscopic level, particularly at the dynamic interface between resorption and repair. Nevertheless, it remains a reliable diagnostic tool compared with conventional imaging modalities. Three‐dimensional analysis revealed an association between bone loss and cervical resorption, accounting for both vertical height and circumferential patterns. We observed an association between ECR characteristics and adjacent ABL, suggesting that periodontal involvement should be considered when evaluating the extent and potential clinical implications of cervical resorptive lesions. The subcrestal extension underscores the role of periodontal attachment in disease severity and prognosis. Future longitudinal studies are warranted to clarify the temporal progression of cervical resorption, determine whether pre‐existing ABL serves as a risk indicator, and support the development of radiological surveillance and early intervention strategies to clarify the practical implications of both conditions. 5. Conclusions We can conclude as follows: •The severity of ECR was associated with ABL, particularly in premolars and molars with increased subcrestal lesion height and wider circumferential extension. The severity of ECR was associated with ABL, particularly in premolars and molars with increased subcrestal lesion height and wider circumferential extension. •ABL related to ECR demonstrated a significant aggravating factor in multifactorial periodontal breakdown. ABL related to ECR demonstrated a significant aggravating factor in multifactorial periodontal breakdown. •Subcrestal extension into the periodontal tissues may contribute to ABL, even in early ECR in the absence of pulp involvement. Subcrestal extension into the periodontal tissues may contribute to ABL, even in early ECR in the absence of pulp involvement. Author Contributions Ana Paula de Medeiros Silva: conceptualization, data curation, formal analysis, investigation, methodology, writing – original draft, writing – review and editing. Thalles Gabriel Germano Lima: conceptualization, formal analysis, investigation, methodology, writing – original draft, writing – review and editing. Ana Beatriz Leitão de Albuquerque Wanderley: conceptualization, investigation, methodology, writing – original draft, writing – review and editing. Monikelly do Carmo Chagas Nascimento: conceptualization, data curation, formal analysis, investigation, methodology, writing – review and editing. Daniela Siqueira Lopes: conceptualization, data curation, formal analysis, investigation, methodology, writing – original draft, writing – review and editing. Funding This study did not receive any specific funding. Disclosure All authors have read and approved the final version of the manuscript. The corresponding author had full access to all of the data in this study and takes complete responsibility for the integrity of the data and the accuracy of the data analysis. Conflicts of Interest The authors declare no conflicts of interest. Supporting Information Additional supporting information can be found online in the Supporting Information section. Supporting information Supporting Information The checklist recommended by the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) checklist is provided as Supporting Information for this cross‐sectional study. Acknowledgments The authors would like to thank Clínica Radiológica Radioface (Recife, PE) for their support, for providing access to the facility, and for granting the imaging data used in this study. Silva, Ana Paula de Medeiros , Lima, Thalles Gabriel Germano , Wanderley, Ana Beatriz Leitão de Albuquerque , Nascimento, Monikelly do Carmo Chagas , Lopes, Daniela Siqueira , Evaluation of Alveolar Bone Loss With ECR Using Cone‐Beam Computed Tomography: A Cross‐Sectional Study, International Journal of Dentistry, 2026, 5186337, 16 pages, 2026. 10.1155/ijod/5186337 Academic Editor: Hannah Wesley Contributor Information Daniela Siqueira Lopes, Email: daniela.siqueira@upe.br. Hannah Wesley, Email: hwesley@wiley.com. Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request. References Associated Data Supplementary Materials Supporting Information The checklist recommended by the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) checklist is provided as Supporting Information for this cross‐sectional study. Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request.