Vascular-related cone-beam computed tomographic findings in healthy and medically compromised patients: A study based on self-reported medical history data

Abstract

Purpose

This study aimed to evaluate the correlation between incidental vascular calcification—like imaging findings and self-reported medical data, as well as to assess the relationship between reported predisposing factors and imaging findings using cone-beam computed tomography (CBCT) data.

Materials and Methods

A total of 391 CBCT scans from 188 males and 203 females were anonymously analyzed for the presence of extra- and intra-cranial carotid artery calcifications (ECAC and ICAC, respectively) and signs of Mönckeberg medial sclerosis (MMS). The patients were categorized into 4 groups based on their self-reported medical histories. Descriptive statistics were used to evaluate the data, which were subsequently validated through simple univariate logistic regression analysis.

Results

Among the 391 CBCT scans reviewed, 23.27% exhibited ECAC, 42.71% demonstrated ICAC, and 1.8% showed MMS. Statistical analysis revealed a significant correlation (P<0.05) between both ECAC and ICAC and self-reported predisposing factors—including hypertension, cardiovascular disease, dyslipidemia, diabetes mellitus, and sleep apnea/chronic obstructive pulmonary disease—with notable differences among the study categories (P<0.05). In addition, a strong correlation (P<0.001) was found between the presence of ECAC, ICAC, and MMS and increasing age. Men were significantly more susceptible to ECAC than women (P<0.05).

Conclusion

These findings underscore the importance of a thorough pre-treatment medical history assessment in dental patients, particularly when vascular calcification—like signs are observed on CBCT imaging.

Introduction

Previous studies have shown that incidental vascular calcifications are detectable on commonly used dental imaging modalities, including periapical and panoramic radiographs (PRs) as well as cone-beam computed tomography (CBCT), in the form of atherosclerosis or other types of arteriosclerosis.1,2 These terms originate from Greek and refer to the thickening of the intimal and medial layers of arteries, respectively.

Clinically, vascular calcifications can occur in nearly all arterial beds, affecting both the medial and intimal layers.3 In the tunica intima, atherosclerosis is characterized by the formation of fibrofatty plaques composed of apolipoprotein B-containing lipoproteins, with low-density lipoprotein typically being the most prevalent form.3,4 Although calcification was traditionally considered a late-stage event, recent evidence indicates that atherosclerosis may present as a circumferential lesion—with an unobstructed lumen—in which calcification occurs earlier in the disease process.5

Atherosclerosis is typically associated with plaque formation driven by lipoproteins and cytokines that trigger inflammatory responses.4 In contrast, arteriosclerosis comprises a group of pathological conditions with various etiologies—primarily stemming from systemic and local disturbances in calcium and phosphate metabolism, glucose metabolism, and possibly elastin degeneration—that ultimately lead to calcification of the tunica media.6 Mönckeberg medial sclerosis, the most common variant, is usually localized to the arteries of the extremities (with diameters of at least 0.5 mm)7 and is frequently linked with type 2 diabetes mellitus and end-stage renal disease.6 Moreover, both forms of vascular calcifications tend to be more prominent in individuals with traditional Framingham risk factors—such as sex, age, levels of total and high-density lipoprotein cholesterol, blood pressure, cigarette use, and diabetes mellitus—which are used in various prediction models (e.g., Framingham risk prediction models, Heart-Score from the European Association of Preventive Cardiology) to estimate a patient's future risk of coronary artery disease and stroke.8,9,10

As early as 1981, dentists recognized the presence of vascular calcifications, particularly in the region of the carotid bifurcation, and their potential association with stroke.9 Their prevalence on PRs ranges from 2% to 5.7%,11,12,13 while on CBCT it ranges from 10.41% to 30.99%.14,15 Furthermore, research has demonstrated a correlation between vascular calcifications detected on PRs and carotid stenosis, suggesting that these incidental findings should not be overlooked given their potential role in stroke prevention.16,17 Additionally, the detection of vascular calcifications in both the extra- and intra-cranial segments of the carotid artery has prompted further investigation, as significant correlations between these regions have been observed.18 Studies have also correlated these radiographic findings with traditional Framingham risk factors,19,20 and associations with diabetes mellitus19 and sleep apnea21 have been documented.

In daily dental practice, dentists are required to review patients' medical records to identify diseases that may affect treatment planning and rehabilitation.22 Therefore, when evaluating patients' self-reported medical histories in conjunction with incidental hyperdense signs resembling vascular calcifications on PRs and/or CBCT scans, these findings may be anticipated rather than regarded as purely incidental.

Accordingly, the present study aimed to determine whether a correlation exists between incidental vascular calcification-like hyperdense imaging findings and self-reported medical data, and to assess the relationship between reported predisposing factors and these imaging findings using CBCT data.

Materials and Methods

Study sample

A total of 526 consecutive large field-of-view (FOV) CBCT scans (15×12 cm and 15×15 cm) from Caucasian patients of both sexes, obtained between May 8, 2017, and November 22, 2022, were initially included in this study. These patients were referred for clinical and radiographic evaluations—including PRs and periapical radiographs when needed, in accordance with the authors' clinic's protocol—for implant placement, temporomandibular joint and tooth impaction assessments, and evaluations of orthodontic and cranio-maxillofacial deformities at the Department of Oral Diagnosis and Radiology, School of Dentistry, National and Kapodistrian University of Athens, Greece. Given the clinical indications, CBCT examinations were performed, and the resulting imaging data were retrospectively analyzed in relation to the patients' self-reported medical records. Of the 526 scans, 135 were excluded due to poor diagnostic quality (e.g., beam hardening, excessive motion artifacts) or incomplete medical records (Fig. 1). Ultimately, 391 CBCT scans from 188 male and 203 female patients (ages 9–92 years; average 51.3, median 55, mode 65) that met the inclusion criteria were coded along with their corresponding self-reported medical data for further evaluation.

Fig. 1. Eligibility criteria. CBCT: cone-beam computed tomography.

This research protocol was approved by the Ethics and Research Committee of the Department of Dentistry, School of Health Sciences, National and Kapodistrian University of Athens, Greece (protocol number 627/20.02.2024). All participants provided written consent for the use of their personal data for research purposes.

CBCT imaging

All CBCT scans were acquired using a Newtom VGi Dental Volumetric Tomograph (QR, Cefla, Verona, Italy; serial No. VG17004S) at the Department of Oral Diagnosis and Radiology, School of Dentistry, National and Kapodistrian University of Athens, Greece. The imaging parameters included a fixed focal spot of 0.3 mm and a preset kilovoltage peak of 110 kV. The milliampere setting was automatically adjusted using SafeBeam technology, which optimizes milliamperage based on the density of the irradiated volume. The exposure time was 3.6 seconds, and the voxel size was 0.3 mm³ for both standard resolution and enhanced scans. The FOV was either 15×12 cm or 15×15 cm, depending on the examination indication. Images were reviewed on a FlexScan MX210 color LCD monitor under dim lighting conditions, using NNT software version 7.2 (installation package 7.2.0; Newtom-cefla S.C., Bologna, Italy).

Image and medical history evaluation

The image and medical history evaluation was carried out in 3 steps.

The first step was image evaluation: The 391 CBCT scans were meticulously examined for incidental findings (e.g., related to the nose, paranasal sinuses, temporomandibular joint, endocranium, mouth floor, cervical spine, and neck). Among these findings, the presence of calcifications along the course of the carotid artery—whether extra- or intra-cranial (ECACs and ICACs, respectively)—as well as calcifications within the lumen of the extra-cranial branches (including the lingual, facial, maxillary, and superficial temporal arteries) consistent with Mönckeberg medial sclerosis (MMS) were recorded individually for each scan. The volumetric data sets were analyzed in consensus by 2 experienced oral and maxillofacial radiologists (SD and AZ), who were blinded to the patients' social, medical, and dental histories, thus minimizing any potential bias introduced by prior information. A calcification was considered present if it appeared in at least 3 sequential slices (axial, coronal, and/or sagittal)—equivalent to 3 times the thickness of the axial slices (3.0×0.3 mm=0.9 mm)—to minimize false-positive findings due to image noise or artifacts.15

The image analysis was standardized by evaluating all multiplanar reconstructions, and a maximum intensity projection protocol was employed to facilitate 3-dimensional localization and assess the extent of the findings.

The imaging criteria for assessing ECACs, ICACs, and MMS signs were as follows: a) ECACs: On axial projections, ECACs appear as single or multiple “rice grain” opacities with homogeneous density, arranged in either linear or curvilinear patterns. They are typically located within the cervical soft tissue—approximately 0–10 mm anterolateral to the anterior tubercle of the transverse process; lateral, and more often posterolateral, to the greater cornu of the hyoid bone; and consistently posterolateral to the pharyngeal airway space.18,23 On coronal projections, these calcifications are visualized lateral to the anterior tubercle of the cervical vertebrae, while in sagittal sections, they are identified as positioned medial and inferior to the angle of the mandible and anterolateral to the cervical tubercle, with a vertical distribution ranging from C3 to C5 (Fig. 2).18,23 b) ICACs: The intra-cranial segment of the internal carotid artery, extending from the ascending portion of the petrous segment to the cavernous portion, was thoroughly evaluated using multiplanar reconstructions. This assessment began along the petrous portion of the carotid canal in the temporal bone, continued through the lacerum segment, and extended into the adjacent cavernous portion—where the artery ascends toward the posterior clinoid process alongside the body of the sphenoid bone before curving upward on the medial side of the anterior clinoid process.18,24 This S-shaped curvature, known as the “carotid siphon,” results in a characteristic oblique ring-like or “figure eight” (8-shaped) radiopacity in coronal projections (Fig. 3).15 c) MMS: On multiplanar, 3D, and panoramic reconstructions, MMS is identified as multiple tortuous vascular calcifications within the soft tissues of the head and neck. These calcifications display a radiopaque “rail track” or “pipeline” pattern, reflecting linear calcifications along the extra-cranial branches of the carotid artery (Fig. 4).25,26,27

Fig. 2. Cone-beam computed tomographic images of extra-cranial carotid artery calcifications are seen in axial (A), coronal (B), and sagittal (C) projections (white arrows).

Fig. 3. Cone-beam computed tomographic images of intra-cranial carotid artery calcifications are seen in axial (A), coronal (B), and sagittal (C) projections (white arrows).

Fig. 4. Cone-beam computed tomographic images of Mönckeberg medial sclerosis signs are seen in the lingual artery in axial (A), coronal (B), and sagittal (C) projections (white arrows) coexistence with extra-cranial carotid artery calcifications (dotted arrow).

The second step involved evaluating patients' medical histories. Self-reported medical data from the 391 patients' records were categorized as follows: a) patients with no significant medical history (N=112- category 1); b) patients with diseases unrelated to the designated predisposing factors or malignancies, or who had undergone surgery for any reason (e.g., sinusitis, anemia, thyroid disorders, prostate hyperplasia, etc.) (N=75 - category 2); c) patients with predisposing factors for atherosclerosis development (e.g., hypertension, cardiovascular disease [CVD], dyslipidemia, diabetes mellitus, sleep apnea/chronic obstructive pulmonary disease [COPD]) (N=152 - category 3); and d) patients with a history of malignancy (N=52 - category 4). The presence or absence of these conditions was further corroborated by reviewing the medications prescribed (e.g., amlodipine/valsartan for hypertension; atorvastatin for dyslipidemia; fluticasone furoate/vilanterol trifenatate for COPD).

The third step was data matching: The data sets obtained from steps 1 and 2 were then matched using the applied coding system, with the process carried out using worksheets in Microsoft 365 Excel (Microsoft Corp, Redmond, WA, USA).

Statistical analysis

The presence of hyperdense findings consistent with vascular calcifications (ECAC, ICAC, and MMS) was recorded and primarily evaluated using descriptive statistics (frequencies and relative frequencies). Age was categorized into decade-based groups (1–10, 11–20, 21–30, 31–40, 41–50, 51–60, 61–70, 71–80, 81–90, and 91–100 years) for each sex. Simple univariate logistic regression analysis was performed to assess the relationship between individual outcome variables (hyperdense imaging findings within each CBCT data set) and independent variables (sex, age group, and self-reported medical history data). The results are summarized in tables presenting P-values and odds ratios (ORs) with 95% confidence intervals (CIs). P-values<0.05 were considered statistically significant. Descriptive statistics were generated using Microsoft Excel (Microsoft Corporation, Redmond, WA, USA), while the logistic regression analysis was performed using R Studio (R version 4.4.0; 2016, R Foundation for Statistical Computing, Vienna, Austria).

Results

Of the 391 patient records included in this study, 188 were men and 203 were women. Overall, 91 (23.27%) ECACs and 167 (42.71%) ICACs, with either unilateral or bilateral presentation, were recorded. Among men, 54 (28.72%) exhibited ECACs and 82 (43.62%) exhibited ICACs, while among women, 37 (18.23%) showed ECACs and 85 (41.87%) demonstrated ICACs. Additionally, 7 patients (1.8%) presented with hyperdense vascular calcifications that met the imaging characteristics of MMS. The frequencies and relative frequencies of the CBCT imaging findings, categorized according to self-reported medical history, are detailed in Table 1. The discrepancy in the number of patients (7) with MMS compared to the numbers presented in Table 1 is attributable to the presence of multiple hyperdense findings per patient (e.g., 4 versus 6 MMS signs in category 3). Notably, no MMS signs were observed in the maxillary artery within this study cohort. Similarly, the hyperdense findings corresponding to vascular calcifications (ECAC, ICAC, MMS) were meticulously recorded for category 3 patients based on each predisposing factor and associated comorbidity (Table 2).

Table 1. The overall distribution of hyperdense findings within the categories of the total sample (n, f(n)).

Table 2. The distribution of patients according to each self-reported predisposing factor separately and in association with comorbidities vs. cone-beam computed tomography hyperdense imaging findings (N=152, n, f(n)).

COPD: chronic obstructive pulmonary disease

Simple univariate logistic regression analysis demonstrated a strong association (all P<0.05) between the presence of ECAC and ICAC and self-reported predisposing factors (hypertension, CVD, dyslipidemia, diabetes mellitus, sleep apnea/COPD) (Table 3). This association indicates that when ECAC and/or ICAC are present in a CBCT scan, there is a high likelihood that a predisposing factor exists. No results for MMS are presented due to the low number of cases. Moreover, the frequency of ECAC and ICAC in Category 1 patients (with no significant medical history) differed significantly (all P<0.05) from that in Category 3 patients (with predisposing factors). The initial interpretation of this result suggests that ECAC and/or ICAC are typically observed in the category 3 population rather than in healthy individuals, with a similar but less pronounced pattern seen in categories 2 and 4 than in category 3 (Table 4). Finally, the logistic regression analysis revealed strong associations (all P<0.05) between hyperdense imaging findings (ECAC, ICAC, and MMS) and increasing age, with men exhibiting a higher susceptibility to ECAC than women (Table 5).

Table 3. The results of simple univariate logistic regression analysis of hyperdense imaging findings for each risk factor separately in category 3 patients.

CI: confidence interval

Table 4. The simple univariate logistic regression analysis results of the presence of extra- and intra-cranial carotid artery calcifications in category 3 patients with those of the other categories examined.

CI: confidence interval, NS: not significant

Table 5. The results of simple univariate logistic regression analysis of the hyperdense imaging findings according to gender and age in the total sample.

CI: confidence interval, NS: not significant

Discussion

Aiming to provide new insights into the role of medical history in addressing cardiovascular risk factors among dental patients, this study used CBCT datasets to investigate whether an association exists between incidental vascular calcification-like imaging findings—specifically ECAC, ICAC, and MMS—and self-reported medical data from patients' records. It also sought to determine whether self-reported predisposing factors correlate with these hyperdense imaging findings.

ECAC, ICAC, and MMS were recorded across all patient categories, albeit with variable frequency (Table 1). Notably, these hyperdense findings were unexpectedly identified in both category 1 (patients with a free medical history) and category 2 patients (those with diseases other than predisposing factors), with decreasing frequencies in both sexes. The frequency of these findings increases dramatically in category 3 patients (those reporting predisposing factors) and then decreases in category 4 patients (those with a history of malignancy). Overall, 91 ECACs (23.27%) were detected in this study. This frequency falls between that reported by Amarin et al.28 de Onofre et al.,29 and Möst et al.30 (19.65%, 20.07%, and 27.80%, respectively). However, these findings differ significantly from those of 43.60% and 45.25% found by Damaskos et al.15,18 This inconsistency is mainly attributed to the age of the study populations, as most of the aforementioned studies involved patients over 40 years old. This study also documents the age-related prevalence of ECAC (Table 5), showing that these findings are more likely to occur in the elderly. Regardless of age, it is particularly interesting that ECAC relative frequencies varied among the categories from 9.82% to 36.18% (Table 1).

Similarly, a total of 167 ICACs (42.71%) were detected, with relative frequencies varying among the patient categories. Unexpectedly, 22.32% of category 1 patients presented with ICAC, while the frequency increased to 65.13% in category 3 patients. de Onofre et al.29 found an ICAC prevalence of 58.45% in their study, which is comparable to these results. In contrast, Amarin et al.28 found an ICAC prevalence of 17.03%, which does not correspond to the results of the present study. Variations in age distribution among the studies may account for these differences, in addition to the sample characteristics of the populations studied. Furthermore, previous studies15,26 have documented a high prevalence of ICAC, supporting the pertinence of the present results. It is also noteworthy that ICAC can occur in patients who are unaware of any predisposing factors; the same can be said of ECAC. Studies have highlighted the clinical significance of these findings, especially because atherosclerosis—a systemic process—can begin as early as the second or third decade of life, and its diffuse nature may affect other vascular beds, such as the coronary arteries.5,15,18,24,31

One might question the authors' choice to investigate the correlation between these incidentally detected vascular calcifications and self-reported medical data, particularly regarding predisposing factors for atherosclerosis development. However, the results (Tables 2 and 3) support this investigation. Moreover, it is intriguing that patients who did not report these factors still exhibited significant vascular calcification-like findings when compared with category 3 (Table 4). These imaging findings in CBCT datasets support a bidirectional approach that suggests the need for a more accurate reassessment of patients' medical histories.

Therefore, dentists should be made aware of the significance of these imaging findings, as they enhance the early detection of atherosclerosis. Recognizing these findings may create novel opportunities for primary prevention—through lifestyle modifications or drug therapy—especially in patients at high cardiovascular risk.

Regarding MMS, this finding was detected in only 7 patients, with multi-site deposition observed in 9 vascular beds. Despite its rarity, MMS is included among the pathological conditions that initiate the vascular calcification process and is associated with serious underlying conditions, such as end-stage renal disease stemming from calcium and phosphate metabolism disorders.6 Interestingly, 1 patient in category 4 exhibited these findings (Table 2), which is likely related to his history of malignancy and associated treatments.32 The small number of MMS cases led us to exclude this parameter from further evaluation due to the non-significant results when assessing its association with gender and age (Table 5).

As mentioned above, age was found to play a very important role in the detection of both ECAC and ICAC (all P<0.05). This indicates that these findings are more common in older adults. The age-related prevalence of these hyperdense findings is well documented in numerous studies3,8,13,24,26,28 and was thoroughly investigated by age group in the authors' previous work.15

In the present study, the age span was broad due to this study's aim of examining the presence of vascular calcification-like findings in the general population in relation to self-reported medical records, rather than focusing on younger patients, in whom these findings are rare. It has been documented that patients over the age of 65 are 5.01 times more likely to present with vascular pathology than those aged 41–65, and 13.39 times more likely than those aged 16–40.33 The results are thus of particular interest in the context of advanced age, serving as a warning for the presence of these vascular calcifications.

Similar, though not identical, results were observed for sex. Men were found to exhibit ECAC more often than women (P<0.05), while no statistically significant difference was observed for ICAC (P>0.05). This suggests that ICAC depiction does not differ between the sexes in the patient categories studied. The findings related to ECAC align with previous studies.15,34 The absence of a sex difference for ICAC may be attributed to the intrinsic role of hormones and tissue factors in maintaining vascular endothelial function in females.34

Furthermore, this study focused on self-reported predisposing factors and their correlation with ECAC and ICAC. Strong associations were observed between these factors-namely hypertension, CVD, dyslipidemia, diabetes mellitus, and sleep apnea/COPD-and the presence of ECAC and ICAC (all P<0.05) (Table 4). These results are consistent with the extensively documented role of these predisposing factors in the development of atherosclerosis.5,6,7,8,13,24,26,28 In particular, studies have shown that patients with sleep apnea and COPD share common inflammatory mediators (e.g., TGFβ, LTB4, and MMP-9) with atherosclerosis, as well as oxidative stress, reduced antioxidant capacities, and leukotriene activation.35,36

As previously noted, this association suggests that when an ECAC and/or ICAC is evident in a CBCT dataset, there is a high probability that a predisposing factor is present. Notably, in category 4 patients (those with a history of malignancy), the findings of this study showed that the presence of ECAC was not significantly correlated with that in category 3 patients (P>0.05), whereas ICAC was (P<0.05). It is well documented that malignancy and its treatments can contribute to the progression of atherosclerosis, with the inflammatory microenvironment exerting profound effects on malignancy development and outcomes.32 Interestingly, malignancy and atherosclerosis share common risk factors, and cancer patients often exhibit metabolic abnormalities. Moreover, the proliferation of cancer cells combined with the apoptosis of normal cells disrupts homeostasis, leading to the development of atherosclerosis.37

At the same time, age, sex, and other predisposing factors (e.g., obesity, menopause, smoking) that were not evaluated in the present study should be co-evaluated.8 In particular, smoking was not evaluated because data regarding tobacco use varied within this cohort (e.g., rolling tobacco, vaping/e-cigarettes), and the time of cessation was unclear. Considering the above, the presence of these hyperdense findings in an imaging modality used to screen dental patients—primarily PR, and CBCT when needed—should prompt clinicians to thoroughly examine or refer their patients for further evaluation, even in cases where the medical history appears unremarkable. This rationale similarly applies to patients with medical issues other than the predisposing factors, making it essential to rule out any underlying conditions.

One might argue that this study did not include factors such as periodontitis or missing teeth, as other studies have.29 However, the vast majority of these conditions are of inflammatory origin and, apart from the traditional risk factors, inflammation is recognized as a “non-traditional risk factor” for atherosclerosis and vascular disease.38 Similarly, periodontal disease, an inflammatory condition, has been causally linked to the development and progression of common chronic diseases such as diabetes mellitus, obesity, and CVD, through mechanisms involving low-grade inflammation that is considered a “silent risk factor.”39,40

A limitation of this study is the reliance on self-reported medical history data, which may not always be accurate. However, Okura et al.41 found good agreement between factors with medical record diagnoses for diabetes, hypertension, stroke, and myocardial infarction (Kappa: 0.76, 0.75, 0.80, and 0.71, respectively). They also noted that this agreement was enhanced by younger age, female gender, higher educational levels, and the absence of comorbidities. Moreover, Smith et al.,42 in a study enrolling 37,798 participants, demonstrated that medical record diagnoses generally agreed with self-reported conditions, although medical records more accurately reflected the absence rather than the presence of disease. They argued that researchers relying on self-reported conditions should, when possible, use multiple data sources—particularly in young or healthy populations. The present study addressed this issue by integrating data on medication use with medical record results, achieving an overall agreement of up to 99.2% for coronary artery disease/bypass surgery.43 To maintain the accuracy of the results, 120 CBCT scans were excluded due to incomplete medical data (Fig. 1). Thus, enrolling only patients who met the stringent criteria (i.e., complete medical records and corresponding prescribed medications) helped reduce “noise” in the results. Other studies based on self-reported medical data have reported higher rates of conditions, primarily hypertension and diabetes mellitus, in dental populations treated for various reasons, underscoring the need to advance dental education by adopting a more medically oriented approach.44,45,46

Another limitation of this study is the absence of multivariate analysis. Future research should incorporate multivariate analysis to better control for confounding factors and provide a more comprehensive understanding of the associations between CBCT-detected vascular calcifications and self-reported medical history.

However, this study was structured to emphasize the crucial role of evaluating medical history data during dental treatment planning, particularly for patients presenting with vascular calcifications on CBCT imaging. The authors anticipate that the outcomes of this study will guide clinicians in identifying dental patients who may benefit from further medical or cardiological evaluation and screening—such as color duplex ultrasonography—thereby reducing both the short- and long-term complications associated with their treatment plans.