Calcium quantity in carotid plaques: detection in panoramic radiographs and association with degree of stenosis

Detection of extracranial and intracranial calcified carotid artery atheromas in cone beam computed tomography using a deep learning convolutional neural network image segmentation approach

OBJECTIVE

We leveraged an artificial intelligence deep-learning convolutional neural network (DL CNN) to detect calcified carotid artery atheromas (CCAAs) on cone beam computed tomography (CBCT) images.

STUDY DESIGN

We obtained 137 full-volume CBCT scans with previously diagnosed CCAAs. The DL model was trained on 170 single axial CBCT slices, 90 with extracranial CCAAs and 80 with intracranial CCAAs. A board-certified oral and maxillofacial radiologist confirmed the presence of each CCAA. Transfer learning through a U-Net-based CNN architecture was utilized. Data allocation was 60% training, 10% validation, and 30% testing. We determined the accuracy of the DL model in detecting CCAA by calculating the mean training and validation accuracy and the area under the receiver operating characteristic curve (AUC). We reserved 5 randomly selected unseen full CBCT volumes for final testing.

RESULTS

The mean training and validation accuracy of the model in detecting extracranial CCAAs was 92% and 82%, respectively, and the AUC was 0.84 with 1.0 sensitivity and 0.69 specificity. The mean training and validation accuracy in detecting intracranial CCAAs was 61% and 70%, respectively, and the AUC was 0.5 with 0.93 sensitivity and 0.08 specificity. Testing of full-volume scans yielded an AUC of 0.72 and 0.55 for extracranial and intracranial CCAAs, respectively.

CONCLUSION

Our DL model showed excellent discrimination in detecting extracranial CCAAs on axial CBCT images and acceptable discrimination on full-volumes but poor discrimination in detecting intracranial CCAAs, for which further research is required.

Defined shapes of carotid artery calcifications on panoramic radiographs correlate with specific signs of cardiovascular disease on ultrasound examination

OBJECTIVE

The aim was to optimize diagnostics for carotid artery calcifications (CACs) on panoramic radiographs (PRs) to identify cardiovascular disease (CVD) by investigating how 4 defined CAC shapes are associated with ultrasound (US) findings indicating CVD.

STUDY DESIGN

The study included 414 participants (802 neck sides) from the Malmö Offspring Dental Study, examined with PRs. The PRs were assessed for CAC shapes stratified into 4 categories: single, scattered, vessel-width defining, and vessel-outlining. The carotid arteries were examined with US for signs of CVD: the presence of plaques, largest individual area of a plaque, number of plaques, and percentage reduction of the lumen. Associations between the different CAC categories and US characteristics were analyzed.

RESULTS

All categories of CAC were significantly associated with a higher degree of US findings indicating CVD compared with no CAC (P < .001). The most significant differences were found for vessel-outlining CAC, with the mean of the largest individual plaque area of 17.9 vs 2.3 mm2, mean number of plaques 1.6 vs 0.2, and mean percentage reduction of the lumen 24.1% vs 3.5% (all P < .001).

CONCLUSIONS

Independent of shape, CACs detected on PRs were associated with a higher degree of US findings of CVD. This was most pronounced for vessel-outlining CAC. With refined differential diagnostics of CACs in PRs, dentists may contribute to improved identification of patients in need of cardiovascular prevention.

Comparison of carotid artery calcification between stroke and nonstroke patients using CT angiographic and panoramic images

OBJECTIVES

This study aimed to analyze the characteristics of carotid artery calcification (CAC) in stroke and nonstroke patients using computed tomography angiographic (CTA) and panoramic images.

METHODS

This is a retrospective study on patients who acquired both CTA and panoramic images at the Neurology Department of Kyungpook National University Hospital, Daegu, South Korea, between 2011 and 2016. The patients were divided into stroke (n = 109) and nonstroke (n = 355) groups based on the final diagnosis. CAC was analyzed in each group based on its presence, shape, and severity using the [Formula: see text]2 test. The differences in age and sex between the two groups were examined using a two-sample t-test. A measure of intraobserver reliability was obtained using Cohen's κ index.

RESULTS

CAC was more frequently observed in the stroke group than in the nonstroke group using both CTA (stroke group, 100%; nonstroke group, 23.1%) and panoramic (stroke group, 83.5%; nonstroke group, 16.6%) images. Although scattered CAC shape and mild severity occupied the largest portion in both groups, vessel-outlined CAC was more common in nonstroke patients than in stroke patients. In age and sex analyses, only females patients in their 70 s showed significant differences in CAC shape between the stroke and nonstroke groups.

CONCLUSIONS

On both CTA and panoramic images, although CAC is found more frequently in the stroke group, vessel-outlined-shaped CAC in the nonstorke group shows significant differences compared to other shapes.

Association of high cardiovascular risk and diabetes with calcified carotid artery atheromas depicted on panoramic radiographs

OBJECTIVE

To evaluate whether estimates of risk of future cardiovascular events and death and established or unknown diabetes are significantly associated with calcified carotid artery atheromas (CCAAs) on panoramic radiographs (PRs). The main focus was on men and women without previous myocardial infarction (MI).

METHODS

The PAROKRANK (Periodontitis and its Relation to Coronary Artery Disease) study included patients with a first MI and matched control subjects. In this substudy, 738 patients (138 women) and 744 control subjects (144 women) with available PRs were assessed for CCAA. Cardiovascular risk estimates were determined according to the Framingham Risk Score (FRS) and Systematic COronary Risk Evaluation (SCORE). Established and previously unknown diabetes was also determined.

RESULTS

CCAA was detected on PRs in 206 control subjects (28%) and 251 patients (34%). FRS was significantly associated with CCAA among control subjects (P = .04) and patients (P = .001). SCORE was associated with CCAA among control subjects (P < .01) but not patients (P = .07). Among men, FRS and SCORE were associated with CCAA in both control subjects and patients. Diabetes was not significantly associated with CCAA after adjustments.

CONCLUSIONS

Elevated cardiovascular risk scores were associated with CCAA on PRs among control subjects. Diabetes was not independently associated with CCAA, possibly owing to selection bias.

Associations among Periodontitis, Calcified Carotid Artery Atheromas, and Risk of Myocardial Infarction

Cardiovascular disease is a common cause of morbidity and premature mortality. Cardiovascular disease can be prevented when risk factors are identified early. Calcified carotid artery atheromas (CCAAs), detected in panoramic radiographs, and periodontitis have both been associated with increased risk of cardiovascular disease. This case-control study aimed to 1) investigate associations between periodontitis and CCAA detected in panoramic radiographs and 2) determine the risk of future myocardial infarctions due to CCAA combined with periodontitis. We evaluated 1,482 participants (738 cases and 744 controls) with periodontitis and CCAAs recruited from the PAROKRANK study (Periodontitis and Its Relation to Coronary Artery Disease). Participants were examined with panoramic radiographs, including the carotid regions. Associations between myocardial infarction and periodontitis combined with CCAA were evaluated in 696 cases and 696 age-, sex-, and residential area–matched controls. Periodontitis was evaluated radiographically (as degree of bone loss) and with a clinical periodontal disease index score (from clinical and radiographic assessments). We found associations between CCAA and clinical periodontal disease index score among cases (odds ratio [OR], 1.51; 95% CI, 1.09 to 2.10; P = 0.02) and controls (OR, 1.70; 95% CI, 1.22 to 2.38; P < 0.01), although not between CCAA and the degree of bone loss. In a multivariable model, myocardial infarction was associated with CCAA combined with periodontitis, as assessed by degree of bone loss (OR, 1.75; 95% CI, 1.11 to 2.74; P = 0.01). When the cohort was stratified by sex, only men showed a significant association between myocardial infarction and CCAA combined with periodontitis. Participants with clinically diagnosed periodontitis exhibited CCAA in panoramic radiographs more often than those without periodontitis, irrespective of the presence of a recent myocardial infarction. Participants with combined periodontitis and CCAA had a higher risk of having had myocardial infarction as compared with participants with either condition alone. These findings implied that patients in dental care might benefit from dentists assessing panoramic radiographs for CCAA—particularly, patients with periodontitis who have not received any preventive measures for cardiovascular disease.

Bilateral vessel-outlining carotid artery calcifications in panoramic radiographs: an independent risk marker for vascular events

Background

In odontology, panoramic radiographs (PRs) are regularly performed. PRs depict the teeth and jaws as well as carotid artery calcifications (CACs). Patients with CACs on PRs have an increased risk of vascular events compared to healthy controls without CACs, but this association is often caused by more vascular events and risk factors at baseline. However, the risk of vascular events has only been analyzed based on the presence of CACs, and not their shape. Thus, this study determined if the shape of CACs in PRs affects the risk of future vascular events.

Methods

The study cohort included 117 consecutive patients with CACs in PRs and 121 age-matched controls without CACs. CAC shape in PRs was dichotomized into bilateral vessel-outlining CACs and other CAC shapes. Participants were followed prospectively for an endpoint of vascular events including myocardial infarction, stroke, and vascular death.

Results

Patients with bilateral vessel-outlining CACs had more previous vascular events than those with other CAC shapes and the healthy controls (p < 0.001, χ²). The mean follow-up duration was 9.5 years. The endpoint was reached in 83 people. Patients with bilateral vessel-outlining CACs had a higher annual risk of vascular events (7.0%) than those with other CAC shapes (4.4%) and the controls (2.6%) (p < 0.001). In multivariate analysis, bilateral vessel-outlining CACs (hazard ratio: 2.2, 95% confidence interval: 1.1–4.5) were independent risk markers for the endpoint.

Conclusions

Findings of bilateral vessel-outlining CACs in PRs are independent risk markers for future vascular events.

Diagnostic accuracy of panoramic radiography in the detection of calcified carotid artery atheroma: a meta-analysis

OBJECTIVE

To determine the diagnostic accuracy of panoramic radiography (PR) in detecting calcified carotid artery atheroma (CCAA) compared with Doppler ultrasonography or angiography (the reference standard).

SOURCES

Cochrane, LILACS, PubMed, Scopus, Web of Science, Google Scholar, Open Grey, and ProQuest were searched. The reference lists of the included studies were also screened.

DATA

Observational studies.

METHODS

Only studies comparing the diagnostic accuracy of PR in detecting CCAA to Doppler ultrasonography or angiography (the reference standard) were included. The primary outcome measures were sensitivity and specificity. The secondary outcomes were negative predictive values, positive predictive values, diagnostic odds ratios, likelihood ratios (positive and negative), receiver operating characteristic curves, accuracy, and Youden's index. Two reviewers independently participated in the study selection, data extraction, and risk of bias assessment without language restriction. Risk of bias was assessed thought QUADAS-2, and the level of evidence was assessed through GRADE.

RESULTS

A total of 773 citations were identified after duplicates were removed, and 12 studies including 1002 patients were included in the final study. The sensitivity and specificity of the different selected studies varied substantially, with sensitivity ranging from 0.31 to 0.95 and specificity from 0.19 to 0.99.

CONCLUSIONS

Most studies reported excellent sensitivity and good specificity. The diagnostic accuracy of PR was good or excellent in 50% of the studies.

CLINICAL SIGNIFICANCE

The identification of CCAA by PR can be a risk predictor for stroke when used as a secondary screening tool.

Association of calcified carotid artery plaque in panoramic images and diagonal earlobe crease

OBJECTIVES:

Brazilians who are free of traditional atherogenic risk factors frequently suffer myocardial infarctions (MI). Calcified carotid artery plaque (CCAP) on panoramic images (PIs) has been shown to be a validated "risk indicator" of future MIs. The diagonal earlobe crease (DELC) is likewise a validated "risk indicator" of future MI. We sought to determine the prevalence of DELC+ among patients having CCAP+ on their PIs.

METHODS:

In a masked cross-sectional study, three dentists evaluated PIs of patients (N = 481) over age 55 for presence of CCAP+. Among the CCAP+ cohort, the presence of DELC+ was determined by two additional dentists masked to the presence of CCAP+ on PIs. A control group (CCAP-) comprised of individuals with PIs devoid of plaque was similarly evaluated for evidence of DELC+.

RESULTS:

A mixed-gender group (8.6%) of patients (N = 41; mean age 67.5 ± 6.8) evidenced CCAP+ on their PIs and approximately 88% of these individuals (N = 36) had concomitant DELC+. Among members of the control group (N = 41), 17 patients were DELC+. Our results demonstrated that among the full mixed-gender cohort (N = 82), the association between CCAP+ and DELC+ was statistically significant (p = 0.0001). However, when adjusting for gender, only the CCAP+ men were significantly (p = 0.00011) more likely to evidence DELC+ on clinical examination.

CONCLUSIONS:

Males with atherosclerosis of their carotid arteries as evidenced by PIs displaying calcified atheromas are significantly more likely to evidence a DELC+ than those without an imaged atheroma.

Calcified carotid artery atheromas in panoramic radiographs are associated with a first myocardial infarction: a case-control study

OBJECTIVE

The aim of this case-control study was to investigate whether patients with a first myocardial infarction (MI) had a higher prevalence of calcified carotid artery atheromas (CCAAs) on panoramic radiographs (PRs) than age-, gender-, and residential area-matched controls without MI.

STUDY DESIGN

Six hundred ninety-six cases with a first MI and 696 controls were included in this substudy of the Swedish multicentre PAROKRANK study. All participants underwent panoramic radiography, and the PRs were evaluated for CCAAs.

RESULTS

The prevalence of CCAAs detected by PR was 33.8% (235 of 696) in cases and 27.6% (192 of 696) in controls (odds ratio [OR] 1.24; 95% confidence interval [CI] 1.04-1.44; P = .012). Among males, 32.7% of cases (184 of 562) and 26.5% of controls (149 of 562) displayed CCAAs on PRs (OR 1.24; 95% CI 1.03-1.48; P = .022). Among both genders, bilateral CCAAs were significantly more common among cases than among controls (P = .002).

CONCLUSIONS

Cases with recent MIs had a significantly higher prevalence of CCAAs on PRs compared with controls without MIs. This difference between groups was more pronounced for bilateral CCAAs. These findings supported the hypothesis that CCAA detection could serve as a risk indicator for future MIs.

Carotid calcification in panoramic radiographs: radiographic appearance and the degree of carotid stenosis

OBJECTIVES:

Approximately 7% of patients examined with panoramic radiographs (PRs) for odontological reasons, and with incidental findings of carotid artery calcification (CAC), have significant (≥50%) carotid stenosis (SCS). The aim of this study was to determine if the radiographic appearance of CACs in PRs could be categorized such that we could improve the positive-predictive value (PPV) for SCS detection.

METHODS:

This cross-sectional study included 278 patients with CACs identified by PRs, 127 with SCS and 151 without SCS. CACs were categorized based on the following appearances: (1) single, (2) scattered or (3) vessel-outlining. Correlates were derived for each category according to positivity or not for SCS in the corresponding neck sides. For a representative adult population (in dentistry) presenting with CACs in PRs, PPVs for SCS prediction were then calculated for three subgroups based on different combinations of appearances (2) and/or (3).

RESULTS:

Vessel-outlining CACs corresponded to 65% of neck sides with SCS vs 47% without SCS (p < 0.001). Single CACs corresponded to 15% of neck sides with SCS vs 27% without SCS (p = 0.006). Scattered CACs corresponded to 20% of neck sides with SCS vs 26% without SCS (p = 0.127). In the representative population, the PPVs for SCS detection were comparable (7.2-7.8%) for all three subgroups and when the presence of a CAC is the sole criterion (7.4%).

CONCLUSIONS:

Stratifying the radiographic appearance of CACs in PRs does not improve the PPV for SCS detection. Whether different radiographic appearances are associated with future risk of stroke, or other cardiovascular events, remains unknown.

Atherosclerotic Calcification Detection: A Comparative Study of Carotid Ultrasound and Cone Beam CT

BACKGROUND AND AIM

Arterial calcification is often detected on ultrasound examination but its diagnostic accuracy is not well validated. The aim of this study was to determine the accuracy of carotid ultrasound B mode findings in detecting atherosclerotic calcification quantified by cone beam computed tomography (CBCT).

METHODS

We analyzed 94 carotid arteries, from 88 patients (mean age 70 ± 7 years, 33% females), who underwent pre-endarterectomy ultrasound examination. Plaques with high echogenic nodules and posterior shadowing were considered calcified. After surgery, the excised plaques were examined using CBCT, from which the calcification volume (mm3) was calculated. In cases with multiple calcifications the largest calcification nodule volume was used to represent the plaque. Carotid artery calcification by the two imaging techniques was compared using conventional correlations.

RESULTS

Carotid ultrasound was highly accurate in detecting the presence of calcification; with a sensitivity of 88.2%. Based on the quartile ranges of calcification volumes measured by CBCT we have divided plaque calcification into four groups: <8; 8-35; 36-70 and >70 mm3. Calcification volumes ≥8 were accurately detectable by ultrasound with a sensitivity of 96%. Of the 21 plaques with <8 mm3 calcification volume; only 13 were detected by ultrasound; resulting in a sensitivity of 62%. There was no difference in the volume of calcification between symptomatic and asymptomatic patients.

CONCLUSION

Carotid ultrasound is highly accurate in detecting the presence of calcified atherosclerotic lesions of volume ≥8 mm3; but less accurate in detecting smaller volume calcified plaques. Further development of ultrasound techniques should allow better detection of early arterial calcification.

Carotid calcifications on panoramic radiographs: a 5-year follow-up study

OBJECTIVE

The aim of this study was to determine whether people with calcifications in the area of the carotid arteries on panoramic radiographs (CALPANs) have an increased prevalence of vascular risk factors or an increased risk of future vascular events.

MATERIALS AND METHODS

We included 113 consecutive patients with CALPANs and 116 gender- and age-matched controls without CALPANs. Vascular risk factors were generally known in the study population, since it was recorded in a population-based community-screening program.

RESULTS

Patients with CALPANs had a higher prevalence of vascular risk factors than controls independent of previous vascular events. During the 5.4-year follow-up, patients with CALPANs had a higher risk of a combined endpoint of vascular events (5.6%/yr) compared with the controls (2.4%/yr) (P = .004 by log rank test; unadjusted hazard ratio (HR) 2.4; 95% CI 1.3-4.3). This difference was not significant when previous vascular events and risk factors were taken into account (adjusted HR; 1.2; 95% CI 0.6-2.3; P = .62; Cox regression).

CONCLUSIONS

People with CALPANs are very likely to have vascular risk factors, but these factors might be unknown when CALPANs are detected. Accordingly, patients with CALPANs should be advised to have their vascular risk factors regularly checked in order to receive advice on preventive lifestyle modifications and medical treatment when indicated. However, it remains unknown whether CALPANs add information about the independent risk of future vascular events. Therefore, further studies are warranted to investigate whether the detection of CALPANs indicates a need for additional or more intense vascular treatment.