Assessment of relationship between extracranial and intracranial carotid calcifications—a retrospective cone beam computed tomography study

Abstract

Objectives:

Cone beam CT scans in current day dental practice are highly collimated yet involve areas along the course of the extracranial carotid artery. Detecting an extracranial carotid calcification on small volume scans leaves the dentist with two questions: whether the patient is likely to have intracranial carotid calcifications and whether the patient warrants further medical attention. This study aimed to assess the presence of intracranial carotid artery calcifications (ICAC) in the presence of extracranial carotid artery calcifications (ECAC).

Methods:

450 CBCT scans were retrospectively evaluated for ECAC and ICAC. Erby et al’s classification was modified to classify calcifications as mild, moderate, and severe. The presence of ICAC when ECAC were present was evaluated in all three orthogonal planes. The risk of ICAC in the presence of ECAC was calculated as odds ratio and the association between the two was calculated using a χ² test.

Results:

The odds ratio for bilateral ICAC in the presence of bilateral ECAC was 15.09. The odds ratio for left ICAC/right ICAC in the presence of left/ right ECAC was 0.833 and 2.564, respectively. The number and severity of calcifications increased with age. The χ² test showed that there was a strong association (p < 0.001) between bilateral ECAC with bilateral ICAC.

Conclusions:

The results of this group of patients showed that there is an increased presence of ICAC in the presence of ECAC.

Introduction

Carotid calcifications are often detected as radiopacities in the cervical area and are termed as extracranial carotid calcifications. The calcifications along the course of the internal carotid artery in the cranium are termed as intracranial carotid calcifications. There are many different ways carotid calcifications are detected and assessed. These methods include CT, MRI, ultrasound and angiography.^1–3 The incidence and prevalence rates of carotid calcifications vary with race, gender, and age of an individual along with genetic predilection and several lifestyle-related comorbidities.^3–5

Close to a million people suffer from stroke and about 140,000 deaths result each year from stroke in United States annually according to the CDC.⁶ It is the second most common cause of death around the world.⁷ It has also been noted as the most common cause of morbidity that leads to dependent or assisted living in adults.^6,8 The presence of carotid calcifications are thought to reflect the atherosclerotic burden of an individual.³ Furthermore, atherosclerotic disease is a significant cause of morbidity and mortality.⁹ The risk factors for carotid calcification are similar to coronary artery calcifications.⁹These risk factors include hyperlipidemia, smoking, aging, and diseases that affect the microvasculature, such as diabetes, hypertension and chronic kidney disease.^9,10 Intracranial carotid calcifications have been implicated as a cause of ischemic stroke by many studies.^1,4,10–13 They have been shown to be associated with Magnetic Resonance Imaging (MRI) documented lacunar infarcts which in turn are associated with cognitive impairments in individuals.^7,14 Studies have also shown that carotid calcification scores on non-enhanced CT correlate with coronary calcium scores and hence serve as independent predictors of asymptomatic coronary artery disease.⁸ Therefore, any attempt to detect vascular calcifications would mean early detection that could save a patient’s life and ultimately reduce the economic burden towards our healthcare system.

Carotid calcifications are of particular interest to maxillofacial radiologists and dentists as they are often detected as incidental findings on small, medium, and large volume cone beam CT (CBCT) scans. The significance and legal implications of such findings have been discussed by various authors.^15–18 In general, the literature states that carotid calcifications warrant either informal or formal referral to the patient’s physician based on the individual’s cardiovascular risk status.¹⁹ Owing to the small voxel size and higher resolution of CBCT, the calcifications detected are comparable to the bone window of the multidetector CT.¹⁷ Therefore, it is imperative that the dentists identify these calcifications in the CBCT scans of their patients referred for dental therapy. The situation gets further complicated when extracranial carotid calcifications (carotid bifurcation) are detected on small volume CBCT images that do not cover the cranium. Such a finding leaves the dentist with two questions: whether the patient is likely to have intracranial carotid calcifications and whether it warrants further medical attention. Studies have also suggested that the calcium scores in the carotid bifurcation significantly correlate with the stenosis of the internal carotid artery as detected on CT angiography.^20,21 The association between intra- and extracranial calcifications has been established by previous studies.¹⁷ The risk for intracranial carotid calcifications and its possible association to stroke is well studied but internal carotid artery calcifications in the presence of extracranial calcifications has not been adequately studied or understood. In the context of an increasing number of these findings being reported and growing concernsregarding these calcifications, the current study aimed to evaluate the presence of intracranial carotid calcifications in the presence of extracranial carotid calcifications.

Methodology

The present study is a retrospective evaluation of CBCT scans of patients referred to UConn School of Dental Medicine. The study was approved by the institutional ethics review board. Large field of view CBCT scans [140 (Length) x 100 (Height) mm] with areas of intracranial segments of internal carotid arteries and cervical segments of carotid arteries were included in the study. The scans with motion and metallic artifacts, age groups below 40, and age groups above 85 years were excluded. The age range was selected based on the higher prevalence of carotid calcification in this group. In general, large volume scans are acquired for multiple implant placements in both jaws as per the imaging protocol of the institute. Other reasons for large volume scans include adult orthodontics, evaluation of jaw pathologies, and evaluation of temporomandibular joints.

The scans were performed at 90 kVp, 7 mA, and 17.5 seconds. The scans were de-identified by a research assistant prior to initiation of the study. A single operator (SM) evaluated all the scans. To calculate inter-rater reliability, an experienced board-certified radiologist (AT) evaluated a sample of 100 scans and a consensus via discussion of the scoring was established. The number of the scans to establish this correlation was decided after consultation with a statistician. While true inter-rater reliability is obtained by evaluating the whole sample by another observer, it was suggested that more than 20% of the total sample be used for inter-rater reliability based on the expected agreement and variance. The researchers were blinded to the patient records such as medical and dental history, as well as clinical and laboratory findings. The age and gender of the patient was available for investigators. A total of 1000 CBCT scans were screened for the above selection criteria and 450 CBCT scans were selected for the study. The scans were imported into Invivo (Anatomage, SA, CA), a third-party CBCT reconstruction software, for evaluation.

The scans were evaluated for the presence of internal and external carotid artery calcifications. The calcifications were evaluated in axial, coronal and sagittal planes. The slice thickness was set to 0.1 mm and maximum intensity projections and three-dimensional reconstruction images were used when necessary. The vascular calcifications were identified as described in the literature by Damaskos et al.²² The calcifications of the cavernous and carotid segments of the internal carotid artery were evaluated. The cavernous segments of the vessel were identified as calcifications lateral to pituitary fossa and extending from anterior to posterior clinoid processes. The carotid segment calcifications were identified as calcifications within the carotid canal. The extracranial carotid artery calcifications were identified as calcifications in the cervical soft tissues lateral to the anterior tubercle of the transverse processes and posterolateral to the greater cornu of the hyoid bone and pharyngeal airway. The detailed anatomic scheme of vascular structures and calcifications are shown in Table 1. As described in the table, all three slices (axial, coronal and sagittal) were used to detect the calcifications. The calcifications were then classified as mild, moderate, and severe (Figures 1–6) based on Erbay et al’s (Table 2) scoring system¹ for CT. A sample of 25 scans was initially evaluated by both of the observers jointly to reach an agreement on the severity and presence of calcifications. This sample was not used in the study later for analysis.

Table 1.

Radiographic criteria for identifying vascular calcifications

Vessel	Radiographic slices	Site for identifying the calcifications
External carotid artery	Axial	Calcifications are noted in the cervical soft tissues anterolateral to the anterior tubercle of the transverse process. Lateral or posterolateral to the greater cornu of the hyoid almost always posterolateral to the airway.
	Coronal	Posterolateral to anterior tubercle of cervical vertebrae.
	Sagittal	Inferomedial to angle of the mandible and anterolateral cervical tubercle. The calcifications are located at the levels of C-3 to C-5 vertebrae and thyroid cartilage.
Internal carotid artery	Axial	Lateral to hypophysial fossa and between the clinoid processes. Also visualized in the carotid canals.
	Coronal	The calcifications are visualized posterolateral, anterolateral and anteromedial to the anterior clinoid processes. The coronal sections allows the visualization of S-shaped curve of cavernous segment.
	Sagittal	Within the carotid canal in the petrous temporal bone to foramen lacerum, medial to anterior clinoid processes.

Figure 1.

Axial CBCT view depicting mild extracranial carotid calcification. CBCT,cone beam CT.

Figure 2.

Axial CBCT view depicting moderate extracranial carotid calcification. CBCT,cone beam CT.

Figure 3.

Axial CBCT view depicting severe extracranial carotid calcification. CBCT, cone beam CT.

Figure 4.

Axial CBCT view depicting mild intracranial carotid calcification. CBCT, cone beam CT.

Figure 5.

Axial CBCT view depicting moderate intracranial carotid calcification. CBCT, cone beam CT.

Figure 6.

Axial CBCT view depicting severe intracranial carotid calcification. CBCT, cone beam CT.

Table 2.

Modified Erbay’s scoring system for vascular calcifications

Current study	Erbay et al scores	Definition
None	1	No calcifications
Mild	2	Tiny, dispersed calcifications
Moderate	3	Thick discontinuous calcifications, or thin continuous calcifications
Severe	4	Thick continuous calcifications

The calcifications were categorized as left, right and bilateral for both extra and intracranial carotid calcifications. The sample of 100 scans was re-evaluated 6 weeks after the study by (SM) to calculate intra-rater reliability through κ statistics. Descriptive statistics were done to assess the distribution of calcification among different age groups and genders. The risk for intracranial carotid calcifications in the presence of extracranial carotid calcifications was calculated as odds ratio. The risk was calculated for each corresponding vessel (e.g. Odds for left intracranial in the presence of left extracranial) and combination of vessels [e.g. Odds for any (left, right or bilateral) intracranial vessel calcification in the presence of left extracranial]. The associations between the vessel calcifications were assessed using the χ² test of independence. The severity of calcification and absence of calcification in each vessel were plotted against the age of an individual to analyze age-related calcifications.

Results

There were 255 (56.7%) females and 195 (43.3%) male patients in the study. The gender distribution of both intra- and extracranial carotid calcifications have been depicted in Tables 3 and 4. Among 450 patients, 171 (38%) had extracranial carotid calcifications which comprised of 78 (40%) males and 93 (36.5%) females. A total of 197 (43.8%) patients had intracranial carotid artery calcifications which consisted of 88 (45.1%) males and 109 (42.7%) females. Overall, 127 (28.22%) patients had at least one of the combinations of both intra and extracranial carotid calcifications. As indicated in the methodology, the age of the patients ranged between 40 and 85 years. The agewise distribution of intra- and extracranial carotid calcifications have been depicted in Tables 5 and 6. It is evident from the table that the percentage of patients having calcifications increases as the age group of the patient increases.

Table 3.

Gender distribution of extracranial carotid calcification

		EBilateral	ELeft	ERight	Total	None	Total
Gender	Females	40	23	30	93 (36.5)	162	255
	Males	40	16	22	78 (40)	117	195
Total		80	39	52	171 (38)	279	450

Values in the parenthesis indicate percentages

Table 4.

Gender distribution of intracranial carotid calcification

		IBilateral	ILeft	IRight	Total	None	Total
Gender	Females	81	11	17	109 (42.7)	146	255
	Males	64	16	8	88 (45.1)	107	195
Total		145	27	25	197 (43.8)	253	450

ICA, intracranial(internal) carotid artery.

Values in the parenthesis indicate percentages

Table 5.

Age distribution of extracranial carotid calcification

		EBilateral	ELeft	ERight	Total	None	Total
Age	41–50	0	3	3	6 (13.6)	38	44
	51–60	18	7	8	33 (27.5)	87	120
	61–70	18	18	28	64 (37.9)	105	169
	71–85	44	11	13	68 (58.1)	49	117
Total		80	39	52	171 (38)	279	450

Values in parenthesis indicate percentages

Table 6.

Age distribution of intracranial carotid calcification

		IBilateral	ILeft	IRight	Total	None	Total
Age	41–50	4	1	2	7 (15.9)	37	44
	51–60	21	6	7	34 (28.3)	86	120
	61–70	49	15	12	76 (44.9)	93	169
	71–85	71	5	4	80 (68.4)	37	117
Total		145	27	25	197 (43.8)	253	450

Values in parenthesis indicate percentages

The κ values for intra-rater reliability for the different types of vascular calcifications ranged from 0.80 to 0.91, indicating a good intra rater reliability. The κ values for inter-rater reliability ranged from 0.79 to 0.86, indicating a good inter-rater reliability. The majority of disagreement was seen in the mild category of the calcifications, while there was near total agreement for the other two categories. The odds ratios (Table 7) for bilateral intracranial carotid calcifications and for any of the combinations of intracranial carotid vessels in the presence of bilateral extracranial carotid calcifications relative to no vessel calcification was 15.9 [confidence interval (CI) 8.056–28.26] and 9.382 (CI 5.171–17.021), respectively. The odds ratios for left intracranial carotid calcifications and for any of the combinations of intracranial carotid vessels in the presence of left extracranial carotid calcifications relative to no vessel calcification was 0.833 (CI 103–6.71) and 3.95 (CI 2.00–7.80), respectively. The odds ratios for right intracranial carotid calcifications and for any of the combinations of intracranial carotid vessels in the presence of right extracranial carotid calcifications relative to no vessel calcification was 2.56 (CI 668–9.84) and 7.59 (CI 3.84–14.98), respectively. The calcifications of both extra- and intracranial calcifications were classified as none, mild, moderate, or severe. These classifications were plotted against the age of the patients (Figures 7–10). The results of the χ² test of significance are depicted in Table 7 against each vessel. There was a strong association (p < 0.001) between bilateral vessel calcifications in the external and internal carotid vessels. Similarly, there was a strong association between bilateral, right and left external carotid calcifications to at least one of the other calcifications (p < 0.001). Such an association suggests that it is more likely a clinician will detect intracranial carotid calcifications if calcifications are detected in either one or both extracranial carotid vessels. No significant association (p > 0.05) was present between left and right external carotid vessels with corresponding internal carotid arteries. It can be noted from the figures that the severity of calcifications increases with age.

Table 7.

Pearson χ², odds ratio and confidence intervals

	Odds ratio	Upper bound	lower bound	Pearson’s χ2	p-value
EBilteral vs IBilateral	15.090	8.056	28.262	91.762	<0.001
Ebilateral vs any vessel	9.382	5.171	17.021	66.145	<0.001
Eleft vs Ileft	.833	.103	6.719	0.029	>0.05
Eleft vs any vessel	3.956	2.005	7.803	17.382	<0.001
Eright vs Iright	2.564	.668	9.840	2.009	>0.05
Eright vs any vessel	7.595	3.849	14.985	41.641	<0.001

Figure 7.

Age vs severity left external carotid artery calcifications.

Figure 8.

Age vs severity right external carotid artery calcifications.

Figure 9.

Age vs severity right intracranial carotid artery calcifications

Figure 10.

Age vs severity left intracranial carotid artery calcifications

Discussion

The current study evaluated the presence of intracranial carotid calcifications relative to the presence of extracranial carotid calcifications. The findings of this study show that, it is reasonably evident that there is an increased presence of intracranial carotid calcifications in the presence of extracranial carotid calcifications. There was a strong association between bilateral vessel calcifications in external and internal carotid vessels.

Previous studies have reported a prevalence of calcifications as high as 42.88% in extracranial vessels and up to 60.1% in intracranial vessels from analysis of CBCT scans.^3,22 Some authors have reported a 2.4–17% prevalence in intracranial carotid calcifications.^23,24 Lower prevalence is noted when younger patients were evaluated in the study. The prevalence in the current study was similar to Damaskos et al²² since the age group selected was similar.

Aging leads to metabolic and physiologic changes in an individual.²⁵ These changes are reflected in the arterial smooth muscles and activity of inflammatory cells, which in turn act as independent risk factors for arterial sclerosis.²⁵ As discussed in the results, the severity of calcification also increased with the age of the patient. In simple terms, arterial aging leads to changes in the mechanical and structural properties of vascular walls, which reduces arterial compliance.²⁶ There are several systems for visually scoring intracranial carotid calcification on brain CTs.²⁶ These methods include binary, 3-, 4-, and 5-point scoring systems.²⁶ In this study, the authors attempted to adapt the four point visual scoring systems of Erby et al to CBCT. Using the severity of vascular calcifications on CT as an indicator of vascular stenosis is a method that has been established by angiographic studies.¹² The severity of vascular calcifications determined by a CT bone window correlated with carotid siphon stenosis in more than 50% of cases determined by angiography.¹² When calcifications of a severe grade are detected on CBCT, it is possible that they likely indicate vascular stenosis. As suggested by Damaskos et al¹⁷ in their study, we evaluated the intra- and inter-rater reliability of vascular calcifications in the current study. We observed good intra- and inter-rater reliability, which reaffirms that our findings were consistent and reproducible.

In general, male patients had higher prevalence of vascular calcifications compared to female patients. There are no established reasons for such a difference and it is merely observational in scope. However, differences in hormones, lifestyle, and vascular beds have been implicated for such differences.

As described above, the risk for bilateral intracranial vessels increased significantly in the presence of bilateral extracranial vessels calcifications and both had very strong associations (p < 0.001). Bilateral external carotid artery calcifications were noted in 80 scans. In about 59 of these, the patients had bilateral internal carotid artery calcifications. The risk for right intracranial carotid calcifications was also not significantly high in patients with extracranial carotid calcifications and no association was found between the two (p > 0.05). Right external carotid vessel calcification was noted in 52 scans, of which only three scans had right internal carotid artery calcifications. Out of the 52 scans with right external carotid vessel calcification, 39 scans had at least one other type of calcification(s). Damaskos et al also showed that extracranial vessel calcifications correlate significantly with intracranial vessel calcification.¹⁷These findings suggest that the presence of extracranial calcification can be used as an independent risk indicator for the presence of intracranial vessel calcification when other indicators are not available. However, it is difficult to derive a hypothesis from these cross-sectional studies as to why the risk increases or why the correlation exists. The risk for left intracranial vessel calcification was <1, which indicates that the risk is reduced in the presence of left extracranial vessel calcification. Additionally, there was no association between the two vessel calcifications. About 39 scans showed left external carotid artery calcifications and only 1 scan among these showed internal carotid vessel calcification. Among the 39 scans, only 25 showed at least one other type of calcification. Such a finding could be due to a small sample size and lesser prevalence of both left extra (22%) and intracranial vessel calcification (13%) in the study sample. Another important finding is the increased risk of contralateral and bilateral intracranial vessel calcification in the presence of single or bilateral extracranial vessel calcification. This finding suggests that vascular calcifications do not follow any specific patterns.

A good way to take this concept forward would be to include patients with a lesser risk of carotid calcification (typically under the age of 40 years) to assess the true prevalence of vascular calcifications. Furthermore, medical histories of patients would be conducive to a more holistic understanding of the patient.

In conclusion, given the limitations of this study, it is reasonable to conclude that the presence of intracranial carotid calcifications increases when extracranial calcifications are detected. Typically, the volumes acquired in dental practice are highly collimated but often cover the extracranial carotid vessel area. When calcifications are detected in extracranial carotid vessels, the clinician should inform the patient that the presence of an intracranial carotid calcification needs to be ruled out. There is clear evidence in the literature that detecting an intracranial carotid calcification on a scan of an individual with cardiovascular risk factors definitely warrants a medical referral. The person who requests the scan is responsible for the findings noted in the entire volume irrespective of the clinical concern. Hence, practitioners should be trained to identify and report such calcifications when noted or refer the scans to be read by an oral and maxillofacial radiologist for a comprehensive evaluation of the scan. Future studies that highlight the correlation of patient’s medical history with the presence of calcifications on CBCT are necessary to understand this concept better. Such studies are likely to provide valuable insights into true risk estimates of more complex medical conditions and the clinical relevance of carotid calcifications.