Introduction

Carotid atherosclerosis is an important prognostic indicator due to its close relationship to the incidence of stroke. An increasing severity of stenosis is related to a higher risk of stroke1,2. Arterial remodeling occurs as atherosclerosis progresses3. During this process, the artery initially enlarges as plaque accumulates, to ensure an adequate blood flow. This enlargement of the artery is known as expansive remodeling. The geometry of the carotid artery influence hemodynamics, so it has been hypothesized as a risk factor for the initiation of atherosclerosis4,5,6,7,8,9,10. Recent studies have reported that vessel volume, diameter and the bifurcation angle of the carotid artery and internal carotid artery (ICA) increased with age11. Studies have shown that expansive remodeling of the carotid artery correlates with the risk of ischemic stroke12.

According to a population-based study, atherosclerotic plaque in the carotid artery is not necessarily characterized by bilateral symmetry13. Further, plaque in the left carotid artery is more vulnerable than in the right artery; however, the study did not compare the geometry of carotid artery. Therefore, to better understand the clinical implications, we compared both geometrical sides of the carotid artery. Furthermore, to date there are a limited number of longitudinal studies that have assessed the carotid artery geometry14. Most studies of carotid geometry compared young patients with older healthy subjects9. Accordingly, existing studies regarding age-related carotid artery geometry is insufficient. Therefore, we designed a longitudinal study to analyze both the right and left carotid artery diameter and bifurcation angle in each patient and changes that occurred over a decade.

The purpose of our study was to use a longitudinal study method to investigate geometric changes in the carotid artery. We hypothesized that the diameter of the common carotid artery (CCA), ICA, external carotid artery (ECA), and the bifurcation angle would increase over a 10-year period; additionally, we expected that expansive remodeling of the carotid artery would differ between the two sides.

Materials and methods

This study was approved by the Institutional Review Board (IRB) of our institute, and the requirement for informed consent was waived by the IRB due to the retrospective design and imaging analysis approach of the study.

Study population

The study population was taken from the dataset of the Jeonbuk National University Hospital. This present study was designed for the retrospective research that included subjects who underwent magnetic resonance angiography (MRA) at least twice during 10 years from 2008 to 2020, with 300 patients that responded to the baseline surveys. This paper uses a dataset from 263 subjects, who underwent MRA at baseline and after about 10 years after the longitudinal follow-up. The MRAs were performed to assess various clinical issues. The inclusion criteria were as follows: contrast-enhanced MRA examination at baseline and after 10 years with the same MR machine, no interval change in carotid stenosis without plaque, and acceptable imaging quality. Clinical data for the subjects, including basic demographics and risk factors for atherosclerosis, namely diabetes, hypertension, dyslipidemia, current smoking, and history of coronary disease, were also recorded.

MR imaging protocol

MR imaging of the brain and carotid angiography was performed with a 3.0 T scanner (Verio; Siemens, Erlange, Germany) at baseline and after about 10 years. Contrast-enhanced carotid MRA was performed in the coronal plane using a T1-wighted spoiled gradient-echo sequence. The initial parameters of this study were as follows: ratio of repetition time (TR) to time to echo (TE) = 3.8/1.5 ms, flip angle = 30°, slice thickness = 0.90 mm, matrix size = 384 × 180, field of view = 183 × 300 mm, echo train length = 1, average number = 1, and acquisition time = 90 s. The parameters after 10 years were as follows: TR/TE = 3.3/1.2 ms, flip angles = 21°, slice thickness = 1.0 mm, gapless, matrix size = 448 × 280, field of view = 281 × 360 mm, echo train length = 1, number of average = 1, and acquisition time = 90 s.

They assessed image quality by consensus as visualization of vessel wall and degree of venous contamination. Poor imagings with ill-defined margin of carotid artery or no separation between carotid artery and jugular vein due to venous contamination were excluded from the final analysis. Disagreements regarding image quality were resolved by consensus.

Imaging reconstruction

Contrast-enhanced carotid MRA of the carotid artery was performed for all subjects. Both carotid arteries were segmented with semi-automatic software (MIMICS; Materialise, Leuven, Belgium) using coronal source data.

Carotid artery geometry measurements

The geometric measurements for both carotid arteries were initially proposed by Thomas et al.9. Figure 1 presents the imaging reconstruction process and factors for measurement. Using MIMICS software, the arterial central lines were generated individually from the CCA to the ECA and to the ICA. The carotid bifurcation angle using central lines was defined as the angle between the ECA projections and ICA vectors at the carotid artery bifurcation. The diameter and area of the ICA and CCA were measured at a distance of 2 cm from the carotid artery bifurcation. The ECA diameter was measured at a distance of 1 cm from the bifurcation of carotid artery because of the influence of arterial branches. Additionally, the maximum diameter and area of the CCA, including the carotid bulb, were measured. The minimum diameter and area of the ICA were also measured.

Figure 1
figure 1

Geometric parameters of the carotid artery.

Full size image

Statistical analysis

Statistical analysis was performed using SPSS version 24.0 (IBM Corp., Armonk, NY). The normality test was not conducted due to the large size of sample numbers. Geometrical changes of each carotid artery between baseline and after 10 years on MRA were assessed by paired T-tests. Comparison of geometrical parameter of both carotid arteries was performed by paired T-tests. All levels of significance were determined with a p value of < 0.05.

Institutional Review Board Statement

This study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Institutional Review Board of Jeonbuk National Hospital (JUH 2019-10-041).

Informed consent

Our institutional review board approved this prospective study, and patient informed consent was received before the MR examination.

Results

Subjects

We selected three hundred patients who underwent carotid MR angiography during a defined period based on a dataset from our hospital. Of these patients, 36 were excluded from this study due to the use of different MR machines or MRA techniques, and 69 were excluded because of poor imaging quality such as flow artifact or venous contamination for imaging processing using MIMICS. Further, 18 patients were excluded due to carotid stenosis with plaque that was identified during the follow-up period. In total, 177 patients (mean age after 10 years, 70.7 years; 106 females) were included this study (Fig. 2). The basic demographic data of subjects in this study are shown in Table 1. The mean time interval between two MRA examinations was 130.0 ± 9.2 months.

Figure 2
figure 2

Subject selection.

Full size image
Table 1 Subject demographic data (after 10 years).
Full size table

Geometric changes in each carotid artery

The geometric changes in each carotid artery and statistical analysis for the parameters are summarized in Tables 2 and 3. Bilateral bifurcation angles significantly increased over a 10-year period (p < 0.001; right: 47.46 ± 1.26° for baseline vs. 49.9 ± 1.25°, left: 43.98 ± 1.14° for baseline vs. 45.96 ± 1.17°). The diameters and areas of bilateral CCAs significantly increased over a 10-year period (p < 0.001). However, the diameters and areas of bilateral ICAs at the reference site decreased during the follow-up period. Especially, the diameters and areas in the left ICA at the reference site were significantly decreased (p < 0.05), and the diameters and areas of the bilateral ECAs at the reference site were significantly decreased during follow-up period (p < 0.05).

Table 2 Geometric changes in the right carotid artery from baseline and after 10 years of follow-up.
Full size table
Table 3 Geometric changes in the left carotid artery from baseline and after 10 years of follow-up.
Full size table

Geometrical changes in bilateral carotid arteries over 10 years

Comparisons of geometrical changes in the bilateral carotid arteries over 10 years and associated statistical analysis are summarized in Table 4 and Fig. 3. The bifurcation angle of the right carotid artery was significantly increased compared to the left carotid artery (p < 0.001; 2.44 ± 0.53 vs. 1.98 ± 0.55) (Fig. 4). The diameter and area of the left CCA was significantly increased compared to the right CCA (p < 0.001), but the diameter and area of the left ICA at the reference site and the minimum diameter and area of the left ICA were significantly decreased compared to the right side (p < 0.001).

Table 4 Geometric changes in both carotid arteries after a 10-year follow-up period.
Full size table
Figure 3
figure 3

Geometric changes in bilateral carotid arteries over 10 years.

Full size image
Figure 4
figure 4

The representative case (left side) of geometric changes over 10 years.

Full size image

Discussion

In this study, we investigated geometric changes in the bilateral carotid arteries that occurred over the course of a decade and without carotid stenosis. We identified cases with follow-up over a 10-year period, and the bilateral bifurcation angle, diameters of both CCAs and areas of both CCAs all increased significantly. Kamenskiy et al.15 also demonstrated that bulb diameter and bifurcation angle increase in carotid arteries with increasing age. However, we found that the maximum diameter and area of the ICA did not significantly change. We also found that the diameters and areas of the bilateral ICAs at the reference site decreased during the follow-up period. This result is supported by the study by Thomas et al. which reported that old vessels exhibited greater bifurcation angles, as well as smaller ECA/CCA and ICA/CCA diameter ratios9. The results of this study indicated that the bifurcation angle of the right carotid artery was significantly increased compared to the left carotid artery; additionally, the diameter and area of the left CCA were significantly increased compared to the right CCA.

Arteries are frequently exposed to radial, circumferential, and longitudinal forces that are exerted on their wall or as shear stress on the endothelial surface16. With age, vascular smooth muscle cells undergo functional changes that alter the normal structure of the vessel wall, which progresses to atherosclerosis. Elastin fragmentation attributable to mechanical fatigue results in passive dilatation and stiffening of the regional aortic wall17. Kamenskiy et al.15 reported that the continuity and density of the circumferential and longitudinal elastin in the CCA and ICA decrease with age.

Variations in the geometry of the carotid bifurcation increase significantly with age or early atherosclerotic progression8,9. However, there are currently limited data that have assessed longitudinal impacts on carotid artery geometry14. These data typically come from studies that compare carotid geometry in young people with older healthy subjects rather than directly studying an individual person9. Further, carotid bifurcation geometry is significantly correlated with the incidence of stroke18. Therefore, measuring the ICA angle may help assess the risk of ischemic stroke and the ICA angle could be used as a possible risk factor for assessing ischemic stroke. The ICA angles were significantly larger in patients with history of a stroke than in controls19. Consequently, we hypothesize that the longitudinal study of carotid bifurcation geometry can help predict the risk of stroke.

The geometry of carotid arteries has been considered as an important potential risk factor for the early development of atherosclerosis because of its significant influence on local hemodynamics20. Likewise, the bifurcation structure can cause various local hemodynamics in carotid segments and therefore may significantly influence the difference and changes in segment atherosclerosis. A previous study reported that the increase of the carotid bifurcation angle was related to carotid remodeling because of aging. With normal aging, the bifurcation angle gradually widens as the arteries dilate and become more tortuous. The carotid bifurcation angle became broader with age and increased by 10° over each decade of life14,15. In our study, the bifurcation angle of the bilateral carotid arteries significantly increased over 10 years. Additionally, the bifurcation angle of the right carotid artery significantly increased compared to the left carotid artery. However, the degree of increase of the bifurcation angle was approximately 2° over the 10-year follow-up period.

Atherosclerotic plaques in the left carotid artery were more vulnerable to conditions such as intraplaque hemorrhage and thin fibrous caps, compared with the right13,21. Our study did not identify geometrical changes in both carotid arteries over the 10-years of follow-up. The bifurcation angle of the right carotid artery significantly increased, but the diameter and area of the left CCA significantly increased. The diameter and area of the left ICA at the reference site and the minimum diameter and area of left the ICA were significantly decreased compared to the right side (p < 0.001). Further, some studies have shown that the diameter and circumference of the ICA increase with age11, while other studies have not reported significant changes over 10 years20. In our study, the changes in the diameter and area of the ICA were different in the left and right sides. The area and diameter of the ICA decreased in the left side, but significant changes were not noted in the right side. Differences in these changes between both carotid arteries could be attributable to geometric factors, such as the bifurcation angle, the configuration of the left carotid artery with the aortic arch or because of the direct connection of the left carotid artery to the aortic arch, as opposed to the right carotid artery that arises from the brachiocephalic artery8,13,22. Vessel anatomy influences hemodynamic forces, and as such, the left carotid artery may be exposed to higher arterial pressures8,13,22. Therefore, the right carotid artery with an increased bifurcation angle can experience hemodynamic forces that cause changes in the carotid bulb, and the left carotid artery, with the increased area and diameter of the CCA, can experience the hemodynamic forces that are changed in the CCA.

This study had some limitations. First, it was a retrospective study, which includes the limitation of the inherent possibility of selection bias. Second, because the carotid MRAs were performed for clinical indications, the included subjects were more likely than the general population to have vascular risks. However, the inclusion criteria of subjects with only normal VBA systems may have helped diminish this effect. Third, during the 10 years of follow-up, different parameters of the carotid MRA sequence were used, due to advances in techniques. However, we analyzed the data set based on examinations performed with one MR machine. Fourth, MRA was used for geometrical analysis of carotid artery instead of using CT angiography. Although MRA often had poor imaging quality due to flow artifact or venous contamination, we utilized contrast-enhanced MRA which showed relatively excellent visualization of vessel lumen. Moreover, in our hospital, MR examination is more common examination compared to CT angiography due to radiation dose and additional information of brain. Finally, because our study was not intended to be an epidemiological study, our patients do not necessarily represent characteristics of the general population.

Conclusion

Our longitudinal study was based on changes in individuals during a long-term follow-up and showed that the bifurcation angle, diameter and area of both CCAs significantly increased over a decade of life. Additionally, geometrical change of both carotid arteries over 10 years was not symmetrical. Changes in the bifurcation angle over a 10-year period were dominant in the right carotid artery and changes in the area and diameter of the CCA were dominant in the left carotid artery. Further studies are required to evaluate the association between geometrical changes and atherosclerotic risk factors.