If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Address correspondence to Kohei Chida, MD, PhD, Department of Neurosurgery, Iwate Medical University, 1-1, Idai-dori 1-chome, Yahaba-cho, Shiwa-gun, Iwate, Japan 028-3694.
Background: Complete removal of the distal end of the plaque is an important requirement in carotid endarterectomy (CEA) to avoid postoperative complication. Preoperative identification of the distal end of plaque contributes to complete plaque removal. Three-dimensional (3D) magnetic resonance (MR) plaque imaging has been widely used to evaluate carotid plaque characterization. The purpose of the present study was to determine whether preoperative 3D fast spin echo (FSE) T1-weighted MR plaque imaging could identify the distal end of carotid plaque. Methods: This study was designed as a prospective cohort study. We examined 50 patients with cervical internal carotid artery (ICA) stenosis who underwent CEA. 3D-FSE T1-weighted MR plaque imaging of the affected carotid bifurcation was preoperatively performed using a 1.5-T scanner. Identification of the distal end of plaque (DEMRI) on MR plaque imaging was performed and the distance from the baseline (DistanceMRI) was measured. Intraoperatively, the superimposed distal end of carotid plaque (Esim) was marked on the ICA according to the measurement on MR plaque imaging. The actual distal end of plaque (DECEA) was then identified after arteriotomy and the difference (DifferenceCEA-MRI) between Esim and DECEA was measured. Contrast ratio of carotid plaque and tortuosity of the ICA were calculated using MR plaque imaging. Results: Interobserver agreements in measurement of DistanceMRI were excellent (intraclass correlation coefficient, .955; 95% confidence interval, .922-.974). In 28 patients (56%), Esim was identical to DECEA. Mean DifferenceCEA-MRI was 1.32 ± 1.77 mm. DifferenceCEA-MRI was significantly greater with fibrotic plaque (4.14 ± 1.21 mm) than with lipid-rich or necrotic plaque (.43 ± .87 mm; P < .05) or hemorrhagic plaque (1.27 ± 1.64 mm; P < .05). Mean DifferenceCEA-MRI was significantly greater in the group with tortuosity of the ICA less than 120° (3.86 ± 1.77 mm) than in the group with greater than or equal to 120° but less than or equal to 150° (1.15 ± 1.51 mm; P < .05) or greater than150° (0.50 ± 1.10 mm; P < .05). No patients showed residual stenosis after surgery on postoperative MR angiography. Conclusions: Using 3D-FSE T1-weighted MR plaque imaging allowed identification of the distal end of carotid plaque and contributed to complete removal of the plaque, although it may be reduced for cases with low-signal-intensity plaque or severe tortuosity of the ICA.
International guidelines require that overall morbidity and mortality rates associated with CEA should be less than 6% in symptomatic patients and less than 3% in asymptomatic patients to justify the intervention.
Complete removal of plaque is important during CEA to avoid postoperative ischemic complications, occlusion of the carotid artery, and future re-stenosis.
Intraoperative identification of the distal end of carotid plaque contributes to complete removal of plaque. Such identification is also needed to ensure a sufficient area of the internal carotid artery (ICA) for appropriate shunt placement.
exposing the distal ICA are not facile in some cases in Japan. Therefore, preoperative identification of the distal end of carotid plaque is important to prevent from exposing the distal ICA excessively.
Several modalities have been used for such identification, including intraoperative ultrasonography and indocyanine green (ICG) videoangiography.
The former is widely used, but does not depict a calcified or high-positioned plaque. The latter does not accurately identify the distal end of plaque, particularly for ICA pseudo-occlusion.
Preoperative angiography using arterial catheterization or computed tomography also does not accurately identify the distal end of plaque, because it essentially images the lumens of the carotid arteries rather than plaque itself.
Recently, 3-dimensional (3D) fast spin echo (FSE) T1-weighted magnetic resonance (MR) plaque imaging has been widely used to identify carotid plaque, because it can minimize partial volume effects and motion artifacts.
Accuracy of preoperative three-dimensional fast spin echo T1-weighted magnetic resonance plaque imaging for severely stenotic cervical internal carotid artery in predicting the development of artery-to-artery emboli during exposure of the carotid arteries in carotid endarterectomy.
In addition, 3D techniques of MR plaque imaging are suitable for assessing carotid plaque, because the carotid artery is tortuous and pulsatile, and plaques are typically small, complex in shape and composition, and elongated in the superoinferior direction.
Accuracy of preoperative three-dimensional fast spin echo T1-weighted magnetic resonance plaque imaging for severely stenotic cervical internal carotid artery in predicting the development of artery-to-artery emboli during exposure of the carotid arteries in carotid endarterectomy.
The purpose of the present study was thus to determine whether preoperative 3D-FSE T1-weighted MR plaque imaging could identify the distal end of carotid plaque and contribute to complete removal of plaque.
Methods
Study Design
This study was designed as a prospective cohort study. This protocol was reviewed and approved by the institutional ethics committee, and written informed consent was obtained from all patients or their next of kin before participation.
Patient Selection
The present study prospectively included patients with ipsilateral ICA stenosis greater than or equal to 70% according to the methods of the North American Symptomatic Carotid Endarterectomy trial (NASCET) on angiography with arterial catherization who had useful residual function (modified Rankin disability scale, 0-2) and underwent CEA of the carotid bifurcation in our institution. CEA was scheduled at least 2 weeks after the ischemic events for symptomatic patients.
Preoperative Determination of the Distal End of Plaque on MR Imaging
Preoperative 3D-FSE T1-weighted MR plaque imaging of affected carotid bifurcation was performed within 1 week prior to CEA using a 1.5-T MR imaging scanner (Signa HDxt; GE Healthcare, Milwaukee, WI) with an 8-channel neurovascular coil under a previously described imaging protocol.
An investigator blinded to other data processed the 3D FSE T1-weighted MR plaque imaging data using a free software package (OsiriX; Pixmeo, Geneva, Switzerland) as follows: on the MR plaque imaging, the distal end of carotid plaque (DEMRI) was visually determined based on the difference in signal intensity between carotid plaque and carotid intima (Fig 1A). On the same image, the arterial branch from the external carotid artery that was displayed most clearly was also determined. The plane being perpendicular to the long axis of the carotid artery and through the mid portion of the arterial branch was defined as axial baseline (Fig 1B). Of lines parallel to the axial baseline, a line through the distal end of carotid plaque was determined (defined as DEMRI line). Finally, the distance between these 2 lines (DistanceMRI) was measured.
Figure 1Preoperative 3-dimensional fast spin echo T1-weighted magnetic resonance (3D-FSE T1-weighted MR) plaque imaging of a 67-year-old man with right cervical ICA stenosis who underwent carotid endarterectomy (CEA). High-intensity plaque is evident at the right cervical internal carotid artery (ICA). The distal end of carotid plaque (DEMRI) is identified based on the difference in signal intensity between carotid plaque (white arrow) and carotid intima (white arrowheads) (A). Since the superior thyroid artery (yellow arrow) is displayed most clearly (A), the plane being perpendicular to the long axis of the carotid artery and through the mid portion of the arterial branch was defined as the baseline (yellow line) (B). Of lines parallel to the axial baseline, a line through DEMRI was defined as DEMRI line (orange line). The distance between these 2 lines (DistanceMRI) was then measured (red arrow). A circular region of interest (yellow circle) is manually traced between the DEMRI line and 5-mm proximal to the DEMRI to measure the signal intensity of the distal end of carotid plaque. (Color version of figure is available online.)
To assess interobserver variability on MR findings, 2 experienced neuroradiologists independently performed determination of the distal end of plaque. Both observers were blinded to the clinical and imaging findings, other than the side on which CEA was performed. Each observer was also blinded to assessments from the other.
Classification of Signal Intensities of Carotid Plaque and Tortuosities of the ICA
Carotid plaque between the DEMRI line and the line 5 mm proximal to the DEMRI was manually traced to measure signal intensity (Fig 1B). Signal intensity of the sternocleidomastoid muscle was also measured. Contrast ratio (CR) of plaque was calculated by dividing plaque signal intensity with the sternocleidomastoid muscle signal intensity in each patient, as described previously.
Accuracy of preoperative three-dimensional fast spin echo T1-weighted magnetic resonance plaque imaging for severely stenotic cervical internal carotid artery in predicting the development of artery-to-artery emboli during exposure of the carotid arteries in carotid endarterectomy.
CRs were classified into 3 groups as less than 1.3 (plaque containing fibrotic component), greater than or equal to 1.3 but less than or equal to 1.6 (plaque containing lipid-rich or necrotic component), or greater than 1.6 (plaque containing hemorrhagic component) as a previous report demonstrated that 3D-FSE T1-weighted MR plaque imaging accurately characterized carotid plaque components.
In addition, the inflection point of the ICA was also defined as the center of the curvature on 2D phase contrast MR angiography performed with MR plaque imaging. Tortuosity of the ICA was defined as the angle of anterior wall of the artery formed by the 2 tangential lines drawn on the ICA side and the common carotid artery side starting from the inflection point on a view in which curvature of the ICA was most visible as previously described.
We classified tortuosities of the ICA into 3 groups as less than 120°, greater than or equal to 120° but less than or equal to 150°, or greater than150°.
Intraoperative Determination of the Distal End of Plaque
All patients underwent surgery under general anesthesia with an operative microscope from skin incision. One surgeon (K.C.) identified the origin of the branch originating from the external carotid artery that corresponded to the origin of the branch selected on MR plaque imaging. The baseline through the origin of the arterial branch was drawn so that the line was perpendicular to the long axis of the common carotid artery or ICA. A line with DistanceMRI determined preoperatively was then manually drawn on the ICA parallel to the baseline. The resulting line was defined as the superimposed distal end of carotid plaque (Esim), which was intraoperatively marked on the ICA using 8-0 nylon (Fig 2A).
Figure 2Intraoperative photos of the same case show measurement of the superimposed distal end of carotid plaque (Esim) during CEA (A). The baseline through the origin of the superior thyroid artery (yellow arrow) is drawn (yellow line). A line with the DistanceMRI determined preoperatively is drawn parallel to the baseline (orange line). The resulting line is defined as Esim, intraoperatively marked on the ICA using 8-0 nylon (white arrow). During endarterectomy, the actual distal end of plaque (DECEA) is determined based on the boundary between the distal end of plaque and the carotid intima (black arrow) (B). The point of the DECEA is located just behind the point of Esim (white arrow). (Color version of figure is available online.)
After arteriotomy, the actual distal end of plaque (DECEA) was visually determined based on the boundary between the distal end of plaque and the carotid intima by another surgeon (K.O.) (Fig 2B). Arteriotomy was extended on the distal end as necessary. The difference between Esim and DECEA was then measured and the absolute value of this difference was defined as DifferenceCEA-MRI. When Esim was identical to DECEA, DifferenceCEA-MRI was 0.
Definition of Residual Stenosis After CEA
Residual carotid stenosis after CEA was defined as the ipsilateral ICA stenosis greater than or equal to 25% with the method of the NASCET on postoperative MR angiography based on a previous study.
All data are presented as mean ± standard deviation. Interobserver agreements in measurement of DistanceMRI between first and second observers were assessed using the intraclass correlation coefficient (ICC). ICC is interpreted as follows: 0-.20, poor agreement; .21-.40, fair agreement; .41-.60, moderate agreement; .61-.80, good agreement; and .81-1.00, excellent agreement. When ICC was greater than or equal to .61, values determined by the first observer were used for analyses. DifferenceCEA-MRI was compared among the 3 groups of CRs or tortuosities of the ICA using the Kruskal-Wallis test. Bonferroni correction was used as post hoc analysis. The difference in DifferenceCEA-MRI between patients with and without ICA pseudo-occlusion was determined using the Mann-Whitney U test. Receiver operating characteristic curves was used to assess the accuracy of CR and tortuosity of ICA in predicting DifferenceCEA-MRIgreater than or equal to 4 mm. For all statistical analyses, significance was set for values of Pless than .05.
Results
During the 17 months, a total of 50 patients satisfied the inclusion criteria and were entered into the present study. The distal end of plaque was determined on MR plaque imaging in all 50 patients.
Mean age of the 50 patients (48 men, 2 women) was 70.6 ± 6.9 years (range, 52-86 years). Twenty-nine patients (58.0%) had symptomatic carotid stenosis. Mean overall degree of ICA stenosis was 88.6 ± 9.2% (range, 70%-99%), including 5 patients with ICA pseudo-occlusion, as per the NASCET on angiography with arterial catheterization performed prior to MR plaque imaging.
Interobserver agreements in measurement of DistanceMRI were excellent (ICC, .955; 95%CI, .922-.974). The range of discrepancy of interobserver agreements was 0-3 mm.
DECEA was located distal and proximal to Esim in 9 (18%) and 13 (26%) patients, respectively. DECEA was identical to Esim in the remaining 28 patients (56%). DifferenceCEA-MRI ranged from 0 to 6 mm (1.32 ± 1.77 mm) and was less than or equal to 3 mm in 43 patients (86%). No significant difference in DifferenceCEA-MRI was evident between patients with ICA pseudo-occlusion (1.6 ± 2.2 mm) and those without (1.3 ± 1.7 mm; P =.830). Preoperative 3D-FSE T1-weighted MR plaque imaging and intraoperative photo of a case with pseudo-occlusion are shown in Figure 3.
Figure 3Preoperative 3D-FSE T1-weighted MR plaque imaging of a 65-year-old man with left cervical ICA pseudo-occlusion who underwent CEA (A). While distal ICA seems attenuated and opacified due to deteriorated blood flow (black arrows), the distal end of carotid plaque is clearly identified (white arrow). Intraoperative photo of the same case shows the point of the DECEA (black arrow) is located right behind the point of Esim (white arrow) (B).
Seven (14%), 20 (40%), and 23 (46%) patients had fibrotic carotid plaque, lipid-rich or necrotic carotid plaque, or hemorrhagic carotid plaque, respectively. Figure 4 shows comparisons of DifferenceCEA-MRI between each carotid plaque component. DifferenceCEA-MRI was significantly greater with fibrotic plaque (4.14 ± 1.21 mm) than with lipid-rich or necrotic plaques (.43 ± .87 mm; P < .05) or hemorrhagic plaque (1.27 ± 1.64 mm; P < .05). DifferenceCEA-MRI did not differ significantly between lipid-rich or necrotic plaque and hemorrhagic plaque.
Figure 4Relationship between carotid plaque components and DifferenceCEA-MRI. DifferenceCEA-MRI is significantly greater with fibrotic plaque (4.14 ± 1.21 mm) than with lipid-rich or necrotic plaque (.43 ± .87 mm; P < .05) or hemorrhagic plaque (1.27 ± 1.64 mm; P < .05). Closed and open circles indicate DECEA located distally and proximally to Esim, respectively.
Seven (14%), 24 (48%), and 19 patients (38%) showed tortuosity of the ICA less than 120°, greater than or equal to 120° and less than or equal to 150°, and greater than150°, respectively. Figure 5 shows comparisons of DifferenceCEA-MRI between each tortuosity of the ICA. DifferenceCEA-MRI was significantly greater in the group with tortuosity of the ICA less than 120° (3.86 ± 1.77 mm) than in those with greater than or equal to 120° but less than or equal to 150° (1.15 ± 1.51 mm; P < .05) or greater than 150° (.50 ± 1.10 mm; P < .05). DifferenceCEA-MRI did not differ significantly between the group with tortuosity of the ICA greater than or equal to 120° but less than or equal to 150° and the group with tortuosity of the ICA greater than 150°.
Figure 5Relationship between tortuosities of ICA and DifferenceCEA-MRI. DifferenceCEA-MRI is significantly greater in the group with tortuosity of the ICA less than 120° (3.86 ± 1.77 mm) than in the group with tortuosity greater than or equal to 120° but less than or equal to 150° (1.15 ± 1.51 mm; P < .05) or greater than 150° (.50 ± 1.10 mm; P < .05). Closed and open circles indicate DECEA located distal and proximal to Esim, respectively.
The relationship among CR, Tortuosity of ICA, and DifferenceCEA-MRI is shown in Figure 6. The area under the receiver operating characteristic curves for predicting DifferenceCEA-MRIgreater than or equal to 4 mm for the CR and the tortuosity of ICA were .627 (95%CI, .356-.898) and .894 (95%CI, .761-1.000), respectively. Sensitivity and specificity for the CR at the cutoff point lying closest to the upper left corner of the receiver operating characteristic curve to predict DifferenceCEA-MRI greater than or equal to 4 mm were 57.1% and 88.0%, respectively (cutoff point = 1.25). Sensitivity and specificity for the tortuosity of ICA at the cutoff point lying closest to the upper left corner of the receiver operating characteristic curve to predict DifferenceCEA-MRI greater than or equal to 4 mm were 85.7% and 86.1%, respectively (cutoff point = 125°).
Figure 6Relationship among CR, Tortuosity of ICA, and DifferenceCEA-MRI. Closed and open circles denote DifferenceCEA-MRI greater than or equal to 4 mm and DifferenceCEA-MRI less than 4 mm, respectively. Vertical and horizontal lines denote the cutoff points lying closest to the upper left corner of the receiver operating characteristic curves for CR and Tortuosity of ICA to detect DifferenceCEA-MRI greater than or equal to 4 mm.
During endarterectomy, arteriotomy were extended distally in addition to the initial arteriotomy in 3 patients (6%) as DECEA was distal to the expected location. No patients showed residual stenosis after surgery on postoperative MR angiography.
Discussion
Our study demonstrated that preoperative 3D-FSE T1-weighted MR plaque imaging identified the distal end of carotid plaque. Although the distal end of carotid plaque was visually identified using a subjective process, interobserver agreements for these measurements were excellent.
While intraoperative ICG angiography seems useful for detecting the distal end of plaque, Okawa et al suggested that intensity of ICG would decrease in cases with ICA pseudo-occlusion due to deteriorated blood flow.
In our study, 5 patients showed ICA pseudo-occlusion. However, DifferenceCEA-MRI was not significantly different between those 5 patients and the remaining patients without pseudo-occlusion. In addition, they suggested decreased ICG intensity where plaque exists but is unable to be detected, if plaque is mainly on the opposite side to exposure.
Since signal intensity on 3D MR plaque imaging does not depend on blood flow or plaque location, this modality would be useful for detecting plaque location in such cases. As for comparisons with ultrasonography, no patients were encountered for whom the distal end of plaque could not be identified according to calcified or high-positioned plaque in this study.
Since extending exposure of the carotid artery after arteriotomy is a risky proposition, the problem is that the actual distal end of plaque is located distal to the expected extent. In the present study, DECEA was located distal to Esim in only 9 patients (18%) and arteriotomy was extended distally in addition to the initial arteriotomy in only 3 patients (6%) during endarterectomy. None of the patients whose DifferenceCEA-MRI were less than 4 mm needed additional arteriotomy during endarterectomy, which is the reason 4 mm was set as the cutoff point of the receiver operating characteristic analysis to assess the accuracy of CR and tortuosity of ICA. Esim was located distal to DECEA for several reasons. Since carotid arteries were entirely exposed during surgery, their position may change from those of preoperative MR plaque imaging. DECEA may shift to distal to Esim depending on the tortuosity of the ICA when the arterial branch from the external carotid artery defined as baseline is proximal to Esim. The result that only 3 patients with DECEA not identical to Esim was seen in the group with ICA tortuosity greater than 150° was thus supported. On the other hand, more patients with DECEA not identical to Esim were seen in the group with hemorrhagic plaque rather than those with lipid-rich or necrotic plaque. The distal end of plaque could be accurately detected only if the intensity was above a certain level unlike tortuosity of the ICA. For the same reason, the accuracy for tortuosity of ICA in detecting DifferenceCEA-MRI greater than or equal to 4 mm was greater than that for CR based on the cutoff point of the receiver operating characteristic analysis. Cervical extension during surgery could be another reason that DECEA was located distal to Esim. Since carotid arteries were also stretched, the actual distal end of plaque would shift distally compared with preoperative MR plaque imaging. In contrast, when the arterial branch from the external carotid artery defined as baseline was distal to Esim, the actual distal end of plaque would shift to proximal to Esim, as the baseline would shift distally compared with preoperative MR plaque imaging for the same reasons. Thus, combination with intraoperative identification after exposure of entire carotid arteries (such as ultrasonography or ICG videoangiography) may lead to more accurate identification of the distal end of carotid plaque. Several investigators have reported detection of the distal end of plaque as measured from carotid bifurcations.
However, carotid bifurcations are not always completely exposed during surgery and may not be well defined, so the origin of some branches of the external carotid artery is suitable as the baseline for measuring the distance to the distal end of plaque.
Several limitations must be included when considering the results of this study. First, this method is not available for the patients with implanted electronic devices. Although patients with artificial teeth would not seem indicated for this method due to imaging artifacts, no such patients showed carotid plaque that was unable to be clearly depicted because of artifacts even in previous studies using 3D-FSE MR plaque imaging.
Accuracy of preoperative three-dimensional fast spin echo T1-weighted magnetic resonance plaque imaging for severely stenotic cervical internal carotid artery in predicting the development of artery-to-artery emboli during exposure of the carotid arteries in carotid endarterectomy.
Second, the imaging technique for the present study was unsuitable for depicting vessels. Although this method requires definition of an arterial branch from the external carotid artery as a baseline, no patients showed arterial branches that were undetectable on MR plaque imaging. Third, though residual carotid stenosis after CEA was evaluated on postoperative MR angiography, postoperative 3D-FSE MR plaque imaging would detect residual carotid plaque more accurately. Finally, the small sample size was another limitation.
Conclusions
The present study demonstrated that 3D-FSE T1-weighted MR plaque imaging could identify carotid plaque location and contribute to complete removal of plaque, while identification may deviate for cases with low-signal-intensity plaque or severe tortuosity of the ICA. This technique offers a noninvasive method for identifying the distal end of carotid plaque for CEA. The technique could also contribute to carotid artery stenting to decide the location of stent placement.
Acknowledgement
This work was supported by JSPS KAKENHI Grant Number JP19K09534.
Conflict of Interest
The authors have no personal, financial or institutional interest in any of the drugs, materials or devices described in this article.
Accuracy of preoperative three-dimensional fast spin echo T1-weighted magnetic resonance plaque imaging for severely stenotic cervical internal carotid artery in predicting the development of artery-to-artery emboli during exposure of the carotid arteries in carotid endarterectomy.
30051-3&id=gr1.jpg){kind=link}
30051-3&id=gr2.jpg){kind=link}
30051-3&id=gr3.jpg){kind=link}
30051-3&id=gr4.jpg){kind=link}
30051-3&id=gr5.jpg){kind=link}
30051-3&id=gr6.jpg){kind=link}