Usefulness of cone-beam computed tomography to predict residual stenosis after carotid artery stenting

Abstract

Objectives

The long-term durability of carotid artery stenting (CAS) may be determined by various factors; however, residual stenosis is a known risk factor for in-stent restenosis. The authors of this article utilized cone-beam computed tomography (CBCT) in angiosuite to investigate plaque features affecting the character and quality of stent expansion after CAS.

Methods

Forty-two CAS cases with both pre- and post-CAS CBCT evaluations were included in this retrospective analysis. Five features derived from pre-CAS images were tested: (1) eccentricity, (2) overballoon, (3) maximum plaque thickness, (4) calcification barrier, and (5) stenotic degree. For post-CAS CBCT, stent configuration was assessed if the stent was expanded and oval or round in shape as well as outward or inward in orientation. Variables were tested if they were associated with oval expansion, outward expansion, and 20% residual stenosis after CAS.

Results

Oval or outward expansion is directly related to residual stenosis. The oval expansion was associated with maximum plaque thickness, and outward expansion was associated with the presence of a calcification barrier. Variables related to > 20% residual stenosis were the maximum plaque thickness, calcification barrier, and pre-CAS stenotic degree.

Conclusions

CBCT for carotid stenosis may provide valuable information about plaque features, especially calcification features that may interfere with the angioplasty effect, as well as the characteristics and quality of stent expansion. Residual stenosis > 20% was associated with calcification barrier, maximum plaque thickness, and pre-CAS stenotic degree.

Introduction

Carotid artery stenting (CAS) has emerged as a major treatment for carotid artery stenosis and may be a safe and effective alternative to carotid endarterectomy (CEA). The long-term durability of both treatment modalities was comparable. There was no significant difference between the CAS and CEA groups in restenosis rate according to the long-term results of previous large RCTs.^1,2

In-stent restenosis (ISR) may be determined by many variables, and some are yet to be clarified; however, many articles have reported that residual stenosis is a risk factor for restenosis after CAS.^3–10 Although the goal of CAS is to decrease the risk of stroke and a residual stenosis of 30%–40% is acceptable, ¹¹ residual stenoses may be closely related to ISR, especially in the early post-treatment phase. In our institution, we routinely perform both pre- and post-stenting balloon angioplasty, and our team's CAS protocol has been consistent for the last several years. Even though aggressive angioplasty using 6–7 mm balloons, some lesions showed suboptimal expansion and ended up with significant waist at the stenotic segment.

Magnetic resonance imaging (MRI) provides tissue characteristics of the plaque components; therefore, it is believed to be the best modality for detailed plaque evaluation. However, not all plaques are imaged with magnetic resonance before a treatment procedure in daily practice, partly because treatment decisions for high-degree carotid stenosis are made based on the degree of stenosis and symptoms rather than lesion stability depicted by plaque imaging, at least for the time being.

Diagnostic angiography plays a significant role in the preoperative evaluation of candidates for CEA or CAS; it offers precise information about the stenotic degree and anatomical environment of the vessel lumen. Catheter angiography is basically a luminal imaging technique that opacifies the vessel by filling it with contrast media; it is not a wall imaging technique. Cone-beam computed tomography (CBCT) with the excellent spatial resolution is now readily available for angiosuite during diagnostic or therapeutic procedures. Small field-of-view (FOV) CBCT with contrast infusion can produce superb images in patients with carotid stenosis, not only providing information about the geometry of the lumen and plaque features such as calcifications, but also the configuration of deployed stents after CAS.

The purpose of this retrospective study was to investigate the usefulness of CBCT in demonstrating certain plaque features on pre-CAS images and the quality of the stent expansion on post-CAS images, as well as to determine if features noted on pre-CAS images affect the suboptimal expansion of stents.

Materials and methods

Patients

Between January 2019 and March 2021, 134 carotid stenting procedures were performed at our institution. Among these cases, the authors retrospectively collected 46 cases with small field-of-view (FOV) CBCT images in both the pre- and post-procedure states. Four cases were excluded because of severe motion artifacts and subsequent poor image quality. Consequently, 42 carotid stenosis lesions in 38 patients (29 men and nine women) were included in the analysis. Informed consent from the study subjects may be waived because this study is a retrospective, a data analysis study, and the analysis had been performed with anonymized data after completing data collection.

Image acquisition

Pre-procedure Dyna computed tomography (CT) images were acquired on the day of diagnostic angiography or on the treatment procedure day. A 20-s cone beam CT protocol (20 s 109 kV Dyna CT Head, Artis Q, Siemens, Enlargen, Germany) with a small FOV (22 cm) was used with intra-arterial contrast media infusion (Visipaque 320, GE Healthcare, Milwaukee, WI, USA). Contrast media were diluted with saline (10% or 20% concentration) at an injection rate of 2 mL/s for 20 s and a 2-s scan delay. In some cases, when diluting the contrast was not possible, 4–8 mL of undiluted contrast media was injected at a rate of 0.2–0.4 mL/s.

All CBCT images were reconstructed with a thickness of 0.2–0.5 mm in the axial, coronal, sagittal, and oblique sagittal planes, where oblique images were intended to be as parallel to the vessel’s course as possible, and the external and internal carotid arteries were to be displayed on the same plane.

CAS procedure

In the authors’ institution, carotid stenting is performed as a separate session from diagnostic angiography, unless emergency revascularization is required. Premedication with daily 100 mg of aspirin and 75 mg of clopidogrel was administered for at least one week prior to the procedure. An antiplatelet agent resistance test was performed on the day of the treatment procedure (VerifyNow^TM, Instrumentation Laboratory, Bedford, MA, USA), and post-procedure antiplatelet medication was adjusted based on the test results. All patients were treated under local anesthesia with systemic heparinization of a bolus intravenous injection of 50 IU/kg at the beginning of the procedure. Our routine carotid stenting procedure consists of the delivery of a distal embolic protection device (EPD), pre-stenting balloon angioplasty, stent deployment, and post-stenting balloon angioplasty. Three types of EPDs were used: Emboshield NAV6 (Abbott Vascular, Santa Clara, CA, USA), Filterwire EZ (Boston Scientific, Marlborough, MA, USA), or Spider FX (Medtronic, Minneapolis, MN, USA). For pre-stenting balloon angioplasty, 3- to 5-mm monorail balloon catheters were used (Sterling, Boston Scientific, Marlborough, MA, USA); specifically, 3-mm in three cases, 5-mm in six cases, and 4-mm in all other cases. The vast majority of the patients were treated with an open-cell type stent, 39 Precise Pro RX (Cordis, Santa Clara, CA, USA), whereas two patients were treated with the Acculink stent (Abbott Vascular, Santa Clara, CA, USA). A closed-cell stent was used in only one case (Wallstent, Boston Scientific, Marlborough, MA, USA). Post-dilatation was performed with 6- or 7-mm devices (26 cases with 7-mm devices, 15 with 6-mm devices), except for one case that used a 5-mm balloon. Some patients demonstrated bradycardia with or without hypotension during and after post-stenting balloon angioplasty. Patients were instructed to cough, and intravenous dopamine infusion was initiated as necessary. Ten-minute delayed angiography was performed in every case to check if acute thrombosis or other immediate complications emerged.

Image analysis

Quantitative assessment

An interventional neuroradiologist (INR) with 5 years of experience retrospectively reviewed the pre- and post-stenting CBCT images for quantitative measurements. Stenotic degrees were recorded according to the North American Symptomatic Carotid Endarterectomy Trial (NASCET) method ¹² on oblique sagittal images for both pre- and post-procedure images.

The eccentricity of the plaque was assessed according to the location of the lumen. It was considered “central” when the center of the lumen was located within an area of half the diameter of the vessel, and “eccentric,” when located in other, areas (Figure 1).

Figure 1.

Eccentricity of the plaque. This schematic demonstrates the relationship between the center of the lumen and the center of the vessel. When r₂ is assumed to be 1/2 of r₁, it was considered to be “eccentric” if the center of the lumen was located between inner and outer circle (A) and “central” if it was located within the inner circle.

Outer diameters of the vessel were recorded and used to determine whether the post-stent angioplasty balloon was “overballooned,” which was defined by the authors when the balloon size was bigger than the outer diameter of the vessel at the most stenotic portion (measured on the oblique sagittal plane) or when the diameter difference was less than 1 mm. For example, a 6-mm balloon for a 6.2-mm-diameter vessel was considered overballooned.

Maximum plaque thickness was measured as the longest diameter from the lumen to the vessel wall on axial images at the level of the most stenotic portion.

Qualitative assessment

All qualitative assessments were conducted by two INRs, with 5 and 25 years of experience, respectively, in consensus. In addition to evaluating the presence of calcification, the morphological and geometrical characteristics of the calcifications were recorded. Calcifications were categorized into six groups: focal dot-like, linear, bulky, luminal bulky, rim-like, or circumferential (Figure 2). We hypothesized that luminal bulky/linear calcification or circumferential calcification may improve the expansion force of angioplasty balloons, resulting in suboptimal luminal gain after CAS. Those features were integrated into a variable called the “calcification barrier,” and the presence of this feature was assessed during the image analysis.

Figure 2.

Calcifications. (A–F) Demonstrates the representative schema of the calcification categories: dot-like (A), linear (B), bulky-adventitial (where “bulky” means it has a ≥ 2 mm thickness in any direction) (C), bulky-luminal (D), rim-like (complete or nearly complete rim outside of the lumen) (E), and circumferential (F). (D–F) were considered a “calcification barrier.”

For the post-CAS CBCT analysis, the final arrangement of stent expansion was examined in multiplanar-reconstructed images and axial images. Observations included whether it involved an oval or round expansion and an outward or inward expansion. When the stent was symmetrically expanded in all directions, the configuration of the stent was round on the axial plane. The asymmetrical expansion would result in an oval-shaped configuration, which was defined as a > 20% difference in the long-axis diameter and short-axis diameter (perpendicular to the line of the longest diameter). It was considered as an “outward” expansion when the stent struts were stretched beyond the boundary of the vessel recognized during the pre-treatment CBCT, or if there was an indentation between the interface of the boundary of the stented lumen and vessel wall, which may result in a number 8-like (or snowman-like) figure. Most of the carotid lesions after CAS showed a mixed type of orientation in terms of inward versus outward expansion; therefore, the reviewers decided on which was the more dominant direction for each lesion (Figure 3 and 4).

Figure 3.

Examples of characteristics of stent expansion, with pre-carotid artery stenting (CAS) plaque findings at the same level.

A1-2: Round and inward expansion in a non-calcified plaque.

B1-2: Oval and inward expansion in a non-calcified but severely eccentric plaque (the margin of the vessel outer wall is demarcated with white arrowheads and the eccentrically located remnant vessel lumen is annotated with a blank arrow).

C1-2: Round and outward expansion in a plaque with rim-like calcification. Note the indentation between the contours of the stent struts and the presumed vessel outer wall (black arrows), resulting in a snowman-like figure.

Figure 4.

A representative case of oval and outward expansion.

A 73-year-old male patient had 72% stenosis on the right side that showed ulceration and a calcification barrier (circumferential calcification, visible in (C)). Post-CAS angiography of the lateral view (B) shows that the waist remains at the most stenotic portion compared to pre-CAS angiography (A). Residual stenosis was 32%, measured using sagittal CBCT images (not shown). On the pre- and post-CAS CBCT images (C, D), the stent was not only showing an oval-shaped configuration but also an outward protrusion compared to the pre-CAS image. There was an indentation between the border of the stent and the parent vessel (long black arrow in D) and the distance from the external carotid artery was decreased (white arrows in C and D). CAS: carotid artery stenting; CBCT: cone-beam computed tomography.

Clinical information

History of diabetes mellitus (DM) and hypertension (HTN), smoking status, and medication status were collected by reviewing the electronic medical record at our institution. Comorbidity information was collected by reviewing admission records at the time of diagnostic angiography, and only the presence of DM or HTN was included, not considering the duration or severity of the disease or medication status. Ex-smokers and current smokers were considered to be ever-smokers. Whether or not the patient was receiving antiplatelets or statins was determined at the time of diagnostic angiography, and pre-medications in preparation for the CAS procedure were not considered as underlying antiplatelet medications. Patients with ipsilateral stroke, transient ischemic attack, or silent diffusion-weighted image-positive lesions in the ipsilateral corresponding territory within 6 months of the CAS procedure were considered symptomatic.

Statistical analysis

Expansion characteristics were compared to the degree of residual stenosis evaluated on post-CAS CBCT sagittal images. Five features assessed on pre-CAS CBCT (eccentricity, overballooning, calcification barrier, maximum plaque thickness, and stenotic degree) were analyzed for their relationship with oval expansion, outward expansion, and > 20% residual stenosis. The analysis was performed using SPSS version 24 (IBM, Armonk, NY, USA).

Results

Patient demographics and clinical information are summarized in Table 1. Patient demographic data were collected, including basic comorbidities such as a history of DM, HTN, smoking history, and status of antiplatelet and statin medication at the time of diagnostic angiography evaluation. The demographic variables were not significantly different between the groups when tested for oval versus round expansion, outward versus inward expansion, and > 20% residual stenosis or less, except for the presence of DM, which was significantly less frequent in the > 20% stenosis group than in the others (p < 0.05, Chi-square tests with Yate's continuity correction).

Table 1.

Patient demographics.

Mean age		70.7 (range: 54–88)
Male		29 (76.3%)
Symptomatic a		25 (59.5%)
HTN		34 (89.5%)
DM		23 (60.5%)
Smoking	Current smoker	10 (26.3%)
	Ex-smoker	10 (26.3%)
	Non-smoker	18 (47.4%)
Any anti-platelet agent use		28 (73.7%)
Statin use		28 (73.7%)

DM: diabetes mellitus; HTN: hypertension; CAS: carotid artery stenting; TIA: transient ischemic attack; DWI: diffusion-weighted imaging.

^a Stroke, TIA, or asymptomatic acute ischemic lesion documented on DWI within 6 months prior to CAS procedure.

In all 42-carotid stenosis in 38 patients, carotid stenting was technically successful, and no intraprocedural complications were noted. The pre-CAS stenotic degree was 70.4% on average (ranging from 41.6% to 88.8%, including four near-total occlusions). The mean residual stenosis after CAS was 15.5% (range: 0–45.7%). Seven (7/42, 16.7%) showed more than 30% residual stenosis, and 15 (15/42, 35.7%) displayed residual stenosis exceeding 20%.

Angioplasty balloons were “overballooned” in 16 cases, meaning that the post-stenting balloon size exceeded the outer diameter of the vessel at the most stenotic segment (measured on oblique sagittal or sagittal images), or if the difference between vessel diameter and balloon size was smaller than 1 mm. There were some calcified portions in the plaques of 37 cases of carotid stenosis (37/42, 88.1%), and 13 carotid plaques demonstrated a “calcification barrier” on pre-CAS CBCT (13/42, 31.0%).

Twenty out of 42 carotid stenoses showed oval expansion after CAS (20/42, 47.6%), and these were closely related to > 20% residual stenosis (Chi-square, p = 0.00). It appears that the oval-shaped configuration of deployed stents indicates their suboptimal expansion in a certain direction. Thirteen patients showed outward expansion (13/42, 31.0%) that resulted from the dominant effect of normal vessel wall stretching rather than stretching on the plaque side. Both oval and outward expansions were seen in 10 carotid arteries, and the average residual stenosis for these 10 lesions was significantly higher than that in the other 32 cases (30.5% vs. 10.8%, t-test, p = 0.00).

Five parameters were tested with univariate logistic regression for their association with oval expansion, outward expansion, or > 20% residual stenosis (Table 2). Statistically significant variables were maximum plaque thickness for oval expansion and calcification barrier for outward expansion (p < 0.05). More than 20% residual stenosis was related to the calcification barrier, maximum plaque thickness, and pre-treatment stenotic degree (p < 0.05).

Table 2.

Univariate logistic regression for factors contributing to oval expansion, outward expansion, and more than 20% residual stenosis.

	Oval expansion			Outward expansion			> 20% residual stenosis
	OR	95% CI	p-value	OR	95% CI	p-value	OR	95% CI	p-value
Eccentricity	0.24	0.03, 1.16	0.101	1.15	0.21, 5.35	0.862	0.17	0.01, 1.08	0.112
Overballoon	0.78	0.22, 2.72	0.694	0.63	0.14, 2.44	0.514	1.13	0.30, 4.14	0.85
Calcification barrier	2.27	0.61, 9.18	0.231	4.47	1.11, 19.5	0.038	5.03	1.28, 22.0	0.024
Maximum plaque thickness	1.75	1.11, 3.08	0.028	0.98	0.62, 1.52	0.923	2.49	1.42, 5.25	0.005
Pre-Tx stenotic degree	1.05	0.01, 1.11	0.985	2.42	0.02, 471	0.73	921	3.76, 777,991	0.026

When we examined seven carotid stenoses with more than 30% residual stenosis, 100% (seven of seven) showed oval expansion and 71.4% (five of seven) showed outward expansion. Five displayed a calcification barrier on pre-CAS CBCT, and the average maximum plaque thickness was 6.81 mm, which was higher than the overall mean value (5.55 mm). Only one out of those seven had “overballooning” during CAS. Although the small number of cases in this subset (together with a small number of our total cases included) makes statistical analysis unreliable, there is a notable tendency in lesions with suboptimal CAS results.

Discussion

In this study, plaque characteristics and quality of stent expansion were evaluated through pre- and post-CAS CBCT, and parameters measured on pre-CAS images were analyzed for their relationship with the status of the deployed stents. To the best of our knowledge, there have been no reports comparing CT-based imaging before and after CAS in terms of the quality of stent expansion.

According to a multi-society expert consensus on carotid stenting published in 2007, ¹¹ pursuing perfect angiographic results is not a goal for CAS, and moderate (30%–40%) residual stenosis is acceptable. Nevertheless, residual stenosis is a risk factor for ISR,^3–9 higher risk of periprocedural events,^13,14 and recurrent ipsilateral stroke in the long term. ¹⁵

The mean residual stenosis after CAS was only 15% in our study group, ranging from 0% to 45.7%. When residual stenosis degree exceeding 30% was considered to be significant in general, only seven cases showed more than 30% residual stenosis (16.7%). Therefore, we applied a slightly stricter standard of 20% residual stenosis for statistical analysis. Our CAS procedure seems to be effective in terms of luminal gain, despite the fact that a high proportion of the plaques contained calcification.

Among the evaluated factors, it can be observed that certain calcification features and severe maximal plaque thickness showed a close relationship with suboptimal stent expansion after CAS. Lesion eccentricity was expected to be related to asymmetric expansion or a higher degree of residual stenosis, but it did not show a statistically significant correlation. However, the eccentric location of the vessel lumen means a larger maximum plaque thickness, which is associated with oval expansion.

We also expected that balloon size may play some role in adequate dilatation after balloon angioplasty and that “overballooning” could be inversely related to residual stenosis; however, our data failed to show statistical significance. It is possible that our initial assumption that the size of a balloon compared to the vessel’s outer wall at the most stenotic part would matter is wrong; instead, perhaps we should have assessed relative balloon size compared to other parameters such as distal normal vessel diameter or pre-CAS lumen diameter.

Many previous reports have demonstrated the relationship between calcification and residual stenosis after CAS in various aspects. Katano et al. reported that heavily calcified plaque showed a disadvantage in terms of stent expansion when treated with CAS. ¹⁶ In another study, the arc of calcification was associated with the degree of residual stenosis; in their data, there was no correlation between total calcification volume and residual stenosis. ¹⁷ Tao et al. ¹⁸ classified calcification according to its location: superficial, basal, and internal calcifications. In addition, the rate of residual stenosis was the highest in the basal group.

The high proportion of carotid stenosis (88.1%) treated in this study had some calcifications, and 31.0% had a calcification barrier as defined by the authors (luminal bulky/rim-like/circumferential calcifications). We speculated that rather than the presence of calcification or a large volume of it, certain types of calcifications can sabotage angioplasty results. The author's assessed features of plaque calcification in a more indirect and subjective way, categorizing them according to their morphology and location. In this study, there were limitations on stent expansion in lesions with rim-like and circumferential calcifications, even when they exhibited relatively smaller volumes or internal soft plaque components.

Luminal gain after stenting and angioplasty is achieved as a combined effect of wall stretching and plaque compression. ¹⁹ Interestingly, in cases with a calcification barrier, stent expansion occurred toward the normal wall (or predominantly toward the normal wall, at least), and even beyond the expected boundary of the vessel wall demarcated on pre-CAS images. Several studies have mentioned that lumen can be gained by stretching the normal vessel wall of the non-calcified segment as opposed to the calcified plaque.^17,19–21 These findings may have been derived from the fact that the vessel wall in the region without plaque is more flexible and capable of stretching than the region with plaque. Due to the different elasticity properties between the calcified plaque and its opposite normal wall, the stent expansion becomes an outward eccentric expansion rather than a concentric expansion, and the probability of residual stenosis increases.

Several studies investigated the plaque features using multidetector CT angiography (MDCTA)^22–29; however, a CBCT-based plaque study has yet to be reported. CBCT can provide similar information about carotid plaques compared with MDCTA, with even higher spatial resolution. The most remarkable advantage of CBCT is that it can be readily applied during diagnostic or treatment procedures in angiosuite. Additional radiation exposure is required to acquire CBCT images during angiography. However, MDCTA may be replaced by CBCT during diagnostic angiography because it provides similar information with higher resolution while requiring a smaller amount of contrast injection. Vulnerability to motion and beam-hardening artifacts are the disadvantages of CBCT.

The main limitation of this retrospective study was the small number of patients. Larger prospective investigations will be useful to validate independent risk factors that affect suboptimal stent expansion after CAS using cone beam CT. Our assessment may be biased and limited not only because we did not quantify plaque calcification, but the reviewers were also not blinded to the CAS results during the image analysis. Another limitation is that we only evaluated stent expansion based on axial plane images, despite the fact that delivery of radial force during angioplasty occurs in multiple directions. The elongation of the device and changes in the vessel's course may be important variables that affect CAS results; however, these were not included in this study. We tried to simplify the analysis and focus on the quality of the stent expansion.

In conclusion, > 20% residual stenosis measured post-CAS CBCT was related to maximum plaque thickness, calcification barrier, and the initial stenosis degree assessed on pre-CAS CBCT. The maximum plaque thickness was correlated with the oval expansion of the stent, and the calcification barrier was associated with outward expansion. The use of CBCT in angiosuite is expected to be a useful tool for evaluating the characteristics of plaque and predicting the characteristics and quality of stent expansion.