Predicting embolic potential during carotid angioplasty and stenting: Analysis of captured particulate debris, ultrasound characteristics, and prior carotid endarterectomy
Article Outline
Introduction
Extracranial carotid stenoses exhibit significant variance in embolic potential, with restenotic lesions having a particularly low propensity for embolization. This study sought to identify characteristics associated with increased generation of embolic debris during carotid angioplasty and stenting (CAS).
Methods
Captured particulate was available for analysis in 56 consecutive patients. Demographics were mean age, 74 years (range, 60-94 years); mean stenosis, 88% (range, 70%-99%); symptomatic, 27%; prior carotid endarterectomy (CEA), 27%; prior radiotherapy, 7%. Plaque echogenicity, heterogenicity, ulceration, and irregularity were assessed with B-mode duplex ultrasound analysis. Gray scale median (GSM) was calculated from normalized B-mode VHS video recordings. Calcification and degree of stenosis were determined angiographically. Captured particulate debris was evaluated for total number; number >200 μm, >500 μm, >1000 μm; mean and median size. Hematoxylin and eosin, trichrome, and von Kossa stains were used for histologic analysis of captured material.
Results
Restenotic carotid stenoses after prior CEA generated minimal embolic debris compared with primary stenoses. Four of 15 patients (27%) with restenotic lesions demonstrated embolic particles; all debris was <500 μm. All 41 patients with primary stenoses had some embolic debris; particulate size was >200 μm in 91%, >500 μm in 72%, and >1000 μm in 43%. In primary lesions, the number and size of captured particulate correlated with GSM and with the combined ultrasound findings of echogenicity, heterogenicity, and luminal irregularity/ulceration (P < .02, 95% confidence interval, 4.5-27.6). None of these ultrasound factors correlated independently with embolic particulate (P = NS). Patients aged >70 years exhibited more total particles (8.1 vs 2.3, P = .008) and increased mean particle size (370 vs 157 μm, P = .02). No significant correlation was observed between the number and size of captured embolic particulate and any other variable (stenosis percentage, prior radiotherapy, preprocedural symptoms, periprocedural symptoms, and calcification). Histologically, the embolic debris consisted of extensive amorphous, acellular proteinaceous material. Calcium debris in the embolic particulate was associated with heavily and moderately calcified lesions.
Conclusions
Considerable variation exists in the number and size of embolic particles generated during CAS. Embolic potential is positively correlated with lesion GSM and the combination of lesion echogenicity, heterogenicity, and irregularity. Restenosis after prior CEA is associated with minimal embolic particulate generation, suggesting that embolic protection may not be necessary for CAS of restenotic lesions.
Atherosclerotic disease of the carotid artery is a frequent cause of ischemic strokes, resulting in substantial morbidity and death each year in the United States. Each year >700,000 new strokes are diagnosed, and nearly 150,000 patients die as a result of stroke.1 Approximately 20% to 30% of these strokes are caused by atheromatous disease of the carotid artery.2
Carotid endarterectomy (CEA) emerged during the past 50 years as the standard treatment to reduce the risk of a cerebrovascular event (CVA) in patients with high-grade occlusive disease of the extracranial carotid artery. Carotid angioplasty and stenting (CAS) is currently being used as an alternative to CEA, particularly in patients considered at increased risk for standard surgery; however, the safety of CAS with regard to periprocedural cerebrovascular events is the subject of ongoing investigation. Embolization of atheromatous and thrombotic debris during angioplasty and stent deployment represents one of the significant risks of this procedure.
Multiple factors increase the risk of embolization from carotid lesions. Investigators have shown that anatomic considerations and plaque characteristic are likely the patient-dependent factors that lead to the generation of debris.3 Tortuosity, severe angulation, anomalous vessel origins, and diffuse plaque burden are all potential sources of emboli. Most of these factors are correlated with increasing age.4, 5 A more important finding is that the characteristics of the carotid lesion itself have been correlated with embolic potential. Intraplaque hemorrhage and echolucent appearance on ultrasound imaging have been associated with increased rates of symptomatic presentation in patients undergoing CEA. In addition, restenotic intimal hyperplastic carotid lesions that occur after CEA are associated with lower rates of symptomatic presentation compared with primary atherosclerotic carotid stenosis.
This study evaluated the relationship between the size and number of embolic debris retained in the embolic filter during CAS procedures with lesion type (primary or restenotic), lesions characteristics, and patient age.
Materials and methods
Patients
Institutional Review Board approval was obtained prior to the commencement of this study. An analysis was done of the captured embolic debris in 56 consecutive unselected embolic protection filters used during CAS procedures performed by a single vascular surgeon. The patients for whom captured particulate material was available were a mean age of 74 years (range, 60-94 years). Procedures were performed for primary atherosclerotic occlusive disease in 41 of the 56 patients (73%), and 15 (27%) underwent CAS for restenosis after a prior CEA (all patients had undergone CEA ≤2 years of developing restenosis).
Medical comorbidities placing patients at high risk for CEA were present in 79%, and 7% had undergone external beam radiotherapy for the treatment of a cervical malignancy. Symptoms were present in 15 of the 56 patients (26%), including transient ischemic attack in 3, amaurosis fugax in 5, minor CVAs in 8, and 1 patient had combined symptoms of amaurosis fugax and CVA. There was no increase in the percentage of symptomatic patients aged >70 years.
CAS procedure
All procedures were in a hybrid angiographic operating room (Siemens AG, Munich, Germany). Patients were pretreated with clopidogrel (75 mg/d) for 5 days or received a single loading dose (450 mg) 4 hours before the procedure. The procedure was performed with the aid of local anesthesia, under fluoroscopic guidance through a common femoral arterial puncture site. All patients received systemic heparinization (100 U/kg), and activated clotting time values were monitored throughout the course of the procedure, with supplemental heparin given for values <250 seconds.
Embolic protection devices were used in all patients, including EPI FilterWire (Boston Scientific, Natick, Mass), Accunet (Guidant, Minneapolis, Minn), and Angioguard (Cordis, Somerville, NJ). Stents implanted included the Wallstent (Boston Scientific), NexStent (EndoTex, Cupertino, Calif), Acculink (Guidant), and Precise (Cordis). The choice of embolic protection device and stent was determined by vessel anatomy, operator preference, and clinical trial participation.
The embolic protection devices and stents were delivered through a 6F transfemoral sheath placed in the common carotid artery. The lesion was crossed initially with the embolic protection device and was predilated with a 4- × 40-mm angioplasty balloon (Long VIVA, Boston Scientific). Postdilation was performed using a 5- to 6-mm × 20-mm angioplasty balloon (Gazelle, Boston Scientific).
Filter preparation
The exterior surface of the embolic protection device was wiped with moistened gauze after retrieval (Fig 1). Filters were then rinsed with saline, blotted dry, and fixed in a 10% neutral buffered formalin solution for 24 hours. Photomicrographs were taken of the interior of the filters and used to determine the number and size of individual particles. Irregularly shaped particles were measured along their longest axis. Histologic analysis of recovered material was performed with hematoxylin and eosin and trichrome stains (Fig 2). Most plaques were atheromatous—not fibrous—as indicated by preprocedural ultrasound analysis. Von Kossa staining was used to determine the presence of calcification in the specimen (Fig 3). Specimens were also analyzed using scanning and transmission electron microscopy.6 Energy-dispersive spectroscopy was used to confirm the presence of calcium in the captured embolic debris (Fig 4).
Fig 1.
Captured particulate debris in filter.
Fig 2.
Photomicrographs of hematoxylin and eosin staining under the dissecting microscope (left, bar, 300 μm; right, bar, 60 μm) of the captured particulate show amorphous, largely acellular proteinaceous material.
Fig 3.
The presence of calcium in the captured embolic debris was correlated with preprocedural ultrasound imaging and angiographic evidence of heavily and moderately calcified lesions (bar, 60 μm).
Fig 4.
Energy dispersive spectroscopy confirmed presence of Ca2+ (3.69 keV).
Gray scale median calculation
All patients underwent a preprocedural duplex ultrasound (DU) examination. B-mode DU analysis was performed to assess plaque echogenicity, heterogeneity, ulceration, and irregularity. Echolucency can be measured using the gray scale median (GSM), which is an objective and quantitative computer-assisted grading of the echogenicity of carotid plaques. The GSM was calculated using Photoshop (Adobe, San Jose, Calif) from normalized B-mode DU VHS video recordings. Pixel analysis of the digitized plaque image (0 representing black and 255 representing white)7 was used to generate a median value for all pixels in the image. Lesions with higher GSM values have increased echogenicity, and those with lower GSM values are more echolucent. Calcification and degree of stenosis were determined angiographically.
Statistical analysis
Patient demographics, lesion, and captured particulate matter characteristics were prospectively collected in a computerized database. Statistical analysis was performed using SPSS software (SPSS, Chicago, Ill). Statistical significance was defined as P < .05. The t test was used for continuous variables, and Fisher's exact test was used for discrete variables.
Results
Patient and lesion characteristics
Technical success was achieved in all patients. The mean angiographically determined internal carotid artery stenosis was 88% (range, 70%-99%). There was no outcome difference in relation to the filter or stent used. Two patients aged >70 years had mild intraoperative symptoms that resolved. There were no postprocedural CVAs.
Primary vs restenotic lesions
Particulate matter from 56 filters was analyzed. The procedures were for primary atherosclerotic occlusive disease in 41 patients (73%) and for restenosis after prior CEA in 15 (26%). All patients with restenosis presented ≤2 years of the primary procedure. Embolic debris was recovered from 100% of primary lesions and 27% of restenotic lesions (P < .001). The mean number of particles recovered in each specimen correlated with lesion type (13.2 for primary vs 0.7 for restenosis; P < .005; Table I).
Table I. Embolic debris size correlates with lesion type
| Lesion type | Total No. | Patients with embolic debris | ||
|---|---|---|---|---|
| No. (%) | >500 μm, % | >1000 μm, % | ||
| Primary | 41 | 41 (100) | 72 | 43 |
| Restenosis | 15 | 4(27) | 0 | 0 |
| P | <.001 | .001 | .004 | |
A broad spectrum of particle sizes was recovered from the filters. Embolic debris ranged in size from <200 μm to >1000 μm. The mean captured particulate size among primary atherosclerotic lesions was 382 μm compared with 62 μm from restenotic lesions (P < .003; Table II). Owing to limitations in filter pore sizes, small size particles might have entered the brain, but the effect of this was unknown. The distribution in number of particles closely correlated with lesion type. A significant difference in particulate size was observed between primary and restenotic lesions. Debris sizes >500 μm and >1000 μm were strongly correlated with the primary atherosclerotic lesion type.
Table II. Number of debris particles correlates with lesion type
| Lesion type | No. | Particles, No. | Mean size | |||
|---|---|---|---|---|---|---|
| Total | >200 μm | >500 μm | >1000 μm | μm | ||
| Primary | 41 | 13.2 | 8.5 | 4.6 | 1.2 | 382 |
| Restenosis | 15 | 0.7 | 0.3 | 0 | 0 | 62 |
| P | .005 | .006 | .01 | .02 | .003 | |
GSM analysis revealed that a score of <20 correlated with increased echolucency and was associated with an increase in the number and size of embolic particles (P = .007). Prior studies have identified a GSM of 20 as the threshold for increased embolic potential and decreased plaque stability. In primary lesions, the number and size of captured particulate correlated with GSM and with the combined DU findings of echogenicity, heterogenicity, and luminal irregularity/ulceration (95% confidence interval, 4.5-27.6; P < .02). The restenotic lesions tended to be morphologically smoother, without ulceration or irregularity. There was also a significant correlation between captured debris calcium and preprocedural and angiographic evidence of calcium (P = .002).
Age and lesion characteristics
The patients were a mean age of 74 years (range, 60-94 years). When these data were correlated with lesion characteristics for the primary atherosclerotic lesions for the 15 patients aged <70 years, the mean number of particles per patient was 2.3, with a size of 157 μm (P = .008), whereas for the 26 patients aged >70 years, the mean was 8.1 particles per patient, with an increased mean size of 370 μm (P =.02; Table III). The age of 70 was selected as the analysis point because this was shown to be a clinically relevant threshold in previous studies such as Carotid Revascularization Endarterectomy vs Stent Trial (CREST). Mild intraoperative symptoms in two patients aged >70 years resolved, and their filters had large pieces of debris or were full of debris.
Table III. Age correlates with particulate number and size
| Age, y | No. | Mean particulate | |
|---|---|---|---|
| No./patient | Size, μm | ||
| <70 | 38 | 2.3 | 157 |
| >70 | 18 | 8.1 | 370 |
| P | .008 | .02 | |
Discussion
The periprocedural success of CAS depends on the rate of embolization and thereby neurologic events. Plaque characterization to determine embolic potential could be an important first step in a process that may help guide physicians in choosing appropriate procedures for patients with certain plaque features. Several modalities could potentially guide this characterization, including ultrasound imaging and magnetic resonance imaging.
The results of studies that have looked at visible debris in filter devices after CAS support routine filter protection.8 To our knowledge, no studies have correlated embolic debris captured during CAS with preprocedural plaque analysis. In this report, we have provided a microscopic analysis of filter debris in correlation with preprocedural ultrasound imaging, type of lesion, and patient age. We hope that this will improve the understanding of the risk factors for embolization and aid in patient selection for CAS.
Atherosclerosis is the pathologic entity responsible for primary carotid disease in about 90% of patients, and entities such as such as fibromuscular dysplasia, kinking, extrinsic compression, dissections, and radiotherapy are the cause in the remaining 10%. The atherosclerotic plaque consists of a nodular deposition of fat that is primarily cholesterol and highly prone to embolization. Restenotic plaques, on the other hand, have two distinct pathologic plaque characteristics, depending on the time interval. Early (<2 years) restenotic plaques result from intimal hyperplasia and are characterized by the presence of smooth muscle cells, with absence of atherosclerotic plaque characteristics such as calcifications and a large lipid or necrotic core.9 This type of plaque is thought to be more stable and less emboligenic. In contrast, late (>2 years) carotid stenosis is similar to primary carotid artery stenosis, with marked macrophage infiltration, calcifications, and a lipid core, and these plaques are considered as emboligenic as primary plaques.10
Our data analysis showed patients with primary atherosclerotic disease had embolic debris 100% of the time compared with 27% of those with restenotic lesions, a difference that was highly statistically significant. The patients with primary disease had an increased number of particles (13.2) and a larger mean particle size (382 μm) compared with patients with restenotic lesions, which was again highly statistically significant. This relates to the theory that because primary lesions are more friable and unstable, there is a larger risk of embolization during the procedure compared with the early restenotic lesions.
Primary plaques in our study were mostly atheromatous, whereas restenotic plaques tended to be smooth and without ulceration or irregularity. There was a low incidence in our restenotic group because they were early presenters (<2 years after their primary procedure), and their plaque would be expected to be less emboligenic. Although patients with primary plaques tended to have larger-sized debris particles and a greater number, no correlation could be made with neurologic effect due to the low number of intraoperative or postprocedural events.
Preoperative evaluation of carotid plaque using ultrasound imaging may prove to be a valuable tool in planning for the procedure. Previous studies have evaluated plaque echolucency as measured by the GSM score in relation to neurologic events and plaque characteristics ex vivo after CEA. Grogan et al3 demonstrated with B-mode ultrasound imaging that symptomatic plaques are more echolucent and less calcified than asymptomatic plaques and are associated with a greater degree of histopathologic plaque necrosis. Echolucent plaques are also established as a risk factor for neurologic events.11
An inverse relationship exists between carotid plaque GSM values and necrotic core volume on histology, which means that plaques with high necrotic core volume have low GSM values. Prior studies have identified a GSM of 20 as the threshold for increased embolic potential and decreased plaque stability. This study showed that a GSM <20 correlated significantly with increased number and size of embolic particles (P = .007). This supports previous studies that have shown that a low GSM value (more echolucent) correlated with an increased embolic risk and possible neurologic events.12
Interestingly, patients aged ≥70 years had a greater number and larger-sized embolic particles than patients aged <70 years (8.1 vs 2.3 and 370 vs 157 μm), which was statistically significant. There was no significant correlation with the percentage of stenosis, prior radiotherapy, calcium, and preprocedural or periprocedural symptoms (given the low number of neurologic events). This implies that embolization is more likely to occur during the procedure in patients aged >70 years, and this could potentially translate to increased neurologic events and periprocedural stroke risk. Again, we selected an age of 70 as the analysis point because previous studies, such as CREST, have shown this is a clinically relevant threshold.
Conclusion
Our study shows considerable variation exists in the number and size of embolic particles generated during CAS. Embolic potential is positively correlated with lesion GSM values and the combination of lesion echogenicity, heterogeneity, and irregularity. Restenosis after prior CEA is associated with minimal embolic particulate generation, suggesting that embolic protection may be optional for CAS of restenotic lesions. No significant correlation was observed between the number and size of captured embolic particulate and any other variable such as percentage of stenosis, prior radiotherapy, preprocedural symptoms, periprocedural symptoms, or calcification.1
Author contributions
References
- . Impact of stroke. http://www.strokeassociation.org/presenter.jhtml?identifier=1033Accessed Jun 14, 2008
- Early clinical differentiation of cerebral infarction from severe atherosclerotic stenosis and cardioembolism. Stroke. 1992;23:486–491
- B-mode ultrasonographic characterization of carotid atherosclerotic plaques in symptomatic and asymptomatic patients. J Vasc Surg. 2005;42:435–441
- . A critical look at “high-risk” in choosing the proper intervention for patients with carotid bifurcation disease. Sem Vasc Surg. 2007;20:199–204
- . The impact of increasing age on anatomic factors affecting carotid angioplasty and stenting. J Vasc Surg. 2007;45:875–880
- Determining the quantity and character of carotid artery embolic debris by electron microscopy and energy dispersive spectroscopy. J Vasc Surg. 2007;45:716–725
- . Computer-assisted carotid plaque characterization. Eur J Vasc Endovasc Surg. 1995;9:389–393
- . The capture of visible debris by distal cerebral protection filters during carotid artery stenting: is it predictable?. J Vasc Surg. 2005;41:950–955
- The histologic characteristics of primary and restenotic carotid plaque. J Surg Res. 1998;74:27–33
- . Histological characterization of restenotic carotid plaques in relation to recurrence interval and clinical presentation: a cohort study. Stroke. 2008;39:1029–1032
- . Echolucent plaques are associated with high risk of ischemic cerebrovascular events in carotid stenosis: the Tromsø Study. Circulation. 2001;103:2171–2175
- Carotid plaque echolucency increases the risk of stroke in carotid stenting: the Imaging in Carotid Angioplasty and Risk of Stroke (ICAROS) study. Circulation. 2004;110:756–762
Competition of interest: none.
The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a competition of interest.
PII: S0741-5214(09)01782-0
doi:10.1016/j.jvs.2009.08.063
© 2010 Society for Vascular Surgery. Published by Elsevier Inc. All rights reserved.
