Article Text
PDF
Abstract
Introduction Despite various measures to protect against distal embolization during carotid artery stenting (CAS), periprocedural ischemic lesions are still encountered.
Objective To evaluate the periprocedural cerebral diffusion weighted imaging (DWI) lesion burden after CASPER stent placement.
Methods Patients who underwent CAS using the CASPER stent system were reviewed. Degrees of carotid stenosis and plaque configuration were determined. All patients were pretreated with dual antiplatelet agents and cerebral pre- and postprocedural MRI was obtained. All CAS procedures were performed by a single operator.
Results A total of 110 patients with severe carotid artery stenosis (median degree of stenosis 80%, median length of stenosis 10 mm) were treated with CAS. Hypoechogenic or heterogeneous, mostly hypoechogenic, plaques were documented in 48.6% (52/107) of patients. Carotid ulceration was present in 15.9% (17/107). Postprocedurally, 7.3% (8/110) of patients were found to have ischemic DWI lesions. They were asymptomatic in all patients. Follow-up at 90 days was available in 88.2% (97/110) of patients with excellent functional outcome (modified Rankin Scale score 0–1) in 95.9% (93/97).
Conclusion Carotid artery stenting using the new CASPER stent in combination with a distal embolic protection device is safe and results in a lower rate of periprocedural DWI lesion burden compared with reported results for historic controls.
- DWI lesion
- CASPER
- carotid stent
- CAS
Statistics from Altmetric.com
Introduction
Atherosclerotic stenosis of the extracranial carotid artery (CA) may result in ischemic stroke from embolization or reduced cerebral perfusion. Twenty percent of ischemic strokes are associated with CA stenosis.1
Carotid artery stenting (CAS) has been established as an alternative, minimally invasive technique for the treatment of symptomatic and asymptomatic CA stenosis to complement carotid endarterectomy.2
Self-expanding stents have been developed for the treatment of carotid lesions, including open-cell and closed-cell designs.3 The carotid stents available are made of various materials and have different designs and geometry characteristics.3 Several studies have reported that closed-cell stents have some obvious advantages over open-cell stents, such as improved plaque coverage, lesser chance of plaque protrusion through the interstices, and consecutively decreased risk for distal embolization.4 The main disadvantage of a closed-cell stent design is their inability to conform to tortuous vessels.3
The ideal carotid stent should cover the entire plaque, provide good vessel wall apposition to the parent vessel, and be flexible and with good radial force. The body of the stent, with its characteristic design and radial force, provides an intrinsic anti-embolic effect as the stent struts prevent the escape of the disrupted plaque material through the interstices while maintaining the caliber and integrity of the vessel wall.5
The problem of periprocedural ischemic brain lesions on magnetic cerebral imaging (MRI) after CAS is well known, and remains a significant safety concern with CAS. Studies report diffusion weighted imaging (DWI) lesion rates on periprocedural MRI of 22–54%.6 7 The use of embolic protection devices (EPDs) during CAS has reduced the number of periprocedural DWI lesions.8
Over the past 2 years, we have used a new CAS system, the CASPER stent system (MicroVention, Inc., Tustin, California, USA). This nitinol self-expanding stent has a double-layer micromesh design to prevent plaque prolapse.
In this single-center retrospective study, we aim to evaluate the incidence of periprocedural DWI lesion burden associated with the CASPER stent system.
Materials and methods
Consecutive patients with carotid artery stenosis treated with the CASPER stent system between January 2014 and May 2017 were reviewed. Data collected included demographic characteristics, severity of stroke as measured by National Institute of Health Stroke Scale (NIHSS), and functional outcome by modified Rankin Scale (mRS) score at admission, discharge, and at 90 days' follow-up.
Patients eligible for CAS
Patients with neurological symptoms related to a proven stenosis ≥70% of diameter (or >50% diameter with ulceration) or asymptomatic patients with ≥80% stenosis at the carotid bifurcation or at the proximal internal cerebral artery were treated with CAS. All patients had internal CA configuration favorable for EPD placement. Patients with a life expectancy of <12 months or with contraindications to aspirin, clopidogrel, or with a history of severe disabling stroke, and patients with a hemorrhagic stroke were excluded. Informed consent for the procedure was obtained from all patients. CAS performed within 24 hours was classified as acute, whereas CAS performed within 2–10 days was classified as subacute. The other CAS procedures were performed electively.
Antiplatelet regimen
Patients were started on clopidogrel 75 mg and aspirin 100 mg daily at least 10 days before CAS. In all patients platelet function testing, using adenosine diphosphate-induced single-platelet count aggregation in clopidogrel-treated patients and acetyl acid-induced single-platelet count aggregation in aspirin-treated patients, was carried out. In clopidogrel non-responders, the dosage of clopidogrel was increased to 150 mg daily. Alternatively, prasugrel 10 mg daily or ticagrelor 90 mg twice a day were administered. Dual antiplatelet medication was continued for at least 6 months and aspirin lifelong. Patients with acute CA stenosis received either an aspirin and clopidogrel loading dose or tirofiban (initial bolus 0.4 µg/kg/min and then continuously 0.1 µg/kg/min over 24 hours).
Radiological investigation
Ultrasound
The verification and quantification of CA stenosis was determined first by high-resolution color-coded duplex ultrasound, equipped with a 9–3 MHz linear transducer. Carotid artery stenosis was measured using the NASCET study method.9
The sonographic characteristics of the plaques were described using the Gray-Weale classification10: type 1: uniformly anechoic or hypoechoic; type 2: predominantly (>50%) hypoechoic; type 3: predominantly (>50%) hyperechoic; type 4: uniformly hyperechoic. We also used another plaque classification score: no calcification (GW 1), mild to moderate calcification (GW 2), and high calcification (GW 3 and 4). Patients with plaque ulceration were also registered.10 11
Ultrasound examinations were performed before CAS and on postprocedure days 1, 14, and 90. In patients with acute CA stenosis, ultrasound was not performed before the procedure.
MRI
A brain MRI scan was performed first with a 1.5 or a 3.0 Tesla scanner (Siemens, Erlangen, Germany) including high-resolution DWI sequences with b-values from 0 to 1000 and gradient echo T2-weighted axial sequences. Axial fluid attenuation inversion recovery (FLAIR) sequences were performed to measure cerebral white matter change.
DWI sequences were used to detect acute ischemic brain lesions. A corresponding apparent diffusion coefficient map confirmed restricted diffusion signal on the DWI sequences. The number, size (<5 mm, 5–10 mm, >10 mm), distribution (cortical, subcortical, or deep nuclei), and location (ipsilateral, contralateral or bilateral to stenosis) of those signals were recorded.
White matter grading was classified using the graduation score from 0 to 8. Periventricular and subcortical white matter signal abnormalities on MRI were scored from barely detectable (grade 1) to extensive and confluent (grade 8).12 Ischemic lesion location (ipsilateral, contralateral and bilateral hemisphere) was recorded.
Brain MRI was performed before the procedure, 24 hours postprocedurally, and after 90 days. Only the preprocedural MRI included a contrast-enhanced MR angiogram (or alternatively a CT angiogram of the head and neck) for further characterizing the degree of stenosis.
The primary imaging outcome was the presence of any new DWI lesions on the 24-hour postprocedure MRI scan not present preprocedurally. Secondary imaging outcome measures were hyperintensity on FLAIR images at 90 days at the site of at least one postprocedural DWI lesion that was not present on the pretreatment MRI scan; and the presence of any new DWI lesions at 90 days that was not present on the postprocedural MRI scan.
Two independent neuroradiologists (EB and VS) evaluated the images.
Carotid stenting procedure
CAS was performed using a biplane high-resolution angiography system (Artis zee; Siemens, Erlangen, Germany) via a femoral approach. All procedures were performed under general anesthesia and by the same experienced neurointerventionalist (MK-O). A 6 French Envoy guiding catheter (DePuySynthes/Codman Neuro, Raynham, Massachusetts, USA) or a Neuron MAX 088 long sheath, depending on the vascular anatomy (Penumbra, Inc., Alameda, California, USA), was advanced into the target common CA using an 0.035" wire (Terumo, Somerset, New Jersey, USA). When the long sheath was used, an additional 5F Select Catheter (Penumbra, Inc) was also used to gain access to the common CA.
A distal EPD was then guided using a 0.014" microguidewire (FilterWire EZ Embolic Protection System, Boston Scientific, Marlborough, Massachusetts, USA) through the internal CA stenosis into the distal part of the internal CA. The EPD was preferably deployed in a straight segment of the extracranial cervical internal CA, according to the manufacturer’s instructions. After EPD deployment, the vessel diameter was measured. In the case of high-grade stenosis, a prestenting balloon angioplasty was performed. Next, the CASPER stent was implanted. For postdilatation, a balloon angioplasty catheter with a medium-high pressure of 7 atm was used. After obtaining a control angiogram, the EPD was retrieved with the retrieval sheath and the groin puncture was closed with a vascular closure device (Angio Seal St. Jude Medical/Daig, Minnetonka, Minnesota, USA or Star Close Abbott Vascular, Redwood City, California, USA).
Technical aspects of the CASPER stent system
The CASPER stent system has a closed-cell design with open-cell mechanical performance. The dual-layer micromesh design has the smallest stent cell size area of all carotid stents. The average pore size of 390–700 µm is built to contain plaque, acting like a metallic covered stent. The stent is available as a 5 French rapid exchange system, and is made up to 50% deployment fully resheathable and repositionable. The appropriate implanted length is relative to the vessel diameter and the desired landing zones. The stent has flared ends, which add about 5 mm to the dual layer length on each end.
Data analysis
Statistical analysis was performed with STATA 13.0 (Texas, USA) statistical package. We used the non-parametric Kruskal-Wallis test owing to non-normality of continuous data. Categorical variables were examined using the two-tailed Fisher exact test and Pearson χ2. A level of P<0.05 was considered as statistically significant. Patients with and without new DWI lesions after CAS were compared.
Results
Demographics and baseline characteristics
The median age of patients was 71 years (22–85). Fifty-nine percent (65/110) of patients had symptomatic CA stenosis. Symptomatic patients had a median NIHSS score of 2 (range 0–25) on admission. All patients included for CAS had a valid platelet function testing. In two patients (1.8%, 2/110), clopidogrel was raised to 150 mg and three patients (2.7%, 3/110) were identified as clopidogrel non-responders and switched to ticagrelor. Three patients (2.7%, 3/110) received acute CAS with two of them (66.7%, 2/3) treated with tirofiban and the other 1 (33.3%, 1/3) with a clopidogrel and aspirin loading dose. Subacute CAS treatment within 2–10 days was performed in 36.4% of patients (40/110) (table 1).
Patients demographics and procedural characteristics
Preprocedural imaging characteristics
The median degree of stenosis was 80% (70–90%) and the median length of stenosis was 10 mm (2–30 mm). In 51.4% (55/107) of patients, hyperechoic high calcification plaques were detected while hypoechoic and heterogeneous plaques were found in 25.2% (27/107) and 23.4% (25/107), respectively. Additional ulceration was documented in 15.9% (17/107).
Most patients (75.5%, 83/110) were diagnosed with white matter grading of a maximum of a thin, continuous periventricular rim with a few patches of subcortical disease (0–2). A prior chronic infarction was documented in 21.8% (24/110) and a new ischemic infarction in 30% (33/110) (table 2; figure 1).
Imaging characteristics
(A and B) Preprocedural brain MRI (apparent diffusion coefficient (ADC) and DWI) shows no fresh ischemic lesions (MRI was performed at an external MRI institution). (C) Angiogram shows reduced flow intracerebrally, particularly in the vascular territory of the anterior cerebral artery. (D) Angiogram shows high-grade carotid stenosis (80%). (E) After implantation of a CASPER stent system the angiogram shows improved flow in the territory of the middle and anterior cerebral arteries. (F) A CASPER stent system (8×30×143) was implanted. After angioplasty with a balloon catheter (4.5×20×135), residual, hemodynamically non-relevant, stenosis of 20% was present. (G and H) Postprocedural brain MRI (ADC and DWI) shows a very small cortical fresh ischemic lesion. However, the patient remained clinically asymptomatic.
Periprocedural and postprocedural complications
There were four (3.6%; 4/110) non-ischemic procedural complications. One patient (0.9%, 1/110) had a dissection of the external iliac artery from groin puncture while a different patient had excessive groin bleeding. One patient (0.9%, 1/110) had postprocedural paresthesia of the leg at the groin puncture site. One patient (0.9%, 1/110) had an asymptomatic small cerebral hemorrhage inside a pre-existing infarction. This patient remained clinically asymptomatic. None of the patients was documented with thromboembolic complications.
Postprocedural outcome and imaging
The median NIHSS score at discharge remained 0 (range 0–14). Ultrasound control at the time of discharge did not show any restenosis.
Postprocedural MRI at 24 hours showed new DWI lesions in 7.3% (8/110) of patients, including five ipsilateral, two contralateral, and one in bilateral hemispheres. Fifty percent (4/8) of the lesions were <5 mm and 50% (4/8) between 5 and 10 mm. Location of the lesions was subcortical, in the deep nuclei, and cortical areas in 62.5% (5/8), 25% (2/8) and 12.5% (1/8) of patients, respectively. In four patients, there was only a single DWI lesion (50%, 4/8). Two patients (25%, 2/8) had two DWI lesions, one patient had three lesions (12.5%, 1/8) and one had five lesions (12.5%, 1/8). Patients with new ischemic lesions were documented with echolucent or heterogeneous in 37.5% (3/8) and with hyperechoic plaques in 62.5% (5/8). All patients with DWI lesions were clinically asymptomatic. A comparison of the 7.3% (8/110) patients with new DWI lesions with those without is presented in tables 1 and 2. A detailed analysis of the DWI lesions of those 7.3% (8/110) of patients is shown in table 3.
Overview of size and distribution of new DWI lesions after stent placement
Follow-up at 90 days
At 90 days, 88.2% (97/110) patients were available for follow-up, of whom 95.9% (93/97) had a mRS score 0–1, 3.1% (3/97) a mRS score 2, and 1.0% (1/97) a mRS score of 3. MRI was available in 94.8% (92/97). MRI did not show any new DWI lesions compared with the postprocedural MRI. In 50.0% (4/8) of the postprocedural DWI-positive patients, the lesions resolved and in another 50.0% (4/8) lacunar parenchymal defects were detected.
Two patients (2.1%, 2/97) received a CT of the head (one of those owing to placement of a cardiac pacemaker in the interim). In these two patients no fresh cerebral ischemia was documented. In 3.1% (3/97) of patients no cerebral imaging was available.
Ultrasound control at 90 days did not show restenosis.
Discussion
This is a comprehensive study of 110 patients to assess DWI lesions after CASPER stenting. It demonstrated a rate of 7.3% DWI lesions within 24 hours of stenting. No restenosis occurred during follow-up and none of the patients had a new cerebrovascular event. All patients with ischemic DWI lesions on postprocedural MRI remained asymptomatic. These results are similar to reports from randomized clinical trials such as the SAPPHIRE and CREST trials,13 14 which reached the conclusion that CAS was not inferior to carotid endarterectomy).13 14 According to prior studies,8 15 all patients had an EPD placed during the CAS. There was no technical difficulty in combining the CASPER stent with an EPD device. This was a problem we experienced with the Cristallo stent (Meditronic Invatec Corporation, Brescia, Italy) in the past, where the EPD sheath was caught at the open-cell and closed-cell junction of the Cristallo stent on removal.
Technical aspects, like resheathing and stent deployment were excellent, and successful in all patients.16–19
The detected DWI lesions were in the downstream vascular territory and located at the corticosubcortical junction, as demonstrated in former studies.20 All lesions were <10 mm in size and similar to prior studies.20 Likewise, all patients remained asymptomatic.21
In our data, there was a difference in age in patients with CA stenosis with or without cerebral embolization. There is a trend towards ischemic lesions after CAS in older patients, as described in a former study.22
Despite the obvious tolerance of the brain to microembolization after CAS, these DWI lesions may cause subtle neurologic and neurocognitive dysfunction as previously reported.23 However, our study did not perform sophisticated neurocognitive testing.
Other CAS studies using conventional stents such as Carotid Wallstent (Boston Scientific, Natick, Massachusetts, USA), Acculink (Abbott Vascular, Santa Clara, California, USA), or Herculink (Abbott Vascular, Abbott Park, Illinois, USA), documented higher rates of postprocedural DWI ischemic cerebral lesions, ranging from 22.7% (n=42),6 to 54% (n=41).24 Additional CAS studies with conventional stents recorded also higher rates of fresh cerebral lesions, ranging from 68% (n=56) for the MRI International Carotid Stenting Study subgroup,25 to even 87.1% (n=31) for PROFI.26 Studies that analyzed CA stenosis with non-conventional stents showed lower rates of new ischemic lesions after CAS. The CARENET (Carotid Embolic Protection Using MicroNet) study that tested the non-conventional CGuard stent (InspireMD, Boston, Massachusetts, USA) in 26 procedures with an EPD showed a rate of new ischemic lesions of 34.6% (n=26).27 Another study tested the Roadsaver (Terumo, Leuven, Belgium), a double-mesh stent system for CAS, and found new ischemic lesions in 30.4%, almost equal to the CARNET study.17 This Roadsaver study also implemented an EPD but a large amount of the patients with DWI lesions (57.1%) was documented with echolucent plaques.17
Other available studies of double-layer stents did not perform postprocedural brain MRI, but reported good technical and functional outcomes in line with our results.4 16 18 19
Rates of new ischemic lesions are lower by far in patients treated with double-layer stents, like the CASPER stent, than in patients treated with conventional stents. Important contributors to cerebral periprocedural embolization in self-expanding stents are the pattern of the stent struts and their interconnections.28 All available stents have differences of stiffness, radial force, flexibility, adaptability, and conformability to the vessel wall, and in the scaffolding effect to reduce plaque prolapse and embolization.29
Studies observed that closed-cell stents have obvious advantages over open-cell stents in plaque coverage and consecutively cerebral embolization, but their main disadvantage is their lack of flexibility in tortuous vessel anatomy.28
The CASPER double-layer micromesh stent is a low-profile braided stent with increased mesh density and reduced pore size compared with other stents. The dual-layer micromesh design helps plaque coverage. This was proved by a recently published study that confirmed a low rate of plaque protrusion in nine patients treated with the CASPER stent, examined by optical frequency domain imaging.4 Umemoto et al analyzed the Roadsaver stent with optical coherence tomography and reported an incidence of plaque prolapse in 20.7% in five patients after stenting.30 No neurological complications (stroke or transient ischemic attack) occurred during the procedural and postprocedural periods. In comparison with conventional stents, mesh-covered stents, like the CASPER, showed less frequent plaque prolapse.30 The CASPER stent is comparable in flexibility to open-cell stents with an excellent conformability to the vessel wall. It appears to reduce the burden of cerebral DWI lesions owing to its advanced engineering technology according to a recently published study about braided stents and as demonstrated in our study.31 The thin layer of endothelial cells is seen earlier than in braided stents and braided stents are also associated with less neointimal development than single-layer stents.31
The low rate of cerebral DWI lesions in the present study, even in comparison with the recently published Roadsaver study, may be explained by the antiplatelet regimen. Patients of the Roadsaver study were treated inconsistently with dual antiplatelets preprocedurally.17 For patients with no preprocedural clopidogrel medication a loading dose of 375 mg was given, also dual antiplatelet therapy was prolonged for only 1 month.17
All patients of the present study with elective and subacute CA stenosis received dual antiplatelet treatment for at least 4 days preprocedurally, except those who were treated with acute CAS. Platelet function testing was performed routinely. Although the CASPER stent has an excellent mesh design, it requires judicious antiplatelet management maintained for at least 3 months.
In this series, 48.6% of patients were documented as having plaques with a higher risk for embolization.
Plaque prolapse through the stent cells, intraluminal manipulation of the lesion, and echolucent plaques are risk factors for distal embolization after stenting. Here, we had a relatively high rate of non-calcified (25.2%), low calcified plaques (23.4%), and ulcerated carotid stenosis (15.9%). Several studies have reported that carotid ulceration and carotid plaque echolucency are associated with the development of neurologic events and lead to an increased number of emboli after CAS.16 32 Also, the ICAROS study with 418 patients found that carotid plaque echolucency increased the risk of stroke in CAS.32 Although the rate of echolucent plaques was relatively high in our cohort, we documented asymptomatic new DWI lesions in only 7.3% of patients after CAS. This may be a testament to the innovative design of the CASPER stent along with the use of an EPD during CAS, which is by itself protective against distal embolization. In a future study, this analysis should be solidified. In high-volume neurointerventional centers where CAS represents the standard of care, this new hybrid stent concept is very promising.
A limitation of our study is the lack of randomization and that only one very experienced interventionalist performed all the procedures.
Conclusion
Carotid artery stenting using the new CASPER stent in combination with a distal embolic device is safe and results in a lower rate of periprocedural DWI lesion burden compared with reported historic controls.
References
Footnotes
Contributors All authors gave their final approval for publication. EB: data collection, manuscript preparation. CG, SM: manuscript preparation and critical review. SP: data manuscript preparation, statistics and critical review. HJ: data manuscript preparation and critical review. VS: data collection, image review. MK-O: study design, manuscript preparation and critical review.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests The senior author of this study receives independent research grant funding from Microvention/Terumo.
Patient consent Not required.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement No data from this study are available elsewhere.