Subclinical embolization after carotid artery stenting: New lesions on diffusion-weighted magnetic resonance imaging occur postprocedure
Article Outline
Objectives
The reported rate of subclinical brain injury after carotid artery stenting (CAS) seen on diffusion-weighted magnetic resonance imaging (DWI) varies from 10% to >40%. Data from transcranial Doppler after CAS indicate that embolization may continue for several days, suggesting that that at least some lesions seen on DWI occur postprocedure. Because DWI lesions appear ≤1 hour of embolization, we used DWI to prospectively study patients before CAS, 1 hour after, and 48 hours after CAS to answer this question.
Methods
The study participants were 48 male patients aged 59 to 83. All patients were examined by a neurologist before and after the procedure and had DWI preprocedure and 48 hours postprocedure. In addition, 23 patients had a DWI 1 hour post-CAS. Magnetic resonance imaging exams, including axial and coronal DWI and fluid-attenuated inversion recovery images, were read by two neuroradiologists blinded to the study timing. The embolic protection device was obtained from all patients, washed, and the contents examined under a digital microscope for fragments ≥60 μm.
Results
There were two periprocedural strokes and one transient ischemic attack (TIA), but no strokes or TIAs occurred during follow-up. In the 23 patients imaged 1 hour postprocedure, new lesions were found in two (9%), and 18 (78%) had new lesions at 48 hours (P < .001). For the entire study group, the incidence of new lesions at 48 hours was 67% (36/54). The median number of DWI lesions was four (range, 1 to 17). Every protection device examined had atherosclerotic debris, with a mean of 135 ± 73 fragments (range, 18 to 310) sized >60 μm and a mean of eight fragments (range, 2 to 21) sized >500 μm. Findings on postprocedure DWI did not correlate with the degree of stenosis, size of angioplasty balloon, or number of inflations, nor with the number or size of fragments retrieved from the protection device.
Conclusions
CAS can be performed with a very low incidence of clinically evident neurologic events; however, it is associated with embolization during and after the procedure. Protection devices effectively prevent clinical and subclinical events during the procedure. Significant embolization continues for at least 48 hours postprocedure, causing lesions on DWI when there is no mechanism for cerebral protection. These data correlate with transcranial Doppler reports of continued embolization after CAS and indicate that DWI should be done as late as possible to accurately assess the rate of subclinical brain injury with CAS procedures.
Ex vivo experiments1, 2 and in vivo transcranial Doppler (TCD) monitoring3 have shown that carotid angioplasty can release large numbers of plaque fragments into the cerebral circulation. Thousands of plaque fragments were retrieved downstream after ex vivo angioplasty, whether or not the lesion was stented.4 To protect the brain from this embolic barrage, carotid angioplasty is now routinely performed with a sieving, or embolic protection device, placed in the distal artery. There is concern about the degree of protection afforded by these devices, however.5 Plaque material may be dislodged as the device is maneuvered through the stenotic lesion before deployment. The devices have a pore size of approximately 100 μm, a particular concern because our ex vivo data suggest that most particulates released with angioplasty have a maximum diameter <100 μm.1
TCD monitoring of patients undergoing carotid angioplasty has confirmed that particulates enter the cerebral circulation even when protection devices are in place.6 Although in most cases the number of particulates detected did not correlate with new neurologic symptoms, tens to several hundred solid emboli were detected.3, 6
Because of our concern about these cerebral emboli, a protocol was established as a collaboration between the Vascular Surgery, Radiology, and Neurology Services when the carotid stenting program was initiated at our facility, the San Francisco VA Medical Center, to examine the causes and consequences of these cerebral emboli with the goal of reducing their number to a minimum. This required a prospectively accumulated database, which included documenting the incidence of both clinically evident and subclinical brain injury using diffusion-weighted magnetic resonance imaging (DWI, MRI).
Previous reports using DWI to examine subclinical brain injury after carotid stenting have looked at only one study postprocedure. When studies have been done ≤24 hours, new lesions were observed in 17% to 53% of cases.3, 7, 8, 9 Reports of DWI done at 48 hours show a trend toward a higher incidence,10, 11 possibly because of the continued embolization seen on TCD monitoring.12 To ensure that we included the full extent of subclinical injury post-CAS, we initiated our series by imaging patients as late as practical, at 48 hours. When we also observed a significant number of new lesions on DWI at this time point, we realized that to determine what might be done to reduce the incidence, the first step was to clarify whether the lesions were the result of intraprocedural or postprocedural embolization. Therefore, a third study at 1 to 2 hours postprocedure was added to the protocol.
Methods
The study was done as a prospective, nonrandomized examination of carotid angioplasty and stenting (CAS) conducted from February 2005 through August 2006. Patient selection was done as follows. A discussion was undertaken with each patient by a member of the research team or the referring practitioner about the relative risks and benefits of CAS vs endarterectomy, which included a review of the aortic arch and proximal carotid anatomy to assess the feasibility of CAS. Thereafter, patients who chose to undergo CAS were approached to participate in the study. The study population consisted of 48 male patients with a mean age of 71 years (range, 59 to 83 years). They underwent 54 procedures for asymptomatic critical stenoses (n = 25) or symptoms of transient ischemic attack (TIA; hemispheric in 15, amaurosis fugax in 5) or cerebrovascular accident (n = 9). During the same period, nine patients undergoing CAS either declined to participate or could not undergo MR scanning, and 38 patients underwent carotid endarterectomy. The protocol and consent form were approved by the Committees on Human Research at University of California, San Francisco, and San Francisco Veterans Affairs Medical Center.
There were two phases of the protocol for identifying new lesions on DWI. In the first 31 cases, we obtained DWI at 48 hours only. The procedure was then changed to include DWI at 1 to 2 hours postprocedure and at 48 hours in 23 cases. All 54 cases undergoing CAS and DWI are reported.
To ensure that the lesions seen on the postprocedure DWI were new, each subject had a study ≤72 hours before the procedure and postprocedure studies at 1 hour and 48 hours, as noted above (Figure 1). The DWI included axial and coronal DWI and fluid-attenuated inversion recovery (FLAIR) images (DWI: echoplanar spin-echo, TR/TE = 5000/100 milliseconds, b = 0.500, 1,000, 20 5-mm thick slices with a 1.5-mm gap, matrix size-128 × 128; FLAIR: TR/TE/TI = 8000/120/2000 milliseconds). Average diffusion coefficients maps were calculated from the DWI images.
Fig 1.
Diffusion-weighted images before, 1 hour, and 48 hours after carotid stenting show three small, new lesions in the right hemisphere.
The MRI studies were independently read by two neuroradiologists (M. W. and C. G.) who were blinded to study timing. An acute ischemic lesion was diagnosed when it was seen on either axial or coronal DWI images and confirmed on the corresponding average diffusion coefficients maps, or was seen on both planes of DWI. To be considered indicative of acute injury, the lesion could not appear on preprocedure imaging and no corresponding FLAIR abnormality could be present for the lesion to be considered as acute. New (acute) lesion location and size were recorded. In cases of initial disagreement between readers, consensus was reached by joint review of the cases.
Every patient was examined by a neurologist, both before and after the procedure, using the National Institutes of Health Stroke Scale.
The carotid stenting procedure was performed as follows. In addition to the DWI, a preprocedure magnetic resonance angiography was done that included a three-dimensional display of the aortic arch; therefore, arch angiography was avoided in most cases (81%). After achieving access to the proximal thoracic aorta, a telescoping technique using the Shuttle Select Sheath (Cook, Bloomington, Ind) and the JB1 (right carotid; Cook) or V-Tek (left carotid; Cook) catheters was used to cannulate the appropriate great vessel.
Angiography to confirm the extracranial and intracranial anatomy was performed, and the embolic protection device was placed into the distal internal carotid artery. The AccuNet device (Guidant, Mountain View, Calif) was used in 51, and the Angioguard (Abbott, Abbott Park, Ill) was used in three. Predilation angioplasty, stent deployment (AccuLink, Guidant; or Xact, Abbott), and postdilation angioplasty were performed (Figure 2).
Fig 2.
A, Preprocedural and B, postprocedural views of a critical internal carotid lesion. Note the continued presence of a small area of luminal irregularity extending outside the stent.
After angiography to confirm satisfactory treatment of the stenosis, the protection device was removed and completion angiograms were obtained. Periprocedural anticoagulation consisted of pretreatment with clopidogrel in all patients, which was continued postprocedure for 6 weeks; intraprocedural heparin given to achieve an activated clotting time of >300 seconds and then continued for 12 hours postprocedure.
An embolic protection device was used in all cases. This was retrieved for fragment analysis in 44 cases. The collapsed filter basket was rinsed with saline to remove adherent blood. After re-expansion, the contents were obtained by washing with approximately 20 mL of saline, which was then centrifuged, the supernatant decanted, and the fragments examined at ×120 under a Zeiss Digital Microscope (Carl Zeiss, Oberkochen, Germany). Photos were taken and fragments measured and counted. With each specimen, a sham control was also done to monitor the debris that may have been included in the saline and glassware. This was subtracted from the fragment count.
Results
One TIA occurred and two periprocedural strokes (stroke rate, 3.7%). One stroke occurred 6 hours after a left internal carotid artery CAS and affected the fine motor function of the patient’s right hand. The second occurred after plaque fragments released during postdilation of the stent occluded the embolic protection device, requiring passage of a second wire and catheter to aspirate the atherosclerotic debris and restore flow. The symptoms were confusion and a mild facial droop and were associated with an area of brain infarction on MRI, the only infarct on MRI in the series. In both patients symptoms had completely resolved by their 2-week follow-up visit. There were two myocardial infarctions, but no deaths. One femoral artery pseudoaneurysm required thrombin injection, and two groin hematomas requiring evacuation occurred early in the series.
Four patients had areas of acute injury on DWI before their carotid artery stent procedure. In the 23 cases where DWI studies were performed 1 to 2 hours postprocedure, only two (9%) had new lesions immediately postprocedure, and 18 (78%) had new lesions when imaged at 48 hours (P < .001; Table I). In the entire group there were new lesions at 48 hours after 36 (67%) of the 54 procedures. The median number of new lesions on DWI was four (range, 1 to 17). Most measured 1 to 2 mm in diameter, with the largest lesion measuring 7 mm. There were new lesions ipsilateral to the procedure in 97% of cases. In 28%, lesions were seen in the contralateral hemisphere as well. In one case, there were two lesions in the contralateral hemisphere and none ipsilateral to the CAS procedure.
Table I. New lesions on diffusion-weighted imaging
| Group | Patients (n) | Positive at 1 h | Positive at 48 h |
|---|---|---|---|
| Imagined at | 54 | — | 36 (67%) |
| 48 h | 54 | — | 36(67%) |
| 1 h and 48 h | 23 | 2(9%)⁎ | 18(78%)⁎ |
⁎P < .001. |
When viewed with the digital microscope and the appropriate sham subtractions were done, every protection device examined had atherosclerotic fragments, with a mean of 135 ± 73 (range, 18 to 310) fragments sized >60 μm, and a mean of eight fragments (range 2 to 21) sized >500 μm (Table II) (Figure 3). We examined the relationship of the number and size of the fragments retrieved from the embolic protection device with the degree of plaque stenosis, conduct of the procedure, and findings on postprocedure DWI. There was no correlation of the number or size of fragments retrieved from the embolic protection device and degree of stenosis, size of angioplasty balloon, number of inflations, or number of new lesions on DWI.
Table II. Number of fragments >60 μm captured in the embolic protection device
| Size (μm) | Mean | SD |
|---|---|---|
| 60-99 | 84.4 | 45.7 |
| 100-199 | 24.2 | 31.0 |
| 200-499 | 11.1 | 11.0 |
| 500-999 | 3.9 | 4.5 |
| >1000 | 1.6 | 2.8 |
| Total | 135.2 | 73.5 |
Fig 3.
Debris recovered from an embolic protection device. The range of sizes includes atherosclerotic tissue fragments above and below 100 μm in size.
Discussion
Our data indicate that as carotid stenting is currently practiced at our institution, nearly all embolic brain injury occurs postprocedure. Although some late ischemic events were expected to occur, we can only speculate why, during a time period when thousands of plaque fragments are released,1, 3 injury to the brain is rare, while later, when the embolic rate is considerably lower, brain injury is virtually the norm.
Our first thought was that the embolic protection device was highly effective. Although the pore size of the AccuNet that was used for most our cases is 120 μm, it should entrap larger fragments that are likely to cause ischemia.4 Furthermore, it is more effective than one might predict in trapping fragments sized <100 μm.
Second, the published TCD data gleaned from only 20 or 30 minutes of monitoring may underestimate both the incidence and frequency of emboli postprocedure. A more thorough examination of embolic burden after carotid stenting will require longer periods of TCD monitoring.
Third, there could be late thrombus formation on the stent or propagation of clot from a nonoccluding embolus in the microvasculature. Heparin was discontinued 12 hours after the procedure, although daily clopidogrel was prescribed for 6 weeks. We have found in an animal model of embolic stroke that even high concentrations of heparin have no impact on the incidence or number of brain lesions from microemboli (unpublished observations), making this explanation unlikely.
Finally, it is also unlikely that there may have been areas of injury that were not imaged by the 1-hour postprocedure DWI study. DWI has been shown to be positive within minutes of injury in both animals and humans,13, 14 and no scan was done less than 1 hour after stent deployment and postdilation angioplasty.
New lesions were seen on DWI after two of every three CAS procedures in this series, which is nearly double the rates reported after 24 hours.7, 9, 15 It is also somewhat higher than other reports of DWI done 48 hours postprocedure.10, 11 Some of the difference may be technical. Most studies report only single-plane imaging, but our DWI was done in both the coronal and axial planes. If only the coronal sections were read, the number of lesions was reduced by one third, placing our data in the mid-range of the other reported incidences. Clearly, the more complete evaluation of brain injury performed at least 48 hours postprocedure is to be preferred if one is examining the incidence of new lesions on DWI.
CAS is being introduced as an alternative to carotid endarterectomy, which rarely is accompanied by new lesions on DWI. We16 and others17 have shown that new lesions are rarely seen on DWI after carotid endarterectomy, and in a comparison of CAS and endarterectomy, Roh et al18 found that both neurologic events and new lesions on DWI were far more common with CAS.
In contrast to the incidence of subclinical injury, the clinical outcomes in this series are comparable with carotid endarterectomy, raising the question of the importance of these lesions seen only on DWI. It would be naive to reflexly respond that all brain injury is to be avoided. In the short term, these lesions seem to have no measurable consequences, and by 6 months post-CAS, most have resolved without residual effects.9
Repetitive embolic injury may have a cumulative effect, however. Recently, a link has been established between the numbers of emboli found during TCD monitoring in patients with vascular dementia and Alzheimer disease compared with nonaffected controls.19 Although much work needs to be done to understand the mechanism of repetitive emboli and dementia, it is premature to consider these subclinical events unimportant, and until an approach is developed that will reduce their incidence, they remain a cause of concern.
We used a digital microscopy with ×120 power to examine the material washed from the embolic protection devices. With this level of scrutiny, all devices contained plaque fragments, and many had maximal diameters smaller than the device pore size. This, combined with the lack of injury while the device is in place, seems to leave little debate as to its utility. Even with our aggressive examination of the embolic protection devices, we could not correlate fragment number or size with clinical events, or with new lesions on DWI. Given that the emboli causing new lesions occurred after the procedure, this lack of correlation should be expected.
Conclusion
We are reporting a series of 48 patients undergoing 54 CAS procedures with excellent clinical outcomes but a concerning number of new lesions on DWI. These subclinical brain injuries did not occur during the procedure but in the ensuing 48 hours, when transcranial Doppler studies have confirmed an ongoing number of embolic events. We submit that the CAS procedure itself is safe, but new approaches should be considered to prevent microembolization postprocedure.
Author contributions
References
- Cerebral ischemia and infarction from atheroemboli <100 micron in size. Stroke. 2003;34:1976–1980
- . Embolic risk of the different stages of carotid bifurcation balloon angioplasty: an experimental study. J Vasc Surg. 2000;31:550–557
- The amount of solid cerebral microemboli during carotid stenting does not relate to the frequency of silent ischemic lesions. AJNR Am J Neuroradiol. 2006;27:157–161
- Atheroemboli to the brain: Size threshold for causing acute neuronal cell death. J Vasc Surg. 2000;32:68–76
- . Is there any evidence that cerebral protection is beneficial: experimental data. J Cardiovasc Surg. 2006;47:127–136
- Analysis of emboli during carotid stenting with distal protection device. Cerebrovasc Dis. 2006;21:223–228
- Silent ischemia after neuroprotected percutaneous carotid stenting: a diffusion-weighted study. Am J Neuroradiol. 2006;27:1338–1345
- . Diffusion-weighted imaging abnormalities after percutaneous transluminal angioplasty and stenting for intracranial atherosclerotic disease. Am J Neuroradiol. 2005;26:385–389
- MR and clinical follow-up of diffusion-weighted cerebral lesions after carotid artery stenting. Am J Neuroradiol. 2005;26:2336–2341
- Incidence of new brain lesions after carotid stenting with and without cerebral protection. Stroke. 2006;37:2312–2316
- Diffusion-weighted MR imaging lesions after filter-protected stenting of high-grade symptomatic carotid artery stenoses. Am J Neuroradiol. 2006;27:1321–1325
- . Carotid angioplasty: detection of embolic signals during and after the procedure. Stroke. 1994;25:2403–2406
- Early detection of ischemic injury: comparison of spectroscopy, diffusion-, T2-, and magnetic susceptibility-weighted MRI in cats. Acta Neutochir Suppl. 1990;51:207–209
- . Diffusion-weighted magnetic resonance imaging: detection of eschemic injury 39 minutes after onset in a stroke patient. Ann Neurol. 1999;45:794–797
- Cerebral microembolization after protected carotid artery stenting in surgical high-risk patients: results of a 2 year prospective study. J Vasc Surg. 2005;42:847-33
- . Detection of clinically silent infarcts after carotid endarterectomy by use of diffusion-weighted imaging. Am J Neuroradiol. 2001;22:646–649
- . Assessment of silent embolism from carotid endarterectomy by use of diffusion-weighted imaging. Am J Neuroradiol. 2001;22:650–653
- Prospective analysis of cerebral infarction after carotid endarterectomy and carotid artery stent placement by using diffusion-weighted imaging. Am J Neuroradiol. 2005;26:376–384
- Cerebral emboli as a potential cause of Alzheimer’s disease and vascular dementia: case-control study. BMJ. 2006;332:1119–1124
Competition of interest: none.CME article
PII: S0741-5214(07)00007-9
doi:10.1016/j.jvs.2006.12.058
© 2007 The Society for Vascular Surgery. Published by Elsevier Inc. All rights reserved.